Voting rights restoration in
Florida: Amendment 4 -
Analyzing electoral impact &
its barriers
By Alexander Klueber and Jeremy Grabiner (MPPs 2020)
submitted in partial fulfillment of the requirements for the degree of Master in Public Policy
Harvard Kennedy School of Government
Master in Public Policy
April 2020
Policy Analysis Exercise
Client: Marc Mauer | Executive Director of The Sentencing Project
Advisor: David C. King | Senior Lecturer in Public Policy. HKS
Advisor: Miles Rapoport | Senior Practice Fellow in American
Democracy. HKS
Seminar Leader: Thomas Patterson | Bradlee Professor of Government
and the Press. HKS
This PAE reflects the views of the authors and should not be viewed as representing
the views of the PAE's external client, nor those of Harvard University or any of their
faculty.
CONTENTS
I. Acknowledgments ................................................................................................................
II. Executive Summary ........................................................................................................... 1
III. Introduction ....................................................................................................................... 3
IV. Background ........................................................................................................................ 4
V. Methodology ...................................................................................................................... 6
A. Quantitative analysis on electoral impact of Amendment 4 ....................................... 6
a. Data sources ............................................................................................................... 6
b. Data manipulation ..................................................................................................... 8
c. Key assumptions ........................................................................................................ 9
B. Qualitative analysis on voting barriers and modulation strategies ............................ 9
VI. Findings ........................................................................................................................... 10
A. Electoral impact assessment ..................................................................................... 10
a. Quality of correctional data ..................................................................................... 10
b. Size of population eligible to vote after Amendment 4 and SB 7066 ....................... 11
c. Voter registration patterns ...................................................................................... 13
d. Estimation of electoral impact ................................................................................. 16
e. Spill-over effects ...................................................................................................... 23
B. Barriers to reaching the ballot box for the enfranchised .......................................... 24
a. Informational Barriers ............................................................................................. 26
b. Financial barriers ..................................................................................................... 28
c. Mobilizing barriers .................................................................................................. 28
VII. Recommendations ...........................................................................................................30
A. Informational Recommendations .............................................................................30
B. Financial Recommendations .................................................................................... 32
C. Mobilizing Recommendations .................................................................................. 34
D. Summary of recommendations ................................................................................. 35
VIII. Appendix .......................................................................................................................... 36
A. Appendix A – Data sources ....................................................................................... 36
B. Appendix B – Assumptions ....................................................................................... 39
C. Appendix C – Detailed predictions ........................................................................... 57
D. Appendix D – Code for reproduction ........................................................................ 65
ACKNOWLEDGMENTS
We thank Marc Mauer and The Sentencing Project for providing us with an opportunity to
explore this research question. Their guidance and expertise on this were extraordinary and
we are deeply grateful.
In addition, we thank our seminar leader, Thomas Patterson and our advisors, Miles Rapoport
and David King. We are grateful for their willingness to help us on every step of the way.
We want to thank Desmond Meade, Neil Volz, and the FRRC team for hosting us and providing
us with important information from the field. As well as Rep. Sean Shaw, Professor Kathryn
DePalo-Gould, Devon Crawford, and Bruce Riley for their input on the project.
Furthermore, we want to thank the Ash Center and the Program in Criminal Justice Policy and
Management for the funding they provided to conduct this research.
1
EXECUTIVE SUMMARY
The passage of Amendment 4 in Florida marked a monumental event for more than 1.4 million
Floridians who could regain their voting rights. However, our findings suggest that only 3.3%
or ~49.000 of them will turn out to cast their ballot in 2020 if all associated legislation stays
in place. We would expect it to rise to 146,000 or 13% in case SB 7066 is repealed.
This paper estimates the impact Amendment 4 and its associated legislation will have on
Florida’s 2020 general election. It uses these estimates to quantify and explore the barriers
that diminish turnout and puts forward recommendations to modulate them.
Our findings suggest no partisan impact by the Amendment and associated legislation on the
electoral map in 2020. Although most counties in our model become increasingly Democratic,
we predict no change in partisan control in any county or at the state level.
Our predictive model allows us to quantify the relevance of the barriers during the journey
from post-sentencing disenfranchisement to the ballot box on election day. Most notable is
the requirement to repay all Legal Financial Obligations (LFOs), as set forth by SB 7066. It
single-handedly disenfranchises more than 1 million citizens who could participate in our
2
democracy. Second to this is the barrier that enfranchised individuals will not exercise their
right to vote. We predict this to be the case for ~300,000 returning citizens, demonstrating
that uncertainty around the right to vote is depressing turnout.
For greater exploration, we reframed the barriers into three broader categories: Informational
Barriers, Financial Barriers, and Mobilizing Barriers. Informational Barriers center around
the lack of a centralized process to determine what an individual owes, difficulty in discovering
one’s eligibility, and lack of awareness of legal options afforded. Financial Barriers stem from
an individual’s inability to pay their LFOs. Mobilizing Barriers stem from apathy towards the
democratic process, lack of engagement with the population, and standard turnout issues.
For each of these categories of barriers, we then outlined recommendations that we hope will
help inform The Sentencing Project’s goals of returning rights to the disenfranchised in this
country.
Recommendations
• Centralized Informational Process
• Change Burden of Proof
• Change Registration Language
• Information Packet at Last Contact
Recommendations
• Repeal SB 7066
• Change Sentencing
• Increase Fundraising
• Partner with Counties
• Exempt 10+ years from LFO requirement
Recommendations
• Streamlined Engagement Strategy
• RC to RC outreach
• Spillover Community outreach
Post-
sentencing
disenfran-
chised voters
Qualified
offenses acc.
to literal
Amendment
Qualified
offenses acc.
to SB 7066
Qualified by
having
fulfilled all
LFOs
Individuals in
Florida
1,487,847
1,454,156
1,435,033
362,614
Total share
100%
98%
96%
24%
3
INTRODUCTION
In 2016 an estimated 6.1 million people were disenfranchised because of felony convictions.
1
This is almost 5 million more than in 1976.
2
The consequences of these policies go beyond the
prohibition to vote - they institutionalize racism, disempower communities and hinder
reintegration.
Through the close ties between felony disenfranchisement and the criminal justice system, the
racist nature of the latter manifests itself in the former. This is to say that a disproportionate
number of Black and Hispanic Americans being incarcerated results in a disproportionate
number being disenfranchised. In line with that, Black Americans over 18 are four times as
likely as the rest of the US population to lose their voting rights as 1 in 13 or ~ 2.2 million black
adults are disenfranchised.
3
Further, disenfranchisement is a painful mechanism of exclusion at the community and the
personal level and often contributes to political apathy for both the individual and the people
around them. Reversing disenfranchisement laws and engaging returning citizens is therefore
crucial to the inclusiveness of American democracy and individual reintegration.
On the back of these insights, states have started to move to less punitive disenfranchisement
legislation. The 2018 passage of Amendment 4 in Florida is the single largest attempt to
reverse these laws. We therefore seek to answer:
1. What is the projected electoral impact of Amendment 4 and associated legislation in
Florida’s 2020 general elections?
2. What barriers are diminishing its electoral impact and how can they be modulated?
From these a series of sub-questions arise. These include: how reliable is the data in our
sample from the Florida Department of Corrections? What is the size of the population eligible
to vote and how many of them are going to turn out on election day? Who is it that will turn
out to vote and how will that impact the governance of the state? What are the barriers
diminishing turnout and how significant are they? How do these barriers manifest themselves
in the journey from post-sentencing disenfranchisement to the ballot box on election day?
What actions can state legislatures and organizations like The Sentencing Project take to
modulate these barriers? What are the spillover effects in the rest of the community?
While we use Amendment 4 in Florida and the partisan nature of the 2020 general election as
the prism of our analysis, our concern is the persistence of disenfranchisement and the
wellbeing of the US democracy. By using a case of particular relevance at this moment in time,
we hope to elucidate systemic considerations of enfranchisement efforts while encouraging
more concern for the newly enfranchised.
1
Uggen, Christopher. 6 Million Lost Voters: State-Level Estimates of Felony Disenfranchisement,
2016. (The Sentencing Project, 2016), 3. https://www.sentencingproject.org/publications/6-million-
lost-voters-state-level-estimates-felony-disenfranchisement-2016/.
2
Ibid.
3
Uggen, 6 million, 3.
4
BACKGROUND
By 2016, Florida accounted for 27% of the national disenfranchised population, with Black
Americans making up 21% of this (only 16% of general population).
4
Florida’s estimated 1.4
million post-sentencing disenfranchised voters were greater than the population of 11 states
and the District of Columbia, highlighting the consequential impact of Florida’s
disenfranchisement laws on our democracy.
5
At the end of the Civil War, the Florida state legislature began enacting laws that would prevent
freed Black men from participating in the democratic process while maintaining white
supremacy as the order of society. After initially refusing to adopt the 14
th
Amendment, Florida
was forced to draft a new constitution in 1868.
6
However, the new constitution maintained
means of excluding or minimizing the power of Black American citizens.
Under Article XIV Section 2, the constitution instituted its provision of felony
disenfranchisement by stating, “No person under guardianship noa compos mentis, or insane,
shall be qualified to vote at any election, nor shall any person convicted of felony be qualified
to vote at any election unless restored to civil rights.”
7
While the Florida constitution
underwent a variety of changes over the last century, the clause disenfranchising those with
felony convictions remained on the books until 2018 when Amendment 4 was passed through
a referendum.
After an extensive campaign by the Florida Rights Restoration Coalition (FRRC) and partners,
voters in Florida passed Amendment 4 with 64% of the vote.
8
Amendment 4 stated that “any
disqualification from voting arising from a felony conviction shall terminate and voting rights
shall be restored upon completion of all terms of sentence including parole or probation”.
9
The
amendment however kept in place restrictions on those “convicted of murder and felony sex
offenses”.
10
The passage of the amendment restored voting rights to more than 1.4 million
Florida residents.
Yet the history of right marginalization repeated itself after the election, as the legislature and
newly elected Governor DeSantis introduced and passed SB 7066 which instituted an LFO
therefore significantly limiting the number of people included in the amendment.
The ACLU brought a lawsuit against the state in opposition to this modern poll tax. In a
statement Julie Ebenstein of the ACLU said, “Over a million Floridians were supposed to
reclaim their place in the democratic process, but some politicians clearly feel threatened by
greater voter participation. They cannot legally affix a price tag to someone's right to vote.”
11
Most recently the U.S. Court of Appeals for 11th Circuit in Atlanta ruled in favor of the 17
4
Ibid.
5
Data Access and Dissemination Systems. “American FactFinder - Results.” American FactFinder -
Results. October 5, 2010.
https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?src=bkmk#.
6
Wood, Erika. “Florida: An Outlier in Denying Voting Rights.” Brennan Center for Justice, 2016.
https://www.brennancenter.org/sites/default/files/publications/Florida_Voting_Rights_Outlier.pdf
7
Florida constitution Art XIV Section 2 retrieved from
https://www.floridamemory.com/items/show/189095?id=23
8
Alejandro de la. Garza. “Florida Passes Amendment 4, Restoring Voting Rights for Felons.” Time.
Time. November 7, 2018. https://time.com/5447051/florida-amendment-4-felon-voting/.
9
Florida Constitution. Amendment 4. Article VI Section 4. Retrieved from
https://www.aclufl.org/en/voter-restoration-amendment-text
10
Ibid.
11
“Groups Sue to Block New Florida Law That Undermines Voting Rights Restoration.” 2019. ACLU of
Florida. July 17, 2019. https://www.aclufl.org/en/press-releases/groups-sue-block-new-florida-law-
undermines-voting-rights-restoration.
5
plaintiffs to be allowed to vote.
12
However, it is possible that this case may end up in the U.S
Supreme Court.
Courts can modify an individual’s sentences, eliminate or reduce their fines and convert fines
into community service. The implementation of this provision however has not been done
throughout the state. Only four counties, Miami-Dade, Broward, Palm Beach and
Hillsborough have implemented processes to enact it.
13
Therefore, as the government and electoral officials navigate this implementation process,
thousands of residents will be left unable to decide who will represent them in the upcoming
Presidential election.
12
Periera, Ivan. “Federal Appeals Court Rules against Florida's Restriction on Former Felons from
Voting over Fines.” ABC News. ABC News Network, February 19, 2020.
https://abcnews.go.com/Politics/federal-appeals-court-rules-floridas-restriction-felons-
voting/story?id=69073124.
13
Rivero, Daniel. “People Across Florida Are Getting Their Voting Rights Back. Few Republicans Could
Benefit.” WLRN, January 5, 2020. https://www.wlrn.org/post/people-across-florida-are-getting-
their-voting-rights-back-few-republicans-could-benefit#stream/0.
6
METHODOLOGY
QUANTITATIVE ANALYSIS METHODOLOGY
DATA SOURCES
To build a predictive model of the electoral impact Amendment 4 will have on the 2020 general
election in Florida, we analyzed more than 12.5 million individuals in Florida, combining
various sources. These included:
• Official Florida voter registration and unofficial voting history information by the
Florida Division of Elections (last updated: 03/20)
14
• Public records of OBIS offender data base in Florida by the Florida Department of
Corrections (last updated: 01/20)
• Historic Florida election results on county and precinct level by the MIT election lab
(last updated: 12/19)
• 2016 Actuarial Life Table by the Social Security Administration
15
• Statute table by the Florida Department of Law Enforcement (last updated: 12/19)
16
• Statistics on imposed sanction by offense type in the 2019 Florida’s Criminal
Punishment Code: A Comparative Assessment by the Florida Department of
Corrections (last updated: 10/19)
17
Please refer to Appendix A for a detailed description of the sources, the data and the variables
included.
We encountered significant data limitations due to legal and financial restrictions. These
limitations may be split into two groups:
a) Not included in the sample but relating to individuals who are returning citizens and
who may be eligible to vote in Florida:
o Individuals convicted of a felony, but released from state probation /
community control without serving a custodial sentence
o Individuals who served their sentence outside the Florida state prison system
(e.g. federal prisons, county jails) and reside in Florida
o Future residence of inmates who are currently in prison, but will have served
their sentence in time to register for the general elections
o Outstanding LFOs (court-ordered fees, fines and restitutions) of individuals in
sample
o No modelling of potential recidivism among individuals in the sample
b) Limitations on returning citizens in other states (shortlisted based on demographic,
cultural and voting restrictions) to estimate expected turnout and community spill-
over effects: offender data and/or voter registration data inaccessible in Alabama,
Georgia, Louisiana, South Carolina, Tennessee and Virginia
14
“Voter Extract Disk File Layout.” Florida Department of State, October 18, 2018.
https://dos.myflorida.com/media/696057/voter-extract-file-layout.pdf.
