America’s Broken Criminal Legal System Contributes to Wealth Inequality

An estimated 70 million to 100 million Americans—roughly 1 in 3 U.S. adults—have an incarceration, conviction, or arrest record, which is a direct consequence of decades of mass incarceration and overcriminalization.¹ Meanwhile, according to a Center for American Progress analysis, nearly half of U.S. children now have at least one parent with such a record.² In addition, failed criminal legal policies have saddled an ever-expanding swath of the nation’s population with the stigma of a criminal record. America’s failed experiment with mass incarceration and overcriminalization—compounded by the proliferation of criminal background checks in the digital era—has upended countless lives. Moreover, it has birthed a nationwide criminal records crisis that, in turn, has become a significant driver of poverty and racial inequality.³

Over the past decade, a large and growing body of research has documented a host of negative, long-term economic consequences—often called collateral consequences—associated with interaction with the criminal legal system.⁴ Appropriately, particular attention has been paid to the dramatic toll that a conviction and/or incarceration record takes on an individual’s employment and earnings prospects in an era when roughly 9 in 10 U.S. employers use background checks in hiring.⁵ In one recent watershed analysis, researchers at the Brennan Center for Justice found that formerly incarcerated Americans see their subsequent earnings reduced by an average of 52 percent—and that, in the aggregate, Americans with criminal convictions face lost wages in excess of $372 billion every year.⁶ Research has also documented an alarming rise in fines, fees, and other criminal legal debts that can total thousands or even tens of thousands of dollars, diminishing families’ ability to save for the future and protect against financial shocks and uncertainty.⁷

However, scant attention has been paid to the effects of incarceration⁸ on savings and ownership in the United States. This report seeks to add to the existing research literature⁹ by highlighting the consequences of criminal legal system involvement for household wealth. Moreover, given that America’s failed criminal legal policies disproportionately harm Black and Hispanic individuals, families, and communities, this report also examines mass incarceration as an underappreciated driver of America’s racial wealth gap.

Several of the key findings in this report document a link between criminal justice interactions and household wealth:

• Households with a currently or previously incarcerated family member have about 50 percent less wealth than households not affected by incarceration, on average.
• Households with criminal legal system interactions face more obstacles to saving and end up deeper in debt. For example, 21.5 percent of households with a currently or previously incarcerated family member had legal debt in 2019, compared with only 2.1 percent of households without an incarcerated family member.
• Financial insecurity is significantly more widespread among households affected by incarceration. For example, 22.6 percent of households with a currently or previously incarcerated family member could not afford to pay all their bills in 2019, while the same was true for 14.5 percent of households unaffected by incarceration.
• Households affected by incarceration have fewer chances for longer-term wealth building. For instance, 23.4 percent of households with a currently or previously incarcerated family member were denied a loan application in 2019, while 11.2 percent of households unaffected by incarceration were denied a loan.

In addition, this report shows that incarceration correlates with and exacerbates the persistent U.S. racial wealth gap. Further compounding this troubling trend, families of color affected by incarceration face more significant negative economic consequences than white families affected by incarceration. Key findings from this report, further detailed in the tables and figures throughout, include:

• Black and Hispanic households affected by incarceration have a fraction of the wealth of white households affected by incarceration. For example, in 2019, Black households with an incarcerated family member had a median wealth of $1,101, and similarly situated Hispanic households had a median wealth of $3,200. In comparison, white households with an incarcerated family member had $15,330 in median wealth. (All these numbers are exclusive of imputed defined-benefit pension wealth.) Similarly, in 2019, 70.8 percent of Black households and 66.5 percent of Hispanic households with an incarcerated family member had assets—financial and nonfinancial investments—of less than $50,000. The respective share for their white counterparts was 55.3 percent.
• A substantial share of the racial wealth gap can be attributed to larger shares of Black and Hispanic households having an incarcerated family member. Statistical models of the correlates of household wealth attribute about 20 percent of the captured difference between Black and white households and 40 percent of the gap between Hispanic and white households to criminal justice system interactions.
• Black and Hispanic households with a currently or previously incarcerated family member experience greater financial insecurity than white households affected by incarceration. For example, 29.2 percent of Black households and 26.3 percent of Hispanic households with a currently or previously incarcerated family member could not pay all their bills in 2019. For their white counterparts, the share was 19.1 percent.
• Black and Hispanic households affected by incarceration face greater barriers to economic mobility than their white counterparts. For instance, 31.5 percent of Black households and 28.3 percent of Hispanic households were denied a loan in 2019, while this was the case for just 18.1 percent of white households.
• Black and Hispanic families with a currently or previously incarcerated family member face greater obstacles to building wealth than their white counterparts. For example, in 2019, more than half of Black households affected by incarceration—51.4 percent—and 44.8 percent of their Hispanic counterparts reported being dissatisfied with their neighborhood, school, safety, or other amenities; this likely reflects that they live in areas with fewer and worse amenities, thus adding costs, such as for schooling, transportation, and other things. In comparison, only 34.7 percent of white households in this situation indicated that they were dissatisfied.

