untitled

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 1 17-APR-17 10:20

TAXI REGULATION IN THE

AGE OF UBER

Katrina M. Wyman*

Uber and other apps for e-hailing taxis have transformed the taxi busi-

ness in recent years. However, to date, U.S. regulators have not been espe-

cially innovative in responding to the emergence of Uber and the other

apps. This article analyzes what taxi regulators should be doing in the age

of taxi apps, for both e-hailed taxis and traditional taxis obtained by phone,

by outstretched arm on the street, and at cab stands.

I make three main points. First, regulators should establish regulatory

standards for e-hailed and traditional taxis as a unit because they are sub-

stitutes, and in fact the same vehicle could be picking up people through

street hails, phone calls, at cab stands, and by app. Second, I address the

level of regulatory standards. Both traditional taxis and e-hailed taxis

should be regulated to address market failures for the most part, and this

means that there is a strong case for regulating traditional and e-hailed

taxis less onerously than traditional taxis historically have been. Different

standards might be applied to traditional and e-hailed taxis, but only if dif-

ferential standards are justified based on the benefits exceeding the costs.

Third, I analyze whether the incumbents in the traditional taxi industry that

bear the costs of the restructuring of the taxi industry should be compen-

sated by governments. I argue that there is no general economic case for

government compensation for the incumbents, who are principally the own-

ers of formerly valuable taxicab licenses. However, there may be a fairness

argument for compensating a select number of license owners. If govern-

ments opt to compensate license owners for fairness or political reasons,

they should do so through monetary compensation, not continued regulatory

protections.

To my knowledge, this is the first article offering a comprehensive,

theoretically grounded analysis of the appropriate role for taxi regulation in

the age of taxi apps, and whether incumbent operators should be compen-

sated to cushion the effects of the advent of taxi apps. Yet how to regulate

taxis, and whether to compensate traditional taxi operators for the losses

they face due to the new app entrants, are now pressing issues for taxi

regulators in the U.S. and around the world.

* Sarah Herring Sorin Professor of Law, New York University School of Law.

This article benefitted from presentations at New York University School of Law,

Seton Hall University School of Law, Kobe University Faculty of Law, the New York

City Bar Association Transportation Committee, and the American Law and Econom-

ics Association annual meeting; and from comments and suggestions from Daniel

Ackman, Yun-chien Chang, Matthew Devlin, Damien Geradin, Harry First, Scott

Hemphill, Marcel Kahan, Michael E. Levine, Jonathan Mazer, Mitchell Moss,

Jonathan Nash, Richard Revesz, Daniel Rubinfeld, Bruce Schaller, Frank Upham, and

many others. Sarah Krame, Aaron Lichter, Dana Rubin, and Andr´e Smith provided

research assistance. The editors of the N.Y.U. Journal of Legislation and Public Policy

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 2 17-APR-17 10:20

2 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

NTRODUCTION

.............................................. 2

I. C

HANGES

TTRIBUTABLE TO

AXI

PPS

.............. 8

II. R

EGULATORY

ESPONSES TO

AXI

PPS

.............. 16

III. T

COPE OF

AXI

EGULATION IN THE

GE OF

BER

............................................... 19

A. Joint Regulation of App and Traditional Taxis .... 20

B. Rethinking the Pillars of Taxi Regulation ......... 31

Entry Regulation ................................ 32

Fare Regulation ................................. 40

Consumer Safety Regulation ..................... 49

Worker Protections .............................. 57

Universal Service Requirements .................. 67

The New Anti-Monopoly Function ............... 75

Another Topic: Who Should Regulate Taxis? ..... 76

IV. T

RANSITION

ELIEF FOR

RADITIONAL

EDALLION

AXI

WNERS

....................................... 77

A. Whether Medallion Owners Should Be

Compensated ................................... 84

Efficiency Analysis.............................. 84

Fairness Arguments ............................. 93

Reliance Argument ............................ 93

Distributional Argument ....................... 96

B. The Form of Any Transition Relief .............. 98

ONCLUSION

................................................ 99

NTRODUCTION

In February 2014, New York City auctioned off a new batch of

168 individual taxicab licenses.

Often called medallions, these li-

censes allow those holding them to operate taxis that pick up passen-

gers on the street. At the time, the City’s auction seemed like a great

success: the number of bids was far higher than the number of licenses

that the City was selling, and the successful bidders committed to pay-

ing far more than the City’s minimum bid of $650,000 per medallion.

One euphoric buyer, an immigrant taxi driver from Bangladesh, told

provided excellent editorial assistance. The Filomen D’Agostino and Max E. Green-

berg Research Fund at New York University School of Law provided financial assis-

tance. All errors are mine. This article was last revised in late 2016 and early 2017.

1. Antonio Antenucci, Taxi Medallion Auctioned for Record-Setting $965,000,

N.Y. P

OST

(Feb. 26, 2014), http://nypost.com/2014/02/26/taxi-medallion-auctioned-

for-record-setting-965000/.

2. Id. (“[T]he city received 297 bids and the winners ranged from $965,000 to

$805,201.97.”). The auction was for individual accessible medallions, which must be

attached to vehicles that are accessible to people using wheelchairs.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 3 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 3

the New York Post that in buying a medallion with a partner for

$910,013, he had purchased an asset that would finance his retire-

ment.

Like many medallion owners before him, this driver assumed

that medallions would continue to increase in value, and that he would

be able to lease the medallion to other drivers to generate cash when

he no longer wanted to drive. But by the end of 2014, medallion val-

ues were falling precipitously and a good number of those who pur-

chased medallions at the City’s recent auctions were suffering from

buyer’s remorse.

Almost a year later, in November 2015, some me-

dallion owners suggested in court papers that the private secondary

market for medallions had deteriorated so much that it was “frozen.”

What happened? In 2014, Uber,

then only five years old, began

making its mark on the New York taxi industry, “poach[ing] . . . [me-

3. Id. (“‘I’ve been waiting for this for years. I missed out the last time,’ said Sadar

Jaman, 44, who emigrated from Bangladesh 22 years ago and has been a cabbie for

ten years . . . . Now, Jaman can think about his post-cabbie days. ‘If I want to retire in

ten years it’s easier if I have a medallion.’”).

4. In fact, “58 prospective buyers who bid $47 million in” the auctions “back[ed]

out” of completing their purchases. Jennifer Fermino, Yellow Taxi Industry Loses 58

Prospective Buyers in Auction for Medallions,

N.Y. D

AILY

EWS

(Dec. 10, 2014),

http://www.nydailynews.com/new-york/yellow-taxi-industry-loses-58-bidders-medal

lions-article-1.2039989. It is unclear from Fermino, supra, which auction (or auc-

tions) the buyers backed out from. In addition to the February 2014 auction mentioned

in the text, New York City auctioned minifleet accessible medallions in 2014 and

2013. Current and Past Auction Results,

N.Y.C. T

AXI

& L

IMOUSINE

OMM

’

, http://

www.nyc.gov/html/tlc/html/industry/medal-

lion_auction_current_and_past_auction_results.shtml (last visited Jan. 22, 2017).

5. Complaint at 12, Melrose Credit Union v. City of New York, No. 1:15-cv-

09042 (S.D.N.Y. Nov. 17, 2015) (“[T]he value of the medallion has plummeted by

more than 40%, and the once vibrant market for the purchase and sale of New York

City taxicab medallions is now frozen.”); see also Amended Complaint at 12, Melrose

Credit Union v. City of New York, No. 1:15-cv-09042 (S.D.N.Y. Mar. 7, 2016)

(same).

6. This article uses the terms Uber, taxi apps, app-dispatched taxis, app-hailed

taxis, taxis summoned by apps, and transportation network companies to refer to the

new entrants to the taxi business that use apps to match taxi drivers and passengers.

As my choice of terminology indicates, I consider Uber and the other apps to be in the

taxi business, providing point-to-point transportation like traditional taxis, which until

recently have been summoned by street hailing (passengers putting up their arms),

waiting at cab stands, and phoning ahead. In using the term “taxi” to refer to Uber and

other app-dispatched vehicles, I am departing from the tendency among regulators in

cities such as New York to refer to these vehicles as “for-hire vehicles,” while keep-

ing the term “taxi” for traditional taxis that are street hailed, or summoned at cab

standards, or by phoning ahead. I do so because, as the article explains, app-dis-

patched taxis and traditional taxis are best regarded as being in the same business

these days given the congruence of technologies. In suggesting that Uber and similar

apps are in the taxi business, I also implicitly am rejecting the argument that the apps

are not in the taxi business, but rather in the technology or the logistics business. It is

true that these app-based services rely on sophisticated technology, and that some

offer more services than traditional taxi companies, such as food delivery. This article

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 4 17-APR-17 10:20

4 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

dallion] cab customers by lowering prices and expanding its fleet,

from 7,000 to 16,000.”

For the first time in decades, medallion taxis

now face competition from new upstarts not required to hold medal-

lions, because the vehicles dispatched by Uber and other apps operate

under a different, less stringent, regulatory framework.

Today, regulators in New York City and many other places in the

U.S. and around the world are struggling to recast taxi regulation,

given the ways that Uber and other taxi apps have fundamentally

transformed the market for “point-to-point” transportation. Reacting to

the arrival of Uber on their doorsteps, U.S. regulators to date have not

been nearly as innovative in their responses to the emergence of the

taxi apps as the apps have been in changing the taxi business. This

article analyzes what taxi regulators should be doing in the age of taxi

apps, for both taxis summoned by apps and traditional taxis obtained

by phone, by hailing on the street or at cab stands. I make three main

points.

First, regulators should regulate app-dispatched and traditional

taxis as a unit because they are substitutes, and in fact the same vehi-

cle could be picking up people through street hails, phone calls, at cab

focuses only on the business of facilitating the provision of point-to-point

transportation.

The article uses the terms traditional taxis, incumbents, incumbent taxi industry,

and traditional taxi operations or companies to refer to the participants in the taxi

business before the creation of Uber. Until recently, traditional taxis provided taxi

service in response to street hails, by waiting at cab stands, or through prearranged

rides booked by phone. Traditional taxi companies now also are deploying apps.

7. Simon Van Zuylen-Wood, The Struggles of New York City’s Taxi King,

LOOMBERG

(Aug. 27, 2015), http://www.bloomberg.com/features/2015-taxi-medal

lion-king/; see

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

RANSP

. R

ESEARCH

., S

PECIAL

EPORT

319,

ETWEEN

UBLIC

& P

RIVATE

O-

BILITY

: E

XAMINING THE

ISE OF

ECHNOLOGY

-E

NABLED

RANSPORTATION

ERVICES

21 (2015) [hereinafter

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

]

(“As of March 2015, the number of Uber vehicles in New York had overtaken the

number of medallion cabs.”). Uber’s push into New York City also coincided with the

licensing of green taxis that are allowed to pick up street hails in New York City’s

outer boroughs, in competition with yellow taxis. Fermino, supra note 4. New York

City has suggested that the City’s licensing of green taxis starting in 2013 may have

contributed to the decline in medallion values. Brief for Respondents at 33, 40, Glyka

Trans LLC v. City of New York, No. 2015-11661 (N.Y. App. Div. April 15, 2016).

However, it is highly likely that Uber’s arrival has had a much larger effect on medal-

lion values than the green taxis because the green taxis operate in areas that the me-

dallion taxis historically rarely served. In addition, after some initial success, the

green taxi business itself has declined due to competition from Uber. Matthew Flamm,

Facing Uber and Accessibility Hurdles, Green Taxi Operators Are Seeing Red,

RAIN

’

N.Y. B

, (Sept. 25, 2016), http://www.crainsnewyork.com/article/2016

0925/SMALLBIZ/160929932/the-great-green-cab-experiment-is-on-the-brink-of-

crumbling-against-uber-and-wheelchair-accessibility-requirements-that-many-green-

cab-operators-say-randomly-change.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 5 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 5

stands and by app. This is not what U.S. regulators currently are do-

ing; instead, they often are reacting to the arrival of app-hailed taxis

by establishing distinct, relatively light levels of regulation for them

without simultaneously revising the burdensome regulations gov-

erning traditional taxis, leaving traditional taxis at a competitive

disadvantage.

8. This differential treatment is leading to many claims from the traditional taxi

industry that it is unfairly burdened by a more stringent regulatory regime than the

new entrants, some of which are formulated as legal claims invoking the Equal Pro-

tection Clause. See, e.g., Ill. Transp. Trade Ass’n v. City of Chicago, 839 F.3d 594

(7th Cir. 2016) (reversing district court and ordering dismissal of plaintiffs’ equal

protection claims); VTS Transp. v. Palm Beach County, No. 9:15-cv-80560 (S.D. Fla.

Mar. 8, 2017) (granting summary judgment to defendant Palm Beach County in plain-

tiffs’ equal protection challenge to County’s Temporary Operating Agreement with

Uber); Cambridge Taxi Drivers & Owners Ass’n. v. City of Cambridge, No. 16-

11357-NMG, 2017 WL 373491 (D. Mass. Jan. 25, 2017) (dismissing equal protection

claim because state law preempts local regulation of transportation network compa-

nies); Bos. Taxi Owners Ass’n v. Baker, No. 16-11922-NMG, 2017 WL 354010 (D.

Mass. Jan. 24, 2017) (dismissing equal protection claim against Massachusetts defen-

dants); Newark Cab Ass’n v. City of Newark, No. 2:16-cv-04681, 2017 WL 214075

(D.N.J. Jan. 17, 2017) (dismissing equal protection claim); Desoto Cab Co. v. Picker,

No. 15-cv-04375-EMC, 2017 WL 118810 (N.D. Cal. Jan. 12, 2017) (dismissing plain-

tiff taxi company’s equal protection claim); Bos. Taxi Owners Ass’n v. City of Bos-

ton, No. 15-10100-NMG, 2016 WL 7410777 (D. Mass. Dec. 21, 2016) (dismissing

equal protection claim against Boston because state legislation largely preempts local

regulation of transportation network companies); Bos. Taxi Owners Ass’n v. City of

Boston, No. 15-10100-NMG, 2016 WL 1274531 (D. Mass. Mar. 31, 2016) (refusing

to dismiss equal protection claims against City of Boston, but denying preliminary

injunction); Gebresalassie v. District of Columbia, 170 F. Supp. 3d 52 (D.D.C. 2016)

(granting defendant’s motion to dismiss); Melrose Credit Union v. City of New York,

No. 1:15-cv-09042 (S.D.N.Y. Jan. 26, 2016) (denying plaintiffs’ motion for prelimi-

nary injunction); Joe Sanfelippo Cabs, Inc. v. City of Milwaukee, 46 F. Supp. 3d 888

(E.D. Wis. 2014), aff’d on other grounds, 839 F.3d 613 (7th Cir. 2016) (denying

plaintiffs’ motion for preliminary injunction); Taxicab Paratransit Ass’n of Cal. v.

Cal. Pub. Utils. Comm’n, No. S220982 (Cal. Nov. 12, 2014) (denying the petition for

review); Second Amended Class Action Complaint, Ruffino v. Broward County, No.

0:15-cv-62312 (S.D. Fla. Sept. 20, 2016); Order of Supplemental Motion to Dismiss,

Ruffino, No. 0:15-cv-62312 (granting defendants’ motion to dismiss); Order of Dis-

missal Without Prejudice, Ruffino, No. 0:15-cv-62312 (dismissing the case without

prejudice pursuant to plaintiffs’ request for voluntary dismissal); Second Amended

Complaint, Checker Cab v. Phila. Parking Authority, No. 2:16-cv-4669 (E.D. Pa. Jan.

4, 2017); Amended Complaint, Checker Cab v. Phila. Parking Auth., No. 2:16-cv-

04669 (E.D. Pa. Nov. 4, 2016); Amended Class Action Complaint and Demand for a

Jury Trial, Miadeco Corp. v. Miami-Dade County, No. 16-4244 (Fla. Cir. Ct. May 4,

2016).

European taxi drivers are making similar arguments that Uber is engaging in

“‘unfair competition’” because it “does not comply with taxi regulations.” Damien

Geradin,

HOULD

BER

LLOWED TO

OMPETE IN

UROPE

? A

? 3

(Competition Policy Int’l June 2015

)

https://www.competitionpolicyinternational.

com/assets/Europe-Column-New-Format.pdf.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 6 17-APR-17 10:20

6 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

Second, I argue that traditional taxis and e-hailed taxis generally

should be regulated only to address market failures. This means that

while there is still a place for regulation to protect the public, regula-

tors should ease the regulatory burden on traditional taxis and avoid

imposing unnecessary regulation on app-dispatched taxis. There still

may be a basis for different regulatory standards for traditional and e-

hailed taxis but any differences in the standards should be justified

based on the benefits of differential standards exceeding the costs.

The changes that I endorse will mean large losses for segments of

the incumbent taxi industry in the short term, and my third point con-

cerns whether governments should compensate these incumbents for

their losses. Traditional taxi operations currently are arguing for com-

pensation for the losses that they are suffering in U.S. jurisdictions

where Uber is being allowed to operate, sometimes in legal claims

invoking the Takings Clause in the U.S. Constitution.

I argue that

there is no compelling economic case for compensating the incum-

bents in the traditional taxi industry. There may be a case on fairness

grounds for compensating some owners of traditional taxicab licenses.

If regulators choose to compensate traditional taxicab license holders

on fairness grounds or for political reasons, regulators should do so

through monetary payments, not regulatory protections that inhibit

competition from the new apps.

9. To date the courts have rejected the traditional taxi industry’s claims for com-

pensation under the Takings Clause. See Ill. Transp. Trade Ass’n, 839 F.3d 594 (af-

firming district court’s dismissal of Takings claim); Joe Sanfelippo Cabs, Inc., 839

F.3d 613 (same); Cambridge Taxi Drivers & Owners Ass’n, No. 1:16-cv-11357-

NMG, 2017 WL 373491 (dismissing Takings claim); Bos. Taxi Owners Ass’n v.

Baker, No. 16-11922-NMG, 2017 WL 354010 (dismissing Takings claim against

Massachusetts defendants); Newark Cab Ass’n, No. 2:16-cv-04681, 2017 WL 214075

(dismissing Takings claim); Bos. Taxi Owners Ass’n v. City of Boston, No. 15-

10100-NMG, 2016 WL 1274531 (dismissing Takings claim); Bos. Taxi Owners

Ass’n v. City of Boston, 84 F. Supp. 3d 72 (D. Mass. 2015) (denying plaintiffs’ mo-

tion for preliminary injunction); Abramyan v. State, No. 2015CV262742 (Ga. Super.

Ct., Mar. 22, 2016) (granting defendants’ motion to dismiss Takings claim); Glyka

Trans LLC v. City of New York, No. 8962/15, 2015 WL 5320868 (N.Y. Sup. Ct.

Sept. 8, 2015) (dismissing Takings claims); Second Amended Class Action Com-

plaint, Ruffino, No. 0:15-cv-62312; Order on Supplemental Motion to Dismiss, Ruf-

fino, No. 0:15-cv-62312 (granting defendants’ motion to dismiss); Order of Dismissal

Without Prejudice, Ruffino, No. 0:15-cv-62312 (dismissing the case without prejudice

pursuant to plaintiffs’ request for voluntary dismissal); see also Gebresalassie, 170 F.

Supp. 3d at 52 (rejecting the substantive due process claim that District of Columbia

had deprived plaintiffs of a protected property interest by legalizing “digital

dispatch”).

There has been no adjudication yet of the Takings claims from the taxi industry

or its creditors in: Amended Complaint, Melrose Credit Union, No. 1:15-cv-09042;

Amended Class Action Complaint and Demand for a Jury Trial, Miadeco Corp., No.

16-4244; Amended Complaint, Checker Cab, No. 2:16-cv-04669.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 7 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 7

This article’s principal contribution is to offer a theoretically

grounded proposal for taxi regulation in the age of apps. To my

knowledge, there is no existing legal scholarship offering a compre-

hensive assessment of the appropriate role for taxi regulation in the

age of taxi apps, and whether incumbent operators should be compen-

sated to cushion the effects of the advent of taxi apps.

Yet how to

10. However, legal and other scholars are beginning to address some of the regula-

tory implications of taxi apps. See, e.g.,

RUN

UNDARARAJAN

, T

HARING

CON-

OMY

: T

ND OF

MPLOYMENT AND THE

ISE OF

ROWD

-B

ASED

APITALISM

(2016);

Jordan M. Barry & Elizabeth Pollman, Regulatory Entrepreneurship, 90

S. C

. L.

(forthcoming 2017); Molly Cohen & Arun Sundararajan, Self-Regulation and

Innovation in the Peer-to-Peer Sharing Economy, 82

U. C

. L. R

. D

IALOGUE

116

(2015); Benjamin G. Edelman & Damien Geradin, Efficiencies and Regulatory Short-

cuts: How Should We Regulate Companies Like Airbnb and Uber?, 19

TAN

. T

ECH

L. R

293 (2016); Shu-Yi Oei & Diane M. Ring, Can Sharing Be Taxed?, 93

ASH

. U. L. R

. 989

(2016); Sofia Ranchord´as, Does Sharing Mean Caring? Regu-

lating Innovation in the Sharing Economy, 16

INN

. J.L. S

. & T

ECH

. 413 (2015);

Daniel E. Rauch & David Schleicher, Like Uber, but for Local Governmental Policy:

The Future of Local Regulation of the Sharing Economy, 76

HIO

. L.J. 901

(2015)

; Brishen Rogers, The Social Costs of Uber, 82

U. C

. L. R

. D

IALOGUE

(2015); Kellen Zale, Sharing Property, 87

U. C

OLO

. L. R

501 (2016); Richard

Darb´era, Principles for the Regulation of For-Hire Road Transport Passenger Ser-

vices (Int’l Transp. Forum, Corp. P’ship Bd. Discussion Paper, Oct. 11, 2015), http://

www.itf-oecd.org/sites/default/files/docs/1-discussion_paper_darbera_rd15-itf-oecd.

pdf; Geradin, supra note 8, at 10–11; E. Glen Weyl & Alexander White, Let the Best

‘One’ Win: Policy Lessons from the New Economics of Platforms (Univ. of Chi. Law

Sch. Coase-Sandor Inst. for Law & Econ. Working Paper No. 709, 2014); Eric Pos-

ner, Why Uber Will—and Should—Be Regulated,

LATE

(Jan. 5, 2015), http://

www.slate.com/articles/news_and_politics/view_from_chicago/2015/01/uber_surge

_pricing_federal_regulation_over_taxis_and_car_ride_services.html; K. Sabeel

Rahman, Curbing the New Corporate Power,

. R

. (May 4, 2015), http://bos-

tonreview.net/forum/k-sabeel-rahman-curbing-new-corporate-power (and accompany-

ing responses); Cass R. Sunstein, Uber Cab App Threatens Death of Taxi Dinosaurs,

LOOMBERG

(June 10, 2013), https://www.bloomberg.com/view/articles/2013-06-10/

uber-cab-app-threatens-death-of-taxi-dinosaurs.

Students have published notes on some of the issues raised by taxi apps. See,

e.g., Katie Barglind, Note, Innovation, Technology, and Transportation: The Need to

Address On-Demand Ridesharing and Modernize Outdated Taxi Regulations in the

U.S., 33

. I

’

L.J.

701 (2015); Emily Dobson, Note, Transportation Network

Companies: How Should South Carolina Adjust Its Regulatory Framework?, 66

S.C.

L. R

. 701 (2015); Rebecca Elaine Elliott, Note, Sharing App or Regulation

Hack(ney)? Defining Uber Technologies, Inc., 41

J. C

ORP

. L.

727 (2016); Caleb Hol-

loway, Comment, Uber Unsettled: How Existing Taxicab Regulations Fail to Address

Transportation Network Companies and Why Local Regulators Should Embrace

Uber, Lyft, and Comparable Innovators, 16

AKE

OREST

J. B

. & I

NTELL

. P

ROP

. L.

20 (2015); Josh Krauss, Note, The Sharing Economy: How State and Local Govern-

ments are Failing and Why We Need Congress to Get Involved, 44

. L. R

365

(2014); Hannah A. Posen, Note, Ridesharing in the Sharing Economy: Should Regula-

tors Impose Uber Regulations on Uber?, 101

OWA

L. R

. 405 (2015); K. Casey

Strong, Comment, When Apps Pollute: Regulating Transportation Network Compa-

nies to Maximize Environmental Benefits, 86

U. C

OLO

. L. R

1049 (2015); David K.

Suska, Regulatory Takings and Ridesharing: “Just Compensation” For Taxi Medal-

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 8 17-APR-17 10:20

8 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

regulate taxis, and whether to compensate traditional taxis for the

losses they face due to the new app entrants, are now pressing issues

for taxi regulators in the U.S. and around the world.

This article has four main parts. Part I briefly describes the tech-

nological and organizational changes introduced by the taxi apps. Part

II categorizes the ways that U.S. taxi regulators have responded to the

emergence of the taxi apps. Part III analyzes the appropriate role of

taxi regulation in the age of apps, for app-dispatched taxis and tradi-

tional taxis. Part IV analyzes whether actors in the incumbent taxi in-

dustry should be compensated, and if so whether compensation should

take the form of monetary compensation or regulatory protection. I

then briefly conclude.

HANGES

TTRIBUTABLE TO

AXI

PPS

Uber and the other taxi apps have brought two major changes to

the taxi industry.

The first is new technology that reduces the transaction costs of

taxi riding. The taxi apps offer a new way of matching taxi drivers

with riders that significantly reduces the “search costs”

involved in

finding a taxi. To quote Brishen Rogers, “[r]ather than calling a dis-

patcher and waiting, or standing on the street, users can hail a car from

indoors and watch its progress toward their location.”

In addition,

lion Owners?, 19 N.Y.U. J.

EGIS

. & P

. P

’

183 (2016); see also Daniel Gross-

baum, An Uber Issue: Regulating Ridesharing Companies in the Modern Day

Economy (unpublished manuscript) (on file with author).

There also are some recent reports on the appropriate role of taxi regulation in

the age of Uber, or issues related to this topic. See, e.g.,

CHALLER

ONSULTING

NFINISHED

USINESS

: A B

LUEPRINT FOR

BER

, L

YFT AND

AXI

EGULATION

(2016)

http://www.schallerconsult.com/rideservices/blueprint.pdf

; C

OMM

FOR

EVIEW OF

N-

NOVATIVE

RBAN

OBILITY

ERVS

supra note 7;

OMPETITION

UREAU

, G

’

TOF

., M

ODERNIZING

EGULATION IN THE

ANADIAN

AXI

NDUSTRY

(2015), http://

www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/04007.html; Transcript of Work-

shop, The “Sharing” Economy: Issues Facing Platforms, Participants, and Regula-

tors,

. T

RADE

OMM

’

(June 9, 2015), https://www.ftc.gov/system/files/

documents/public_events/636241/sharing_economy_workshop_transcript.pdf [herein-

after FTC Workshop].

Edelman & Geradin, cited near the start of this footnote, is likely the closest

article to mine in scope and ambition. Geradin, supra note 8, at 11–12, briefly argues

against compensating the owners of taxi licenses.

11. For other discussions of the changes, see, for example, Rogers, supra note 10,

at 86–90, and Geradin, supra note 8, at 5–6 (discussing “Uber’s disruptive business

model”).

12. See, e.g., Rogers, supra note 10, at 88.

13. Id. See also Edelman & Geradin, supra note 10, at 297.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 9 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 9

Uber simplifies the process of paying for the ride by automatically

billing a credit card account stored with the app.

The second change is to the industrial organization of the taxi

industry. Historically the taxi industry in many places has been

“highly fragmented,” both horizontally and vertically.

Taxi opera-

tions have tended to be city-specific, with little cross-ownership across

cities, at least in recent times.

Moreover, within cities taxi operations

have been segmented. For example, in New York City, city and state

regulations have segmented the industry into medallion (also called

“yellow”) taxis with the right to pick up riders through street hails and

by prearrangement;

“community liveries” and “black cars” (collec-

14. Dobson, supra note 10, at 703–04.

15. Rogers, supra note 10, at 89.

16.

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note 7, at

60–61. I do not mean to imply there is no common ownership across cities. See, e.g.,

Jared Meyer, New York’s Taxi King is Going Down,

EDERALIST

(Oct. 26, 2015),

http://thefederalist.com/2015/10/26/new-yorks-taxi-king-is-going-down (profiling

Gene Friedman, “who at one point owned more than 1,000 New York City taxi me-

dallions” and also owned medallions in “New Orleans, Philadelphia, and Chicago”).

Early in the twentieth century, there were efforts to create “multicity taxi empires.”

ORMAN

ILBERT

& R

OBERT

E. S

AMUELS

, T

AXICAB

: A

RBAN

RANSPORTA-

TION

URVIVOR

72 (1982).

17. These taxis are called medallion taxis because they are required to have one of a

limited number of medallions to operate. A medallion is a physical object that is on

the hood of the taxi. The medallion is the physical evidence that the taxi has one of a

limited number of licenses to operate as a taxi picking up people on the street. Medal-

lion taxis are also called yellow taxis because the taxis are painted yellow.

From the late 1980s until the last few years, the ability of medallion taxis to pick

up passengers through prearrangement was constrained by regulatory limits on the

technology that medallion taxi drivers could use. “Starting in the mid-sixties, taxis

began forming associations and serving customers through radio calls. . . . As the

number of taxis responding to radio calls grew, it became increasingly difficult for

passengers to hail a taxi on the street.” Brief for Respondents at 12–13, Glyka Trans

LLC v. City of New York, No. 2015-11661 (N.Y. App. Div. Apr. 15, 2016). In the

1980s, to increase the number of taxis responding to street hails, the Taxi and Limou-

sine Commission urged “taxi owners” to remove two-way radios from taxis and put

them in “non-medallion vehicles known as ‘black cars.’” Id. at 14. Eventually, in the

mid-1980s, the Commission prohibited taxis from “arranging rides through ‘an actual

two-way radio device in their vehicle’” through a “regulation” that remains on the

books. Id. at 29 (referring to 35 RCNY § 58-34(b)(3)); see also id. at 15–16. The

Commission created another regulatory barrier to medallion taxis prearranging rides

in 1999, when it “prohibited all drivers—taxi drivers and for-hire vehicle drivers—

from using cell phones while driving.” Id. at 21. The regulatory limits on the use of

cell phones subsequently were expanded to cover “all portable electronic devices, not

just cell phones.” Id. at 21. But recognizing that such devices could assist in the

dispatch of for-hire vehicles, the Commission allowed for-hire vehicle drivers to make

limited use of devices “in connection with a dispatch from a base,” while maintaining

the prohibition on medallion taxi drivers using such devices unless their vehicles were

“lawfully standing or parked.” Id. at 22 (citing 35 RCNY § 6-16(u); 35 RCNY § 2-

25). Even within the confines of these limitations, medallion taxis might have prear-

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 10 17-APR-17 10:20

10 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

tively called “for-hire vehicles”) that are allowed to pick up riders

through prearrangement but not street hails; and green taxis allowed to

pick up riders through street hails and prearrangement but only in

northern Manhattan and the outer boroughs.

Nothing I have found in

my research for this article suggests that cross-ownership is common

across the various segments within the City.

In cities limiting the number of taxi vehicles, the industry also

has been vertically fragmented. Consider New York City’s yellow me-

dallion taxi sector as an example. The drivers of yellow taxis are

independent contractors, not employees.

Because they are indepen-

dent contractors, they receive few benefits, and standard employment

ranged rides, for example, by drivers informally agreeing to pick up passengers at

appointed times and places. When smartphone apps first began to be used to summon

transportation in the 2010s, medallion taxis could not use them due to the prohibition

on drivers using “electronic device[s] while driving.” Id. at 27. The Commission al-

lowed medallion taxis to respond to app requests for service on a pilot basis in

2013–2014, and in 2015 formally changed its rules to allow medallion taxis to re-

spond to apps. Id. at 27–30; see also Brief for Respondents at 39, Melrose Credit

Union v. City of New York, No. 2016-02214 (N.Y. App. Div. Aug. 17, 2016).

18. For a description of the various segments of the New York City taxi industry,

see Glyka Trans LLC v. City of New York, No. 8962/15, 2015 WL 5320868, at *1–2

(N.Y. Sup. Ct. Sept. 8, 2015). Green taxis are allowed to pick up passengers by prear-

rangement at the airports, as well as in northern Manhattan and the outer boroughs.

Your Guide to Boro Taxis,

N.Y.C. T

AXI

& L

IMOUSINE

OMM

’

, http://www.nyc.gov/

html/tlc/html/passenger/shl_passenger.shtml (last visited Jan. 22, 2017).

19. Ahmed v. City of New York, 129 A.D.3d 435, 437 (N.Y. App. Div. 2015). U.S.

taxi drivers generally are classified as independent contractors. See

IRECTORATE FOR

. & E

NTER

. A

FFAIRS

OMPETITION

OMM

., W

ORKING

ARTY

. 2

OMPETI-

TION

& R

EGULATION

, T

AXI

ERVICES

EGULATION AND

OMPETITION

— U

NITED

TATES

3 n.6 (Oct. 15, 2007), https://www.ftc.gov/sites/default/files/attachments/us-

submissions-oecd-and-other-international-competition-fora/ustaxis.pdf (“Gilbert et

al. . . . report, based on their survey, that 91% of U.S. taxi drivers are now indepen-

dent contractors rather than employees.”) (citing

ORMAN

ILBERT ET AL

., T

RANSP

ESEARCH

., N

’

ESEARCH

OUNCIL

, T

OLE OF THE

RIVATE

FOR

-H

IRE

E-

HICLE

NDUSTRY IN

UBLIC

RANSIT

23 (2002));

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

supra note 7, at 19 (“[M]ost taxi drivers are independent

contractors.”); Seth D. Harris & Alan B. Krueger, A Proposal for Modernizing Labor

Laws for Twenty-First-Century Work: The “Independent Worker” 26 (Hamilton Pro-

ject

Discussion Paper No. 2015-10, 2015) (referring to the taxi industry as an industry

“where most of the work is already conducted by independent contractors”). For indi-

cations that taxi drivers in cities other than New York are classified as independent

contractors, see Sebago v. Bos. Cab Dispatch, Inc., 28 N.E.3d 1139 (Mass. 2015)

(Boston “licensed taxicab drivers” are independent contractors); Parks Cabs Co. v.

Annunzio, 412 Ill. 549, 555 (1952) (cab drivers that lease licenses are not employees

under Unemployment Compensation Act);

OLO

. P

. U

TIL

. C

OMM

’

AXI

N-

DUSTRY IN THE

ENVER

ETROPOLITAN

REA

4–5 (2008), http://www.taxi-library.

org/denver-2008-puc-report.pdf (Denver taxi drivers are independent contractors).

Taxi drivers may be classified as employees for some purposes but not others.

Yellow Cab Coop., Inc. v. Workers’ Comp. Appeals Bd., 226 Cal. App. 3d 1288,

1296 n.5 (1991) (citing case law holding that Chicago cab drivers are not employees

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 11 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 11

protections such as minimum wage laws do not protect taxi drivers.

The drivers often lease the medallion

and the vehicle required to

drive a taxi. Sometimes these are leased directly from owners of me-

dallions and taxicabs; sometimes drivers lease medallions (and/or ve-

hicles) from brokers who are agents for medallion owners.

Uber is a new, more horizontally integrated business model for

the taxi industry.

Uber is not the typical single-city local taxi opera-

tor. It is a “virtual global fleet manager”

with many drivers, in many

cities around the world, available to serve the growing numbers of

riders who use its app.

Within a single city like New York, Uber

upends the traditional segmentation of taxi services. An app-dis-

patched taxi is a hybrid between a street-hailed and prearranged taxi.

It often arrives with almost the same immediacy as a street-hailed taxi,

but the use of the app to obtain the taxi and the short wait time be-

tween the request and the arrival of the vehicle echo the steps involved

in getting a taxi by calling ahead.

But an Uber vehicle also might

under the National Labor Relations Act, and case law holding that drivers are covered

by workers’ compensation).

20.

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note 7, at

84, 86–87.

21. Katrina Miriam Wyman, Problematic Private Property: The Case of New York

Taxicab Medallions, 30

ALE

125, 137 (2013).

22. Rogers, supra note 10, at 89. See also the description of the “[m]odels of opera-

tion” of New York medallion taxis in

N.Y.C. T

AXI

& L

IMOUSINE

OMM

’

, 2016 TLC

ACTBOOK

2 (2016), http://www.nyc.gov/html/tlc/downloads/pdf/2016_tlc_fact

book.pdf; and the description of the “four business models under which a taxicab may

be put into service” under Boston taxi regulations in Sebago, 28 N.E.3d at 1144.

23. See Dobson, supra note 10; Rogers, supra note 10, at 90.

24. I am borrowing this term from Justin Jenk, Theory Meets Practice in the Taxi

Industry: Coase and Uber 2 (Munich Pers. RePEc Archive, Working Paper No.

63206, 2015) (describing mobile apps as “manag[ing] virtual fleets”).

25. See

N.Y.C. C

OUNCIL

OMM

RANSP

., R

EPORT OF THE

UMAN

ERVICES

IVISION

6 (June 30, 2015) [hereinafter

EPORT OF THE

UMAN

ERVICES

IVISION

]

;

Dana Rubinstein, Uber’s Kalanick Makes a Pitch for a More ‘Prosperous City,’

O-

LITICO

(June 3, 2015), http://www.politico.com/states/new-york/city-hall/story/2015/

06/ubers-kalanick-makes-a-pitch-for-a-more-prosperous-city-023469; cf.

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

supra note 7, at 106 (noting that

after midnight, more Uber drivers are available than are taxis).

26. See

. P

. U

TILS

. C

OMM

’

, R. 12-12-001,

ECISION

DOPTING

ULES AND

EGULATIONS TO

ROTECT

UBLIC

AFETY

HILE

LLOWING

NTRANTS TO THE

RANSPORTATION

NDUSTRY

20–21 (2013) [hereinafter CPUC] (explaining why

“Transportation Network Companies” provide prearranged service).

Whether e-hailing should be classified as hailing (or street hailing) on the one

hand, or prearrangement on the other, is currently disputed in New York City. Under

existing legislation and regulations, medallion taxis have a monopoly on street hails

(except in northern Manhattan and the outer boroughs, where green taxis also are

allowed to pick up street hails). See Act of Feb. 17, 2012, 2012 N.Y. Sess. Laws 9,

§ 11 (McKinney) (amending Ch. 602 of Laws of New York 2011); N.Y.C. Admin.

Code §§ 19-502(l), 19-504(a)(1), 19-507(a)(4), 19-516(a) (West 2016); 35 RCNY

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 12 17-APR-17 10:20

12 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

§ 55-19(a); see also Brief for Respondent Eric T. Schneiderman at 4–5, Melrose

Credit Union v. City of New York, No. 2016-02214 (N.Y. App. Div. Aug. 22, 2016)

(noting that the “supposed exclusive right of the taxicab industry to serve the street-

hail market in New York City . . . arises from three sources,” citing the HAIL Act, the

New York City Administrative Code, and Taxi and Limousine Commission Rules). If

e-hailing were considered street hailing, only medallion taxis (and green taxis in

northern Manhattan and the outer boroughs) would be able to respond to e-hails, and

black cars, which include Uber cars, would no longer be able to respond to e-hails

unless they complied with the same rules as medallion taxis. Thus, not surprisingly,

the traditional taxi industry argues that e-hailing through Uber and other taxi apps is

the equivalent of hailing or street hailing, not prearrangement. Classifying e-hailing as

street hailing would prevent Uber from operating under its current business model.

See Complaint at 19, CGS Taxi LLC v. City of New York, No. 653264/2015 (N.Y.

Sup. Ct. Sept. 30, 2015) (“An ‘e-hail’ is a hail, not a pre-arrangement.”); Complaint at

18, Singh v. City of New York, No. 701402/2017 (N.Y. Sup. Ct. Jan. 30, 2017)

(same); Brief for Petitioners-Appellants at 37, Glyka Trans LLC v. City of New York,

No. 2015-11661 (N.Y. App. Div. Nov. 30, 2015) (“E-hails are equivalent to the tradi-

tional hail and are therefore exclusively in the domain of medallion taxicab ser-

vices.”); Brief for Petitioners-Appellants at 2, Glyka Trans LLC, No. 2015-11661

(“An e-hail is a hail, not a prearrangement . . . . ”); Reply Brief for Petitioners-Appel-

lants at 1, Glyka Trans LLC, No. 2015-11661 (“[I]f an e-hail is a hail, then an e-hail

transmitted from the street is a street hail . . . .”); Amended Complaint at 41, Melrose

Credit Union v. City of New York, No. 1:15-cv-09042 (S.D.N.Y. Mar. 7, 2016)

(“[E]lectronic ‘hails’ constitute a modern form of traditional street hail—and thus,

belong exclusively to medallion taxicabs.”).