15
“Social Security.” Actuarial Life Table. Accessed March 22, 2020.
https://www.ssa.gov/oact/STATS/table4c6.html.
16
FDLE's Statute Table. Florida Department of Law Enforcement. Accessed December, 2019.
https://web.fdle.state.fl.us/statutes/about.jsf.
17
“Florida’s Criminal Punishment Code: A Comparative Assessment .” Florida Department of
Corrections, October 2019. http://www.dc.state.fl.us/pub/scoresheet/Criminal Punishment Code
2019.pdf.
7
To address the data restrictions under b) we decided to rely on assumptions from expert
interviews and literature. To address the data restrictions under a), we took the following
(imperfect) measures:
• Treat available data points as a sample of the overall post-sentencing disenfranchised
population (estimated to be at 1,487,847). When reduced to post-sentencing
disenfranchised voters (excl. individuals who will still be serving non-financial legal
obligations of their sentence at general election voter registration date, do not reside
in Florida, are fugitives, deported, likely dead, etc.), this sample includes 396,104
individuals.
18
To reflect the differences of offense types between the population
convicted of a felony who had to serve a custodial sentence and the population that
didn’t have to serve a custodial sentence, we adjust the scaling from our sample to the
population based on the data published by the Florida Department of Corrections on
the sanctions imposed by offense type.
19
o There is additional concern around the representativeness of the sample for the
overall population (an example is pointed out by the Crime and Justice
Institute (2019)): “One of the main principles of the CPC is neutrality with
respect to race, gender, and social and economic status. Despite this stated goal
of fairness, defendants with similar criminal conduct and criminal histories
experience vastly different outcomes.”
20
Therefore, we must expect the
population that is convicted of a felony, but released without serving a custodial
sentence to also differ from the one that is not released in terms of judicial
circuit, county, economic status, county of residence, race, etc.).
o Only the 2019 publication of Florida’s Criminal Punishment Code included the
sanctions imposed by offense type on a level granular enough to match it to the
offenses excluded under Amendment 4 and SB 7066. The split for that
particular year may deviate from the historic average.
• Approximate county of future residence for currently active prisoners as the county in
which most felonies have been committed (if an equal number of felonies have been
committed in multiple counties, we go by alphabetical order of counties)
o The proxy may be flawed, because individuals move, have committed felonies
in counties in which they don’t live, etc.
o Note: this limits the granularity of our analysis to the level of counties (no
longer possible to have analysis at level of individual elections)
• Estimate outstanding LFOs at county level based on racial and county metrics outlined
for 58 counties in Expert testimony by Daniel A. Smith in September 2019.
21
o This forces us to make a concerning oversimplification of individual
circumstances based on county affiliation and race
o Average unlikely representative for 9 outstanding counties, because of size and
enfranchisement efforts (Broward, Miami-Dade, Palm Beach, Hillsborough
with systematic effort to reinstate enfranchisement by waiving/transferring
fees and fines in exchange for community service)
18
Uggen, Christopher. 6 Million Lost Voters: State-Level Estimates of Felony Disenfranchisement,
2016. (The Sentencing Project, 2016), 3. https://www.sentencingproject.org/publications/6-million-
lost-voters-state-level-estimates-felony-disenfranchisement-2016/.
19
“Florida’s Criminal Punishment Code: A Comparative Assessment.” Florida Department of
Corrections, October 2019. http://www.dc.state.fl.us/pub/scoresheet/Criminal Punishment Code
2019.pdf.
20
Marguiles, Lisa, Sam Packard, and Len Engel. “An Analysis of Florida’s Criminal Punishment Code.”
Crime and Justice Institute, June 2019. https://www.crj.org/assets/2019/06/An-Analysis-of-Florida-
CPC-June-2019.pdf.
21
Dan A. Smith, on behalf of plaintiffs Consolidated Case No. 4:19-cv-300. August 2, 2019.
https://www.aclu.org/sites/default/files/field_document/gruver_v_barton_-
_expert_report_of_daniel_a._smith_ph.d.pdf
8
o Note: this limits the granularity of our analysis to the level of counties (no
longer possible to have analysis at level of individual elections)
In addition, there is a real concern around the validity of the addresses available for individuals
from the department of correction, as shall be demonstrated under “Quality of correctional
data”.
DATA MANIPULATION
Using the entire offender database (active, released, supervised), we selected the population
that would gain enfranchisement via Amendment 4 as postulated by SB 7066 by excluding:
a) Individuals that will not have served their non-financial sentence by Oct. 5 2020
(Florida general election registration date)
b) Individuals that are expected to be deceased
c) Individuals that have committed disqualifying offenses (murder or sexual offenses) at
any point in time (as defined literally by the Amendment and as specified in SB 7066)
d) Individuals that no longer reside in Florida
e) Individuals who are expected to have outstanding LFO’s by Oct. 5 2020
We then combined this data with the Florida voter registration data to determine which
enfranchised individuals have registered to vote. We did so by harmonizing naming
conventions and adapting the notation (lowercase, removing all punctuation) of a person’s
first name, last name, name suffix, date of birth, race code, and sex code in both databases and
concatenating them to match them across the data sets.
Using these inputs, we leveraged a series of assumptions (detailed in the next chapter) to
model at the county and state level the number of enfranchised voters, the number of expected
votes and the partisan allocation of these votes and then compared that to the margin of victory
in each county in the 2016 presidential elections.
Lastly, we translated this information into four tables and four maps, estimating the electoral
impact of Amendment 4 for the various counties in Florida and statewide elections:
• Potential electoral significance of Amendment 4 in 2020
• Scenario 1 – medium turnout: expected electoral significance of Amendment 4 with
and without SB 7066
• Scenario 2 – low turnout: expected electoral significance of Amendment 4 with and
without SB 7066
• Scenario 3 – high turnout: expected electoral significance of Amendment 4 with and
without SB 7066
9
KEY ASSUMPTIONS
ID
Variable
Assumptions
Source
a)
Population down-
scaling
According to sample with
corrections for felons with /
without custodial sentence
Table.1
b)
Mapping offense type
to sanction imposed
for felons
See Appendix B
Florida Department of
Corrections, Florida’s Criminal
Punishment Code: A Comparative
Assessment, October 2019.
c)
Disqualifying offenses
acc to literal
interpretation of
Amendment
See Appendix B
Dara Kam, Meaning of 'murder'
key in Florida felons' voting
rights, January 2019.
d)
Disqualifying offenses
acc to SB 7066
See Appendix B
Florida Senate Bill No. 7066.
e)
Legal Financial
Obligations
See Appendix B
Dan A. Smith, on behalf of
plaintiffs Consolidated Case
No.4:19-cv-300. September 17,
2019.
f)
Life expectancies
See Appendix B
Social Security Administration,
Actuarial Life Table 2018.
g)
Voter turnout
High: 35%
Medium (Expected): Black:
16%, Others: 12%
Low: 5%
See Appendix B
h)
Spill-over effect
1.72
Expert interviews
i)
Party predilection
pattern 1
Observations among 150,000
ex-felons for which governor
Crist restored voting rights in
2007 convicted of less serious
offenses
Marc Meredith and Michael
Morse, Why letting ex-felons vote
probably won’t swing Florida,
November 2018.
j)
Party predilection
pattern 2
Party affiliation of matched
registered enfranchised voters
Table.4
k)
Total population
estimate
1,487,847
Uggen, Christopher. 6 Million
Lost Voters: State-Level
Estimates
of Felony Disenfranchisement,
2016.
Please refer to Appendix B for a detailed description of the assumptions.
QUALITATIVE ANALYSIS METHODOLOGY
A portion of our qualitative work relied on interviews. These interviews were influential in
providing a landscape of the issues, comparative analysis to other states, and a means of
verifying certain legal aspects regarding felony disenfranchisement. Academics, advocates,
and elected officials were the three categories of individuals that we interviewed. Interviews
were conducted by phone as well as face-to-face and were recorded with the permission of the
interviewees. Interviews were conducted both in a semi-structured and unstructured format.
10
FINDINGS
ELECTORAL IMPACT ASSESSMENT
The analysis found interesting findings in five areas:
a. Quality of correctional data (used as a source by majority of researchers on this topic)
b. Size of population eligible to vote
c. Voter registration patterns
d. Estimation of electoral impact
e. Spill- over effects
QUALITY OF CORRECTIONAL DATA
A comparison of the addresses from registered enfranchised individuals in the Florida
Department of Correction OBIS Offender Database and the voter registration data, indicated
that their overlap is low. Of the 23,843 individuals that we could match across the data sets,
only 76% had a matching county, 38% a matching Zip Code and 12% a matching address. Of
these 23,843 individuals, 13,092 registered to vote after Amendment 4 was enacted. This is
55% of the matched individuals thus raising concerns on the reliability of the correctional data
that the majority of researchers in this space rely on.
11
SIZE OF POPULATION ELIGIBLE TO VOTE AFTER AMENDMENT 4 AND SB
7066
In the media and most academic research, the population to be enfranchised by Amendment
4 is estimated to stand at around 1.5 million. This number originates from an estimate by the
Sentencing Project on the number of post-sentencing disenfranchised individuals in Florida.
22
As detailed before, Amendment 4 outlines 3 limitations to the right to vote:
• Completion of all terms of the sentence including parole or probation
• Doesn’t apply to those convicted of murder
• Doesn’t apply to those convicted of sexual offenses
The almost 1.5 million does not yet consider the latter two. They may be interpreted literally
according to the text in Amendment 4 or more stringently as outlined in SB 7066. In the latter,
case murder includes only first-degree murder and sexual offenses include only rape and sex
offenses against children. In the broader interpretation according to SB 7066 murder also
includes second degree murder and homicide, and sexual offenses include anything that leads
to a listing on the sex offender list. (Please see Appendix B Disqualifying offenses acc to SB
7066 for details).
As Table.1 below indicates, a literal interpretation will exclude around 2% or ~32,000
individuals from political participation. This number rises by another ~20,000 individuals or
1.5% when broadening the extent of the interpretation along the lines of SB 7066. These
estimates result from classifying individuals based on the description in their adjudication
charge (see appendix A c) and d) for details).
However, the main point of contention is how to interpret “all terms of the sentence”. The
ballpark around how many individuals may regain their right to vote changes significantly
when LFOs are included. The share remaining of the sample drops from around 96% to just
24%. If applied as an estimate for the entire population, it disenfranchises more than 1 million
individuals. This estimate follows when we allocate the individuals in our sample to counties
based on their addresses (in voter registration files, in OBIS database or if neither was
available for current inmates approximated as described under “Data Sources”) and then apply
the county level assumptions for outstanding LFO’s and ability to repay (see Appendix B e) for
details).
An expected turnout rate of around 14% among this population (see medium scenario) would
then result in 48,680 additional votes, or about 3.3% of the people who originally had the
prospect of regaining their right to vote.
As we are working with a sample that covers around 27% of the entire estimated population,
there is significant uncertainty around these estimates.
22
Uggen, Christopher. 6 Million Lost Voters: State-Level Estimates of Felony Disenfranchisement,
2016. (The Sentencing Project, 2016), 3. https://www.sentencingproject.org/publications/6-million-
lost-voters-state-level-estimates-felony-disenfranchisement-2016/.
12
Table 1: Barriers to electoral impact for enfranchised
Post-
sentencing
disenfran-
chised voters
Qualified
offenses acc.
to literal
Amendment
Qualified
offenses acc.
to SB 7066
Qualified by
having
fulfilled all
LFOs
Individuals in
Florida
1,487,847
1,454,156
1,435,033
362,614
Total share
100%
98%
96%
24%
Sample
individuals
396,104
363,210
350,136
59,602
Share of
sample
100%
92%
88%
15%
Non-sample
individuals
1,091,743
1,090,946
1,084,897
303,012
Share of non-
sample
100%
100%
99%
28%
13
VOTER REGISTRATION PATTERNS
Uncertainty around the right to vote is a significant depressor of voter turnout among the
enfranchised. The punishments for voting illegally is too severe to risk when an individual is
uncertain of their voter status.
The legal obscurity around the interpretation of Amendment 4 is a prime source of
uncertainty. This also expresses itself when we look at the voter registration patterns among
the 13,092 individuals that we matched as having registered since the Amendment was
enacted on a timeline together with the major legal decisions around the bill:
1. 2018-11-06: Amendment 4 passes
2. 2019-01-08: Amendment 4 takes effect
3. 2019-05-03: SB 7066 passes
4. 2019-07-01: SB7066 is signed into law
5. 2019-10-18: Federal District Court declares inclusion of FLOs in SB 7066 unconstitutional
and allows the 17 individuals plaintiffs with FLOs in the case to register
6. 2020-01-16: Florida Supreme Court declares SB 7066 and inclusion of FLOs constitutional
7. 2020-02-19: Eleventh Circuit Court declares inclusion of FLOs in SB 7066 unconstitutional
Graph 1: Timeline of matched enfranchised voter registrations
14
The graph shows a first smaller rise in registrations upon the passing of the Amendment (1).
This can be interpreted as an increased interest resulting from the public attention around the
Amendment by individuals convicted of less serious offenses that had regained their right to
vote in 2007 with Governor Crist’s executive order. There was an immediate spike in
registrations after the Amendment took effect in January 2019 (2) that decreased over the
ensuing months. However, as the public discussion around the Amendment escalated prior to
SB 7066 being enacted, the number of registrations increased again until then (4). The circuit
court’s decision to challenge the inclusion of LFOs in “serving a sentence” and extend
protection to individuals potentially enfranchised under Amendment 4 was followed by a
further increase in registration numbers (5). An overall upwards trend remained (with strong
fluctuations) until the Florida Supreme Court declared that LFO’s would be included in their
interpretation of “serving a sentence” in January 2020 (6), which resulted in a drop of
registrations. As the primary registration season began ramping up, so did the number of
registrations, which shows the correlation between voter registrations and legislation
uncertainty.
A comparison of characteristics among the individuals in our sample who registered and the
individuals in our sample who have not registered indicates a high proportion of African
American registrants. While only 44% of the sample population is African American, 60% of
the matched registrants are. The opposite is true for the White population (see Table.2), where
52% in the sample compare with 39% among the registered matched individuals. In terms of
gender the pattern of registrants roughly matches that of the overall sample (See Table.3). It
also stands out that there is a predilection for the Democratic party (55%) among the
enfranchised registered to vote of more than 2 to 1 compared to the Republican party (21%).
The block of no or other affiliations (24%) is significant (See Table.4).
