Estimating the Earnings Loss Associated with a Criminal Record and Suspended Driverâ•Žs License

Santa Clara Law Santa Clara Law

Santa Clara Law Digital Commons Santa Clara Law Digital Commons

Faculty Publications Faculty Scholarship

3-7-2022

Estimating the Earnings Loss Associated with a Criminal Record Estimating the Earnings Loss Associated with a Criminal Record

and Suspended Driver’s License and Suspended Driver’s License

Colleen Chien

Santa Clara University School of Law

, [email protected]

Alexandra George

Srihari Shekhar

Santa Clara University

Robert Apel

Rutgers University - Newark

Follow this and additional works at: https://digitalcommons.law.scu.edu/facpubs

Part of the Law Commons

Automated Citation Automated Citation

Colleen Chien, Alexandra George, Srihari Shekhar, and Robert Apel,

Estimating the Earnings Loss

Associated with a Criminal Record and Suspended Driver’s License

ARIZ. L. REV. (2022),

Available at: https://digitalcommons.law.scu.edu/facpubs/994

This Article is brought to you for free and open access by the Faculty Scholarship at Santa Clara Law Digital

Commons. It has been accepted for inclusion in Faculty Publications by an authorized administrator of Santa Clara

Law Digital Commons. For more information, please contact [email protected].

Estimating the Earnings Loss Associated with a Criminal

Record and Suspended Driver’s License

Colleen Chien, Alexandra George, Srihari Shekhar, and Robert Apel

As states pass reforms to reduce the size of their prison populations, the number of Americans

physically incarcerated has declined. However, the number of people whose employment and

related opportunities are limited due to their criminal records continues to grow. Another

sanction is the loss of one's driver's licenses for reasons unrelated to driving. While many states

have laws on the books to redress these harms, a growing body of research has documented large

“second chance gaps” between eligibility and delivery of expungement and restored license

relief due to their poor administration. This paper is a first attempt to measure the cost of these

“paper prisons” of limited economic opportunity, in terms of annual lost earnings. Analyzing the

literature, we estimate the annual earnings loss associated with misdemeanor and felony

convictions to be $5,100 and $6,400, respectively, and that of a suspended license to be $12,700.

We use Texas as a case study for comparing the cost of “paper prisons” with the cost of physical

prisons. In Texas, individuals with criminal convictions may seal their records after a waiting

period and people that have lost driver’s licenses may get restored occupational driver’s licenses

to drive to work or school. Analyzing administrative data, this paper finds that approximately

5% of people eligible for relief have had their records sealed. The 670K people in the second

chance sealing gap translates to an annual earnings loss of about $3.5 billion annually. Using a

similar approach, we find that about 20% of the people that appear eligible for occupational

drivers’ licenses (ODLs) in Texas have received them, leaving about 430,000 people who could

have them without licenses; an earnings loss of about $5.5 billion. Based on these figures, we

find the cumulative annual earnings loss associated with Texas’s “paper prisons” to be

comparable with, and likely more than, the yearly cost to Texas of managing its physical prisons,

of around $3.6 billion.

graduate from Santa Clara University with degrees in Philosophy and Political Science, Srihari Shekhar is a graduate

student in the Masters of Information Science program at SCU, and Robert Apel is a criminologist and Professor at

Rutgers University-Newark. Corresponding author: Colleen Chien, [email protected]

Electronic copy available at: https://ssrn.com/abstract=4065920

TABLE OF CONTENTS

INTRODUCTION 3

PART I: OVERVIEW OF TEXAS’ RECORD SEALING AND LICENSE REINSTATEMENT LAWS 6

Record Relief: Order of Non-Disclosure 8

Driver’s License Restoration: Occupational Driver’s License 11

Barriers to Relief Under Texas’s Second Chance Laws 13

PART II: ESTIMATING THE SIZE OF TEXAS’S SEALING AND DRIVER’S LICENSE

RESTORATION SECOND CHANCE GAPS 14

The Texas Criminal Population That Could Benefit from Sealing Relief 15

Table 1: The Population of People in Texas with Convictions 17

The Texas Population That Could Benefit from License Restoration Relief 18

Table A: Demographic Information About Population With Suspended Driver’s Licenses 20

Sizing Texas’s Second Chance Sealing Gap 21

Table 2: Estimated Eligibility and Uptake of Texas Record Sealing and Drivers Restoration 22

Sizing Texas’s Second Chance Driver’s License Restoration Gap 22

PART III: ESTIMATING THE ANNUAL EARNINGS IMPACTS OF A CRIMINAL CONVICTION

AND LOST DRIVER’S LICENSE AND THE EARNING IMPACTS OF TEXAS’ SECOND CHANCE

GAPS 25

Summary of Results 25

Table 3: The Size and Annual Earnings Losses Associated with Texas’s Second 25

Chance Sealing and Driver’s License Gaps 25

Estimating The Earnings Effect of Incarceration and Conviction 26

Table B. Estimates of Lost Annual Earnings from Conviction and Incarceration 29

Estimating the Earnings Impact of the Texas Second Chance Expungement (Sealing) Gap 30

The Earnings Effect of a Suspended Driver’s License 31

Comparing the Cumulative Earnings Effect of Texas’s Paper Prisons with the Out-of-Pocket Cost of

Texas’s Physical Prisons 34

PART IV: AUTOMATION AND POLICY PILOTS 34

CONCLUSION 40

APPENDIX 41

Sizing the Texas Sealing Second Chance Gap 41

Electronic copy available at: https://ssrn.com/abstract=4065920

INTRODUCTION

In 2002, Christian Watts was charged with felony drug possession after purchasing

MDMA (ecstasy) from a friend.

Watts pleaded guilty to a lower misdemeanor charge, and the

judge sentenced him to three months of house arrest and 36 months of probation.

At an average

cost of $9.17 per day, Watts’s supervision cost federal taxpayers $9,904.

But the consequences

of his conviction did not end there. Since Watts completed his sentence, he has been denied

civilian and military employment opportunities due to his record.

Despite earning Associates

and Bachelors degrees, and the praise of a judge for his self-rehabilitation efforts,

Watts has

been able to find work only as a dog walker and cross-fit trainer, jobs with average annual

incomes of around $35,000.

Commenting that “my life is stuck in a standstill,”

Watts has

abandoned plans to become a lawyer, a profession with an average annual income of about

$90,000.

These figures imply a potential earnings gap of $55,000 per year, and associated with

it, a gap in productivity, skills, and tax revenue.

In 2002, Demetrice Moore, a Certified Nursing Assistant, was convicted of grand larceny

and sentenced to jail and costs, including the cost of the lawyer appointed to represent her

because she was indigent.

Moore served her time, but could not repay the costs.

As a result,

her driver’s license was automatically suspended.

This interfered with her work because “[a]s a

Id.

Mario Salas & Angela Ciolfi, Driven by Dollars: A State-By-State Analysis of Driver’s License Suspension Laws

for Failure to Pay Court Debts, Lᴇɢᴀʟ Aɪᴅ Jᴜsᴛɪᴄᴇ Cᴇɴᴛᴇʀ, 3, (Fall 2017),

https://www.justice4all.org/wp-content/uploads/2017/09/Driven-by-Dollars.pdf.

Prosecutor Salary in Las Vegas, NV, ZɪᴘRᴇᴄʀᴜɪᴛᴇʀ, (last visited November 5, 2021),

https://www.ziprecruiter.com/Salaries/Prosecutor-Salary-in-Las-Vegas,NV. See also U.S. ᴠ. Wᴀᴛᴛs, supra note 4

(Watts explaining that “I want to continue to further my education and have an active application into Boyd school

of law for this fall semester. I'm hoping to pursue a career as a prosecutor. Law school is costly and at 37 years old

loans of that magnitude can be daunting. I did not want this to deter me from accomplishing my goal. I came to

discover if I served my country in the military, not only can I satisfy my sense of duty, contribute to the greater

good, but the government will help me with the cost of my education. I want very much to enlist as an officer and

serve, possibly in the National Guard. However, no branch of the military will accept me with my current federal

drug conviction”).

Rʜᴏᴅᴀɴ, supra note 1.

Professional Dog Walker Salary in Las Vegas, NV, ZɪᴘRᴇᴄʀᴜɪᴛᴇʀ, (last visited November 5, 2021),

https://www.ziprecruiter.com/Salaries/Professional-Dog-Walker-Salary-in-Las-Vegas,NV and Fitness Trainer Salary

in Las Vegas, NV, ZɪᴘRᴇᴄʀᴜɪᴛᴇʀ, (last visited November 5, 2021),

https://www.ziprecruiter.com/Salaries/Fitness-Trainer-Salary-in-Las-Vegas,NV (listing average salaries at $30,830

and $37,602, respectively).

Id. and Rʜᴏᴅᴀɴ, supra note 1 (At Watts’ hearing, the judge said, “I wish I had far more people before me who

show the kind of self rehabilitation and effort that you’ve demonstrated” and even shook Watts’ hand.)

U.S. v. Watts, 2:04-CR-00146-PMP-RJJ (D. Nev. Apr. 25, 2011).

$9.17 per day x 30 days per month x 36 months = $9,903.60. For per day probation supervision costs, see

Supervision Costs Significantly Less than Incarceration in Federal System, Uɴɪᴛᴇᴅ Sᴛᴀᴛᴇs Cᴏᴜʀᴛs, (July 18, 2013),

https://www.uscourts.gov/news/2013/07/18/supervision-costs-significantly-less-incarceration-federal-system.

Id. and United States of America v. Watts. Case no: 2:04-0146-PMP-RJJ. Motion to Terminate Supervised Release.

June 3, 2009.

Maya Rhodan, Misdemeanor Conviction Is Not a Big Deal, Right? Think Again, Tɪᴍᴇ, (April 24, 2014),

https://time.com/76356/a-misdemeanor-conviction-is-not-a-big-deal-right-think-again/.

Electronic copy available at: https://ssrn.com/abstract=4065920

CNA, she had to drive extensively to care for elderly and disabled patients in their homes.”

Despite this, Moore attempted to keep working and was consequently convicted several times

and jailed for driving on a suspended license.

In the end, Moore had to “stop working as a CNA

because of the required driving.”

An average-to-experienced CNA makes around $33,000 to

$40,000 annually.

Many have commented on the massive size of the American criminal justice system and

celebrated reforms to reduce it.

But while the number of people put behind bars declines,

those who have old convictions and criminal records

continue to encounter structural barriers to

work or the deprivation of a driver’s license.

Among the freedoms curtailed by such actions,

economic liberty stands out. As the Supreme Court commented about suspended licenses in the

case of Bell v. Burson: “Once licenses are issued. . . their continued possession may become

essential in the pursuit of a livelihood. Suspension of issued licenses thus involves state action

that adjudicates important interests of the licensees.”

To remove these barriers, nearly every state has laws on the books that allow old,

generally minor, convictions to be expunged.

In many states, lost licenses can be restored in

In this article we use the word “expunged” to refer generally to the shielding of records from public view through

records remediations strategies such as sealing, orders of non-disclosure expungement, and expunction.

402 U.S. 535, 539, 1971.

See Kansas v. Glover, 589 U.S. ___ (2020) (in her concurring opinion, Justice Kagan notes that “several studies

have found that most license suspensions do not relate to driving at all; what they most relate to is being poor.” See

also Suspended/Revoked Working Group, Best Practices to Reducing Suspended Drivers, Aᴍᴇʀɪᴄᴀɴ Assᴏᴄɪᴀᴛɪᴏɴ ᴏғ

Mᴏᴛᴏʀ Vᴇʜɪᴄʟᴇs, 34-37, (February 2013),

https://www.aamva.org/Suspended-and-Revoked-Drivers-Working-Group/ (building off of the Department of

Transportation’s H.S. 811 092, “Reasons for Drivers License Suspension, Recidivism and Crash Involvement among

Suspended/Revoked Drivers,” this study estimates that, based on a sample of drivers from six states in the U.S., the

number of people who had licenses suspended for reasons unrelated to driving increased from 21% to 29% between

2002 and 2006). Some states have reported declines, however; see, e.g. Nina R. Joyce et al., Individual and

Geographic Variation in Driver’s License Suspensions: Evidence of Disparities by Race, Ethnicity, and Income, 19

Jᴏᴜʀɴᴀʟ ᴏғ Tʀᴀɴsᴘᴏʀᴛ ᴀɴᴅ Hᴇᴀʟᴛʜ 1, 4-5, 2020 (finding that, based on a random sample of about 7.6M drivers in

New Jersey, the prevalence of people with non-driving related suspensions between 2004 and 2018 decreased from

7.9% to 5%).

Becki Goggins, New Blog Series Takes Closer Look at Findings of SEARCH/BJS Survey of State Criminal History

Information Systems, 2016, Sᴇᴀʀᴄʜ, (March 29, 2018),

https://www.search.org/new-blog-series-takes-closer-look-at-findings-of-search-bjs-survey-of-state-criminal-history-

information-systems-2016/ (showing the growth in the number of subjects in state criminal history files, from ~81M

in 2006 to ~110M in 2016. These numbers, which represent biometric (fingerprint) data, contain duplicates).

John Gramlich, America’s incarceration rate falls to lowest level since 1995, Pᴇᴡ Rᴇsᴇᴀʀᴄʜ Cᴇɴᴛᴇʀ, (August 16,

2021), https://www.pewresearch.org/fact-tank/2021/08/16/americas-incarceration-rate-lowest-since-1995/ (noting

that “[a]t the end of 2019, there were just under 2.1 million people behind bars in the U.S., including 1.43 million

under the jurisdiction of federal and state prisons and roughly 735,000 in the custody of locally run jails”)

See Overcrowding and Overuse of Imprisonment in the United States, Aᴍᴇʀɪᴄᴀɴ Cɪᴠɪʟ Lɪʙᴇʀᴛɪᴇs Uɴɪᴏɴ, (May

2015), https://www.ohchr.org/Documents/Issues/RuleOfLaw/Overincarceration/ACLU.pdf (discussing the causes of

the growth in the U.S.’s incarcerated population).

Nurse Journal Staff, Certified Nursing Assistant Salary Guide, NᴜʀsᴇJᴏᴜʀɴᴀʟ, (January 20, 2022),

https://nursejournal.org/cna/salary/.

Id.

Electronic copy available at: https://ssrn.com/abstract=4065920

order to support going to work or school.

But as an emerging literature has documented, the

poor administration of “second chance” policies means that many of the people, and often, the

majority, that stand to benefit from, e.g., expunged records or restored licenses, are not receiving

the benefits of these laws. One of us has defined this difference as the “second chance gap,” and

estimated its size across a number of realms, including expungement, restoration of the right to

vote, and resentencing.

An associated project, the Paper Prisons Initiative, has documented

uptake rates of records-expungement across over a dozen states, finding rates of less than 10%

— implying that over 90% of eligible people are not taking advantage of the law —- to be

common.

The economic impacts of paper prisons are more difficult to quantify than the

out-of-pocket costs of physical prisons, but they are still consequential. Though the total

unemployment rate in January 2022 was less than 4%, the last estimate of unemployment of

formerly incarcerated people living in the United States, published in 2018, reported an

unemployment rate of “over 27% — higher than the total U.S. unemployment rate during any

historical period, including the Great Depression,”

which figure was also close to triple the

national unemployment rate at the time.

But the fact that the cost of paper prisons is largely unquantified and unknown makes it

difficult to know how much to prioritize closing the gap. It also obscures the cost of the poor

drafting of second chance laws, which, in many cases, are extremely complicated and become

applicable only after numerous conditions have been satisfied.

Efforts to pass “Clean Slate”

bills that would help narrow the second chance gap have stalled across the country due to the

See Appendix A for a description of Texas's sealing law.

Kimberly Amadeo, Unemployment Rate by Year Since 1929 Compared to Inflation and GDP, Tʜᴇ Bᴀʟᴀɴᴄᴇ, (last

updated January 28, 2022), https://www.thebalance.com/unemployment-rate-by-year-3305506 (reporting a national

unemployment rate of 10% in 2009)

Lucius Couloute & Daniel Kopf, Out of Prison & Out of Work: Unemployment Among Formerly Incarcerated

People, Pʀɪsᴏɴ Pᴏʟɪᴄʏ Iɴɪᴛɪᴀᴛɪᴠᴇ, (July 2018), https://www.prisonpolicy.org/reports/outofwork.html. (describing the

findings based on the National Former Prisoner Survey, conducted in 2008.)

Id. (noting that across a variety of second chance programs in the U.S., “ only a small fraction (less than 10

percent) of those eligible for relief actually received it”); See also J.J. Prescott and Sonja B. Starr, Expungement of

Criminal Convictions: An Empirical Study, 133 Hᴀʀᴠ. L. Rᴇᴠ. 2460, 2461 (2020) (finding that in Michigan “among

those legally eligible for expungement, just 6.5% obtain it within five years of eligibility”); Pᴀᴘᴇʀ Pʀɪsᴏɴs

Iɴɪᴛɪᴀᴛɪᴠᴇ, (last visited September 19, 2021), https://www.paperprisons.org/ (documenting the “uptake rate” of

expungement relief in states across the U.S.).

See Colleen Chien, America’s Paper Prisons: The Second Chance Gaps, 119 Mɪᴄʜ. L. Rᴇᴠ. 519, 519 (2020)

(analyzing a variety of second chance programs in the U.S. including clemency, compassionate release,

resentencing, and nonconvictions expungement, finding that in many cases “ only a small fraction (less than 10

percent) of those eligible for relief actually received it”).

See What is a hardship license vs. restricted license comparison, Iɴᴛᴏxᴀʟᴏᴄᴋ, (June 30, 2021),

https://www.intoxalock.com/blog/post/difference-between-hardship-and-restricted-license/ (describing that states

including Florida, Indiana, Wisconsin, Arkansas, and Kentucky all have“hardship licenses” that allow certain

individuals to have licenses for going to work, and California, Texas, Washington, Virginia, and Iowa have

“restricted licenses” that serve a similar purpose). Such licenses are also available to drivers in Pennsylvania

(Occupational Limited Licenses, Pᴇɴɴsʏʟᴠᴀɴɪᴀ Dᴇᴘᴀʀᴛᴍᴇɴᴛ ᴏғ Tʀᴀɴsᴘᴏʀᴛᴀᴛɪᴏɴ & Vᴇʜɪᴄʟᴇ Sᴇʀᴠɪᴄᴇs,

https://www.dmv.pa.gov/Information-Centers/Suspensions/Pages/Occupational-Limited-Licenses.aspx).

Electronic copy available at: https://ssrn.com/abstract=4065920

cost of updating records.

As the problem of paper prisons persists unabated, it is important to

quantify the economic costs of second chance gaps. This article does so by estimating the

earnings and employment consequences of old expungable convictions and lost driver’s licenses

that are available for restoration under existing law.