New York City maintains that “all trips arranged by smartphone app [are] . . .

prearrangements.” Brief for Respondents at 43, Glyka Trans LLC, No. 2015-11661;

see also Brief for Respondents at 53, Melrose Credit Union, No. 2016-02214 (“[T]he

use of an app to arrange a ride is a prearrangement . . . . ”). The City denies that an e-

hail is a “street hail,” but nonetheless maintains “that an e-hail is a hail” (as well as a

prearrangement). Brief for Respondents at 57, Melrose Credit Union, No. 2016-

02214. The City argues that “hail” is a generic term for summoning a taxi that encom-

passes street hails and prearrangement, and that e-hails are a sub-category of prear-

rangement. Brief for Respondents at 53, Glyka Trans LLC, No. 2015-11661 (“The

Commission’s rules . . . describ[e] several ways to hail a taxi. One is a street hail,

which involves ‘a verbal (audio) action such as calling out, yelling, or whistling, and/

or a visible physical action such as raising one’s hand or arm.’ 35 R.C.N.Y. § 51-03.

Another way to hail a taxi is ‘through an electronic method’ involving a smartphone

app. Id. Both are hails, but only the first is a street hail; the second is a prearrange-

ment. . . . [S]treet hails and prearrangements are two distinct subsets of ‘hails . . . .’”).

By characterizing e-hails as a form of prearrangement, rather than a street hail, the

City presumably is trying to protect its decision to allow for-hire vehicles (which

include black cars), as well as medallion and green taxis, to respond to e-hails. Neither

City nor New York State law grants any industry segment a monopoly on providing

prearranged service.

In the ongoing litigation, the traditional industry argues that the City has taken

inconsistent positions on whether e-hails are a form of pre-arrangement. Brief for

Petitioners-Appellants at 17–18, Glyka Trans LLC, No. 2015-11661 (“Without any

explanation, however, the TLC has now taken the bizarre position that an e-hail is a

hail when it comes to yellow taxis, but that an e-hail is a pre-arrangement when it

comes to black cars.”); see also Brief for Petitioners-Appellants at 42–43, 47–53,

Melrose Credit Union, No. 2016-02214. The traditional industry argues that the Taxi

and Limousine Commission categorized e-hails as a type of hail, “not a pre-arrange-

ment,” in 2015 when it allowed medallion taxis to respond to e-hails, and earlier

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 13 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 13

arrive after having been prearranged some time earlier, as app compa-

nies such as Uber and Lyft move to allow passengers to schedule rides

in advance, similar to the way that consumers historically have called

ahead for a cab or for-hire vehicle to take them to the airport or other

locations.

Despite the above, the Uber business model is not necessarily

more vertically integrated than the traditional taxi industry. Uber is in

“a direct contractual relationship with its drivers”—there is no role for

a broker leasing medallions (and/or taxi vehicles) as there is in some

components of New York’s yellow taxi sector.

However, Uber clas-

sifies its drivers as independent contractors, not employees, much as

defended a pilot program allowing medallion taxis to respond to e-hails when black

car companies challenged the authority of the Commission to approve the pilot. Brief

for Petitioners-Appellants at 22–23, Glyka Trans LLC, No. 2015-11661 (referring to

Black Car Assistance Program v. City of New York, No. 100327/13, 2013 WL

1808082 (N.Y. Sup. Ct. 2013), aff’d 110 A.D. 3d 618 (N.Y. App. Div. 2013)); see

also Amended Complaint at 40–43, Melrose Credit Union, No. 1:15-cv-09042. The

City insists that it has not changed its position, and has “consistently” maintained that

e-hails are a form of prearrangement. See Brief for Respondents at 53–54, Glyka

Trans LLC, No. 2015-11661 (“[T]he Commission has consistently treated app-ar-

ranged rides as prearrangements, regardless of which type of vehicle is involved. . . .

The petitioners’ claim that the Commission has ‘expressly’ argued otherwise demon-

strates a singular disregard for the facts.”); Defendants’ Memorandum of Law in Sup-

port of Defendants’ Motion to Dismiss the Amended Complaint at 28 n.9, Melrose

Credit Union, No. 1:15-cv-09042 (“Plaintiffs’ intentional mischaracterization of state-

ments made by TLC and counsel in prior court proceedings . . . , as a means of

reaching their desired legal conclusion that rides provided by FHVs using electronic

app technology are legally the same as ‘street hails,’ cannot be credited.”). The City’s

classification of prearrangements as a subset of hails might be interpreted as an effort

to reconcile its position in the current litigation against the medallion industry that e-

hailing is a form of prearrangement, with the position that the City took before that e-

hailing is hailing when the City was defending against litigation by black car compa-

nies that sought to block medallion taxis from responding to e-hails.

27. Russell Dicker, Flight Booked, Bags Packed, Ride Scheduled,

BER

(

June 9,

2016), https://newsroom.uber.com/scheduledrides/ (Uber will allow customers to

“schedule” rides between “15 minutes to 30 days in advance”); see also Jelisa Cas-

trodale, Uber Launches Scheduled Rides, So You Can (Finally) Book a Car in Ad-

vance,

USA T

ODAY

(June 14, 2016), http://www.usatoday.com/story/travel/

roadwarriorvoices/2016/06/14/uber-scheduled-rides/85866022/; Dan Rivoli, Uber

Now Offers Reserved Rides in NYC, Infringing Further On Conventional Taxi Ser-

vices,

N.Y. D

AILY

EWS

(Sept. 22, 2016), http://www.nydailynews.com/new-york/

uber-offers-reserved-rides-new-york-city-article-1.2801464.

28. Rogers, supra note 10, at 90. Rogers takes the absence of an intermediary be-

tween Uber and its drivers as evidence that Uber’s business model is more vertically

integrated than that of the traditional taxi industry. However, there were a number of

different relationships between medallion owners and taxi drivers in the traditional

taxi industry, and medallion owners did not necessarily rely on an intermediary be-

tween them and drivers to lease out medallions. In some instances, medallion owners

themselves were leasing out medallions, meaning that, like Uber, they were in a direct

contractual relationship with their drivers. In other instances, the medallion owner

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 14 17-APR-17 10:20

14 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

traditional taxi drivers usually are independent contractors.

Because

Uber classifies its drivers as independent contractors, they, like tradi-

tional taxi drivers, do not benefit from many standard employment

protections. Uber charges drivers a service fee on fares, usually con-

sisting of twenty percent of the fare.

Uber drivers provide their own

cars, and pay for “gas, equipment maintenance, and repairs.”

Uber

drove the taxi. Sebago v. Bos. Cab Dispatch, Inc., 28 N.E.3d 1139, 1144–55 (Mass.

2015).

29. The decisions of Uber and Lyft to classify their drivers as independent contrac-

tors are being challenged in the public arena and the courts. See, e.g., Lia Eus-

tachewich, Uber Drivers Sue “Hire”-Ups,

N.Y. P

OST

, Sept. 11, 2015, at 9; Kevin

McGowan, Uber Sued Again Over “Wage Theft” in New York,

LOOMBERG

BNA

(Oct. 27, 2016), http://www.bna.com/uber-sued-again-n57982079217/. The Cotter

suit against Lyft was settled, and the trial judge first rejected and then, after renegoti-

ation, granted preliminary approval to the settlement. Cotter v. Lyft, No. 13-cv-04065,

2016 WL 3561742 (N.D. Cal. June 23, 2016). Uber recently settled two lawsuits chal-

lenging the classification of its drivers as independent contractors (O’Connor in Cali-

fornia, and Yucesoy in Massachusetts); under the settlements, Uber drivers continued

to be classified as independent contractors. Mike Isaac & Noam Scheiber, Uber Set-

tles Cases with Concessions, but Drivers Stay Freelancers,

N.Y. T

IMES

(Apr. 21,

2016), http://www.nytimes.com/2016/04/22/technology/uber-settles-cases-with-con-

cessions-but-drivers-stay-freelancers.html; see also Growing and Growing Up,

BER

(Apr. 21, 2016), https://newsroom.uber.com/growing-and-growing-up/. However, in

August 2016, Federal District Court Judge Edward Chen rejected the settlement

agreement as “not fair, adequate, and reasonable.” See O’Connor v. Uber Techs., No.

13-cv-03826, 2016 WL 4398271 (N.D. Cal. Aug. 18, 2016); Mike Isaac, Judge Over-

turns Uber’s Settlement with Drivers,

N.Y. T

IMES

(Aug. 18, 2016), http://

www.nytimes.com/2016/08/19/technology/uber-settlement-california-drivers.html. A

subsequent ruling from the Ninth Circuit may complicate the efforts of drivers to

challenge their classification status, because the ruling holds that the drivers are sub-

ject to arbitration clauses. See Mike Isaac, Ruling Tips Uber Drivers Away from

Class-Action Suits,

N.Y. T

IMES

(Sept. 7, 2016), http://www.nytimes.com/2016/09/08/

technology/ruling-tips-uber-drivers-away-from-class-action-suits.html; Joel Rosen-

blatt, Uber’s Drivers Win the Backing of U.S. Labor Watchdog,

LOOMBERG

(Nov. 2,

2016), https://www.bloomberg.com/news/articles/2016-11-02/u-s-labor-panel-backs-

uber-drivers-in-class-action-fight.

There already is a prominent administrative agency decision in California hold-

ing that an Uber driver is an employee, not an independent contractor as Uber argues.

Berwick v. Uber Techs., No. 11-46739 EK (Cal. Labor Comm’r’s Office June 3,

2015). This decision and another decision from the Oregon Bureau of Labor, holding

that “Uber drivers are employees,” are discussed in

OMM

FOR

EVIEW OF

NNOVA-

TIVE

RBAN

OBILITY

ERVS

, supra note 7, at 92–93. State agency decisions going

the other way, and finding that Uber drivers are not employees, are discussed in Da-

vey Alba, Florida Says Uber Driver Isn’t an Employee After All,

IRED

(Oct. 1,

2015), http://www.wired.com/2015/10/florida-uber-decision-reversal/. In the UK, an

employment tribunal recently held that Uber drivers are “workers” under national em-

ployment legislation. Assam & Farrar v. Uber, No. 2202550/2015 (Emp’t Tribunal,

Oct. 28, 2016) (UK).

30. Ellen Huet, Uber Tests Taking Even More from Its Drivers with 30% Commis-

sion,

ORBES

(May 18, 2015), http://www.forbes.com/sites/ellenhuet/2015/05/18/

uber-new-uberx-tiered-commission-30-percent/.

31. Oei & Ring, supra note 10, at 1002.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 15 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 15

drivers may lease the cars that they are driving from a third party,

which Uber may facilitate.

The technological and industrial organization changes just de-

scribed are interrelated. Uber’s technology is a “two-sided platform”

that matches riders and drivers, and has network effects that help to

explain why Uber is creating such a large virtual fleet. Riders benefit

when more drivers use the app, because more drivers make it easier to

get a taxi quickly, while drivers also benefit from the presence of more

riders, because this makes it easier to get fares. This contributes to the

rise of a large fleet like Uber.

To be sure, Uber is not the first two-

sided market in the history of the taxi industry—taxi companies in

many jurisdictions historically have operated as two-sided platforms,

dispatching taxis by radio in response to calls from customers.

How-

ever, Uber’s powerful and easy-to-use app technology has facilitated

the formation of large collections of drivers and passengers on a

global geographic scale that radio taxi companies were not able to

realize.

In thinking about the changes wrought by Uber and other taxi

apps, it is important to keep in mind the persistence of traditional taxi

operations. In New York and other places, the historical incumbent

providers of taxi services still remain active market players, although

they certainly have been harmed by the advent of apps.

32. Nitasha Tiku, Uber and Its Shady Partners Are Pushing Drivers into Subprime

Loans,

AWKER

(Nov. 4, 2014), http://valleywag.gawker.com/uber-and-its-shady-

partners-are-pushing-drivers-into-su-1649936785; Rent a TLC Vehicle,

BER

, http://

www.driveubernyc.com/get-a-vehicle/ (last visited Nov. 2, 2016); Uber NYC Market-

place,

BER

, http://www.ubernycmarketplace.com/ (last visited Dec. 1, 2016). Uber

does not own cars or provide financing for buying, leasing, or renting cars. Dobson,

supra note 10, at 716.

33. For a typology of network effects from platforms, see FTC Workshop, supra

note 10, at 174–75 (remarks of Maurice Strucke). Arun Sundararajan argues that

“two-sided markets network effects” and “learning by doing” are the most common in

the sharing economy. Id. at 175 (remarks of Arun Sundararajan).

Rauch and Schleicher refer to the potential that there might be diseconomies of

scale in two-sided platforms such as taxi apps. Rauch & Schleicher, supra note 10, at

918. More generally on the “determinants of” the “size and structure” of two-sided

platforms, see David S. Evans & Richard Schmalensee, The Industrial Organization

of Markets with Two-Sided Platforms, 3

OMPETITION

’

151, 163–66

(2007).

34.

ARK

W. F

RANKENA

& P

AUL

A. P

AUTLER

CONOMIC

NALYSIS OF

AXI-

CAB

EGULATION

(1984).

35. Associated Press, Yellow Cabs Now Outnumbered by Uber Cars on NYC

Streets,

WNYC

(Mar. 19, 2015), http://www.wnyc.org/story/yellow-cabs-now-out-

numbered-uber-cars-nyc-streets/. For indications that New York City’s yellow taxis

still command significant market share, notwithstanding the arrival of Uber, Lyft and

other apps, see infra note 121 and accompanying text.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 16 17-APR-17 10:20

16 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

II.

EGULATORY

ESPONSES TO

AXI

PPS

As mentioned at the outset, regulatory responses to Uber and

other apps have not been especially innovative. Below is a preliminary

sketch of the regulatory responses in the United States.

First, there remain jurisdictions that are not allowing Uber to op-

erate legally. They are, by definition, not innovating to facilitate adop-

tion of the new technology. Some of these jurisdictions are enforcing

regulations against Uber’s operations.

Other jurisdictions are tolerat-

ing Uber’s operating in apparent contravention of regulatory

requirements.

Second, there are what I call the “welcoming jurisdictions” that

are allowing Uber to operate legally. There are now many welcoming

jurisdictions in the United States. Over sixty U.S. governments, in-

cluding over thirty states and the District of Columbia, have passed

legislation or regulations to legalize the use of taxi apps such as

36. See, e.g., Madeline Stone, East Hampton Officials Are Pushing for Jail Time

for Uber Drivers, Says Their Lawyer,

. I

NSIDER

(June 15, 2015), http://

www.businessinsider.com/east-hampton-montauk-uber-drivers-could-face-jail-time-

2015-6 (describing the plight of Uber in East Hampton, New York). In the European

Union, “Uber activities have been banned or subject to serious restrictions in Member

States, such as Belgium, Germany, Italy and Spain.” Geradin, supra note 8, at 3.

37. St. Louis, Missouri might be an example. See, e.g., Danny Wicentowski, Uber

Driver Arrested at Lambert Airport After Dropping Off Passenger,

IVERFRONT

IMES

(July 20, 2016), http://www.riverfronttimes.com/newsblog/2016/07/20/uber-

driver-arrested-at-lambert-airport-after-dropping-off-passenger (“[A]n uneasy truce

has persisted while Uber and the [St. Louis Metropolitan Taxicab Commission] slug it

out in federal court.”); see also Wallen v. St. Louis Metro. Taxicab Comm’n, No.

4:15-cv-1432, 2016 WL 5846825 (E.D. Mo., Oct. 6, 2016) (refusing to dismiss action

brought by Uber and others against St. Louis Metropolitan Taxicab Commission for

violating the Sherman Act based on treatment of Uber and UberX); Verified First

Amended Complaint for Injunctive and Other Relief at 2, Wallen v. St Louis Metro.

Taxicab Comm’n, No. 4:15-cv-1432 (E.D. Mo. Oct. 20, 2016) (alleging “anticompeti-

tive conduct” by St. Louis Metropolitan Taxicab Commission and others “to protect

the entrenched taxicab industry” from transportation network companies, including

Uber and others). Miami-Dade County was an example of a jurisdiction tolerating

Uber’s operations until recently, when Miami-Dade passed an ordinance legalizing

transportation network entities. See Douglas Hanks, Taxis Suing Miami-Dade for $1

Billion Over New Uber Law,

IAMI

ERALD

(May 4, 2016), http://

www.miamiherald.com/news/local/community/miami-dade/article75555187.html; Pa-

tricia Mazzei, Lawsuit Reignites Miami-Dade Policy Fight Over Uber, Lyft,

IAMI

ERALD

(Feb. 10, 2015), http://www.miamiherald.com/news/local/community/miami-

dade/article9704705.html.

Barry and Pollman discuss Uber as a paradigmatic example of a “regulatory en-

trepreneur,” for which “changing the law [is] a material part of its business plan.”

Barry & Pollman, supra note 10, at 2.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 17 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 17

Uber.

The jurisdictions that are legalizing Uber and other taxi apps

each appear to be taking one of two paths.

First, many jurisdictions are legalizing Uber and other taxi apps

by creating a new regulatory category under which they are regulated

more lightly than traditional taxis.

In a rulemaking in 2013, the Cali-

fornia Public Utilities Commission established a new category of

“Transportation Network Company,” under its statutory authority over

“charter-party carriers,” to regulate services such as UberX and Lyft.

Other jurisdictions are following California’s lead and enacting provi-

sions to regulate taxi app companies using the term “Transportation

Network Company.”

Second, some jurisdictions, such as New York City, are allowing

Uber to operate under a regulatory category that pre-existed the devel-

opment of apps, and adding regulatory requirements incrementally to

address emerging issues raised by the apps.

New York City has long

had a much more lightly regulated “for-hire vehicle” sector in addition

38. See Associated Press, Uber, Lyft Campaign to Expand Operations into Upstate

New York,

. J.

(Apr. 8, 2016), http://www.insurancejournal.com/news/east/2016/

04/08/404672.htm (“More than 40 states now allow ride-hailing companies to com-

pete with traditional taxis.”); Transportation Network Company States with Enacted

Legislation,

ROP

. C

ASUALTY

NSURERS

’

NOF

, http://viewer.zmags.com/publi

cation/60841263#/60841263/1 (last visited Jan. 22, 2017).

39. See Strong, supra note 10, at 1054 (noting that regulators have either applied

“existing taxi or ‘private carrier’ regulations . . . or creat[ed] a distinct set of rules” for

“Transportation Network Companies”); id. at 1077–78.

40. See

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note

7, at 17 (“TNCs are being regulated, but much more lightly than traditional taxi and

other for-hire transportation services.”).

41. CPUC, supra note 26, at 7–8. For background on CPUC’s creation of the TNC

category, see FTC Workshop, supra note 10, at 90–101 (remarks of Catherine J.K.

Sandoval).

Uber offers several levels of service. UberX is a low-cost service. See What is

UberX?

, U

BER

STIMATE

, http://uberestimate.com/what-is-uberx/ (last visited Jan. 19,

2017).

The Commission “defines a TNC as an organization, whether a corporation, part-

nership, sole proprietor, or other form, operating in California that provides prear-

ranged transportation services for compensation using an online-enabled application

(app) or platform to connect passengers with drivers using their personal vehicles.”

CPUC, supra note 26, at 2. The Commission’s rulemaking was not the last word in

California on regulating taxi apps. In 2014, the California legislature passed a statute

defining a TNC in roughly the same terms, and to increase the insurance coverage that

taxi apps must provide, in the wake of a tragic accident involving an UberX driver.

See

. U

TIL

ODE

§§ 5431(a), 5433 (West 2017); Dobson, supra note 10, at

705–09.

42.

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note 7, at

52–53.

43. New York City’s approach to regulating Uber and other app-based services is

described in Glyka Trans LLC v. City of New York, No. 8962/15, 2015 WL 5320868

(N.Y. Sup. Ct. Sept. 8, 2015).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 18 17-APR-17 10:20

18 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

to the heavily regulated yellow taxis. Uber and other taxi app vehicles

operate in New York City as licensed for-hire vehicles, usually as

“black cars,” which are a sub-category of for-hire vehicles.

The jurisdictions that are allowing Uber to operate, whether

under existing or new categories, generally have not also been system-

atically revisiting the existing regulatory frameworks for historical in-

cumbent taxi operators to examine whether these should be altered in

light of the advent of the new apps and other changes over time.

44. “Black car” is a defined term in N.Y.C. Admin. Code § 19-502(u): “ ‘Black car’

means a for-hire vehicle dispatched from a central facility whose owner holds a

franchise from the corporation or other business entity which operates such central

facility, or who is a member of a cooperative that operates such central facility, where

such central facility has certified to the satisfaction of the commission that more than

ninety percent of the central facility’s for-hire business is on a payment basis other

than direct cash payment by a passenger.” See also 35 RCNY § 51-03 (“All Black Car

Vehicles are owned by franchisees of the Base or are members of a cooperative that

operates the Base”) (defining “Black Car Base”); 35 RCNY § 55-03(d)(2) (same); 35

RCNY § 59A-03(c)(2) (same); 35 RCNY § 59B-03(c)(2) (same).

On Uber’s use of the “black car” category, see Glyka Trans LLC, No. 8962/15,

2015 WL 5320868, at *3.

Incumbents in the yellow taxi sector argue that the City has improperly licensed

Uber vehicles as black cars, because they do not satisfy the legal requirements for

base licensing—Uber vehicles are not “owned by franchisees of [Uber] base[s] or . . .

members of a cooperative that operates a base.” Complaint at 16, CGS Taxi v. City of

New York, No. 653264/2015 (N.Y. Sup. Ct. Sept. 30, 2015); see also Complaint at

17, Singh v. City of New York, No. 701402/2017 (N.Y. Sup. Ct. Jan. 30, 2017); Brief

for Petitioners-Appellants at 32, Glyka Trans LLC v. City of New York, No. 2015-

11661 (N.Y. App. Div. Nov. 30, 2015) (“Uber’s drivers are neither franchisees nor

cooperative members.”). They also assert that Uber vehicles do not receive rides

through prearrangement from a base, as required for black cars. Complaint at 18–19,

CGS Taxi LLC, No. 653264/2015 (Sept. 30, 2015); Complaint at 17–18, Singh, No.

701402/2017 (Jan. 30, 2017); Reply Brief for Petitioner-Appellants at 17–18, Glyka

Trans LLC, No. 2015-11661; see also Brief for Petitioners-Appellants at 17, Melrose

Credit Union v. City of New York, No. 2016-02214 (N.Y. App. Div. May 17, 2016)

(“§ 55-03(h), § 55-03(g) and § 55-03(d)(1) of the TLC Rules state that for all black

car trips, the car must be dispatched through and ‘from’ the ‘physical location’ of its

affiliated base station, and for a passenger that has scheduled a pick-up in the future.

35 RCNY §§ 55-03(h), (d)(1) (2015). Uber is violating these rules.”). Moreover, the

industry argues that Uber does not comply with the black car rules because Uber

vehicles accept street hails, which are within “the exclusive province of yellow taxis

(or green taxis, if outside the exclusionary zone).” Complaint at 19, CGS Taxi LLC,

No. 653264/2015; see also Complaint at 18, Singh, No. 701402/2017 (Jan. 30, 2017).

The New York Taxi and Limousine Commission recently established a new reg-

ulatory category of “Dispatch Service Provider.” A “Dispatch Service” “is dispatch-

ing, reserving, or referring trips to Drivers on behalf of TLC-licensed Bases through a

publicly available, Passenger-facing booking tool.” 35 RCNY §51-03; see also 35

RCNY § 77 (Licensing Rules for Dispatch Service Providers).

45. In researching this article, I had two research assistants review legislation

passed in many of the U.S. states that have legalized Uber and other transportation

network companies. Their survey showed that of twenty-seven states (including for

these purposes the District of Columbia) that have legalized taxi apps like Uber, the

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 19 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 19

other words, regulators generally do not appear to be comprehensively

rethinking taxi regulation, but rather reacting ad hoc to the arrival of

the new entrants.

III.

COPE OF

AXI

EGULATION IN THE

GE OF

BER

This part analyzes the appropriate objectives of taxi regulation in

the age of apps. In re-envisioning taxi regulation to meet the needs of

the current era, I generally assume that the taxi sector should be regu-

lated to address market failures. Market failures “refer[ ] to situations

in which free markets do not result in an efficient allocation of re-

sources, resulting in a loss of economic and social welfare. Markets

can fail for a variety of reasons, including the presence of a natural

monopoly, large sunk costs, information asymmetries, and negative or

positive externalities.”

vast majority have not revisited traditional taxi regulation when legislating to legalize

“Transportation Network Companies.” Summary Memoranda from Aaron Lichter &

Andr´e Smith (Jan.–Mar. 2016) (on file with author); see also Suska, supra note 10, at

194–97 (describing regulatory changes in Chicago and Massachusetts that legalized

the operation of Uber and other ridesharing services).

To be sure, there are some jurisdictions that are beginning to re-examine tradi-

tional taxi regulation. In Seattle, in 2014, local regulators legalized “Transportation

Network Companies” and reduced some restrictions on traditional cab companies, and

authorized the issuance of additional taxi cab licenses. See Strong, supra note 10, at

1086, 1088; Lynn Thompson, Seattle Council Gives Nod to Compromise Rules for

Ride Services,

EATTLE

IMES

(July 14, 2014), http://www.seattletimes.com/seattle-

news/seattle-council-gives-nod-to-compromise-rules-for-ride-services/. In Los Ange-

les, the Taxicab Commission has proposed regulatory changes for the taxi industry—

including requiring traditional taxis “to become accessible via a mobile application

similar to the ones used by Uber and Lyft,” after the California Public Utilities Com-

mission legalized “Transportation Network Companies.” See Maria Bustillos, How

L.A.’s Taxi Boss Plans to Take On Uber,

ORKER

(Dec. 12, 2014), http://

www.newyorker.com/business/currency/city-los-angeles-plans-make-taxis-like-uber.

In legalizing taxi apps in 2014, Milwaukee eliminated the cap on the number of per-

mits for taxis, but in other respects left in place more stringent requirements for tradi-

tional taxis. See Complaint at 18–19, Joe Sanfelippo Cabs, Inc. v. City of Milwaukee,

46 F. Supp. 3d 888 (E.D. Wis. Sept. 12, 2014) (No. 14-CV-1036). Miami-Dade

County significantly overhauled the regulation of traditional taxicabs when it estab-

lished a framework for legalizing and regulating transportation network companies in

2016. See Motion to Dismiss Amended Class Action Complaint at 5–7, 11, Miadeco

Corp. v. Miami-Dade County, No. 1:16-cv-21976 (S.D. Fla. June 29, 2016). Cam-

bridge, Massachusetts is revisiting its regulation of taxis since the Massachusetts leg-

islature enacted legislation regulating transportation network companies. See Adam

Vaccaro, To Take On Uber, Cambridge Taxi Operators to Ask City to Ease Rule,

LOBE

(Sept. 19, 2016), https://www.bostonglobe.com/business/2016/09/19/take-

uber-cambridge-taxi-operators-ask-city-ease-rules/4XlXRt6eS9FL4ZJVzFlGIL/

story.html.

46.

OMPETITION

UREAU

, supra note 10, at n.40; see also Edelman & Geradin,

supra note 10, at 309 (arguing that regulation of transportation and other platforms is

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 20 17-APR-17 10:20

20 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

Because this article is focused on the normative question of what

taxi regulation should be doing in the age of apps, the article does not

address the politics of taxi regulation. It is worth underscoring the im-

portance of the politics, though, because in practice politics affects the

substance of taxi regulation. There is ample evidence that powerful

interest groups, such as taxi medallion owners and financiers, histori-

cally have distorted local taxi regulation for their benefit.

There is

also a strong likelihood that large, well-financed new entrants like

Uber will become as politically powerful as the incumbents that they

currently are battling, if the new entrants have not already.

This part begins by emphasizing the need for taxi regulators to

design regulations for traditional taxis and app-hailed taxis in tandem.

Then I outline the need to streamline taxi regulation, generally by

eliminating existing regulations that do not address market failures,

and focusing on regulations that do address such failures. In advocat-

ing a scaling back of taxi regulation for the existing taxi sector and the

new app-hailed taxis, I draw on established literature, going back de-

cades, on the appropriate scope for taxi regulation.

A. Joint Regulation of App and Traditional Taxis

The failure of U.S. jurisdictions to update the regulations gov-

erning incumbent taxi operators when legalizing the new entrants, as

explained in Part II, is understandable to a degree given the urgency

that regulators likely feel to legalize the new entrants in the face of

consumer demand for their services, and regulatory resource con-

straints that make it difficult to address multiple issues at once. How-

ever, this failure is not defensible in economic terms.

justified to address market failures). Cohen and Sundararajan also ground their analy-

sis of where regulation is warranted in market failures, but they adopt a somewhat

unusual definition of market failure, that includes situations where the market leads to

“inequitable” outcomes, not just “inefficient . . . outcomes,” because of “asymmetric

information, the problem of public goods, the threat of monopoly, or the existence of

externalities that are not naturally internalized by market participants.” Cohen & Sun-

dararajan, supra note 10, at 120.

47. See Edmund W. Kitch et al., The Regulation of Taxicabs in Chicago, 14

J.L. &

CON

. 285 (1971); Wyman, supra note 21.

48. See, e.g.,

RANKENA

& P

AUTLER

, supra note 34.

49. Early in 2016, it appeared that some courts might provide an impetus for revisit-

ing taxi regulation. As of then, there were some judicial decisions that suggested that

some courts might be willing to countenance claims from the incumbent taxi industry

that regulating it more stringently than the transportation network companies violates

the Equal Protection Clause. See Desoto Cab Co. v. Picker, No. 15-cv-04375, 2016

WL 3913643 (N.D. Cal. July 20, 2016) (denying motion to dismiss); Bos. Taxi Own-

ers Ass’n v. City of Boston, No. 15-10100-NMG, 2016 WL 1274531 (D. Mass. Mar.

31, 2016) (refusing to dismiss equal protection claims against City of Boston, but

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 21 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 21

denying preliminary injunction); see also VTS Transp. v. Palm Beach County, No.

9:15-cv-80560, 2016 WL 4250684 (S.D. Fla. June 21, 2016) (refusing to dismiss

plaintiffs’ equal protection claim for damages for the period in which Palm Beach

County allowed Uber to operate under a Temporary Operating Agreement “without

complying with” Vehicle For Hire ordinance). But see VTS Transp. v. Palm Beach

County, No. 9:15-cv-80560, (S.D. Fla. July 1, 2016) (rejecting plaintiffs’ motion seek-

ing leave to file class certification).

However, the “weight of authority” now solidly rejects arguments from the tradi-

tional taxi industry that a dual regulatory regime for traditional and app-dispatched

taxis violates the Equal Protection Clause. Desoto Cab Co. v. Picker, No. 15-cv-

04375-EMC, 2017 WL 118810, at *8 (N.D. Cal. Jan. 12, 2017); see Ill. Transp. Trade

Ass’n v. City of Chicago, 839 F.3d 594 (7th Cir. 2016); VTS Transp. v. Palm Beach

County, No. 9:15-cv-80560 (S.D. Fla. Mar. 8, 2017) (granting summary judgment to

defendant Palm Beach County); Cambridge Taxi Drivers & Owners Ass’n. v. City of

Cambridge, No. 16-11357-NMG, 2017 WL 373491 (D. Mass. Jan. 25, 2017) (dis-

missing equal protection claim because state law preempts local regulation of trans-

portation network companies); Bos. Taxi Owners Ass’n v. Baker, No. 16-11922-

NMG, 2017 WL 354010 (D. Mass. Jan. 24, 2017) (dismissing equal protection claim

against Massachusetts defendants); Newark Cab Ass’n v. City of Newark, No. 2:16-

cv-04681, 2017 WL 214075 (D.N.J. Jan. 17, 2017) (dismissing equal protection

claim); Bos. Taxi Owners Ass’n v. City of Boston, No. 15-10100-NMG, 2016 WL

7410777 (D. Mass. Dec. 21, 2016) (dismissing equal protection claim against City

because state legislation largely preempts local regulation of transportation network

companies); Gebresalassie v. District of Columbia, 170 F. Supp. 3d 52 (D.D.C. 2016);

Melrose Credit Union v. City of New York, No. 1:15-cv-09042 (S.D.N.Y. Jan. 26,

2016) (denying plaintiffs’ motion for preliminary injunction); Joe Sanfelippo Cabs,

Inc., 46 F. Supp. 3d 888; Taxicab Paratransit Ass’n of Cal. v. Cal. Pub. Utils.

Comm’n, No. S220982 (Cal. Nov. 12, 2014) (denying petition for review); Second

Amended Class Action Complaint, Ruffino v. Broward County, No. 0:15-cv-62312

(S.D. Fla. Sept. 20, 2016); Order of Supplemental Motion to Dismiss, Ruffino, No.

0:15-cv-62312 (granting defendants’ motion to dismiss); Order of Dismissal Without

Prejudice, Ruffino, No. 0:15-cv-62312 (dismissing the case without prejudice pursuant

to plaintiffs’ request for voluntary dismissal).

A Seventh Circuit Court of Appeals decision authored by the influential Judge

Richard Posner rejecting equal protection claims from incumbents in the Chicago taxi

industry seems to have played an important role in shifting the judicial tide against the

industry’s equal protection claims. Ill. Transp. Trade Ass’n, 839 F.3d 594; see VTS

Transp. v. Palm Beach County, No. 9:15-cv-80560, at 9, 19–20 (citing Ill. Transp.

Trade Ass’n); Bos. Taxi Owners Ass’n v. Baker, 2017 WL 354010, at *6 (same);

Newark Cab Ass’n, 2017 WL 214075, at *6 (same); Desoto Cab Co. v. Picker, No.

15-cv-04375-EMC, 2017 WL 118810, at *7–8 (N.D. Cal. Jan. 12, 2017) (same). The

Seventh Circuit ruling upholding a dual regulatory regime is perfectly defensible as a

legal matter, since the governing rational basis standard is highly deferential to gov-

ernments. However, the ruling is problematic to the extent that it can be interpreted as

suggesting that it is desirable for governments to establish different regulatory stan-

dards for incumbent taxis and transportation network vehicles because they provide

“[d]ifferent products.” Ill. Transp. Trade Ass’n, 839 F.3d at 598. As discussed in the

text above, some of the differences between traditional taxis and the new entrants

likely are the result of decisions by regulators; the incumbents and the new entrants

might not differ so greatly if regulatory requirements were otherwise. For example,

one distinction between transportation network company vehicles and traditional taxis

is that the former do not accept street hails. But this distinction likely relates to the

fact that transportation network company vehicles are legally prohibited from ac-

cepting street hails and so this kind of service is legally provided only by incumbent

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 22 17-APR-17 10:20

22 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

The regulation of traditional and app taxis poses a “joint op-

timization” problem.

Regulators should be simultaneously establish-

ing standards for app-dispatched taxis and revisiting standards for

traditional taxis because they are substitutes competing in the same

“product market.”

Both perform the same service of providing con-

sumers with on-demand transportation for a price: a traditional taxi

and an app-based taxi both transport passengers from one point to an-

other, with the consumer selecting the point of departure, the destina-

tion, and sometimes even the route if the driver obliges.

Underscoring the extent to which traditional taxis and new en-

trants like Uber operate within the same product market, a recent New

York City report refers to yellow taxis and “e-dispatch[ed]” vehicles

such as Uber’s as substitutes, specifically indicating that “[i]ncreases

in e-dispatch trips are largely substituting for yellow taxi trips in the”

city’s Central Business District.

Although Uber sometimes argues

taxis. Id. at 596–97. It seems “circular” for governments to regulate two providers

differently on the basis of distinctions that the government itself has created. Bos. Taxi

Owners Ass’n, No. 15-10100-NMG, 2016 WL 1274531, at *6. In addition, as dis-

cussed above, the products offered by traditional taxis and the new entrants are grow-

ing more similar as the former are adapting innovations pioneered by the new entrants

and the new entrants are increasing their offerings. Indeed, in the companion case of

Joe Sanfelippo Cabs, Inc., 839 F.3d 613, decided on the same day as Ill. Transp.

Trade Ass’n, Judge Posner recognizes the similarity between traditional taxis and

Uber, in rejecting a Takings claim from the traditional taxi industry in Milwaukee

after the entry of Uber. Notably, while Judge Posner insists on the differences be-

tween Uber and traditional taxis in Ill. Transp. Trade Ass’n, in Joe Sanfelippo Cabs,

Inc., he describes Uber as “provid[ing] [a] close though not identical substitute[ ] for

conventional taxi services,” and later refers to Uber and other “app-based . . . compa-

nies” as “substitutes” and “taxi substitutes.” 839 F.3d at 614–15. In upholding the

authority of governments to establish less stringent regulatory standards for transpor-

tation network companies, the Seventh Circuit panel in Ill. Transp. Trade Ass’n por-

trayed itself as enabling governments to introduce competition in the for-hire sector.

839 F.3d at 597–98. But given the presence that Uber has established in many juris-

dictions, it may be time to think about promoting competition by relaxing the restric-

tions on the traditional taxi industry that are inhibiting it from competing with Uber

and other new entrants.

50. I borrow the use of the term “joint optimization problem” to describe the need

to regulate new and old sources together from Richard L. Revesz & Allison L. West-

fahl Kong, Regulatory Change and Optimal Transition Relief, 105

. U. L. R

1581, 1618 (2011).

51. Verified First Amended Complaint for Injunctive and Other Relief at 20, Wal-

len v. St. Louis Metro. Taxicab Comm’n, No. 4:15-cv-1432 (E.D. Mo. Oct. 20, 2016).

52.

FFICE OF THE

AYOR

, C

ITY OF

ORK

, F

-H

IRE

EHICLE

RANSPORTA-

TION

TUDY

5 (2016), http://www1.nyc.gov/assets/operations/downloads/pdf/For-

Hire-Vehicle-Transportation-Study.pdf.

Federal District Court Judge Gorton nicely captured the need to see through the

effects of existing regulations in examining the relationship between traditional taxis

and transportation network companies in refusing to dismiss an equal protection claim

from the traditional taxi industry:

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 23 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 23

that it is not “a transportation provider,”

Uber itself suggests that it

competes in the same “product market” as traditional taxi operations

in a complaint filed in St. Louis, Missouri.

In this case, Uber and

other plaintiffs are suing the St. Louis taxi regulatory commission for

acting anticompetitively to protect the incumbent industry from com-

petition from Uber and other transportation network companies.

[P]laintiffs have stated at least a plausible claim that the Equal Protection

Clause requires that the two groups [i.e., “taxi operators” and “transporta-

tion network companies” (TNCs)] be treated alike.

First, the Court finds persuasive plaintiffs’ argument that many of the

obvious differences between taxis from TNCs, such as the kind of vehicle

used and the fact that taxicabs must be clearly labeled, are caused by the

City’s application of the requirements of [City] Rule 403 to taxi operators

but not to TNCs. The City may not treat the two groups unequally and

then argue that the results of that unequal treatment render the two groups

dissimilarly situated and, consequently, not subject to equal protection

analysis. Such circular logic is unavailing.

Bos. Taxi Owners Ass’n, No. 15-10100-NMG, 2016 WL 1274531, at *6. For a con-

trasting perceptive, which insists that the transportation network companies and the

traditional taxi industry offer “[d]ifferent products,” justifying different regulatory re-

gimes notwithstanding the Equal Protection Clause, see the decision of a panel of the

Court of Appeals of the Seventh Circuit rejecting the Chicago taxi industry’s equal

protection claims in Ill. Transp. Trade Ass’n, 839 F.3d 598. In December 2016 and

January 2017, Judge Gorton dismissed equal protection claims brought by the incum-

bent Boston taxi industry against the City of Boston and a set of Massachusetts defen-

dants. In dismissing the case against the Massachusetts defendants, Judge Gorton

cited Ill. Transp. Trade Ass’n, among other cases, as “persuasive authority.” Bos. Taxi

Owners Ass’n v. City of Boston, No. 15–10100–NMG, 2016 WL 7410777; see also

Bos. Taxi Owners Ass’n v. Baker, No. 16-11922-NMG, 2017 WL 354010, at *6 (re-

ferring to Ill. Transp. Trade Ass’n).