Table 2: Racial split of registered and total enfranchised population
Racial split b/w matched registered enfranchised voters and all enfranchised voters in
sample (March)
All enfranchised voters in
sample
Registered enfranchised voters
in sample
Race
Absolute
number
Share
Absolute
number
Share
White
170,764
52%
5,276
39%
Black
146,870
44%
8,145
60%
Hispanic
12,042
4%
216
2%
All others/unknown
872
0%
3
0%
AAPI
265
0%
6
0%
Asian or pacific islander
51
0%
2
0%
Total
330,864
100%
13,648
100%
Source: Florida Division of Elections, Florida Department of Corrections, own analysis.
15
Table 3: Gender split of registered and total enfranchised population
Gender split between registered enfranchised voters and all enfranchised voters in sample
(March)
All enfranchised voters in
sample
Registered enfranchised voters in sample
Sex
Absolute
number
Share
Absolute
number
Share
Male
279,632
85%
11,330
83%
Female
51,232
15%
2,318
17%
Total
330,864
100%
13,648
100%
Source: Florida Division of Elections, Florida Department of Corrections, own analysis.
Table 4: Party affiliation of registered enfranchised voters
Party affiliation for registered enfranchised voters in sample (March)
Party
Absolute
Share
DEM
7,522
55%
NPA
2,952
22%
REP
2,916
21%
IND
225
2%
LPF
20
0%
CPF
7
0%
REF
5
0%
PSL
1
0%
Total
13,648
100%
Source: Florida Division of Elections, Florida Department of Corrections, own analysis.
16
ESTIMATION OF ELECTORAL IMPACT
Potential impact
Amendment 4 has the potential to change the future of Florida’s electoral landscape.
Regardless of the restrictions outlined in SB 7066, we estimate that more than 360,000 new
votes could be cast. The newly enfranchised will be able to swing votes both on the state level
and in 5 of the 67 counties (all of which are controlled by the Republican party). This uses the
2016 Presidential elections as a baseline and works under the assumption that all other
variables (voter turnout, voting location, which party to vote for, etc.) would remain the same.
What could further strengthen the impact of the Amendment are the implications for the
communities that the enfranchised individuals are part of. Many of these tend to be
communities of low voting propensity. A Randomized Control Trial in the Orlando Mayoral
election suggests that the spill-over effects combined with an outreach effort could be as high
as 1.72 times the original vote.
23
When we consider this very optimistic spill-over effect, we
would be looking at almost 1,000,000 additional votes in Florida, enough to change the
outcome of the last 7 elections for president in Florida and flip 11 counties (2 of which
controlled by the Democratic party and 9 of which controlled by the Republican party) in
relation to their 2016 outcomes (See Table.5 for details).
Table 5: The electoral significance of Amendment 4 in 2020 – Potential
The electoral significance of Amendment 4 in 2020
2016 results
2020 estimate
County
Incumbent
Victory
margin
Enfranchi-
sed voters
Enfranchised
& spillover
Potential to
swing
(enfranchised)
Potential to
swing
(enfranchised
& spillover)
Duval
Republican
5,968
16,348
44,467
Yes
Yes
Gadsden
Democratic
8,292
3,145
8,554
No
Yes
Hendry
Republican
1,580
803
2,184
No
Yes
Hillsborough
Democratic
41,026
28,753
78,208
No
Yes
Jefferson
Republican
389
785
2,135
Yes
Yes
Monroe
Republican
2,933
1,703
4,632
No
Yes
Pinellas
Republican
5,500
21,641
58,864
Yes
Yes
Polk
Republican
39,997
17,315
47,097
No
Yes
Seminole
Republican
3,529
8,329
22,655
Yes
Yes
St. Lucie
Republican
3,408
11,401
31,011
Yes
Yes
Wakulla
Republican
6,164
3,237
8,805
No
Yes
Statewide
Republican
112,911
362,611
986,305
Yes
Yes
Source: See assumptions and own analysis.
23
Desmond Meade and Neil Volz, January 2020.
17
We subsequently explore three scenarios of voter turnout among returning citizens, while we
keep all other assumptions stable. These assumptions are based on various papers and expert
interviews that suggest a turn out range between 5% and 35%. However, these papers fail to
consider the impact the presence of FRRC will have on voter turnout. Their statewide
organizing efforts to support the newly enfranchised population’s rights to vote is unique and
therefore not reflected in the academic papers drawing inferences from felon turnout in past
elections.
Graph 2: The electoral significance of Amendment 4 in 2020 – Potential
18
Table 6: Overview of assumptions on turnout scenarios
Scenarios
Turnout
Sources
High
35%
1. RCT by FRRC
2. Christopher Uggen and Jeff Manza in Democratic Contraction?
Political Consequences of Felon Disenfranchisement in the US
Medium
Black: 16%
White: 12%
1. Meredith and Morse in Why letting ex-felons vote probably
won’t swing Florida
2. Traci Burch in Turnout and Party Registration among Criminal
Offenders in the 2008 General Election
Low
5%
Michael V. Haselswerdt in Con Job: An Estimate of Ex-Felon
Voter Turnout Using Document-Based Data
For each of these scenarios we are comparing two potential patterns of party affiliation. The
first is based on the voting pattern among the 150,000 returning citizens convicted of less
serious offenses that regained their right to vote in 2007 by executive clemency of Governor
Charlie Crist. 87% of black voters registered as Democrats while 40% of non-Black voters
registered as Republicans, 34% as Democrats and 26% with neither of the two parties. Just
16% of Black and 12% of nonblack returning citizens voted.
24
The second is based on the
pattern among the matched registrants (See Table.4 in f).
24
Meredith, Marc, and Michael Morse. “Why Letting Ex-Felons Vote Probably Won't Swing Florida.”
Vox. Vox, November 2, 2018. https://www.vox.com/the-big-idea/2018/11/2/18049510/felon-voting-
rights-amendment-4-florida.
19
Scenario: 1 – Medium Turnout: Expectation of Amendment 4 to swing the vote
In the medium scenario (the most likely), with a turnout between 12% and 16%, our
predictions show that no counties would change control. In this scenario just 3.3% or ~49,000
of the almost 1.5 million post-sentencing disenfranchised voters actually turn out to cast their
votes. The governance of the state would remain with the incumbent party. Looking at the
predictions on a county level according to party affiliation pattern 1 (modelling affiliation after
150,000 returning citizens enfranchised in 2007), the marginal vote would turn more
Republican in 2 counties, would remain stable in 5 counties and would become more
Democratic in the remaining 60 counties as well as at state level. The root of this can be traced
back to the higher registration rate among matched African American registrants. Graph 3
illustrates these predictions by visualizing the additional votes through the Amendment
(enfranchised & spillover as a % of the decisive margin in 2016).
It is noteworthy that state-wide elections are more prone to flip than most county level races.
Graph 3: The electoral significance of Amendment 4 in 2020 - Medium
20
The electoral consequences of SB 7066 are highlighted by Table 7. The table depicts our
predictions without SB 7066 in place. We would expect ~195,000 returning citizens to vote
instead of 49,000, an increase of 146,000 votes. This would constitute ~13% of the post-
sentencing disenfranchised population in the state instead of 3.3%. In terms of partisan
politics this would translate into 5 counties flipping from Republican to Democratic control.
In addition, the statewide results would change from Republican to Democratic (Table 7).
Table 7: The electoral significance of SB 7066 in 2020 – Medium scenario
The electoral significance of SB 7066 in 2020
2016 results
2020 estimate
County
Incum-
bent
party
Victory
margin
Enfran-
chised
voters
Enfran-
chised &
spillover
Shift
in
votes
(pat. 1)
Advan-
tage
(pat. 1)
Partisan
swing
(pat. 1)
Shift
in
votes
(pat. 2)
Advan-
tage
(pat. 2)
Partisan
swing
(pat. 2)
Duval
Rep
5,968
9,222
25,084
10,486
Dem
Yes
8,429
Dem
Yes
Jefferson
Rep
389
439
1,194
530
Dem
Yes
397
Dem
Yes
Pinellas
Rep
5,500
11,565
31,457
7,964
Dem
Yes
10,619
Dem
Yes
Seminole
Rep
3,529
4,489
12,210
3,650
Dem
Yes
4,113
Dem
Yes
St. Lucie
Rep
3,408
5,977
16,257
2,921
Dem
No
5,508
Dem
Yes
Statewide
Rep
112,911
195,205
530,953
146,464
Dem
Yes
179,062
Dem
Yes
Source: See assumptions and own analysis.
21
Scenario: 2 – Low Turnout: Expectation of Amendment 4 to swing the vote
In the low scenario, with a turnout of just 5%, we would expect no county to change control
and the governance of the state to remain with the incumbent party. Employing party
affiliation pattern 1 (modelling affiliation after 150.000 returning citizens enfranchised in
2007), the vote would turn more Republican in 2 counties, would remain stable in 12 counties
and would become more Democratic in the remaining 53 counties as well as at state level. Of
the post-sentencing disenfranchised voters only ~1.45% would cast their votes.
If SB 7066 were repealed in this context, the number of enfranchised voters that cast their vote
would go up to ~5% or ~73,000. There would be no partisan impact on county or the state
level.
Graph 4: The electoral significance of Amendment 4 in 2020 – Low
22
Scenario: 3 – High Turnout: Expectation of Amendment 4 to swing the vote
Even in a high scenario the expected voter turnout among the enfranchised population would
reach only 35%. With SB 7066 in place, two counties and the statewide elections might flip
from Republican to Democratic control. Depending on the party pattern employed (true only
for pattern 2) these predictions differ (Table 8). We would predict ~127,000 new voters to cast
their votes or 8.5% of the post-sentencing disenfranchised population in Florida. Looking at
the predictions on a county level according to party affiliation pattern 1 (modelling affiliation
after 150,000 returning citizens enfranchised in 2007), the marginal vote would turn more
Republican in 2 counties, would remain stable in 2 counties and would become more
Democratic in the remaining 63 counties as well as at state level.
Table 8: The electoral significance of Amendment 4 in 2020 – High scenario
The electoral significance of Amendment 4 in 2020
2016 results
2020 estimate
County
Incum-
bent
party
Victory
margin
Enfran-
chised
voters
Enfran-
chised &
spillover
Shift
in
votes
(pat. 1)
Advan-
tage
(pat. 1)
Partisan
swing
(pat. 1)
Shift
in
votes
(pat. 2)
Advan-
tage
(pat. 2)
Partisan
swing
(pat. 2)
Pinellas
Rep
5,500
7,574
20,601
3,988
Dem
No
6,952
Dem
Yes
St. Lucie
Rep
3,408
3,985
10,839
1,390
Dem
No
3,656
Dem
Yes
Statewide
Rep
112,911
126,841
345,003
74,617
Dem
No
116,483
Dem
Yes
Source: See assumptions and own analysis.
23
Graph 5: The electoral significance of Amendment 4 in 2020 - High
The electoral consequences of SB 7066 are highlighted by Table 9. The table depicts our
predictions without SB 7066 in place. We would expect ~509,000 returning citizens to vote
instead of ~127,000 an increase of ~382,000 votes. This would constitute ~34% of the post-
sentencing disenfranchised population in the state instead of 8.5%. In terms of partisan
politics, this would translate into 5 counties flipping from Republican to Democratic control.
In addition, the statewide results would change from Republican to Democratic (Table 9).
24
Table 9: The electoral significance of SB 7066 in 2020 – High scenario
The electoral significance of SB 7066 in 2020
2016 results
2020 estimate
County
Incum-
bent
party
Victory
margin
Enfran-
chised
voters
Enfran-
chised &
spillover
Shift
in
votes
(pat. 1)
Advan-
tage
(pat. 1)
Partisan
swing
(pat. 1)
Shift
in
votes
(pat. 2)
Advan-
tage
(pat. 2)
Partisan
swing
(pat. 2)
Duval
Rep
5,968
22,958
62,446
10,486
Dem
Yes
21,102
Dem
Yes
Jefferson
Rep
389
1098
2,987
530
Dem
Yes
995
Dem
Yes
Pinellas
Rep
5,500
30,375
82,620
7,964
Dem
Yes
27,872
Dem
Yes
Seminole
Rep
3,529
11,687
31,789
3,650
Dem
Yes
10,750
Dem
Yes
St. Lucie
Rep
3,408
15,981
43,468
2,921
Dem
Yes
14,666
Dem
Yes
Statewide
Rep
112,911
508,644
1,383,512
299,164
Dem
Yes
467,111
Dem
Yes
Source: See assumptions and own analysis.
See Appendix D for detailed predictions.
In conclusion, our analysis indicates that the impact of Amendment 4 on the electoral map of
2020 is significantly dampened if SB 7066 remains in place. If however it is repealed, we
expect the Amendment to swing the vote at the state level in the 2020 Presidential election.
25
SPILL-OVER EFFECTS
A key assumption in our modelling is the large spill-over effect that voter participation of
returning citizens will have on the people around them. The current literature suggests that
enfranchised individuals who live in communities with a high percentage of disenfranchised
individuals have a lower than state average voter turnout. This dampening effect was first
studied by Marc Mauer and Ryan King for the Sentencing Project. Their analysis focused on
specific districts in Georgia where Black Males had a 5% lower turnout rate compared to their
White Male counterparts.
25
Building on this research, Browers and Preuhs used a statistical
analysis to further demonstrate that the negative effect of felony disenfranchisement on the
political participation of non-felons was statistically significant in Black communities.
26
Anecdotally, Desmond Meade said this effect was common sense to him. “Back in the Civil
Rights Era, when dad went to vote he took his whole family. That civic engagement was part
of the conversation at the dinner table for the family… When you strip dad and mom the right
to vote then you’re not having those conversations.” This emphasizes the point that the impact
on felony disenfranchisement laws expands beyond those who are directly impacted
(returning citizens) but includes those in their families and communities and specifically has
an increased impact on those in Black communities.
Understanding the literature and contextual background to the dampening effects of felony
disenfranchisement laws on the broader community and electorate, we expected the opposite
to occur through Amendment 4. This is to say, that with the restoration of voting rights for a
specific population, that their family members and community will find themselves more
responsive to the democratic process than they previously were. Sean Shaw, former State Rep
and Democratic AG candidate in 2018, commenting on this effect:
“If people go through the effort of getting their rights restored then it certainly seems to me
that you’re going have that inverse kind of thing be true where people in that household are
now going to take voting that much more serious because they got someone who had it
stripped away and now have it restored… If an individual has jumped through the hoops to get
their rights restored they are going to be anything but apathetic [about voting] and that affects
the people around them.”
27
This underscores that the impact is not limited to returning citizens alone but includes family
and community level impact.