Part I of the article provides an overview of two of Texas’s second chance laws,

governing the grant of orders of non-disclosure (ONDs) and restoration of occupational driver’s

licenses (ODLs). It details the processes required to obtain each form of second chance relief and

explores how various administrative factors may contribute to gaps in their uptake. Part II then

describes the populations of people eligible for convictions and driver’s license relief, and our

methodology for estimating each second chance gap, measured by the number of individuals

with records who qualify for relief (“the current gap”), the share of people eligible for a given

second chance that have obtained it (“the uptake gap”), and the number of years it would take to

clear each second chance backlog based on the current pace of relief. Part III presents estimates

of the lost earnings and employment consequences associated with a criminal record and a lost

driver’s license. We then use these estimates and the findings from Part II to calculate the

earnings loss associated with Texas’s paper prisons.

We find that approximately 5% of people eligible for sealing relief have accessed it,

leaving a gap of about 670,000 people (with people in the gap having a last conviction, on

average, of 17 years ago). We estimate the earnings loss associated with this gap to be

approximately $3.5 billion annually. Using a similar approach, we find that about 20% of people

eligible for occupational drivers licenses (ODLs) in Texas have accessed them, leaving a gap of

430,000 people eligible for ODLs who have not gotten one, which translates into an earnings

loss of about $5.5 billion. Based on these figures, we find the cumulative annual earnings loss

associated with Texas’s “paper prisons” to exceed the yearly cost of funding physical prisons in

Texas, which is around $3.6 billion.

PART I: OVERVIEW OF TEXAS’ RECORD SEALING AND LICENSE

REINSTATEMENT LAWS

For our exploration of the economic impacts of the second chance gap, we used Texas as

a case study for a few reasons. First, Texas has the 9th largest economy in the world by GDP, and

prides itself on being business-friendly and a reliable source of skilled workers.

Secondly, like

many states, Texas has been under fiscal pressure to reform and reduce the costs of its criminal

https://businessintexas.com/news/texas-enters-2021-as-worlds-9th-largest-economy-by-gdp/ (describing the

benefits of Texas for businesses as including “highly competitive tax climate, world-class infrastructure, a skilled

workforce of 14 million people, business-friendly economic policies and abundant quality of life,”

See, e.g. Washington state, which in 2020 passed a “Clean Slate” bill that would have narrowed but which the

Governor vetoed due to cost,. Iin part exacerbated by the anticipated cost of COVID expenditures. Rachel M.

Cohen, Washington Governor Vetoes Bill that Would Have Automatically Cleared Criminal Records, Tʜᴇ Aᴘᴘᴇᴀʟ,

(May 19, 2020), https://theappeal.org/politicalreport/washington-governor-vetoes-clean-slate-bill/.

Electronic copy available at: https://ssrn.com/abstract=4065920

justice system and has led the country in many respects in doing so.

As part of this reform,

policymakers and politicians have widely celebrated the cost savings associated with closing

Texas’s physical prisons.

However, while the number of Texans who were incarcerated in state

prisons and jails decreased by 27% between fiscal years 2005 and 2020,

individuals with

criminal history in the state doubled over the same period, according to repository consortium

SEARCH.

There are 5 million people in Texas’s database of people with convictions, and,

before the law was reformed in 2020, 1.4 million Texans, or 5% of the population, had

suspended licenses.

As such, the extent to which people with records and suspended licenses

are integrated — or not — into the workforce has significant consequences for the Texas

economy.

At least 400,000 licenses remain suspended though the elimination of a controversial program restored 1.4M

licenses. Letter from Department of Public Safety to Karly Jo Dixon from the Texas Fair Defense Project. Letter on

file with authors. For more on the significance of the repeal of the controversial program, see Emily Gerrick & Mary

Mergler, Commentary: Lawmakers Need to fix another problem that buries Texas drivers in fines, Sᴛᴀᴛᴇsᴍᴀɴ, (last

updated July 5, 2020), https://www.austintexas.gov/edims/document.cfm?id=340640 (noting that “[we]e cannot

overstate how significant the repeal of this program is. When the law goes into effect in September, 1.4 million

license suspensions will be lifted, and nearly $2.5 billion of surcharge debt will be wiped clean. Huge numbers of

people will escape the cycle of suspensions and get back on the road driving legally. This repeal will help vulnerable

Texans achieve financial stability, save taxpayer dollars and boost the Texas economy.”)

Data from bi-annual Survey of State Criminal History Information Systems, SEARCH (2020, 2008, 2009, 2011,

2014, 2015, 2018, 2020) at Table 1. (2020 report available e.g. at

https://www.ncjrs.gov/pdffiles1/bjs/grants/255651.pdf). In 2006, the repository had 7,986,000 records (Bureau of

Justice Statistics, Survey of State Criminal History Information, 2003, U.S. Dᴇᴘᴀʀᴛᴍᴇɴᴛ ᴏғ Jᴜsᴛɪᴄᴇ, (February

2006), https://bjs.ojp.gov/content/pub/pdf/sschis03.pdf), in 2018 that number was 15,437,000 (Becki R. Goggins &

Dennis A. DeBacco, Survey of State Criminal History Information Systems, 2016: A Criminal Justice Information

Policy Report, Sᴇᴀʀᴄʜ, (February 2018), https://www.ncjrs.gov/pdffiles1/bjs/grants/251516.pdf).

From 152,213 people to 142,169 people. This excludes participants in the Substance Abuse Felony Program. See

Statistical Reports, Tᴇxᴀs Dᴇᴘᴀʀᴛᴍᴇɴᴛ ᴏғ Cʀɪᴍɪɴᴀʟ Jᴜsᴛɪᴄᴇ (2020, 2019, 2018, 2017, 2016, 2015, 2014, 2013,

2012, 2011, 2010, 2009, 2008, 2007, 2006, 2005) at 8 (all reports as of January 21, 2022 available at

https://www.tdcj.texas.gov/publications/statistical_reports.html).

In 2020, for example, the Texas Department of Criminal Justice projected that closing two Texas prisons would

“free up about $20 million in its budget.” Julie McCullogh, As the Texas prison population shrinks, the state is

closing two more lockups, Tʜᴇ Tᴇxᴀs Tʀɪʙᴜɴᴇ, (February 20, 2020),

https://www.houstonpublicmedia.org/articles/news/2020/02/21/361405/as-the-texas-prison-population-shrinks-the-st

ate-is-closing-two-more-lockups/. This followed a 2017 estimate that Texas could “eliminate more than 2,000

beds...[and] save the state some $49.5 million” from closing four prisons. Brandi Grissom, With crime,

incarceration, rates falling, Texas closes record number of prisons, Tʜᴇ Dᴀʟʟᴀs Mᴏʀɴɪɴɢ Nᴇᴡs, (July 5, 2017),

https://www.dallasnews.com/news/politics/2017/07/05/with-crime-incarceration-rates-falling-texas-closes-record-nu

mber-of-prisons/.

Michael Haugen, Ten Years of Criminal Justice Reform in Texas, Rɪɢʜᴛ ᴏɴ Cʀɪᴍᴇ, (August 1, 2017),

https://rightoncrime.com/2017/08/ten-years-of-criminal-justice-reform-in-texas/ (describing states adopting justice

reinvestment packages similar to the ones in Texas).

Electronic copy available at: https://ssrn.com/abstract=4065920

Sources: Texas Department of Criminal Justice's Annual Statistical Reports for fiscal years 2005 through 2020 (for

jail and prison population data)

; SEARCH (for criminal subjects data).

Third, over the past several decades, Texas has introduced laws that advance both

criminal justice and workforce-related objectives, allowing individuals with old convictions to

get them sealed and individuals that have lost their licenses a chance to regain their right to drive

to work or school. The scale of Texas’s criminal justice system and its adoption of many second

chance reforms, as well as the lack of attention paid to their implementation, make the state a

good subject for study and analysis.

This section provides an overview of Texas’s second chance laws, describing their

legislative history and the processes set forth by the law for obtaining relief. These laws share the

goals of advancing economic interests and removing barriers to work, as well as preserving

public safety.

A. Record Relief: Order of Non-Disclosure

In Texas, every time a person is convicted of a crime, this event is memorialized in the

Based on data obtained from Becki Goggins et al; Survey of State Criminal History Information Systems, 2020: A

Criminal Justice Information Policy Report, available at https://www.ncjrs.gov/pdffiles1/bjs/grants/255651.pdf,

Table 1 (listing the total number of records in the state repository as of Dec 2018) (2020) and previous versions of

the report, at Table 1.

Data about the size of the on hand state jail and prison population obtained from Annual Statistical Reports, Tᴇxᴀs

Dᴇᴘᴀʀᴛᴍᴇɴᴛ ᴏғ Cʀɪᴍɪɴᴀʟ Jᴜsᴛɪᴄᴇ, 8 (2020, 2019, 2018, 2017, 2016, 2015, 2014, 2013, 2012, 2011, 2010, 2009,

2009, 2007, 2006, 2005) (reports last available on January 23, 2022,

https://www.tdcj.texas.gov/publications/statistical_reports.html). Even though the TDCJ places people in jail, prison,

and Substance Abuse Felony Punishment Facility (SAFP) Program in the same “on hand” category in its annual

reports, we exclude the number of people in the SAFP Program from our estimate of the size of the state's prison and

jail population.

Electronic copy available at: https://ssrn.com/abstract=4065920

person’s criminal record, which sets off various “collateral consequences,” or civil punishments

that follow a person long after time has been served. The National Inventory of Collateral

Consequences of Conviction has cataloged over 1,600 civil sanctions in Texas alone for people

with criminal records spanning child support, employment, volunteering, civic participation, real

estate, visitation and parental rights, and housing residency.

People with certain criminal

records are disqualified from numerous top jobs for people without college degrees. “Private

security” on the Texas Labor Analysis’ list of the top 25 projected occupations for individuals

with a maximum education of postsecondary non-degree,

with wages from ~$21,000 to

~$37,000,

is one of them.

In an effort to give people a second chance and allow them to more easily find

employment, the Texas Legislature passed Texas Gov. Code Chap. 411, creating two pathways

for individuals to remove their past criminal records from public access. The first pathway

provides relief via an expunction, which occurs when all “information about an arrest, charge, or

conviction [is removed] from [one’]s permanent records.”

Expunction is only available for

felony and Class A, B, and C misdemeanor nonconvictions.

In contrast, the second pathway, an

“Order of Nondisclosure” (OND), seals records from the general public, while allowing certain

employers and government agencies to “see through” the OND.

Sealing via OND is available

to people convicted of first-time low-level misdemeanor convictions as well as to those who

completed deferred adjudication community supervision (“deferred adjudication”) for low-level

offenses. This article focuses only on sealing of convictions because the earnings impact of a

conviction is recognized to be much more significant than a non-conviction.

As such, we will

use “sealing” throughout the remainder of the article to refer exclusively to record relief granted

using the OND pathway.

In 2003, S.B. 1477 put the OND pathway on the books in Texas,

in an effort to remove

Id.

See Chien, supra note ___ at ___. (providing an overview of this impact literature) [pin/add’l parenthetical

needed]

Expunctions in Texas, Tᴇxᴀs Yᴏᴜɴɢ Lᴀᴡʏᴇʀs Assᴏᴄɪᴀᴛɪᴏɴ & ᴛʜᴇ Sᴛᴀᴛᴇ Bᴀʀ ᴏғ Tᴇxᴀs, 5, (2010 and 2019),

https://www.texasbar.com/AM/Template.cfm?Section=Our_Legal_System1&Template=/CM/ContentDisplay.cfm&

ContentID=23459 and Tex. Gov. Code Sec. 411.0765. Thank you to Derek Cohen of the Texas Policy Lab for

raising this to us in a comment on a previous draft.

Id.

Texas Code of Criminal Procedure Art. 55.01 (2005). For more on the legislature’s intent for expunction to redress

the harm associated with criminal justice involvement, see also State v. T.S.N, 547 S.W.3d 617 (Tex. 2018) (noting

that the state’s expunction statute allows for expunction “in limited, specific circumstances…[with the] intent to,

under certain circumstances, free persons from the permanent shadow and burden of an arrest record, even while

requiring arrest records to be maintained for use in subsequent punishment proceedings and to document and deter

recidivism.”

Collateral Consequences Inventory, Nᴀᴛɪᴏɴᴀʟ Iɴᴠᴇɴᴛᴏʀʏ ᴏғ Cᴏʟʟᴀᴛᴇʀᴀʟ Cᴏɴsᴇᴏ̨ᴜᴇɴᴄᴇs, supra note ___ (specify

“private security, investigations, and locksmiths” as a keyword.)

Demand Analysis - Occupational Detail - Texas - Security Guards (SOC 33-9032), Tᴇxᴀs Lᴀʙᴏʀ Aɴᴀʟʏsɪs, Report

generated August 27, 2021.

Top Statistics - Texas - Projections - No Education Requirement, High School Diploma, or Postsecondary

Non-Degree, Tᴇxᴀs Lᴀʙᴏʀ Aɴᴀʟʏsɪs, Report generated August 27, 2021.

Collateral Consequences Inventory, Nᴀᴛɪᴏɴᴀʟ Iɴᴠᴇɴᴛᴏʀʏ ᴏғ Cᴏʟʟᴀᴛᴇʀᴀʟ Cᴏɴsᴇᴏ̨ᴜᴇɴᴄᴇs, (last visited November 5,

2021), https://niccc.nationalreentryresourcecenter.org/consequences (select “Texas”).

Electronic copy available at: https://ssrn.com/abstract=4065920

impositions on a “person's ability to obtain a desired job or position for many years after the

offense.”

In 2015, the 84th Texas Legislature passed S.B. 1902 to “[give] reformed offenders a

second chance, creating a safer Texas, and increasing the workforce with individuals who are no

longer limited by their minor criminal histories” by making sealing of deferred adjudication

dismissals “automatic.”

This was followed two years later by H.B. 3016 to further expand

eligibility.

Despite these revisions, the scope of Texas’s record sealing law remains narrow. Offenses

that are given deferred adjudication up to a maximum duration

are eligible for relief following

successful completion of community supervision and dismissal.

First offense misdemeanor

convictions, after a two-year waiting period following sentence completion, are also generally

eligible as long as there have been no prior convictions or deferred adjudications.

First-time

driving while intoxicated (DWI) offenses are eligible after a two-to-five year waiting period.

Offenses involving violence besides simple assault, sex crimes, or a handful of other crimes,

that are committed by individuals who have been convicted of certain crimes, are disqualified or

have longer waiting periods.

There are two processes people can use to apply for record sealing. The first process, the

petition route, is used for convictions, most misdemeanors and felonies given deferred

adjudication.

The petition process starts with an individual submitting a petition and a fee to the

“clerk of the court. . . that sentenced [them] or placed [them] on community supervision. . . or

Tex. Gov’t Code §§411.0735 and 411.0725. See also Nondisclosure - Procedure for Deferred Adjudication - for

Felonies and Certain Misdemeanors - Under Section 411.0725, Tᴇxᴀs Lᴀᴡ Hᴇʟᴘ, (last updated January 22, 2022),

https://texaslawhelp.org/article/nondisclosure-procedure-for-deferred-adjudication-for-felonies-and-certain-misdeme

anors-under; Nondisclosure - Procedure for Conviction for Certain Misdemeanors - Under Section 411.0735, Tᴇxᴀs

Lᴀᴡ Hᴇʟᴘ, (January 21, 2022),

https://d9.texaslawhelp.org/article/nondisclosure-procedure-for-conviction-for-certain-misdemeanors-under-section-

4110735.

Tex. Gov. Code §411.0735(c-1) (2015). People are ineligible to have their records sealed if they have ever been

convicted of or received deferred adjudication for offenses including homicide, human trafficking, aggravated

kidnapping, child or elder abuse, stalking, and offenses that require registration as a sex offender. Texas Government

Code §411.074(b) (2015). Payment of legal financial obligations, if required for sentence completion, is also

required. Texas Government Code §411.0735(b) (2015); Texas Government Code §411.0736(b) (2017).

Including Boating while intoxicated (Penal Code §49.06), Flying while intoxicated (Penal Code §49.05),

Assembling or operating an amusement ride while intoxicated (Penal Code §49.065), or Organized Crime (Penal

Code Chapter 71)

Tex. Gov’t Code §411.0736 (2017).

Tex. Gov’t Code §§ 411.073, 411.0735. Some convictions have a shorter waiting period, but for simplicity and to

be conservative, we do not model these shorter periods, as described in the Appendix.

Texas Government Code §42A.102, Texas Government Code §411.0725.

A maximum of 2 years of community supervision following misdemeanors and 10 years of community

supervision for a felony. Art. 42A.103

Senfronia Thompson et al., C.S.H.B. 3016 Bill Analysis, Sᴇɴᴀᴛᴇ Rᴇsᴇᴀʀᴄʜ Cᴇɴᴛᴇʀ, (May 18, 2017),

https://capitol.texas.gov/tlodocs/85R/analysis/pdf/HB03016S.pdf#navpanes=0.

Charles Perry, S.B. 1902 Bill Analysis, Sᴇɴᴀᴛᴇ Rᴇsᴇᴀʀᴄʜ Cᴇɴᴛᴇʀ, 1, (April 17, 2015),

https://capitol.texas.gov/tlodocs/84R/analysis/pdf/SB01902I.pdf#navpanes=0.

Royce West, C.S.S.B. 1477 Bill Analysis, Sᴇɴᴀᴛᴇ Rᴇsᴇᴀʀᴄʜ Cᴇɴᴛᴇʀ, 1, (May 8, 2003),

https://capitol.texas.gov/tlodocs/78R/analysis/pdf/SB01477S.pdf#navpanes=0.

Electronic copy available at: https://ssrn.com/abstract=4065920

deferred adjudication. . . .”

Although the filing fee can vary by county,

“the petition must be

accompanied by payment of a $28 fee to the clerk of the court in addition to any other fee that

generally applies to the filing of a civil petition.”

After submitting a petition, the court holds a

trial to determine whether the petitioner is eligible for sealing and whether sealing the

petitioner’s record “is in the best interests of justice.”

If the court answers both questions

affirmatively, the relief is granted and the record is sealed.

The second process of sealing one’s record, the submission route, is used only for

nonviolent first-time misdemeanor offenses that received a deferred adjudication community

supervision sentence that was completed and dismissed on or after September 1, 2017.

The

submission route does not require a petition.

63 64

Instead, to initiate the process, the applicant

must “[p]resent evidence necessary to establish that [they] are eligible to receive an order under

Section 411.072,” which typically involves filing a Letter Requesting an Order of Nondisclosure

Under Section 411.072.

When submitting the Letter to the clerk, the applicant must also pay a

$28 fee or request a fee waiver.

A judge will then review the evidence and seal the record if the

applicant meets the eligibility criteria.

Despite the procedural differences between the two

record sealing processes, both involve legal fees, court fees, and a petition initiated by the

applicant, which requires awareness of both eligibility and the possibility of sealing.