53. Sam Schechner, Uber’s ‘Not a Taxi Company’ Defense on Trial in EU,

ALL

TREET

(Nov. 29, 2016), http://www.wsj.com/articles/ubers-not-a-taxi-company-de-

fense-on-trial-in-eu-1480427094 (“Uber Technologies Inc. has long tried to fight local

transportation laws by saying it isn’t a transportation company . . . . It is arguing at the

EU’s Court of Justice that [‘national transportation regulations’] don’t apply to Uber

because it is an online-service provider, rather than a transportation provider—a des-

ignation that it says should give it protection under existing EU laws.”); see also Mark

Scott, In Europe, Is Uber a Transportation Service or a Digital Platform?,

N.Y.

IMES

(Nov. 27, 2016), http://www.nytimes.com/2016/11/27/technology/uber-europe-

court-ecj.html. For a U.S. case where Uber denies it is “a transportation company,”

see Meyer v. Kalanick, 174 F. Supp. 3d 817, 820 (S.D.N.Y. 2016) (“Uber states that it

is not a transportation company.”). An employment tribunal in the UK recently held

that it is “unreal to deny that Uber is in business as a supplier of transportation ser-

vices.” Assam & Farrar v. Uber, No. 2202550/2015, at 27 (Emp’t Tribunal, Oct. 28,

2016) (UK).

54. Verified First Amended Complaint for Injunctive and Other Relief at 20, Wal-

len, No. 4:15-cv-1432.

55. Id. at 1–2; see also Wallen v. St. Louis Metro. Taxicab Comm’n, No. 4:15-cv-

1432, 2016 WL 5846825, at *1 (E.D. Mo. Oct. 6, 2016) (refusing to dismiss action

brought by Uber and others against St. Louis Metropolitan Taxicab Commission for

violating the Sherman Act based on its treatment of Uber and UberX).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 24 17-APR-17 10:20

24 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

Uber and the other plaintiffs assert that “taxicab companies, livery

companies, and independent drivers using TNCs [transportation net-

work companies, such as Uber] or associations of independent drivers

who share dispatch services” are “competitors” in “the market for pas-

senger motor vehicle transportation services” in which “consumers

purchase point-to-point transportation services as riders.”

Uber and

other plaintiffs maintain that “[e]xcept for the fact that the MTC [St.

Louis Metropolitan Taxicab Commission] has restrained competition,

. . . consumers would view all of these competitors as offering services

that are reasonably interchangeable substitutes.”

Uber and the traditional taxi industry are not only substitutes, but

their services also could evolve to more closely resemble each other.

Indeed we are already seeing changes to the services that are making

56. Verified First Amended Complaint for Injunctive and Other Relief at 20, Wal-

len, No. 4:15-cv-1432.

57. Id.

The full two paragraphs from the Complaint are as follows:

63. The relevant product market is the market for passenger motor vehi-

cle transportation services (the “Transportation Market”). In this market,

consumers purchase point-to-point transportation services as riders. Com-

petitors in this market include individuals and entities that offer or could

offer point-to-point transportation services to consumers, such as taxicab

companies, livery companies, and independent drivers using TNCs or as-

sociations of independent drivers who share dispatch services.

64. Except for the fact that the MTC has restrained competition, as fur-

ther alleged below, consumers would view all of these competitors as

offering services that are reasonably interchangeable substitutes. There is

a high cross-elasticity of demand among those competing services, such

that, for example, an increase in the price of traditional taxi services will

result in a relatively high increase in demand for drivers using the uberX

platform.

Id.

Later, the complaint states:

118. Within the St. Louis area, for-hire transportation service providers

(including dispatch and/or TNC services that are used to arrange such for-

hire transportation services) directly compete with one another. In the

Transportation Market, riders who hail taxicabs from the street may in-

stead use the Uber App to arrange for transportation, and vice-versa. Sim-

ilarly, riders who call taxicab or livery vehicle dispatchers to arrange for

taxicab or livery vehicle rides may instead use the Uber App to arrange

for transportation, and vice-versa.

Id. at 40.

As discussed above, Judge Posner also refers to Uber and other “app-based . . .

companies” as “substitutes for conventional taxicab service” and “taxi substitutes” in

Joe Sanfelippo Cabs, Inc. See supra note 49.

In arguing that the plaintiff’s antitrust suit for violations of the Sherman Act and

its New York State equivalent should be dismissed in Meyer v. Kalanick, Uber argued

that the relevant product market for the antitrust inquiry includes not only “taxis,” “car

services,” and ridesharing services, “but also . . . public transit such as subways and

buses, personal vehicle use, and walking.” 174 F. Supp. 3d at 827.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 25 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 25

them more like each other where regulatory rules permit. Perhaps the

principal meaningful difference between the services is how the pas-

senger initiates the request for a vehicle: by app, by outstretched arm,

by waiting in line at a cab stand, or by phone. Uber and other transpor-

tation network company vehicles usually are summoned by app; tradi-

tional taxis are summoned by street hails, waiting in line at a cab

stand, by phone, and may be called by app. But the differences in how

the vehicles are summoned likely are related to regulators segmenting

the market for taxi services: transportation network company vehicles

are often not legally allowed to accept street hails,

and there are

regulatory barriers to them waiting in the same cab stand lines as

traditional taxis.

There is evidence that when there is no regulatory

58. In researching this article, I asked two research assistants to review legislation

that has been passed in multiple states to legalize “Transportation Network Compa-

nies.” They found that sixteen of twenty-seven states (including the District of Colum-

bia) that had legalized Uber and other “Transportation Network Companies,” as of the

time of their survey, prohibited drivers for these companies from picking up street

hails. Lichter & Smith, supra note 45.

Although transportation network company vehicles are not legally allowed to

accept street hails, they may be illegally picking up street hails. There are reports that

some Uber drivers in New York City are illegally picking up passengers on the street.

Dan Rivoli, Uber Cars Rack Up Violations for Illegal Street Hails as NYC Proposes

Crackdown,

N.Y. D

AILY

EWS

(Apr. 6, 2016), http://www.nydailynews.com/new-

york/nyc-proposes-crackdown-uber-cabs-picking-street-fares-article-1.2591233; see

also Amended Complaint at 53, Melrose Credit Union v. City of New York, No. 1:15-

cv-09042 (S.D.N.Y. Mar. 7, 2016) (referring to “the open and notorious proliferation

of e-hailing FHVs trolling the streets of Manhattan attempting to pick up traditional

street hails”).

59. Consider the 2009 rules of the Port Authority of New York and New Jersey,

which regulates the provision of ground transportation at New York area airports.

Under these rules, “ground transportation services shall be provided . . . pursuant to

specific pre-arrangement.”

ORT

UTH

N.Y.

AND

N.J. A

IRPORT

, R

ULES AND

EGU-

LATIONS

§ V(D)(1), at 12 (Aug. 4, 2009), http://www.panynj.gov/airports/pdf/

Rules_Regs_Revision_8_04_09.pdf. The only vehicles that are allowed to pick up

passengers who have not pre-arranged rides are vehicles “licensed” to pick up passen-

gers by “hails” by the municipality in which the airport is located. Id. § V(D)(2). At

airport terminals within the boundaries of New York City, vehicles licensed as “taxi-

cab[s]” by the New York City Taxi and Limousine Commission—in other words,

medallion or yellow taxis—count as vehicles allowed to pick up passengers who have

not pre-arranged rides. Id. The rules of the New York City Taxi and Limousine Com-

mission also prohibit for-hire vehicles (including black cars, which is the category

under which many Uber vehicles are licensed) from picking up passengers at cab

stands at the New York airports. See 35 RCNY §55-19(b) (“A For-Hire Driver must

not pick up a Passenger at an authorized taxi stand.”); 35 RCNY § 55-19(c) (“A

Driver must not accept a dispatch while parked or otherwise located at Kennedy Air-

port or La Guardia Airport unless the Driver is parked in an area other than the air-

port’s designated passenger pick up locations.”); 35 RCNY § 59B-25(g) (“A Base

Owner must not dispatch a Vehicle parked or otherwise located at Kennedy Airport or

La Guardia Airport unless the Vehicle is parked in an area other than the airport’s

designated passenger pick up locations.”).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 26 17-APR-17 10:20

26 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

segmentation of the market, the same vehicle will use multiple tech-

nologies to obtain passengers. In New York City, for example, yellow

medallion taxis can now be hailed on the street by an outstretched

arm, by waiting in line at cab stands, and through prearrangement—

including by apps such as Arro and Curb, introduced in response to

the competition from Uber and other transportation network compa-

nies, and changes in regulatory rules in New York.

As mentioned

above, Uber itself also is evolving to allow riders to obtain Uber vehi-

cles in multiple ways, when not inhibited by regulatory constraints.

Uber’s app now provides not only immediate, on-demand service

analogous to street hails, but also allows riders to schedule trips well

in advance, similar to traditional taxi and livery services. Sometimes,

riders also may be able to obtain Uber vehicles through lines that re-

60. 35 RCNY § 78 (“Licensing and Rules for Providers of E-Hail Applications”);

N.Y.C. Taxi & Limousine Comm’n, Notice of Promulgation of Rules on Licensure of

E-Hail Applications (2015), http://www.nyc.gov/html/tlc/downloads/pdf/

newly_passed_rules_ehail.pdf; Brief for Respondents at 27–30, Glyka Trans LLC v.

City of New York, No. 2015-11661 (N.Y. App. Div. Apr. 15, 2016) (describing the

evolution of the New York Taxi and Limousine Commission’s e-hail rules allowing

yellow taxis to deploy apps); Cristian Salazar, Taxi App Test-Drive: Uber, Lyft, Gett,

Arro vs. Hailing a Yellow Cab by Hand,

ORK

(Dec. 6, 2015), http://

www.amny.com/transit/uber-lyft-gett-arro-vs-hailing-a-yellow-cab-by-hand-taxi-app-

test-drive-1.11134578. For examples of medallion-taxi apps, see Arro—Safe. Fast.

Easy.

RRO

, I

, https://www.goarro.com/ (last visited Jan. 22, 2017);

URB

, https://

gocurb.com/curb-here/ (last visited Nov. 2, 2016).

According to the New York City Taxi and Limousine Commission, however,

“the overwhelming majority of [yellow] taxi rides [in New York City] . . . are still

street hails,” with few yellow taxi rides arranged by app, notwithstanding the ability to

arrange rides by app. Brief for Respondents at 57, Glyka Trans LLC, No. 2015-11661,

at 57. In contrast, medallion owner and financier plaintiffs in Melrose Credit Union v.

City of New York suggest that the share of medallion taxi trips originating through e-

hails is much larger than the City estimates—indeed, they state that “[o]n information

and belief, approximately one-third of medallion taxicab hails on average are now

being made through E-Hails—i.e., approximately 131,000 hails per day are E-Hails

and approximately 262,000 per day are made through traditional taxi street hails.”

Amended Complaint at 52–53, Melrose Credit Union, No. 1:15-cv-09042. If the

City’s statistics are accurate, it would be interesting to know why few yellow taxi

rides are arranged by app, and whether regulatory restrictions (such as regulated fare

levels that keep yellow taxis from competing on price) are part of the reason.

The incumbent taxi industry in other cities also is adopting new technology. For

example, in Milwaukee, traditional taxis also can now be obtained by phone, using an

app, by outstretched arm on the street, and at cab stands. Joe Sanfelippo Cabs Inc. v.

City of Milwaukee, 46 F. Supp. 3d 888, 891 (E.D. Wis. 2014), aff’d, 839 F.3d 613

(7th Cir. 2016) (noting that traditional taxis can also operate as network vehicles

under Milwaukee ordinance); see also

OMPETITION

UREAU

, supra note 10 (report-

ing that four Vancouver taxi companies “have . . . jointly launched a software applica-

tion that allows passengers to request and track taxis, pay with their credit card and

rate their driver”).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 27 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 27

semble cab stands at a transportation hub like a train station.

In a

part of Florida, Uber is now “experimenting” with allowing customers

to call a central dispatch line and obtain an Uber ride by phone.

Other distinctions between traditional taxis and new entrants like

Uber also are not fixed. For example, the Curb app for obtaining yel-

low taxis in New York requires passengers to set up an account,

which, echoing the Uber experience, enables riders to follow their trip

on a map as it progresses, bill their trip automatically to a credit card

and receive an emailed receipt, rate their trip experience afterwards,

and reduce the anonymity of the rider-driver interaction.

On the flip

side, the Uber trip experience also could become more like that of a

traditional taxi cab trip. While Uber is associated with seamless credit

card payment in the United States, in some developing countries Uber

drivers take cash, similar to traditional taxis in the United States.

Because app and traditional taxis are substitutes competing in the

same market, the stringency of regulations governing each type of taxi

will affect the supply of, and demand for, the other type of taxi. Less

stringent regulation of app-dispatched taxis will lead to less consumer

demand for traditional taxis, because app-dispatched taxis will be

61. See, e.g., Carl Bialik, How Uber’s Truce with Regulators Played Out During

the DNC,

IVE

HIRTY

IGHT

(July 29, 2016), http://fivethirtyeight.com/features/how-

ubers-truce-with-regulators-played-out-during-the-dnc/ (referring to Uber’s “dedi-

cated pickup spot on the 30th Street side of the station” during the Democratic Con-

vention in Philadelphia); Drive with Uber Philadelphia,

BER

, http://

philadelphia.ubermovement.com/uber-at-30th-st-station/ (last visited Jan. 19, 2017).

62. Spencer Woodman, Uber Is Experimenting with a Phone Dispatch System,

ERGE

(June 14, 2016), http://www.theverge.com/2016/6/14/11933140/uber-phone-

dispatch-system-florida (“The call-in feature is part of a local program that will grant

publicly subsidized Uber rides to low-income residents who do not have cars or easy

access to the area’s bus-based public transit system.”)

63. See Andrew J. Hawkins, Taxi-Hailing App Curb to Relaunch with a New Attack

Plan Against Uber,

ERGE

(Mar. 23, 2016), http://www.theverge.com/2016/3/23/

11294758/curb-app-taxi-hail-uber-nyc-verifone;

URB

, supra note 60; see also

OM-

PETITION

UREAU

, supra note 10, at 3 (four Vancouver taxi companies “have . . .

jointly launched a software application that allows passengers to request and track

taxis, pay with their credit card and rate their driver”). As mentioned above, however,

New York City maintains that few yellow taxi trips are booked through apps such as

Curb in New York City today. Taxi industry plaintiffs in ongoing litigation in the City

suggest otherwise. See supra note 60.

64. Jon Russell, Uber Begins to See the Payout from Accepting Cash Payments,

ECH

RUNCH

(Feb. 8, 2016), https://techcrunch.com/2016/02/08/uber-begins-to-see-

the-payout-from-accepting-cash-payments/. In New York City, Uber would have dif-

ficulty operating under “black car” base licenses if it moved to accept cash payments

from riders, because black car bases are required by regulation to be paid for over 90

percent of their business through non-cash means. See supra note 44. Also, allowing

passengers to use transportation network company vehicles without credit cards might

reduce the safety of drivers, because passengers could no longer be easily identified.

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note 7, at 142.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 28 17-APR-17 10:20

28 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

cheaper and consequently attract more consumers. It will also lead to a

reduction in the supply of traditional taxis, as the number of traditional

taxis will fall to reflect the erosion of demand for them.

In major cities where apps are well-established, the taxis affili-

ated with them are already providing a growing number of taxi trips in

response to growing consumer demand.

The supply of traditional

taxis is falling as drivers are migrating from the traditional to the app-

based sector, and driving for Uber and its competitors.

Driving for

Uber may be more profitable for drivers, because they avoid paying

medallion leasing fees; an Uber vehicle does not require a medallion,

though drivers pay Uber a service fee on fares.

Medallion owners

are responding by reducing the rates that they charge drivers to lease

medallions, and by keeping taxis off city streets due to the erosion in

demand for their services.

It may be that app-dispatched taxis are the wave of the future and

that they will entirely supplant traditional taxis. But as of now tradi-

tional taxis are still being demanded and supplied, albeit it at reduced

levels. This may be because not all consumers have the credit cards

that are required to use taxi apps such as Uber in the United States.

65. See infra note 121 and accompanying text.

66.

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note 7, at

89. New York City Council has enacted legislation that will make it easier for drivers

to switch between driving yellow taxis and for-hire vehicles (such as Uber), by creat-

ing a “universal driver’s license.” N.Y.C., N.Y., Law 2016/051 (Apr. 21, 2016) (con-

cerning a universal driver’s license for taxicab and for-hire vehicle drivers). The

legislation “codif[ied]” the Taxi and Limousine’s “practice” since 2015 “of issuing a

‘Universal License.’” Hearing on Intros Nos. 658, 1080, 1092, 1095 & 1096 Before

the N.Y.C. Council Transp. Comm. (Feb. 29, 2016) (testimony of Meera Joshi, Com-

missioner & Chair, N.Y.C. Taxi & Limousine Comm’n) [hereinafter Joshi

Testimony].

67. Complaint at 21, CGS Taxi LLC v. City of New York, No. 653264/2015 (N.Y.

Sup. Ct. Sept. 30, 2015); Complaint at 20, Singh v. City of New York, No. 701402/

2017 (N.Y. Sup. Ct. Jan. 30, 2017).

68. Jennifer Bain et al., Why Uber Is to Blame for Yellow Cabs Hogging Street

Parking,

N.Y. P

OST

(Aug. 26, 2015), http://nypost.com/2015/08/26/why-uber-is-to-

blame-for-yellow-cabs-clogging-street-parking/.

69. One possible proxy for the number of people without credit cards is the number

without bank accounts, which is almost ten percent of the population in New York

State. John Aldan Byrne, 1 in 10 New Yorkers Doesn’t Have a Bank Account,

N.Y.

OST

(June 7, 2014), http://nypost.com/2014/06/07/1-in-10-new-yorkers-dont-have-

bank-accounts/. For national estimates, see

OMM

FOR

EVIEW OF

NNOVATIVE

R-

BAN

OBILITY

ERVS

, supra note 7, at 142. We might think that people without bank

accounts—and credit cards—would lack the means to take taxis, but these individuals

also may be less likely to own cars, which might leave them dependent on taxis when

public transit is not an option. There are studies indicating that taxi riders include low-

income persons, including in New York City.

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note 7, at 135 (“Low-income households use taxis

more often than middle-income households and at about the same rate as high-income

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 29 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 29

Also, cab stands, in which traditional taxis line up for passengers, re-

main an entrenched mechanism of obtaining taxis at hubs such as air-

ports and train stations.

Lightly regulating app-dispatched taxis,

without reducing the regulatory burden on traditional taxis, could im-

pose social losses, as unnecessary regulatory burdens translate into

higher taxi fares for consumers of traditional taxis.

households.”) (citations omitted); Id. at 140; Wyman, supra note 21, at 161 n.183

(citing sources).

As mentioned above, the mandatory use of credit cards is not integral to the app-

technology, and Uber might be able to devise ways to incorporate people without

access to such cards into its platforms. Abroad, Uber does accept payment in cash,

where there is limited access to credit cards. Russell, supra note 64.

70. Geradin, supra note 8, at 9. However, the persistence of cab stands may be

partly the result of decisions by regulators, not a reflection of consumer demand.

Regulators may be choosing to continue to maintain cab stands to steer passengers to

traditional taxis, which may be the only vehicles legally authorized to pick up passen-

gers at these stands. See supra note 59 (describing the rules of the Port Authority of

New York and New Jersey and the New York Taxi & Limousine Commission).

71. See also

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

supra

note 7, at 68 (“[I]f TNCs weaken or bankrupt many taxi services, then those without

credit cards and smartphone access may find themselves with fewer mobility options

than before.”).

My argument that new and old actors need to be jointly regulated when they are

substitutes extends an argument made in environmental law.

Revesz & Kong analyze whether, when establishing more stringent environmen-

tal standards for new sources such as new power plants, regulators should establish

the higher standards for the new sources and then determine in a second step whether

to apply them to existing sources, or establish standards for new and old sources in

one step. Revesz & Kong, supra note 50, at 1618. They argue that “the socially opti-

mal approach is to jointly determine the new source standard and the” rule for existing

sources. Id. at 1582–83. The reason is “the disparity between the regulatory strin-

gency that applies to new sources and” existing sources “has on the rate at which”

existing sources “close down and are replaced by new sources.” Id. at 1617. As the

history of regulating power plants under the Clean Air Act demonstrates, establishing

stringent standards for new sources, while not requiring existing sources to meet these

higher standards, may mean that the environmental benefits of the more stringent

standards are not realized, because few new sources will be built and the owners of

the old sources will prolong their useful lives. Id. at 1582.

There are differences between taxi regulation and the environmental regulatory

context that concerns Revesz & Kong. In the environmental context, regulators are

establishing more stringent standards for new sources that have yet to be built, with

the risks described in the above paragraph. In the taxi context, regulators are establish-

ing less stringent standards for new sources than for the incumbent sources, both of

which are already driving on city streets. So, unlike environmental regulators, taxi

regulators do not have to worry that the standards they are establishing will discour-

age the entry of new sources (assuming they establish standards for the new sources at

the optimal level). They face the risk of social welfare losses from excessive regula-

tion of the traditional taxi fleet, which seems likely to have some market share for the

foreseeable future.

Though the risk concerns new sources in the environmental context and existing

sources in the taxi context, the source of the risk is the same: any disparity between

the standards applicable to new and old sources will affect entry and exit of the substi-

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 30 17-APR-17 10:20

30 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

Nonetheless, it is important to recognize that differential stan-

dards for traditional taxis and transportation network companies still

could be welfare-enhancing, even though traditional taxis and trans-

portation network companies are substitutes, and the distinctions be-

tween them are—and could be increasingly—blurry. Different

standards would be justified if the benefits of such standards exceed

the costs from a societal perspective.

For example, as discussed be-

low, there may be reasons to regulate the fare levels of street-hailed

taxis but not e-hailed taxis. However, there should be compelling evi-

dence that differential standards are cost-justified before allowing

them, because of the likelihood that different regulatory standards for

different taxi services could create opportunities for “regulatory arbi-

tute source in unintended ways. The best way of addressing the risk is to simultane-

ously establish standards for new and old sources, in the taxi context as in the

environmental context.

72. Revesz & Kong recognize—as I do, above—that establishing standards for old

and new sources simultaneously does not require that the standards for both be identi-

cal. Id. at 1620–21 (examples three and four recommend different standards for new

and existing sources). The issue in the environmental context that concerns Revesz &

Kong is often that regulators are establishing new, more stringent sources for new

sources; the question is whether to apply the more stringent standard to an old polluter

that already has invested in pollution control equipment to satisfy a pre-existing regu-

latory standard. Id. at 1582. Steven Shavell argues that it might not be justified on

social welfare grounds to apply the new standard to the old source, if the benefits of

applying the new standard (the incremental reduction in pollution it will yield) are

lower than the costs (what the firm will have to pay for additional pollution control

equipment to meet the new standard). See Steven Shavell, On Optimal Legal Change,

Past Behavior, and Grandfathering, 37

J. L

EGAL

TUD

37, 37–39 (2008). Revesz &

Kong agree with Shavell that less stringent standards might be justified for old

sources compared with new sources on social welfare grounds, but they suggest that it

would in a narrower range of circumstances than Shavell argues, because of strong

assumptions in his model. Revesz & Kong, supra note 50, at 1615.

As Richard Revesz pointed out to me, the taxi context is the mirror image of the

environmental context that drives the analysis by Shavell and by Revesz & Kong. In

the taxi context, regulators are establishing a less stringent regime for new sources

(app-dispatched vehicles), not a more stringent regime for old sources. The issue is

whether the old sources (traditional taxis) should benefit from the new, less stringent

regime, or continue to be regulated under the pre-existing more burdensome regime.

As indicated in the text, under a social welfare analysis, the question should be framed

in terms of whether the benefits of retaining the pre-existing regime outweigh the

costs of operating under it. Id. at 1593–94. Even if the old regulatory standard might

not be justified in cost-benefit terms at this point, it still might be desirable to retain it

for old sources because the benefits of retaining the standard exceed the costs. As

explained above, I think that we should be careful in assuming that the social benefits

of retaining the old regulatory regime for the traditional taxi industry exceed the social

costs.

Thank you to Richard Revesz for his comments on the relevance of his work

with Kong, and Shavell’s work, for the taxi context.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 31 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 31

trage” by market players and inhibit meaningful competition.

Thus

in determining regulatory standards for taxis in the age of Uber, regu-

lators should treat all vehicles providing point-to-point transportation

in response to customer requests as a unit, and establish differential

standards for vehicles providing different types of services only if

such standards can be justified.

B. Rethinking the Pillars of Taxi Regulation

I now consider more concretely how we should be regulating

both traditional and e-hailed taxis.

There historically have been five pillars of taxi regulation.

First,

regulators have limited entry to the taxi business by limiting the num-

ber of taxis. Limits on entry often have been enforced by requiring

taxis to have one of a limited number of licenses, often called medal-

lions, to operate. Second, regulators have regulated taxi fares, often by

73. Edelman & Geradin, supra note 10, at 327 (“[W]hen these services take regula-

tory shortcuts, it is difficult to know whether the services gain traction through genu-

ine excellence and efficiency, or through regulatory arbitrage.”).

While I lean toward thinking that the same standards should apply to traditional

and e-hailed taxis, as the text indicates, I recognize that differential standards might be

cost-justified. Others are more insistent that the same standards should apply to tradi-

tional and e-hailed taxis. See, e.g., FTC Workshop, supra note 10, at 153 (remarks of

Adam Thierer) (advocating “deregulating down to give everybody an equal level

playing field”).

Although my concern is what is best from a societal point of view, it is interest-

ing to note that parts of the incumbent taxi industry now seem to think that from their

point of view, it would be better to reduce the stringency of the regulations imposed

on them to the level imposed on app-based services, to enable the incumbents to

compete. After initially seeking to block the entry of Uber and other e-hailing services

or to require that they are regulated to the same extent as traditional taxis, the incum-

bent taxi industry in some places is now arguing that the regulatory burden it faces is

unfairly excessive given the less stringent regulations for apps like Uber. See, e.g.,

Plaintiff’s Opposition to Defendants’ Motion to Dismiss at 9, Cambridge Taxi Drivers

& Owners Ass’n v. City of Cambridge, No. 1:16-cv-11357-NMG (D. Mass. Jan. 25,

2017), 2017 WL 373491 (case dismissed) (“It is within Defendants’ discretion and

ability to revisit existing Taxi Regulations, for example, to lessen the burden, particu-

larly financially, to which medallion owners and taxicab operators are subject.”); Vac-

caro, supra note 45 (Cambridge taxi companies “want Cambridge to lighten the

regulatory load on taxis,” for example by allowing taxis to establish “their fares”).

74. Different authors refer to different numbers of pillars (or categories) of regula-

tion. See, e.g.,

RANKENA

& P

AUTLER

, supra note 34, at 16–28 (five pillars); Chris-

tian Seibert, Taxi Deregulation and Transaction Costs, 26

CON

. A

71 (2006)

(three pillars); Geradin, supra note 8, at 4 (four pillars).

In some jurisdictions taxi regulators also have sought to advance environmental

objectives by encouraging taxis to purchase hybrid or energy efficient vehicles. These

efforts have met with mixed success due to legal impediments. See Christina Ma,

Hybridizing Federal and State Regulation of Clean Taxis Introduction, 42

NVTL

. L.

. N

EWS

& A

NALYSIS

10840 (2012); Sarah E. Light, Precautionary Federalism and

the Sharing Economy, 66

MORY

L.J. 333, 374–76 (2017).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 32 17-APR-17 10:20

32 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

establishing uniform fares that prevent “fare competition.”

Third,

they have imposed health and safety regulations to protect consumers,

such as background checks and insurance requirements for drivers,

and standards or model types for vehicles. Fourth, regulators have im-

posed some regulations to protect the economic interests and health

and safety of taxi drivers, such as requirements that drivers be covered

by workers’ compensation insurance. Fifth, regulators have imposed

universal service requirements, which require taxis to serve customers

unless there are justifiable reasons for not doing so.

I recommend

restructuring all five pillars to varying degrees, and adding a new pil-

lar, which would charge regulators with guarding against the forma-

tion of a new taxi monopoly.

Entry Regulation

Since before World War II, many cities have limited the number

of taxis.

Historically, a number of reasons have been offered for lim-

iting entry, including reducing traffic congestion, controlling pollu-

tion, protecting driver incomes, and reducing accidents caused by taxis

competing aggressively with each other for fares.

The most salient

of these justifications today is the idea that we need to limit (or cap)

the number of taxis to reduce the congestion that taxis cause on city

streets.

With the growing popularity of taxi apps, the number of vehicles

providing taxi services in cities like New York and London is increas-

ing, leading to recent calls in both cities to limit the number of app-

dispatched taxis in the name of addressing traffic congestion.

For

example, in the summer of 2015, New York City Mayor Bill de Blasio

advocated severe constraints on the growth in the number of for-hire

vehicles, including Uber vehicles, although he then dropped this pro-

posal after Uber launched a highly public campaign against it.

75.

RANKENA

& P

AUTLER

, supra note 34, at 22–23.

76. See, e.g., N.Y.C. Admin. Code § 19-507(a)(2) (West 2016) (“No driver of a

taxicab shall refuse, without justifiable grounds, to take any passenger or prospective

passenger to any destination within the city.”); 35 R.C.N.Y. § 54-20 (“Operations—

Refusing Passengers”).

77. See also Edelman & Geradin, supra note 10, at 308 (“Whatever the difficulty of

revisiting applicable regulation, the task appears to be compulsory.”).

78.

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note 7, at

41.

79. See Wyman, supra note 21, at 149, 168.

80. See Rory Cellan-Jones, Will London’s Mayor Put the Brakes on Uber?,

BBC

EWS

(May 22, 2015), http://www.bbc.com/news/technology-32842332.

81. Colleen Wright, Uber Says Proposed Freeze on Licenses in New York City

Would Limit Competition,

N.Y. T

IMES

(June 30, 2015), http://www.nytimes.com/

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 33 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 33

January 2016, the City released a report that concluded that increasing

numbers of Uber vehicles are not “driv[ing] the recent increase in con-

gestion in the” city’s Central Business District, because trips by Uber

and other for-hire vehicles “are largely substituting for yellow taxi

trips.”

Traffic congestion is an externality because a taxi does not expe-

rience the full cost of the space it is taking up on the street.

However

there is little basis for limiting the number of taxis to reduce traffic

congestion.

If we want to reduce congestion, we should start by defining a

targeted level of congestion, and then introducing policy changes to

achieve that goal “‘in the least-cost way,’” by enlisting sources to

reduce congestion in line with their costs of congestion reduction.

There are many different sources of congestion in addition to taxis:

other vehicles, such as private cars and bicycles; other uses of the

streets, such as bike lanes, and pedestrian plazas; and “[p]opulation

and job growth,” among other factors.

Most likely, there is signifi-

2015/07/01/nyregion/uber-says-proposed-freeze-on-licenses-would-limit-

competition.html.

Technically, the proposal to limit the growth in for-hire vehicles was embodied

in a bill introduced in the City Council, which was supported by Mayor de Blasio. The

bill would have severely interfered with the ability of taxi apps like Uber to expand in

the next year by “limiting the issuance of new for-hire vehicle licenses,” pending the

completion of an analysis of the pollution and congestion impacts of the recent in-

crease in the number of for-hire vehicles on New York City streets.

EPORT OF THE

UMAN

ERVICES

IVISION

, supra note 25, at 2.

82.

FFICE OF THE

AYOR

, C

ITY OF

ORK

supra note 52, at 5. The report

emphasizes that it “does not recommend a cap on for-hire vehicles at this time.” Id. at

11. An even more recent report, from one of the authors of the City’s study, finds that

TNC ridership has continued to increase since the period examined in the City’s re-

port.

CHALLER

ONSULTING

NSUSTAINABLE

? T

ROWTH OF

-B

ASED

IDE

ERVICES AND

RAFFIC

, T

RAVEL AND THE

UTURE OF

ORK

ITY

9 (2017)

, http:/

/www.schallerconsult.com/rideservices/unsustainable.pdf (“Fall 2016 [TNC] ridership

averaging 15 million passengers per month was more than triple ridership levels in

Spring 2015, the period studied in the City FHV report.”); see also id. at 16 (compar-

ing the most recent analysis to that in the City’s report).

83. Tizra S. Wahrman, Breaking the Logjam: The Peak Pricing of Congested Ur-

ban Roadways Under the Clean Air Act to Improve Air Quality and Reduce Vehicle

Miles Traveled, 8

UKE

NVTL

. L. & P

’

. 181, 196 (1998) (“Congestion is a

classic negative externality. As additional road users occupy the road, the quality of

service provided to all users declines.”); see also Jonathan Remy Nash, Economic

Efficiency Versus Public Choice: The Case of Property Rights in Road Traffic Man-

agement, 49

B.C. L. R

673, 685–86 (2008).

84. Wyman, supra note 21, at 149 n.126 (excerpting Richard B. Coffman, The Eco-

nomic Reasons for Price and Entry Regulation of Taxicabs: A Comment, 11

J. T

RANS-

PORT

CON

. & P

’

288, 295 (1977)).

85.

FFICE OF THE

AYOR

, C

ITY OF

ORK

, supra note 52, at 5. This report

usefully distinguishes the factors that contribute to congestion in Manhattan’s Central

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 34 17-APR-17 10:20

34 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

cant variation among jurisdictions in the relative contributions of these

different sources of congestion and accordingly, in the costs and bene-

fits of curtailing those sources.

There are reasons to believe that taxis, at least in certain places,

might alleviate, rather than exacerbate, congestion.

Taxis can substi-

tute for private cars if there are enough taxis to eliminate the need to

own a car.

Multiple passengers travelling in the same taxi at the

same time can further reduce congestion by reducing the number of

private cars and taxis required to provide transportation services—and

Uber, Lyft and other companies are now offering taxi pooling services

in major U.S. cities that enable multiple customers to share a ride.

Then there is the effect of taxi availability on the use of mass transit. If

taxis are cheap and accessible relative to mass transit, people might

substitute taxis for mass transit, thereby increasing congestion.

But

Business District into “those that affect the volume (demand) of travel and those that

affect the capacity (supply) of the roadway.” Id. at 4.

86. There is very limited empirical evidence about the impact of Uber and other

apps on congestion, “although at least one study has shown that TNCs may be substi-

tuting for both transit and driving trips.”

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

supra note 7, at 29 (citing Lisa Rayle et al., App-Based, On-De-

mand Rise Services: Comparing Taxi and Ridesourcing Trips and User Characteris-

tics in San Francisco (presented at the 2015 Annual Transportation Research Board

Meeting, Washington, D.C. Jan. 2015)). As already mentioned, New York City re-

leased a study showing that the growth in for-hire vehicles attributable to Uber and

other companies is not a major contributor to increasing congestion in the City’s Cen-

tral Business District.

FFICE OF THE

AYOR

, C

ITY OF

ORK

, supra note 52, at

87. Wyman, supra note 21, at 152 n.136 (citing sources). As one attentive reader

pointed out to me, reducing private car ownership also might reduce congestion by

reducing the number of cars driving around searching for parking places in crowded

places such as Manhattan.

88. A report for the Transportation Review Board refers to pooling as “concurrent

sharing of vehicles among strangers,” a phenomenon it contrasts with “sequential

sharing of vehicles.”

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

supra note 7, at 2.

The text assumes that concurrent sharing displaces private cars and taxis, but if

passengers shift from mass transit to concurrent sharing of taxis, then congestion

could increase because taxis hold fewer passengers than mass transit vehicles.

FFICE

OF THE

AYOR

, C

ITY OF

ORK

, supra note 52, at 6.

For a humorous article discussing the social counters facilitated by car-pooling

services offered by Uber, Lyft and especially Via in New York City, see Caroline

Tell, With Via, Sharing More Than Just a Ride,

N.Y. T

IMES

(Dec. 30, 2015), http://

www.nytimes.com/2015/12/31/fashion/with-via-sharing-more-than-just-a-ride.html.

89. Wyman, supra note 21, at 151; see also Dobson, supra note 10, at 703 (discuss-

ing a study that “found that TNCs both compete with and supplement public transit”)

(citing Lisa Rayle et al., App-Based, On-Demand Ride Services: Comparing Taxi and

Ridesourcing Trips and User Characteristics in San Francisco (Univ. of Cal. Transp.

Ctr., Working Paper UCTC-FR-2014-08, Nov. 2014));

CHALLER

ONSULTING

, supra

note 82, at 18 (noting that in 2016, in New York City, as “taxi/for-hire ridership

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 35 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 35

the availability of cheap and accessible taxis also might increase mass

transit use, and therefore reduce congestion, because taxis can be used

to transport people from transit stops to their homes or offices.

Manhattan, for example, seventy-six percent of households do not

own a car,

likely due partly to the ease of access to public transpor-

tation, as well as taxis.

Assume that taxis are a major contributor to congestion in a juris-

diction, and that the costs of curtailing taxi use make it worthwhile to

target taxis as part of a package of policies to reduce congestion. Mid-

town Manhattan on a business day might be a place where it makes

sense to curtail taxi use to reduce congestion: there is evidence that

before Uber, yellow medallion taxis accounted for forty to sixty per-

cent of the vehicles on city streets in Manhattan’s central business

district.

Airports might be another place where taxis are major con-

tributors to congestion, and restrictions might be justified as a low-

cost way of addressing it, given limited space at airports. Nonetheless,

a city-wide cap on the number of medallion taxis, like New York

City’s current cap of 13,587 medallion taxis,

is unlikely to be the

lowest-cost method of curtailing taxi use in a place like Manhattan’s

central business district or at an airport. Jurisdiction-wide caps are

blunt tools that limit the number of taxis even in places and at times

where there are no congestion concerns, such as the city outside the

increase[ed],” “[s]ubway ridership declined for the first time in years and bus rider-

ship dropped for the third consecutive year”).

90. Rayle et al., supra note 89, at 17 (providing “evidence,” based on survey and

other data, “that ridesourcing both complements and competes with public transit”).

Some transit agencies are taking steps to facilitate transit riders using Uber and Lyft to

travel to and from transit stops, including incorporating the app services into the

transit agency app. See Ellen Powell, MBTA, Uber and Lyft Join Forces: Is This the

Future of Paratransit?,

HRISTIAN

. M

ONITOR

(Sept. 17, 2016), http://

www.csmonitor.com/USA/2016/0917/MBTA-Uber-and-Lyft-join-forces-Is-this-the-

future-of-paratransit-video. A recent report recommends that transit agencies embrace

transportation network services and other “‘emerging mobility’ services” as a “com-

plement” to mass transit and provides examples of such cooperation, but also ac-

knowledges that “there can . . . be tensions.”

RANSIT

ENTER

, P

RIVATE

OBILITY

UBLIC

NTEREST

: H

UBLIC

GENCIES

ORK WITH

MERGING

OBILITY

ROVIDERS

20 (2016)

91. Wyman, supra note 21, at 127 n.2 (citing

ENNINGSON

, D

URHAM

& R

ICHARD-

SON

RCHITECTURE

& E

’

, P.C., T

AXI

EDALLION

NCREASE

: D

RAFT

NVIRON-

MENTAL

MPACT

TATEMENT

ES-3 (May 2012)).

92. Wyman, supra note 21, at 149 & n.123. I use the term “might” advisedly. As

mentioned above, New York City recently released a report that concluded that

growth in the number of Uber vehicles is not “driv[ing] . . . the recent increase in

congestion in the [Central Business District].”

FFICE OF THE

AYOR

, C

ITY OF

ORK

, supra note 52, at 5.

93. Van Zuylen-Wood, supra note 7.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 36 17-APR-17 10:20

36 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

central business district or the airports.

Perversely, these caps often

lead taxis to concentrate in the areas of the city that are the most con-

gested, because these also are the places where it is easiest for taxis to

find passengers.

The caps also create barriers to entry that may in-

crease taxi fares, and reduce service quality and innovation.

Congestion charges likely will be a lower-cost way of reducing

congestion than limiting the number of taxis. They can be applied only

to the areas of the city prone to congestion, such as the airport, and

vary depending on the levels of congestion at a given place and time.

Economists have advocated congestion charges for all vehicles (not

just taxis) for decades.

They have been implemented in some places

such as London, though in a relatively crude manner.

In New York

City and other places, there have been powerful political impediments

to implementing congestion charges for vehicles.