Leaders at all levels of government continue to reckon with the toll that decades of failed mass incarceration and overcriminalization policies have taken on tens of millions of American families, including these policies’ role in driving widespread economic insecurity and persistent racial inequality. Public policies that remove barriers to economic security and upward mobility for Americans with records and their families are needed now more than ever. Meanwhile, as the nation continues to work to recover and rebuild from the far-reaching effects of the COVID-19 pandemic, it is essential that policymakers enact reforms to remove barriers to employment, housing, education, and wealth building for families affected by mass incarceration. These policy reforms must include instituting fair-chance hiring, licensing, housing, and education; expanding access to criminal record-clearing by making it automatic; and reforming regressive fines-and-fees policies—all key to ensuring the economy no longer leaves tens of millions of system-affected individuals and families behind and closing the racial wealth gap in this country.

Wealth is the difference between what people own, such as a house, retirement savings, and bank accounts, and what they owe—for instance, their mortgage, credit card, or student loan debt. Households need substantial savings to maintain economic stability and security. This means having the ability to cover financial emergencies such as job loss, a broken-down car, or other unexpected expenses or income losses. It also means being able to take advantage of opportunities for upward economic mobility, such as buying a house in a safe and secure neighborhood with good job opportunities; starting and growing a business; investing in children’s education and healthy development; and saving for a secure retirement.

Incarceration does not affect just the individual who spends time in prison or jail; it also has long-term, far-reaching negative consequences for the entire household’s chances at building wealth. As prior CAP research has found, parents’ criminal records have enormous deleterious effects on their children’s future economic chances.¹⁰

The wealth-stripping effects of incarceration are multifaceted and exacerbate the already deeply unequal consequences of the American criminal legal system.¹¹ Families who have or had a member in jail or prison generally face a sharp drop in household income during the period when a wage earner is incarcerated, and nearly 2 in 3 families with an incarcerated member report being unable to meet their family’s basic needs such as food and housing.¹² These families also are far more likely to incur debt related to interaction with the criminal legal system, including lawyer and other legal fees and court costs. Or it can come from families needing to turn to payday loans and other forms of high-cost credit to meet monthly bills and to cover the often usurious costs charged by prison and jail commissaries for hygiene products and other basic needs of loved ones behind bars.¹³ Keeping in contact with an incarcerated loved one also strips wealth from families, with the high cost of visitation and phone calls alone pushing more than 1 in 3 families affected by incarceration into debt.¹⁴

Even after a loved one returns home, families typically continue to face steep economic challenges that make it difficult to make ends meet, let alone build savings. These families often serve as the primary resource and source of support for their returning family members, stepping into the breach when their loved ones face discrimination and structural barriers to housing, employment, and even public assistance upon reentry. In addition, families with an incarcerated member may face higher health care costs due to the mental and physical stresses that they experience. For these and myriad other reasons, mass incarceration actively strips wealth not just from the individuals swept up into the criminal legal system but also from entire families and communities. These families and communities are disproportionately people of color, LGBT people, and people with disabilities, due to these groups’ overrepresentation in America’s prisons and jails.¹⁵ These factors can make it substantially harder for already economically marginalized households to achieve financial stability, let alone build wealth.