B. Driver’s License Restoration: Occupational Driver’s License

In Texas, individuals can lose their licenses for a variety of reasons ranging from minor

driving-related offenses (e.g., not having insurance, outdated registration, or not signaling

), to

Chris Abel, Failure to Appear and Traffic Violations, Aʙᴇʟ Lᴀᴡ Fɪʀᴍ, (last visited September 20, 2021),

https://www.flowermoundcriminaldefense.com/failure-appear-and-traffic-violations.

Id., 11.

Id., 11-12.

Id., 11.

Texas Criminal Law: Orders of Nondisclosure, Sᴀᴘᴜᴛᴏ Lᴀᴡ, (last visited September 18, 2021),

https://saputo.law/criminal-law/record-clearing/orders-of-nondisclosure/ (also explaining that although some call

this type of OND “automated,” this is misleading because an individual must still initiate the process).

Id.

Nondisclosure - Deferred Adjudication Community Supervision for Certain Nonviolent Misdemeanors - Under

Section 411.072, Tᴇxᴀs Lᴀᴡ Hᴇʟᴘ, (last updated January 21, 2022),

https://texaslawhelp.org/article/nondisclosure-deferred-adjudication-community-supervision-for-certain-nonviolent-

misdemeanors-under.

Id.

Collateral Consequences Resource Center, Texas: Restoration of Rights & Record Relief, Rᴇsᴛᴏʀᴀᴛɪᴏɴ ᴏғ Rɪɢʜᴛs

Pʀᴏᴊᴇᴄᴛ, (last updated May 20, 2021),

https://ccresourcecenter.org/state-restoration-profiles/texas-restoration-of-rights-pardon-expungement-sealing/.

Texas Administrative Code §411.0745(b) (2015).

Orders of Nondisclosure Overview, Tᴇxᴀs Oғғɪᴄᴇ ᴏғ Cᴏᴜʀᴛ Aᴅᴍɪɴɪsᴛʀᴀᴛɪᴏɴ, 5, (last updated April 2017),

https://www.txcourts.gov/media/821650/order-of-nondisclosure-overview.pdf.

An Overview of Orders of Nondisclosure, Tᴇxᴀs Oғғɪᴄᴇ ᴏғ Cᴏᴜʀᴛ Aᴅᴍɪɴɪsᴛʀᴀᴛɪᴏɴ, 2, (last updated January 2020),

https://www.txcourts.gov/media/1445464/overview-of-orders-of-nondisclosure-2020.pdf.

Electronic copy available at: https://ssrn.com/abstract=4065920

serious driving offenses (e.g., “a habitually reckless or negligent operator of a motor vehicle”

Individuals can also lose their driver’s license for non-driving related offenses, such as “fail[ing]

to appear for a citation (FTA) or fail[ing] to satisfy a judgment ordering the payment of a fine” or

fee.

Once an individual loses their driver’s license, that person is faced with a difficult choice:

find alternatives for basic tasks like driving to work or school, or risking further criminalization

if they are caught driving without a license.

To redress the productivity-related harms associated with not having a driver’s license,

Texas Transportation Code Section 521 Subchapter L allows certain individuals to apply for an

Occupational Driver’s License (ODL) to regain their rights to drive to work and school. The

legislative history suggests that the goal of providing ODLs was to support employment. For

instance, in 1969 Senator William T. Moore noted that “[t]he law relating to driver’s licenses is

now discriminatory in that it deprives many persons of the privilege of following their

occupations and earning a living.”

S.B. 743, which “provide[d] for the issuance of an

occupational license to certain people who have had their license suspended” followed.

In 2015

the 84th Legislature passed H.B. 2246 to both balance the need for public safety, vis-a-vis

individuals with licenses suspended due to past intoxication, with the desire to help such

individuals “continue to support themselves and their families.”

To effect this goal, ODLs were

made available following the installation of an ignition interlock device.

A more recent, related development to restore driver’s licenses was the 2019 repeal of the

Driver's Responsibility Program (DRP), which controversially imposed large fines for often

minor traffic offenses (e.g., speeding or driving without insurance) for the funding of trauma

centers in rural areas.

The program’s end resulted in the restoration of thousands of licenses.

Described by Morgan Smith, To pay for trauma centers, state program sinks thousands of Texas drivers into deep

debt, Tʜᴇ Tᴇxᴀs Tʀɪʙᴜɴᴇ, (August 27, 2018),

https://www.texastribune.org/2018/08/27/pay-trauma-centers-texas-sinks-thousands-drivers-deep-debt/ (describing

how the program, originally intended to "hold bad drivers responsible for the damage they caused, with the license

suspensions having the added benefit of keeping them off the roads" was eventually seen as a “massive failure”).

See also Matthew Menendez et al., Fees and Fines: A Fiscal Analysis of Three States and Ten Counties, Bʀᴇɴɴᴀɴ

Cᴇɴᴛᴇʀ ғᴏʀ Jᴜsᴛɪᴄᴇ, 26, (November 21, 2019),

https://www.brennancenter.org/our-work/research-reports/steep-costs-criminal-justice-fees-and-fines.

Texas Transportation Code §521.244(e). (creating OND eligibility based on evidence of financial responsibility

and proof of the installation of an ignition interlock device on each motor vehicle operated by the individual.)

Jason Villalba et al., H.B. 2246 Bill Analysis, Sᴇɴᴀᴛᴇ Rᴇsᴇᴀʀᴄʜ Cᴇɴᴛᴇʀr, (May 18, 2015),

https://capitol.texas.gov/tlodocs/84R/analysis/pdf/HB02246E.pdf#navpanes=0.

Id.

William T. Moore, S.B. 753 Bill Analysis, Sᴇɴᴀᴛᴇ Rᴇsᴇᴀʀᴄʜ Cᴇɴᴛᴇʀ, 8, (May 7, 1969),

https://lrl.texas.gov/LASDOCS/61R/SB743/SB743_61R.pdf#page=8.

As recognized by the U.S. Supreme Court, see Bell v. Burson (1971), supra note __.

Driven By Debt, Tᴇxᴀs Aᴘᴘʟᴇsᴇᴇᴅ ᴀɴᴅ Tᴇxᴀs Fᴀɪʀ Dᴇғᴇɴsᴇ Pʀᴏᴊᴇᴄᴛ, (December 13, 2018),

https://report.texasappleseed.org/driven-by-debt/.

Failure to Appear/Failure to Pay Program, Tᴇxᴀs Dᴇᴘᴀʀᴛᴍᴇɴᴛ ᴏғ Pᴜʙʟɪᴄ Sᴀғᴇᴛʏ, (last visited September 20,

2021), https://www.dps.texas.gov/section/driver-license/failure-appearfailure-pay-program.

Texas Government Code §521.292(a)(2) (1999). See also Driver’s License Enforcement Actions, Tᴇxᴀs

Dᴇᴘᴀʀᴛᴍᴇɴᴛ ᴏғ Pᴜʙʟɪᴄ Sᴀғᴇᴛʏ, (last visited September 20, 2021),

https://www.dps.texas.gov/internetforms/Forms/DL-176.pdf (listing the reasons individuals can have their licenses

revoked, suspended, and/or disqualified).

Electronic copy available at: https://ssrn.com/abstract=4065920

However, “an estimated 500,000 individuals’ licenses remained suspended after their DRP

surcharges were eliminated,”

a figure consistent with the data reported in this study.

To obtain an ODL under Texas Transportation Code Section 521 Subchapter L, one must

demonstrate “an essential need” for the ODL, for example as evidenced by the need to drive to

work or to school, and a lack of an alternative transportation option.

However, individuals are

ineligible for ODLs if they have “lost [their] driving privileges because of a mental or physical

disability,” have “lost [their] driving privileges for failure to pay child support,” have “received

two ODLs in the past 10 years after a conviction,” or “have a ‘hard suspension’ waiting period

due to a prior DWI arrest or conviction.”

ODLs are also unavailable to individuals who need a

license to drive a commercial motor vehicle.

To obtain an ODL, an eligible individual must complete a petition, in accordance with

local court requirements.

Next, the applicant must file the petition and submit a filing fee.

their license was “automatically suspended or canceled following a conviction, [the applicant

should] file the Petition in the same court that convicted [them].”

If not, they can choose to file

the petition “in the county where [they] live or where the offense occurred.”

Following filing,

there is a hearing where a judge reviews the petition and other paperwork of the petitioner,

including a Certified Abstract of the petitioner’s full driving record, an SR-22 Proof of Insurance

from the petitioner’s insurance company, and evidence that the petitioner needs the license to go

to work, attend school, etc.

The judge will then decide whether to grant an ODL.

C. Barriers to Relief Under Texas’s Second Chance Laws

As discussed above, Texas enacted its record sealing and driver’s relicensing laws with

the goals of reducing the size of the criminal justice system and increasing accessibility to

second chances and workforce opportunities. However, three administrative burdens placed on

applicants by the second chance laws limit the legislature’s success in meeting these goals.

A related area of literature about administrative burdens is growing quickly. For instance, see Julian Christensen

et al., Human Capital and Administrative Burden: The Role of Cognitive Resources in Citizen-State Interactions, 80

Pᴜʙʟɪᴄ Aᴅᴍɪɴɪsᴛʀᴀᴛɪᴏɴ Rᴇᴠɪᴇᴡ 127, (January/February 2020) (describing the ways in which “citizens with lower

levels of human capital” experience greater administrative burdens, which contributes to reinforcing inequality);

Pᴀᴍᴇʟᴀ Hᴇʀᴅ & Dᴏɴᴀʟᴅ P. Mᴏʏɪɴʜᴀɴ, Aᴅᴍɪɴɪsᴛʀᴀᴛɪᴠᴇ Bᴜʀᴅᴇɴ: Pᴏʟɪᴄʏᴍᴀᴋɪɴɢ ʙʏ Oᴛʜᴇʀ Mᴇᴀɴs (2018) (arguing

that administrative burdens are conscious policy choices); Elizabeth Linos et al., Nudging Early Reduces

Administrative Burden: Three Field Experiments to Improve Code Enforcement, 39 Jᴏᴜʀɴᴀʟ ᴏғ Pᴏʟɪᴄʏ Aɴᴀʟʏsɪs ᴀɴᴅ

Id.

Supra note 75, 3.

Id, 2.

Id.

Id. For more on hard suspension waiting periods, see Occupational Driver’s License, Tᴇxᴀs Lᴀᴡ Hᴇʟᴘ, (last

updated June 2, 2021), https://texaslawhelp.org/guide/occupational-drivers-license/?tab=0.

Texas Occupational Driver’s License, 1, (last visited January 20, 2022),

https://www.co.chambers.tx.us/upload/page/0100/docs/Occupational%20DL/BrochureODL.pdf.

Tᴇxᴀs Lᴀᴡ Hᴇʟᴘ, supra note 72.

Electronic copy available at: https://ssrn.com/abstract=4065920

First, the petition process in both cases requires an individual not only to prove they deserve a

second chance, but also to ascertain the law and fill out potentially confusing paperwork.

Second, the requirement that individuals seeking a second chance attend a hearing similarly

burdens people with challenges ranging from getting a hearing on the calendar (which requires

successfully submitting a petition) to attending a hearing (which can involve taking time off of

work, traveling, etc.). Third, complex criteria that frequently evolve make it difficult for

individuals to keep up with the law and determine their eligibility. These administrative barriers

to petitioners not only contrast with the automatic removal of rights and collateral consequences

after a misdemeanor conviction, incarceration, or the suspension of a driver’s license, but also

contribute to the gap in delivery of record cleaning and license restoration, as quantified in the

next part.

PART II: ESTIMATING THE SIZE OF TEXAS’S SEALING AND DRIVER’S LICENSE

RESTORATION SECOND CHANCE GAPS

While Texas legislatures have passed second chance laws to advance a variety of goals,

the benefits of second chance relief depend on its delivery. In this part, we start by profiling the

people in each target population. Analyzing criminal convictions data from the state, we find

individuals with convictions on average to be of working age (mid-40s) and on average, have a

last conviction from over a decade ago. The available evidence suggests that people with

suspended licenses appear to be a bit younger (30-40), and have their licenses suspended on

average for five years and seven months.

To ascertain the number of people eligible to have their convictions sealed,

we applied a

simplified version of the sealing law to criminal convictions records to estimate the size of

Texas’s “second chance sealing gap” and “second chance driver restoration gap.” Based on our

analysis, we estimate that about 670,000 people are able to have their records cleared, 59,000

people completely, to achieve a “clean slate” under existing law (Table 1). For this eligible

population, the average number of years since the last conviction is about 17 (median = 15

years). Furthermore, 430,000 people appear eligible to apply for an ODL to drive to work or

school. These numbers translate into a 5% and 17% uptake rate of sealing and driver’s license

reinstatements, respectively. (Table 1)

A. The Texas Criminal Population That Could Benefit from Sealing Relief

Cʜɪᴇɴ, supra note 20.

Language modified and adapted from Colleen Chien et al., The Texas Second Chance Non-Disclosure/Sealing

Gap, Pᴀᴘᴇʀ Pʀɪsᴏɴs Iɴɪᴛɪᴀᴛɪᴠᴇ, (last visited January 21, 2022), https://www.paperprisons.org/states/TX.html.

Mᴀɴᴀɢᴇᴍᴇɴᴛ 243, (Winter 2020) (uses a field experiment to demonstrate that learning costs, compliance costs, and

psychological costs help to explain why residents do not always take up programs for which they are eligible”); Cass

R. Sunstein, Sludge and Ordeals, 68 Dᴜᴋᴇ L. J. 1843, (2019) (argues that deregulation driven by data and behavioral

information should be undertaken due to the 9.78 billion hours of “sludge” paperwork Americans completed in 2018

for the government, but notes that such deregulation will be filled with numerous tradeoffs).

Electronic copy available at: https://ssrn.com/abstract=4065920

Before estimating the size of the population entitled to second chance relief, it is worth

considering the current size and characteristics of the existing Texas criminal population, as

reflected in the dataset upon which we rely for our records sealing analysis: the Texas

Computerized Criminal History System (CCH). Maintained by the Texas Department of Public

Safety, the CCH is a database containing all publicly available convictions for adults from 1976

to the date of extraction.

This database is quite large, containing over 5.2 million Texans who

have publicly available convictions records. However, the true size of Texas's conviction

population is smaller because the CCH data likely includes individuals who are deceased. To

account for this, we removed all individuals over the age of 80 years old from the dataset on the

basis that the average life expectancy for Americans in 2019 was 78.8 years.

After doing so, we

estimate that approximately 4.8 million Texans (~16.4% of the state’s adult population in 2020)

have publicly available convictions records.

This estimate is at best an approximation because

it fails to account for the thousands of people who move in and out of the Lone Star State each

year. This database also does not include people with non-conviction and deferred

adjudication-only records, who are also eligible for records relief under the law.

Analyzing a random sample selected from the database of 150,000 people, we find that

about 80% have felony convictions and 62% have misdemeanor convictions. The most common

charges include drug possession, driving while intoxicated, and felony burglary as well as

misdemeanor assault causing bodily injury to a family member (Appendix Table 1). Consistent

with general trends, the population is overwhelmingly male. But while the average

See Appendix __ for details.

America Counts Staff, Texas Added Almost 4 Million People Last Decade, Uɴɪᴛᴇᴅ Sᴛᴀᴛᴇs Cᴇɴsᴜs Bᴜʀᴇᴀᴜ,

(August 25, 2021),

https://www.census.gov/library/stories/state-by-state/texas-population-change-between-census-decade.html. ~16.4%

= 4.8M divided by Texas's adult population in 2020, which we calculated by multiplying the size of Texas's

Population in 2020 (29,145,055 people) by the percentage of the population over the age of 18 (75%).

National Center for Health Statistics, Life Expectancy, Cᴇɴᴛᴇʀs ғᴏʀ Dɪsᴇᴀsᴇ Cᴏɴᴛʀᴏʟ ᴀɴᴅ Pʀᴇᴠᴇɴᴛɪᴏɴ, (January 8,

2022), https://www.cdc.gov/nchs/fastats/life-expectancy.htm.

About CCH, Tᴇxᴀs Dᴇᴘᴀʀᴛᴍᴇɴᴛ ᴏғ Pᴜʙʟɪᴄ Sᴀғᴇᴛʏ, (last visited August 20, 2021),

https://publicsite.dps.texas.gov/DpsWebsite/CriminalHistory/AboutCch.aspx (also see this source for more about the

CCH). Per the Department of Public Safety, “Computerized Criminal History (CCH) was created in 1976 and we

began sending the conviction database in 1998. A person’s criminal history is retained for 125 years from their date

of birth.” See email from Texas Department of Public Safety, on file with editors. < for editors:

https://app.sparkmailapp.com/web-share/sW1CF6EOzhRBYaJTOh-6TdDpbW6qQLxd8W2CwfVq >

Electronic copy available at: https://ssrn.com/abstract=4065920

Table 1: The Population of People in Texas with Convictions

Estimated Number of People with

Convictions

4.8M

% Male

85%

Top Convictions - Felonies

poss cs pg 1 <1g (11.4%), DWI 3rd

or more (4.2%), burglary (3.7%)

Top Convictions - Misdemeanors

poss marij <2oz (8.3%), DWI

(5.6%), assault (3.7%)

Avg Years since last Conviction

12.6 (median = 11)

Average Age at First Conviction

28 (median =27)

Average Current Age of People with

Convictions

White and Latinx % (share in pop =

82%)

69%

Black % (share in pop = 13%)

31%

Asian % (share in pop = 5%)

Source: Authors’ analysis based on the Texas CCH database.

age at first conviction is 28, the average current age of people in the sample is 45. On average,

the last conviction of each person was 12.6 years ago. From an earnings perspective this implies

that the average age of people who live with convictions in Texas overlaps with the years in

which workers typically hit their peak earnings.

Previous research finds that the racial disparities in Texas's criminal justice system are

significant. For example, although Black people account for 13% of the state population,

they

make up 33% of the Texas prison population.

In contrast, white people make up 44% of the

state population but account for just 33% of the prison population.

Thus, as of 2020, Black

Id.

Incarceration Trends in Texas, Vᴇʀᴀ Iɴsᴛɪᴛᴜᴛᴇ ᴏғ Jᴜsᴛɪᴄᴇ, (December 2019),

https://www.vera.org/downloads/pdfdownloads/state-incarceration-trends-texas.pdf.

Ashley Nellis, The Color of Justice: Racial and Ethnic Disparities in State Prisons, Tʜᴇ Sᴇɴᴛᴇɴᴄɪɴɢ Pʀᴏᴊᴇᴄᴛ.

(October 13, 2021),

https://www.sentencingproject.org/publications/color-of-justice-racial-and-ethnic-disparity-in-state-prisons/ at 7.