100

There also may be

legal impediments to local governments implementing congestion

charges, as local governments may require state approval to imple-

ment such charges.

101

The advent of e-hailing services such as Uber might provide the

occasion for implementing congestion charging for taxis in specific

areas such as an airport where traffic congestion is a major issue, and

where taxis are a major contributor and a low-cost source of reduc-

tions. The technology that Uber uses for “surge pricing” could be

adapted to implement a very sophisticated form of congestion charge

94. Wyman, supra note 21, at 150.

95. In New York City, for example, yellow taxis concentrate in the Central Busi-

ness District of Manhattan, a few parts of Brooklyn, and the airports. Id. at 160.

96. Id. at 159–60.

97. Id. at 149–52 (citing sources).

98. For an interesting intellectual history of congestion pricing focusing on the Brit-

ish context, but discussing the American contributions to the idea of congestion pric-

ing, see David Rooney, The Political Economy of Congestion: Road Pricing and the

Neoliberal Project, 1952-2003, 25

WENTIETH

ENTURY

RIT

. H

IST

628 (2014).

99. Darb´era, supra note 10, at 11 (“[T]he present London congestion charge

scheme is far from being economically optimal . . . .”).

100. See, e.g., Nash, supra note 83; Bruce Schaller, New York City’s Congestion

Pricing Experience and Implications for Road Pricing Acceptance in the United

States, 17

RANS

. P

’

266 (2010). William Vickerey, who “offered the first de-

tailed account of a practical congestion charging system,” anticipated that congestion

pricing would be politically difficult to implement. Rooney, supra note 98, at 632.

101. In analyzing why then New York City Mayor Michael Bloomberg’s 2007 pro-

posal for congestion pricing failed, Schaller observes that “congestion pricing in New

York City needed approval by three legislative bodies (the City Council and each

house of the [New York State] Legislature) as well as the Governor [of New York

State].” Schaller, supra note 100, at 271.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 37 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 37

for taxis.

102

The charge could be varied based not only on where the

vehicle is driving and when to reflect its contribution to congestion,

but also the number of passengers, to credit individuals who are reduc-

ing the number of vehicles on the road by pooling. Revenues from the

charge must flow to either state or local governments or airport au-

thorities, not taxi drivers, lest the charge incentivize drivers to use

congested city streets to earn higher fares. The time is ripe for moving

from using a relatively crude cap on the number of taxis, implemented

through property-like taxi medallions, to using a sophisticated pricing

mechanism that builds on Uber’s surge pricing.

103

But regulators

should be cautious about imposing a congestion charge on taxis in the

absence of concrete evidence that taxis contribute to congestion and

that reducing the number of taxis is a low-cost way of reducing con-

gestion. In particular, congestion pricing for taxis only might discour-

age the use of taxis, which might increase congestion by increasing

the use of private cars and/or reducing the use of public transportation.

It likely would be preferable to apply congestion charges to vehicles

generally within the congested areas, at the congested times.

To be clear, the above analysis suggests that the number of app-

dispatched taxis should not be limited, even if it can be demonstrated

that they contribute to congestion in a particular community, because

congestion charges are a more efficient way of reducing congestion

from these vehicles. Congestion charges would be more efficient if

applied to vehicles generally,

104

but subject to the just mentioned ca-

veats, a congestion charge for taxis only in a specific area might be an

102. Surge pricing is a form of dynamic pricing. Under surge pricing, Uber fares

increase when passenger demand for vehicles is high, for example on a busy Friday

night. What is Surge

BER

, https://help.uber.com/h/e9375d5e-917b-4bc5-8142-

23b89a440eec (last visited Nov. 3, 2016). Surge pricing may bring demand and sup-

ply into balance, by attracting more drivers to the areas where prices are surging and

reducing demand for rides in these areas. On the reasons why surge pricing can work

to bring demand and supply into balance, see

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

supra note 7, at 74–76. For an interesting discussion of

whether Uber’s surge-pricing encourages more drivers to drive for Uber, see Nicolas

Diakopoulus, How Uber Surge Pricing Really Works,

ASH

. P

OST

(Apr. 17, 2015),

https://www.washingtonpost.com/news/wonk/wp/2015/04/17/how-uber-surge-pric-

ing-really-works/.

103. Darb´era also supports using a more sophisticated form of congestion pricing.

Darb´era, supra, note 10, at 11. For a recent, relatively crude, proposal to apply a form

of congestion charge on traditional New York taxis (and app-dispatched taxis), see

OVE

NY F

AIR

LAN

19–20, 22, 26 (Feb. 2015).

104. For a discussion of how congestion pricing for vehicles generally would work,

and some examples of congestion pricing in the U.S., see Nash, supra note 83, at

708–15, 723–25. For a call for congestion pricing in New York City in light of the

growth in the number of for-hire vehicles, see

CHALLER

ONSULTING

, supra note 82,

at 23–24.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 38 17-APR-17 10:20

38 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

efficient way of reducing congestion if taxis are contributors, and if

curtailing their use is a low-cost way of reducing congestion. Limits

on the number of traditional taxis also should be eliminated. These

limits will not reduce congestion in a world where other forms of taxis

are unlimited; limits on traditional taxis will merely induce consumers

to substitute app-hailed for street-hailed and other pre-booked taxis,

because app-hailed taxis will be more available as their numbers are

unlimited.

This last proposal to remove limits on the number of traditional

taxis, including street-hailed taxis, might strike some observers as un-

wise. For example, a 2015 report from the Transportation Research

Board’s Committee for Review of Innovative Urban Mobility Services

cautioned against removing regulatory barriers to entering the street-

hail taxi industry (or other regulations imposed on street-hailed

taxis).

105

The Committee seemed to be of the view that because there

are few natural barriers to entering the street-hail industry, if the num-

ber of street-hailed taxis is not regulated,

106

removing caps on the

number of these taxis would increase the number of taxis, leading to

problems of “oversupply,” and greater traffic congestion in already-

congested areas of the city, such as in “dense downtowns [and] air-

ports.”

107

The Committee pointed to the increase in the number of

taxis in the relatively small number of jurisdictions that removed lim-

its on the number of taxis in the late 1970s and early 1980s in the

United States.

108

These arguments ignore the fact that Uber and other

app-dispatched vehicles currently are essentially substitutes for street-

hailed taxis because these vehicles often arrive with virtually the same

immediacy as street-hailed taxis. So while New York City, for exam-

ple, still legally limits the number of yellow taxis, it has functionally

105.

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

supra note 7, at

80 (“Lessons from taxi deregulation suggest that allowing TNCs [transportation net-

work companies] to compete in the street-hail market would lead to oversupply and

excess competition; thus, the street-hail business does not appear be a promising area

for lighter regulation.”). But see

OMPETITION

UREAU

, supra note 10 (recom-

mending the elimination of “restrictions on the number of taxi plates” and that “all

drivers” be allowed to “respond to street hails, regardless of whether they work for a

taxi company or ride-sharing service, unless there is a compelling policy reason” to

restrict picking up street hails).

106.

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

supra note 7, at

70 (“The street-hail business—if unregulated—entails very low barriers to entry. A

person with a vehicle can drive along crowded streets and find riders.”).

107. Id. at 71, 77–79.

108. Compare id. at 49–51 (providing a generally negative account of experiments

with taxi deregulation in the U.S.), with Wyman, supra note 21, at 147–48 n.121

(discussing efforts to deregulate the taxi industry in the U.S. and other countries).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 39 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 39

removed the limit on the number of street-hailed vehicles by allowing

for-hire vehicles to pick up passengers through e-dispatch.

Technology might be harnessed to address concerns that formally

removing the existing legal limits on the number of street-hailed taxis

might lead to oversupply in certain geographic areas. As mentioned

above, variable congestion charges might be used to signal to street-

hailed and other taxis the areas of the city that they should avoid, and

the app provider might be charged with collecting the congestion

charge on behalf of the governmental authority. Also, all taxis ac-

cepting street hails might be required to affiliate with at least one taxi

app.

109

App companies with taxis picking up street hails might be re-

quired to provide affiliated drivers with constantly updated informa-

tion about where and when taxis are needed in a jurisdiction, given the

current demand for, and supply of, taxis.

110

These requirements would

enable street-hailed taxis to have information to tailor their work to

areas and times of unmet passenger demand, and avoid adding to con-

gestion by cruising in already well-serviced areas or times. More gen-

erally, the affiliation requirement would enable the City to recruit the

app companies to aid in enforcing taxi regulations to protect public

safety. For example, app companies could be required to ensure that

vehicles affiliating with their apps meet certain standards. The tech-

nology that companies such as Uber have pioneered means that the

situation is very different from the late 1970s and early 1980s. Regula-

109. On the idea of requiring taxis to affiliate with an app, see generally

UREAU DU

AXI DE

ONTR

EAL

& W

INDELS

ARX

ANE

& M

ITTENDORF

, LLP

TUDY FOR A

ENTRALIZED

PPLICATION FOR

AXIS IN

ONTR

EAL

(Apr. 2016), http://www.wind

elsmarx.com/resources/documents/Study%20for%20a%20Centralized%20Applica

tion%20for%20Taxis%20in%20Montreal%20-%20April%202016.pdf. This report in-

cludes a discussion of the experiences of jurisdictions that have considered requiring

taxis to join an app.

A precedent for an app-affiliation requirement might be New York City’s ex-

isting requirement that for-hire vehicles affiliate with a base.

N.Y.C. T

AXI

& L

IMOU-

SINE

OMM

’

, supra note 22, at 2 (“TLC-Regulated Industries”).

As implied in the idea that street-hailed taxis could be required to affiliate with at

least one app, I would not require that street-hailed taxis be required to affiliate with

only one app. Requiring them to affiliate with only one app would encourage them to

affiliate with the app with the most customers and drivers, because that is the app

through which they are most likely to gain business. In turn, limiting taxis to a single

app would enhance the dominant app’s market power.

110. The Committee for Review of Innovative Urban Mobility Services makes an

important point when it observes that, to avoid drivers “clustering in areas” where

they are not needed, it is necessary to provide drivers “with real-time information on

places where the ratio of drivers to customers [is] . . . high and where it [is] . . . low.”

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

supra note 7, at

78–79. Simply providing information on where demand is high, without information

on the supply of taxis, could lead taxis to drive to areas where there are many passen-

gers, but already enough taxis to serve them.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 40 17-APR-17 10:20

40 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

tors should not dismiss proposals for removing limits on entry now

based on past experience, but rather should consider how technologi-

cal changes might be enlisted to further public policy goals.

111

Fare Regulation

Many cities historically have regulated the fare levels of taxi ser-

vices.

112

The standard justification for regulating the level of taxi fares

for street-hailed taxis is imperfect information.

113

When passengers

hail taxis on the street, they are poorly positioned to assess whether a

fare that a taxi is proposing is reasonable because riders lack essential

information. The passenger will not know when the next taxi will

come by and what it will charge, and searching for additional taxis to

compare the prices that they would charge will be costly.

114

Regulated

fares avoid these search costs. Importantly, imperfect information

does not justify regulating the level of taxi fares when taxis are

booked by phone or summoned by app. When passengers call ahead,

they can price shop by calling multiple taxi companies, provided the

taxi companies will provide fare estimates that can be compared. Sim-

ilarly, passengers using an app can compare prices of different compa-

nies using the companies’ apps, assuming again that the companies

will provide fare estimates that can be compared.

115

111. See also

OMPETITION

UREAU

, supra note 10 (“[N]ew technologies which pro-

vide real-time data on the taxi industry may mitigate [congestion] problems which

previously arose in deregulation experiments.”).

112. “In nearly all cities, taxi fares are set by regulation . . . . The regulations most

commonly set a fixed fare rate that applies uniformly across all companies . . . .

[F]ares are calculated on the basis of an initial charge (the ‘drop’), along with mileage

and time charges.”

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

supra note 7, at 40. For the current regulated rates for New York City medallion

taxicabs, see 35 RCNY § 58-26.

113. Seibert, supra note 74, at 72 (discussing imperfect information and imperfect

coordination).

114. The driver also may lack information to negotiate a fare, because he or she will

not know when the next passenger will come by or his or her destination, and search-

ing for other passengers will be costly to the driver. Reflecting the potential that the

rider’s and the driver’s information base could be improved by seeking out other

drivers and riders, Gallick and Sisk argue in a well-known article that search costs are

the basis for fare regulation. The idea is that a uniform fare avoids the need to engage

in costly searches that could be used to establish a reasonable fare. Edward C. Gallick

& David E. Sisk, A Reconsideration of Taxi Regulation, 3

J. L. E

CON

. & O

117

(1987); see also Edelman & Geradin, supra note 10, at 303 (explaining why “dy-

namic pricing” is not generally permitted “for curbside hails”).

115. In June 2016, Uber announced that it will be moving to providing riders with

“upfront fares” before riders book a ride, to enable them to know the cost of their ride

before they book, not just to obtain an estimate of the likely fare. See Douglas Mac-

Millan, Uber Customers Will Get Upfront Pricing in New App Version,

ALL

TREET

(June 23, 2016), http://www.wsj.com/articles/uber-customers-will-get-upfront-pric

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 41 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 41

Economies of scale in the dispatch market may justify regulating

the level of taxi fares for taxis that are obtained by phoning ahead, and

potentially app-dispatched taxis. As mentioned above, traditional taxi

companies that dispatch taxis by radio, and e-hailing services such as

Uber, are two-sided markets.

116

Several decades ago, economists rec-

ognized that there are economies of scale in dispatching taxis by radio

that might enable a small number of companies to dominate the mar-

ket in a jurisdiction, and that justify regulating the level of taxi fares.

Customers benefit when a taxi company has more drivers, because it

is easier to get a taxi quickly; drivers benefit when the company has

more customers, because it is easier to get a fare.

117

Likewise today,

many argue that there are economies of scale in using apps to match

passengers and drivers, because passengers benefit when a platform

like Uber has large numbers of drivers, and drivers benefit when the

platform has a large network of passengers. Some commentators even

suggest that Uber is a natural monopoly because of these economies of

scale, and that regulating the level of its fares is warranted given the

potential that it may charge “above cost” fares using its market

power.

118

ing-in-new-app-version-1466731781; Arundhati Singh & Dennis Zhao, Upfront

Fares: No Math and No Surprises,

BER

(June 23, 2016), https://newsroom.uber.com/

upfront-fares-no-math-and-no-surprises/.

It is now possible to compare online the prices that Uber and Lyft charge for the

same trip. See Yoni Heisler, You Can Now Figure Out if Uber or Lyft Is More Expen-

sive Before You Ride,

N.Y. P

OST

(Mar. 31, 2016), http://nypost.com/2016/03/31/you-

can-now-figure-out-if-uber-or-lyft-is-more-expensive-before-you-ride/; Aimee Picchi,

Uber vs. Taxi: Which Is Cheaper?,

ONSUMER

(June 10, 2016), http://

www.consumerreports.org/personal-finance/uber-vs-taxi-which-is-cheaper/.

116. See supra note 33 and accompanying text.

117.

RANKENA

& P

AUTLER

, supra note 34, at 54–55; see also

OMM

FOR

EVIEW

NNOVATIVE

RBAN

OBILITY

ERVS

supra note 7, at 70 (stating that “dispatch

taxi firms [have] some qualities of a natural monopoly” because of “high barriers to

entry”); Darb´era, supra note 10, at 12 (“The market for taxis booking by telephone is”

a natural monopoly in Paris, where it is controlled by the “G7 monopoly.”).

118. Posner, supra note 10. For other commentators who suggest that Uber is a natu-

ral monopoly, see Weyl & White, supra note 10; Darb´era, supra note 10, at 12,

15–16; FTC Workshop, supra note 10, at 21 (remarks of Glen Weyl). On the other

hand, Joshua Gans doubts that Uber is a monopoly. FTC Workshop, supra note 10, at

35, 38 (remarks of Joshua Gans). Edelman and Geradin maintain that “the growth of

software platforms [including Uber] seems to trigger few competition law concerns.”

Edelman & Geradin, supra note 10, at 304. Liran Einav suggests that there are power-

ful network effects at play that may diminish competition. FTC Workshop, supra note

10, at 38 (remarks by Liran Einav). For a nuanced discussion of the potential for a

transportation network company to evolve into a monopoly, see

OMM

FOR

EVIEW

NNOVATIVE

RBAN

OBILITY

ERVS

supra note 7, at 76–77.

Viscusi et al. state that “[a] market is a natural monopoly if, at the socially opti-

mal quantity, industry cost is minimized by having only one firm produce.”

W. K

ISCUSI ET AL

., E

CONOMICS OF

EGULATION AND

NTITRUST

376 (2005).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 42 17-APR-17 10:20

42 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

It is important to recognize that the case for regulating the level

of taxi fares based on economies of scale in the dispatch service is

contingent on the dispatch service having sufficient market power to

dictate prices. Before the advent of Uber, there may have been econo-

mies of scale that enabled a single taxi company to exert market

power in some jurisdictions. But with Uber, these traditional taxi com-

panies face new competition and they are no longer monopolies,

which suggests that there is less justification for regulating fare levels.

Likewise, if the levels of these companies’ fares are not set by regula-

tors, these companies could compete with Uber on price, and therefore

reduce the likelihood that Uber itself will evolve into a monopoly.

Moreover, some jurisdictions might have “good substitutes [for taxis],

such as public transit,”

119

that compete with Uber and traditional taxi

companies for passengers, based partly on price.

What does this analysis imply for the regulation of taxi fares?

First, it suggests that we should not be regulating the fare levels of

apps such as Uber, at least at this point in time.

120

There is no ratio-

nale for doing so based on imperfect information. Also, Uber is not

yet a monopolist dictating market prices. In many places it seems to

still have many competitors that collectively have significant market

share, including both traditional taxis and competing apps, such as

Lyft. As an indication, in New York City, as of April 2016, yellow

taxis made 65 percent of trips, compared with 27 percent for Uber, 4

percent for Lyft, 3 percent for Via, and 1 percent for Gett, though the

share of trips by yellow taxis was down from “84 [percent] . . . in

April 2015.”

121

Moreover, it is not yet clear that Uber will evolve into

119.

RANKENA

& P

AUTLER

, supra note 34, at 55.

120. As others have noted, Uber sets the prices that its drivers charge. It “is a critical

price-setting intermediary,” Rogers, supra note 10, at 89, that “uses a version of aver-

age-cost pricing; all Uber drivers charge you the same amount, based on congestion as

well as distance,” Posner, supra note 10. The way that Uber sets prices for rides

through contracts with drivers recently has led to allegations that it is engaging in

price-fixing, contrary to the Sherman Act. First Amended Complaint, Meyer v.

Kalanick, 174 F. Supp. 3d 817 (S.D.N.Y. 2016) (No. 1:15-cv-09796), 2016 WL

950376. Judge Jed Rakoff denied the defendant’s motion to dismiss in this case.

Meyer, 174 F. Supp. 3d 817.

121. Elena Holodny, Uber and Lyft Are Demolishing Yellow Taxi Drivers,

. I

N-

SIDER

(Oct. 12, 2016), http://www.businessinsider.com/nyc-yellow-cab-medallion-

prices-falling-further-2016-10 (reporting data shared by Morgan Stanley); see also

CHALLER

ONSULTING

, supra note 82, at 9 (noting that in the fall of 2016, Uber

provided “72 percent . . . of all TNC trips in” New York City, Lyft provided “12

percent,” and “Via, Juno and Gett” provided the rest);

OMM

FOR

EVIEW OF

NNO-

VATIVE

RBAN

OBILITY

ERVS

, supra note 7, at 21 (noting that in New York City,

“[a]s of March 2015,” yellow cabs are making “10 times the number of trips made by

Uber cars”); id. at 22 (“Lyft provided 30 percent of trips” by 380 TNC users surveyed

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 43 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 43

a monopoly. New players could enter the app market, as there are

relatively low fixed costs in establishing a taxi app, and it is easy for

passengers and drivers “to join and use several platforms.”

122

Even if

Uber evolves into a monopoly that can dictate taxi fares, the case for

regulating fare levels may be selective, rather than general. Some cit-

ies might be justified in not regulating its fare levels because their

residents have other sources of transportation, such as mass transit.

Other cities may be required to regulate fares because there are no

substitutes.

The levels of fares charged by traditional taxi companies dis-

patching taxis in response to phone calls also should no longer be

regulated. Imperfect information does not justify regulating these

companies’ fares, as passengers can comparison shop by calling

around to different taxi companies or comparing the fares charged by

Uber and traditional taxi companies. If there historically was a single

traditional taxi company because of economies of scale, the advent of

Uber means that the traditional monopoly now has competition and so

economies of scale no longer argue in favor of regulating fare levels.

Indeed, as mentioned above, it may be helpful to deregulate the fare

levels of traditional taxi companies to enable them to compete on price

with Uber, to reduce the likelihood that it will evolve into a

monopoly.

123

The most difficult question is whether fare levels should continue

to be regulated for street-hailed taxis. Requiring street-hailed taxis to

use government determined fares inhibits these taxis from competing

with transportation network companies on price,

124

and leaves these

in San Francisco, while UberX provided fifty-three percent of trips and “other Uber

services . . . represented another 8 percent.”).

However, there are indications that Uber may be becoming a significant provider

of taxi services in some places in the United States. See, e.g., Meyer, 174 F. Supp. 3d

at 821 (“Uber’s own experts have suggested that in certain cities in the U.S., Uber

captures 50% to 70% of business customers in the combined market of taxis, cars for

hire, and mobile-app generated ride-share services.”).

In China, Didi Chuxing, not Uber, is the leading taxi app, and Uber sold its

Chinese operations to Didi Chuxing. Paul Masur & Mike Isaac, Uber to Sell to Rival

Didi Chuxing and Create New Business in China,

N.Y. T

IMES

(Aug. 1, 2016), http://

www.nytimes.com/2016/08/02/business/dealbook/china-uber-didi-chuxing.html.

122. Evans & Schmalensee, supra note 33, at 166. Switching between platforms is

often referred to as multihoming.

123. Consistent with what I am suggesting, New York City historically has not regu-

lated the fare levels of for-hire vehicles not providing street-hailed service, because

passengers using such vehicles have the opportunity to comparison shop. Brief for

Respondents at 18–19, Glyka Trans LLC v. City of New York, No. 2015-11661 (N.Y.

App. Div. Apr. 15, 2016).

124. The traditional taxi industry argues that government fare regulation inhibits it in

the current competitive environment. Amended Complaint at 8–9, 55, Melrose Credit

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 44 17-APR-17 10:20

44 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

companies free to set their fares below the regulated fare levels of

street hailed taxis to attract riders away from them.

125

But the problem

of imperfect information remains today, as passengers hailing a cab on

the street still may be poorly positioned to evaluate the reasonableness

of the fare that the driver offers or to negotiate a fare with the driver.

The problem is much less acute than in the past because many of the

people hailing taxis on the street have smartphones that they easily can

use to check when an Uber vehicle might arrive at their location, and

know the fare that the Uber vehicle would charge.

126

Since search

costs now are lower for passengers hailing from the street, they are

better positioned to evaluate the reasonableness of fares proposed by

drivers. Drivers have access to the same information, and presumably

would be disciplined in the fares they propose by competition from

transportation network vehicles such as Uber cars.

127

Nonetheless, un-

til everyone has a smartphone, there still will be some potential taxi

Union v. City of New York, No. 1:15-cv-09042 (S.D.N.Y. Mar. 7, 2016) (referring to

“inflexible metered fare restrictions”). Medallion owners suggest that government-

determined fares impede the industry from competing for drivers as well as passen-

gers. See, e.g., Complaint at 8, Newark Cab Ass’n v. City of Newark, No. 2:16-cv-

04681 (D.N.J. Jan. 17, 2017), 2017 WL 214075 (case dismissed) (“By requiring tradi-

tional taxis to operate under burdensome regulations and to charge only City-dictated

prices, while allowing the de facto taxis to pay lower fees and charge fares that esca-

late well above taximeter fares at times of high demand, the City has given the de

facto taxi companies an unfair advantage in recruiting drivers. Transportation Plain-

tiffs have, as a result, lost many drivers to the de facto taxi companies, resulting in

lost revenue and a higher percentage of idle taxis.”); id. at 27–28 (“The absence of

required, uniform rates gives Uber a substantial competitive advantage over Transpor-

tation Plaintiffs . . . . These pricing differences help Uber attract customers (and build

brand loyalty) at certain times and give them an edge over Transportation Plaintiffs in

the recruitment of drivers.”).

125. For an interesting article comparing the fares of Uber and traditional taxis, see

Picchi, supra note 115. In arguing that New York City medallion taxis should have

greater pricing flexibility, the medallion owner and financier plaintiffs in Melrose

Credit Union note that “Uber announced on January 26, 2016, that it was going to

dramatically slash its fares in New York City to further accelerate ridership gains,

rendering them as much as 35% lower than regulated medallion taxicab fares.”

Amended Complaint at 64, Melrose Credit Union, No. 1:15-cv-09042.

126. “Currently, [sixty-four] percent of Americans own smartphones, a percentage

that reflects rapid growth (from 35 percent in 2011) across all common demographic

categories (income, gender, age, and race).”

OMM

FOR

EVIEW OF

NNOVATIVE

R-

BAN

OBILITY

ERVS

, supra note 7, at 143. For a discussion of the ease of obtaining

fare information about Uber rides, see supra note 115.

127. Also, if regulators require street hailed taxis to affiliate with at least one app,

then one or more of the apps may determine the fare that the driver charges. As

mentioned above, Uber today determines the fares that drivers charge. See supra note

120. If an app company determines the fare, even when the taxi is street hailed, then

there is no scope for bargaining between the passenger and the driver, and no room for

the driver to take advantage in fare-setting of the passenger’s lack of information

about when the next taxi will be available and what it might charge.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 45 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 45

riders who lack the ability to compare competing fares and who could

be vulnerable to high fares from taxi drivers if drivers and riders were

left to determine fares on an individualized basis. Taxi fares increased

in jurisdictions that deregulated taxi fares (and entry) in the late 1970s

and early 1980s, before the spread of smartphone technology.

128

The idea of drivers and riders determining fares in an unregulated

environment points to another reason to retain fare level regulation for

street-hailed taxis: in the absence of the government establishing fare

levels, it will be necessary for drivers and riders to bargain. This will

make it more cumbersome for passengers to use street hailed taxis and

possibly encourage a shift to Uber and other transportation network

companies with standardized pricing, and thus increase the likelihood

that Uber will gain market dominance. Allowing bargaining over fares

also might lead to discrimination against passengers based on prohib-

ited grounds such as their destination or race, as drivers could insist on

higher fares to take different passengers.

129

Moreover, the need for

riders and drivers to bargain could give rise to congestion externalities

on the streets and at cab stands as taxis take up space on the street

while riders and drivers bargain.

There are intermediate options that regulators might consider to

provide the street-hailed sector with greater latitude to compete with

transportation network company vehicles on price, but not go so far as

to completely delegate fare-setting to driver-passenger negotiations.

For example, government fare level regulation might take the form of

establishing maximum fares, rather than prescribing uniform fares.

130

128. Wyman, supra note 21, at 160 n.178 (discussing the effects of experiments with

the removal of entry and fare regulation); see also

OMM

FOR

EVIEW OF

NNOVA-

TIVE

RBAN

OBILITY

ERVS

supra note 7, at 50.

129. On the universal service requirement for taxis, see infra notes 201–203 and

accompanying text.

130. Cf.

OMPETITION

UREAU

, supra note 10 (recommending the “eas[ing] of price

controls, such as regulated taxi fares, to allow fares to be adjusted during periods of

varying demand, such as weekends, evenings and bad weather”). The Canadian Com-

petition Bureau seems to think that “maximum fares” might be desirable, as it refers

positively to the introduction of maximum fares in Australia.

OMPETITION

UREAU

supra note 10, at n.53.

There are longstanding arguments that taxi fare level regulation should take the

form of maximum fares, not uniform fares.

RANKENA

& P

AUTLER

, supra note 34, at

37–66. Frankena and Pautler’s analysis is somewhat dated, however, and additional

economic analysis would be beneficial of the merits of using maximum fare regula-

tion given today’s market conditions in the taxi industry.

Miami-Dade County is an example of a jurisdiction using maximum fare regula-

tion for incumbent taxis in the age of Uber. Miami-Dade County allows traditional

taxis “to establish their own rates, with the exception of the airport or seaport, as long

as those rates do not exceed a maximum rate established by the County Commission.”

Motion to Dismiss Amended Class Action Complaint at 6–7, n.3, Miadeco Corp. v.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 46 17-APR-17 10:20

46 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

Street-hailed taxis could be allowed to charge fares below the maxi-

mum levels, but obligated to disclose their fare structures and poten-

tially how they depart from the maximum fare levels. Still, allowing

taxis to set their own fares as long as they remain below a maximum

fare would lead to drivers and riders bargaining with the risks identi-

fied above, unless drivers somehow were aggregated into fleets that

adopted a common fare structure, perhaps by joining an app, and

found a way to signal to customers the group’s fare structures before

the customers entered the vehicle. Another intermediate option would

be for regulators to set a small number of regulated fare levels—say,

“high,” “medium,” and “low.”

131

Regulators could require street-

hailed drivers to choose among these fare levels, and signal their

choice to consumers, perhaps by displaying a colored light on the

outside of the cab. Different colored lights could be used to signal

each of the regulated fare levels, with the number of possible fare

levels (and therefore light colors) limited to reduce consumer confu-

sion. Such a system would give street-hailed taxis some, but not total,

flexibility to compete based on price without requiring drivers and

passengers to engage in costly fare negotiations for every ride. How-

ever, allowing drivers to choose a fare level might facilitate discrimi-

nation against passengers based on prohibited grounds such as

destination or race, because drivers might insist on the highest fare

level if they do not want to drive a passenger.

If the fare levels of street-hailed taxis remain regulated, even

through maximum fares or a limited menu of fare levels from which

drivers select, regulators will need to grapple with the fact that taxis

that take street hails also may be picking up passengers by phone and

by app. This raises the question of whether to regulate the fare levels

of street-hailed taxis not only when they are picking up street hails,

but also when they are picking up passengers by phone and by app. As

discussed above, there is not a strong argument for regulating the fare

levels of taxis summoned by phone or app based on information costs

or economies of scale. Also, regulating the fare levels of taxis that

take street hails when they are summoned in these ways will inhibit

the ability of taxis to compete on price with other vehicles, such as

Uber vehicles currently, that pick up passengers only through app. On

the other hand, it might seem administratively simpler to regulate all

the fares of taxis that can be street hailed, regardless of whether the

taxi is in fact picking up passengers through street hails. Regulating

Miami-Dade County, No. 1:16-cv-21976 (S.D. Fla. June 29, 2016) (referring to Ordi-

nance No. 16-43).

131. Thank you to Marcel Kahan for this suggestion.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 47 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 47

the fares of taxis only when they are engaged in a street-hailed ride,

but not when driving a passenger met through an app, raises the ques-

tion of how to set the regulated fare given that the driver potentially

has the option of picking up by app and earning the prevailing market

fare instead of the regulated fare for street hails. It is possible that the

behavior of taxi drivers will not be heavily influenced by the potential

to charge different fares depending on how they pick up passengers

and so regulators could ignore the market fare option in establishing

the regulated fare for street hails. Drivers may take the first passenger

that they encounter, regardless of whether they encounter that person

through a street hail or by app, because they prefer a “bird in the hand

to two in the bush.”

132

However, in theory, taxis could behave strate-

gically if their fares are regulated for street-hailed passengers, but not

other passengers. Taxis required to charge a regulated fare when pick-

ing up a street hail might avoid such hails if the regulated fare is be-

low the market price that they can obtain by picking up a passenger by

app. Conversely, if the regulated fare is higher than the market price,

taxis might prefer to pick up passengers through street hails, even if

consumers prefer to summon taxis by app.

New York City currently takes two different approaches to the

issue of regulating fares of taxis legally authorized to pick up by street

hail and other means. As mentioned above, yellow taxis are allowed to

pick up passengers by street hail and through pre-arrangement, includ-

ing by app. The Taxi and Limousine Commission currently sets the

fares for all trips taken by yellow taxis, a decision that it is defending

in litigation.

133

The City points out that while “[yellow t]axis make

hundreds of thousands of trips every day,” “[a]pp-arranged rides con-

stitute a tiny fraction of these rides.”

134

It argues that “if it were to

132. Thank you again to Marcel Kahan for this observation.

133. The plaintiffs in Melrose Credit Union argue that “medallion taxicabs should be

free to engage in flexible rate structures, just as e-hailing companies such as Uber are

permitted, at least for those passengers who utilize E-Hails to secure medallion taxi-

cabs, in order to ensure that these similarly situated services are not treated dispa-

rately.” Amended Complaint at 54, Melrose Credit Union v. City of New York, Case

No. 1:15-cv-09042 (S.D.N.Y. Mar. 7, 2016); see also id. at 70 (“[T]here is no justifi-

cation for the disparate regulatory treatment in fare restrictions, such that when a

passenger raises their arm or swipes a button to E-Hail a medallion taxicab, that pas-

senger must not only know the rate of fare in advance, but also that this rate of fare be

determined by Defendants; but if the same passenger decides to swipe that

smartphone button in the opposite direction and E-Hails a company such as Uber, Lyft

or Gett, those rules no longer apply.”).

134. Brief for Respondents at 56, Glyka Trans LLC v. City of New York, No. 2015-

11661 (N.Y. App. Div. Apr. 15, 2016). In support of this statement, the City indicates

that “[d]uring the two-year e-hail pilot program, only 0.39 percent of yellow taxi rides

were arranged by a smartphone app.” Id. The program to which the City is referring

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 48 17-APR-17 10:20

48 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

permit a two-tiered pricing system that allowed app-arranged rides to

be more lucrative, drivers might focus on app-arranged rides to the

detriment of street hailers.”

135

However, New York City does allow a

“two-tiered pricing system” for the green taxis that service upper

Manhattan and New York City’s outer boroughs through street hails

and prearrangement, including by app.

136

The Taxi and Limousine

Commission now “sets the fare when the [taxi] . . . is hailed” (includ-

ing e-hailed through an app), while “the base [dispatching the vehicle]

sets the fare when the fare is pre-arranged.”

137

The Commission-set

fares for green taxis are the same as the regulated fares for yellow

taxis.

138

The Commission’s practice suggests that it is not concerned

about the incentive effects of the dual pricing regime on green taxi

drivers, even while it is arguing that incentive effects justify maintain-

ing a single pricing structure for yellow taxis.

139

In summary, fare levels should not be regulated for taxis that pro-

vide only pre-arranged taxi service, whether that service is arranged

was implemented in 2013–14. Id. at 27–30; see also Black Car Assistance Program v.

City of New York, No. 100327/13, 2013 WL 1808082 (N.Y. Sup. Ct. 2013), aff’d 110

A.D. 3d 618 (N.Y. App. Div. 2013);

N.Y.C. T

AXI

& L

IMOUSINE

OMM

’

, E-H

AIL

ILOT

ROGRAM

INAL

EPORT

(Jan. 28, 2015), http://www.nyc.gov/html/tlc/down

loads/pdf/ehail_q5_report_final.pdf. In contrast, medallion owner and financier plain-

tiffs in Melrose Credit Union suggest that the share of medallion taxi trips originating

through e-hails is much larger that the City estimates—indeed, they state that “[o]n

information and belief, approximately one-third of medallion taxicab hails on average

are now being made through E-Hails—i.e., approximately 131,000 hails per day are

E-Hails and approximately 262,000 per day are made through traditional taxi street

hails.” Amended Complaint at 52–53, Melrose Credit Union, No. 1:15-cv-09042.

135. Brief for Respondents at 57, Glyka Trans LLC, No. 2015-11661.

136. Matthew Flamm, Taxi-Hailing App Takes On Uber with a New Weapon,

RAIN

’

N.Y. B

(Mar. 23, 2016), http://www.crainsnewyork.com/article/20160323

/TECHNOLOGY/160329946/verifones-taxi-hailing-app-curb-takes-on-uber-with-ad

vance-booking; Andrew J. Hawkins, Meet the Taxi Industry’s Last, Best Hope to Sur-

vive Uber Age,

RAIN

’

N.Y. B

(Aug. 27, 2015), http://www.crainsnewyork.com/

article/20150827/BLOGS04/150829899/meet-arro-the-yellow-taxi-industrys-last-best

-hope-to-compete-with-uber; Your Guide to Boro Taxis,

N.Y.C. T

AXI

& L

IMOUSINE

OMM

’

, http://www.nyc.gov/html/tlc/html/passenger/shl_passenger.shtml (last vis-

ited Jan. 22, 2017).

137.

N.Y.C. T

AXI

& L

IMOUSINE

OMM

’

supra note 22, at 2 (“TLC-Regulated Ve-

hicles”). The Curb and Arrow apps charge the regulated fares for green taxis sum-

moned by the app. See Flamm, supra note 136; Hawkins, supra note 136.

138. Your Guide to Boro Taxis,

N.Y.C. T

AXI

& L

IMOUSINE

OMM

’

, http://www.

nyc.gov/html/tlc/html/passenger/shl_passenger.shtml (last visited Jan. 22, 2017).

139. In arguing for greater pricing flexibility, medallion taxi owners and financiers in

Melrose Credit Union v. City of New York indicate that Chicago has amended its taxi

ordinances to allow “the use of ‘surge pricing’ when taxis are dispatched from an app

or digital platform.” Amended Complaint at 73, Melrose Credit Union, No. 1:15-cv-

09042. From the context, it appears that traditional Chicago taxis are given the flexi-

bility to engage in surge pricing.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 49 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 49

by app or by phone call.

140

If Uber or another app emerges as a mo-

nopoly and uses its market power to dictate high fares, then regulators

could institute fare level regulation for pre-arranged service. At this

time, regulators likely are justified in requiring the disclosure of fares

for pre-arranged service and regulating the format in which fares are

disclosed, to address information problems that consumers may face

in comparison shopping among services.

141

Regulators also should re-

visit fare level regulation of street-hailed taxi service. Requiring street

hailed taxis to charge a regulated fare inhibits them from competing

with pre-arranged service providers, such as the transportation net-

work companies, on price. Nonetheless, such fare level regulation still

may be justified given the information problems burdening passengers

hailing cabs on the street or at cab stands, the potential for fare dis-

crimination to emerge on prohibited grounds if fares are not standard-

ized, and the externalities that individualized fare negotiations

between riders and drivers might generate for other users of street

space. However, it might be desirable to allow taxis that pick up street

hails to charge unregulated fares when they are picking up passengers

in other ways, for example by app or phone call, to enable these taxis

to compete on price with the providers of only prearranged taxi ser-

vice when the taxis are providing similar services.

Consumer Safety Regulation

Cities regulate the attributes of taxi drivers, for example by re-

quiring that they are a certain age; that they undergo training, drug

140. Posner may acknowledge that it is premature to regulate Uber fares now. See

Posner, supra note 10; see also Rauch & Schleicher, supra note 10, at 937 n.204

(“The firms do not appear to have any substantial market power yet, and while there

are some economies of scale and network effects, two-sided markets do not, as a

general matter, regularly result in monopolies.”).

141. In April 2016, the New York City Council passed legislation requiring that

“black car base[s],” “luxury limousine base[s],” and “dispatch service provider[s]”

“allow prospective passengers to request a fare quote prior to booking transportation”;

the legislation also prohibits the passenger from being charged “more than 120 per-

cent of the price quoted unless such passenger takes any action to alter the estimated

route.” N.Y.C., N.Y., Law 2016/049 (Apr. 21, 2016). The legislation built on existing

Taxi and Limousine Commission rules, and followed similar legal requirements as

those in Chicago. See 35 RCNY §§ 77-15, 77-20(a), (d);

EPORT OF THE

UMAN

ERVICES

IVISION

supra note

, at 7; Joshi Testimony, supra note 66, at 2–4. As

the chair of the Taxi and Limousine Commission noted in testimony on the bill, legis-

lating that fare quotes be binding may not have been necessary given “that market-

driven customer service concerns [should] . . . largely prevent companies from charg-

ing above a fare estimate.” Joshi Testimony, supra note 66, at 3. As previously noted,

Uber now provides upfront fares, and it is possible to compare online the prices that

Uber and Lyft charge for the same trip. See supra note 115.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 50 17-APR-17 10:20

50 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

testing and criminal background checks; and that they are certified as

fit for the job of taxi-driving by a physician.

142

Cities also require that

taxis are insured in case of accidents,

143

and that the vehicles meet

certain specifications.