25
King, Ryan S., and Marc Mauer. 2004. The Vanishing Black Electorate: Felony Disenfranchisement
in Atlanta, Georgia. The Sentencing Project. Available at hhttp://www.sentencing- project.org
26
Bowers, M. and Preuhs, R.R. (2009), Collateral Consequences of a Collateral Penalty: The Negative
Effect of Felon Disenfranchisement Laws on the Political Participation of Nonfelons*. Social Science
Quarterly, 90: 722-743. doi:10.1111/j.1540-6237.2009.00640.x
27
Sean Shaw, in expert interview. February 2020
26
BARRIERS TO REACHING THE BALLOT BOX FOR THE ENFRANCHISED
Table 10: Barriers to electoral impact for enfranchised
Post-
sentencing
disenfran-
chised voters
Qualified
offenses acc.
to literal
Amendment
Qualified
offenses acc.
to SB 7066
Qualified by
having
fulfilled all
LFOs
Individuals in
Florida
1,487,847
1,454,156
1,435,033
362,614
Total share
100%
98%
96%
24%
Sample
individuals
396,104
363,210
350,136
59,602
Share of
sample
100%
92%
88%
15%
Non-sample
individuals
1,091,743
1,090,946
1,084,897
303,012
Share of non-
sample
100%
100%
99%
28%
Based on our estimate of the population eligible to vote after Amendment 4, SB 7066 and
expert interviews, we further detailed the barriers to the impact of Amendment 4. The
subsequent infographic illustrates the road for returning citizens to reclaim their vote, placing
the above in the larger context while extracting the barriers in more detail based on the
interviews.
We identified three cross-cutting barriers:
a. Informational barrier
b. Financial barrier
c. Mobilizing barrier
27
28
INFORMATIONAL BARRIERS
The informational barrier persists throughout the entire “road” to reach the ballot box. The
difficulty accessing relevant information or the ignorance on the options available are main
focal points that need to be addressed.
What is Owed
In the Financial Barrier section of the roadmap, we mention that LFOs are an automatic
indicator of whether an individual is eligible to have their voting rights restored. As such,
accessing how much is owed is an integral step in the realization of one’s rights being restored.
But, in the state of Florida there is no centralized method of accessing this information. Most
of the LFOs are owed to and collected by the county in which an individual was convicted.
Each county therefore has an account of what is owed to it, although county records are often
incomplete. While many returning citizens will be aware of their debt due to collection
solicitations, accessing that information in lieu of these solicitations are a significant barrier,
especially if it is spread out across multiple counties. This barrier is compounded for
individuals who completed the terms of their sentence prior to the digitization of this
information. Recovering the paper trail is particularly difficult in these cases.
As LFOs are often the responsibility of a county, if an individual no longer lives in that
particular county, was convicted in federal court, or convicted in a different state, contacting
the relevant officials to obtain a precise tabulation of what is owed can be difficult.
Eligibility
Beyond LFOs, there are two eligibility related barriers that currently exist in the process. The
first is based on what convictions are excluded from the restoration of voting rights under
Amendment 4. The text of the Amendment kept in restrictions for those convicted of murder
and felony sex offenses. However, as was referenced in the development of the predictive
model and shown in the appendices, there are a significant number of convictions that may
fall under each. The second is the uncertainty around the completion of the terms of one’s
sentence. The current laws lay out that the completion of prison, parole, probation, and most
recently fines and fees are required before the restoration of voting rights. Yet, an individual
may be concerned whether their court mandated substance abuse counseling is included in
that.
These two informational components stand as barriers for returning citizens due to their
limited access of information, especially in the absence of nonprofit advocates or legal aid. A
demonstration of how difficult this process can be is shown below.
An initial google search of “how do I get my voting rights restored in Florida” does not reveal
any up to date information on what the process is post Amendment 4 and SB7066. As such,
we expect someone to decide to go to the Secretary of State’s website as they are responsible
for elections in Florida.
On the Secretary of State website, finding information on if one is eligible to vote is not an
easy process. On the Elections page of the website, there is no clear tab or option targeted
towards returning citizens on information related to their voting rights. If one uses the
website’s search bar and searches for phrases such as “how do I get my voting rights restored?”
“voting restored” or “disenfranchised”, no hits are found. It is only when one searches the word
“felon” that a link for a voting FAQ which contains a paragraph for those seeking to have their
rights restored (below).
29
Clicking on the first link provided simply takes you to the home page of the Florida
Department of Corrections. There is no indication on the DOC website of how to see one’s
eligibility regarding voting rights. In fact, a search in the “offender search” tab, which requires
a name and DC number, reveals only a person’s specific convictions and the release date from
prison.
The next step would be to visit the county clerk’s website, which neither provides concrete
information on one’s status. This means that one must visit the county clerk’s office, that also
could be a barrier dependent on an individual’s ability to visit the office during business hours.
As was hopefully demonstrated in this brief example, the process of determining your
eligibility can be a discouraging and seemingly insurmountable process.
Options available
The last major subcategory of informational barriers deals with an individual’s awareness of
the specific legal options in restoring their rights. This is particularly relevant regarding the
options surrounding LFOs. As previously mentioned, an individual can petition the court to
have their fines and fees waived, reduced or converted into community service. While this
option is open to all returning citizens, only four counties (Miami-Dade, Broward, Palm Beach
and Hillsborough) have instituted a process of grouping petitioners and processing the
requests simultaneously to expedite the process. Although these counties represent around
35% of Florida’s total population, those who do not live in these counties must wait indefinitely
for their case is heard. More importantly, knowledge of this legal option is not widespread,
especially for returning citizens who may live in rural or other areas not covered by advocacy
groups.
Uncertainty created
Each of these informational hurdles increases the uncertainty around voting right
restorations. While reclaiming your voting right is a significant step in restoring an individual’s
citizenship within society the returning citizen’s efforts to have her/his rights restored are
inhibited by this uncertainty. This is exacerbated by the fact that the benefit associated with
voting is outweighed by the punishment of voting illegally, that is the impact of a single vote
is generally small however, should an individual erroneously cast a vote the punishment is
30
third-degree felony carrying a maximum sentence of 5 years.
28
In 2018, a Texas woman who
believed her rights had been restored, voted in the election and was sentenced to 5 years in
prison.
29
This imbalance is compounded by placing the responsibility to know your right to
vote on the returning citizen. Election officials will not be able to confirm eligibility at the time
of registration. As such, the uncertainty surrounding the information of one’s status and the
outsized impact of being wrong may keep many from attempting to register to vote.
Additionally, the current legal battle that is ensuing around SB 7066 has served only to
increase the uncertainty around who is eligible or not. With every ruling and appeal, Floridians
are left unsure and confused on what the state of the legislation is. The impact of this
uncertainty can have a permanent effect on an individual and is what Bruce Riley of Voice of
the Experienced, in Louisiana, describes as the power of “word on the street”. “Word on the
street actually has power. If word on the street says you don’t have the right to vote it actually
doesn’t matter if you have the right to vote or not because you are going to listen to word on
the street.”
30
Once individuals are convinced that they are not eligible to vote, it is difficult to
change their minds. This barrier highlights that regardless of the outcome of the legal battle
surrounding SB 7066, we expect that the uncertainty created by the law will have a lasting
impact.
FINANCIAL BARRIERS
Perhaps the most intuitive barrier is the financial cost due to SB 7066. As previously
mentioned, we expect the enactment of this law to result in the disenfranchisement of ~70%
of the previously enfranchised individuals. This cohort holds billions of dollars in outstanding
fines and fees. For most returning citizens this financial burden is prohibitive.
31
This barrier is
binary in nature – you’re either disenfranchised because you have LFOs or you are not. This
can only be resolved by one’s ability to pay off their fines and fees or by having them reduced
or converted into community service by the courts.
It should be noted, irrespective of SB 7066, that Florida’s LFO system has been one of the most
punitive in the nation. The current system makes no exemptions for indigent individuals and
traps them in a vicious cycle of debt to the state. For an in-depth analysis of Florida’s LFO
system, we suggest referencing the Brennan Center’s publication on the topic.
32
MOBILIZING BARRIERS
Once an individual has determined their eligibility, there are still barriers that exist in getting
out the vote. They will manifest themselves at different stages throughout the process. The
three main subcategories identified were apathy, lack of engagement, and standard turnout
issues.
Apathy
28
s. 775.082
29
Romo, Vanessa, and Sasha Ingber. “Texas Woman Sentenced To 5 Years For Illegal Voting.” NPR.
NPR, March 31, 2018. https://www.npr.org/sections/thetwo-way/2018/03/31/598458914/texas-
woman-sentenced-to-5-years-for-illegal-voting.
30
Bruce Riley, in expert interview. March 2020
31
Sweeney, Dan. “South Florida Felons Owe a Billion Dollars in Fines - and That Will Affect Their Ability
to Vote.” sun, May 31, 2019. https://www.sun-sentinel.com/news/politics/fl-ne-felony-fines-broward-
palm-beach-20190531-5hxf7mveyree5cjhk4xr7b73v4-story.html.
32
Diller, Rebekah. “The Hidden Costs of Florida’s Criminal Justice Fees.” Brennan Center for Justice,
2010. https://www.brennancenter.org/sites/default/files/2019-08/Report_The Hidden-Costs-
Florida's-Criminal-Justice-Fees.pdf.
31
A significant barrier is apathy towards the political system. Many Americans believe that
participation in the democratic process does not have an impact on governance or are
uninterested in the governance because they perceive it as hostile towards them. However, for
many returning citizens the apathy is deeper than a simple distaste for politics. Desmond
Meade, founder of FRRC and a returning citizen himself, had an acute description of the
psychological factors at play.
“There is a level of belonging that is damaged when you take someone’s right to vote away and,
in my experience, its personal and professional. When a person can’t vote that is something
painful. Whether they want to vote or not, the fact that they’re told they can’t vote is a stark
reminder that you are not a part of our society. And that’s painful. Because the human instinct,
the natural human instinct, is to be a part of this group, to be a part of something and to be
told you’re that you’re not right is something painful. So we mask that, with an indifference
we find the way to dull or nullify the pain. What comes out of this is “I don’t give a damn about
voting don’t matter who gets in office, we’re still gonna be messed up, my vote don’t count.”
33
This barrier does not have a technical solution but requires significant engagement on an
individual level to be overcome.
Lack of Engagement
Lack of engagement efforts and the nature of these efforts hinder the turnout of returning
citizens. Many returning citizens are concentrated in districts with low voting propensity. This
often leads to reduced attention by electoral engagement efforts. Additionally, with returning
citizens as a new voting population, the engagement infrastructure is often not yet in place.
The efforts around this may increase as organizations and political parties recognize the
potential impact the population can have.
The method of engagement determines its success. It is important to recognize that the
restoration of voting rights is greater than the transactional nature of turning out to the polls
and voting for a specific candidate. In conversations with advocates that are working in this
area, there was an emphasis that the restoration of voting rights was about restoring an
individual’s dignity and their status as a first-class citizen.
This sentiment was echoed with individuals who worked with voter engagement of returning
citizens from Louisiana to Alabama. Restoration of citizens’ rights and one’s dignity must be
front and center to any strategy aimed at mobilizing and organizing returning citizens. Neil
Volz, FRRC board member, explained that for their mission “it is about returning citizens lives,
it’s about getting people plugged into the community, it’s about people educated on the
issues.”
34
Any engagement and organizing effort that isn’t centered around issues of dignity and
citizenship, will not be successful.
Standard Turnout Issues
The previously mentioned barriers were all specific to returning citizens, they however also
face the same barriers as the general population. This includes a lack of transportation, voter
suppression, long voting lines, lack of early voting, and myriad of other factors. Resolving
these issues will increase the impact of both returning citizens and the general population.
33
Desmond Meade, in expert interview, January 2020.
34
Neil Volz, in expert interview, January 2020.
32
RECOMMENDATIONS
Our research shows that Amendment 4 will have a positive impact on democratic participation
in Florida by returning citizens and their communities. Below are recommendations for the
Florida legislature, the Sentencing Project, and other advocates to push that even further. We
want to caveat that our position is that there should be no limitations on an individual’s right
to vote, however, we recognize that some changes must be done incrementally to be successful
politically. As such some recommendations may not be fully in line with that ethos on
disenfranchisement but are rather pragmatic solutions.
Our recommendations are categorized in line with the barriers. There is some overlap across
categories, demonstrating the interplay between them
INFORMATIONAL RECOMMENDATIONS
Create a centralized informational process
As was demonstrated previously, obtaining information on one’s eligibility can be a difficult
and complicated process. This informational barrier necessitates the creation of a streamlined
process that combines the requirements of the Secretary of State, the Department of
Corrections and county clerks. Providing clear and actionable steps for returning citizens to
have their voting rights restored will decrease the concerns that are associated with the current
system.
We recommend that the State of Florida develop a centralized system with which an individual
can input their name and DC number and immediately find out what is owed and if they are
eligible. With the state potentially being hesitant to invest in such a process, this is also an
opportunity for a third party, although administration through a non-governmental
organization has obvious draw backs. Nonetheless, the data Dan Smith’s team has collected is
a fantastic starting point although it is currently being treated as proprietary information.
It is important to emphasize that the creation of a database is addressing two issues – LFOs
and conviction eligibility. These two issues have inputs from very different sources. Gaining
access and combining these inputs comes with a degree of difficulty, that must be thoughtfully
managed.
Change the burden of proof
A significant barrier preventing individuals from registering to vote is that the burden of proof
for eligibility falls on the returning citizen as Florida does not maintain a system of
determining eligibility at the point of registration. This creates huge uncertainty that will
discourage returning citizens.
We recommend that the burden of proof should be changed from the returning citizen to the
state. Registration to vote may only be successfully completed should the individual legally
qualify, thus protecting the returning citizen from prosecution if it is later determined they
were not eligible. It would be the fault of the system not the individual.
Change language in registration process
When an individual decides to register to vote, whether online or with a physical form, they
are met with the following language:
33
“If I have been convicted of a felony, I affirm my voting rights have been restored pursuant to
s. 4 , Art. VI of the State Constitution upon the completion of all terms of my sentence,
including parole or probation.”
We recommend that the Secretary of State changes the language of the registration form from
“including parole or probation” to “including parole, probation, and the payment of fines and
fees.” This change is necessary for individuals who might not be aware of the details postulated
in SB 7066. It is possible for a returning citizen to be aware of the passage of Amendment 4
granting them their voting rights and when they go online to register see only the mention of
parole or probation. This would suggest to them that they are eligible, despite the possibility
that they still owe fines and fees. Changing the language is a simple way to ensure that
returning citizens are not registering despite being ineligible.