Julia Carpenter, Millennials’ High-Earning Years Are Here, but It Doesn’t Feel That Way, Tʜᴇ Wᴀʟʟ Sᴛʀᴇᴇᴛ

Jᴏᴜʀɴᴀʟ, (August 12, 2021),

https://www.wsj.com/articles/millennials-high-earning-years-are-here-but-it-doesnt-feel-that-way-11628769603

(indicating, based on figures from the Bureau of Labor Statistics, that workers typically experience peak earnings

“between the ages of 35 and 54”).

Electronic copy available at: https://ssrn.com/abstract=4065920

people are 3.4 times more likely to be in prison than white people in Texas.

In contrast, Latinx

and Asian people are underrepresented in the prison population relative to their representation in

the population in general.

100

The data we analyzed, from the Texas Department of Public Safety, suggests similar

disparities in the breakdown of felony and misdemeanor convictions by race. For instance, even

though Black people make up just 13% of Texas's population, we find that Black people make up

30% of people with misdemeanor convictions and 31% of people with felony convictions in our

database. Whites and Asians appear to be underrepresented in felony and misdemeanor

convictions relative to their representation in the population in general. (Table 1)

B. The Texas Population That Could Benefit from License Restoration Relief

In contrast to the population of people with criminal records, less is known about the

demographic characteristics of people with suspended licenses. The populations are distinct,

however, as driver’s license suspensions are administrative penalties that, in Texas, generally

follow non-compliance with court-ordered fines and fees or requests to appear.

101

This means that

license suspensions often impact people who haven’t committed serious crimes,

102

or even been

accused of them.

Other studies have considered license suspension programs in Texas, North Carolina, and

New Jersey. Though the details of each suspension program are unique, the available evidence

suggests that individuals with suspended licenses tend to draw disproportionately from

low-income, urban communities and particularly harm the Black community.

Carnegie et al.’s study of New Jersey drivers from 2007 reports that “only 16.5 percent of

New Jersey licensed drivers reside in lower income zip codes, while 43 percent of all suspended

drivers live there.”

103

A later study by Joyce et al.

104

of all suspended licenses in New Jersey from

2004 to 2018 found that the median household income for people with non-driving related

suspensions was about $78,000,

105

which is about $14,000 lower than the median household

105

Quick Facts: New Jersey, Uɴɪᴛᴇᴅ Sᴛᴀᴛᴇs Cᴇɴsᴜs Bᴜʀᴇᴀᴜ, (last visited January 23, 2022),

https://www.census.gov/quickfacts/fact/table/NJ/SBO001212.

104

Nina R. Joyce et al., Individual and Geographic Variation in Driver’s License Suspensions: Evidence of

Disparities by Race, Ethnicity, and Income, 19 Jᴏᴜʀɴᴀʟ ᴏғ Tʀᴀɴsᴘᴏʀᴛ ᴀɴᴅ Hᴇᴀʟᴛʜ 1, 6 (2020).

103

Jon A. Carnegie et al., Driver’s License Suspensions, Impacts and Fairness Study, Nᴇᴡ Jᴇʀsᴇʏ Dᴇᴘᴀʀᴛᴍᴇɴᴛ ᴏғ

Tʀᴀɴsᴘᴏʀᴛᴀᴛɪᴏɴ, 66 (August 2007).

102

Justin Wm. Moyer, More than 7 million people may have lost driver’s licenses because of traffic debt,

Wᴀsʜɪɴɢᴛᴏɴ Pᴏsᴛ (May 19, 2018),

https://www.washingtonpost.com/local/public-safety/more-than-7-million-people-may-have-lost-drivers-licenses-be

cause-of-traffic-debt/2018/05/19/97678c08-5785-11e8-b656-a5f8c2a9295d_story.html (noting that “[d]river’s

license suspensions were criticized by anti-poverty advocates after a 2015 federal investigation, focused on

Ferguson, Mo., revealed that law enforcement used fines to raise revenue for state and local governments”).

101

100

See Table 1 for a snapshot of the Texas criminal population, by race.

Ashley Nellis, The Color of Justice: Racial and Ethnic Disparities in State Prisons, Tʜᴇ Sᴇɴᴛᴇɴᴄɪɴɢ Pʀᴏᴊᴇᴄᴛ, 21

(October 13, 2021),

https://www.sentencingproject.org/publications/color-of-justice-racial-and-ethnic-disparity-in-state-prisons/.

Electronic copy available at: https://ssrn.com/abstract=4065920

income of $82,000 in 2019.

106

In contrast, the median household income for people who did not

have suspensions was nearly $105,000, about $23,000 above New Jersey’s median household

income.

107

Although estimates about the average or median income of people with suspended

licenses is unavailable for Texas, researchers for the Texas Fair Defense Project and Texas

Appleseed found a negative correlation between the number of license suspensions and the

household income in both Houston

108

and Dallas: “as zip code income increased, the number of

holds decreased.”

109

Both studies from New Jersey find people with suspended licenses are disproportionately

from urban areas. Carnegie et al. note that “[a]lthough only 43 percent of New Jersey licensed

drivers reside in urban areas, 63 percent of suspended drivers live there.”

110

More recent research

similarly reports that 4% of people who had any non-driving related suspensions lived in rural

areas, compared to 5.7% of people with no suspension.

111

Racial disparities among drivers with suspended licenses are significant. The Texas

studies referred to earlier, for example, find that in Dallas, Black people account for only 11% of

the driving population and yet account for 28.6% of people who cannot get their licenses

renewed due to OmniBase holds.

112

A similar trend is present in Houston, where Black people

make up only 22% of the city population, but comprise 40% of the people with OmniBase

license holds from the Houston Municipal Court.

113

Couzier and Garrett’s study of suspensions in

North Carolina finds that Black people make up 21% of drivers and account for 50.5% of license

113

Driven by Debt Houston, Tᴇxᴀs Fᴀɪʀ Dᴇғᴇɴsᴇ Pʀᴏᴊᴇᴄᴛ ᴀɴᴅ Tᴇxᴀs Aᴘᴘʟᴇsᴇᴇᴅ, 5 (July 2020),

https://www.texasappleseed.org/sites/default/files/DrivenByDebt-Houston-July2020.pdf.

112

Driven by Debt Dallas, Tᴇxᴀs Fᴀɪʀ Dᴇғᴇɴsᴇ Pʀᴏᴊᴇᴄᴛ ᴀɴᴅ Tᴇxᴀs Aᴘᴘʟᴇsᴇᴇᴅ, 6 (November 2019),

https://www.texasappleseed.org/sites/default/files/Driven%20By%20Debt%20Dallas.pdf.

111

Nina R. Joyce et al., Individual and Geographic Variation in Driver’s License Suspensions: Evidence of

Disparities by Race, Ethnicity, and Income, 19 Jᴏᴜʀɴᴀʟ ᴏғ Tʀᴀɴsᴘᴏʀᴛ ᴀɴᴅ Hᴇᴀʟᴛʜ 1, 6 (2020). (also finding that

6.7% of people with driving-related suspensions lived in rural areas)

110

Jon A. Carnegie et al., Driver’s License Suspensions, Impacts and Fairness Study, Nᴇᴡ Jᴇʀsᴇʏ Dᴇᴘᴀʀᴛᴍᴇɴᴛ ᴏғ

Tʀᴀɴsᴘᴏʀᴛᴀᴛɪᴏɴ, 66 (August 2007).

109

Driven by Debt Dallas, Tᴇxᴀs Fᴀɪʀ Dᴇғᴇɴsᴇ Pʀᴏᴊᴇᴄᴛ ᴀɴᴅ Tᴇxᴀs Aᴘᴘʟᴇsᴇᴇᴅ, 5 (November 2019),

https://www.texasappleseed.org/sites/default/files/Driven%20By%20Debt%20Dallas.pdf and Driven by Debt

Houston, Tᴇxᴀs Fᴀɪʀ Dᴇғᴇɴsᴇ Pʀᴏᴊᴇᴄᴛ ᴀɴᴅ Tᴇxᴀs Aᴘᴘʟᴇsᴇᴇᴅ, 4 (July 2020),

https://www.texasappleseed.org/sites/default/files/DrivenByDebt-Houston-July2020.pdf.

108

For more specific data on Houston, see Driven by Debt Houston, Tᴇxᴀs Fᴀɪʀ Dᴇғᴇɴsᴇ Pʀᴏᴊᴇᴄᴛ ᴀɴᴅ Tᴇxᴀs

Aᴘᴘʟᴇsᴇᴇᴅ, 5 (July 2020), https://www.texasappleseed.org/sites/default/files/DrivenByDebt-Houston-July2020.pdf

(“The zip code with the most holds per resident is 77026, an area in Northeast Houston covering the Kashmere

Gardens neighborhood, with 344 holds per every 1,000 residents. This zip code has more than one third of its

residents living below the poverty level and a majority of its residents (52%) are black. The median income is over

$20,000 less than the citywide median income. Profiles of other heavily affected zip codes are similar. The ten zip

codes with the highest rates of holds all have people living in poverty at higher rates than the city’s overall poverty

rate and most have median incomes below the city’s median income. Six of these ten zip codes have a population

that is more than 50% people of color).

107

Nina R. Joyce et al., Individual and Geographic Variation in Driver’s License Suspensions: Evidence of

Disparities by Race, Ethnicity, and Income, 19 Jᴏᴜʀɴᴀʟ ᴏғ Tʀᴀɴsᴘᴏʀᴛ ᴀɴᴅ Hᴇᴀʟᴛʜ 1, 6 (2020).

106

Jon A. Carnegie et al., Driver’s License Suspensions, Impacts and Fairness Study, Nᴇᴡ Jᴇʀsᴇʏ Dᴇᴘᴀʀᴛᴍᴇɴᴛ ᴏғ

Tʀᴀɴsᴘᴏʀᴛᴀᴛɪᴏɴ, 66 (August 2007).

Electronic copy available at: https://ssrn.com/abstract=4065920

suspensions, whereas white people account for 65% of drivers and 36.3% of license

suspensions.

114

Insofar as lost earnings due to suspended licenses is concerned, it is also worthwhile

considering the age of people with a suspended license and the average number of years their

license was suspended. Crozier and Garrett’s North Carolina study finds that the average age of

people at the time of license suspension was 28 years to 29 years, and the average length of hold

ranges from five to ten years.

115

Joyce et al.’s study of all New Jersey drivers with a suspended

license from 2004 to 2018 documented an average driver age of 39.4. Most relevant for our

purposes, the reports by the Texas Fair Defense Project and Texas Appleseed report an average

hold length of five years and seven months based on 2018 data acquired from the Department of

Public Safety. A summary of this information is available in Table A.

Table A: Demographic Information About Population With Suspended Driver’s Licenses

Study

Jurisdiction

Age

Average Length of

Hold

Source of

Estimates

Texas Fair Defense Project

and Texas Appleseed

(2019 and 2020)

Dallas and

Houston

– –

5 years and 7 months

(statewide estimate

from 2018 DPS data)

Pages 5-6

Joyce et al. (2020)

New Jersey

39.4 (mean age of

people with any

non-driving related

suspension)

– –

Table 1

Crouzier and Garrett

(2020)

North

Carolina

28.67 (median age at

time of offense)

10.1 years (median)

Table 2

C. Sizing Texas’s Second Chance Sealing Gap

115

Driven by Debt Dallas, Tᴇxᴀs Fᴀɪʀ Dᴇғᴇɴsᴇ Pʀᴏᴊᴇᴄᴛ ᴀɴᴅ Tᴇxᴀs Aᴘᴘʟᴇsᴇᴇᴅ, 5-6 (November 2019),

https://www.texasappleseed.org/sites/default/files/Driven%20By%20Debt%20Dallas.pdf; Driven by Debt Houston,

Tᴇxᴀs Fᴀɪʀ Dᴇғᴇɴsᴇ Pʀᴏᴊᴇᴄᴛ ᴀɴᴅ Tᴇxᴀs Aᴘᴘʟᴇsᴇᴇᴅ, 4-5 (July 2020),

https://www.texasappleseed.org/sites/default/files/DrivenByDebt-Houston-July2020.pdf; William E. Couzier &

Brandon L. Garrett, Driven to Failure: An Empirical Analysis of Driver’s License Suspension in North Carolina, 69

Dᴜᴋᴇ L.J. 1585, 1607 (2020); Nina R. Joyce et al., Individual and Geographic Variation in Driver’s License

Suspensions: Evidence of Disparities by Race, Ethnicity, and Income, 19 Jᴏᴜʀɴᴀʟ ᴏғ Tʀᴀɴsᴘᴏʀᴛ ᴀɴᴅ Hᴇᴀʟᴛʜ 1, 6

(2020).

114

William E. Couzier & Brandon L. Garrett, Driven to Failure: An Empirical Analysis of Driver’s License

Suspension in North Carolina, 69 Dᴜᴋᴇ L.J. 1585, 1607-1608 (2020).

Electronic copy available at: https://ssrn.com/abstract=4065920

Having provided an overview of Texas’s criminal population, we begin by calculating the

second chance sealing gap — the difference between eligibility and delivery of sealing relief to

people with criminal convictions. We use the gap-sizing methodology described in Chien,

America's Paper Prisons: The Second Chance Gap, Michigan Law Review (2020) to determine

several numbers: (1) the current gap — the number and share of individuals with records that

could qualify for relief, (2) the uptake gap — the share of people eligible for a given second

chance that have obtained it,

116

and (3) based on the same data used to calculate the current and

uptake gaps, how many years, at current rates, it would take to clear the existing backlog.

First, we ascertained and modeled Texas’s OND law. Next, we applied the model of the

laws to a sample of criminal histories obtained from the state to identify the number of

individuals eligible for a given second chance. Once we had estimated the number of people

eligible for a given second chance, we calculated the “current gap,” the uptake rate, and the pace

of record relief using the following steps. To estimate the current gap, we divided the number of

people eligible for a given second chance by the number of people in our sample. To estimate the

population eligible for relief, we multiplied the current gap by the total population, which was

estimated using state data. Next, using historical data obtained from the state, we calculated the

estimated relief granted over the past five to ten years by adding the number of second chances

granted to the product of (i) the number of second chances granted in the earliest year of data and

(ii) the number of years left to reach five or ten years of data (whichever was closest to the actual

number of years of data the state provided). Then, to calculate the uptake rate, we divided the

estimated historical relief rate by the number of people eligible for relief plus the estimated

historical relief rate. After calculating the current gap and the uptake gap, we estimated the

number of years it would take to clear the backlog by dividing the population eligible for relief

by the number of people who were granted a second chance in the most recent full year of data.

There are several weaknesses with our methodology. First, we do not account for

eligibility requirements related to fines and fees due to a lack of data, making our eligibility

estimates generous. In the other direction, we also do not include eligibility for expungements of

non-convictions, which depress, potentially dramatically, our estimates of who falls into the

records relief gap. Second, the underlying criminal history provided by the state at times was

missing sentence expiration dates. When that data was missing, we inferred expiration dates

based on data where expiration dates were present.

117

Finally, as detailed in the appendix, certain

eligibility provisions contained ambiguities that we were unable to resolve despite multiple

consultations with local criminal law experts. These challenges introduce inaccuracies that cause

our estimates to be both over- and under-inclusive.

Table 2: Estimated Eligibility and Uptake of Texas Record Sealing and Drivers Restoration

Order of Non-Disclosure

Occupational Driver’s

117

Based on our analysis, we assumed expiration date = sentence_start_date + 2.9 years for misdemeanors and 3.2

years for felonies. Additional details are available in the Appendix.

116

Colleen Chien, America’s Paper Prisons: The Second Chance Gaps, 119 Mɪᴄʜ. L. Rᴇᴠ. 519 (2020) at 542-543.

Electronic copy available at: https://ssrn.com/abstract=4065920

(Sealing)

License (Restoration)

Total Population (Eligible + Ineligible)

4,826,860

N/A

Population Eligible for Relief

676,845

438,000

Population Eligible for a Clean Record (no

Conviction)

58,501

N/A

Relief Granted in Last Available Year

2,650 (2019)

16,350 (2019)

Estimated Relief Granted

36,409 (10 years)

87,027 (5 years)

Uptake Rate (estimated share of people eligible

for a given second chance that have obtained it)

~5%

~17%

Years It Would Take to Clear the Backlog

255

Current Gap (estimated share of all people with

records that are eligible for relief)

14%

N/A

Based on taking the steps described above,

118

we estimate that around 677,000 people

with misdemeanor convictions or deferred adjudications are eligible for sealing relief under

Texas Gov. Code Chapter 411, and 59,000 for a clean record. We further find that 18,593 people

had their records sealed between fiscal years 2014 and 2019. Based on this, we project that, at

most, 36,000 people sealed their records over the past ten years. Combining these historical

figures with our eligibility calculations, we estimate that ~5% of people eligible for relief have

received it, leaving 95% of people in the “Texas Second Chance Sealing Gap.” Based on

administrative data, 2,650 people sealed their records in the last year of available data (2019).

119

At this rate, it would take 255 years to clear the sealing backlog. The profile of individuals that

could get a “cleaner” or “completely cleaned” record is similar to that of the average profile of a

person with a conviction, except that the average years since the last conviction is 17 and 19

years, respectively.

120

D. Sizing Texas’s Second Chance Driver’s License Restoration Gap

We applied a similar “second chance gap” approach to quantifying the number of people

in Texas who appear eligible for but have not received an occupational driver’s license

“restoration” based on Texas Transportation Code Section 521 Subchapter L. Doing so requires

an understanding of how driver’s licenses are suspended in Texas in the first place. Practitioners

have generally described two main, non-driving-related causes of a license suspension: (i) failure

120

See Appendix

119

Data was acquired from the Texas Department of Public Safety. See Appendix E for a breakdown of sealings by

year for every year of data the Department of Public Safety gave us.

118

Further details are provided in the Appendix.

Electronic copy available at: https://ssrn.com/abstract=4065920

to pay criminal court-related fines, fees, and surcharges; and (ii) failure to appear (FTA).

121

FTAs

in Texas result, in turn, when a person is summoned to court for a range of offenses, ranging

from traffic citations to felonies, but does not appear for their trial or hearing.

122

The

consequences can be severe. For instance, failure to appear can result in the issuance of an arrest

warrant.

123

Individuals can also lose their driver’s license under Texas Transportation Code 706,

which allows the state to “deny renewal of the driver's license of a person who fails to appear for

a complaint or citation or fails to pay or satisfy a judgment ordering payment of a fine and cost. .

. in a matter involving any [criminal] offense.”

124

Using this as a foundation, we focus on people who lost their licenses under the

OmniBase Program, which suspends licenses for failure to pay court-related expenses or failure

to appear.

125

We acquired data provided by the Texas Department of Public Safety to ascertain

the number of people that presumably met the “FTA” criteria for ODL relief by virtue of having

a FTA on their driver’s license, which was 438,000 individuals.

126

To complement our analysis,

we relied on statistics from the Department of Public Safety on the number of people who

received an ODL under the OmniBase Program, which reports that 47,449 ODLs were granted

between September 2017 and December 2020.