144

Regulating attributes of drivers and vehicles

for health and safety is justified based on asymmetric information.

Taxi riders presumably want to ride in taxis that are well maintained,

driven by safe drivers, and insured in case of accidents. But it is costly

for riders to verify whether drivers and vehicles have these attributes

(information which likely is known to the driver).

145

Regulating some

driver and vehicle attributes also may be justified based on externali-

ties. Unsafe drivers and improperly maintained vehicles can endanger

third parties, such as pedestrians and other drivers injured in

accidents.

146

There is a strong case, justified by asymmetric information and

potential externalities, that the same minimum standards that apply to

traditional taxis and their drivers should apply to the drivers of app-

dispatched taxis and their vehicles.

147

As with traditional taxis, con-

sumers presumably want to know that Uber drivers are safe drivers

and do not have criminal backgrounds, that their vehicles are mechani-

142. See

35 RCNY § 54-04

; Ill. Transp. Trade Ass’n v. City of Chicago, 134 F.

Supp. 3d 1108, 1110–11 (N.D. Ill. 2015), aff’d in part & rev’d in part, 839 F.3d 594

(7th Cir. 2016) (“[T]axis must undergo Chicago Police Department background check

. . . [and] submit annual drug tests conducted by authorities approved by the Commis-

sioner . . . . Taxi drivers must have a chauffeur license, requiring a background check

by the Chicago Police Department, training, safety courses and continuing

education.”).

143. See 35 RCNY § 58-13 (detailing the insurance requirements for New York City

medallion taxicabs); Ill. Transp. Trade Ass’n, 134 F. Supp. 3d at 1110 (noting the

insurance requirements for Chicago taxicabs). See generally

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

supra note 7, at 114–15 (describing local and

state insurance requirements for traditional taxi industry); Darb´era, supra note 10, at

144. Ill. Transp. Trade Ass’n, 134 F. Supp. 3d at 1111 (explaining that Chicago

“[t]axis must meet certain vehicle requirements under the Ordinance, including age

and condition”); see also

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

supra note 7, at 104–05 (describing vehicle inspection requirements).

145. See Wyman, supra note 21, at 152–53 (citing sources);

OMPETITION

UREAU

supra note 10 (“[P]assengers are not well-placed to judge the mechanical safety of a

vehicle or rate the quality of insurance.”).

146. See Edelman & Geradin, supra note 10, at 311; CPUC, supra note 26, at 3.

Taxi passengers—and taxi drivers—also may have cognitive biases that lead them to

underestimate the risks of unsafe vehicles and driving and to undervalue safety.

Edelman & Geradin, supra note 10, at 317–18.

147. See

OMPETITION

UREAU

, supra note 10. But see Edelman & Geradin, supra

note 10, at 305, 317–18 (suggesting that there might be a case, on cost-benefit

grounds, for less stringent safety standards for drivers driving commercially for only a

minimal amount of time).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 51 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 51

cally sound, and that there is insurance in the event of an accident.

Riders still will have difficulty verifying these attributes, even though

Uber allows passengers to rate drivers, because riders cannot easily

check the conditions of cars and driver qualifications. In addition, peo-

ple on the street, who are not in the taxi, should not have to bear the

cost of dangerous drivers or vehicles. Not surprisingly, a considerable

amount of the initial regulation introduced by jurisdictions like Cali-

fornia that have been open to the taxi apps is focused on establishing

minimum standards for drivers, vehicles, and insurance in the case of

accidents.

148

The insurance requirements for app-dispatched taxis

have proven to be an especially contentious issue, as Uber and other

app companies have been accused of providing insurance with gaps in

coverage. Legislative interventions in many states have required com-

panies like Uber and Lyft to increase the insurance coverage that they

provide.

149

Whether transportation network company drivers should

148. See, e.g., Ill. Transp. Trade Ass’n, 134 F. Supp. 3d at 1115 (describing regula-

tory requirements imposed on “Transportation Network Providers” as including back-

ground checks, drug tests, vehicle inspection and insurance requirements, but

emphasizing that the requirements imposed on TNPs are much less stringent than the

requirements imposed on traditional taxis); CPUC, supra note 26; FTC Workshop,

supra note 10, at 94–96 (remarks of Catherine Sandoval).

149. FTC Workshop, supra note 10, at 130 (remarks of Ashwini Chhabra) (“[N]ow

there’s model legislation that’s being implemented in several states that puts the onus

on Uber and any other TNC to carry coverage for all instances where the driver hasn’t

purchased his or her own TNC-specific policy. . . . [The insurance industry is] devel-

oping insurance products specifically for TNC drivers.”).

For a lucid discussion of the insurance issues raised by the transportation net-

work companies, the innovations that these companies have prompted in insurance,

and a comparison of the insurance requirements for the traditional taxi industry and

the new entrants, see

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

supra note 7, at 114–20; see also

ATIONAL

’

NOF

. C

OMM

’

, T

RANSPORTA-

TION

ETWORK

OMPANY

NSURANCE

RINCIPLES FOR

EGISLATORS AND

EGULA-

TORS

(2015), http://www.naic.org/documents/committees_c_sharing_econ_wg

_exposure_adopted_tnc_white_paper_150331.pdf.

The insurance coverage of taxi app drivers varies depending on the jurisdiction’s

local legal requirements. In New York City, for-hire vehicle owners, which include

Uber drivers, must carry a certain amount of liability insurance under Taxi and Lim-

ousine Commission rules. See 35 RCNY § 59A-12(c)(1) (requiring “$200,000 per

person, payable for those expenses specified in paragraphs 1, 2 and 3 of subdivision a

of section 5102 of the New York State Insurance Law” and “$100,000 minimum

liability and $300,000 maximum liability for bodily injury and death, as those terms

are described and defined in section 370(1) of the Vehicle and Traffic Law”). These

insurance coverage requirements, which apply twenty-four hours a day, seven days a

week, mirror the amounts for New York City medallion taxis. See 35 RCNY § 58-

13(a)(1), (d). These New York City rules make Uber’s generic insurance coverage

irrelevant in New York City. See Certificates of Insurance—U.S. Ridesharing,

BER

(Jan. 11, 2015), http://newsroom.uber.com/2015/01/certificates-of-insurance-u-s-

ridesharing/.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 52 17-APR-17 10:20

52 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

be fingerprinted also has been a flashpoint.

150

Notwithstanding the market failure justifications for consumer

safety regulation of traditional and app-based taxis, the existing regu-

latory regime for protecting passengers and third parties needs to be

rethought along two dimensions. The ambit of consumer safety regu-

lation needs to be carefully targeted to address only market failures.

The stringency of consumer safety regulation also needs to be cali-

brated to ensure that it is set at levels that are cost-beneficial.

As part of ensuring that consumer safety regulation targets mar-

ket failures, there is scope for narrowing the ambit of such regula-

tion.

151

The incumbent taxi industry likely is burdened with

150. Writing generally, and not referring to practices in specific jurisdictions such as

New York City, the Transportation Research Board’s Committee for Review of Inno-

vative Urban Mobility Services explained that traditional taxi drivers may be subject

to fingerprint checks as well as other criminal background checks. On the other hand,

“the background checks” done for Uber and Lyft “do not include fingerprinting but do

include checks of government criminal records in drivers’ counties of residence.”

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

supra note 7, at 101.

The Committee did not locate “any careful empirical studies on the effectiveness” of

the different methods of vetting drivers in protecting “passenger safety.” Id. at 104;

see also id. at 108–09.

In Austin, TX, in May 2016, a majority of voters voted to maintain municipal

requirements that Uber and Lyft drivers be fingerprinted, like taxi drivers, even

though Uber and Lyft threatened to leave Austin if the fingerprinting requirement was

maintained. The referendum result is evidence of public support for similar consumer

protection measures for traditional taxi drivers and Uber and Lyft drivers. See Mike

McPhate, Uber and Lyft End Rides in Austin to Protest Fingerprint Background

Checks, N.Y.

IMES

(May 9, 2016), http://www.nytimes.com/2016/05/10/technology/

uber-and-lyft-stop-rides-in-austin-to-protest-fingerprint-background-checks.html;

Richard Parker, Opinion, How Austin Beat Uber,

N.Y. T

IMES

(May 12, 2016), http://

www.nytimes.com/2016/05/12/opinion/how-austin-beat-uber.html. Uber and Lyft

withdrew from Austin after the referendum, but Uber has expressed interest in re-

turning to Austin. Alex Samuels, Uber Wants to Return to Austin, Spokesman Says,

. T

RIB

(Nov. 17, 2016), https://www.texastribune.org/2016/11/17/uber-spokes-

man-austin-says-ride-hailing-app-may-re/. In New York City, transportation network

company drivers, like other taxi and for-hire drivers, “are fingerprinted for . . . crimi-

nal background check[s].” Karizza Sanchez, How Safe Is Uber in New York City?,

OMPLEX

(May 20, 2016), http://www.complex.com/life/2016/05/uber-new-york-city-

safety.

For a report recommending the fingerprinting of “all for-hire ground transporta-

tion drivers,” including taxi drivers and transportation network company drivers, us-

ing “biometric fingerprints,” see

ATTHEW

W. D

AUS

& P

ASQUALINO

USSO

, O

TANDARD FOR

: C

RIMINAL

ACKGROUND

HECKS FOR

AXICAB

, F

-H

IRE

AND

RANSPORTATION

ETWORK

OMPANY

(TNC) 4 (2015). The report advocates finger-

printing, not just name checks, because “[t]he fingerprint . . . is the one true identifier

that cannot be stolen or falsified by the applicant.” Id. at 11. However, the report also

states that “[i]t is important to run both name and fingerprint checks for maximum

reliability of the background check record.” Id. at 18–19.

151. See Alex Tabarrok & Tyler Cowen, The End of Asymmetric Information,

ATO

NBOUND

(Apr. 16, 2015), https://www.cato-unbound.org/2015/04/06/alex-tabarrok-

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 53 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 53

regulations justified in the name of consumer safety, which are unre-

lated to information asymmetries or externalities. Unjustified regula-

tions should be eliminated for the incumbent industry and not applied

to the new entrants. Consider, for example, the New York Taxi and

Limousine Commission’s ongoing “Taxi of Tomorrow” initiative to

require that most yellow taxis in the future use a specific vehicle, a

Nissan NV200 designed specifically to be a New York taxi. The Com-

mission defends the decision to mandate that most taxis going forward

use this vehicle as promoting “safety, health and comfort.”

152

But the

decision seems at least partly rooted in a governmental desire to estab-

lish a new “iconic” taxi that will become as famous as the old Checker

cab that many people still associate with New York City.

153

The ambit

of consumer safety regulation also might be reduced because require-

ments that were once established to protect safety now no longer are

needed due to technological advances that address the informational

asymmetries that formerly necessitated regulation.

154

tyler-cowen/end-asymmetric-information; FTC Workshop, supra note 10, at 151–55

(remarks of Adam Thierer);

HRISTOPHER

OOPMAN ET AL

ERCATUS

., T

HARING

CONOMY AND

ONSUMER

ROTECTION

EGULATION

: T

ASE FOR

OL-

ICY

HANGE

(2014), https://www.mercatus.org/system/files/Koopman-Sharing-

Economy.pdf.

152. Emma G. Fitzsimmons, Top State Court Backs New York City’s Taxi of To-

morrow,

N.Y. T

IMES

(June 25, 2015), http://www.nytimes.com/2015/06/26/nyregion/

top-state-court-backs-new-york-citys-taxi-of-tomorrow.html.

153. Greater N.Y. Taxi Ass’n v. N.Y. City Taxi & Limousine Comm’n, 25 N.Y.3d.

600, 602 (2015) (upholding the Commission’s authority to prescribe use of single

vehicle).

154. See Cohen & Sundararajan, supra note 10 (discussing potential for self-regula-

tion to deal with asymmetric information and externalities in the sharing economy);

FTC Workshop, supra note 10, at 85, 87–88, 158–59, 166–68 (remarks of Arun Sun-

dararajan). Sundararajan argues that platforms “seem like they would be quite effec-

tive” in addressing market failures due to information asymmetry, but not

externalities, which a third party may need to address. FTC Workshop, supra note 10,

at 87–88 (remarks of Arun Sundararajan). He notes, however, that self-regulation is

not a “panacea.” Id. at 166 (remarks of Arun Sundararajan).

In April 2016, the New York City Council passed legislation eliminating the

requirement that black cars be retired once they reach a certain age. Under the new

legislation, black cars can continue to be used if they pass inspections. N.Y.C., N.Y.,

Law 2016/050 (Apr. 21, 2016). In 2008, the New York Taxi and Limousine Commis-

sion had introduced rules requiring the retirement of the black cars “with the purpose

of improving vehicle quality and service” but already had relaxed these rules consid-

erably in 2015 because the increasingly competitive market for black car services now

provided incentives to retire cars in line with consumer preferences.

EPORT OF THE

UMAN

ERVICES

IVISION

supra note 25, at 8–9 (Apr. 7, 2016). As the Chair of the

Commission told the Council’s Committee on Transportation, the upshot of the Com-

mission’s rule changes in 2015 is that the legislation will “eliminate the retirement

requirement for the approximately 28% of black cars in service today that are Model

Year 2012 or earlier.” Joshi Testimony, supra note 66, at 2.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 54 17-APR-17 10:20

54 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

To some extent, the ambit of consumer safety regulation may

need to be expanded to address new issues, such as privacy protection

for consumer data. Service providers have been collecting credit card

and trip data electronically for traditional taxis since the 2000s in New

York City, but this data collection historically did not enable the ser-

vice providers or taxi drivers to know the identity of taxi passengers in

real time.

155

Uber collects location and financial information about its

passengers that enables it to identity its passengers and their location

in real time, and Uber has been criticized for misusing consumer

data.

156

Regulations to protect passenger privacy might be grounded in

information problems such as imperfect information (because riders

and possibly the app companies themselves may not foresee uses of

the data that they are collecting),

157

or asymmetric information (be-

cause the apps have better information about how they use consumer

data than consumers, perhaps because it is too costly for riders to un-

derstand firm privacy policies),

158

or the idea that privacy is a public

good that is likely to be under-provided by the market.

159

155. See 35 R.C.N.Y. § 75 (“Rules for Authorization of Taxicab Technology Service

Providers”); Alexandre v. N.Y.C. Taxi & Limousine Comm’n, No. 07-cv-8175, 2007

WL 2826952, at *1 (S.D.N.Y. Sept. 28, 2007); Taxicab Passenger Enhancements

Project Archive,

N.Y.C. T

AXI

& L

IMOUSINE

OMM

’

, http://www.nyc.gov/html/tlc/

html/industry/taxicab_serv_enh_archive.shtml (last visited Jan. 22, 2017).

156. Electronic Privacy Info. Ctr., Complaint, Request for Investigation, Injunction,

and Other Relief in the Matter of Uber Technologies, Inc. (Fed. Trade Comm’n June

22, 2015), https://epic.org/privacy/internet/ftc/uber/Complaint.pdf; see also Geradin,

supra note 8, at 11; Natasha Singer & Mike Isaac, Uber Data Collection Changes

Should Be Barred, Privacy Group Urges,

N.Y. T

IMES

(June 22, 2015), http://

www.nytimes.com/2015/06/23/technology/uber-data-collection-changes-should-be-

barred-privacy-group-urges.html.

Concerns also have been raised about the use of data collected from the tradi-

tional taxi industry. For example, the New York Taxi and Limousine Commission has

been criticized for releasing data about taxi drivers that could be easily de-

anonymized. See Paul Ducklin, New York City Made a Hash of Taxi Driver Data

Disclosure,

AKED

ECURITY

(June 24, 2014), https://nakedsecurity.sophos.com/

2014/06/24/new-york-city-makes-a-hash-of-taxi-driver-data-disclosure/.

157. Megan Cox asks: “[I]f Uber or Lyft decided to offer grocery delivery service of

the future, should it be able to use data collected from users at its transportation ser-

vice?” FTC Workshop, supra note 10, at 171 (remarks of Megan Cox).

158. Maurice Strucke refers to the related potential that a platform could “use[ ] its

market power to extract more data than what consumers might otherwise want.” FTC

Workshop, supra note 10, at 162 (remarks of Maurice Strucke).

159. Glen Weyl suggests that privacy might be a public good. FTC Workshop, supra

note 10, at 28 (remarks of Glen Weyl).

For differing views on whether regulation is necessary to protect consumer pri-

vacy, see Rogers, supra note 10, at 94 (arguing that “privacy issues will largely self-

correct” because “[u]nlike Facebook and Google, sale or exploitation of user data

does not seem to be a major revenue source for Uber”); FTC Workshop, supra note

10, at 171–72 (remarks of Sofia Ranchord´as) (advocating platforms be treated as fidu-

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 55 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 55

Second, the level of consumer protection likely needs to be cali-

brated to reflect the costs and benefits of safety regulation. There is a

lack of empirical evidence about the costs and benefits of safety regu-

lations applied to the taxi industry, both the traditional industry and

the new transportation network companies.

160

But some of the con-

sumer safety regulations currently applicable to the incumbent taxi in-

dustry may be excessively stringent, given the benefits that they yield

relative to their costs. For example, some of the requirements that

taxis install certain forms of technology may be excessively stringent,

given the benefits of the technology, or the possibility that newer tech-

nologies could yield the same benefits at lower cost.

161

It is also likely

ciaries for information, drawing on ideas of Jack Balkin); and FTC Workshop, supra

note 10, at 172–73 (remarks of Adam Thierer) (underscoring the “tension” between

protecting privacy and addressing information asymmetries, and arguing against im-

posing fiduciary obligations on platforms, because they need access to information to

address information asymmetries). For a survey of scholarship on the economics of

information, highlighting that there are costs, and benefits, to regulating for the pro-

tection of privacy, see Alessandro Acquisti et al., The Economics of Privacy, 54

CON

. L

ITERATURE

442 (2016).

In April 2016, the New York City Council passed legislation requiring “[a]ll

entities licensed by the [Taxi and Limousine] [C]ommission, or authorized by the

commission to provide services regulated by the commission, that collect or maintain

passenger personal information or passenger geolocation information” to “file with

the commission an information security and use of personal information policy” that

includes certain minimum requirements. N.Y.C., N.Y., Law 2016/043 (Apr. 21,

2016). As the Taxi and Limousine Commission chair noted in testimony on the bill,

“the Council [was] not writing on a blank slate. TLC licensees are already subject to a

complex set of federal and state laws, as well as TLC rules, governing the use of

personal and credit card information.” Joshi Testimony, supra note 66, at 8.

160. Edelman & Geradin, supra note 10, at 311–12;

OMM

FOR

EVIEW OF

NNOVA-

TIVE

RBAN

OBILITY

ERVS

, supra note 7, at 104.

161. See, e.g., Amended Complaint at 35, Melrose Credit Union v. City of New

York, No. 1:15-cv-09042 (S.D.N.Y. Nov. 17, 2015) (stating that “TLC [Taxi and

Limousine Commission] specifications for taximeters, partitions, in-vehicle camera

systems, credential holders, and ‘T-PEP’ taxicab technology” are among the rules that

taxis, but not for-hire vehicles such as Uber vehicles, must follow). However, some of

these tools, such as partitions and in-vehicle camera systems, may be more necessary

to protect drivers in traditional taxis than they would be in transportation network

company vehicles, because the identity of passengers in these vehicles is known to the

TNC, unlike the identity of passengers in traditional taxis if the passengers are picked

up through street hails.

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

supra note 7, at 98–100. As the Committee reports, “[d]riving a taxi . . . is quite

dangerous relative to other occupations” because drivers are allowing “strangers into

their vehicles.”

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

supra

note 7, at 99; see also Complaint at 10, Bos. Taxi Owners Ass’n v. Baker, No. 1:16-

cv-11922-NMG (D. Mass. Jan. 24, 2017), 2017 WL 354010 (case dismissed) (noting

that partitions “not only protect drivers from assault and theft but also protect the

driver from impaired passengers that may interfere with the driver’s ability to operate

the taxi”); Second Amended Complaint at 11, Bos. Taxi Owners Ass’n v. City of

Boston, No. 1:15-cv-10100-NMG (D. Mass. Dec. 21, 2016), 2016 WL 7410777 (case

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 56 17-APR-17 10:20

56 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

that some existing consumer safety regulations applicable to the in-

cumbent taxi industry are insufficiently stringent in cost-benefit terms,

and that these regulations should be made more stringent for both in-

cumbents and new app-entrants. In particular, the minimum insurance

requirements applicable to traditional taxis in New York City and

other cities arguably are insufficiently protective in the event of a car

accident.

162

Notably, Uber’s insurance coverage for passengers when

a vehicle is “engaged in a prearranged ride” is more protective of pas-

senger well-being than existing regulatory requirements for taxicabs in

some cities.

163

The difference suggests the need to review the mini-

mum insurance requirements that should apply to all forms of taxis,

based on an analysis of the costs and benefits of different levels of

insurance coverage.

An interesting question is whether the means of implementing

and enforcing consumer safety regulations should be changed, given

the emergence of large scale providers of taxi services such as Uber

that rely on consumer rating systems and that have an incentive to

protect their brands.

164

For example, cities might consider allowing

dismissed) (same). The New York Taxi and Limousine Commission recently changed

its rules to allow the owners of all medallion taxis to install in-vehicle camera systems

instead of partitions, but black cars (which include many Uber cars) are not required

to install either device.

N.Y.C. T

AXI

& L

IMOUSINE

OMM

’

, Notice of Promulgation

of Rules on In-Vehicle Camera Systems and 496 Waivers (2016), http://

www.nyc.gov/html/tlc/downloads/pdf/newly_passed_rule_ivcs_496_waiver.pdf.

162. 35 RCNY § 58-13(a)(1), (d) (describing the insurance requirements for New

York City medallion taxis, which are the same as the insurance requirements for New

York City for-hire vehicles, which include Uber vehicles (35 RCNY § 59A-

12(c)(1))); see also

OMM

. F

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

supra note 7, at 119–120 (referring to the potential for innovation in insurance cover-

age for traditional taxis).

163. FTC Workshop, supra note 10, at 111 (remarks of Ashwini Chhabra); see also

Edelman & Geradin, supra note 10, at 313; Christian Denmon, Ride Sharing vs.

Traditional Taxis: How Do Injury Insurance Claims Compare?,

UFFINGTON

OST

(May 7, 2014), http://www.huffingtonpost.com/christian-denmon/ride-sharing-vs-tra-

dition_b_5273964.html (“While there are still concerns over Lyft and Uber’s personal

property coverage for the driver, the bottom line is that the passenger in a Lyft or

Uber vehicle often has more coverage available than if he or she was injured in an

identical situation but in a taxi cab. In some states like Florida, that coverage can be

more than 400 percent greater than the taxi cab’s comparable coverage.”).

164. Apps like Uber enable consumers to rate their drivers, which they presumably

do based on features that are observable to consumers, such as the politeness of the

driver and the cleanliness of the vehicle. Uber removes drivers from its system if their

ratings fall below a certain level. See Berwick v. Uber Techs., No. 11-46739 EK (Cal.

Labor Comm’r’s Office June 3, 2015); James Cook, Uber’s Internal Charts Show

How Their Driver-Rating System Actually Works,

. I

NSIDER

(Feb. 11, 2015), http:/

/www.businessinsider.com/leaked-charts-show-how-ubers-driver-rating-system-

works-2015-2; see also Edelman & Geradin, supra note 10, at 315–17 (analyzing

ratings systems).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 57 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 57

taxi services meeting a certain threshold of size to ensure compliance

with driver and vehicle requirements established by regulators, subject

to audits and spot checks by regulators.

165

Large virtual fleet manag-

ers like Uber are more likely to have the resources to establish inspec-

tion facilities and systems, and the reputational incentive to do so, than

the historically fragmented taxi industry.

166

A drawback of allowing

the large operators to self-regulate, albeit subject to oversight from

regulators, is that self-regulation may provide the large operators like

Uber with a cost advantage over smaller players. Thus allowing the

larger operators to self-regulate might be in tension with guarding

against the emergence of monopoly, a new function I discuss below

for taxi regulators.

Worker Protections

Many taxi drivers currently are classified as independent contrac-

tors for many purposes.

167

Nonetheless, legislators and regulators have

in some instances adopted limited measures to protect the health,

safety and economic interests of taxi drivers. Medallion owners in

some jurisdictions are required to purchase workers’ compensation in-

surance for taxi drivers to whom they lease medallions, even though

the drivers are otherwise considered independent contractors.

168

Cities

165. See, e.g., CPUC, supra note 26, at 26–33, 40–44 (imposing safety requirements

on TNCs); FTC Workshop, supra note 10, at 159 (remarks of Arun Sundararajan)

(advocating “delegated regulation by data” under which platforms enforce standards);

FTC Workshop, supra note 10, at 126–27 (remarks of Matthew Daus) (referring to

the Federal Motor Carrier Safety Administration approach for interstate trucking and

other interstate transportation industries as an example of self-regulation subject to

enforcement).

166. See Cohen & Sundararajan, supra note 10 (discussing the potential for self-

regulation to deal with asymmetric information and externalities in the sharing

economy).

167. See supra note 19.

168. See Yellow Cab Coop., Inc. v. Workers’ Comp. Appeals Bd., 226 Cal. App. 3d

1288, 1296 (1991) (stating that ten state courts have held that lessee cabdrivers “are or

may be” employees for workers’ compensation, while four state courts have held that

they are not);

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra

note 7, at 87 (“In several major jurisdictions in which taxi drivers are classified as

independent contractors—including the states of Washington, Colorado, and New

York, plus San Francisco and Chicago, among others—they nonetheless receive

workers’ compensation.”) (citation omitted).

In New York City and Chicago, legislation and regulations require medallion

owners to purchase workers’ compensation insurance. See

N.Y. W

ORKERS

’ C

OMP

§§ 2(3), 2(4), 18-c (McKinney 2014); 35 RCNY §§ 58-03(y), 58-14(a); Ill.

Transp. Trade Ass’n v. City of Chicago, 134 F. Supp. 3d 1108, 1110 (N.D. Ill. 2015),

aff’d in part & rev’d in part, 839 F.3d 594 (7th Cir. 2016) (noting that taxis must have

“workers’ compensation insurance, typically costing well over $4,000 per year per

taxi for coverage”); FAQs – Medallion Owners & Agents,

N.Y.C. T

AXI

& L

IMOUSINE

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 58 17-APR-17 10:20

58 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

such as New York also cap the rates at which medallion owners can

lease their medallions to taxi drivers.

169

Lease caps likely have less

effect now because taxi drivers are switching to Uber and away from

driving medallion taxis,

170

but when the demand from drivers to lease

medallions routinely exceeded the supply of medallions, the lease caps

likely often set lease prices. In doing so, the lease caps allocated fare

revenues between medallion owners and taxi drivers, and effectively

reserved some minimum share of the revenues for drivers, much like a

collective agreement would have done if the drivers bargained collec-

tively. The lease caps, which New York City introduced in 1996,

171

functioned this way because they were combined with regulated taxi

fare levels: through fare level regulation, the Taxi and Limousine

Commission set the fares that drivers collected from passengers, and

through the lease caps the Commission ensured that drivers retained

some share of the fares, by preventing medallion owners from charg-

ing drivers all of the fares for the medallion.

172

As a historical matter, the health, safety and economic protec-

tions for taxi drivers were likely introduced due to concerns about

inequality of bargaining power between taxi drivers and medallion

owners.

173

This inequality of bargaining power is likely related to the

OMM

’

http://www.nyc.gov/html/tlc/html/faq/faq_medallion_owners_and_agents.

shtml (last visited Nov. 13, 2016) (“You need Workers’ Compensation if you lease

your medallion out to a driver.”); Workers’ Compensation Coverage: Taxi Cabs—

Most Taxi Cab Operators Are Considered Employees,

N.Y. S

. W

ORKERS

’ C

OMPEN-

SATION

http://www.wcb.ny.gov/content/main/onthejob/CoverageSituations/taxi-

Cabs.jsp (last visited Nov. 12, 2016). New York City medallion owners also must

provide drivers leasing medallions with disability benefits. See Ahmed v. City of New

York, 129 A.D.3d 435, 437 (N.Y. App. Div. 2015).

In contrast, Boston taxi drivers are not covered by workers’ compensation (or

unemployment insurance or income tax withholding requirements). Sebago v. Bos.

Cab Dispatch, Inc., 28 N.E.3d 1139, 1153 (Mass. 2015).

169. 35 RCNY § 58-21(c) (setting caps on standard lease rates). In Boston, police

department rules established “flat lease rate[s]” for drivers leasing medallions and

taxis. Sebago, 28 N.E.3d at 1145.

170. New York medallion owners seem to be reducing lease rates below the lease

caps to attract drivers, although there may still be lucrative shifts (such as Fridays)

when the lease caps set leasing rates. See Amended Complaint at 13, 33, 61, Melrose

Credit Union v. City of New York, No. 1:15-cv-09042 (S.D.N.Y. Mar. 7, 2016).

171. See

N.Y.C. T

AXI

& L

IMOUSINE

OMM

’

, Notice of Final Rule Promulgation

Adjusting the Maximum Lease Rates a Medallion Owner May Charge a Licensed

Taxicab for Various Shifts (2004), http://www.nyc.gov/html/tlc/downloads/pdf/

lease_rates.pdf.

172. See 35 RNCY § 58-21(c)(7) (codifying that lease caps do not apply if a collec-

tive agreement governs).

173. For indications that lease caps were intended to protect drivers against eco-

nomic exploitation by medallion owners, see, for example,

N.Y.C. T

AXI

& L

IMOUSINE

OMM

’

, Notice of Final Rule Promulgation Adjusting the Maximum Lease Rates a

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 59 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 59

high costs of collective action that drivers face. Taxi drivers are not

currently easy to organize in large cities like New York, where there

are many taxi drivers, they work in individual cars that are constantly

in motion, and there is the legal obstacle to unionizing if drivers are

independent contractors.

174

Since many other low-income workers

face prohibitively high costs of collective action, we might ask if there

are special reasons for protecting the health, safety and economic in-

terests of taxi drivers in particular.

Some of the existing protections for taxi drivers might be justi-

fied as a response to externalities, although I am unaware of empirical

evidence tying protections for drivers to increased public safety. Theo-

Medallion Owner May Charge a Licensed Taxicab for Various Shifts (2004), http://

www.nyc.gov/html/tlc/downloads/pdf/lease_rates.pdf (“[A] substantial increase in

lease rates would have an adverse impact upon a driver’s income.”); Press Release,

N.Y. State Office of the Attorney Gen., A.G. Schneiderman and TLC Secure First-of-

Its-Kind Agreement Protecting Rights of Taxicab Drivers (Dec. 19, 2013), http://

www.ag.ny.gov/press-release/ag-schneiderman-and-tlc-secure-first-its-kind-agree

ment-protecting-rights-taxicab-0; Richard Perez-Pena, Cab Fleet Coalition Seeks In-

crease of 27% in Fares,

N.Y. T

IMES

, Sept. 20, 1995, at B4; Richard Perez-Pena, Taxi

Board Votes Age Limit for Cabs, and a Rise in Fares,

N.Y. T

IMES

, Jan. 26, 1996.

174. The National Labor Relations Act (NLRA) does not apply to independent con-

tractors. 29 U.S.C. § 152(3) (1947); see also NLRB v. Friendly Cab, 512 F.3d 1090

(9th Cir. 2008) (upholding an NLRB order that Oakland taxi drivers are covered by

the NLRA); AAA Transportation/Yellow Cab, No. 28-RC-106979 (N.L.R.B. Oct. 23,

2015) (finding that taxi drivers in Tuscon are employees, not independent contractors,

who can unionize);

LLEN

ICHNER

, A

. B

’

NDEPENDENT

ONTRACTORS

AND

ONTINGENT

ORKERS

NDER THE

ATIONAL

ABOR

ELATIONS

3 (2010),

http://www.americanbar.org/content/dam/aba/administrative/labor_law/meetings/2010

/annualconference/018.authcheckdam.pdf (discussing how the NLRB approaches the

question of whether taxi drivers are employees or independent contractors); Jon Wein-

berg, NLRB Finds Tucson Taxi Drivers Employees, and Uber Drivers Could Be Next,

ABOR

(Dec. 14, 2015), https://onlabor.org/2015/12/14/nlrb-finds-tucson-taxi-dri

vers-employees-and-uber-drivers-could-be-next/ (analyzing the NLRB decision). For

a brief discussion of the difficulties of unionizing taxicab drivers, see

ILBERT

AMUELS

, supra note 17, at 94–96.

There is a New York Taxi Workers Alliance, which describes itself as “a mem-

bership based non-profit union fighting for the rights of NYC’s 50,000+ taxi drivers.”

N.Y. T

AXI

ORKERS

, http://www.nytwa.org/ (last visited Nov. 12, 2016). His-

torically, before leasing of medallions became widespread, many taxi drivers were

employees of large fleets and drivers were unionized in New York from “the mid-

1960s until the late 1970s or the early 1980s.” Wyman, supra note 21, at 158.

The International Brotherhood of Electrical Workers has filed a petition before

the National Labor Relations Board to represent Uber drivers at LaGuardia airport. To

succeed, the union would have to establish that Uber drivers are employees, not

independent contractors, under the NLRA. Conor Skeling, Uber Files to Represent

Uber Drivers Serving LaGuardia,

OLITICO

(Feb. 3, 2016), http://www.politico.com/

states/new-york/city-hall/story/2016/02/union-files-to-represent-uber-drivers-serving-

laguardia-030906; see also Jon Weinberg, Gig News: Union Files NLRB Petition to

Represent Uber Drivers in New York,

ABOR

(Feb. 3, 2016), http://onlabor.org/

2016/02/03/gig-news-union-files-nlrb-petition-to-represent-uber-drivers-in-new-york/.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 60 17-APR-17 10:20

60 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

retically, protecting drivers from medallion owners through lease caps

might have discouraged dangerous driving by taxi drivers. In the ab-

sence of lease caps, drivers might have been more inclined to drive

extended hours or compete more aggressively with each other for

fares, resulting perhaps in more accidents involving taxis.

175

Injured

taxi drivers without workers’ compensation also might be more likely

to return to taxi driving before fully recovering, thereby jeopardizing

public safety.

Like many players in the incumbent taxi industry, Uber and Lyft

treat their drivers as independent contractors. The companies’ classifi-

cation of their drivers as independent contractors is controversial.

Worker protection advocates have criticized Uber and Lyft for classi-

fying their drivers in this way,

176

transforming Uber in particular into

a symbol of the rise of the “contingent workforce,”

177

populated by

175. See, e.g., Defendants’ Memorandum of Law in Support of Defendants’ Motion

to Dismiss the Amended Complaint at 32–33, Melrose Credit Union, No. 1:15-cv-

09042 (“Lease caps were enacted as a means of protecting the riding public, as well as

taxi drivers from untoward leasing practices by medallion owners and agents. Because

drivers previously working on commission-based models would be compelled to work

long hours and drive aggressively to maximize the number of trips in any given shift,

TLC passed the lease cap rule to improve taxi service.”).

The recent Seattle ordinance creating a framework for for-hire vehicle drivers to

bargain collectively draws a link between working conditions for drivers and public

safety, stating: “Establishing the drivers’ contractual terms through a collective nego-

tiation process will also help ensure that the compensation drivers receive for their

services is sufficient to alleviate undue financial pressure to provide transportation in

an unsafe manner (such as by working longer hours than is safe, skipping needed

breaks, or operating vehicles at unsafe speeds in order to maximize the number of

trips completed) or to ignore maintenance necessary to the safe and reliable operation

of their vehicles.” Seattle, Wash., Ordinance 124968, Version 4 § 1.I.2

(

Dec. 23,

2015) (

relating to taxicab, Transportation Network Company, and for-hire vehicle

drivers).

It should be noted that there is empirical evidence that taxi drivers are safer

drivers than “noncommercial drivers.” Edelman & Geradin, supra note 10, at 310

(citing data from Schaller Consulting); see also

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

supra note 7, at 116 (same). Also, there may be other ways

of promoting driver—and passenger and pedestrian—safety than through workers’

compensation and lease caps. For example, New York City regulates the maximum

number of hours that medallion taxi and for-hire drivers can drive, to reduce the likeli-

hood of accidents due to driver fatigue. N.

Y.C. T

AXI

& L

IMOUSINE

OMM

’

, Notice

of Promulgation of Rules to Reduce the Risks of Fatigued Driving by Licensed Driv-

ers (2016), http://www.nyc.gov/html/tlc/downloads/pdf/newly_passed_rule_fatigued_

driving.pdf.

176. See, e.g.,

’

ROJECT

IGHTS ON

EMAND

: E

NSURING

ORK-

PLACE

TANDARDS AND

ORKER

ECURITY IN THE

-D

EMAND

CONOMY

(Sept.

2015), http://www.nelp.org/content/uploads/Rights-On-Demand-Report.pdf.

177. Alison Griswold, Hillary Clinton Is Skeptical About the “Sharing” Economy,

LATE

(July 13, 2015), http://www.slate.com/blogs/moneybox/2015/07/13/hillary

_clinton_economy_speech_i_ll_crack_down_on_sharing_economy_exploitation.html;

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 61 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 61

economically vulnerable workers without adequate social benefits. On

the other hand, Uber argues that drivers prefer to be classified as

independent contractors because they value the freedom to choose

their hours and to work for others.

178

It can point to evidence (from an

Uber-funded study) that over half of Uber drivers have other jobs and

drive for Uber for under 15 hours a week, which suggests that many

drivers are relying on Uber as a supplementary source of income and

may not need the benefits that would flow from being classified as

employees.

179

Irrespective of these arguments, classifying Uber and

see also Steven Greenhouse, Uber: On the Road to Nowhere

, A

. P

ROSPECT

(Dec. 7,

2015), http://prospect.org/article/road-nowhere-3 (stating that, “[b]y some estimates,”

Uber’s classification of its drivers as independent contractors “cuts Uber’s compensa-

tion costs by more than 20 percent per driver”); Steven D. Solomon, Uber Case High-

lights Outdated Worker Protection Laws,

N.Y. T

IMES

(Sept. 15, 2015), http://www.ny

times.com/2015/09/16/business/dealbook/uber-case-highlights-outdated-worker-pro

tection-laws.html. There are varying estimates of the size and significance of the con-

tingent workforce. See, e.g., Jonathan V. Hall & Alan B. Krueger, An Analysis of the

Labor Market for Uber’s Driver-Partners in the United States 3–7 (Princeton Univ.,

Indus. Relations Section, Working Paper No. 587, Jan. 22, 2015); Orly Lobel, The Gig

Economy & the Future of Employment and Labor Law 4–5 (Univ. of San Diego Sch.

of Law, Legal Studies Research Paper Series, Research Paper No. 16-223, Mar.

2016).

178. Growing and Growing Up,

BER

(Apr. 21, 2016), https://newsroom.uber.com/

growing-and-growing-up (“Drivers value their independence.”).

The fact that Uber drivers choose their hours does not necessarily mean that they

are independent contractors as a matter of law. In denying Uber’s motion for summary

judgment in a California case challenging its classification of its drivers as indepen-

dent contractors, the district court “cited cases holding that workers may set their own

hours and still be legal employees, so long as the putative employer exerts substantial

control while they are on the clock.”

RISHEN

OGERS

, A

. C

ONSTITUTION

’

Y FOR

& P

OLICY

, R

EDEFINING

MPLOYMENT FOR THE

ODERN

CONOMY

4 (2016),

https://www.acslaw.org/sites/default/files/Redefin-

ing_Employment_for_the_Modern_Economy.pdf (citing O’Connor v. Uber Techs.,

Inc., 82 F. Supp. 3d 1133, 1140 (N.D. Cal. 2015)).

179. Hall & Krueger, supra note 177, at 10 (“Uber’s driver-partners fall into three

roughly equal-sized groups: driver-partners who are partnering with Uber and have no

other job (38 percent), driver-partners who work full-time on another job and partner

with Uber (31 percent), and driver-partners who have a part-time job apart from Uber

and partner with Uber (30 percent).”); id. at 18 (“In the combined set of 20 areas

surveyed by BSG [Benenson Strategy Group], more than half of uberX driver-partners

chose to drive for less than 15 hours a week, and fully 85 percent chose to drive less

than 35 hours a week.”).