Substantial research literature suggests that the inverse relationship also holds: Low household wealth increases the chance of incarceration. Households with less wealth are also more likely to live in neighborhoods with higher rates of police presence and/or surveillance and are thus more likely than wealthier households to face frequent interactions with the criminal legal system. Additionally, limited savings diminish a family’s ability to afford legal representation and bail money in the event a family member is arrested or incarcerated. The inability to afford bail can substantially lengthen the amount of time a loved one is behind bars awaiting trial. Once a household is affected by incarceration, the associated negative economic effects can further diminish opportunities for wealth building. This sets up a vicious cycle of massive financial insecurity.¹⁶

For Black and Hispanic families, this vicious cycle is particularly pronounced. These families are more likely to face arrest, conviction, and incarceration, even for the same crimes as white people—a consequence of widespread systematic racial biases in the criminal justice system. At the same time, Black and Hispanic families have much less wealth than white families due to centuries of violent oppression and widespread institutionalized discrimination in labor, housing, mortgage, health care, and financial markets, among many other realms. It should thus come as little surprise that mass incarceration not only corresponds with but also has become a major driver of racial economic disparities in the United States—and specifically, America’s persistent and unacceptable racial wealth gap.

The endnotes and Appendix to this report discuss the data and methodology in greater detail. But in short: This report uses data from two Federal Reserve bank surveys—the Survey of Household Economics (SHED) and Decisionmaking and the Survey of Consumer Finances—to provide summary data and some regression-based statistics to explore the nexus of criminal justice interaction and household wealth, with special consideration for how this link breaks down by race and ethnicity. The SHED delineates race and ethnicity using the mutually exclusive demographic labels “white, non-Hispanic”; “Black, non-Hispanic”; and “Hispanic.” These labels apply to all data with a racial and ethnic breakdown but are shortened in the accompanying text to “Black,” “white,” and “Hispanic” for ease of reading.

In 2019, the average wealth for households with an incarcerated family member—not counting the imputed wealth of defined benefit (DB) pensions—was $227,582, compared with $925,006 for households without a criminal justice system interaction. The respective median wealth levels were $7,350 and $177,800. (see Table 1)¹⁷ That is, households with an incarcerated family member had a little less than one-fourth the average wealth and only about 4 percent of the median wealth of households without an incarcerated family member.

Table 1

The wealth gap associated with criminal legal interactions largely follows from the fact that affected households have much lower asset values than but debt levels similar to those of households with an incarcerated family member. (see Table 1) For example, households with an incarcerated family member had $54,800 in median assets, or 19.4 percent of the $282,700 in median assets of households without an incarcerated family member.¹⁸ In contrast, households with an incarcerated family member had a median debt level of $25,000, while households without an incarcerated family member had a similar median debt level of $25,900. The lack of wealth for households with an incarcerated family member is therefore likely associated with greater obstacles to saving and investing and a similar need, even amid lower earnings, to borrow money.

The wealth differences reported in Table 1 reflect an underlying, systematic negative relationship between wealth and incarceration.¹⁹ The results show that a 1 percent increase in the average chance of having a family member incarcerated reduces household wealth by 1.82 percent. For households with DB pensions, a 1 percent increase in the chance of incarceration correlates with a 1.63 percent reduction in wealth. The bottom line is that wealth without DB pensions is 46.6 percent lower for households with an incarcerated family member than for households without one, and wealth with DB pensions is 41.6 percent lower.

People of color, people with less education, and single individuals bear higher costs of interactions with the criminal justice system

The chances of interacting with the criminal legal system are higher among people of color, single women and men, and individuals with less education. People of color are more likely to have an immediate family member incarcerated, as shown in Table 2, and they face more obstacles to building wealth when a family member has been incarcerated than is the case for white households, as summarized later in this report. Table 2 shows that Black households make up 19.5 percent of all households with an incarcerated family member but only 9.7 percent of households without an incarcerated family member. This is the largest demographic disparity; Hispanic households make up another 19.5 percent of households with an incarcerated family member but account for 15.4 percent of households without an incarcerated family member. Moreover, households with less than a college degree are overrepresented among households with an incarcerated family member, as are single women and men. Criminal legal system interactions are higher among these groups, which often face greater obstacles to building wealth because of lower incomes; greater income instability; and higher costs, including for health care, education, and family caregiving.²⁰