127

There are a few limitations to our approach. First, we were unable to obtain data by

127

See Driver License Division High Value Set, Dᴇᴘᴀʀᴛᴍᴇɴᴛ ᴏғ Pᴜʙʟɪᴄ Sᴀғᴇᴛʏ (February 2020, January 2020,

December 2019, November 2019, October 2019, September 2019, August 2019, July 2019, June 2019, May 2019,

April 2019, March 2019, February 2019, January 2019, December 2018, November 2018, October 2018, September

2018, August 2018, July 2018, June 2018, May 2018, April 2018, March 2018, February 2018, January 2018,

December 2017, November 2017, October 2017, September 2017) (all reports as of January 21, 2022 available at

https://www.dps.texas.gov/section/driver-license/driver-license-division-high-value-data-sets).

126

Email from Texas Department of Public Safety to Texas Fair Defense Project. Email is on file with authors. [for

editors: https://cloudhq.net/s/e398c60edff3f1]

125

Failure to Appear/Failure to Pay Program, Tᴇxᴀs Dᴇᴘᴀʀᴛᴍᴇɴᴛ ᴏғ Pᴜʙʟɪᴄ Sᴀғᴇᴛʏ, (last viewed January 22, 2022),

https://www.dps.texas.gov/section/driver-license/failure-appearfailure-pay-program. See also Texas Transportation

Code, Chapter 706.

124

Texas Transportation Code, Chapter 706.

123

Id., 7-8.

122

Notably, “[m]ost often, people are charged with a fine-only offense when they receive a ticket written by a law

enforcement officer. The ticket instructs them to pay the fine and court costs, or alternatively, to appear in court on

or by a certain date. Only people who cannot pay immediately or want to contest the ticket must show up in court.”

Deborah Fowler et al., Pay Or Stay: The High Cost of Jailing Texans for Fines & Fees, Tᴇxᴀs Aᴘᴘʟᴇsᴇᴇᴅ ᴀɴᴅ Tᴇxᴀs

Fᴀɪʀ Dᴇғᴇɴsᴇ Pʀᴏᴊᴇᴄᴛ, 7, (February 2017),

https://www.texasappleseed.org/sites/default/files/PayorStay_Report_final_Feb2017.pdf.

121

See Driven By Debt, Tᴇxᴀs Aᴘᴘʟᴇsᴇᴇᴅ ᴀɴᴅ Tᴇxᴀs Fᴀɪʀ Dᴇғᴇɴsᴇ Pʀᴏᴊᴇᴄᴛ, (December 13, 2018),

https://report.texasappleseed.org/driven-by-debt/; Driven by Debt: The Failure of the Omnibase Program, Tᴇxᴀs

Aᴘᴘʟᴇsᴇᴇᴅ ᴀɴᴅ Tᴇxᴀs Fᴀɪʀ Dᴇғᴇɴsᴇ Pʀᴏᴊᴇᴄᴛ, 3, (August 2021),

https://www.texasappleseed.org/sites/default/files/OmniBaseRevenueReport-Aug11-Final.pdf (explaining that

“Holds on license renewals are triggered when license holders either fail to pay fines and costs or fail to appear in

court, usually for traffic offenses. However, eventually all OmniBase Holds are incurred due to an inability to pay,

because the only way to lift a hold is to completely pay off all underlying debt”); Driven by Debt: Dallas, Tᴇxᴀs

Aᴘᴘʟᴇsᴇᴇᴅ ᴀɴᴅ Tᴇxᴀs Fᴀɪʀ Dᴇғᴇɴsᴇ Pʀᴏᴊᴇᴄᴛ, 1, (November 2019),

https://www.texasappleseed.org/sites/default/files/Driven%20By%20Debt%20Dallas.pdf (explaining that “[t]he vast

majority of criminal cases in Texas are fine-only misdemeanors, which are the lowest level of criminal offenses in

Texas and intended to be punished by fines alone and no jail time. Fine-only misdemeanors include most traffic

offenses, city ordinance violations and other Class C misdemeanors such as public intoxication”).

Electronic copy available at: https://ssrn.com/abstract=4065920

which we could determine the bases for license suspension and so therefore could not model the

“inability to pay” criteria. We excluded people qualifying for occupational licenses based on this

criteria from our estimate, depressing it. That being said, we note that there is likely substantial

overlap between the universe of individuals who qualify under each criteria, particularly because

a major cause of a FTA is lack of transportation to attend one’s hearing.

128

Second, we do not

model the requirement for receiving an ODL of having driver’s insurance as required in the

court. In operation, this requirement is likely to act as a significant hurdle to ODL restoration by

the subset of people with FTAs whose FTAs are based on poverty.

129

Additionally, although

individuals with FTAs are unlikely to have the traits in general which would disqualify them for

ODL relief,

130

it remains the case that “a small percentage [of the people who have an FTA] may

have other suspensions that make them ineligible for an ODL, such as [that their license was]

also revoked for a medical reason, a hard suspension/waiting period because of an alcohol or

drug offense, or their driver license is revoked because of non-payment of child support.”

131

Once again, these methodological weaknesses cause us to both underestimate and overestimate

the second chance gap.

Based on the methods, data sources, and caveats described above and further elaborated

in the Appendix, we estimate that around 438,000 people with suspended licenses due to FTA are

eligible for an ODL under Texas Transportation Code Section 521 Subchapter L. Using the data

sources described above, we project that the Texas Department of Public Safety granted 87,027

ODLs over the past five years. Combining these historical ODL figures with our eligibility

calculations, we estimate that 16.58% of people eligible for relief have received it, leaving 83.2%

of people in the “Texas Second Chance Driving Relicensing Gap.” Based on reported records,

the State granted 16,350 ODLs in the last year of fully available data (2019). At this rate, it

would take nearly three decades (27 years) to clear the ODL backlog.

131

Email with contact at Texas Fair Defense Project. Email is on file with authors.

130

For example, on the basis of hard suspension waiting periods due to DWIs, which are Class B misdemeanors or

more serious offenses. See Know Your Rights: Traffic Tickets and Other Class C Misdemeanors, ACLU Tᴇxᴀs, (last

visited August 26, 2021),

https://www.aclutx.org/en/know-your-rights/know-your-rights-traffic-tickets-and-other-class-c-misdemeanors and

Tex. Penal Code §49.06.

129

Deborah Fowler et al., Pay Or Stay: The High Cost of Jailing Texans for Fines & Fees, Tᴇxᴀs Aᴘᴘʟᴇsᴇᴇᴅ ᴀɴᴅ

Tᴇxᴀs Fᴀɪʀ Dᴇғᴇɴsᴇ Pʀᴏᴊᴇᴄᴛ, 17-18, (February 2017),

https://www.texasappleseed.org/sites/default/files/PayorStay_Report_final_Feb2017.pdf.

128

Deborah Fowler et al., Pay Or Stay: The High Cost of Jailing Texans for Fines & Fees, Tᴇxᴀs Aᴘᴘʟᴇsᴇᴇᴅ ᴀɴᴅ

Tᴇxᴀs Fᴀɪʀ Dᴇғᴇɴsᴇ Pʀᴏᴊᴇᴄᴛ, 7, (February 2017),

https://www.texasappleseed.org/sites/default/files/PayorStay_Report_final_Feb2017.pdf. See also Driven by Debt:

The Failure of the Omnibase Program, Tᴇxᴀs Aᴘᴘʟᴇsᴇᴇᴅ ᴀɴᴅ Tᴇxᴀs Fᴀɪʀ Dᴇғᴇɴsᴇ Pʀᴏᴊᴇᴄᴛ, 3, (August 2021),

https://www.texasappleseed.org/sites/default/files/OmniBaseRevenueReport-Aug11-Final.pdf (explaining that

“Holds on license renewals are triggered when license holders either fail to pay fines and costs or fail to appear in

court, usually for traffic offenses. However, eventually all OmniBase Holds are incurred due to an inability to pay,

because the only way to lift a hold is to completely pay off all underlying debt”).

Electronic copy available at: https://ssrn.com/abstract=4065920

PART III: ESTIMATING THE ANNUAL EARNINGS IMPACTS OF A CRIMINAL

CONVICTION AND LOST DRIVER’S LICENSE AND THE EARNING IMPACTS OF

TEXAS’ SECOND CHANCE GAPS 25

A. Summary of Results

Employment and earning opportunities are limited for persons who have criminal records

or who lack driver’s licenses, compared to those without such restrictions. Though second

chance laws make it possible to remove these barriers to work through records sealing or license

restoration, our analysis of Texas’s criminal justice system in the previous part finds that only a

fraction of those apparently eligible for relief are accessing it, with over a million people falling

into either the second chance sealing or occupational driver’s license gap. While second chance

reforms are often generally motivated by a desire to expand economic opportunity, no study of

which we are aware has yet attempted to specify the cost to society, in aggregate earnings and

employment losses, associated with the “second chance gap.” The following paragraphs detail

our attempts to do so, with the results of our analysis shown in Table 2, below.

Table 3: The Size and Annual Earnings Losses Associated with Texas’s Second

Chance Sealing and Driver’s License Gaps

Orders of Non-Disclosure

Drivers License Restoration

National Estimate of Annual Wage

Loss

$5,100

National Estimate of Annual

Wage Loss (lower bounds

estimate)

$12,696

People estimated in the second

chance ODL gap (total clearance)

676,845

Estimate of Annual Wage Loss

(upper bounds estimate)

$23,552

People estimated in the second

chance ODL gap (total clearance)

438,000

Estimate of Annual Earnings Loss

$3,451,909,500

Estimate of Annual Earnings Loss

(lower bounds)

$5,560,848,000

Estimate of Annual Earnings Loss

(upper bounds)

$10,315,776,000

B. Estimating The Earnings Effect of Incarceration and Conviction

A great deal of empirical work on the impact of contact with the criminal legal system on

employment and earnings is limited to incarceration, although a growing body of studies is

Electronic copy available at: https://ssrn.com/abstract=4065920

beginning to push the point of contact up to conviction and even arrest.

132

There are challenges

with these kinds of studies, namely, adjusting for the distinct possibility that individuals who

have contact with the criminal legal system would experience employment problems or wage

erosion even in the absence of that contact. This is known as the selection problem, and it is

exemplified by socioeconomic profiles indicating that individuals who have contact with the

criminal legal system tend to be drawn from lower rungs of the socio-economic ladder. For

example, Harlow reports that just 32% of state prison inmates have a high school diploma,

compared to 82% of the general population.

133

It is for this reason that Wakefield and Uggen

characterize the criminal legal system partly as a social filter that absorbs marginal and

marginalized populations.

134

Furthermore, individuals who experience more punitive forms of

contact with the criminal legal system (e.g., conviction with incarceration as opposed to

conviction with probation) tend to have even more disadvantaged profiles.

135

This would seem to

call into question whether employment difficulty could be interpreted as the causal effect of

contact with the criminal legal system experience, or else a spurious artifact. Despite that

challenge, many studies have taken great care to deal with the selection problem using several

kinds of quasi-experimental designs. There are many such studies that currently exist, but we

focus our attention on the few that are directly relevant for our purpose.

Studies of large national surveys in the U.S. consistently indicate incarceration, even a

very short spell of confinement, is highly disruptive for an individual’s employment prospects.

Apel and Ramakers review many of them and conclude that formerly incarcerated individuals

have a 10-20% lower likelihood of employment than their non-incarcerated peers.

136

This means

that if the baseline employment-to-population ratio is 60%, individuals who have been

incarcerated have a comparable ratio of 48-54%.

137

Among individuals who are employed,

moreover, the wage gap tends to be on the order of 5-30%, meaning when judged against a

baseline wage of $10 per hour, otherwise similar but formerly incarcerated individuals who find

employment earn $7 - $9.50 per hour.

138

Assuming a 40-hour work week, this would amount to a

penalty of $1,040 to $6,240 per year, compared to an annual baseline of $20,800.

139

Id.

138

Id.

137

Id.

136

Robert Apel & Anke Ramakers, Impact of incarceration on employment prospects, 85-104, in Hᴀɴᴅʙᴏᴏᴋ ᴏɴ ᴛʜᴇ

Cᴏɴsᴇᴏ̨ᴜᴇɴᴄᴇs ᴏғ Sᴇɴᴛᴇɴᴄɪɴɢ ᴀɴᴅ Pᴜɴɪsʜᴍᴇɴᴛ Dᴇᴄɪsɪᴏɴs (Beth M. Huebner & Natasha A. Frost, eds., 2019).

135

Id., 389-393. See also Reducing Racial Disparity in the Criminal Justice System: A Manual for Policymakers and

Practitioners, Tʜᴇ Sᴇɴᴛᴇɴᴄɪɴɢ Pʀᴏᴊᴇᴄᴛ (2008) (discussing causes of racial disparity in the criminal justice system,

how these disparities manifest, and what practitioners and policymakers can do to address these disparities).

134

Sara Wakefield & Christopher Uggen, 36 Incarceration and Stratification, Aɴɴᴜᴀʟ Rᴇᴠɪᴇᴡ ᴏғ Sᴏᴄɪᴏʟᴏɢʏ 387

(2010).

133

Caroline Wolf Harlow, Education and Correctional Populations (NCJ 195670), Bᴜʀᴇᴀᴜ ᴏғ Jᴜsᴛɪᴄᴇ Sᴛᴀᴛɪsᴛɪᴄs

(January 2003), https://bjs.ojp.gov/content/pub/pdf/ecp.pdf.

132

For instance, see Robert Apel & Kathleen Powell, Level of Criminal Justice Contact and Early Adult Wage

Inequality, 5 Rᴜssᴇʟʟ Sᴀɢᴇ Fᴏᴜɴᴅᴀᴛɪᴏɴ Jᴏᴜʀɴᴀʟ ᴏғ ᴛʜᴇ Sᴏᴄɪᴀʟ Sᴄɪᴇɴᴄᴇs 198, 199-200 (January 2019) (presenting a

literature review about key studies related to lost earnings and criminal justice contact, and noting “[t]his rich

research tradition is mixed with respect to legal jurisdictions, types of contact, age and representativeness of the

samples, measurement sources, research designs, and methodological rigor”).

Electronic copy available at: https://ssrn.com/abstract=4065920

The “disruption” effect of incarceration is distinct from, but in many cases, further

compounded by, the “marking” effect of having a criminal record, and, in particular, a

conviction, on earnings and employment. Craigie and her coauthors provide the most recent

estimates of the impact of misdemeanor and felony conviction and incarceration on earnings,

which we use as a starting point.

140

They rely on self-reported criminal justice contact in a large,

representative, and longitudinal sample of individuals who were in their 30s the last time they

were interviewed.

141

Their design entails matching individuals who experienced criminal justice

contact to their peers who share the same demographic profile and regional economic

characteristics, but who did not experience criminal justice contact.

142

They estimate lost annual

earnings due to misdemeanor convictions to be $5,100, which is a 16% difference relative to

baseline annual earnings of $32,000, and lost earnings from a felony conviction to be $6,400 (a

22% decline relative to baseline of $29,400).

143

These represent the effects of misdemeanor and

felony conviction relative to no conviction, the latter mostly reflecting no criminal justice

contact, although it could include some number of individuals who experienced forms of

criminal justice processing that did not culminate in a conviction (e.g., arrest, booking). Slightly

larger are the lost earnings from incarceration, which the authors estimate to be $7,100 per year

(a 52% decline relative to baseline of $13,800).

144

This estimate compares individuals who

experienced any post-conviction sentence of incarceration to individuals who did not experience

criminal justice contact as well as individuals who might have experienced other forms of

criminal justice processing (possibly even those held in jail awaiting processing) but were not

sentenced to incarceration.

145

The evidence that is most relevant for our purpose comes from Harris County, Texas.

Mueller-Smith takes advantage of the fact that misdemeanor and felony defendants are randomly

assigned to a courtroom, which determines both the judge that presides over their case as well as

the prosecution team.

146

Because randomization ensures there is no systematic tendency for

defendants with certain characteristics to be assigned to a particular courtroom, variation in

incarceration arises solely from the assigned courtroom. In that sense, variation in punishment

due to the preferential leanings of individual judges can be thought of as being as good as

randomly assigned. The analysis merges administrative data from law enforcement, the court, the

correctional system, and the labor department to pool together individuals who are incarcerated

for the first time with individuals who have prior incarceration spells, and to compare them to

those who could have been incarcerated but were not because they were randomly assigned to a

146

Michael Mueller-Smith, The Criminal and Labor Market Impacts of Incarceration, Unpublished manuscript,

University of Michigan, Ann Arbor (2015).

145

Id., 17-19.

144

Id., 17.

143

Id., 15.

142

Id., 31-33.

141

Id., 31-32.

140

Terry-Anne Craigie et al., Conviction, Imprisonment, and Lost Earnings: How Involvement with the Criminal

Justice System Deepens Inequality, Bʀᴇɴɴᴀɴ Cᴇɴᴛᴇʀ ғᴏʀ Jᴜsᴛɪᴄᴇ, (September 15, 2020),

https://www.brennancenter.org/our-work/research-reports/conviction-imprisonment-and-lost-earnings-how-involve

ment-criminal.

Electronic copy available at: https://ssrn.com/abstract=4065920

less punitive courtroom.

147

Mueller-Smith estimates that, during incarceration, individuals

sentenced for misdemeanors and felonies are less likely to be employed by more than 30

percentage points.

148

Following incarceration, their quarterly employment probability remains

lower by 4-5 percentage points, and their quarterly earnings are lower by 30-40% relative to their

peers who were not incarcerated.

149

In terms of lost earnings, he estimates the annual earnings

gap to be $1,864 due to incarceration for a misdemeanor and $4,706 due to a two-year

incarceration spell for a felony.

150

As a further sign of their difficulty in attaining financial

self-sufficiency following incarceration, formerly incarcerated individuals rely more heavily on

government programs (e.g., food stamps, cash welfare, etc.) after their release.

151

A more recent study by Mueller-Smith and Schnepel focuses on two policy changes in

Harris County related to court use of diversion for certain groups of criminal defendants — one

in 1994 that reduced the use of diversion (for certain property and drug defendants) and another

in 2007 that expanded the use of diversion (for certain low-risk defendants).

152

In this study,

diversion refers to deferred adjudication of guilt, whereby defendants evade a felony conviction

by completing a period of community supervision.

153

This type of sanction is akin to probation,

but whereas probation is an alternative to prison that follows a formal conviction for low-level

offenses, deferred adjudication avoids a conviction outright. The study speaks directly to the

impact on employment of a felony conviction record, since the marginal defendant is one who is

most likely to be sentenced to community supervision, meaning the only difference is whether

they acquire a felony conviction in the process.

154

Their analysis takes advantage of the fact that

these two deferred adjudication policies were implemented abruptly—abrupt changes like this

are frequently referred to as natural experiments, because they allow comparison of individuals

who would have received the same sanction but for the policy change, and who instead receive

different sanctions that are arguably as good as being randomly assigned.