On the other hand, the Hall & Krueger study shows that a fair number of Uber

drivers rely on Uber as a source of income, and that many lack health insurance,

which suggests that Uber drivers potentially might benefit from being classified as

employees. See id. at 11 (“Driving on the Uber platform provides an important source

of income for driver-partners. For nearly one-quarter of driver-partners (24 percent),

Uber is their only source of personal income, and for another 16 percent Uber is their

largest but not only source of income. More than one-third of driver-partners view

income earned on the Uber platform as a supplement to their income but not a signifi-

cant source (38 percent).”); id. at 12 (“About half (49 percent) of Uber’s driver-part-

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 62 17-APR-17 10:20

62 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

Lyft drivers as independent contractors may not be legally sound and

the classification is now being litigated in multiple jurisdictions,

180

just as the incumbent taxi industry’s classification of its drivers as

independent contractors has been litigated over the years.

181

The fact

that the incumbent taxi industry often classifies its drivers as indepen-

dent contractors tends to be overlooked in the debate about whether

Uber drivers are legally independent contractors, and whether they

should be classified in this way.

The debate about the classification of Uber drivers is related to

the broad public policy debate about income inequality and economic

insecurity.

182

There are many proposals to address these concerns,

some of which would decouple social benefits from employment and

have governments, rather than employers, provide many of these ben-

efits.

183

Other less radical proposals involve extending some of the

ners currently receive employer-provided health insurance from their employer at

another job or from a spouse or other family member’s job”). By not providing bene-

fits such as health insurance to its drivers, Uber might be regarded as free-riding off

the expenditures of firms that are financing health insurance for the drivers that have

health insurance.

Uber has recently taken a number of steps to address the concerns that its drivers

are being exploited. Uber and the International Association of Machinists and Aero-

space Workers union recently announced that they will establish an association of

Uber drivers in New York City. In addition, Uber indicated that it plans to draw on the

advice of the Freelancers Union in the City on increasing the availability of benefits to

its drivers in the City, many of whom are full-time. See Josh Eidelson, Uber Found an

Unlikely Friend in Organized Labor,

LOOMBERG

USINESSWEEK

, (Oct. 27, 2016),

https://www.bloomberg.com/news/articles/2016-10-27/uber-found-an-unlikely-friend-

in-organized-labor; Noam Scheiber & Mike Isaac, Uber Recognizes New York Driv-

ers’ Group, Short of a Union,

N.Y. T

IMES

(May 10, 2016), http://www.nytimes.com/

2016/05/11/technology/uber-agrees-to-union-deal-in-new-york.html.

180. See supra note 29.

A small number of the state statutes legalizing “Transportation Network Compa-

nies” state, explicitly or implicitly, that the drivers for these companies are not em-

ployees of the companies. See Jon Weinberg, Gig News: Uber Successfully Pursuing

State Legislation on Independent Contractor Status,

ABOR

(Dec. 11, 2015), http://

onlabor.org/2015/12/11/gig-news-uber-successfully-pursuing-state-legislation-on-inde

pendent-contractor-status/; Summary Memoranda, supra note 45.

181. See supra note 19 (including case law challenging the classification of taxi driv-

ers as independent contractors).

182. Brishen Rogers, Whether and How to Regulate Uber and Lyft,

OLUM

. L. S

LUE

LOG

(

Aug. 25, 2015), http://clsbluesky.law.columbia.edu/2015/08/25/

whether-and-how-to-regulate-uber-and-lyft/ (noting that the “existing debate around

Uber [is] caught up in debates about inequality more generally”).

183. See, e.g.,

UNDARARAJAN

, supra note 10, at 177–202 (discussing a number of

options for providing benefits to the sharing economy workforce); Lobel, supra note

177, at 6–15 (outlining four paths for reforming employment and labor law to address

the rise of the contingent workforce, including “rethinking some of the historical links

between work and welfare”). One idea that has been much discussed recently is in-

creasing the minimum wage, a measure that might benefit low-income workers, al-

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 63 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 63

protections traditionally provided to employees to some workers now

treated as independent contractors. These include a recent proposal to

create an intermediate category of “independent worker” that would

enjoy some of the protections currently accorded only to employees—

such as “the freedom to organize and collectively bargain, civil rights

protections, tax withholding, and employer contributions for payroll

taxes”—but not others, such as “hours-based benefits, including over-

time or minimum wage requirements,” or “unemployment insur-

ance.”

184

The proponents of the independent worker category describe

Uber drivers and many taxi drivers as examples of the kinds of work-

ers that they believe should come within the category.

185

In the midst of the broad public policy debate about whether to

increase worker protections to address income inequality and eco-

nomic insecurity, actors regulating taxis must decide what kinds of

protections should be available to incumbent industry drivers and driv-

ers for the new transportation network companies. In contemplating

protections for drivers, taxi regulators must be mindful of legal limits

on their jurisdiction, which in some places constrain regulators to in-

troducing worker protection measures that can be tied to improving

taxi safety or managing taxis in some way.

186

Moreover, taxi regula-

though not taxi drivers if they are independent contractors. See, e.g., Alan B. Krueger,

The Minimum Wage: How Much Is Too Much?,

N.Y. T

IMES

(Oct. 9, 2015), http://

www.nytimes.com/2015/10/11/opinion/sunday/the-minimum-wage-how-much-is-too-

much.html.

184. Harris & Krueger, supra note 19, at 2. The proposal could be implemented by

Congress and state legislatures. Id. at 5.

With respect to workers’ compensation, the one benefit provided to some taxi

drivers otherwise categorized as independent contractors, including New York medal-

lion taxi drivers, Harris and Krueger propose that “intermediaries [such as Uber] be

permitted to opt to provide expansive workers’ compensation insurance policies to the

independent workers with which they work without transforming these relationships

into employment.” Id. at 20.

There are many criticisms of the proposal for a new “independent worker” cate-

gory. See, e.g.,

OGERS

, supra note 178, at 5–6; Ben Sachs, Do We Need an

“Independent Worker” Category?,

ABOR

(Dec. 8, 2015), http://onlabor.org/2015/

12/08/do-we-need-an-independent-worker-category/#more-6978; Noam Scheiber, A

Middle Ground Between Contract Worker and Employee,

N.Y. T

IMES

(Dec. 10,

2015), http://www.nytimes.com/2015/12/11/business/a-middle-ground-between-con-

tract-worker-and-employee.html. Lobel nicely situates the Harris and Krueger propo-

sal for an independent worker category within existing literature for addressing the

economic vulnerability of workers not classified as employees, including other pro-

posals for intermediate categories. Lobel, supra note 177, at 11.

185. Harris & Krueger, supra note 19, at 5, 14, 22–23, 27.

186. Several years ago, before the rise of Uber, the New York Taxi and Limousine

Commission attempted to establish a charge on taxi fares to create a fund enabling

taxi drivers to obtain health and disability insurance. A court invalidated the Commis-

sion’s efforts, partly on the basis that they “exceeded [the Commission’s] authority.”

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 64 17-APR-17 10:20

64 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

tors may face other legal obstacles to extending worker protections to

taxi drivers, such as federal antitrust law, which may be an obstacle to

ordinances such as that passed in Seattle to extend the right to collec-

tively bargain to traditional taxi and app-drivers.

187

However sympa-

thetic we might be to the desire to address the income inequality and

insecurity affecting many workers, including many taxi and Uber driv-

ers, the ability of taxi regulation specifically to deal with the issue may

be limited.

188

At a minimum, taxi regulators should maintain protections for

traditional taxi drivers that also may protect the public at large from

unsafe driving, and extend these protections to app drivers. These

worker protection requirements include the requirements in some ju-

Ahmed v. City of New York, 129 A.D.3d 435, 440 (N.Y. App. Div. 2015) (“TLC

[Taxi and Limousine Commission] has certain delineated powers to ensure that driv-

ers are capable of driving safely . . . . However, nothing in the Charter or the enabling

Code provisions contemplates the establishment and outsourcing of a miniature health

insurance navigation and disability insurance department.”). In October 2016, several

New York City Council members introduced legislation directing the Taxi and Lim-

ousine Commission to provide taxi and for-hire vehicle drivers with medical and other

benefits, funded from a surcharge on taxi fares. N.Y.C., N.Y., Int. 1301-2016 (intro-

duced Oct. 13, 2016). The legislation “would effectively negate” the just-described

court decision. Dana Rubinstein, Council Bill Would Provide Health Benefits for Taxi,

Uber Drivers,

OLITICO

(Sept. 6, 2016), http://www.politico.com/states/new-york/

city-hall/story/2016/09/city-council-aims-to-revive-health-care-fund-for-taxi-uber-

drivers-105199.

187. See Chamber of Commerce v. City of Seattle, No. C16-0322RSL, 2016 WL

4595981 (W.D. Wash. Aug. 9, 2016) (dismissing, based on lack of standing, action to

enjoin Seattle ordinance as violating federal antitrust law and preempted by the Na-

tional Labor Relations Act); Harris & Krueger, supra note 19, at 15–17 (describing

barriers that antitrust law creates for independent workers organizing to bargain col-

lectively); Mike Isaac et al., Seattle Considers Measure to Let Uber and Lyft Drivers

Unionize,

N.Y. T

IMES

(Dec. 13, 2015), http://www.nytimes.com/2015/12/14/technol-

ogy/seattle-considers-measure-to-let-uber-and-lyft-drivers-unionize.html (reporting on

potential that Seattle ordinance violates antitrust law); Caroline O’Donovan, Drivers

Win the Right to Bring Uber and Lyft to the Negotiating Table,

UZZ

EED

(Dec. 14,

2015), http://www.buzzfeed.com/carolineodonovan/drivers-win-the-right-to-bring-ube

r-and-lyft-to-the-negotiat/.

Seattle is attempting to invoke the “state action immunity defense.” See Seattle,

Wash., Ordinance 124968, Version 4 § 1.I.2

(

Dec. 23,

2015)

(“As the City is acting

under specific state statutory authority, it is immune from liability under antitrust

laws.”); Greenhouse

supra note 177. Litigation over the legality of the Seattle ordi-

nance has been delayed by the slow pace of implementing the ordinance. See Mike

Richards, Seattle Push for Uber Union Vote Slowed; Automation Is Coming, in Time,

ENS

(Aug. 29, 2016), http://thelens.news/2016/08/29/seattle-push-for-uber-union-

vote-slowed-automation-is-coming-in-time/.

188. See Harris & Krueger, supra note 19, at 15 (arguing for a “comprehensive solu-

tion” implemented by Congress and state legislatures to address “the emergence of

independent workers,” rather than “courts or administrative agencies us[ing] their ex-

isting authority to address a few of the problems created by the emergence of indepen-

dent workers” because courts and agencies lack sufficient authority).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 65 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 65

risdictions that medallion owners or app companies ensure that drivers

are covered by workers’ compensation insurance.

189

What about the lease caps that some jurisdictions introduced to

protect drivers leasing medallions from exploitation by medallion

owners? At this juncture, it is not clear how much benefit traditional

taxi drivers who lease medallions are deriving from the lease caps.

190

Taxi drivers in cities like New York with Uber and other apps likely

enjoy more market power than they have in decades.

191

Unemploy-

ment is low in many places, and so the pool of potential drivers likely

is relatively stable. Drivers no longer need to lease medallions; they

now have the choice of driving either traditional taxis with medallions

or vehicles for Uber, Lyft, and the other apps without medallions.

Many taxi drivers have been abandoning the traditional medallion in-

dustry and choosing to drive for Uber, which seems to have been forc-

ing medallion lessors to lower the leasing fees for medallions.

192

189. In New York City, Uber drivers generally are covered by the Black Car Fund,

which provides the equivalent of workers’ compensation for black car drivers, be-

cause many Uber drivers are licensed black car drivers. See Workers’ Comp. Law

§§ 2(3), 2(4), 18-c; 35 RCNY § 59B-12(b)(1); Rebecca Harshbarger, Uber’s Claim of

$90K Average Pay for Drivers Is Overestimated,

N.Y. P

OST

(Nov. 2, 2014), http://

nypost.com/2014/11/02/ubers-claim-of-90k-average-pay-for-drivers-is-overestimated/

; Workers’ Compensation Coverage: Taxi Cabs—Most Taxi Cab Operators Are Con-

sidered Employees,

N.Y. S

. W

ORKERS

’ C

OMPENSATION

http://www.wcb.ny.gov/

content/main/onthejob/CoverageSituations/taxiCabs.jsp (last visited Nov. 12, 2016).

190. In a recent New York lawsuit, the traditional taxi industry argues that the lease

caps that apply to it, but not for-hire vehicles including Uber vehicles, put the tradi-

tional industry at a disadvantage. Amended Complaint at 33, Melrose Credit Union v.

City of New York, No. 1:15-cv-09042 (S.D.N.Y. Mar. 7, 2016) (“FHVs [for-hire ve-

hicles] are not subject to lease caps, which allows FHVs to be leased to drivers at rates

determined by supply and demand. This disparate treatment forces Plaintiffs to oper-

ate at a disadvantage as compared to similarly situated FHVs by restricting Plaintiffs

from charging higher rates during times of high demand (e.g. on Friday), while charg-

ing lower rates during times of low demand (e.g. on Monday).”). This statement sug-

gests that the lease caps may still be binding, and therefore may still benefit taxi

drivers, at times.

191. The last time taxi drivers enjoyed as much market power in New York may

have been in the heyday of the taxi drivers’ union, in the 1960s and early 1970s.

192. See, e.g.,

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra

note 7, at 89 (“Many taxi drivers have shifted to ride service companies.”);

FFICE OF

THE

AYOR

, C

ITY OF

ORK

, supra note 52, at 3 (“In the past two years, the

number of active yellow taxi drivers [in New York City] has declined by five percent.

Over the same two-year period, new for-hire vehicle registrations by former yellow

taxi drivers have increased many times over.”); Rebecca Harshbarger, Yellow Taxis

Launch $1 Million Campaign to Lure Back Drivers, Passengers from Uber, Lyft,

ORK

(Dec. 14, 2015), http://www.amny.com/transit/yellow-taxis-launch-1-

million-campaign-to-lure-back-drivers-passengers-from-uber-lyft-1.11228657

(describing “$1 million campaign” that fleet owners have launched “to bring back

drivers who left yellow cabs for new options like Uber”). For a discussion of why taxi

drivers are switching from traditional taxis to Uber and other for-hire vehicle compa-

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 66 17-APR-17 10:20

66 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

Driving for Uber may be more profitable for drivers, because they

avoid paying medallion leasing fees since an Uber vehicle does not

require a medallion, though drivers pay Uber a service fee on fares.

193

Moreover, it is difficult to argue at this juncture that measures

analogous to lease caps should be introduced to protect Uber drivers

from exploitation by Uber.

194

In cities such as New York where Uber,

Lyft, and other apps are seeking to establish themselves, the apps are

recruiting drivers and they appear to be offering drivers relatively

good terms (compared with the traditional taxi industry, if not in abso-

lute terms). The evidence for this is that, as mentioned above, tradi-

tional taxi drivers in cities like New York have been leaving the

traditional taxi industry to drive for Uber.

195

However, the favorable conditions for drivers may not last, and

they already appear to be disappearing in some places. Thus worker

protections may be warranted if not now, then in the future, to avoid a

diminution in working conditions that might jeopardize public

safety.

196

Uber’s expansion strategies include reducing fares to attract

passengers, increasing the portion of the fare that it takes from driv-

ers,

197

and increasing the number of its drivers, all of which reduce

drivers’ take-home earnings if Uber’s expansion is not accompanied

by an offsetting increase in consumer demand for service.

198

Moreo-

nies, based on a survey of drivers, see

FFICE OF THE

AYOR

, C

ITY OF

ORK

supra note 52, at 9.

193. Complaint at 21, CGS Taxi LLC v. City of New York, No. 653264/2015 (N.Y.

Sup. Ct. Sept. 30, 2015); Complaint at 20, Daler Singh v. City of New York, No.

701402/2017 (N.Y. Sup. Ct. Jan. 30, 2017).

194. Isaac et al., supra note 187.

195. Uber has been subsidizing drivers to grow the number of drivers in its network.

Eric Newcomer, Uber Loses at Least $1.2 Billion in First Half of 2016,

LOOMBERG

ECHNOLOGY

(Aug. 25, 2016), https://www.bloomberg.com/news/articles/2016-08-

25/uber-loses-at-least-1-2-billion-in-first-half-of-2016 (“Uber’s losses in the first half

of 2016 totaled at least $1.27 billion. . . . Subsidies for Uber’s drivers are responsible

for the majority of the company’s losses globally.”). But see

reenhouse

supra note

177 (referring to Uber ads in New York City in fall 2015 “saying anyone who signed

up to drive would earn a minimum of $7,000,” but emphasizing that Uber’s other

polices, such as fare reductions, are harming Uber drivers and describing an Uber

driver as “cut[ting] back his Uber hours to part-time so he could also drive for a

friend’s black-car service”).

196. See generally Greenhouse, supra note 177.

197. See Isaac et al., supra note 187 (reporting that in Seattle, “Uber and Lyft have

cut the rates they charge passengers for rides and ended the incentives used to recruit

drivers”); Greenhouse

supra note 177 (referring to an Uber fare decrease in Los

Angeles).

198. See James Covert, Uber’s NYC Fare Cuts Haven’t Helped Drivers,

N.Y. P

OST

(Apr. 25, 2016), http://nypost.com/2016/04/25/uber-is-really-feeling-the-wrath-of-ci-

tys-rate-cuts/ (stating that January 2016 fare cuts have not increased driver pay);

Greenhouse

supra note 177 (referring to the effects of increasing the number of

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 67 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 67

ver, down the line, there is the possibility that Uber or another app

may emerge as the dominant provider of taxi services, by eclipsing the

other apps and the traditional taxi industry. If that occurs, drivers

where it happens could be economically vulnerable to the dominant

app, which might attempt to significantly increase the fees that it

charges drivers for matching them with customers, in order to increase

the app company’s share of fares.

199

Paralleling the lease caps imposed to protect traditional taxi driv-

ers, the service fee that Uber charges drivers on fares could be simi-

larly capped. However, effective implementation of a cap on the

service fees might require regulating the underlying fares that app

companies charge passengers, as app companies might circumvent any

caps on service fees charged to drivers by altering the fare structure.

“[M]inimum mile charges for riders” might be used to protect Uber

drivers from exploitation by Uber,

200

though these also likely would

entail regulating fare levels.

Universal Service Requirements

The last traditional pillar of taxi regulation is the requirement that

taxis agree to serve all riders.

201

This universal service requirement is

justified based on a non-discrimination principle, and reflects a com-

mitment to avoiding inequitable outcomes, as opposed to inefficient

UberX drivers and Uber’s reducing fares by “30 percent” on incomes of drivers in

New York); Marc Santora & John Surico, Uber Drivers in New York City Protest

Fare Cuts,

N.Y. T

IMES

(Feb. 1, 2016), http://www.nytimes.com/2016/02/02/nyregion/

uber-drivers-in-new-york-city-protest-fare-cuts.html (reporting on a protest by Uber

drivers of a fare cut by Uber, which the company defends on the basis that a past fare

cut increased driver earnings by inducing more passengers to take Uber).

199. See

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note

7, at 77; Darb´era, supra note 10, at 13.

200. Isaac et al., supra note 187.

201. See, e.g., FAQs – Passenger: Refusal Law,

NYC T

AXI

& L

IMOUSINE

OMM

’

http://www.nyc.gov/html/tlc/html/faq/faq_pass.shtml (last visited Nov. 12, 2016) (“It

is against the law to refuse a person based on race, disability, or a destination in New

York City. A taxicab driver is required to drive a passenger to any destination in the

five boroughs.”). The prohibition is in the New York City Administrative Code and

Rules of the City of New York. N.Y.C. Admin. Code § 19-507(a)(2) (West 2016); 35

R.C.N.Y. § 54-20; see also Ill. Transp. Trade Ass’n v. City of Chicago, 134 F. Supp.

3d 1108, 1111 (N.D. Ill. Sept. 22, 2015), aff’d in part & rev’d in part, 839 F.3d 594

(7th Cir. 2016) (noting that Chicago “[t]axis are prohibited by . . . Ordinance to refuse

service to a passenger based on destination” and that “taxis are required to abide by all

federal, state, and City non-discrimination laws”). Edelman and Geradin also use the

term “universal service” to refer to the obligation to “serve all customers.” Edelman &

Geradin, supra note 10, at 320.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 68 17-APR-17 10:20

68 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

ones, as efficiency is conventionally understood.

202

For decades, the

concern animating the enforcement of the requirement has been that

drivers discriminate against riders based on their race or their destina-

tion (which, due to residential segregation based on race in major cit-

ies, is often correlated with race).

203

Anti-discrimination norms justify requiring app-dispatched taxis

to accept rides irrespective of the destination or the characteristics of

the rider, just as traditional taxis are required to do. There has been

relatively little controversy about the idea that taxi apps should be

prohibited from discriminating against riders based on their race or

destination.

204

Some argue that app-based taxis are less likely to dis-

criminate based on riders’ destination or race than street-hailed taxis.

The apps’ ability to match drivers with riders may reduce drivers’ in-

centives to turn down rides based on destination because the apps’

ability to match drivers and passengers may make it easier for drivers

to get return trips from less densely populated areas.

205

Apps like Uber

also monitor the percentage of trips that drivers accept. Uber encour-

ages its drivers to accept at least 80% of trips, “but ‘the closer to 100%

the better,’” and this may discourage drivers from turning down trips

based on race or destination.

206

Also, “Uber drivers don’t see where

202. I add the caveat “as efficiency is conventionally understood” advisedly. See the

discussion in terms of efficiency of “prohibition[s] against service refusal” in

RANKENA

& P

AUTLER

, supra note 34, at 61–63. They suggest that such prohibitions

“may be second best efficient” in certain circumstances. Id. at 65.

203. See, e.g., Booth Now Will Prosecute Cab Discrimination Cases,

N.Y. T

IMES

(Dec. 10, 1966), http://timesmachine.nytimes.com/timesmachine/1966/12/10/8297

4302.html?pageNumber=28; Equal Taxi Opportunity,

N.Y. T

IMES

(Mar. 14, 1978),

https://timesmachine.nytimes.com/timesmachine/1978/03/14/110803095.html?page

Number=34; Negroes Accept Wallander Word: Commissioner Won’t Tolerate Race

Bias, Will Investigate Charges of Brutality,

N.Y. T

IMES

(Aug. 10, 1946), http://times-

machine.nytimes.com/timesmachine/1946/08/10/91621786.html?pageNumber=28

(noting that a police commissioner “promised that his department would look into the

practice of certain taxicab drivers in refusing to drive fares to Harlem,” and that of-

fending taxicab drivers would be “dealt with by prompt action”); State Warns Taxis to

Accept Negroes,

N.Y. T

IMES

(Mar. 23, 1961), http://timesmachine.nytimes.com/

timesmachine/1961/03/23/118029313.html?pageNumber=25.

For a useful discussion of the requirement that taxis “serve all geographic areas

of the regulating jurisdiction” that recognizes that it implements a form of cross-subsi-

dization, see

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra

note 7, at 44–46 (citing P.S. Dempsey, Taxi Industry Regulation, Deregulation, & Re-

regulation: The Paradox of Market Failure, 24

RANSP

. L.J.

73 (1996)).

204. Arun Sundararajan advocates delegating to platforms responsibility for ensuring

“that there is no discrimination in transactions conducted on their platforms.” FTC

Workshop, supra note 10, at 159 (remarks of Arun Sundararajan).

205. Rogers, supra note 10, at 95.

206. Cook, supra note 164 (quoting an Uber email newsletter to drivers).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 69 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 69

passengers want to go until after they’ve accepted a request.”

207

This

makes it harder to discriminate based on destination, although not im-

possible because drivers still can cancel trips after accepting them. On

the other hand, a recent academic study found evidence that UberX

drivers are discriminating against passengers based on race, presuma-

bly using the passenger information that drivers get after accepting a

trip to identify passenger race, and the driver’s ability to cancel a trip

after initially accepting it. In particular, the study “found that UberX

drivers are nearly three times as likely to cancel a ride on a male pas-

senger upon seeing that he has a ‘black-sounding’ name.”

208

Drivers

also might discriminate against passengers based on race through the

ratings that drivers give passengers at the completion of trips, but the

study did not find that driver ratings varied depending on the race of

207. Johana Bhuiyan, Uber and Lyft Position Themselves as Relief from Discrimina-

tion,

UZZ

EED

(Oct. 7, 2014), http://www.buzzfeed.com/johanabhuiyan/app-based-

car-services-may-reduce-discrimination-issues-tk.

208. Yanbo Ge et al., Racial and Gender Discrimination in Transportation Network

Companies 19 (Nat’l Bureau of Econ. Research, Working Paper No. 22776, Oct.

2016). The study indicates that the finding “seems to be driven primarily by behavior

in areas with low population densities.” Id. The study also found discrimination based

on gender, with female passengers “driven farther.” Id. at 18. The finding of driver

discrimination echoes a similar finding that Airbnb hosts discriminate against guests

based on race. See Mike McPhate, Discrimination by Airbnb Hosts Is Widespread,

Report Says,

N.Y. T

IMES

, (Dec. 11, 2015), http://www.nytimes.com/2015/12/12/busi-

ness/discrimination-by-airbnb-hosts-is-widespread-report-says.html (“Fictional guests

set up by the researchers with names like Lakisha or Rasheed were roughly 16 percent

less likely to be accepted than identical guests with names like Brent or Kristen.”). In

a co-authored article, one of the authors of the study about discrimination on Airbnb

presciently predicted that “if a transportation platform made a passenger’s race salient

(through name and face), it might facilitate the same discrimination passengers previ-

ously faced offline.” Edelman & Geradin, supra note 10, at 322.

There is some evidence that Uber is better serving areas of New York City that

have traditionally been under-served by medallion taxis. See Jared Meyer, Bringing

Uber to All New Yorkers,

EDERALIST

(Dec. 22, 2015), http://thefederalist.com/

2015/12/22/bring-uber-to-all-new-yorkers/ (“In December 2014, 27 percent of UberX

pickups were outside Manhattan or city airports. For comparison, less than 6 percent

of yellow taxi rides begin in the outer boroughs (excluding airports) and the northern

tip of Manhattan.”). Also, there is evidence that Uber is serving low-income and Afri-

can-American neighborhoods, as well as higher-income neighborhoods in New York

City. Id.

For a discussion of the potential for transportation network companies to reduce

or exacerbate racial discrimination, and service low-income areas, see generally

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note 7, at

132–34, 141. The report notes that there is “a lack of data . . . to assess [the] . . .

arguments” about the impact of these companies on racial discrimination, and only

“preliminary” evidence about whether the companies are improving service in low-

income areas. Id. at 134, 141; see also id. at 145.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 70 17-APR-17 10:20

70 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

the passenger.

209

Customers also may be discriminating against Uber

drivers based on race in rating drivers, something that the apps cur-

rently are not required to police because they have not been deemed

employers bound by “Title VII of the [federal] Civil Rights Act of

1964.”

210

Even if racial discrimination remains an issue with the Uber

platform, the data that Uber collects about passengers and drivers po-

tentially could be used to identify patterns of discrimination and ad-

dress them in novel ways that were not possible in the pre-app era.

211

Recently, people with mobility disabilities have argued that all

taxis should be accessible to people with such disabilities, and sought

to broaden the ambit of the universal service requirement.

212

Some

cities have introduced policies that require that a certain percentage of

their traditional taxi fleets be accessible to persons with mobility is-

sues by a target date.

213

In 2014, the New York City Taxi and Limou-

209. Ge et al., supra note 208, at 18 n.9. For the idea that driver ratings of passen-

gers might discriminate based on race, see

OMM

FOR

EVIEW OF

NNOVATIVE

R-

BAN

OBILITY

ERVS

, supra note 7, at 133.

210. Ben Sachs, Uber: A Platform for Discrimination?,

ABOR

(Oct. 22, 2015),

http://onlabor.org/2015/10/22/uber-a-platform-for-discrimination/; see also

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note 7, at 133 (referring

to potential for passenger ratings to discriminate against drivers based on race); Dan

Adams, Boston-Based Attorney Argues Uber’s Star Ratings Are Racially Biased,

LOBE

(Oct. 7, 2016), https://www.bostonglobe.com/business/2016/10/06/attorney-

for-uber-drivers-says-star-ratings-are-racially-biased/R28mqWL6ShjMFB5xAr3uGL/

story.html (noting that a Boston attorney has filed a complaint with the EEOC arguing

that Uber’s system for passenger ratings of drivers is “racially discriminatory”). Uber

and other transportation network companies already may face liability if their drivers

suffer discrimination based on race, even if the drivers are considered independent

contractors.

OGERS

, supra note 178, at 9.

211. Ge et al., supra note 208, at 20; Lobel, supra note 177, at 10.

212. See, e.g., Jan Garrett, On the Move: Increasing the Wheelchair Accessible Taxis

Around the Country, 34

. R

. 14 (2007) (briefly describing history of efforts to

increase the accessibility of taxicabs to people with mobility disabilities in New

York);

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note 7, at

46–47 (describing efforts to increase the accessibility of taxis and the modest resulting

increases in accessibility).

213. See, e.g.,

ISABILITY

DVISORY

OMM

., D.C. T

AXICAB

OMM

’

, C

OMPREHEN-

SIVE

EPORT AND

ECOMMENDATIONS ON

CCESSIBLE

AXICAB

ERVICE

4 (Feb. 20,

2014), http://dfhv.dc.gov/sites/default/files/dc/sites/dc%20taxi/page_content/attach

ments/DC%20Taxicab%20Comission%20Disability%20Advisory%20Committee%20

Comprehensive%20Report%20022014%20FINAL%20w%20Addendum.pdf (“Under

the DC Taxi Act, each taxi company with [twenty] or more taxicabs in its fleet as of

July 1, 2012 will be required to dedicate a portion of its fleet to wheelchair accessible

taxis: [six] percent by December 31, 2014; [twelve] percent by December 31, 2016;

and [twenty] percent by December 31, 2018.”); id. at 14 (“Chicago recently passed

legislation with the following requirement: ‘Any single licensee that owns or controls

[twenty] or more licenses must place into service wheelchair accessible vehicles as

taxicabs on five percent of its taxicab vehicle fleet.’”) (excerpting

. M

. C

ODE

§ 9-112-070); see also Matthew W. Daus,

XPANDING

RANSPORTATION

ET-

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 71 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 71

sine Commission required that “[b]y 2020, half of all yellow taxis—

7,500 total—will be wheelchair accessible,” consistent with the terms

of the settlement of a lawsuit brought by disability rights advocates

under the Americans with Disabilities Act.

214

“The transformation

will be funded through a 30-cent per-ride Taxi Improvement

Surcharge integrated into the yellow medallion taxicab fare as of Janu-

ary 1, 2015.”

215

The requirement that a certain percentage of the fleet

be accessible does not apply to black car companies such as Uber in

New York City, but black car company bases are required to “provide

‘equivalent service’ to persons with disabilities,” an obligation that the

bases can meet by “[d]ispatching an [a]ccessible [v]ehicle from” their

base or “[c]ontracting with another base.”

216

In 2014, the Taxi and

Limousine Commission found a general lack of compliance with this

WORK

OMPANY

“E

QUITY

”

26–27, http://www.whosdrivingyou.org/wp-content/

uploads/2016/08/Equity-Report-FINAL-11232642.pdf (summarizing local policies to

increase the accessibility of taxicab service in various municipalities in the U.S. and

Canada).

214. Press Release,

N.Y.C. T

AXI

& L

IMOUSINE

OMM

’

, City Poised for Major Ex-

pansion of Accessible Taxis Under New TLC Rules (Mar. 26, 2014), http://www.

nyc.gov/html/tlc/downloads/pdf/press_release_03_26_14.pdf; see also 35 RCNY

§ 58-50 (2015) (listing taxicab accessibility requirements); 35 RCNY ch. 53 (regard-

ing accessible taxicabs); Singh v. Joshi, 152 F. Supp. 3d 112, 119–120 (E.D.N.Y.

2016) (describing implementation of taxicab accessibility requirements). On the litiga-

tion that led to the policy commitment, see

ISABILITY

DVISORY

OMM

, supra note

213, at 3. For an unsuccessful attempt to enjoin the implementation of regulations to

make New York City’s taxi fleet more accessible, see Singh, 152 F. Supp. 3d 112; see

also Singh v. Joshi, No. 15-CV-5496-FB-VMS, 2016 WL 4272349 (E.D.N.Y. Aug.

15, 2016) (denying plaintiffs’ motion for reconsideration and granting defendants’

motion for summary judgment). On the New York Taxi and Limousine Commission’s

accessible dispatch system, see Find an Accessible Ride,

N.Y. T

AXI

& L

IMOUSINE

OMM

’

http://www.nyc.gov/html/tlc/html/passenger/accessible.shtml (last visited

Nov. 12, 2016).

215. Press Release,

N.Y.C. T

AXI

& L

IMOUSINE

OMM

’

, supra note 214; see also

Singh, 152 F. Supp. 3d at 122 (describing New York City’s argument that the Taxicab

Improvement Fund will offset the cost to medallion owners of changing to accessible

vehicles, but also noting that as of the date of the decision the City had not imple-

mented a timetable for distributing the Fund); Meera Joshi, Commissioner &

Chairperson, N.Y.C. Taxi & Limousine Comm’n, Speech to Crain’s Breakfast 7–8

(Jan. 14, 2016) (Exhibit B to Letter from Karen B. Selvin, Assistant Corporation

Counsel, New York City Law Department, to Honorable Analisa Torres, United

States District Court Southern District of New York (Aug. 18, 2016)) (describing the

Taxi and Limousine Commission’s “financial incentive programs for drivers and own-

ers of wheelchair accessible taxis”).

216. 35 RCNY § 59B-17(c). The term “Equivalent Service” is defined in 35 RCNY

§ 59B-17(c)(2); see also Singh, 152 F. Supp. 3d at 118 (noting that the New York

Taxi and Limousine Commission “require[s black car] licensees to provide a wheel-

chair-accessible vehicle on request”); Singh, 2016 WL 4272349 (denying plaintiffs’

motion for reconsideration and granting defendants’ motion for summary judgement).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 72 17-APR-17 10:20

72 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

regulatory requirement.

217

Advocates for people with disabilities in

New York City also have criticized Uber for attempting to serve peo-

ple with disabilities by dispatching yellow and green taxis, which oth-

erwise compete for business with Uber vehicles.

218

App companies

such as Uber also are being sued for poorly serving people with mo-

bility disabilities.

219

The requirements that a certain percentage of taxi fleets must be

accessible need to be rethought with the advent of taxi apps.

220

When

the traditional taxi industry was insulated from competition, it was

logical to think that service to people with mobility disabilities could

be improved by requiring the industry to purchase and operate accessi-

ble taxi vehicles, with assistance from subsidies to offset the cost of

purchasing and operating accessible vehicles. Requiring that a certain

percentage of taxi fleets be accessible might not by itself ensure that

217. Joshi, supra note 215, at 8 (“The pre-arranged car service world . . . is subject

to a more general requirement to provide wheelchair users with service that is

equivalent to what’s provided to other passengers, a requirement that has not always

created the desired reality. In the fall of 2014 the TLC did an industry wide enforce-

ment action, and 90% of the care [sic] service industry was unable to provide a wheel-

chair passenger with equivalent service.”).

218. See Johana Bhuiyan, Disability Rights Advocate Files Discrimination Com-

plaint Against Uber,

UZZ

EED

(Jan. 19, 2016), https://www.buzzfeed.com/

johanabhuiyan/disability-rights-advocate-files-discrimination-complaint-ag; Rosa

Goldensohn, Wheelchair Users’ Advocacy Group Demands Crackdown on Uber,

RAIN

’

N.Y. B

(Jan. 13, 2016); Issie Lapowsky, Uber’s Business Isn’t Built to

Help Disabled People,

IRED

(Aug. 14, 2015), http://www.wired.com/2015/08/uber-

disability/; Sarina Trangle, Taxi Medallion Owners, Disability Advocates Call for

Wheelchair Accessible Mandate for Uber,

ITY

& S

TATE

N.Y. (June 22, 2016), http://

cityandstateny.com/articles/politics/new-york-city/taxi-medallion-owners,-disability-

advocates-call-for-wheelchair-accessible-mandate-for-uber.html.

Uber’s use of accessible yellow and green taxis to provide accessible service

could be regarded as free-riding off the traditional taxi industry’s investments (albeit,

subsidized investments) in providing accessible service. In any event, relying on the

traditional taxi industry to provide accessible service may not be sustainable to the

extent that it is displaced by Uber and other new entrants.

219.

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note 7, at

136 (discussing litigation involving transportation network companies by advocates

for people with disabilities); Daus, supra note 213, at 29-30 (same). However, there

are disability rights advocates who support Uber, arguing that it has improved the

transit options of people with disabilities. See, e.g., Chris Bragg, Disability Rights

Groups Support Legalizing Uber in New York,

IMES

NION

(Dec. 21, 2015), http://

www.timesunion.com/tuplus-local/article/N-Y-taxi-operator-subsidizes-work-of-disa-

bility-7972052.php. It may be that Uber has increased the transit options for people

with some types of disabilities, but not others. See

OMM

FOR

EVIEW OF

NNOVA-

TIVE

RBAN

OBILITY

ERVS

, supra note 7, at 136, 145; FTC Workshop, supra note

10, at 141 (remarks of Ashwini Chhabra).

220. Jim Dwyer, Accessibility as Challenge in Age of Uber,

N.Y. T

IMES

(Dec. 22,

2015), http://www.nytimes.com/2015/12/23/nyregion/accessibility-as-challenge-in-

age-of-uber.html.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 73 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 73

accessible taxis are available when and where needed by people with

disabilities, because the taxis might be servicing other passengers or

off-duty. But the odds generally were that fleet requirements for ac-

cessibility would improve service to people with disabilities, espe-

cially if a large share of the fleet was required to be accessible.

Moreover, while it remained largely insulated from competition, the

traditional taxi industry likely could afford to cross-subsidize the pro-

vision of services to people with mobility disabilities from the indus-

try share of taxi fares, to the extent that any additional costs in

providing these services were not covered by taxi riders through spe-

cial surcharges like New York City’s or taxpayer-funded subsidies.

221

The picture has changed now that the traditional industry is com-

peting against Uber and other apps, which are not similarly required to

ensure that a certain percentage of the vehicles on their platforms are

accessible to persons with mobility disabilities. In New York City at

least, the accessible vehicle fleet requirements may no longer be func-

tioning to increase the number of accessible taxis on city streets. Ac-

cording to a recent complaint filed in federal court by medallion

owners and credit unions that finance medallion purchases, drivers are

refusing to lease accessible vehicles, because they are less profitable

for drivers to operate (even with the subsidies) than conventional taxi

vehicles, and drivers now have the choice to lease taxis that are not

accessible or to drive with Uber and other apps in vehicles that are not

accessible.

222

With drivers refusing to lease accessible taxicabs, there

221. See, e.g., Amended Complaint at 40, Melrose Credit Union v. City of New

York, No. 1:15-cv-09042 (S.D.N.Y., Mar. 7, 2016) (“[T]he only conceivable justifica-

tion for requiring medallion taxicabs to comply with the Accessible Conversion Rules

while not imposing comparable rules on FHVs [for-hire vehicles] was the medallion

owners’ exclusive right to hails.”). For a description of the New York City Taxi and

Limousine Commission’s “financial incentive[s]” for medallion taxi drivers and vehi-

cles owners to deploy wheelchair accessible vehicles, see Joshi, supra note 215.

222. See Amended Complaint at 25–26, Melrose Credit Union, No. 1:15-cv-09042

(“According to drivers, accessible vehicles rattle, are too big, guzzle gas, and are

equally disliked by passengers. According to drivers, accessible vehicles are simply

not worth leasing given the additional regulatory burdens imposed on them and the

alternative leasing options available, including abundant unrestricted medallion vehi-

cles and an endless supply of unrestricted Uber vehicles.”); id. at 39 (“[T]he driver of

a wheelchair-accessible cab may pick up non-disabled fares, but must always respond

to calls for a disabled pickup if they’re the closest wheelchair-accessible cab to the

call and are dispatched. As a result, drivers sometimes travel multiple unpaid blocks

to pick up a disabled fare.”) (excerpting Daniel Fitzsimmons, Fare Access,

N.Y.