Provide information at last official contact with the system
Access to information has been shown to be a difficult process for many returning citizens.
Since some individuals might not be aware of the options available to them or the restrictions
they are under, it is important that this information is provided in an accessible method. We
recommend that returning citizens are provided an information packet regarding their voting
rights at their last official contact with the criminal justice system. Whether this is released
from prison or the last meeting with a parole or probation officer, an individual should be
provided with relevant information on how they can get their rights restored.
34
FINANCIAL RECOMMENDATIONS
Repeal SB 7066
While it is unlikely, unless by court order, we recommend that Governor DeSantis and the
Florida legislature repeal SB 7066. The implementation of SB 7066 has had an outsized impact
on how many people are disenfranchised. Repealing the law will not only grant back the voting
rights of the approximately 1,000,000 disenfranchised but will also result in a positive spill
over effect in their communities.
Change in Sentencing
The possibility to convert or reduce LFOs has been prioritized by previously mentioned
counties. Streamlining this process provides an opportunity to mitigate some of the
repercussions of the de-facto poll tax. The willingness of certain counties to petition the
modification of one’s sentence post-release to help restore voting rights provides an
opportunity to change sentencing. This is to say, that if the court knows that the individual will
be unable to pay their fines and fees post-release, they might advocate to have the fines and
fees dropped or changed to community service at the time of sentencing. Currently in Florida,
there are certain categories of offenses in which the assessment of fines is discretionary.
35
Leveraging both the discretionary nature and the provisions allotted in SB 7066, will allow the
reduction of fines and fees for any individual prior to serving their sentence. Addressing their
LFOs at this juncture will ensure the process of restoring their voting rights is less complicated
post-release.
Increase Fundraising Efforts
With the implementation of SB 7066, the financial burden to voting is the most limiting factor
to the impact of Amendment 4. As Floridians wait for the legal battle to be settled in the courts,
an effective action in restoring individual’s rights is paying off their fines and fees. Currently,
FRRC created a “Fines and Fees Fund” that has raised $500,000 and has paid the legal
obligations for 200 individuals.
36
With billions of outstanding fines and fees, addressing this
financial burden can be a daunting task.
37
It is therefore crucial to approach this effort
strategically. This can mean developing a methodology that prioritizes the repayment of LFOs
in a way that maximizes the number of individuals the fund helps. For instance, financial
assistance is particularly influential for returning citizens who don’t live in the counties that
have made the transformation of LFOs a priority.
Given the high cost associated with paying one’s LFOS, it should be noted that using those
funds to support organizing efforts may result in higher democratic participation from
returning citizens. We suggest that future study is conducted on the turnout rates for returning
citizens who have their fines and fees paid for. Tracking this information would better inform
decisions on how to allocate the limited resources.
35
Diller, Rebekah. “The Hidden Costs of Florida’s Criminal Justice Fees.” Brennan Center for Justice,
2010. https://www.brennancenter.org/sites/default/files/2019-08/Report_The Hidden-Costs-
Florida's-Criminal-Justice-Fees.pdf.
36
Perry, Mitch. “Early Christmas Gift for 31 Hillsborough Co. Felons: Right to Vote Restored.” Early
Gift for 31 Hillsborough Felons Right to Vote Restored, December 20, 2019.
https://www.baynews9.com/fl/tampa/news/2019/12/20/in-hillsborough-county--an-early-xmas-
gift-for-31-felons---their-right-to-vote-is-restored.
37
Sweeney, Dan. “South Florida Felons Owe a Billion Dollars in Fines - and That Will Affect Their Ability
to Vote.” sun, May 31, 2019. https://www.sun-sentinel.com/news/politics/fl-ne-felony-fines-broward-
palm-beach-20190531-5hxf7mveyree5cjhk4xr7b73v4-story.html.
35
Partisan interests in counties
Our impact map allows us to identify the county parties that have a partisan interest in
facilitating the process of translating LFOs into community service. This may result in partisan
interest groups pushing the process streamlining of LFO transformation as modelled by
Broward, Miami-Dade, Palm Beach and Hillsborough along.
The respective counties for the Democratic party are:
• Alachua
• Broward
• Gadsden
• Hillsborough
• Leon
• Miami-Dade
• Orange
• Osceola
• Palm beach
The respective counties for the Republican party are:
• Citrus
• Dixie
• Gilchrist
• Hardee
• Holmes
• Okeechobee
• Santa Rosa
Exempt 10yrs + from LFO Requirement
During our research, we discovered the added difficulty of discovering what is owed for
individuals who completed their sentences prior to their records being digitized. As a result of
this added barrier, we recommend that Florida enact an exemption of the LFO requirement
for those who have completed their sentencing over 10 years ago. This legislation would not
remove the requirement to pay one’s LFOs but would mean the LFO requirement assessed to
the restoration of voting rights would no longer apply. In the immediate term, this would
enfranchise a significant number of returning citizens who may find it difficult to determine if
they have LFOs.
A critique of this recommendation centers around its usefulness beyond the initial
implementation. The proposed recommendation’s primary objective is providing a means of
mitigating the issue of returning citizens whose documentation may not have been digitized
and potentially lost. Yet, for future returning citizens there is no basis for the argument since
their information will be available in new processes. Long term, we recommend that this
legislation would serve as an incentive or reward for those who have avoided interactions with
the criminal justice system. This would be similar to Louisiana’s law that restores voting rights
to returning citizens who have been out of prison for five years but remain under probation or
parole.
38
38
Crisp, Elizabeth. “Gov. John Bel Edwards Signs Law Restoring Felon Voting Rights after Five
Years.” The Advocate, May 31, 2018.
https://www.theadvocate.com/baton_rouge/news/politics/legislature/article_8f432008-6515-11e8-
a42c-4f773f2862c5.html.
36
MOBILIZING RECOMMENDATIONS
Create a streamlined engagement strategy
The quantitative and statistical modelling is useful in helping us predict the impact of the
Amendment on future elections. However, this impact can only be realized, and expanded, if
individuals are engaged, their rights restored, and turned out to vote. Therefore, we
recommend creating a playbook that can be used by various groups to engage returning
citizens. This strategy should be created in coordination with various advocacy groups and
stakeholders to ensure that it is centered around dignity and citizenship. The creation of a
strategy template will allow for those who wish to engage with returning citizens to do so in an
effective manner.
Increase use of returning citizens to returning citizen canvassing
As previously mentioned, engaging returning citizens in a purely transactional form may lead
to pushback therefore an effective strategy must be centered around dignity and citizenship.
Combined with this is ensuring that returning citizens are a central component to any
engagement campaign. During their campaign to gather signatures to get Amendment 4 on to
the ballot in 2018, FRRC was intentional about having returning citizens be a driving force in
these efforts. According to Neil Volz, FRRC has continued to strive to have returning citizens
be the main contacts to other returning citizens in their voter registration and mobilizing
efforts.
39
While seemingly intuitive that those closest to the issue will be able to speak most
convincingly about dignity, citizenship, and the importance of voting, this principle has basis
in voter engagement literature. Experimental research on door-to-door canvassing found that
a canvasser simply being from the same zip code had a statistically significant impact.
40
Mobilizing and organizing campaigns around voter registration and turnout for returning
citizens must employ and support returning citizens as a central component. This is
particularly influential in combatting the apathy that many returning citizens feel towards the
democratic process.
Also focus on individuals not convicted of a felony
Given the limiting factor of SB 7066 on the direct impact of Amendment 4, spillover effects
pose a unique opportunity to expand the impact of the Amendment. For those interested in
expanding democratic participation, there is a benefit to focus on these often-neglected
communities as they are positioned for a resurgence in political involvement. Often, political
campaigns create their mobilization and turnout strategies based on past turnout trends in
specific districts or precincts, thus allocating resources and attention to areas that traditionally
have high turnout. However, this can result in a self-fulfilling prophecy, as communities that
have low turnout rates become neglected leading to less political engagement. Because of the
impact of spillover effects, both positive and negative, we expect that a majority of these low
turnout precincts will see a rise in willingness to engage in the political process due to
individuals in their families and communities who get their rights restored.
39
Neil Volz, in expert interview, January 2020.
40
Green, Donald P., and Alan S. Gerber. Get out the Vote: How to Increase Voter Turnout. Washington,
D.C.: Brookings Institution Press, 2019.
37
SUMMARY OF RECOMMENDATIONS
Recommendations
• Centralized Informational Process
• Change Burden of Proof
• Change Registration Language
• Information Packet at Last Contact
Recommendations
• Repeal SB 7066
• Change Sentencing
• Increase Fundraising
• Partner with Counties
• Exempt 10+ years from LFO requirement
Recommendations
• Streamlined Engagement Strategy
• RC to RC outreach
• Spillover Community outreach
38
APPENDIX
APPENDIX A – DATA SOURCES
a) Official Florida voter registration and unofficial voting history information
by the Florida Division of Elections (last updated per 03/20)
41
Official voter registration information: Data is extracted from the Florida Voter
Registration System and includes information on voters who are officially registered as of the
end of the prior month. Public information is included except in those cases in which an
exemption applies, including if a registered voter requested exemption from public disclosure
pursuant toSection 119.071, Fla. Stat. (Section 97.0585, Fla. Stat.).All information on
registered voters who are participants in the Attorney General’s Address Confidentiality
Program pursuant to 741.401 –741.465, Fla. Stat., is excluded.
Unofficial voting history information: Data is extracted from independently reported
voting history from the 67 county supervisors of elections. The data is captured at a fixed point
in time and may not always correspond to information in a county’s voter registration file
because the voter may have moved to another county after voting. You must look to the voter’s
registration information in that county-specific voter registration file. Disclaimer: The
Division of Elections cannot warrant the content or accuracy of the unofficial voting history
extract. The most accurate and current official voting history resides with the supervisors of
elections. Any discrepancies in this extract must be resolved in favor of the official voting
records in the county where the voter actually voted.
For details on contact and variables, please visit the Florida Division of Elections voter extract
disk file layout at: https://dos.myflorida.com/media/696057/voter-extract-file-layout.pdf .
b) Public records of OBIS offender data base in Florida by the Florida
Department of Corrections (last per 01/20)
42
The OBIS offender database compiles Florida Department of Correction data for current
inmates, released and supervised populations. This data is accessible at:
http://www.dc.state.fl.us/pub/obis_request.html .
Inmate Population Information
In the interest of public safety, the Florida Department of Corrections makes available to the
Department of Corrections staff and the criminal justice community, information and
photographs of inmates who are incarcerated. This information is completely refreshed
weekly, with the exception of release dates and location changes which are nightly. For
additional information, please contact the Department of Corrections, Bureau of Classification
and Central Records, Policy and Systems Development Section, at (850) 488-9859 or
central.classificati[email protected].com.
Inmate Release Information
In the interest of public safety, the Florida Department of Corrections makes available to the
Department of Corrections staff and the criminal justice community, information and
photographs of inmates who either have been released or are scheduled to be released. This
information is updated nightly. For additional information, please contact the Department of
Corrections, Bureau of Admission and Release, at (850) 488-9167 or
dc.release@fdc.myflorida.com.
41
https://dos.myflorida.com/media/696057/voter-extract-file-layout.pdf
42
http://www.dc.state.fl.us/pub/obis_request.html
39
Supervised Population Information
In the interest of public safety, the Florida Department of Corrections makes available to the
Department of Corrections staff and the criminal justice community, information and
photographs of offenders who are under supervision. This information is completely refreshed
weekly, with the exception of termination dates and location changes which are updated
nightly. For additional information, please contact the Department of Corrections, Probation
and Parole Field Services, at (850) 717-3444 or co[email protected]rida.com.
An overview of the variables included in the respective data bases, can be accessed at:
https://drive.google.com/file/d/1TbBF-
FhQ7fj3JwogTBBZqzLwVPT0iVzQ/view?usp=sharing .
c) Historic Florida election results on county and precinct level by the MIT
election lab (last updated: 12/19)
43
The MIT election lab supports advances in election science by collecting, analyzing, and
sharing core data and findings.
The data on the results of the 2016 presidential election that we used as our baseline is
available at: https://electionlab.mit.edu/data .
d) 2016 Actuarial Life Table by the Social Security Administration
44
A life table is a concise way of showing the probabilities of a member of a particular population
living to or dying at a particular age. In this study, the life tables are used to examine the
mortality changes in the Social Security population over time. For this table, the period life
expectancy at a given age is the average remaining number of years expected prior to death for
a person at that exact age, born on January 1, using the mortality rates for 2016 over the course
of his or her remaining life.
The table is accessible at: https://www.ssa.gov/oact/STATS/table4c6.html .
e) Statute table by the Florida Department of Law Enforcement (last updated:
12/19)
The FDLE Statute Table is a comprehensive legislative resource containing current and past
Florida criminal offenses. It was developed for use in various criminal justice applications in
conjunction with the Office of State Courts Administrator, the Florida Court Clerks and
Comptrollers, the Department of Juvenile Justice, and the Department of Corrections.
The Clerk and Arrest versions of the Florida Statute Tables are updated annually by laws that
create, amend, or repeal statutory material. The offenses and violations are listed by Florida
statute number.
The Arrest Statute Table used by law enforcement agencies contains only the current
arrestable offenses. The Clerk Statute Table used by the Clerks of Court contains the arrestable
offenses in addition to statutory history, such as when a statute was repealed or renumbered.
For more information on the 2019 release of the Statute Table, please refer to the CJIS
Memorandum that was distributed by FDLE.
The table used in our analysis to establish the connection between statue and code value, is
accessible via: https://web.fdle.state.fl.us/statutes/about.jsf .
43
https://electionlab.mit.edu/data
44
https://www.ssa.gov/oact/STATS/table4c6.html
40
f) 2019 Florida’s Criminal Punishment Code: A Comparative Assessment by the
Florida Department of Corrections (last updated: 10/19)
45
The intent of the report is to address the requirement set forth in Florida Statute 921.002(4)(a)
to analyze sentencing events under the Florida Criminal Punishment Code. Each year, the
Department of Corrections is required to report on trends in sentencing practices and
sentencing score thresholds, and provide an analysis of the sentencing factors considered by
the courts. On October 1, 2018, the digitized criminal punishment code scoresheet system was
implemented statewide. This report details the 92,730 scoresheets received with sentence
dates in FY 2018-2019 received by the Florida Department of Corrections, for felony offenses
committed on or after October 1, 1998.
The information relevant to this analysis can be found on page 32 and the following pages in
the table titled “Offense Type Based on Best Match to Statute or Primary Offense Code on
Scoresheet using Offense Categories Specific to the Department of Corrections”.