155

The authors estimate

that individuals who benefit from deferred adjudication have a higher quarterly employment

likelihood by at least 15 percentage points (an almost 50% increase over baseline in both

years).

156

They also have substantially higher total earnings — the authors estimate the average

annual earnings loss from a felony conviction to be as high as $8,500 (1994) and at least $4,100

(2007).

157

These employment and earnings effects are long lasting, as they persist for 10 years

after deferred adjudication and, after the 1994 change, for an astonishing 20 years.

158

Id., 883, 902.

157

Id., Table 4 at 901.

156

Id., 900.

155

Id.

154

Id., 884-885.

153

Id., 883.

152

Michael Mueller-Smith & Kevin T. Schnepel, Diversion in the Criminal Justice System, 88 Jᴏᴜʀɴᴀʟ ᴏғ Eᴄᴏɴᴏᴍɪᴄ

Sᴛᴜᴅɪᴇs 883, 884-886 (2021).

151

Id., 30, 32.

150

Id., 47 at Table 7.

149

Id., 28-31.

148

Id., 28.

147

Id., 17-19.

Electronic copy available at: https://ssrn.com/abstract=4065920

In light of their careful design, these two Harris County studies provide credible estimates

of the causal impact of incarceration and conviction, respectively, on employment and earnings.

What is notable about both studies is that, not only do conviction and incarceration worsen short-

and long-term employment and earnings prospects, but they also increase the likelihood of

further entanglement with the criminal legal system. This seems to be especially likely among

individuals with no prior felony convictions, for whom the impacts of a felony conviction on

recidivism and employment are largest.

159

This implies the state of Texas pays not only in terms

of lost wage-related revenue but also in the cost of additional crime and legal processing as well

as the cost of government programs from which justice-impacted individuals seek relief.

Table B provided a summary of the estimates of different forms of criminal justice

sanctions on annual earnings, obtained from the studies described above. The estimates are

comparable even though they derive from different samples and different quasi-experimental

designs. Our particular interest is in lost earnings from conviction, estimates of which are broken

down by misdemeanor and felony conviction.

Table B. Estimates of Lost Annual Earnings from Conviction and Incarceration

Study

Jurisdiction

Impact of Conviction

Impact of Incarceration

Source of

Estimate

Craigie et al. (2020)

National

Misdemeanor: $5,100

Felony: $6,400

$7,100

Table 3

Mueller-Smith

(2015)

Harris County, TX

---

Misdemeanor: $1,864

Felony: $4,786

Table 7

Mueller-Smith and

Schnepel (2021)

Harris County, TX

Felony (1994): $8,536

Felony (2007): $4,144

Table 4

Note: The felony incarceration estimate in Mueller-Smith (2015) assumes a two-year sentence of incarceration, as

he allows the impact to differ by duration of sentence. The estimates from Mueller-Smith and Schnepel (2021) relate

to deferred adjudication policy reforms, through which individuals are able to avoid a conviction record by

completing community supervision.

C. Estimating the Earnings Impact of the Texas Second Chance Expungement (Sealing) Gap

We specifically use the estimates from Craigie et al., since this is the only study to

estimate the impact of both misdemeanor and felony conviction, and since its estimate of

earnings loss from a felony conviction is the midpoint of the two estimates provided by

Mueller-Smith and Schnepel. This leads us to conclude the national average impact of a

159

Id., 902 and 911

Electronic copy available at: https://ssrn.com/abstract=4065920

misdemeanor conviction on earnings is about $5,100 per year, and the impact of a felony

conviction is about $6,400 per year. Although the Mueller-Smith study relies on a control group

that includes not only first-time but also additional convictions, because studies indicate that the

binary step, of going from having to not having a conviction record, or vice versa, is most

significant, we conservatively include in our calculation only people who would be able to seal

their records entirely of convictions, a smaller number than who could receive any relief under

sealing law. It is important to note that these studies, and our figures, are drawn from national

calculations that may not accurately predict the outcomes of Texans with convictions.

160

Thus,

though we use the estimates above to carry out our exercise, we heavily caveat its precise total

for at least this reason.

Multiplying the number of individuals that could entirely clear their records, 58,501, by

$5,100 yields a conservative estimate of approximately $300 million annually in lost earnings. If

we change our estimate to include all in our sample who could get sealing relief (676,845), an

approach that is consistent with the Mueller-Smith and Schnepel research design of comparing

people with various records who received deferred adjudications with those who did not, the

cumulative annual earnings loss balloons to approximately $3.5 billion. These figures represent

forgone earnings due to having one additional criminal conviction relative to no criminal

conviction — in economic parlance, it represents lost earnings on the extensive margin.

One might wonder whether these wages are permanently lost, in light of evidence that a

criminal record stigmatizes individuals in the labor market, and sustained time out of the labor

market erodes human capital even further. This suggests that initial demand-side barriers to work

have the capacity to become self-sustaining as they crystallize into experience gaps that make it

even more difficult to secure gainful employment. With respect to the latter possibility, surveys

of employers indicate they are less likely to want to hire individuals with a “spotty work record”

as opposed to individuals with only a GED, and even compared to individuals who have been

unemployed for a year or more.

161

Although employers are least likely to want to hire someone

with a criminal record, then, criminal record expungement might not suffice to fully recover lost

earnings, since experience gaps that accumulate over time because of the criminal record cannot

be fully remedied.

Even if some employment erosion is permanent, there are compelling reasons to believe

that expungement policy can lead to the recovery of some fraction of lost earnings. This is

especially likely to be the case for lost earnings that are due to, for example, restrictions on

occupational licensing for individuals with a conviction record, employer refusals to interview or

hire individuals who have a conviction record (either by requiring it on a job application or via

background check), or employer channeling of individuals with conviction records into

lower-paying positions. One relevant finding from the Mueller-Smith and Schnepel study is that

individuals who benefitted from deferred adjudication were more likely to find work in

industries that were otherwise closed off to individuals with conviction records (e.g., retail trade,

161

Harry J. Holzer et al., Perceived Criminality, Criminal Background Checks, and the Racial Hiring Practices of

Employers, 49 J.L. & Eᴄᴏɴ. 451 (2006).

160

Thank you to Craigie et al. for raising this point with us.

Electronic copy available at: https://ssrn.com/abstract=4065920

health care, educational services).

162

Additional evidence comes from the immediate employment

and wage gains that accrue to individuals with a criminal record who obtain expungement.

163

These findings suggest some amount of lost earnings because of employer stigma and legal

barriers are indeed recoverable.

D. The Earnings Effect of a Suspended Driver’s License

The research base on the economic impact of driver’s license suspension is less well

developed, but we wish to draw attention to this work despite its provisional status. A study of

driver history records in New Jersey found, remarkably, just 10% of license suspension orders

were for driving-related behavior that could be construed as a road safety concern (e.g., driving

under the influence, points accumulation from moving violations, uninsured driving, reckless

driving, etc.).

164

The lion’s share of suspensions was either for financial reasons (40%, e.g.,

failure to pay motor vehicle authority surcharge, comply with a court-ordered installment plan, or

pay child support) or for non-appearance in court to satisfy a summons (32%).

165

In the same

study, a survey of drivers with suspended licenses sought to understand the collateral effects of

license suspension. Among more than 300 respondents, 40% reported they were unable to keep

their job after their suspension, with low-income individuals far more likely to suffer job loss

(64% of those with income under $30,000).

166

Almost one-fifth (17%) of the sample reported not

only losing their job but also being unable to find a new one (or 32% of low-income

respondents).

167

Among those who were able to find another job, almost all reported negative

effects on their income, but regrettably, the estimated amount of lost income was not included in

the survey.

168

In a separate study, Pawasara and Quinn assessed the impact of having a valid

driver’s license among welfare recipients as part of the administration of a training program

carried out in Milwaukee.

169

They report that among program participants with less than 12 years

of education, individuals with a valid driver's license were “four times more likely to show

earnings above the poverty level, compared to those without a current license.”

170

The single study we found that attempted to specifically quantify the impact on earnings

of lacking a license was of the city of Phoenix’s “Compliance Assistance Program (CAP),”

170

John Pawasrat & Lois M. Quinn, Research Brief on ETI Driver’s License Studies, 186 ETI Pᴜʙʟɪᴄᴀᴛɪᴏɴs 1, 4

(2017).

169

John Pawasrat & Lois M. Quinn, The EARN (Early Assessment and Retention Network) Model for Effectively

Targeting WIA and TANF Resources to Participants, 60 ETI Pᴜʙʟɪᴄᴀᴛɪᴏɴs 1 (2007).

168

Id., 56.

167

Id.

166

Id, 55-57.

165

Id.

164

Jon A. Carnegie et al., Driver’s License Suspensions, Impacts and Fairness Study, Nᴇᴡ Jᴇʀsᴇʏ Dᴇᴘᴀʀᴛᴍᴇɴᴛ ᴏғ

Tʀᴀɴsᴘᴏʀᴛᴀᴛɪᴏɴ, 33 (August 2007).

163

J.J. Prescott & Sonja B. Starr, Expungement of Criminal Convictions: An Empirical Study, 133 Hᴀʀᴠ. L. Rᴇᴠ.

2460 (2020); Jeffrey Selbin et al., Unmarked? Criminal Record Clearing and Employment Outcomes, 108 J. Cʀɪᴍ.

L. & Cʀɪᴍɪɴᴏʟᴏɢʏ 1 (2018).

162

Michael Mueller-Smith & Kevin T. Schnepel, Diversion in the Criminal Justice System, 88 Jᴏᴜʀɴᴀʟ ᴏғ Eᴄᴏɴᴏᴍɪᴄ

Sᴛᴜᴅɪᴇs 883, 900 (2021).

Electronic copy available at: https://ssrn.com/abstract=4065920

which for a period of time reinstated driving privileges of individuals with civil traffic violations

and unpaid fines and fees.

171

The program restored licenses along with waiving some fees in

exchange for enrolling in a payment plan. From in-person and mail surveys of 75 program

enrollees, 31% of those who were employed reported losing their job as a direct consequence of

their suspended license, and 57% reported working fewer hours. Across the sample, the median

income loss was estimated at $36,800 (with a range of $1,200 to $300,000).

172

The estimates provided here should be regarded as descriptive and therefore highly

provisional. We rely on the Phoenix study to estimate the impact of license suspension on

income, and although some of the needed information is not provided in the study, it is the only

study to date that attempts to quantify income loss. Because the sample comprises individuals

who enrolled in the CAP program, they might have been especially motivated to do so because

of job difficulty.

173

Acknowledging these and other limitations, we take at face value the estimate

of lost annual income for the typical (median) person with a suspended license to be $36,800, but

scale this downward to account for individuals who were not working or whose work was not

disrupted by suspension. We note that 36% of the sample was not working at the time of license

suspension, and among the workers, we estimate that 72% experienced disruption (either because

of job loss [31%] or reduction in hours [57%], with 72% representing our effort to adjust for

double counting).

174

Using this information, we conservatively estimate that license suspension

results in about $12,700 in lost annual earnings.

175

Multiplying this by the number of people

eligible for an occupational driver’s license (438,000), yields an estimate of around $5.5 billion

in lost earnings yearly due to not having a driver’s license.

176

While further research is certainly needed, recent qualitative studies and accounts of

“unlicensing” support an impact on earnings that is at least as large, if not larger, than the

estimate we report above. Job loss and job disqualification, and lower-wage job options, appear

to be the main mechanisms. For example, through semi-structured phone interviews conducted in

2020 and 2021, Sartin and her co-authors found suspended license “employment challenges” to

present in several ways: directly, by preventing subjects from applying for the numerous jobs

that mandated licenses, on or off-the job, and indirectly, in terms of the type of work and level of

176

If we instead use $23.5K as our estimate of in lost earnings annually, based on assuming that all 64% of people

working at the time of the suspension had income around the median ($23.5K = $36,800 *.64), our estimate of lost

earnings balloons to $10B.

175

We attempt to be conservative in our use of the information reported in the Phoenix CAP study. Our estimate

assumes 36% of the sample was not working and lost $0; 18% of the sample was working but did not experience

disruption and lost $0; 23% of the sample was working and lost $18,400 (the midpoint of $0 and the median

$36,800); and 23% of the sample was working and lost $36,800. Using the percentages as weights for lost income,

the mean is $12,696.

174

Id.

173

Indeed, the authors shy away from using the analysis “for statistical analysis or inferences.” Id., at 11.

172

The City of Phoenix Municipal Court’s Compliance Assistance Program, 2016: An Economic Assessment, L.

Wɪʟʟɪᴀᴍ Sᴇɪᴅᴍᴀɴ Rᴇsᴇᴀʀᴄʜ Iɴsᴛɪᴛᴜᴛᴇ, 22 (June 2, 2017),

https://finesandfeesjusticecenter.org/content/uploads/2018/11/Phoenix-license-restoration-pilot-THE-CITY-OF-PHO

ENIX-MUNICIPAL-COURT%E2%80%99S-COMPLIANCE-ASSISTANCE-PROGRAM.pdf.

171

Compliance Assistance Program (CAP), Cɪᴛʏ ᴏғ Pʜᴏᴇɴɪx, (last viewed January 22, 2022),

https://www.phoenix.gov/court/cap.

Electronic copy available at: https://ssrn.com/abstract=4065920

pay available to persons without licenses.

177

Dindial and her co-authors, also interviewing

subjects across states, concluded that suspended licenses “creat[ed] obstacles to finding or

keeping paid work.”

178

Jobs unavailable to people without licenses include driving for Uber/Lyft

and delivery and driving-related jobs, automobile sales and services, home healthcare aides, and

the construction trade.

179

Positions including retail security officer (pay: $10.55 an hour), a

caregiver for the disabled ($10 an hour), an eye-care associate, an administrative assistant, and a

deli clerk have also been reported to require a license.

180

Studies show that employers also use

licenses as screens

181

or positive signal: “positions use a driver’s license as a proxy for whether

you’re employable.”

182

In Texas, 20.8% of civilian jobs in 2016 required driving a passenger vehicle.

183

The

state’s size and infrastructure have meant that 80% (+/- 0.2) of workers in Texas at least 16 years

old drove alone (i.e., did not carpool or rideshare) to work in 2016.

184

When a license is

reinstated, it can quickly translate into an earnings opportunity as the experience of 47-year old

John Blackwell illustrates.

185

Once Blackwell got his license back, Blackwell, who works as a

carpenter foreman, “c[ould] make an additional $2 an hour [with the ability to]. . . drive a

company truck.”

186

Assuming that Blackwell works 40 hours per week, the extra $2 per hour

adds up to an extra $4,480 in wages per year.

187

E. Comparing the Cumulative Earnings Effect of Texas’s Paper Prisons with the

Out-of-Pocket Cost of Texas’s Physical Prisons

187

$2/hour x 40 hours per week x 52 weeks per year = $4,480

186

Id.

185

Angie Jackson, Penalized for being poor: Michigan drivers could get break on license suspensions, Dᴇᴛʀᴏɪᴛ Fʀᴇᴇ

Pʀᴇss, (updated August 1, 2020),

https://www.freep.com/story/news/local/michigan/2020/08/01/michigan-drivers-license-suspensions/4819150002/.

184

United States Census Bureau, Percent Of Workers 16 Years And Over Who Traveled To Work By Car, Truck, Or

Van--Drove Alone, Aᴍᴇʀɪᴄᴀɴ Cᴏᴍᴍᴜɴɪᴛʏ Sᴜʀᴠᴇʏ, (last visited August 2, 2021),

https://www.census.gov/acs/www/data/data-tables-and-tools/geographic-comparison-tables/.

183

Bureau of Labor Statistics, U.S. Department of Labor, 30 percent of civilian jobs require some driving in 2016,

Tʜᴇ Eᴄᴏɴᴏᴍɪᴄs Dᴀɪʟʏ, (June 27, 2017),

https://www.bls.gov/opub/ted/2017/30-percent-of-civilian-jobs-require-some-driving-in-2016.htm.

182

Alana Semuels, No Driver’s License, No Job, Tʜᴇ Aᴛʟᴀɴᴛɪᴄ, (June 15, 2016),

https://www.theatlantic.com/business/archive/2016/06/no-drivers-license-no-job/486653/.

181

Jon A. Carnegie et al., Driver’s License Suspensions, Impacts and Fairness Study, Nᴇᴡ Jᴇʀsᴇʏ Dᴇᴘᴀʀᴛᴍᴇɴᴛ ᴏғ

Tʀᴀɴsᴘᴏʀᴛᴀᴛɪᴏɴ, 3, 66 (August 2007).

180

Joseph Shapiro, How Driver’s License Suspensions Unfairly Target the Poor, NPR, (January 5, 2015),

https://www.npr.org/2015/01/05/372691918/how-drivers-license-suspensions-unfairly-target-the-poor.

179

Jon A. Carnegie et al., Driver’s License Suspensions, Impacts and Fairness Study, Nᴇᴡ Jᴇʀsᴇʏ Dᴇᴘᴀʀᴛᴍᴇɴᴛ ᴏғ

Tʀᴀɴsᴘᴏʀᴛᴀᴛɪᴏɴ, 3 (August 2007).

178

Reckless Lawmaking: How Debt-Based Driver’s License Suspension Laws Impose Harm and Waste Resources,

ACLU, 26 (2021),

https://www.aclu.org/sites/default/files/field_document/reckless_lawmaking_aclu_final_4.19.21.pdf.

177

Emma B. Sartin et al., Impacts of Non-Driving Related License Suspensions on Quality of Life: A Qualitative

Study, Aᴘᴘʟɪᴇᴅ Rᴇsᴇᴀʀᴄʜ ɪɴ Qᴜᴀʟɪᴛʏ ᴏғ Lɪғᴇ 1, 11-12 (2022).

Electronic copy available at: https://ssrn.com/abstract=4065920

Aggregating the most conservative estimates of lost earnings associated with the second

chance sealing and second chance relicensing gaps described above yields a total that ranges

from $5.6 billion (assuming complete overlap) to $6 billion (assuming no overlap). By way of

comparison the entire 2022 operating budget for the Texas Department of Criminal Justice was

$3.7 billion.

188

Taken at face value, these calculations would suggest that the cost of Texas’s “paper

prisons,” in terms of lost earnings, is as large as if not larger than the out-of-pocket cost of its

physical prisons. Provided for purposes of illustration, they suggest that, although often

motivated by considerations of fairness, second chance rules have significant and overlooked

earnings and employment consequences.

PART IV: AUTOMATION AND POLICY PILOTS

The previous part considered the order of magnitude effects, in terms of lost earnings,

associated with the failed delivery of second chances in Texas. What this exercise underscores is

not only the large number of people for whom second chances appear to have largely been

missed chances, but also, the large economic and earnings impact associated with the second

chance gap, and potential to reverse, at least partially, these economic losses with policies that

close the gaps. Ways to both narrow the second chance gap and the gap in understanding its

economic impact, are discussed in this part. We start by considering the drivers of Texas’s second

chance gaps, then the reforms that could narrow them. We recommend the use of policy pilots to

both introduce policy interventions and provide the research infrastructure for observing their

outcomes.