RESS

(Aug. 26, 2015)); id. at 13–14, 36–40; see also Singh v. Joshi, No. 15-CV-

5496-FB-VMS, 2016 WL 304761, at *9–10 (E.D.N.Y. 2016); Memorandum of Law

in Support of the Verified Petition and Petitioners’ Motion for a TRO and Preliminary

Injunction at 10, Clair v. City of New York, No. 102277/2015 (N.Y. Sup. Ct. Feb. 18,

2016);

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note 7, at

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 74 17-APR-17 10:20

74 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

likely are fewer of them on city streets to service the needs of people

with mobility disabilities. Also, facing competition for drivers, and

lower profits, incumbent medallion owners now are even less willing

than before to help finance the cost of serving people with mobility

disabilities, even when there are programs in place to offset industry

costs in this regard, as in New York City.

223

We should be looking for ways to use the new app technology to

improve transportation services, including taxi services, for people

with mobility disabilities.

224

How to adapt the new technology to bet-

ter serve people with mobility disabilities is a large topic that is be-

yond the scope of this article. As the number of people with mobility

disabilities increases with the aging of the population, the issues that

need to be considered include how to better integrate taxis and para-

transit services, as there is the potential to provide enhanced and more

cost effective transportation services to many people with disabilities

through new and incumbent taxi services rather than conventional

paratransit services.

225

Another issue is how to fund the provision of

134, 136–37 (listing “obstacles” to the expansion of “accessible vehicle fleets” and

describing drop in availability of accessible taxis and “wheelchair trips” in San Fran-

cisco); Lapowsky, supra note 218.

223. Singh v. Joshi, No. 15-CV-5496-FB-VMS, 2016 WL 4272349 (E.D.N.Y. Aug.

15, 2016) (rejecting plaintiffs’ motion for reconsideration and granting New York

City defendants’ motion for summary judgment in challenge by medallion owners to

wheelchair accessibility requirements).

224. In New York City, Uber is suggesting “that the city levy a small per-trip fee on

‘for-hire vehicles’” that would create a fund that the for-hire industry, including Uber,

could use to “offer sweeteners to bases and drivers to get them to deploy accessible

vehicles.” Dana Rubinstein, Uber and de Blasio Aides Quietly Push Dueling Accessi-

ble-Taxi Proposals,

OLITICO

(Apr. 21, 2016), http://www.politico.com/states/new-

york/city-hall/story/2016/04/uber-and-de-blasio-aides-quietly-push-dueling-accessible

-taxi-proposals-100914. For two other ideas on how platforms such as Uber might be

required to serve people with mobility disabilities, see Edelman & Geradin, supra

note 10, at 321.

225. “The Americans with Disabilities Act (ADA) requires that individuals with dis-

abilities who are unable to use accessible mass transit for some or all of their trips

must be provided with paratransit.” Guide to Access-A-Ride Service,

ETRO

. T

RANSP

UTH

, http://web.mta.info/nyct/paratran/guide.htm (last visited Nov. 12, 2016). Some

transit agencies are experimenting with using Uber and Lyft to provide paratransit

services. Powell, supra note 90. For two reports recommending that paratransit agen-

cies work with transportation network companies to improve paratransit service, and

integrate technology used by transportation network companies, see

ITIZENS

UDGET

OMM

’

CCESS

-A-R

IDE

: W

AYS TO

O THE

IGHT

HING

ORE

FFICIENTLY

(2016);

ARAH

M. K

AUFMAN ET AL

., I

NTELLIGENT

ARATRANSIT

(2016)

. For a report

highlighting the poor service and inefficiency of Access-A-Ride, which provides para-

transit in New York City, see

ARJORIE

ANDA

, O

FFICE OF THE

OMPTROLLER

, C

ITY

ORK

, A

UDIT

EPORT OF THE

ETROPOLITAN

RANSPORTATION

UTHOR-

ITY

’

CCESS

-A-R

IDE

ROGRAM

(2016), http://comptroller.nyc.gov/wp-content/

uploads/documents/FK15_098A.pdf.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 75 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 75

taxi and other transportation services for people with mobility disabili-

ties. As in New York City, there is a tendency to fund taxi services for

people with disabilities from taxi fares and paratransit from transit

fares, but there is a strong case that taxpayers should fund these ser-

vices, not riders, because the benefits accrue to society at large.

226

The New Anti-Monopoly Function

While the advent of taxi apps may provide occasion for scaling

back taxi regulation, the apps also may suggest a new role for regula-

tors: guarding against the creation of a monopoly taxi app. As men-

tioned above, there are a number of commentators who argue that

Uber is a natural monopoly,

227

and the company’s high valuation sug-

gests there are many market players who may share this view.

228

But

not everyone agrees that Uber is a natural monopoly, because it is easy

for drivers and passengers to switch to other apps, and Uber is not a

monopoly at this juncture.

In this transitional period, regulators should be trying to avoid the

creation of a monopoly app such as Uber.

229

This means that regula-

tors must avoid pursuing policies that could have the unintended con-

sequence of making it more likely that Uber will become a monopoly.

For example, in the fall of 2014, the New York City Taxi and Limou-

sine Commission considered—and then dropped—a proposed rule

that would have limited the ability of for-hire vehicle drivers to accept

rides from multiple dispatchers (technically, bases). As a letter from

New York State Attorney General Eric Schneiderman pointed out, the

rule likely would have led to “market consolidation around a small

number of the best capitalized and most well-known services” (i.e.

226.

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note 7, at

134–35. The Committee report also points out that making taxi riders pay for accessi-

ble taxis through surcharges on taxi fares, which is the current practice in New York

City for example, could be regressive because it could mean “low-income taxi riders

would pay disproportionately high per trip and per mile fees to subsidize people in

wheelchairs who might well be more affluent than they are.” Id. at 135; see also

ITIZENS

UDGET

OMM

’

, supra note 225, at 1 (arguing that the cost of funding

Access-A-Ride, the Metropolitan Transportation Authority’s paratransit service,

should “be borne” largely by taxpayers, rather than transit fares, while transit “fares,

tolls and dedicated taxes” now finance fifty-seven percent of the cost of Access-A-

Ride).

227. See supra note 118.

228. Darb´era, supra note 10, at 12–13.

229. Not everyone agrees that regulators should be seeking to avoid the creation of a

monopoly taxi app. Some scholars suggest that it might be preferable for regulators to

facilitate the creation of a monopoly taxi app, and then regulate it as a natural monop-

oly due to the beneficial network effects of the platform. Weyl & White, supra note

10.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 76 17-APR-17 10:20

76 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

Uber).

230

If limited to choosing a single dispatcher, drivers likely

would have chosen the company with the largest network of

customers.

Another Topic: Who Should Regulate Taxis?

The question of which level of government should be regulating

taxis in the age of Uber is a second order issue that is attracting con-

siderable attention. Historically, municipal governments have been

mainly responsible for regulating taxis in the U.S., with relatively mi-

nor involvement from the federal and state governments.

231

However,

state legislatures recently have been extremely active in legalizing the

spread of taxi apps. Over half of the states have passed legislation

creating legal authorization for Uber and other transportation network

companies to operate,

232

and there have been calls for other state leg-

islatures or regulators to do so in the face of entrenched resistance to

taxi apps at the local level.

233

Some scholarship has even called for

230. See Letter from Eric T. Schneiderman, N.Y. Attorney Gen., to the Members of

the N.Y.C. Taxi & Limousine Comm’n (Nov. 7, 2014), https://ag.ny.gov/pdfs/Let-

ter_from_AG_to_TLC.pdf; see also Johana Bhuiyan, In a Win for Uber, Taxi and

Limousine Commission Revises Proposed One-Base Rule,

UZZFEED

(Nov. 19, 2014),

http://www.buzzfeed.com/johanabhuiyan/in-a-win-for-uber-taxi-and-limousine-com-

mission-revises-prop#.ic7d9J5q; Dan Rivoli, Uber, Lyft Fight Proposed City Rule

That Would Alter How Drivers Are Dispatched,

ORK

(Oct. 16, 2014), http://

www.amny.com/transit/tlc-proposes-rules-changes-to-hiring-drivers-1.9512638.

231.

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note 7, at

39 (“In large metropolitan areas . . . taxis [are] commonly regulated by local govern-

ments and sometimes airport authorities as well.”).

232. “Colorado was the first state to legislatively address the [Transportation Net-

work Companies.]” Strong, supra note 10, at 1079; see also Dobson, supra note 10,

at 709–10. Colorado has historically regulated taxis and for-hire vehicles; accord-

ingly, state regulation of the app-dispatched vehicles did not entail preempting local

authority. Strong, supra note 10. Many states that have created a state-level regulatory

infrastructure for transportation network companies have preempted local regulation

of these companies. See, e.g., Light, supra note 74, at 343 (“Uber/Lyft have success-

fully lobbied more than a dozen state legislatures to preempt all local and municipal

governance of such firms.”); id. at 376–81 (describing the approaches in various state

statutes dealing with transportation network companies to supplemental local

regulations).

233. Dobson, supra note 10, at 719–20, 722 (stating that the South Carolina legisla-

ture and public utilities commission should create regulations for “Transportation Net-

work Companies”); see also Editorial, Help New and Old Taxi Services Share the

Road,

EWSDAY

(July 4, 2015), http://www.newsday.com/opinion/editorial/uber-and-

taxis-can-coexist-we-just-need-to-let-them-1.10607020 (stating that New York State

should authorize and regulate new taxi operators). Cities are concerned about state

governments acting to address sharing economy issues. FTC Workshop, supra note

10, at 109–10 (remarks of Brooks Rainwater).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 77 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 77

federal regulation of taxis in the era of taxi apps.

234

Another second-

order issue is which agency—or agencies—should regulate taxis

within whatever level of government is deemed appropriate. There are

various candidates: special purpose taxi regulators, antitrust agencies,

and planning departments. At the moment, I set to the side these sec-

ond-order issues.

IV.

RANSITION

ELIEF FOR

RADITIONAL

EDALLION

AXI

WNERS

For taxi regulators in major cities today, the most pressing issue

after how they should regulate the new app entrants, is whether they

should compensate

235

incumbents in the traditional taxi industry who

are suffering large losses as Uber and other apps eat into their histori-

cal monopoly on the taxi business. The main actors incurring losses

due to the arrival of Uber are medallion owners, and the financial in-

stitutions that loaned money to these owners, using the medallions as

collateral.

236

As mentioned above, unlike medallion owners and finan-

234. See, e.g., Weyl & White, supra note 10; Krauss, supra note 10. But see Posner,

supra note 10. In contrast, Light stresses the need for the states and the federal gov-

ernment to avoid preempting local authority to regulate Uber and Lyft given the “un-

certainty about the potentially significant environmental impacts” of the rise of these

companies. Light, supra note 74, at 382. She argues local governments should have

scope to experiment in addressing the environmental consequences of transportation

network companies. Id. at 384–90.

235. I use the terms compensation and transition relief interchangeably. In using the

terms interchangeably, I am following Louis Kaplow, who uses the term “transitional

relief,” to encompass “compensation[,] . . . grandfathering, . . . phase-ins and delayed

implementation.” Louis Kaplow, An Economic Analysis of Legal Transitions, 99

ARV

. L. R

. 509, 514 (1986); see also id. at 582–92 (discussing “transition mecha-

nisms” which are “analytically equivalent to direct compensation”). The term “transi-

tion relief” comes from literature on “changes in tax policy.” Bruce R. Huber,

Transition Policy in Environmental Law, 35

ARV

. E

NVTL

. L. R

91, 92 n.2 (2011).

236. See, e.g., Amended Complaint at 10–14, Melrose Credit Union v. City of New

York, No. 1:15-cv-09042 (S.D.N.Y. Mar. 7, 2016); Deirdre Fernandes, Banks Step Up

Efforts to Collect on Taxi Medallion Loans,

. G

LOBE

(May 24, 2016), https://www

.bostonglobe.com/business/2016/05/24/banks-stepping-efforts-collect-taxi-medallion-

loans/QPc6vFHd5UDfN3X9dJqe8M/story.html; Danielle Furfaro, Taxi Medallion

Owners Find Their Dreams Dashed by Uber, Lyft, N

.Y. P

OST

(July 5, 2016), http://

nypost.com/2016/07/05/city-lets-uber-and-lyft-cannibalize-the-american-dream/.

In addition to medallion owners, individuals who own vehicles that have been

hacked up to comply with taxi regulations arguably have a claim to compensation.

These special purpose vehicles have less value, as Uber cars do not need to comply

with many of the requirements that regulators impose on taxis. For evidence of the

impact of Uber on vehicle owners (who do not also own medallions) in New York

City, see Verified Petition at 7, Glyka Trans LLC v. City of New York, No. 100578/

2015 (N.Y. Sup. Ct. Mar. 31, 2015) (describing a vehicle owner’s 19.3% loss in reve-

nue from sub-leasing “his taxi vehicle” from February 2014 to March 2015). The

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 78 17-APR-17 10:20

78 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

ciers, taxi drivers actually may be benefitting from the arrival of Uber

in some cities.

237

Medallions are the mechanism that many cities historically have

used to enforce limits on the number of taxis allowed to operate on

city streets. Governments distributed medallions in various ways—in-

cluding at little cost, and through lotteries and auctions.

238

Before the

advent of Uber, there were active secondary markets in many cities’

medallions in which individuals privately bought and sold medallions,

and medallion owners borrowed money from banks and credit unions

using the medallions as collateral. Many cities’ medallions were valu-

able because having one was necessary to operate a taxi, and cities

often limited their supply to a number well below that required to

serve demand for taxis. At the peak in the early 2010s, New York City

plaintiffs’ claims in Glyka Trans LLC were dismissed in Glyka Trans LLC v. City of

New York, No. 8962/15, 2015 WL 5320868 (N.Y. Sup. Ct. Sept. 8, 2015).

237. See supra notes 66–67, 191–93 and accompanying text.

238. Most New York City medallions date from the early twentieth century. “In

1937, the first [New York City] taxi medallions were sold for ten dollars each. Haas

Act § 4.” Brief for Respondents at 30, Glyka Trans LLC v. City of New York, No.

2015-11661 (N.Y. App. Div. Apr. 15, 2016). That ten dollars is the equivalent of

roughly $168 in 2016 dollars, using the Bureau of Labor Statistics CPI Inflation Cal-

culator. Since 1996, the City periodically has auctioned new medallions, garnering

millions of dollars for the City budget. Verified Petition at 20, Melrose Credit Union

v. City of New York, No. 6443/15 (N.Y. Sup. Ct. May 26, 2015) (“New York City

auctioned taxicab medallions in 1996 and 1997, realizing approximately $85 million

in revenue from the sale of 400 medallions. Between 2006 and 2013, New York City

auctioned approximately 1,050 medallions, generating an additional $486 million in

revenue. . . . Most recently, in fiscal 2014, New York City reportedly sold 400 taxicab

medallions, generating nearly $338 million in revenue. New York City has also budg-

eted for medallion sales between 2015 and 2019, anticipating $1.6 billion in revenue

for its budget.”); cf. id. at 12 (“In FY 2014, the City sold 350 of the 2,000 new taxi

medallions that were authorized by State law in 2011, generating $359 million (more

than $1 million per new medallion sold).”) (excerpting New York State Comptroller’s

Office’s March 2015 Review of the Financial Plan of the City of New York). Miami-

Dade County has used lotteries to allocate new medallions, charging the owners of the

new medallions a flat fee of $15,000, as well as auctions. Motion to Dismiss

Amended Class Action Complaint at 4, Miadeco Corp. v. Miami-Dade County, No.

1:16-cv-21976 (S.D. Fla. June 29, 2016). Milwaukee also used a lottery to distribute

new “taxicab permits” in 2013. Joe Sanfelippo Cabs, Inc. v. City of Milwaukee, 839

F.3d 613, 615 (7th Cir. 2016).

New York City has benefitted from medallion values not only through auctions,

but also because the City has collected a transfer tax of five percent on secondary

market sales of medallions since 1980. See N.Y.C. Admin. Code §§ 11-1401, 11-1402

(West 2016); Wyman, supra note 21, at 163. “The TLC establishes the fair market

value of medallions for purposes of the transfer tax based on the average sales price of

the previous month’s transfers.” Verified Petition at 20, Melrose Credit Union, No.

6443/15.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 79 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 79

medallions were selling for over $1 million.

239

Medallion values have

dropped substantially since the spread of Uber and other taxi apps,

because cars affiliated with these companies compete with incumbent

medallion taxis, but do not require medallions.

240

Medallion owners fall into different categories, and the propor-

tion of medallion owners in each category varies depending on the

jurisdiction. Some medallion owners are taxi drivers who own a single

medallion.

241

They may have immigrated to the U.S. and bought their

lone medallion after years of hard work driving taxis using leased me-

dallions and vehicles.

242

To buy that medallion, they may have bor-

239. “Taxi medallions are a transferable commodity that until recently were worth

over $1 million each (about $17 billion in the aggregate) precisely because the hail

exclusivity that the City created and sold is so valuable.” Verified Petition at 33,

Glyka Trans LLC, No. 100578/2015. The plaintiffs’ claims in this case were dis-

missed in Glyka Trans LLC, No. 8962/15, 2015 WL 5320868.

Some suggested that New York medallions were overvalued even before Uber

reduced medallion values. Felix Salmon, How the Taxi-Medallion Bubble Might

Burst,

EUTERS

(Jan. 20, 2012), http://blogs.reuters.com/felix-salmon/2012/01/20/

how-the-taxi-medallion-bubble-might-burst/ (arguing that New York City’s an-

nouncement that it will issue 2000 additional medallions could burst the medallion

bubble).

240. See Amended Complaint at 11–12, Melrose Credit Union, No. 1:15-cv-09042

(“[T]he value of the medallion has plummeted by more than 40% and the once vibrant

market for the purchase and sale of New York City taxicab medallions is now fro-

zen.”); Complaint at 20–22, CGS Taxi LLC v. City of New York, No. 653264/2015

(N.Y. Sup. Ct. Sept. 30, 2015) (describing reductions in medallion values); Complaint

at 19–21, Daler Singh v. City of New York, No. 701402/2017 (N.Y. Sup. Ct. Jan. 30,

2017) (same).

241. See, e.g., Summons at 1, 3, D&P Baidwan v. City of New York, No. 708107/

2015 (N.Y. Sup. Ct. July 31, 2015).

242. For references to immigrant medallion owners, see, for example, Complaint at

6–7, Newark Cab Ass’n v. City of Newark, No. 2:16-cv-04681 (D.N.J. Jan. 17, 2017),

2017 WL 214075 (case dismissed) (referring to an immigrant owner of two medal-

lions); Second Amended Complaint at 13, Bos. Taxi Owners Ass’n v. City of Boston,

No. 1:15-cv-10100-NMG (D. Mass. Dec. 21, 2016), 2016 WL 7410777 (case dis-

missed) (referring to an immigrant owner of one medallion); Erik Engquist, Three

More Taxi-Medallion Buyers Sue the de Blasio Administration,

RAIN

’

N.Y. B

(Aug. 4, 2015), http://www.crainsnewyork.com/article/20150804/BLOGS04/

150809956/three-more-taxi-medallion-buyers-sue-new-york-city.

I have found statistics on the immigration status of taxi drivers, but not medallion

owners. According to the most recent Taxicab Factbook issued by the New York City

Taxi and Limousine Commission, only “4% of medallion taxi drivers” in the City

“were born in the U.S.”

N.Y.C. T

AXI

& L

IMOUSINE

OMM

’

supra note 22, at 9.

According to the 2014 Taxicab Fact Book, the two largest source countries for medal-

lion taxi drivers are Bangladesh (23.1 percent of drivers) and Pakistan (13.2 percent).

N.Y.C. T

AXI

& L

IMOUSINE

OMM

’

, 2014 T

AXICAB

ACT

OOK

9 (2014)

, http://

www.nyc.gov/html/tlc/downloads/pdf/2014_tlc_factbook.pdf. Nationally, “[f]orty-

four percent of taxi and limousine drivers are foreign-born according to the U.S. Cen-

sus Bureau’s American Community Survey for November 2007, the most recent data

available. Among full-time drivers, the percentage of foreign-born rises to 53 per-

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 80 17-APR-17 10:20

80 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

rowed a lot of money from a specialized financial institution, such as

Medallion Financial or a credit union. Other owners own large num-

bers of medallions and taxi vehicles, which they lease to taxi drivers

who cannot afford to buy medallions or taxicabs. These fleet owners

may have accumulated their medallions over decades. Indeed, some of

their medallions may have been bought by family members shortly

after medallions were introduced in the early twentieth century.

243

Still other medallion owners may be professionals such as doctors

who have never driven a taxi, but nonetheless invested in medallions

that they lease to taxi drivers. For example, two of the medallion

owner plaintiffs in an ongoing lawsuit in New York City are oncolo-

gists.

244

According to the complaint in the case, one of these oncolo-

gists “is now being forced to sell his apartment and use the money in

order to attempt to refinance his medallion loan, because leasing pay-

ments . . . no longer cover the monthly payments on [this] . . .

loan.”

245

Then there are the financial institutions that loaned money to

medallion owners to finance medallion purchases, or for other pur-

poses. Medallion loans were once a safe bet, with little risk of default,

but they have since become a source of major trouble for lenders be-

cause of the competition that taxis now face from Uber. Conditions

now are so bad that in September 2015, “the New York State Depart-

ment of Financial Services announced that it had taken possession of

taxicab medallion financier Montauk Credit Union,” and the financial

condition of other credit unions that finance medallions also has dete-

riorated because medallion owners are becoming delinquent on their

loans.

246

cent.”

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note 7, at

84 n.4.

243. For discussion of the history of some of the large fleets of medallion taxis in

New York City, see Wyman, supra note 21, at 156–57 nn.152–53 (referring to news-

paper stories “about large fleet owners whose families have owned medallions since

the late 1930s and 1940s”). According to one source, “[f]ifty-eight percent of New

York City’s cabs are owned by corporate entities . . . . Just a third are owned by the

people who drive them.” Van Zuylen-Wood, supra note 7.

I have not come across a fleet owner suing New York City for compensation for

legalizing Uber. However, Gene Friedman, who “took over his father’s modest yellow

taxi business,” and “at one point owned more than 1,000 New York City taxi medal-

lions,” called for the City to bail him out after the value of his medallions declined.

Meyer, supra note 16; see also Josh Barro, New York Taxi Mogul, Seeking a Bailout,

Says He’s Too Big to Fail,

N.Y. T

IMES

(Apr. 10, 2015), http://www.nytimes.com/

2015/04/11/upshot/new-york-taxi-mogul-seeking-a-bailout-says-hes-too-big-to-

fail.html.

244. Amended Complaint at 22, Melrose Credit Union, No. 1:15-cv-09042.

245. Id.

246. Id. at 11; see also id. at 10–11, 49 (describing increase in loan delinquencies

and troubled debt restructurings at Melrose Credit Union).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 81 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 81

In New York and many other cities, medallion owners, and

sometimes the financial institutions that loaned money against medal-

lions, are bringing Takings claims, seeking compensation for the re-

cent drop in medallion values due to the legalization of Uber and the

other taxi apps.

247

I am not aware of any cases where the plaintiffs

247. See supra note 9 (citing Takings Clause cases for the loss of medallion values).

These claims raise interesting legal questions about whether taxi medallions

count as private property protected by the Takings Clause. This is an issue that has

never been adjudicated in relation to New York City medallions, although there is

recent jurisprudence from other jurisdictions indicating that other cities’ taxicab li-

censes—or the market value of these licenses—are not a private property interest

protected by the Takings Clause. See Ill. Transp. Trade Ass’n v. City of Chicago, 839

F.3d 594 (7th Cir. 2016); Joe Sanfelippo Cabs, Inc. v. City of Milwaukee, 839 F.3d

613 (7th Cir. 2016); Melancon, Inc. v. City of New Orleans, 703 F.3d 262 (5th Cir.

2012); Minneapolis Taxi Owners Coal, Inc. v. City of Minneapolis, 572 F.3d 502 (8th

Cir. 2009); Cambridge Taxi Drivers & Owners Ass’n v. City of Cambridge, No. 16-

11357-NMG, 2017 WL 373491 (D. Mass. Jan. 25, 2017); Bos. Taxi Owners Ass’n v.

Baker, No. 16-11922-NMG, 2017 WL 354010 (D. Mass. Jan. 24, 2017); Newark Cab

Ass’n v. City of Newark, No. 2:16-cv-04681, 2017 WL 214075 (D.N.J. Jan. 17,

2017); Bos. Taxi Owners Ass’n v. City of Boston, No. 15-10100-NMG, 2016 WL

1274531 (D. Mass. Mar. 31, 2016); Bos. Taxi Owners Ass’n v. City of Boston, 84 F.

Supp. 3d 72 (D. Mass. 2015); Melancon, Inc. v. City of New Orleans, No. 12-1337,

2014 WL 1117881 (E.D. La. Mar. 19, 2014); Abramyan v. State, No. 2015CV262742

(Ga. Super. Ct. Mar. 22, 2016).

There is older case law holding that Chicago medallions are protected by the

Takings Clause, but this case law seems poorly reasoned and in any event recently

was distinguished by the Illinois Transportation Trade Association court in dismissing

the taxi industry plaintiffs’ claims. Ill. Transp. Trade Ass’n v. City of Chicago, 134 F.

Supp. 3d 1108, 1111 (N.D. Ill. 2015). The district court’s holding dismissing the Tak-

ings claim was affirmed in Ill. Transp. Trade Ass’n, 839 F.3d 594 (7th Cir. 2016).

In recent litigation, New York City “concede[d]” that taxi medallions are prop-

erty for procedural Due Process purposes, but the test for property under procedural

due process is less demanding than the test for private property under the Takings

Clause. Singh v. Joshi, 152 F. Supp. 3d 112, 124 (E.D.N.Y. 2016). In Singh v. Joshi,

No. 15-CV-5496-FB-VMS, 2016 WL 4272349 (E.D.N.Y. Aug. 15, 2016), the court

denied the plaintiffs’ motion for reconsideration and granted the defendants’ motion

for summary judgment.

In some jurisdictions, medallion owners’ claims may be influenced by the exis-

tence of legislation or regulations describing medallions as property. See, for example,

53 Pa. C.S.A. 5713(a) (“Medallions are property and may not be revoked or canceled

by the authority.”), which is cited in Amended Complaint at 25, Checker Cab v. Phila.

Parking Auth., No. 2:16-cv-04669 (E.D. Pa. Nov. 4, 2016); and Miami-Dade County

Code of Ordinances § 31-81(z) (“Medallion means a plate or decal issued by the

County as the physical evidence of a taxicab license which is affixed to the outside or

inside of such taxicab.”) and § 31-81(aa) (“Medallion system means the system which

deems a taxicab for-hire license to be intangible property.”), which are cited in

Amended Class Action Complaint and Demand for a Jury Trial at 5, Miadeco Corp. v.

Miami-Dade County, No. 16-4244 (Fla. Cir. Ct. May 4, 2016). Nonetheless, these

references to medallions as property may not assure medallion owners of legal vic-

tory. The courts are not legally bound by these legislative descriptions in Takings

Clause cases, which rest on constitutional rights. Moreover, the courts could hold that

even if a medallion is property, the property rights of medallion owners do not encom-

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 82 17-APR-17 10:20

82 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

have prevailed on their Takings claims.

248

Here, I consider the policy

question of whether the owners of these medallions should be com-

pensated for reductions in medallion values since the advent of Uber

and other taxi apps that likely would be exacerbated if my proposal to

pass protection against new competitors and reductions in the market value of medal-

lions. In its motion to dismiss a suit brought by the incumbent taxi industry, Miami-

Dade County describes medallions as conferring “limited intangible personal prop-

erty” to operate for-hire vehicles. The County insists that this right has not been taken

by the legalization of transportation network companies because the property right

does not include “the right to exclude others from the for-hire market” or protection

for the market value of medallions. Motion to Dismiss Amended Class Action Com-

plaint at 13, 27, 29–30, Miadeco Corp. v. Miami-Dade County, No. 1:16-cv-21976

(S.D. Fla. June 29, 2016); see also Joe Sanfelippo Cabs, Inc., 839 F.3d at 616 (“The

taxi permits issued by the Milwaukee city government are property, but have not been

‘taken,’ as they do not confer on the holders a property right in, amounting to control

over, all transportation by taxis and taxi substitutes (such as Uber) in Milwaukee.”).

248. For a list of decisions going against the plaintiffs, see supra note 9. Consistent

with the prevailing case law, Suska recently has argued that “medallion owners lack a

doctrinally plausible Takings claim.” Suska, supra note 10, at 198. He argues that

they will have difficulty establishing that medallions constitute a property interest for

Takings purposes. Id. at 198–201. But even if medallion owners can establish that

medallions are private property protected by the Takings Clause, medallion owners

may not be able to establish that their property has been taken under the Penn Central

factors. Id. at 201–05.

There are two very interesting European cases rejecting arguments from taxi

drivers that opening entry to the taxi business deprived them of property rights by

undercutting the value of their taxicab licenses. Gorman v. Minister for the Env’t

[2001] 2 IR 414 (H. Ct.) (Ir.) held that Irish taxicab licenses are property protected by

the Irish Constitution’s equivalent of the Takings Clause. However, the court also held

that the deregulation of entry did not infringe the license owners’ property rights be-

cause there was a condition inhering in the licenses that the law about them might

change. R (Royden) v. Metro. Borough of Wirral [2002] EWHC (Admin) 2484,

[2002] All ER 256 rejected an argument from taxi drivers that the Borough’s removal

of the numerical limit on taxis deprived them of property rights protected under Arti-

cle 1 of Protocol 1 to the European Convention for the Protection of Human Rights

and Fundamental Freedoms 1950. The court ruled that the drivers’ licenses were not

protected by Article 1, and in the alternative that the removal did not breach Article 1

if it applied. Notwithstanding the holding in Gorman, the Irish government paid a

limited amount of compensation under political pressure from license holders. See

EAN

B. B

ARRETT

EGULATORY

APTURE

, P

ROPERTY

IGHTS AND

AXI

EREGULA-

TION

—A C

ASE

TUDY

RGANISATION

CONOMIC

OPERATION AND

EVEL-

OPMENT

, E

UROPEAN

ONFERENCE OF

INISTERS OF

RANSPORT

, T

RANSPORT

ESEARCH

ENTRE

, (D

EGULATION OF THE

AXI

NDUSTRY

OUND

ABLE

133,

145–46 (2007) (discussing government’s appointment of Taxi Hardship Panel after

Gorman); Ken Johnstone, Letter to the Editor, Minister of Transport and Taxi Market

Deregulation,

RISH

IMES

(Jan. 6, 2004), http://www.irishtimes.com/opinion/letters/

minister-of-transport-and-taxi-market-deregulation-1.1129023. After the economic

crisis in 2008, Ireland subsequently reinstated limits on the number of taxi licenses.

Graeme O’Meara, The Taxi Market in Ireland: To Regulate or Deregulate?,

UB-

LIC

OLICY

(Oct. 23, 2014), http://www.publicpolicy.ie/the-taxi-market-in-ireland-

to-regulate-or-deregulate/.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 83 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 83

eliminate limits on the number of taxis were implemented.

249

Such

compensation could be implemented by legislative decisions by city

councils or state governments, if the courts continue to reject claims

for just compensation under the Takings Clause.

250

In considering

whether compensation is justified, it is worth remembering that gov-

ernments do not normally compensate actors who suffer losses due to

regulatory changes.

251

249. I focus on whether medallion owners should be compensated, and do not ex-

plicitly consider whether the financial institutions that loaned money to medallion

owners should be compensated for the decline in the value of the medallions on which

the institutions rely as collateral. Without having seriously considered the matter, it

seems to me that the arguments against compensating the financial institutions are

even stronger than the arguments against compensating medallion owners, assuming

that the financial institutions were even better positioned than many medallion owners

to diversify.

For a Takings claim brought by financial institutions in New York City, see

Amended Complaint, Melrose Credit Union, No. 15-cv-09042. New York City argues

that the credit union plaintiffs lack standing in the case. Defendants’ Memorandum of

Law in Support of Defendants’ Motion to Dismiss the Amended Complaint at 5–8,

Melrose Credit Union, No. 15-cv-09042. In a separate state court case, the credit

unions are suing the City (and New York State Attorney General) for violating “taxi-

cab medallion owners’ exclusive, statutory right to accept hails” by allowing Uber and

other black car companies to accept e-hails. Brief for Petitioners-Appellants at 2, Mel-

rose Credit Union v. City of New York, No. 2016-02214 (N.Y. App. Div. May 17,

2016). The City is also arguing in this case that the credit unions lack standing and

some of the points the City makes also argue against compensating the financial insti-

tutions as a policy matter. Brief for Respondents at 31–40, Melrose Credit Union v.

City of New York, No. 2016-02214. As a legal practitioner suggested to me, an inter-

esting question to consider is whether the fact that the financial institution plaintiffs in

these two cases are credit unions—as opposed to banks—gives rise to a stronger

claim for compensation as a policy matter, given that credit unions are “not-for-profit

organizations that exist to serve their members.” What Is a Credit Union?, Nat’l

Credit Union Admin., http://www.mycreditunion.gov/about-credit-unions/Pages/

How-is-a-Credit-Union-Different-than-a-Bank.aspx (last visited Jan. 19, 2017).

250. Suska, supra note 10, at 205 (arguing for “a legislative decision to compensate

medallion owners”). A bill passed by the Massachusetts legislature did “include[ ] a

temporary 20-cent surcharge for every Uber or Lyft ride, of which 15 cents would pay

for state and local transportation improvements” and “the remaining 5 cents will gen-

erate money to help taxi firms adapt.” Dante Ramos, Opinion, The Uber War Is Over

— Uber Won,

. G

LOBE

(Aug. 5, 2016), https://www.bostonglobe.com/opinion/

2016/08/04/lessons-from-uber-war/Gq8oNWabciihO4GTx5lbnM/story.html (report-

ing that in legislating “to legitimize Uber and Lyft” Massachusetts “lawmakers looked

into measures such as a medallion buyback, but” decided against them because such

measures “could cost tens of millions of dollars or more”).

251. As Justice Holmes famously recognized in Pennsylvania Coal v. Mahon,

“[g]overnment hardly could go on if to some extent values incident to property could

not be diminished without paying for every such change in the general law.” 260 U.S.

393, 413 (1922). But see Huber, supra note 235, at 106, 110 (suggesting that govern-

ments often provide transition relief when they change environmental laws); Kaplow,

supra note 235, at 534 (“[T]here often is . . . compensation” for “government risks.”);

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 84 17-APR-17 10:20

84 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

This part argues that, in general, there is no compelling economic

argument for compensating medallion owners. However, there may be

a fairness argument for compensating some medallion owners—the

owners of a lone medallion—on the grounds that they are “less so-

phisticated investors.”

252

If regulators decide to offer compensation on

fairness grounds or as a matter of political expediency, I argue on

economic grounds that this compensation should take the form of

monetary compensation rather than continued regulatory protection

from competition. The arguments I make are general, and not specific

to any jurisdiction or its regulatory structure or history.

253

A. Whether Medallion Owners Should Be Compensated

Efficiency Analysis

The anticipated effects of compensation on the investment deci-

sions of private actors are a conventional starting point in analyses of

whether it would be efficient for governments to compensate for legal

ICHAEL

J. T

REBILCOCK

, D

EALING WITH

OSERS

: T

OLITICAL

CONOMY OF

OL-

ICY

RANSITIONS

(2014).

252. I borrow the term “less sophisticated investor” from Kaplow, supra note 235, at

549. He is contrasting “less sophisticated (often, less wealthy) individual investors”

and “institutional investors and corporations.” Id.

253. This leaves open the possibility that there might be jurisdictions where the spe-

cific regulatory history would justify compensating medallion owners.

Compensation would seem to be warranted if, as medallion owners and lenders

currently are alleging in lawsuits in New York City, the taxi regulatory body inten-

tionally posted misleading information about medallion values on its website, and the

owners and lenders justifiably relied to their detriment on this information. Intentional

misrepresentation combined with reliance might constitute fraud, and if the agency

acted fraudulently, it acted illegally and should pay compensation.

The plaintiffs are suing for fraud in Amended Complaint at 17, 74–78, 83–85,

Melrose Credit Union, No. 1:15-cv-09042; and for violation of general business law,

fraudulent inducement and negligent misrepresentation in Complaint at 2, 22, 23–26,

CGS Taxi LLC v. City of New York, No. 653264/2015 (N.Y. Sup. Ct. Sept. 30,

2015); and for violation of general business law, fraudulent inducement, breach of

contractual covenant of good faith and fair dealing, and negligent misrepresentation in

Complaint at 2, 22-25, Singh v. City of New York, No. 701402/2017 (N.Y. Sup. Ct.

Jan. 30, 2017).

For the elements of civil fraud under New York law, see Loreley Fin. (Jersey)

No. 3 Ltd. v. Wells Fargo Sec., LLC, 797 F.3d 160, 170 (2d Cir. 2015) (“Under New

York law, fraud requires proof of: (1) a material misrepresentation or omission of a

fact, (2) knowledge of that fact’s falsity, (3) an intent to induce reliance, (4) justifiable

reliance by the plaintiff, and (5) damages.”).

In the text, I set to the side any allegations that the government acted illegally,

which at any rate currently remain unproven allegations in New York City. I consider

whether government should pay compensation if all it has done is legalize Uber and

other taxi apps or permitted them to operate, and thereby reduced the value of a city’s

medallions.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 85 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 85

changes.

254

A standard argument is that governments should not gen-

erally compensate for legal changes, just as they do not generally

compensate for changes resulting from market forces, such as technol-

ogy change.

255

Government compensation for legal changes would en-

able actors to externalize some of the costs of these changes onto

society, and therefore reduce the incentives for actors to take into ac-

count the risk of such changes in making investment decisions.

256

Kaplow argues that actors should not be compensated for legal

changes, just as they generally are not compensated for market

changes, to ensure that actors “bear all real costs and benefits of their

decisions.”

257

Investors need to face all these costs and benefits to

avoid inefficient “overinvestment.”

258

This economic argument against compensation applies forcefully

to the current situation in the taxi sector, and argues against compen-

sating medallion owners. First, there is a strong argument for treating

the legal changes that have legalized Uber like “market risks,” because

the legal changes result from market forces.

259

To operate legally,

Uber and the other app-based taxi services have required government

approval, because it is generally illegal to offer rides for hire without

government permission.

260

But the advent of Uber, though immedi-

ately facilitated by government authorization, ultimately is the product

of market forces, specifically “[a]dvances in information and commu-

nication technologies, combined with smartphone applications and lo-

cation data from global positioning systems.”

261

New York State

Supreme Court Judge Allan Weiss recently recognized the technologi-

cal changes that underpin the arrival of Uber in denying a claim

254. Kaplow, supra note 235. Kaplow’s discussion of when transition relief is ap-

propriate clearly covers a situation such as the arrival of Uber. Kaplow underscores

“the ubiquity” of transition concerns, arguing that they arise whenever unanticipated

“changes in government policy . . . affect the value of investments made prior to those

changes.” Id. at 518. He identifies deregulation as an example of a policy change

raising transition issues. Id. at 517.

255. Id. at 520, 522, 533–36.

256. Id. at 533.

257. Id. at 529; see also id. at 531, 539–40.

258. Id. at 529.

259. For Kaplow’s definition of market risks, see id. at 533.

260. For the prohibition in New York City, see N.Y.C. Admin. Code §§ 19-

506(b)(1), 19-506(c)(1) (West 2016), and 35 RCNY § 58-02(a), (b). For evidence that

New York City authorized Uber to operate, see, for example, Brief for Respondents at

23, Glyka Trans LLC v. City of New York, No. 2015-11661 (N.Y. App. Div. Apr. 15,

2016) (“In 2011, the [Taxi and Limousine] Commission licensed Uber to operate a

base station in New York City.”).

261.

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note 7, at

6; see also id. at 11–13 (describing technological changes transforming

transportation).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 86 17-APR-17 10:20

86 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

brought by elements of the New York City taxi industry for compen-

sation for the decline in medallion values, holding that “[i]n this day

and age, even with public utilities, investors must always be wary of

new forms of competition arising from technological

developments.”

262

Second, medallion owners should not be compensated because

compensation would set an undesirable precedent of allowing inves-

tors to externalize onto society losses from foreseeable downside

risks. As Kaplow argues, investors need to face all of the costs—and

benefits—of their investments to promote optimal investment deci-

sions; if actors can externalize losses when asset values fall, society

risks promoting overinvestment.

The argument that actors should not be compensated for losses

due to legal changes rests in part on these changes being foreseeable

ex ante to investors.