45
http://www.dc.state.fl.us/pub/scoresheet/Criminal%20Punishment%20Code%202019.pdf
41
APPENDIX B – ASSUMPTIONS
b) Mapping offense type to sanction imposed for felons
46
Source: f) Florida’s Criminal Punishment Code.
Utilizing the 2019 Florida Criminal Punishment Code, we drew out the table below, which
depicts the number of offenses in the FY 2019 by offense types and sanctions imposed. Below
shows the offense types most relevant for our analysis. On this crude level, we differentiated
roughly whether disqualification would be applicable at a literate reading of the Amendment
or as specified in SB 7066. We didn’t level this out over a series of years, because only starting
from the latest report the granularity was available at this level. We used this information to
determine disqualification for the population that was not included in our sample.
Offense Type
County Jail
Probation
Community
Control
Literal
Bill
01-Capital Murder
0
6
1
x
x
02-2nd Degree Murder
2
1
5
x
04-Homicide, Other
2
6
1
x
07-Capital Sexual Battery
1
2
3
x
x
08-Life Sexual Battery
7
3
1
x
x
09-1st Degree Sexual Battery
3
18
3
x
x
10-2nd Degree Sexual Battery
9
49
12
x
11-Sexual Assault, Other
7
8
3
x
12-Lewd/Lascivious Behavior
36
178
49
x
Total
23300
38162
5022
Literal
100.0%
99.9%
99.8%
Bill
99.7%
99.3%
98.4%
The split of sanctions imposed (County Jail, Probation, Community Control) on the post-
sentencing historically disenfranchised population that was not included in our sample fell for
our estimation is based on information in the criminal punishment code. To level out potential
fluctuation between years, we worked with an average over the last three years.
Year
County
Jail
Probatio
n
Community
Control
State
Prison
Othe
r
Total
2018-2019
23300
38162
5022
21536
315
88335
2017-2018
23252
40188
4236
16994
774
85444
2016-2017
21096
38809
4221
16768
631
81525
Average
22549
39053
4493
18433
573
85101
46
http://www.dc.state.fl.us/pub/scoresheet/Criminal%20Punishment%20Code%202019.pdf
42
On this basis we split the post-sentencing historically disenfranchised population not reflected
in our sample into:
County
jail
Probation
Community
Control
Total
34%
59%
7%
66669
43
c) Disqualifying offenses acc. to literal interpretation of Amendment
d) Disqualifying offenses acc. to SB 7066
Source: b) Public records of OBIS offender data base, classification according to own
assessment.
Offenses
Bill
Literal
SEX BATT V12-17, NO INJURY
Yes
Yes
1ST DG MUR/PREMED. OR ATT.
Yes
Yes
SEX BAT BY ADULT/VCTM LT 12
Yes
Yes
POSS PHOTO ETC CHILD SEX PERF
Yes
Yes
USE OF CHILD, PROM.SEX PERFM
Yes
Yes
L/L MOLEST V<12 OFF 18+
Yes
Yes
PROV.OBSCENE MAT.MINOR
Yes
Yes
SECOND DEG.MURDER,COMM.OF
FELO
Yes
No
SEX BAT/MINOR,FAM/CUST AUTH
Yes
No
L&L BATT FORCE/ENTICE ABUSE
Yes
No
L/L EXHIBIT. V<16 OFF 18+
Yes
Yes
ADLT SEX W/16-17 YR OLD
Yes
Yes
2ND DEG.MURD,DANGEROUS ACT
Yes
No
L/L, INDEC.ASLT CHILD U/16
Yes
Yes
SEXUAL BATTERY UNSPECIFIED
Yes
No
SEX BAT/COERCES BY THREAT
Yes
No
SEX BAT/ WPN. OR FORCE
Yes
No
SEX BAT/INJURY NOT LIKELY
Yes
No
RAPE - STRONGARM
Yes
Yes
LEWD ASLT/SEX BAT VCTM<16
Yes
Yes
L/L MOLEST V12-15 OFF 18+
Yes
Yes
L/L CONDUCT V<16 OFF 18+
Yes
Yes
L&L BATT MINOR
Yes
Yes
SX OFFNDR FAIL COMPLY PSIA
Yes
No
KIDNAP V<13/AGG.CHLD ABUSE
Yes
Yes
1ST DEG MUR,COM.OF FELONY
Yes
Yes
SEX BAT/PHYS HELPLESS RESIST
Yes
No
SEX BATT V & O 18+
Yes
No
USE OF CHILD, ENG SEX PERFM
Yes
Yes
INTERNET/LURE CHILD FOR SEX
Yes
Yes
SEND MINOR HARMFUL INFO.
Yes
No
SEX BATT V & O 18+, NO INJURY
Yes
No
FALSE IMPRISON MINOR AGA.WILL
Yes
No
VID VOYEUR-CHLD<16
Yes
Yes
1ST DEG MUR,DEATH FRM DRUGS
Yes
Yes
SEX ASSLT-SODOMY-MAN-OTHER
Yes
No
L/L BATT.CAUSE SXACT V<16
Yes
Yes
SEND CHILD PORN.
Yes
Yes
L/L BATT.SEX W/V12-15
Yes
Yes
44
KIDNAP MINOR AGAINST WILL
Yes
Yes
SEX ACT COERCE CHILD BY ADULT
Yes
Yes
USE OF CHILD, SEX PHOTO ETC
Yes
Yes
SEX BAT/MENTALLY DEFECTIVE
Yes
No
SEX OFFENSE-OTHER
Yes
No
SEX OFF, FALSE INFO/OMIT
Yes
No
SEX BATT V12-17 O18+
Yes
Yes
HOMICIDE,KILL UNBORN CHLD
Yes
Yes
SEX BAT BY JUVEN/VCTM UNDER 12
Yes
Yes
SEX OFF/PRDR POSS ALT DL/ID
Yes
No
SEX BATT/CARNAL INTERC. U/18
Yes
Yes
SX.OFNDR.RE-REGISTR.VIOL.
Yes
No
LURE CHILD INTO BLDG/CAR
Yes
Yes
SEX PRED, TRANS RES NO REP
Yes
No
SEX.BAT/VICT O/12 INCAP
Yes
Yes
SEX OFF, TRANS NO REP
Yes
No
SEX BAT/COERCES BY RETALIATION
Yes
No
KIDNAP V<13/SEX. BATTERY
Yes
Yes
COMM SEX MINOR/MENT INCAP
Yes
Yes
POS PLAC MAN SUB/MINOR EXIST
Yes
No
SEX BATT V12+ O<18, NO INJURY
Yes
Yes
VISUALLY EXP.MINOR/SEX.CONDUCT
Yes
Yes
MANSLAUGHTER/UNBORN CHILD
Yes
No
SEX PRED FAIL TO REGISTER
Yes
No
ABUSE OF HUMAN CORPSE
Yes
No
SEX ASSAULT/BATTERY
Yes
No
SEX PRED, FALSE INFO/OMIT
Yes
No
L/L EXHBT - PRESNCE OF EMPLYEE
Yes
No
L/L COMP. EXHIB. OFF 18+
Yes
No
SX.OFNDR.NOT NOTIF.SHERIFF
Yes
No
LEWD MOLEST.-ELDER/DISABLD
Yes
No
LEWD EXHIBIT-ELDER/DISABLD
Yes
No
INCEST WITH RELATIVE
Yes
No
SEX ASSLT-OTHER/OTHER STATE
Yes
No
WWW/ASK PARENT:SEX W/MINOR
Yes
Yes
KIDNAP MINOR FOR RANSOM
Yes
No
HIV-POS.HAS SEX 2+ VIOLATIONS
Yes
No
HOMICIDE-OTHER/OTHER STATE
Yes
Yes
USE/PROD - MINOR SX MATERIAL
Yes
Yes
L/L MOLEST V12-15 OFF<18
Yes
Yes
KIDNAP V<13/L&L FS800.04
Yes
Yes
L/L MOLEST V<12 OFF<18
Yes
Yes
SEX BATT V <12 FAM CAP
Yes
Yes
SEX TRAFF-VIC<14 OR DEATH
Yes
Yes
SEX BATT V <12 FAM LIFE
Yes
Yes
45
1ST DG.MURDER/UNBORN CHILD
Yes
Yes
SEX BATT V 12+ , PRIOR CONV.
Yes
Yes
SX.PRED.NOT NOTIF.SHERIFF
Yes
No
FELONY MURDER-NONSEX
Yes
No
L/L EXHIBIT. V<16 OFF<18
Yes
Yes
SEX TRAFFICKING
Yes
Yes
SX OFFNDR FAIL RPT.NONMOVE
Yes
No
HUMAN TRFC:SX TRADE,CHILD<18
Yes
Yes
HUMAN TRAFFICKING:SEX TRADE
Yes
No
SEX ACT/SOLICIT OF CHILD
Yes
Yes
USE COMPUTR SEX W/MINOR
Yes
Yes
SEX PRED.FAIL RPT.NONMOVE
Yes
No
DISCLOSE SEX DISEASE INFO
Yes
No
SEX BATT/SOLICITATION OF CHILD
Yes
Yes
LEWD BATT.ON ELDER/DISABLD
Yes
No
SEX BATT V 12+ O<18
Yes
Yes
HIV-POSITIVE HAS SEX
Yes
No
PURC.CUSTODY OF MINOR/SEX.COND
Yes
Yes
FALS.IMP.V<13/SEX. BATTERY
Yes
Yes
SEXUAL BATTERY BY LEO
Yes
No
HUM TFK:SEXUAL ADULT
Yes
No
HARBR/HIDE/LIE RE SX.OFNDR.
Yes
No
FALS.IMP.V<13/L&L FS800.04
Yes
No
STAFF SEX W/INMATE/OFENDER
Yes
No
HUMAN TRFC:SX TRADE,CHILD<15
Yes
Yes
ADLT/CUST VID VOYEUR-CHLD<16
Yes
Yes
L/L CONDUCT V<16 OFF<18
Yes
Yes
SEX BAT/VICTIM DRUGGED
Yes
No
SEND CHILD PORN-OTH STATE
Yes
No
PROVIDE FALSE REPORT STD
Yes
No
HT:SEX TRANSF STATE MINOR
Yes
Yes
SEX BATT V12+, PRIOR CONV
Yes
Yes
SEX MISC.PSYCHOTHERAPIST
Yes
No
AGG.STALK-SXOFDR VIOL.ORDER
Yes
Yes
ADLT/CUST VID VOYUER-CHLD<16
Yes
Yes
SEX OFNDR ILLEGAL RESIDENCE
Yes
Yes
BATTERY CODE INPECTOR
Yes
Yes
HUMAN TRAFFIC - VIOLENT
Yes
No
HUMAN TRFC:SX TRADE,INTRSTATE
Yes
No
STAFF SEX W/APD CLIENT
Yes
No
EMPLOYEE SEX W/DJJ OFFNDR
Yes
No
2ND DG.MURDER/UNBORN CHILD
Yes
Yes
SEX VIO INJUC VIO 2+ PRI CON
Yes
Yes
HOMICIDE-WILLFUL KILL-FMLY-GUN
Yes
Yes
RAPE WITH OTHER WEAPON
Yes
Yes
46
SEX ASSLT - CARNAL ABUSE
Yes
No
SEX ASSLT-SODOMY-BOY-STGARM
Yes
Yes
HOMICIDE
Yes
Yes
SEX ASSLT-SODOMY-BOY-GUN
Yes
Yes
HOMICIDE-WLFL KILL-NONFMLY-
GUN
Yes
Yes
RAPE - GUN
Yes
Yes
HOMICIDE-WILLFUL KILL-GUN
Yes
Yes
HOMICIDE-WILFL KILL-FMLY-OTHER
Yes
Yes
SEX ASSLT-SODOMY-WOMAN-
STGARM
Yes
No
HOMICIDE-WLFL KILL-PUB OFF-OTH
Yes
Yes
HOMICIDE-WILFUL W/VESSEL
Yes
Yes
SEX ASSLT-SODOMY-GIRL-OTHER
Yes
Yes
OBSCENE EXHIBIT/INVOLV MINOR
Yes
Yes
SELL CHILD FOR MONEY
Yes
Yes
KIDNAP/CHILD ABUSE INVES.
Yes
Yes
DETENTION STAFF SEX W/INMATE
Yes
No
SXOFDR NO CONTACT ORD VIOL.
Yes
No
SX.OFNDR.REFUSE PHOTO
Yes
No
HOMICIDE-WLFL KILL-POL OFF-OTH
Yes
Yes
SEX ASSLT-SODOMY-WOMAN-OTHER
Yes
No
OBSCENE PRESENTATION W/MINOR
Yes
Yes
CONT TO DELINQ OF MINOR
Yes
No
SEX ASSLT-SODOMY-GIRL-STGARM
Yes
Yes
SEX ASSLT-SODOMY-BOY-OTHER
Yes
Yes
HOMICIDE-WLFL KILL-POL OFF-GUN
Yes
Yes
SEX CYBERHARRASSMENT
Yes
No
TERRORISM
Yes
No
DCF STAFF SEX W/PATIENT
Yes
No
LURE CHILD PREV. 794 VIOL
Yes
Yes
47
e) Legal Financial Obligations
Assumptions for eligibility based on outstanding LFOs:
Outstanding
LFOs
Assumption of eligibility
0 $
100% expected to be eligible by election
day
Up to 100 $
50% expected to be eligible by election
day
Up to 250 $
20% expected to be eligible by election
day
Up to 500 $
10% expected to be eligible by election day
Up to 1000 $
5% expected to be eligible by election day
More than 1000
$
1% expected to be eligible by election day
Assumptions for LFOs in OBIS data base (sample population) according to county and race:
Source: Dan A. Smith expert testimony
47
County
Race
0
(%)
Up to
100 (%)
Up to
250 (%)
Up to
500 (%)
Up to
1000 (%)
More than
1000 (%)
Total
share of
eligible
voters
Alachua
White
&
others
0.05
0.01
0.02
0.13
0.22
0.57
0.09
Alachua
Black
0.13
0.03
0.03
0.16
0.26
0.40
0.18
Baker
White
&
others
0.18
0.02
0.04
0.16
0.19
0.42
0.23
Baker
Black
0.11
0.02
0.02
0.13
0.28
0.44
0.15
Bay
White
&
others
0.16
0.04
0.03
0.08
0.18
0.51
0.21
Bay
Black
0.11
0.04
0.03
0.08
0.20
0.55
0.15
Bradford
White
&
others
0.29
0.04
0.03
0.13
0.22
0.29
0.35
Bradford
Black
0.25
0.03
0.03
0.15
0.23
0.31
0.30
Brevard
White
&
others
0.27
0.03
0.04
0.06
0.14
0.46
0.31
Brevard
Black
0.16
0.03
0.03
0.06
0.16
0.56
0.20
47
https://www.brennancenter.org/sites/default/files/2019-
10/SupplementalExpertReportofDanielA.SmithPh.D.UniversityofFloridaSeptember172019.pdf .