A. Drivers of Texas’s Second Chance Gaps

Why don’t more people seek to clear their criminal convictions or reinstate their lost

driver’s licenses, given their negative impact on earnings and employment? The processes of

obtaining relief presented in Section I confirm a number of the same barriers to relief

documented by others, including lack of awareness, unclear criteria, burdensome application

processes, and fines and fees

189

to impede uptake of the relief offered in Texas.

Texas law puts the impetus on the individual to initiate the process or relief-seeking

through a petition. But in many cases, there is a lack of awareness about the second chances to

which one may be entitled. The awareness gap is particularly acute when a second chance

opportunity is first introduced, or changes are made to its eligibility. There is typically no

specific advertising or communications budget allocated for alerting eligible citizens that they

can apply for relief — indeed, to specify such a budget would likely reduce the chances of

enactment in a climate of fiscal austerity. This is a problem particularly in the case of records

189

Colleen Chien, America’s Paper Prisons: The Second Chance Gaps, 119 Mɪᴄʜ. L. Rᴇᴠ. 519 (2020)

188

Operating Budget for Fiscal Year 2022, Tᴇxᴀs Bᴏᴀʀᴅ ᴏғ Cʀɪᴍɪɴᴀʟ Jᴜsᴛɪᴄᴇ, 2, (December 1, 2021),

https://www.tdcj.texas.gov/documents/bfd/FY2022_Operating_Budget_LBB.pdf

Electronic copy available at: https://ssrn.com/abstract=4065920

relief provisions, which are often updated frequently. Since record sealing was introduced in

2003, legislative acts in 2015 and 2017 have changed who is eligible, making it difficult for even

the diligent to know when and if the law may apply to them. Similarly, very few people are

aware that ODLs are an option that would allow them to drive to work or take care of other

essential tasks.

190

As one report stated, “most people with. . . [driver’s license] holds do not know

that ODLs are an option and do not apply for them.”

191

Even for those that are aware, the process

of applying for an ODL has been described as “labyrinthine. . . just determining which court to

file in can be impossible for many people to figure out on their own.”

192

Insufficient information

about how to reinstate licenses has also been cited as a barrier to application in a number of

states with programs similar to Texas.’

193

Ambiguous or undefined criteria

194

and discretion-based processes also contribute to the

gap. They discourage uptake because they make the payoff from the application process

uncertain. This is particularly a problem because of the costly, burdensome, and informationally

intensive processes described in Part I. As one of us has previously written, “[i]n the same way

that the accused remain innocent until proven guilty in the U.S. criminal justice system, many

194

For example, people convicted of first-time misdemeanors punishable by fine only and misdemeanors carrying a

sentence of incarceration are ineligible to have their records sealed if the misdemeanor in question was a violent

crime and/or a crime of a sexual nature, with the exception of offenses under §22.01, Penal Code (Tex. Admin. Code

§411.0735). But Code of Criminal Procedure §59.01(3) defines a crime of violence as “...(A) any criminal offense

defined in the Penal Code or in a federal criminal law that results in a personal injury to a victim... [or] (B) an act

that is not an offense under the Penal Code involving the operation of a []vehicle [] that results in injury or death…,”

facts that are impossible to be evaluated without delving into the court record. Additionally, although some offenses

like rape are obviously of a sexual nature, we could not find any clear definition of a crime of a sexual nature in the

penal code. As a result, determining whether an offense is of a violent or sexual nature is not straightforward. The

phrase the “same criminal episode” has also raised statutory interpretation challenges: “A court may not order the

expunction of records and files relating to an arrest for an offense for which a person is subsequently acquitted,

whether by the trial court, a court of appeals, or the court of criminal appeals, if the offense for which the person was

acquitted arose out of a criminal episode, as defined by Section 3.01, Penal Code, and the person was convicted of

or remains subject to prosecution for at least one other offense occurring during the criminal episode” (Tex. Code of

Criminal Procedure Art. 55.01(c) (1977)). Oddly, two offenses can be considered to have occurred under the same

criminal episode regardless of how many years apart the offenses occurred (HB 2684 (Canales): Clean Slate

Eligibility, Cʟᴇᴀɴ Sʟᴀᴛᴇ Tᴇxᴀs, (last visited August 24, 2021),

https://de203503-6f74-4104-bc1d-dc9fde7da5dd.filesusr.com/ugd/413c6c_54294486d6644a2194cd9577f4799d35.p

df). In cases such as In re Expunction of M.T.R. (2020), Texas courts have recently interpreted this to mean that “if a

person is convicted of one offense, then later is acquitted of a ‘same or similar offense,’ the acquitted case is

ineligible for expunction,” an outcome that advocates have described as leading to “absurd results” (Clean Slate

Texas, HB 2684 (Canales): Clean Slate Eligibility, (last visited August 24, 2021),

https://de203503-6f74-4104-bc1d-dc9fde7da5dd.filesusr.com/ugd/413c6c_54294486d6644a2194cd9577f4799d35.p

df). Determining whether an offense is the same or similar insofar as expunction eligibility is concerned makes the

process of determining whether someone should get a second chance more involved and costly.

193

Reckless Lawmaking: How Debt-Based Driver’s License Suspension Laws Impose Harm and Waste Resources,

ACLU, 20 (2021),

https://www.aclu.org/sites/default/files/field_document/reckless_lawmaking_aclu_final_4.19.21.pdf (“Some felt the

steps and procedures were unclear, while others had to do their own research to figure out what to do, contacting

several different agencies and even going back and forth to several locations while legally prevented from driving”).

192

Id.

191

Id.

190

Driven By Debt, Tᴇxᴀs Aᴘᴘʟᴇsᴇᴇᴅ ᴀɴᴅ Tᴇxᴀs Fᴀɪʀ Dᴇғᴇɴsᴇ Pʀᴏᴊᴇᴄᴛ, (December 13, 2018),

https://report.texasappleseed.org/driven-by-debt/.

Electronic copy available at: https://ssrn.com/abstract=4065920

second chance programs require defendants to ‘prove’ that they deserve second chances before

awarding them. As such, getting one’s second chance through petition-based processes may

include enduring a bureaucratic and expensive process, amassing information through a variety

of sources, and being evaluated by an adjudicative or administrative body.”

195

Petition filing fees

196

as well as the requirement to satisfy unpaid fines, fees, and

surcharges also present substantial barriers. Indeed, the requirement that fines and fees be paid

off may be the most important barrier to relief to restored driver’s licenses. One prime example

of this is the OmniBase Program, which has no fee-waiving mechanism.

197

“OmniBase holds

aren’t lifted even when somebody appears in court and gets on a payment plan or begins

assigned community service. The only way to lift a hold is to satisfy the fines and costs in

full.”

198

This lack of fee-waiving mechanism is critical because “[m]ore than one in four

OmniBase holds between 2013 and 2017 were a result of poverty-related offenses such as

driving without insurance, displaying expired license plates, driving with an invalid license, or

no driver’s license at all.”

199

B. Narrowing Texas’s Second Chance Gaps

The most impactful way to improve the administration of second chances would be to

eliminate fines and fees requirements, as other states have done,

200

and to streamline delivery

through automation. Due to different administrative aspects of the laws, automation would look

200

See PA HB440, which in 2020, eliminated the requirement in Pennsylvania that fines and fees be paid off prior to

receiving “Clean Slate” relief. Described at Aaron Moselle, Pa. residents with court debts could have their records

automatically sealed under new bill, WHYY (October 22, 2020),

https://whyy.org/articles/pa-residents-with-court-debt-could-have-their-records-automatically-sealed-under-new-bill/

199

Driven By Debt: How Driver’s Licenses Suspensions for Unpaid Fines and Fees Hurt Texas Families, Tᴇxᴀs Fᴀɪʀ

Dᴇғᴇɴsᴇ Pʀᴏᴊᴇᴄᴛ ᴀɴᴅ Tᴇxᴀs Aᴘᴘʟᴇsᴇᴇᴅ, (March 2, 2019),

https://finesandfeesjusticecenter.org/articles/license-suspensions-hurt-texas-families/. For full report, see Driven By

Debt, Tᴇxᴀs Aᴘᴘʟᴇsᴇᴇᴅ ᴀɴᴅ Tᴇxᴀs Fᴀɪʀ Dᴇғᴇɴsᴇ Pʀᴏᴊᴇᴄᴛ, (December 13, 2018),

https://report.texasappleseed.org/driven-by-debt/.

198

Emily Gerrick & Mary Mergler, Commentary: Lawmakers need to fix another program that buries Texas drivers

in fines, Sᴛᴀᴛᴇsᴍᴀɴ, (updated July 5, 2019),

https://www.statesman.com/opinion/20190705/commentary-lawmakers-need-to-fix-another-program-that-buries-tex

as-drivers-in-fines.

197

Emily Gerrick & Mary Mergler, Commentary: Lawmakers need to fix another program that buries Texas drivers

in fines, Sᴛᴀᴛᴇsᴍᴀɴ, (updated July 5, 2019),

https://www.statesman.com/opinion/20190705/commentary-lawmakers-need-to-fix-another-program-that-buries-tex

as-drivers-in-fines.

196

For example, the cost of expungement $750 to $2,500, not including the cost of traveling to the hearing as well as

filing fees, which, if not waived, can add an additional $200 and $400 Fees and Costs, Dᴀʟʟᴀs Exᴘᴜɴᴄᴛɪᴏɴ

Aᴛᴛᴏʀɴᴇʏs, (last viewed October 14, 2021), https://www.dallas-expunction-attorney.com/fees-and-costs/. Based on

other research, getting an expunction could be slightly less expensive than sealing one’s record. See Ruthie

Goldstein, Moving Forward: Proposals for Expanding and Simplifying Expungements in Texas, 98 T. Lᴀᴡ Rᴇᴠ. 144

(2020) (noting that “Because expunctions can only be granted after filing a petition with the court and because

drafting the petition typically requires the assistance of an attorney, the cost of an expunction can total between $500

and $2,500, not including filing fees. Filing fees typically total between $200 and $400, but can be waived for

indigency.”)

195

Colleen Chien, America’s Paper Prisons: The Second Chance Gaps, 119 Mɪᴄʜ. L. Rᴇᴠ. 519 (2020) at 540-541.

Electronic copy available at: https://ssrn.com/abstract=4065920

slightly different for each type of second chance. For instance, automatically sealing records

would ideally follow the Clean Slate model, which uses computer queries to identify eligible

individuals and then automatically seals those people’s records. A similar program could

automatically look at whether individuals meet the ODL eligibility criteria and invite eligible

individuals to apply.

Automation makes the application process less burdensome in four ways. First,

automation would reduce the awareness gap for individuals seeking second chances because it

would not only identify eligible individuals, but also either (i) invite individuals eligible for relief

to apply, or (ii) provide relief to eligible individuals. Second, automation makes it so that people

seeking to seal their records do not need to complete a petition, thereby “streamlin[ing]

petition-based record clearing— a costly and time-intensive process that prevents the vast

majority of eligible people from ever obtaining needed relief.”

201

Third, automation could reduce

the number of individuals who need an attorney to help them navigate complicated application

processes. Fourth, automation, coupled with the elimination of fines and fees, would remove the

financial obstacles to getting a second chance. Notably, however, the reduction of fines and fees

would apply only to record sealing, at least based on the policy changes this article proposes. As

such, the driver’s relicensing gap might remain large until policymakers find a way to either (i)

make fee waivers more accessible, or (ii) eliminate filing fees.

There are at least three types of concerns that relate to automating second chances. One is

that developing the systems needed to identify and notify eligible individuals requires time,

money, and oversight.

202

Although it is true that Texas will need to invest in technology to

automate second chances, it is worth remembering that we roughly estimate that Texans lose up

to $3.5 billion annually in lost earnings, and associated tax revenue, just from the second chance

sealing gap. This may be why a financial report for H.B. 3601, a proposed clean slate bill from

the 87th Legislature, reported that “[n]o significant fiscal implication to the State is

anticipated.”

203

Indeed, the tax revenue associated with higher earnings would be expected to

save the state money in the long run.

The second concern about automating second chances is that automating the current law

cannot actually be done effectively due to the unclear criteria and data deficiencies highlighted

earlier in this article. For instance, some scholars express concern that automating expunctions

203

87th Regular Legislative Session, Lᴇɢɪsʟᴀᴛɪᴠᴇ Bᴜᴅɢᴇᴛ Bᴏᴀʀᴅ, (April 10, 2021),

https://capitol.texas.gov/tlodocs/87R/fiscalnotes/pdf/HB03601I.pdf#navpanes=0.

202

See Ruthie Goldstein, Moving Forward: Proposals for Expanding and Simplifying Expungements in Texas, 98 T.

Lᴀᴡ Rᴇᴠ. 144 (2020) (explaining that “[u]ltimately, in a state as large as Texas, it might not be financially feasible to

require that expunctions become automatic.” However, recognizing that fees pose a barrier to people seeking

expunction, Goldstein proposes waiving filing fees as an alternative to automation).

201

Clean Slate Toolkit: Frequently Asked Questions, Cᴇɴᴛᴇʀ ғᴏʀ Aᴍᴇʀɪᴄᴀɴ Pʀᴏɢʀᴇss, (November 15, 2018),

https://www.americanprogress.org/issues/poverty/reports/2018/11/15/460907/clean-slate-toolkit/. Several states have

implemented clean slate policies with great success. For example, Pennsylvania’s 2018 clean slate bill sealed over

47 million offenses and helped an estimated 1.1 million people within one year of its launch. Laurie Mason

Schroder, In one year, Pa.’s Clean Slate law has erased 35 million crimes. What’s next?, Tʜᴇ Mᴏʀɴɪɴɢ Cᴀʟʟ, (June

30, 2020),

https://www.mcall.com/news/breaking/mc-nws-pennsylvania-clean-slate-law-one-year-20200630-ges77qb3ffahhizn

bjzjtelq7q-story.html.

Electronic copy available at: https://ssrn.com/abstract=4065920

might not be realistic in Texas, “where expunctions cannot be granted until the statute of

limitations has passed. Statutes of limitation vary by crime and class of crime, and keeping track

of their expiration would put a significant burden on counties, which would most likely be

responsible for processing the automatic expunctions. Additionally, assuming an ‘automatic

expunction’ [or automatic sealing] bill is retroactive, it would be extremely time-consuming for

counties to look back into their old records, determine what can be expunged, and go through the

expunction process.”

204

So concerns are likely true for record sealing and ODLs. However, we

note that although automating second chances “would require significant data normalization and

cleaning efforts,”

205

reforming the law to make it implementable at scale can provide a path

forward. To illustrate, we discuss clearance criteria challenges and legislative drafting

alternatives to overcome these challenges in Appendix F. This could serve as a foundation for

starting to adjust Texas’ laws to the reality of Texas’ data.

The third concern is more general, about the effectiveness of records relief as a way to

achieve the policy aims of public safety, job growth, and equity. The potential harms of statistical

discrimination replacing records-based exclusion, a lack of knowledge of sealed records

rendering the policy impactless, and the risk of automating relatively minor crimes intensifying

the earnings penalty for people with more significant records have all been previously raised.

206

C. Narrowing Knowledge Gaps about the Earnings Impact of Sealing and Driver’s

License Restoration

The introduction of an intervention to narrow the second chance gap at the county or state

level presents a chance not only to narrow the second chance gap, but also to narrow the

knowledge gap about the associated earnings impact. Policy pilots, which temporarily change a

law or policy so lawmakers and scholars can learn from it,

207

can provide an excellent way to

overcome “the knowledge gap and risk aversion in policy development.”

208

There are at least two ways that policy pilots could improve the delivery of second

chances. First, carrying out an automation trial could help lawmakers identify ambiguity in the

existing statute and pinpoint how it needs to be adjusted to support records clearance at scale.

Second, policy pilots could be used to study the impact on earnings of newly available second

chance relief, for example, in order to carry out the cost-benefit to the state of such a policy.

Such policy pilots would not only provide valuable information about the true cost of Texas’s

second chance gaps, but would also provide lawmakers with additional information to ensure

that they pursue cost-effective policy solutions.

208

Id.

207

Colleen V. Chien, Rigorous Policy Pilots: Experimentation in the Administration of Law, 104 Iᴏᴡᴀ L. Rᴇᴠ. 2313

(2019) at 2316.

206

Chien, supra note ___ at ___.

205

Colleen Chien et al., The Texas Second Chance Non-Disclosure/Sealing Gap, Pᴀᴘᴇʀ Pʀɪsᴏɴs Iɴɪᴛɪᴀᴛɪᴠᴇ, (last

visited September 19, 2021), https://www.paperprisons.org/states/TX.html at 2.

204

Ruthie Goldstein, Moving Forward: Proposals for Expanding and Simplifying Expungements in Texas, 98 T. Lᴀᴡ

Rᴇᴠ. 144 (2020).

Electronic copy available at: https://ssrn.com/abstract=4065920

The best prospect would be for a study conducted at the individual level, involving

matching of administrative records from multiple agencies. This might include the state law

enforcement agency that maintains a criminal history repository with arrests and charges and

local and county courts that retain data on convictions and sentences along with non-penal

sanctions such as fines and fees. Other relevant agencies include the state motor vehicle authority

with data on license suspensions and traffic violations and state labor and welfare agencies that

collect data on earnings and transfer payments. To ensure that the impact of closing a second

chance gap can be discerned using experimental or quasi-experimental methods, local policy labs

such as the Texas Policy Lab at Rice University or the Texas Public Policy Foundation think tank

could be appointed to consult on the policy’s roll-out.

209

Upstream attention to implementation

will support downstream learning about interventions to boost the workforce.

CONCLUSION

This article provides an estimate of the annual earnings loss associated with a conviction,

finding it to be on the order of $5,100 to $6,400 per year, as well as a provisional estimate of the

annual earnings loss associated with a suspended driver’s license, conservatively estimating it to

be approximately $12,700. Using Texas as a case study, it quantifies the size of the state’s second

chance sealing and occupational driver’s license gaps, estimating that around 670,000 Texans

could seal their misdemeanor convictions and about 438,000 Texans are eligible for ODLs.

Mechanically aggregating these figures results in a earning loss estimate of $3.5 billion in

connection with the second chance sealing gap and $5.5 billion from the second chance ODL gap

in Texas, in comparison with the operating budget of Texas's correctional system of $3.6 billion.

This suggests not only that the costs of paper prisons could exceed the cost of physical prisons,

but also that closing paper prisons, through automation, can have fiscal benefits that resemble or

far exceed closing physical prisons, although rigorous evaluation and research are needed to see

whether this is the case.

209

Colleen V. Chien, Rigorous Policy Pilots: Experimentation in the Administration of Law, 104 Iᴏᴡᴀ L. Rᴇᴠ. 2313

(2019).