263

If the changes are not foreseeable, then there is

no risk that compensation for the changes when they occur will affect

investment behavior ex ante.

264

As just explained, the decline in me-

dallion values is a product of legal changes that have increased the

number of taxi vehicles (the legalization of Uber), and technological

changes (the introduction of new app technology for matching of pas-

sengers and taxi drivers). Though the precise mechanism by which

technological and legal changes recently have combined to reduce the

value of medallions may not have been foreseeable, government pol-

icy changes to increase the number of taxi vehicles and technological

changes affecting the taxi industry were foreseeable ex ante.

There is a history of regulatory changes increasing the number of

vehicles providing taxi services in major cities, and this history pro-

vided medallion owners with notice of the potential that governments

might authorize additional competitors. Consider the regulatory his-

tory in New York City. While the number of medallion taxis has been

capped since 1937, regulatory changes have gradually expanded the

number of vehicles providing taxi services in the city. The legalization

of “for-hire vehicles” in the 1980s introduced new providers of trans-

portation services that compete to some degree with medallion taxis;

to be sure, prior to the arrival of Uber, the competition was muted

because these vehicles were allowed to provide transportation service

262. Glyka Trans LLC v. City of New York, No. 8962/15, 2015 WL 5320868, at

*12–13 (N.Y. Sup. Ct. Sept. 8, 2015).

263. I say that it rests in part on the changes being foreseeable, because Kaplow

argues that insurance—especially compulsory insurance—might be a better solution

to “risk misassessment” than compensation. Kaplow, supra note 235, at 549.

264. Id. at 548.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 87 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 87

only through prearrangement.

265

Of greater relevance, since 1996,

New York City periodically has increased the number of medallion

taxis, by auctioning additional medallions.

266

In addition, in 2011 and

2012, New York State passed a law that introduced new green taxis

that are allowed to pick up passengers through street hails in northern

Manhattan and the outer boroughs, in competition with medallion

taxis.

267

The introduction of the green taxis limited the geographic

scope of the medallion taxi monopoly to picking up street hails in one

part of Manhattan and at the airports. The increases in the number of

vehicles providing taxi services on city streets through governmental

licensing of for-hire vehicles, the increases in the number of medallion

taxis since 1996, and the licensing of green taxis starting in 2014,

268

mean that medallion owners should have been on notice that there

could be policy changes to increase the number of vehicles providing

taxi services.

269

265. See Wyman, supra note 21, at 172 (discussing legalization of illegal cabs as

for-hire vehicles); Complaint at 9, CGS Taxi LLC v. City of New York, No. 653264/

2015 (N.Y. Sup. Ct. Sept. 30, 2015) (“The pre-arrangement requirement limited the

competition.”); Complaint at 8, Daler Singh v. City of New York, No. 701402/2017

(N.Y. Sup. Ct. Jan. 30, 2017) (same).

266. Since 1996, New York City has increased the number of medallions by fifteen

percent. See Van Zuylen-Wood, supra note 7 (stating that there were 11,787 medal-

lions in 1996, and there are now 13,587, for an increase of about 15 percent).

267. HAIL Act, 2012 N.Y. Sess. Laws 9 (McKinney) (amending 2011 N.Y. Laws

602); Greater N.Y. Taxi Ass’n v. State, 993 N.E.2d 393, 397–98 (N.Y. 2013). The

history of the green taxis is described in Wyman, supra note 21, at 134–35, 179–82,

and Complaint at 6, CGS Taxi LLC, No. 653264/2015. In addition to being allowed to

pick up street hails in northern Manhattan and the outer boroughs, green taxis are

allowed to pick up by prearrangement in these areas and the airports. Your Guide to

Boro Taxis,

N.Y.C. T

AXI

& L

IMOUSINE

OMM

’

, http://www.nyc.gov/html/tlc/html/

passenger/shl_passenger.shtml (last visited Jan. 22, 2017).

268. Complaint at 6, CGS Taxi LLC, No. 653264/2015.

269. Recent judicial decisions reject claims from the traditional taxi industry in New

York, Boston, Chicago, Atlanta, Newark, Cambridge and Milwaukee that govern-

ments have taken private property by legalizing taxi apps. Supra note 9. Some of the

decisions emphasize that the taxi industry could not have reasonably expected that

governments would protect them from competition from taxi apps. See Ill. Transp.

Trade Ass’n v. City of Chicago, 839 F.3d 594, 596–98 (7th Cir. 2016); Bos. Taxi

Owners Ass’n v. City of Boston, 84 F. Supp. 3d 72, 79–80 (D. Mass. 2015) (denying

preliminary injunction); Glyka Trans LLC v. City of New York, No. 8962/15, 2015

WL 5320868, at *12–13 (N.Y. Sup. Ct. Sept. 8, 2015); Abramyan v. State of Georgia,

No. 2015CV262742, at 2 (Ga. Super. Ct., Mar. 22, 2016). In Joe Sanfelippo Cabs,

Inc. v. City of Milwaukee, in affirming the dismissal of a Takings claim, Judge Rich-

ard Posner explicitly states that the plaintiff taxi permit holders “were . . . on notice

that there was no guarantee that the [“no-new-permit ordinance of 1992”] . . . would

remain in force indefinitely” because when it was passed, some “aldermen warned

that the ordinance could be changed to the disadvantage of the taxi companies.” 839

F.3d 613, 613 (7th. Cir. 2016).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 88 17-APR-17 10:20

88 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

It also was foreseeable that technological change might affect the

taxi business. There is a precedent for the way that developments in

communications technology currently are reshaping the taxi business:

the spread of two-way radio technology into the taxi industry after

World War II, when it was commonplace for “households[ ]” to have

telephones.

270

With these technological developments, it was no

longer necessary to go to the street to hail a taxi; the passenger could

telephone for a taxi from the comfort of home, and the taxi company

would dispatch the taxi by two-way radio.

271

The widespread adoption of mobile phone technology is merely

the latest advance in communications technology to affect the taxi

business. No longer do we have to call taxis from home, and dispatch

centers no longer need to communicate with cabs through radios.

Many people, passengers and drivers alike, now have mobile phones

through which we can connect with each other while we are on the

street.

272

Interestingly, the potential that the spread of cell phones

might threaten New York’s medallion taxis was anticipated long

before the creation of Uber in 2009.

273

In 1998, the Sunday New York

Times Magazine published a story in which then Baruch College busi-

ness professor Ed Rogoff predicted that, “‘[i]n a few years . . . when

everyone has a cell phone and the dispatching system is fully comput-

erized, you’ll be able to press a couple of buttons on your phone and

have a nonmedallion cab appear anywhere you want in a minute or

two.’”

274

The story prophetically went on to add that, “[i]f that hap-

pens, the medallion owners might be distressed to find themselves

with pieces of plastic that are suddenly worthless.”

275

270. Darb´era, supra note 10, at 4; see also

ILBERT

& S

AMUELS

, supra note 16, at

83.

271. In Paris, the new technology meant that it was no longer necessary to physically

go to the garage nearby to book a voiture de remise. Darb´era, supra note 10, at 4.

“The voiture de remise had to wait in remises (i.e. garages) until being contracted by

a client to go and pick him up.” Id. at 3.

272. “Currently, [sixty-four] percent of Americans own smartphones, a percentage

that reflects rapid growth (from 35 percent in 2011) across all common demographic

categories (income, gender, age, and race). Among those earning less than $30,000,

50 percent owned a smartphone in 2015, compared with 43 percent in 2013.

Smartphone access, on average, varies more by age than by income: just 27 percent of

adults over age 65 have a smartphone, compared with 18 percent in 2013.”

OMM

FOR

EVIEW OF

NNOVATIVE

RBAN

OBILITY

ERVS

, supra note 7, at 143.

273. Adam Lashinsky, Uber: An Oral History,

ORTUNE

(June 3, 2015), http://for-

tune.com/2015/06/03/uber-an-oral-history/.

274. John Tierney, The Big City: You’ll Wonder Where the Yellow Went,

N.Y. T

IMES

(July 12, 1998), http://www.nytimes.com/1998/07/12/magazine/the-big-city-

you-ll-wonder-where-the-yellow-went.html.

275. Id.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 89 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 89

Government compensation for the medallion owners who are

now suffering due to foreseeable legal and technological changes

would reduce the incentives for other investors to take full account of

legal and technological risks in future investment decisions.

276

Fur-

thermore, it is especially undesirable to compensate those who over-

invest in government-created assets such as medallions because these

assets may arise from inefficient government policies.

277

Taxi medal-

lions are a paradigmatic example of an asset created by an inefficient

government policy: there is widespread agreement among economists

that limiting the number of taxis using medallions harmed society at

large, for example by reducing the quality of taxi service, and that it

was distributionally unfair because it constrained the ability of low-

skilled workers to enter an industry that has few natural barriers to

entry.

278

Compensating the owners of problematic government-cre-

ated assets such as medallions when they lose their value means that

society in effect pays twice for the inefficient policies that created the

medallions: first, through the social costs that medallions and similar

assets create when they limit the number of taxis; and second, by com-

pensating medallion owners when the medallions no longer serve the

function of limiting the number of taxis. Compensating for the elimi-

nation of these assets also may have the pernicious effect of encourag-

ing others to lobby the government to create inefficient monopoly

rights comparable to medallions that limit entry in other fields.

279

In a recent thought-provoking article, David Suska argues, con-

trary to this piece, that economic efficiency dictates that medallion

owners should be compensated for the decline in medallion values.

280

Suska offers three reasons why compensation would be efficient. Po-

tentially the most compelling is his argument that compensation “may

preserve incentives for risk-averse actors to invest in property that

may be regulated, and more generally, to deal with government.”

281

This argument echoes an argument in efficiency analyses of Takings

276. Kaplow, supra note 235, at 548, 557–58.

277. See

ARRETT

, supra note 249, at 146 (raising questions about the compensation

that Irish taxi license holders received after deregulation of entry in the early 2000s).

Kaplow also seems to gesture toward the relevance of the efficiency of the underlying

government policy in the decision about whether compensation is justified. Kaplow,

supra note 235, at 572–74.

278. Wyman, supra note 21, at 147 n.121.

279. See Geradin, supra note 8, at 10; Louis Kaplow, Transition Policy: A Concep-

tual Framework, 13 J.

ONTEMP

. L

EGAL

SSUES

161, 199 (2003) (“[P]romises of pro-

tection ex post make special interest legislation more attractive ex ante, which will

increase rent-seeking activity surrounding original enactments.”).

280. Suska, supra note 10, at 206–09.

281. Id. at 209.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 90 17-APR-17 10:20

90 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

Clause compensation that risk averse property owners should be com-

pensated for “large losses” that they suffer due to regulatory changes

if they are “unable to insure against such losses.”

282

The idea behind

the argument is that if left uncompensated, such property owners

might be discouraged from making socially beneficial investments due

to their aversion to risks that may substantially reduce their wealth,

and their inability to protect against these risks through insurance or

diversification.

283

Homeowners whose wealth is largely tied up in

their house would be a paradigmatic instance. They might be discour-

aged from making socially useful upgrades to their homes, if, for ex-

ample, the government could radically alter the zoning of their land to

preclude residential use without paying compensation, assuming

homeowners cannot buy insurance against zoning changes or diversify

their holdings.

284

Taxi medallion owners, especially if they are taxi

drivers who own a single medallion, might be analogized to individual

homeowners. Due to the reduction in medallion values, driver-owners

282. Lawrence Blume & Daniel L. Rubinfeld, Compensation for Takings: An Eco-

nomic Analysis, 72

. L. R

. 569, 573 (1984).

283. Id. at 588. A similar argument is made by Rose-Ackerman and Rossi. Susan

Rose-Ackerman & Jim Rossi, Disentangling Regulatory Takings, 86

. L. R

1435 (2000). They argue that there should be a “presumption” that the government is

not obligated to compensate due to policy changes, but they support a “presumption”

for compensation when the government acquires assets. Id. at 1440; see also id. at

1477–93. Referring to Blume and Rubinfeld, however, they argue that there might be

a case for government to compensate for policy changes when the “investors” “are

risk averse [because] . . . the uncertainty created by the threat of harm may lead them

to invest less and to hold their assets in a form that is unlikely to be affected by the

public program.” Id. at 1486; see also id. 1488 n.167 (citing Blume & Rubinfeld,

supra note 282, at 582–99).

From an ex ante perspective, Kaplow also recognizes that risk perception and the

ability to diversify also may influence the justifiability of compensation for legal

changes. He speculates that “less sophisticated (often, less wealthy) individuals” are

more likely to improperly estimate risks. Kaplow, supra note 235, at 549. He suggests

that compensation is less problematic when individuals are prone to underestimate

risks, because the compensation is unlikely to affect their investment choices. Id. at

548–49. However, he argues that insurance—especially “compulsory insurance”—

may be a more appropriate response than compensation given the likelihood that risks

will be systematically underestimated. Id. at 549. He also recognizes that “[i]n some

situations . . . certain individuals lack the information to use . . . markets to diversify

risks created by the possibility of changes in government policies.” Id. at 550.

284. Blume and Rubinfeld’s argument is focused on regulatory takings involving

real property, not intangible property like medallions. Blume & Rubinfeld, supra note

261, at 570 n.8. Rose-Ackerman and Rossi argue that “[t]he insurance rationale [for

governments to compensate for takings] is much stronger for government takings of

family homes than for actions that harm broadly-held corporations” because home-

owners are less likely to be able to “insure by holding a diversified portfolio of invest-

ments.” Rose-Ackerman & Rossi, supra note 262, at 1488–89. In addition,

corporations are more likely to be able to purchase “political risk insurance.” Id.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 91 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 91

of single medallions likely have lost a significant portion of their over-

all wealth. Moreover, they likely had little opportunity to insure

against this loss of wealth, for example by purchasing political risk

insurance or diversifying their assets.

285

Suska seems to argue for

compensating all medallion owners on the basis that medallion owners

generally share these characteristics.

286

But his argument justifies

compensating only the subset of risk-averse medallion owners that

suffer significant losses against which they could not have insured.

Medallion owners who own large numbers of medallions and profes-

sional investors are more likely to have had the opportunity to diver-

sify their asset holdings, given that they likelier have greater

resources.

Moreover, even if there is an economic argument for compensat-

ing owners of single medallions, it is not clear that the argument is

strong. Recall that the argument for compensating such risk-averse

owners rests on the idea that we do not want to discourage socially

beneficial investments. As mentioned above, limiting the number of

taxis through medallion systems was an inefficient government policy

that likely reduced service quality,

287

and investments in taxi medal-

lions consequently were socially inefficient. Compensating medallion

owners would not only pay them for socially inefficient investments,

but, as mentioned, force the public to bear the cost of undoing a policy

the costs of which it has shouldered for many years in the form of

higher fares and lack of innovative taxi service. Also, compensation in

this instance might set a precedent that could be invoked in the future

by other interest groups that lose similarly inefficient government pol-

icies that benefit them.

Suska’s other two efficiency arguments for compensation focus

on the incentives that compensation might create for governments, not

private actors.

288

He argues that compensation is desirable to facilitate

the transition to a more efficient regulatory regime that allows app-

dispatched taxis. By paying compensation, governments may be able

to reduce the opposition to changing the regulatory regime.

289

This

argument seems unavailing in many places in the United States, al-

though it might be more relevant in countries where transportation

network companies have met with greater resistance from regulators,

285. Suska, supra note 10, at 210.

286. Suska seems to generalize from the situation in Chicago where “the majority”

of medallion owners “own a single medallion.” Id.

287. See supra note 278 and accompanying text.

288. See Kaplow, supra note 279, at 164 (separately analyzing the impacts of transi-

tion relief on “private actors” and “government behavior”).

289. Suska, supra note 10, at 207–09.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 92 17-APR-17 10:20

92 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

such as in Europe.

290

In the U.S., transportation network companies

have been legalized in many jurisdictions without compensating the

incumbent taxi industry.

291

The issue now in many places is that the

regulatory burden on the traditional taxi industry remains excessively

stringent, and if the burden is not lowered, the existing industry likely

will be further eroded. Since in my view it is in the long-term interests

of the traditional industry to revisit the stringency of the regulations

that currently apply to it, there is less political need to compensate that

industry for overhauling the regulations that likely are inhibiting it,

because that industry will benefit from the overhaul relative to pre-

serving the status quo. In any event, there are strong arguments against

granting relief generally to facilitate transitions to more efficient poli-

cies by reducing the likelihood of opposition. Granting such relief it-

self may encourage inefficient behavior by regulated actors, such as

lobbying for the continuation of transition relief, that may perpetuate

inefficient government policies.

292

Suska’s other argument that governments should pay compensa-

tion is that compensation is desirable to “force government to internal-

ize the costs of its actions.”

293

The idea behind this “fiscal illusion”

argument is that governments must be forced to bear the costs of their

actions so that they recognize that policy changes have costs, as well

as benefits, and act only when the benefits exceed the costs.

294

One

difficulty with the argument that governments should compensate me-

dallion owners to avoid inefficient decisions, is that the argument fo-

cuses narrowly on ensuring that governments internalize the costs of

removing restrictions on the number of taxis, while ignoring the bene-

fits of removing the restrictions on entry. Forcing the government to

bear the costs, but not enabling it to enjoy the benefits, may lead deci-

sion-makers to over-emphasize the costs compared with the benefits.

A more fundamental objection to the argument is that government de-

cision-makers are primarily motivated by political considerations,

which do not necessarily align with economic costs and benefits.

295

Thus even if we want to incentivize governments to make efficient

290. Geradin, supra note 8.

291. See supra note 38 and accompanying text.

292. See, e.g., Kaplow, supra note 279, at 197–200 (doubting the public choice ar-

guments for transition relief); Revesz & Kong, supra note 50, at 1621–32 (rejecting

the public choice arguments for transition relief).

293. Suska, supra note 10, at 206.

294. See, e.g., Kaplow, supra note 279, at 194 (discussing the fiscal illusion argu-

ment for government compensation for legal changes).

295. Daryl J. Levinson, Making Government Pay: Markets, Politics, and the Alloca-

tion of Constitutional Costs, 67 U.

. L. R

. 345, 345 (2000).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 93 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 93

decisions, it is not clear that forcing governments to pay for the eco-

nomic costs of their actions will increase the likelihood of efficient

policy-making. Indeed, given political realities, it is possible that if

compensation is required, the diminishing number of U.S. jurisdic-

tions that have yet to legalize e-hailing could be further delayed in

doing so. Policy-makers might be deterred from legalizing e-hailing

because they fear having to raise taxi fares or taxes, or cut spending, to

pay compensation to medallion owners.

296

Fairness Arguments

If there is a case for compensating medallion owners, it likely

rests on ideas about fairness rather than efficiency. Below I elaborate

and reject a fairness argument that medallion owners are advancing

for compensation based on their reliance on existing laws. Then I turn

to a distributional argument that might justify compensating some me-

dallion owners.

Reliance Argument

Actors seeking government compensation for legal change often

argue that they are entitled to it because they invested relying on the

pre-existing legal framework.

297

Medallion owners currently are mak-

ing such an argument. They assert that governments should compen-

sate medallion owners because they invested in medallions relying on

the stability of the regulatory framework, especially the long-existing

legal restrictions on entering the taxi business that gave rise to medal-

lion values.

298

In legalizing Uber, medallion owners argue, govern-

296. As previously noted, in legislating “to legitimize Uber and Lyft,” Massachusetts

“lawmakers looked into measures such as a medallion buyback, but” decided against

this because “that could cost tens of millions of dollars or more.” Ramos, supra note

250; see also Huber, supra note 235, at 94 (noting that “lawmakers may be less

willing to pursue needed regulatory change if law or custom comes to require relief”).

297. See, e.g., Huber, supra note 235, at 107–108 (discussing arguments for transi-

tion relief rooted in the idea that it is unfair for governments to change laws on which

people have relied, without compensation).

298. See, e.g., Amended Complaint at 5–6, Melrose Credit Union v. City of New

York, No. 1:15-cv-09042 (S.D.N.Y. Mar. 7, 2016) (noting that “medallion purchasers

. . . reasonably relied” on “the statutory right to hail exclusivity that accompanies”

medallions “in making the decision to purchase a medallion and operate their private

businesses”); Complaint at 10, Newark Cab Ass’n v. City of Newark, No. 2:16-cv-

04681 (D.N.J. Jan. 17, 2017), 2017 WL 214075 (case dismissed) (claiming that me-

dallion owners bought medallions “[i]n reliance” on the regulatory “structure”); Reply

Brief For Petitioner-Appellants at 27, Glyka Trans LLC v. City of New York, No.

2015-11661 (N.Y. App. Div. May 13, 2016) (noting that “the City itself sold the

medallions, taking in nearly $1 billion based on medallion purchasers’ reliance on the

promise that they had the exclusive right to accept street hails”).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 94 17-APR-17 10:20

94 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

ments are abandoning their longstanding policy of limiting entry to the

taxi business to entities holding medallions, and undermining the

value of investments predicated on regulatory restrictions on entry.

Governments are arguably engaging in “deregulatory takings”—re-

ducing, and potentially eliminating, the value of medallions by remov-

ing restrictions on entry, or not enforcing them against Uber and other

taxi app services.

299

From the point of view of medallion owners, such

takings are unfair “and tantamount to the government suddenly chang-

ing the terms of the contract”

300

because investors “ma[de] investment

decisions . . . based on the expectation that the law will not

change.”

301

Although medallion owners invested relying on municipal regu-

latory restrictions on entry, there is a counter-argument that fairness

does not require compensating them for the disappearance of their mo-

nopoly. While they may have relied on the regulatory limits on entry,

“it is [not] reasonable to expect laws never to change.”

302

Indeed, as

mentioned above, there is a history of regulatory changes in major

cities such as New York City to increase the availability of taxi ser-

vices, and this history provided medallion owners with notice of the

New York Supreme Court Judge Allan Weiss rejected a reliance-based claim for

compensation under the Takings Clause from elements of New York City’s traditional

taxi industry. Glyka Trans LLC v. City of New York, No. 8962/15, 2015 WL

5320868 (N.Y. Sup. Ct. Sept. 8, 2015).

299. For discussion of deregulatory takings, see, for example, Rose-Ackerman &

Rossi, supra note 283, at 1459 (using the term “ ‘deregulatory takings’ challenges” to

refer to claims by regulated actors that changes in government policies have caused

them a loss in revenues and “interference with investment-backed expectations”). For

example, when the federal government and a number of states restructured the elec-

tricity sector to introduce greater competition in the 1990s and early 2000s, electric

utilities argued that they should be compensated for their “stranded costs” through

“transition surcharges” paid by “consumers or new market participants.” Id. at 1436;

see also id. at 1457–61. Rose-Ackerman and Rossi argue that there should be a “pre-

sumption” that the government is not obligated to compensate due to policy changes,

but they support a “presumption” for compensation when the government acquires

assets. Id. at 1440; see also id. at 1477–93. As of when Rose-Ackerman and Rossi

were writing, “no court ha[d] . . . accepted the sweeping deregulatory takings argu-

ment advocated by the [electricity] industry,” but regulators allowed recovery of

“stranded costs” through other means. Id. at 1466; see also id. at 1467–68.

300. Revesz & Kong, supra note 50, at 1587 (referring to Graetz’s critiques of “the

reliance argument used by old-view scholars”). Newark medallion plaintiffs specifi-

cally argued that “[b]y permitting the de facto taxi companies [such as Uber] to oper-

ate without fully complying with City Taxi Regulations, the City has breached its

contract with the Medallion Owner Plaintiffs.” Complaint at 16, Newark Cab Ass’n,

No. 2:16-cv-04681 (case dismissed).

301. Revesz & Kong, supra note 50, at 1585 (summarizing “the old view” in favor

of transition relief).

302. Kaplow, supra note 235, at 522; see also Revesz & Kong, supra note 50, at

1587–88.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 95 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 95

potential for additional regulatory changes.

303

To use New York City

as an example, medallion owners might respond that while the City

had permitted increases in the number of vehicles providing taxi ser-

vices since 1937, it had never abandoned the policy of limiting the

number of taxis allowed to pick up people by street hails until the

advent of Uber.

304

The difficulty with this argument is that New York

City, like other cities, always might have abandoned limits on the

number of taxis that can accept street hails, not merely expanded the

number of vehicles providing taxi services. Indeed, a number of cities

in the U.S. and abroad experimented with removing limits on the num-

ber of entrants to the taxi industry long before Uber.

305

Thus, on the facts, the argument of medallion owners that they

reasonably relied on longtime restrictions on entry to the taxi business

when investing in medallions seems weak. But even if the facts

pointed in the opposite direction, that would not resolve the normative

question of whether medallion owners should be compensated. As

Kaplow argues, reliance-based arguments for compensation for legal

changes beg the “normative question” of why the expectations of in-

vestors of no change should be honored with compensation in the first

303. See supra notes 265–268 and accompanying text.

304. See Complaint at 4, Singh v. City of New York, No. 701402/2017 (N.Y. Sup.

Ct. Jan. 30, 2017); Complaint at 4, CGS Taxi LLC v. City of New York, No. 653264/

2015 (N.Y. Sup. Ct. Sept. 30, 2015); Brief for Petitioner-Appellants at 46, Glyka

Trans LLC v. City of New York, No. 2015-11661 (N.Y. App. Div. May 13, 2016)

(“The basic quid pro quo in the industry is that yellow taxis are subject to medallion,

fare limit, and other onerous restrictions in exchange for the exclusive right to pick up

passengers immediately, whereas black cars are free from these restrictions but are

only permitted to arrange to pick up passengers in the future. Appellants could have

anticipated that conventional hailing would be replaced by e-hailing, but they could

not have anticipated that Respondents would upend this longstanding industry quid

pro quo and render yellow taxi hail exclusivity all but valueless by allowing everyone,

with or without a medallion, to pick up e-hails.”); Amended Complaint at 3, 7, Mel-

rose Credit Union v. City of New York, No. 1:15-cv-09042 (S.D.N.Y. Mar. 7, 2016)

(“From its inception, the regulatory structure of the for-hire transportation industry in

New York City has been based on ‘hail’ exclusivity. In exchange for the price paid to

New York City for a taxicab medallion . . . medallion bearing taxicabs were granted

the exclusive right to accept hails, while all other for-hire vehicles (‘FHVs’) . . .

operating in New York City, subject to certain limited statutory exceptions, were re-

stricted to accepting passengers solely by pre-arrangement. . . . [T]he now ubiquitous

acceptance of on-demand E-Hails by every category of FHV operating in New York

City, . . . has eviscerated hail exclusivity and destroyed the carefully designed market-

place dichotomy.”).

305. See, e.g., Wyman, supra note 21, at 147–48 n.121 (discussing efforts to deregu-

late the taxi industry in the U.S. and other countries);

OMM

FOR

EVIEW OF

NNOVA-

TIVE

RBAN

OBILITY

ERVS

, supra note 7, at 49–51 (providing a generally negative

account of experiments with taxi deregulation in the U.S.).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 96 17-APR-17 10:20

96 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

place.

306

There are powerful arguments against honoring investor ex-

pectations that laws will never change by compensating owners, be-

cause compensating for changes likely will incentivize owners not to

anticipate legal changes.

307

Distributional Argument

A potentially more compelling fairness argument for compensat-

ing medallion owners draws on the distributional implications of the

decline in medallion values. Medallion owners might argue that they

should be compensated because they (or at least some of them) are

“less sophisticated investors.”

308

Remember that some medallion own-

ers are taxi drivers who drove taxis for years to save money to buy a

single medallion, and took out a large loan to buy the medallion,

maybe when the government auctioned off a new batch, as it did in

New York in 2014. Even if it may be inefficient to compensate these

medallion owners because of the incentives that it may create, is it

really fair to make buyers like these bear the costs of the precipitous

decline in medallion values?

309

They were operating in a world where

governments required taxi drivers to own or lease a medallion, and it

seemed more profitable to be an owner than a lessee.

310

Owners like

these likely have little wealth or ability to diversify, and they invested

large amounts of their own and borrowed money into assets whose

riskiness they mistakenly underestimated. It might seem fairer to shift

the risk of the decline in medallion values to the local or state govern-

ments, which can redistribute this loss to taxpayers generally, each of

whom would pay only a small portion of the loss. The intuition that

governments should absorb the losses of driver-owners may be

306. Kaplow, supra note 235, at 524. Kaplow similarly faults arguments that no

compensation is owed because laws often change, as begging “the normative ques-

tion” of whether compensation is due. Id. at 525.

307. Kaplow, supra note 235, at 513.

308. As noted above, I borrow the term from Kaplow, supra note 235, at 549.

309. On the other hand, New York City insists that “those who bid for medallions

sold at auction must acknowledge in writing that medallion values and Commission

regulations may change.” Brief for Respondents at 71, Glyka Trans LLC, No. 2015-

11661.

310. Medallion owners are arguing that the fact that governments required medal-

lions to operate taxis supports the case for compensation. See, e.g., Complaint at 8,

Newark Cab Ass’n v. City of Newark, No. 2:16-cv-04681 (D.N.J. Jan. 17, 2017),

2017 WL 214075 (case dismissed) (“Principles of fundamental fairness also include

the constitutional right to just compensation when the government takes private prop-

erty. That right is particularly important where, as here, the government itself created

the property, sold it to private parties and developed a system under which hundreds

of private parties were induced and required by government to invest and risk hun-

dreds of millions of dollars as a precondition to engaging lawfully in business.”).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 97 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 97

stronger if the owners bought their medallions from the local govern-

ment in an auction that the government held just before Uber ex-

panded, to fund the city budget.

311

One difficulty with this fairness argument for compensating me-

dallion owners is that not all medallion owners are best regarded as

“less sophisticated investors.” While the taxi driver-owners of a lone

medallion might fit this description, fleet owners and professional in-

vestors are not easily categorized as “less sophisticated investors.” All

of them likely had the resources to diversify their assets. At most, the

argument justifies compensating medallion owners who fit the profile

of “less sophisticated investor.” It would be over-inclusive to compen-

sate all medallion owners because some were “less sophisticated in-

vestors.”

312

In New York City, there might be approximately 4,500

medallion owners in that category, if we assume that the number of

medallion owners who themselves drive taxis is a good proxy.

313

Compensating some medallion owners on the grounds that they

are “less sophisticated investors” would require delineating what that

term means in this context, and sorting medallion owners using that

definition. This likely would be an administratively and politically

complex exercise. It also might raise legal claims, as owners judged

ineligible for compensation might challenge the determination, for ex-

311. Medallion owners are arguing that the fact that New York City benefitted finan-

cially from the sale of medallions supports a case for compensation. See, e.g., Brief

for Petitioner-Appellants at 42, Glyka Trans LLC, No. 2015-11661 (“The character of

the City’s conduct . . . strongly supports the conclusion that it acted unconstitution-

ally. The City purposefully enriched itself, raising nearly $1 billion for its coffers by

creating a marketplace and selling medallion owners the exclusive right to pick up

hails.”); see also id. at 47 (“The fact that Respondents pocketed Appellants’ money in

exchange for the hail exclusivity they promised, and then turned around an[d] allowed

tens of thousands of non-medallion black cars to pick up passengers immediately

through e-hail apps, strongly supports Appellants’ position under the ‘character of the

government action’ prong of the Penn Central test.”).

As noted earlier, the Irish government paid limited compensation to taxi license

holders after entry into the taxi industry was opened in the 2000s. See supra note 248.

312. For literature discussing the over- and under-inclusion problems involved in

addressing concerns about the distribution of income outside the tax and transfer sys-

tem, see, for example,

OUIS

APLOW

& S

TEVEN

HAVELL

, F

AIRNESS

ERSUS

EL-

FARE

34 (2002); Ronen Avraham & Kyle D. Logue, Redistributing Optimally: Of Tax

Rules, Legal Rules, and Insurance, 56

L. R

. 157, 182–88 (2003); Lee A. Fen-

nell & Richard McAdams, The Distributive Deficit in Law and Economics 10–11

(Univ. of Chi. Law Sch. Coase-Sandor Inst. for Law & Econ., Working Paper No. 713

2d Series; Pub. Law & Legal Theory Working Paper No. 498, 2015).

See also Kaplow, supra note 235, at 580 (“[E]quality thus provides a dubious

foundation for a fairness justification of transitional relief.”).

313. Van Zuylen-Wood, supra note 7 (“There are 13,587 yellow cabs in New

York. . . . Fifty-eight percent of New York City’s cabs are owned by corporate entities

. . . . Just a third are owned by the people who drive them.”).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 98 17-APR-17 10:20

98 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

ample on equal protection grounds. Moreover, one might argue

against compensating some medallion owners because they are “less

sophisticated investors” because it would do little to promote distribu-

tive justice across society overall. We might be especially concerned

that compensating medallion owners would be regressive if the funds

used to compensate would be raised from taxing lower-income indi-

viduals, or diverted from government programs that otherwise assist

lower-income individuals. In addition, if we compensate some medal-

lion owners out of a sense of fairness, we are likely to face calls from

other groups harmed by government policy changes to compensate

them. It is not clear which principle governments might use to limit

future compensation obligations if the basis for compensating medal-

lion owners is that governments are better positioned to bear the

losses, as this will be true in many cases in which investors suffer

financial losses.

314

B. The Form of Any Transition Relief

Assume that a municipal or state government that legalizes Uber

and other taxi apps decides to provide transition relief for fairness or

political reasons to some or all medallion owners. What form should

transition relief take?

There are two main options: monetary compensation, or contin-

ued regulatory protection.

315

Monetary compensation could be keyed

to the harm that medallion owners have incurred from Uber’s entry

(the tort measure), the benefits they expected to receive by owning

medallions (the contract measure), or the benefits that municipal gov-

ernments gained by violating the expectations of medallion owners

and legalizing Uber (the restitution measure).

316

Continued regulatory

protection would involve protecting the monopoly of medallion own-

ers to some degree. It could involve tightly restricting the operations

of taxi apps to preserve more of the market for taxi services for the

traditional industry, or limited restrictions on app operations. For ex-

ample, transportation network company vehicles might be allowed to

314. See also Suska, supra note 10, at 211 (identifying, and rejecting, the potential

objection to his argument for compensating medallion owners on the basis that his

argument does not “include[ ] any coherent limiting principle”).

315. For typologies of transition relief, see Kaplow, supra note 235, at 584–92; Hu-

ber, supra note 235, at 95–107; Jonathan R. Nash, The Cathedral of Transition Relief

in Environmental Law 5–6 (May 17, 2011) (unpublished manuscript) (on file with

author) (distinguishing liability rule protection, which is monetary compensation, and

property rule protection, which is grandfathering).

316. Nash refers to the first two possible measures of compensation. Nash, supra

note 315, at 3.

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 99 17-APR-17 10:20

2017] TAXI REGULATION IN THE AGE OF UBER 99

pick up passengers using the apps, but picking up passengers through

street hails and at taxi stands could be reserved to medallion taxis to

preserve some of the value of medallions. Indeed, many jurisdictions

that have legalized transportation network vehicles currently prohibit

them from picking up street hails,

317

perhaps in an effort to preserve

some market share for traditional taxis.

From a social welfare perspective, any transition relief for medal-

lion owners should take the form of monetary compensation, not con-

tinued regulatory protection.

318

Continued regulatory protections will

inefficiently distort the provision of taxi services going forward, re-

sulting in higher prices or poorer services for consumers, or both.

Monetary compensation presumably will involve a one-time payout to

medallion owners.

ONCLUSION

Until the arrival of Uber, the taxi business had remained funda-

mentally unchanged for decades. Many cities limited entry, and im-

posed similar suites of regulatory requirements concerning fares,

health and safety, and universal service. Medallion owners and the

financiers, brokers, and agents that supported them were the principal

beneficiaries of this system. Consumers and taxi drivers suffered, as

limited entry restricted competition and created a barrier to entering

the industry. Uber and other taxi apps have revolutionized the taxi

business by introducing new technology and new business models.

The principal challenge facing taxi regulators is to rethink the regula-

tory framework governing the industry to catch up with developments

in the marketplace.

The taxi industry and taxi regulators largely avoided the deregu-

lation that occurred in major transportation and telecommunications

industries, such as the airline industry, starting in the late 1970s. The

technological and business model changes introduced by Uber and its

competitors now provide the opportunity to remove significant com-

ponents of the antiquated regulatory framework governing the taxi in-

dustry. Though some degree of regulation is still justified, there is

significant room to reduce the scope of regulation of the traditional

industry, and regulators should not seek to impose unnecessarily bur-

densome requirements on the new entrants.

317. See supra note 58.

318. See Kaplow, supra note 235, at 582 (“[T]ransition mechanisms tend to be even

less efficient when they differ from compensation.”).

\\jciprod01\productn\N\NYL\20-1\NYL101.txt unknown Seq: 100 17-APR-17 10:20

100 LEGISLATION AND PUBLIC POLICY [Vol. 20:1

The deregulation of the telecommunications industry in the 1980s

and 1990s was followed by a major technological revolution—the de-

velopment of mobile telephone technology—that was not anticipated

by the deregulators. Deregulation of the taxi industry is now feasible

on a broad scale due in part to the development of technological ad-

vances, including wireless technology introduced in the deregulated

telecom sector. In turn, deregulation of the taxi industry could help to

promote another wave of technological change in the transportation

sector: the widespread adoption of fleets of driverless cars that con-

sumers can rent out as required, instead of buying their own cars.

There is much speculation that driverless cars are the future of ground

transportation.

319

Removing regulatory barriers to entering the taxi

business such as medallion requirements may help to facilitate the im-

plementation of this new mode of transportation. Certainly Uber and

Lyft regard their business models as but a step along the path toward

introducing fleets of driverless cars.

320

319. See, e.g., Jeffery B. Greenblatt & Samveg Saxena, Autonomous Taxis Could

Greatly Reduce Greenhouse-Gas Emissions of US Light-Duty Vehicles, 5

ATURE

LIMATE

HANGE

860 (2015); Rachel Abrams, Self-Driving Cars May Get Here

Before We’re Ready,

N.Y. T

IMES

(Jan. 21, 2016), http://www.nytimes.com/2016/01/

22/business/dealbook/davos-self-driving-cars-may-get-here-before-were-ready.html;

Mike Ramsey & Gautham Nagesh, GM, Lyft to Test Self-Driving Electric Taxis,

ALL

. J.

(May 5, 2016), http://www.wsj.com/articles/gm-lyft-to-test-self-driving-

electric-taxis-1462460094. Some argue that driverless trucks will be deployed com-

mercially sooner than driverless cars. See, e.g., John Markoff, Want to Buy a Self-

Driving Car? Trucks May Come First,

N.Y. T

IMES

(May 17, 2016), http://

www.nytimes.com/2016/05/17/technology/want-to-buy-a-self-driving-car-trucks-

may-come-first.html. Others are skeptical of the likelihood that driverless cars will be

adopted in the near future. See, e.g., Donald N. Dewees, Are Automated Vehicles

Coming at the Right Speed? (Univ. of Toronto Dep’t of Econ., Working Paper No.

564, 2016), https://www.economics.utoronto.ca/public/workingPapers/tecipa-564.pdf

320. See

FTC

Workshop, supra note 10, at 99–101 (remarks of Catherine Sandoval)

(discussing the potential that transportation network companies “will seek to use au-

tonomous vehicles”); Alex Hern, Are Driverless Cars the Future of Uber?,

UARDIAN

(Feb. 3, 2015), http://www.theguardian.com/technology/2015/feb/03/are-

driverless-cars-the-future-of-uber; Mike Isaac, General Motors, Gazing at Future, In-

vests $500 Million in Lyft,

N.Y. T

IMES

(Jan. 4, 2016), http://www.nytimes.com/2016/

01/05/technology/gm-invests-in-lyft.html; Cecilia Kang, No Driver? Bring It On.

How Pittsburgh Became Uber’s Testing Ground, N.Y.

IMES

(Sept. 10, 2016), http://

www.nytimes.com/2016/09/11/technology/no-driver-bring-it-on-how-pittsburgh-be

came-ubers-testing-ground.html; Uberworld: The World’s Most Valuable Startup is

Leading the Race to Transform the Future of Transport,

CONOMIST

(Sept. 3,

2016), http://www.economist.com/news/leaders/21706258-worlds-most-valuable-start

up-leading-race-transform-future; Amy X. Wang, Self-Driving Cars Could Spell the

End of the Taxi Industry. Is That A Good Thing?,

LATE

(July 7, 2015), http://

www.slate.com/blogs/future_tense/2015/07/07/self_driving_cars_a_new_department_

of_energy_sponsored_report_finds_self.html.