48
Broward
White
&
others
0.16
0.04
0.03
0.08
0.18
0.51
0.21
Broward
Black
0.11
0.04
0.03
0.08
0.20
0.55
0.15
Calhoun
White
&
others
0.09
-
0.02
0.02
0.14
0.72
0.11
Calhoun
Black
0.11
0.01
0.02
0.06
0.22
0.58
0.14
Charlotte
White
&
others
0.09
0.09
0.08
0.19
0.25
0.30
0.18
Charlotte
Black
0.07
0.10
0.10
0.22
0.27
0.24
0.17
Citrus
White
&
others
0.17
0.01
0.02
0.04
0.05
0.71
0.20
Citrus
Black
0.12
0.02
0.04
0.06
0.08
0.68
0.15
Clay
White
&
others
0.16
0.04
0.03
0.08
0.18
0.51
0.21
Clay
Black
0.11
0.04
0.03
0.08
0.20
0.55
0.15
Collier
White
&
others
0.15
0.04
0.03
0.11
0.17
0.51
0.20
Collier
Black
0.09
0.03
0.02
0.12
0.19
0.55
0.14
Columbia
White
&
others
0.08
0.02
0.02
0.06
0.19
0.62
0.12
Columbia
Black
0.06
0.02
0.02
0.06
0.21
0.63
0.10
Desoto
White
&
others
0.07
0.02
0.02
0.02
0.15
0.71
0.11
Desoto
Black
0.08
0.02
0.02
0.03
0.10
0.75
0.11
Dixie
White
&
others
0.11
0.03
0.06
0.06
0.18
0.57
0.16
Dixie
Black
0.03
0.05
-
0.09
0.09
0.74
0.08
Duval
White
&
others
0.11
0.03
0.01
0.05
0.44
0.36
0.16
Duval
Black
0.07
0.03
0.02
0.05
0.46
0.38
0.11
Escambia
White
&
others
0.16
0.04
0.03
0.08
0.18
0.51
0.21
49
Escambia
Black
0.11
0.04
0.03
0.08
0.20
0.55
0.15
Flagler
White
&
others
0.10
0.08
0.02
0.07
0.40
0.33
0.17
Flagler
Black
0.09
0.07
0.04
0.08
0.40
0.31
0.17
Franklin
White
&
others
0.13
0.01
0.02
0.11
0.17
0.56
0.17
Franklin
Black
0.05
0.02
0.02
0.09
0.29
0.52
0.10
Gadsden
White
&
others
0.18
0.19
0.17
0.23
0.15
0.08
0.33
Gadsden
Black
0.09
0.15
0.15
0.25
0.21
0.16
0.23
Gilchrist
White
&
others
0.12
0.04
0.02
0.14
0.16
0.51
0.17
Gilchrist
Black
0.06
-
0.01
0.19
0.28
0.46
0.10
Glades
White
&
others
0.19
0.13
0.12
0.13
0.19
0.25
0.30
Glades
Black
0.10
0.10
0.12
0.22
0.26
0.19
0.22
Gulf
White
&
others
0.17
0.00
0.02
0.04
0.13
0.64
0.19
Gulf
Black
0.12
-
0.05
0.05
0.15
0.63
0.15
Hamilton
White
&
others
0.02
0.09
0.05
0.15
0.16
0.54
0.10
Hamilton
Black
0.02
0.03
0.04
0.11
0.16
0.64
0.07
Hardee
White
&
others
0.13
0.02
0.01
0.04
0.15
0.66
0.16
Hardee
Black
0.02
0.04
-
0.06
0.24
0.64
0.07
Hendry
White
&
others
0.09
0.05
0.03
0.17
0.26
0.39
0.16
Hendry
Black
0.08
0.04
0.03
0.17
0.26
0.42
0.14
Hernando
White
&
others
0.16
0.04
0.03
0.08
0.18
0.51
0.21
Hernando
Black
0.11
0.04
0.03
0.08
0.20
0.55
0.15
50
Highlands
White
&
others
0.12
0.01
0.01
0.01
0.08
0.77
0.14
Highlands
Black
0.06
0.01
0.01
0.02
0.06
0.84
0.08
Hillsborough
White
&
others
0.16
0.04
0.03
0.08
0.18
0.51
0.21
Hillsborough
Black
0.11
0.04
0.03
0.08
0.20
0.55
0.15
Holmes
White
&
others
0.09
0.01
0.01
0.02
0.08
0.80
0.11
Holmes
Black
0.04
-
0.01
0.03
0.10
0.82
0.06
Indian river
White
&
others
0.04
0.04
0.02
0.13
0.21
0.56
0.10
Indian river
Black
0.02
0.02
0.03
0.15
0.20
0.57
0.07
Jackson
White
&
others
0.08
0.01
0.01
0.06
0.23
0.61
0.11
Jackson
Black
0.05
0.01
0.01
0.04
0.25
0.64
0.08
Jefferson
White
&
others
0.36
0.05
0.04
0.18
0.26
0.10
0.43
Jefferson
Black
0.17
0.08
0.05
0.29
0.25
0.16
0.26
Lafayette
White
&
others
0.13
0.07
0.04
0.13
0.13
0.50
0.20
Lafayette
Black
0.10
0.07
0.07
0.07
0.31
0.38
0.18
Lake
White
&
others
0.30
0.02
0.01
0.00
0.01
0.66
0.32
Lake
Black
0.25
0.02
0.01
0.01
0.01
0.71
0.27
Lee
White
&
others
0.12
0.12
0.07
0.11
0.18
0.40
0.22
Lee
Black
0.07
0.10
0.07
0.10
0.20
0.46
0.15
Leon
White
&
others
0.24
0.01
0.01
0.02
0.06
0.65
0.26
Leon
Black
0.17
0.01
0.01
0.02
0.06
0.73
0.19
Levy
White
&
others
0.14
0.03
0.07
0.30
0.27
0.18
0.22
51
Levy
Black
0.06
0.01
0.04
0.45
0.25
0.20
0.13
Liberty
White
&
others
0.19
0.05
0.03
0.08
0.25
0.40
0.25
Liberty
Black
0.20
0.03
0.05
0.12
0.34
0.26
0.26
Madison
White
&
others
0.07
0.04
0.03
0.09
0.16
0.61
0.12
Madison
Black
0.07
0.05
0.04
0.10
0.13
0.61
0.13
Manatee
White
&
others
0.10
0.03
0.02
0.09
0.22
0.52
0.15
Manatee
Black
0.06
0.01
0.02
0.10
0.24
0.57
0.10
Marion
White
&
others
0.20
0.04
0.02
0.04
0.15
0.54
0.24
Marion
Black
0.11
0.04
0.02
0.04
0.14
0.65
0.15
Martin
White
&
others
0.18
0.02
0.03
0.07
0.18
0.52
0.22
Martin
Black
0.07
0.02
0.02
0.06
0.23
0.60
0.11
Miami-dade
White
&
others
0.16
0.04
0.03
0.08
0.18
0.51
0.21
Miami-dade
Black
0.11
0.04
0.03
0.08
0.20
0.55
0.15
Monroe
White
&
others
0.06
0.17
0.05
0.12
0.27
0.32
0.19
Monroe
Black
0.03
0.14
0.06
0.12
0.28
0.37
0.14
Nassau
White
&
others
0.14
0.02
0.02
0.17
0.27
0.38
0.19
Nassau
Black
0.08
0.02
0.02
0.24
0.29
0.35
0.14
Okaloosa
White
&
others
0.14
0.27
0.07
0.06
0.26
0.20
0.31
Okaloosa
Black
0.07
0.22
0.09
0.11
0.25
0.25
0.23
Okeechobee
White
&
others
0.06
0.02
0.01
0.03
0.12
0.75
0.09
Okeechobee
Black
0.04
0.02
0.00
0.04
0.21
0.69
0.07
52
Orange
White
&
others
0.19
0.02
0.02
0.05
0.24
0.48
0.22
Orange
Black
0.10
0.01
0.02
0.04
0.24
0.59
0.13
Osceola
White
&
others
0.16
0.04
0.03
0.08
0.18
0.51
0.21
Osceola
Black
0.11
0.04
0.03
0.08
0.20
0.55
0.15
Palm beach
White
&
others
0.16
0.02
0.01
0.10
0.29
0.42
0.20
Palm beach
Black
0.08
0.01
0.01
0.09
0.26
0.55
0.11
Pasco
White
&
others
0.02
0.04
0.03
0.06
0.18
0.68
0.06
Pasco
Black
0.02
0.02
0.03
0.05
0.24
0.65
0.05
Pinellas
White
&
others
0.16
0.04
0.03
0.08
0.18
0.51
0.21
Pinellas
Black
0.11
0.04
0.03
0.08
0.20
0.55
0.15
Polk
White
&
others
0.12
0.08
0.02
0.04
0.10
0.64
0.18
Polk
Black
0.09
0.09
0.02
0.04
0.10
0.67
0.15
Putnam
White
&
others
0.08
0.02
0.03
0.13
0.28
0.46
0.13
Putnam
Black
0.04
0.03
0.04
0.15
0.24
0.50
0.10
Santa rosa
White
&
others
0.25
0.01
0.02
0.03
0.21
0.49
0.28
Santa rosa
Black
0.13
0.02
0.01
0.04
0.30
0.50
0.16
Sarasota
White
&
others
0.13
0.05
0.04
0.18
0.22
0.39
0.19
Sarasota
Black
0.06
0.02
0.03
0.17
0.25
0.48
0.11
Seminole
White
&
others
0.21
0.19
0.12
0.26
0.16
0.07
0.36
Seminole
Black
0.12
0.19
0.14
0.30
0.18
0.08
0.28
St. johns
White
&
others
0.40
0.02
0.02
0.10
0.12
0.34
0.43
53
St. johns
Black
0.34
0.02
0.03
0.12
0.13
0.36
0.37
St. lucie
White
&
others
0.42
0.01
0.01
0.02
0.10
0.45
0.44
St. lucie
Black
0.16
0.01
0.01
0.01
0.08
0.73
0.18
Sumter
White
&
others
0.08
0.01
0.01
0.03
0.10
0.77
0.10
Sumter
Black
0.07
0.01
0.01
0.02
0.09
0.80
0.10
Suwannee
White
&
others
0.04
0.08
0.03
0.05
0.14
0.67
0.10
Suwannee
Black
0.00
0.06
0.02
0.05
0.11
0.75
0.06
Taylor
White
&
others
0.11
0.01
0.03
0.02
0.07
0.77
0.13
Taylor
Black
0.08
0.01
0.01
0.02
0.09
0.78
0.10
Union
White
&
others
0.31
0.06
0.04
0.09
0.17
0.33
0.37
Union
Black
0.23
0.07
0.04
0.10
0.16
0.41
0.29
Volusia
White
&
others
0.08
0.04
0.01
0.07
0.34
0.45
0.13
Volusia
Black
0.05
0.03
0.02
0.07
0.36
0.48
0.10
Wakulla
White
&
others
0.37
0.00
0.01
0.04
0.11
0.46
0.39
Wakulla
Black
0.25
0.01
0.02
0.04
0.11
0.57
0.27
Walton
White
&
others
0.18
0.02
0.02
0.03
0.12
0.62
0.21
Walton
Black
0.14
0.03
0.01
0.03
0.12
0.68
0.17
Washington
White
&
others
0.09
0.02
0.01
0.03
0.06
0.80
0.12
Washington
Black
0.07
-
0.02
0.04
0.06
0.81
0.09
For non-sample population:
County
0
(%)
Up to
100 (%)
Up to
250 (%)
Up to
500 (%)
Up to
1000 (%)
More than
1000 (%)
Share paid
LFOs
54
Total
22%
5%
4%
10%
21%
38%
27.93%
55
f) Life expectancies
Source: Social Security Administration Actuarial Life Table
48
Below table details the expected years of life at a given age differentiated by male and female.
While life expectancy is lower for people that have been incarcerated, this approximation
sufficed for our purposes.
Age
Male
Female
0
76.04
80.99
1
75.52
80.43
2
74.55
79.46
3
73.58
78.48
4
72.59
77.49
5
71.6
76.5
6
70.62
75.51
7
69.63
74.52
8
68.64
73.53
9
67.64
72.54
10
66.65
71.54
11
65.66
70.55
12
64.66
69.56
13
63.67
68.56
14
62.68
67.57
15
61.7
66.58
16
60.73
65.6
17
59.76
64.62
18
58.81
63.63
19
57.86
62.66
20
56.91
61.68
21
55.98
60.71
22
55.05
59.73
23
54.13
58.76
24
53.22
57.8
25
52.3
56.83
26
51.38
55.86
27
50.47
54.9
28
49.55
53.93
29
48.63
52.97
30
47.72
52.01
31
46.8
51.05
32
45.89
50.09
33
44.97
49.14
34
44.06
48.19
35
43.15
47.23
48
https://www.ssa.gov/oact/STATS/table4c6.html
56
36
42.23
46.28
37
41.32
45.34
38
40.41
44.39
39
39.5
43.45
40
38.59
42.5
41
37.69
41.56
42
36.78
40.62
43
35.88
39.69
44
34.98
38.76
45
34.08
37.83
46
33.19
36.9
47
32.3
35.98
48
31.43
35.07
49
30.55
34.16
50
29.69
33.26
51
28.84
32.36
52
27.99
31.48
53
27.16
30.59
54
26.34
29.72
55
25.52
28.85
56
24.72
27.99
57
23.93
27.13
58
23.15
26.28
59
22.37
25.44
60
21.61
24.6
61
20.85
23.76
62
20.11
22.94
63
19.37
22.12
64
18.65
21.3
65
17.92
20.49
66
17.2
19.69
67
16.49
18.89
68
15.78
18.11
69
15.09
17.33
70
14.4
16.57
71
13.73
15.82
72
13.07
15.09
73
12.43
14.37
74
11.8
13.66
75
11.18
12.97
76
10.58
12.29
77
10
11.62
78
9.43
10.98
79
8.88
10.35
80
8.34
9.74
57
81
7.82
9.15
82
7.32
8.58
83
6.84
8.04
84
6.38
7.51
85
5.94
7.01
86
5.52
6.53
87
5.12
6.07
88
4.75
5.64
89
4.4
5.23
90
4.08
4.85
91
3.78
4.5
92
3.5
4.18
93
3.25
3.88
94
3.03
3.61
95
2.83
3.37
96
2.66
3.16
97
2.51
2.96
98
2.37
2.79
99
2.25
2.63
100
2.13
2.48
101
2.02
2.33
102
1.91
2.19
103
1.81
2.06
104
1.71
1.93
105
1.61
1.81
58
g) Voter turnout
Below lists literature that we have consulted to estimate voter turnout among the newly
enfranchised population. We used this as a basis to delineate the three turnout scenarios.