Electronic copy available at: https://ssrn.com/abstract=4065920

APPENDIX

APPENDIX A: RULES

To estimate the number and share of people eligible for but not receiving sealing,

compassionate release, and driver’s license restoration in Texas, we used the Methodology

established in Section II. For more information about our approach, please see Colleen V. Chien,

America's Paper Prisons: The Second Chance Gap, Michigan Law Review (2020).

1. Sizing the Texas Sealing Second Chance Gap

To estimate the number of Texans eligible for records sealing, we focused narrowly on

the subset of people that could, if eligibility criteria were automatically applied to them,

transition from having a misdemeanor conviction or deferred adjudication to not having a

conviction or deferred adjudication under Texas Gov. Code Chap. 411. To conduct our analysis,

we acquired the CCH from the Texas Department of Public Safety.

210

To support our analysis of

the uptake rate of sealing, the DPS also sent us administrative data containing historical

information about the number of people who had their records sealed annually.

We imported and then inner joined the data to create unique person IDs for each person.

Once we had created unique IDs, we randomly selected 153,674 individuals for our sample. To

make sure our data was consistent, we removed about forty individuals whose sentence

expiration dates were earlier than their sentence start dates, leaving us with a random sample of

153,632 individuals. We then turned to cleaning the data, focusing first on categorizing

disposition data.

211

Likewise, we categorized the following dispositions as non-convictions: not

guilty, dismissed, mistrial, abandoned charges, waived, acquitted, community supervision

expired, deferred, unadjudicated with, dead, pending, dismissed, dismissed – mentally

incompetent, and pending – mental incompetence. After categorizing the dispositions, we turned

to cleaning the sentencing data. Though 2,041,483 records in our sample had sentence

completion data, for 554,444 records, the data was missing. Rather than exclude these records

from our analysis, we instead calculated, from the cases where the data was present, the average

lag between sentencing and sentence completion and added these numbers (2.9 years after the

date of sentencing and 3.2 years for misdemeanors and felonies, respectively), and used this to

roughly estimate sentence completion dates.

To be conservative in our estimates about the number of individuals eligible for record

sealing, we made a number of assumptions when conducting our analysis. First, since we could

211

Based on consultation with criminal law experts, we categorized the following dispositions as

convictions: convicted, multiple charges, commuted, amend probation, convicted – appeal

pending, pardoned, probation discharge, probation revocation, sentence modified, convicted -

lesser charge, and mistrial.

210

The Criminalized Computer History (CCH) More information about the CCH is available at Texas Department

of Public Safety, About CCH, (last visited August 20, 2021),

https://publicsite.dps.texas.gov/DpsWebsite/CriminalHistory/AboutCch.aspx.

Electronic copy available at: https://ssrn.com/abstract=4065920

not easily determine whether offenses were of a violent or sexual nature, we manually determine

whether fourteen offenses would be ineligible due to being offenses of a violent or sexual

nature.

212

When in doubt after reading the statutes associated with these fourteen offenses, we

assumed that they were ineligible to be conservative in our estimates.

213

Second, we assume that

offenses that appeared in the data as misdemeanors but had a statute indicating an enhancement

to a state jail felony or a felony were ineligible.

214

Third, if someone has ever been convicted of

an offense listed in Tex. Gov. Code 411.074(b), that person cannot have any of the offenses on

their record sealed. Most of those offenses were easy to model, but some depended on facts of

the case that we did not have access to. In these cases, we assumed that anyone convicted of

these offenses were ineligible.

215

Finally, we assumed that someone was incarcerated if the

number of days between the disposition date and the sentence expiration date was greater than

two years.

216

When the data contains both sentence expiration date and a parole date for the same

person and offense, we assume that any sentence expiration date that is greater than the parole

until date indicates time in prison.

Despite our efforts to be conservative, there is a further noteworthy limitation to our

analysis. While we only model sealing for adults, the sealing data that we acquired from the

Department of Public Safety includes the aggregate number of sealings for juveniles and adults.

As such, it is likely that we overestimate the pace of relief provided under the current,

application-based method, and overestimate the uptake rate and underestimate the current gap.

RULES

217

Orders of Non-Disclosure and Expunction

Primary Sources: Texas Gov. Code Chap. 411 (2017) | Sec. 411.072 | 411.0725 | 411.073 |

411.0731 | 411.0735 | 411.074 (all 2017) | Art. 55.01(2019)

Secondary Sources: University of Texas Guide (2019) | Texas CCRC (12/3/2021) | App | Texas

Law Help OND Prep Guide

A. CONVICTIONS

217

Reprinted with permission. *Thank you to Hithesh Sekhar Bathala, Obie Reynolds, and Ahmanda Lee for

contributing the concise statement as well as the shorter version of this article, published at paperprisons.org

216

Statutory Wrongdoing and its Consequences, Tʜᴇ Tᴇxᴀs Pᴏʟɪᴛɪᴄs Pʀᴏᴊᴇᴄᴛ, (January 8, 2022),

https://texaspolitics.utexas.edu/archive/html/just/features/0201_01/crimeandp.html.

215

For instance, an individual convicted of Section 30.02 (Burglary), Penal Code, is ineligible if “the offense or

conduct is punishable under Subsection (d) of that section and the actor committed the offense or engaged in the

conduct with intent to commit a felony listed in Paragraph (A) or (C).” We do not have access to this data, so we

assume that all individuals convicted of burglaries are ineligible.

214

Examples of such offenses in our dataset include 49.045(b), 49.09(b-4), and 49.07(c).

213

For instance, we assumed that 51.031 (child labor law) and 22.02(a)(2) (aggravated assault with a deadly

weapon) were ineligible.

212

Although we did our best with the data cleaning, it is always worth noting that the dirty nature of the data could

have impacted our findings and is thus a weakness. Also, when identifying whether convictions were of a violent or

sexual nature, we relied on the following resource: Texas Office of Court Administration, What is an Order of

Nondiclosure and How Do I Get One?, Texas Law Help, (Last updated June 28, 2021),

https://texaslawhelp.org/article/what-is-a-nondisclosure-order-and-how-do-i-get-one.

Electronic copy available at: https://ssrn.com/abstract=4065920

1. Any first time misdemeanor conviction if it is the person's only conviction or deferred

adjudication after a two-year waiting period (in which the person cannot get an add’l

convictions or deferred adjudication) following sentence completion (Sec. 411.0735).

218

a. In addition to universal disqualifiers, ineligible offenses under this section include

misdemeanor offenses that are:

i. Violent or sexual in nature (with the exception of §22.01, Penal Code)

ii. Convictions under §106.41 Alcoholic Beverage Code

iii. Driving While Intoxicated (Penal Code §49.04)

iv. Boating while Intoxicated (Penal Code §49.06)

v. Assembling or operating an amusement ride while intoxicated (Alcoholic

Beverage Code §49.065)

vi. Organized crime (Chapter 71, Penal Code)

2. Any DWI conviction under Section 49.04, Penal Code, if it is the person’s only offense

(conviction or deferred adjudication) after a two to five year waiting period (in which the

person cannot get additional convictions deferred adjudications) (Sec. 411.0736).

219

(In

our model, we default to five years in order to be conservative in our estimates.)

a. In addition to universal disqualifiers, ineligible offenses include 49.04(d), Penal

Code (driving while intoxicated with a BAC greater than 0.15).

B. DEFERRED ADJUDICATIONS

1. Any felony that is sentenced to deferred adjudication community supervision after a five

year waiting period starting from community supervision completion and offense

dismissal (in which the person cannot get additional convictions deferred adjudications)

(Sec 411.0725).

a. In addition to universal disqualifiers, ineligible offenses include:

i. Driving While Intoxicated (Penal Code §49.04)

2. Any misdemeanor that is sentenced to deferred adjudication community supervision after

up to a two year waiting period starting from community supervision completion and

offense dismissal (in which the person cannot get additional convictions deferred

adjudications) (Sec 411.0725) including offenses under 49.04 and 49.06 (411.0726).

C. UNIVERSAL DISQUALIFIERS

1. Ineligible per 411.074(b) if individual has any convicted or deferred adjudication offense:

a. Requiring registration as a sex offender (Section 62.001(5))

b. Involving aggravated kidnapping (Section 20.04)

c. Involving homicide (Sections 19.02 , 19.03)

219

Some offenses have a two year waiting period rather than a five year waiting period. However, we default to five

years in order to be conservative in our estimates.

218

Please note that all offenses have a two year waiting period except for those punishable by fine-only, which do

not have a waiting period. We do not model this nuance because we use the longest waiting period to be

cosnervative in our estimates.

Electronic copy available at: https://ssrn.com/abstract=4065920

d. Involving human trafficking (Sections 20A.02 , 20A.03)

e. Involving child/elder abuse (Sections 22.04, 22.041)

f. Involving family violence (Sections 25.07 , 25.072 , 71.004)

g. Involving stalking (Section 42.072)

h. Flying while Intoxicated (Penal Code §49.05)

D. UNMODELED CRITERIA

1. Did not model expunctions eligibility for non-convictions. (Art. 55.01(a))

2. Did not model shorter waiting period for certain deferred adjudication community

supervision under 411.072 (first time non-violent misdemeanor offenses) or 411.0726

(deferred adjudication for DWI)

3. Did not include sealing eligibility for individuals that successfully complete veteran’s

treatment court (Sec. 411.0727)

4. Did not include sealing eligibility by individuals that are victims of trafficking (Sec.

411.0728)

5. Did not model expunction eligibility for Class C Misdemeanors sentenced to community

supervision (Article 55.01(a)(2)).

6. Did not model whether LFOs were repaid

7. Did not model whether certain deferred adjudication offenses eligible for OND/sealing

had “affirmative finding under Article 42A.105(f), Code of Criminal Procedure...filed in

the papers of the case” (411.072(a)(1)(b), 411.0726(a)(2))

8. Did not model whether certain deferred adjudication offenses otherwise eligible for

OND/sealing had findings under “former Section 5(k), Article 42.12, Code of Criminal

Procedure...filed in the papers of the case” (411.072(a)(1)(b)).

9. Did not model certain offenses requiring registration as a sex offender (62.05(D),

62.05(E), 62.05(G), 62.05(H), 62.05(I)), that would be universally ineligible.

Electronic copy available at: https://ssrn.com/abstract=4065920

Appendix Table 1: Detailed View of The Texas Second Chance Gap

Population

People in the

Criminal Population

People Eligible for

Any Records Relief

People Eligible for

Sealing of All

Convictions

Estimated Number of

People with Convictions

4.8M

675,305

59663

% Male

85%

87%

85%

Top Felony Deferred

Adjudication

Community Supervision

poss cs pg 1 <1g

(11.4%), DWI 3rd or

more (4.2%),

burglary (3.7%)

poss cs pg 1 <1g,

driving while

intoxicated 3rd or

more iat

Top Convictions -

Misdemeanors

poss marij <2oz

(8.3%), DWI

(5.6%), assault

(3.7%)

driving while

intoxicated, poss

marij <2oz, driving

while intoxicated

2nd

driving while

intoxicated, poss

marij <2oz, assault

causes bodily injury

family member

Avg Years since last

Conviction

12.6

Share of People whose

Last Conviction was

10+ Years Ago

2,407,274(~49.87%)

446,535

49,537

Average Age at First

Conviction

28.26

27.86

Average Current Age of

People with Convictions

Median Years since last

Conviction

White and Latinx %

(share in pop = 82%)

0.69

0.73

0.69

Black % (share in pop =

13%)

0.31

0.26

0.3

Asian % (share in pop =

5%)

0.01

Electronic copy available at: https://ssrn.com/abstract=4065920

APPENDIX B: DATA SAMPLE DESCRIPTION

220

Data Statistics

Number of People in Dataset

4,826,860

Sample Size

153,674

OND Sample Data Statistics

Number of People with Misdemeanor Convictions in Sample

94,403 (61.45%)

Estimated Number of People with Misdemeanor Convictions in Population

2,966,662

Number of People with Felony Convictions in Sample

122,611 (79.81%)

Estimated Number of People with Felony Convictions in Population

3,853,113

Average Age of Misdemeanor Convictions

46 years old

APPENDIX C: COMMON CHARGES

221

A. Top 10 Charges in our Dataset

B. Top 10 Misdemeanor Convictions in Sample

C. Top 10 Felony Convictions in Sample

D. Top 10 Charges Eligible for OND in our Dataset

Table A. Top 10 Charges in Sample

Charges

Number of

Records

Percentage of

Charges

Number of People

with Charges

poss cs pg 1 <1g

335,265

12.97%

36,232

poss marij <2oz

238,491

9.22%

34,202

driving while intoxicated

147,550

5.71%

27,365

burglary of habitation

116,005

4.49%

13,836

driving while intoxicated 3rd or more iat

111,187

4.3%

12,627

assault causes bodily injury family

member

103,171

3.99%

17,562

poss cs pg 1 >=1g<4g

95,857

3.71%

12,585

221

A shorter version of this article is available at paperprisons.org

220

A shorter version of this article is available at paperprisons.org

Electronic copy available at: https://ssrn.com/abstract=4065920

assault causes bodily inj

75,109

2.9%

12,107

burglary of building

72,598

2.81%

8,004

unauth use of vehicle

67,916

2.63%

8,311

Table B. Top 10 Misdemeanor Convictions in Sample

Charges

Number of

Records

Percentage of

Charges

Number of People

with Charges

poss marij <2oz

215379

8.33%

34166

driving while intoxicated

145373

5.62%

27335

assault causes bodily injury family

member

95485

3.69%

17521

assault causes bodily inj

69788

2.7%

12079

driving while intoxicated 2nd

55722

2.15%

10659

fail to id fugitive intent give false info

53855

2.08%

8688

resist arrest search or transport

53052

2.05%

8812

driving w/lic inv w/prev conv/susp/w/o

fin res

39245

1.52%

6796

poss cs pg 3 < 28g

27195

1.05%

5233

evading arrest detention

26269

1.02%

5500

Table C. Top 10 Felony Convictions in Sample

Charges

Number of

Records

Percentage of

Charges

Number of People

with Charges

poss cs pg 1 <1g

295925

11.44%

35970

driving while intoxicated 3rd or more iat

108898

4.21%

12573

burglary of habitation

95745

3.7%

13789

poss cs pg 1 >=1g<4g

81261

3.14%

12557

burglary of building

64327

2.49%

7864

unauth use of vehicle

61414

2.37%

8236

man del cs pg 1 <1g

54832

2.12%

5957

forgery financial instrument

54563

2.11%

7046

man del cs pg 1 >=4g<200g

50568

1.96%

8240

Electronic copy available at: https://ssrn.com/abstract=4065920

poss cs pg 1 >=4g<200g

41772

1.62%

6936

Table D. Top 10 Misdemeanor Conviction Charges Eligible for OND in our Dataset

Charges Eligible for OND

Number of

Records

Percentage of

Charges

Number of People

with Charges

poss marij <2oz

10,315

0.4%

10,315

assault causes bodily injury family

member

222

4,648

0.18%

4,648

assault causes bodily inj

3,651

0.14%

3,651

driving while intoxicated 2nd

3,169

0.12%

3,169

fail to id fugitive intent give false info

2,372

0.09%

2,372

resist arrest search or transport

2,217

0.09%

2,217

poss cs pg 3 < 28g

1,041

0.04%

1,041

evading arrest detention

1,027

0.04%

1,027

theft prop >=$100<$750

579

0.02%

579

poss dangerous drug

444

0.02%

444

APPENDIX D: ROBUSTNESS TEST AND HISTOGRAM FOR SENTENCE

EXPIRATION DATE ASSUMPTIONS

222

In Texas, assault is not considered a violent felony as listed in 411.0735(c-1):

https://statutes.capitol.texas.gov/Docs/GV/htm/GV.411.htm

Electronic copy available at: https://ssrn.com/abstract=4065920

The two bars on the left show the average number of years in jail when records with missing

sentence expiration date are excluded. The blue bar shows the average number of years for

felony charges and the orange bar shows the average number of years of misdemeanor charges.

The two bars on the right show the average number of years in jail when records with missing

sentence expiration date are assumed to be sentence start date + 2.9 years for misdemeanor

convictions and + 3.2 years for felony convictions.

APPENDIX E: DETAILED RECORD RELIEF STATISTICS

223

We obtained OND statistics from the Texas Department of Public Safety, which reports

that 18,593 ONDs were processed between fiscal years 2014 and 2019. In addition, we obtained

the number of ODLs granted from the Texas Department of Public Safety, which reports that

44,812 ODLs were granted between September 2017 and February 2020.

224

A breakdown of

each type of relief granted by year is available in Table H below.

Table H. Breakdown of Relief Granted Each Year in Texas

Year

ONDs Processed in Fiscal Year

ODLs Granted in Calendar Year

2020

N/A

2,637

2019

2,650

16,350

2018

2,558

15,831

2017

2,313

12,637

2016

2,360

N/A

2015

4,258

N/A

2014

4,454

N/A

*ODLs granted for 2017 are from September 2017 - December 2017

+ODLs granted for 2020 are from January 2020 - February 2020

224

Driver License Division High Value Set, Dᴇᴘᴀʀᴛᴍᴇɴᴛ ᴏғ Pᴜʙʟɪᴄ Sᴀғᴇᴛʏ, (February 2020, January 2020,

December 2019, November 2019, October 2019, September 2019, August 2019, July 2019, June 2019, May 2019,

April 2019, March 2019, February 2019, January 2019, December 2018, November 2018, October 2018, September

2018, August 2018, July 2018, June 2018, May 2018, April 2018, March 2018, February 2018, January 2018,

December 2017, November 2018, October 2017, September 2017) (data last available January 23, 2022 at

https://www.dps.texas.gov/section/driver-license/driver-license-division-high-value-data-sets).

223

A shorter version of this information is available at paperprisons.org

Electronic copy available at: https://ssrn.com/abstract=4065920

APPENDIX F: CLEARANCE CRITERIA CHALLENGES AND LEGISLATIVE

DRAFTING ALTERNATIVES

225

Criteria

Administrative

Challenge

Example

Drafting

Alternative

Sentence

Completion

(OND)

Not tracked in

court data and

hard to infer as

clean sentencing

data is often not

available; it also

is often unclear

whether or not

outstanding fines

and fees must be

paid, and whether

they have been.

“Notwithstanding any other provision of

this subchapter or Subchapter F, a person

described by Subsection (a) who

completes the person's sentence,

including any term of confinement

imposed and payment of all fines, costs,

and restitution imposed, may petition the

court that imposed the sentence for an

order of nondisclosure of criminal

history record information under this

section…” (Tex. Gov. Code 411.0735(b))

Disposition

Date (+ X

Years)

First

conviction;

qualifying

conditions

(OND)

Lack of unique

identifier across

precludes

determination

Bless

commercial

identification

approximation

technique

Personal

demographic

trait such as

age, military

status, or other

condition

(OND)

Information may

not be easily

ascertainable /

available on the

record or charge