Year
Authors
Study
2009
Michael V. Haselswerdt
Con Job: An Estimate of Ex-Felon Voter Turnout Using
Document-Based Data
2002
Christopher Uggen and Jeff Manza
Democratic Contraction? Political Consequences of
Felon Disenfranchisement in the United States
2010
Randi Hjalmarsson and Mark Lope
The Voting Behaviour of Young Disenfranchised Felons:
Would They Vote if They Could?
2014
Meredith and Morse
The Politics of the Restoration of Ex-Felon Voting
Rights: The Case of Iowa
2011
Traci Burch
Turnout and Party Registration among Criminal
Offenders in the 2008 General Election
2014
Meredith and Morse
Do Voting Rights Notification Laws Increase Ex-Felon
Turnout?
2018
Meredith and Morse
Vox: Why letting ex-felons vote probably won’t swing
Florida
2020
Florida Rights Restoration Coalition
Orlando Mayor RCT
59
APPENDIX C – DETAILED PREDICTIONS
a) All enfranchised potential voters
Potential impact prediction on county level through Amendment 4
Including all potential voters
County
Enfranchised
in sample
Share
of
sample
Enfranchised
in
population
Enfranchised
& spillover in
population
Alachua
867
1%
5275
14348
Baker
157
0%
955
2598
Bay
1495
3%
9095
24738
Bradford
247
0%
1503
4088
Brevard
2560
4%
15575
42364
Broward
4650
8%
28290
76949
Calhoun
61
0%
371
1009
Charlotte
379
1%
2306
6272
Citrus
441
1%
2683
7298
Clay
485
1%
2951
8027
Collier
498
1%
3030
8242
Columbia
303
1%
1843
5013
Desoto
125
0%
760
2067
Dixie
71
0%
432
1175
Duval
2687
5%
16348
44467
Escambia
1724
3%
10489
28530
Flagler
168
0%
1022
2780
Franklin
61
0%
371
1009
Gadsden
517
1%
3145
8554
Gilchrist
53
0%
322
876
Glades
47
0%
286
778
Gulf
85
0%
517
1406
Hamilton
33
0%
201
547
Hardee
90
0%
548
1491
Hendry
132
0%
803
2184
Hernando
562
1%
3419
9300
Highlands
234
0%
1424
3873
Hillsborough
4726
8%
28753
78208
Holmes
63
0%
383
1042
Indian river
198
0%
1205
3278
Jackson
150
0%
913
2483
Jefferson
129
0%
785
2135
Lafayette
25
0%
152
413
60
Lake
1335
2%
8122
22092
Lee
1401
2%
8524
23185
Leon
1580
3%
9613
26147
Levy
191
0%
1162
3161
Liberty
59
0%
359
976
Madison
71
0%
432
1175
Manatee
642
1%
3906
10624
Marion
1634
3%
9941
27040
Martin
386
1%
2348
6387
Miami-dade
5108
9%
31077
84529
Monroe
280
0%
1703
4632
Nassau
205
0%
1247
3392
Okaloosa
1023
2%
6224
16929
Okeechobee
102
0%
621
1689
Orange
3423
6%
20825
56644
Osceola
817
1%
4971
13521
Palm beach
2192
4%
13336
36274
Pasco
435
1%
2647
7200
Pinellas
3557
6%
21641
58864
Polk
2846
5%
17315
47097
Putnam
268
0%
1630
4434
Santa rosa
607
1%
3693
10045
Sarasota
697
1%
4240
11533
Seminole
1369
2%
8329
22655
St. johns
936
2%
5695
15490
St. lucie
1874
3%
11401
31011
Sumter
123
0%
748
2035
Suwannee
113
0%
687
1869
Taylor
104
0%
633
1722
Union
105
0%
639
1738
Volusia
1250
2%
7605
20686
Wakulla
532
1%
3237
8805
Walton
223
0%
1357
3691
Washington
90
0%
548
1491
Source: See assumptions and own analysis.
61
b) Scenario 1: Medium
Impact prediction on county level through Amendment 4
Medium: Voter turnout at 16% and 12% for enfranchised black and non-black
voters
Party pattern 1
Party pattern 2
County
Democra
ts
Republica
ns
Other
s
Democra
ts
Republica
ns
Other
s
Alachua
1670
348
133
1191
462
514
Baker
150
117
49
182
65
82
Bay
1572
1093
530
1754
677
745
Bradford
299
166
82
299
117
133
Brevard
2780
1836
895
3044
1175
1292
Broward
7018
2598
1175
5940
2301
2532
Calhoun
65
33
16
65
33
33
Charlotte
348
299
150
430
166
182
Citrus
364
364
182
498
199
215
Clay
462
364
182
563
215
231
Collier
446
381
182
563
215
248
Columbia
364
199
101
364
150
150
Desoto
166
82
33
150
65
65
Dixie
65
65
33
82
33
33
Duval
4104
1491
661
3441
1341
1472
Escambia
2415
1025
479
2168
843
928
Flagler
215
101
49
199
82
82
Franklin
65
49
16
65
33
33
Gadsden
1009
215
82
729
280
316
Gilchrist
49
49
16
65
16
33
Glades
49
33
16
49
16
16
Gulf
101
49
33
101
33
49
Hamilton
49
16
16
49
16
16
Hardee
65
65
33
101
33
33
Hendry
166
101
49
166
65
65
Hernando
498
446
215
645
248
264
Highlands
248
182
82
280
117
117
Hillsborough
6422
2913
1357
5892
2285
2516
Holmes
33
49
33
65
16
33
Indian river
231
133
65
231
101
101
Jackson
182
101
49
166
65
65
Jefferson
199
65
33
166
65
65
Lafayette
33
16
0
33
16
16
Lake
1605
911
430
1621
628
696
Lee
1556
1009
479
1670
645
713
62
Leon
2630
794
348
2067
811
876
Levy
182
150
65
215
82
101
Liberty
49
33
16
65
16
33
Madison
133
33
16
101
33
49
Manatee
745
446
215
778
299
332
Marion
1803
1159
563
1937
762
827
Martin
381
299
150
462
182
199
Miami-dade
7249
3044
1390
6438
2500
2747
Monroe
280
215
101
332
133
133
Nassau
215
150
65
231
101
101
Okaloosa
1110
745
348
1208
462
514
Okeechobee
82
82
33
117
49
49
Orange
4453
2184
1025
4219
1637
1803
Osceola
827
612
299
960
364
413
Palm beach
2796
1406
661
2682
1042
1142
Pasco
364
348
182
498
199
215
Pinellas
4352
2366
1126
4319
1670
1836
Polk
3460
1904
895
3460
1341
1472
Putnam
332
182
82
332
133
133
Santa rosa
479
498
248
677
264
280
Sarasota
677
530
248
811
316
348
Seminole
1771
860
413
1670
645
713
St. johns
1074
661
316
1126
430
479
St. lucie
2067
1341
645
2233
860
960
Sumter
150
82
33
150
49
65
Suwannee
117
82
33
133
49
49
Taylor
133
65
33
133
49
49
Union
117
65
33
133
49
49
Volusia
1455
860
413
1491
579
645
Wakulla
514
413
199
612
248
264
Walton
199
182
82
248
101
117
Washington
82
65
33
101
33
49
Source: See assumptions and own analysis.
63
c) Scenario 2: Low
Impact prediction on county level through Amendment 4
Low: Voter turnout at 5% for enfranchised voters
Party pattern 1
Party pattern 2
County
Democrat
s
Republica
ns
Other
s
Democrat
s
Republica
ns
Other
s
Alachua
530
133
49
397
150
166
Baker
65
49
16
65
33
33
Bay
563
446
215
677
264
280
Bradford
101
65
33
117
49
49
Brevard
1009
745
364
1175
446
498
Broward
2350
1025
479
2119
827
911
Calhoun
33
16
0
33
16
16
Charlotte
133
117
65
166
65
65
Citrus
150
150
65
199
82
82
Clay
166
150
82
215
82
101
Collier
182
166
82
231
82
101
Columbia
133
82
33
133
49
65
Desoto
49
33
16
49
16
16
Dixie
16
16
16
33
16
16
Duval
1374
596
264
1224
479
530
Escambia
811
397
182
778
299
332
Flagler
65
33
16
65
33
33
Franklin
33
16
0
33
16
16
Gadsden
316
65
33
231
82
101
Gilchrist
16
16
0
16
0
0
Glades
16
16
0
16
0
0
Gulf
33
16
16
33
16
16
Hamilton
16
16
0
16
0
0
Hardee
16
33
16
33
16
16
Hendry
49
33
16
65
16
33
Hernando
199
182
82
248
101
117
Highlands
101
65
33
117
49
49
Hillsboroug
h
2217
1159
547
2168
843
928
Holmes
16
16
16
33
16
16
Indian river
82
49
33
101
33
33
Jackson
65
33
16
65
33
33
Jefferson
65
33
16
65
16
33
Lafayette
0
0
0
16
0
0
Lake
563
364
182
596
231
264
Lee
563
397
199
645
248
264
64
Leon
860
299
133
729
280
316
Levy
65
49
33
82
33
33
Liberty
33
16
0
33
16
16
Madison
33
16
0
33
16
16
Manatee
264
182
82
299
117
133
Marion
645
479
231
745
280
316
Martin
150
117
65
182
65
82
Miami-dade
2467
1208
563
2334
911
993
Monroe
117
101
49
133
49
65
Nassau
65
65
33
101
33
33
Okaloosa
397
299
150
462
182
199
Okeechobee
33
33
16
49
16
16
Orange
1523
876
413
1556
596
661
Osceola
299
248
117
364
150
150
Palm beach
976
563
264
1009
397
430
Pasco
150
150
65
199
82
82
Pinellas
1523
960
462
1621
628
696
Polk
1208
762
364
1292
498
547
Putnam
117
65
33
117
49
49
Santa rosa
182
215
101
280
101
117
Sarasota
264
215
101
316
117
133
Seminole
612
348
166
612
248
264
St. johns
381
264
133
413
166
182
St. lucie
745
530
264
843
332
364
Sumter
49
33
16
49
16
16
Suwannee
33
33
16
49
16
16
Taylor
49
33
16
49
16
16
Union
49
33
16
49
16
16
Volusia
514
348
166
579
215
248
Wakulla
199
166
82
248
101
101
Walton
82
82
33
117
49
49
Washington
33
16
16
33
16
16
Source: See assumptions and own analysis.
65
d) Scenario 3: High
Impact prediction on county level through Amendment 4
High: Voter turnout at 35% for enfranchised voters
Party pattern 1
Party pattern 2
County
Democrat
s
Republica
ns
Other
s
Democrat
s
Republica
ns
Other
s
Alachua
3756
895
364
2764
1074
1175
Baker
413
332
166
498
199
215
Bay
4004
3128
1523
4765
1852
2035
Bradford
745
462
215
794
316
332
Brevard
7034
5247
2549
8174
3161
3492
Broward
16483
7148
3294
14843
5758
6338
Calhoun
182
117
49
199
65
82
Charlotte
911
843
413
1208
462
514
Citrus
976
1042
514
1406
547
596
Clay
1191
1074
530
1540
596
661
Collier
1208
1110
547
1588
612
677
Columbia
895
579
280
960
381
413
Desoto
397
231
101
397
150
166
Dixie
166
166
82
231
82
101
Duval
9615
4088
1888
8590
3327
3658
Escambia
5742
2864
1341
5494
2135
2350
Flagler
514
299
150
530
199
231
Franklin
166
133
65
199
82
82
Gadsden
2249
514
215
1637
628
696
Gilchrist
117
117
65
166
65
65
Glades
133
82
49
150
49
65
Gulf
248
166
82
280
101
117
Hamilton
117
49
16
101
33
49
Hardee
199
215
101
280
117
117
Hendry
381
248
117
413
166
182
Hernando
1325
1273
628
1787
696
762
Highlands
628
479
231
745
280
316
Hillsboroug
h
15422
8125
3822
15093
5843
6438
Holmes
133
150
82
199
82
82
Indian river
579
381
182
628
248
264
Jackson
462
264
133
479
182
199
Jefferson
498
182
82
413
166
182
Lafayette
82
49
16
82
33
33
Lake
3939
2549
1224
4254
1654
1820
Lee
3922
2829
1374
4469
1738
1904
66
Leon
6057
2135
960
5048
1953
2152
Levy
498
413
199
612
231
264
Liberty
150
117
49
182
65
82
Madison
299
82
33
231
82
101
Manatee
1836
1240
596
2035
794
860
Marion
4534
3327
1605
5212
2018
2233
Martin
976
843
413
1224
479
530
Miami-dade
17226
8424
3939
16301
6321
6966
Monroe
713
612
299
895
348
381
Nassau
530
446
215
661
248
280
Okaloosa
2812
2103
1009
3261
1257
1390
Okeechobee
248
231
117
332
133
133
Orange
10758
6139
2929
10921
4235
4668
Osceola
2135
1754
860
2614
1009
1110
Palm beach
6819
3988
1904
6999
2715
2978
Pasco
993
1009
498
1390
530
596
Pinellas
10690
6702
3210
11353
4401
4850
Polk
8522
5377
2581
9085
3525
3873
Putnam
811
498
231
860
332
364
Santa rosa
1341
1455
729
1937
745
827
Sarasota
1771
1523
745
2217
860
944
Seminole
4319
2448
1159
4368
1689
1869
St. johns
2647
1852
895
2978
1159
1273
St. lucie
5195
3805
1836
5973
2317
2549
Sumter
381
215
101
397
150
166
Suwannee
299
248
117
364
150
150
Taylor
316
199
101
332
133
133
Union
316
215
101
348
133
150
Volusia
3623
2448
1175
3988
1556
1705
Wakulla
1341
1175
579
1705
661
729
Walton
547
514
248
729
280
316
Washington
231
182
82
280
117
117
Source: See assumptions and own analysis.
67
APPENDIX D – CODE FOR REPRODUCTION
You can find replication code and underlying data at: https://github.com/Alex1005-
stack/PAE