Table 2

Figure 1 displays the same data slightly differently, showing for each population the chance of having or having had an immediate family member in jail or prison in 2019. More than 1 in 5 U.S. households—21.9 percent—indicated that they had an incarcerated family member. For African American households, this share was 36 percent, while it was only 19 percent for white households. Among single women in 2019, the share was 26.9 percent, while it was 19.8 percent for married couples. The shares for households with and without a college degree were 13.3 percent and 35.1 percent, respectively. These data again highlight that interactions with the criminal legal system are more likely to happen among households that have less wealth, lower incomes, and greater income uncertainty, as well as those that face higher costs.

Figure 1

Incarceration poses a financial risk to households, but that risk is higher and costlier for some groups than it is for others. Black and Hispanic households are more likely to have an incarcerated family member (see Figure 1) and, if they do, they are also more likely to end up with legal debt—in 24.8 percent of cases—than was the case for white households, who had an 18.7 percent chance of ending up with legal debt. (see Figure 2) Households with less education and single women households were also more likely to have legal debt if they had an incarcerated family member than was the case for households with a college degree and those made up of married couples. The dual costs of mass incarceration—the widespread chance of criminal legal system interactions and adverse economic consequences—tend to fall more heavily on people of color, single women and men, and less educated people.

Figure 2

These differences in the risk and cost of incarceration are especially pronounced by race and ethnicity. The rest of this report looks more closely at the overall differences in wealth and financial security resulting from incarceration, as well as at the related gaps by race and ethnicity.

Interactions with the criminal justice system are linked to lower assets

Having an incarcerated family member has two likely consequences for wealth: 1) It makes it harder for households to save, reflected in lower asset ownership; and 2) it forces many households to take on more costly debt, such as credit card or legal debt, to pay their bills. Indeed, Table 3 shows that households with an incarcerated family member are less likely to own key assets than are households without an incarcerated family member. For example, 61 percent of households with an incarcerated family member had assets of less than $50,000, while only 43.6 percent of households without an incarcerated family member had assets of less than $50,000. Similar gaps exist for homeownership, retirement benefits, and emergency savings. For instance, only 49 percent of households with an incarcerated family member reported being able to cover a $400 emergency with cash or a cash equivalent, while 67.2 percent of households without an incarcerated family member reported being able to do so. Incarceration goes hand in hand with much less asset ownership.

Table 3

In addition to incarceration, the data breakdown by race and ethnicity highlights persistent inequalities. As shown in Table 3, Black and Hispanic households in the United States have substantially fewer assets than white households. For example, 60.4 percent of Black households without an incarcerated family member have less than $50,000 in assets, compared with 38.4 percent of white households. At the same time, 70.8 percent of Black households with an incarcerated family member have less than $50,000 in total assets, while the same is true for 55.3 percent of white households. Even after accounting for criminal legal system interactions, Black and Hispanic households have fewer assets than white households. In fact, white households with an incarcerated family member are more likely to have more than $50,000 in assets than are either their Black or Hispanic counterparts. In short, asset differences by race and ethnicity dominate asset gaps due to interactions with the criminal legal system.²¹

Households with an incarcerated family member have more costly debt

When it comes to household debt, the amount and composition varies by interaction with the criminal legal system. Households without an incarcerated family member are more likely to have a mortgage (see Table 4), reflecting the greater homeownership rates seen in Table 3. At the same time, households with an incarcerated family member are more likely to have unpaid credit card balances—57 percent compared with 41.6 percent—along with more student loan and legal debt. (see Table 4) These forms of credit are often costlier and riskier than mortgage debt. Credit card and legal debt also are not directly associated with investments in a household’s future, unlike mortgages and student loans.

Table 4

Black and Hispanic households in the United States are also more likely to owe more costly debt than white households, regardless of their interactions with the criminal legal system. (see Table 4) For example, 68.3 percent of Black households and 60.4 percent of Hispanic households with an incarcerated family member had unpaid credit card balances, compared with 53.4 percent of white households. Moreover, even Black and Hispanic households without an incarcerated family member were more likely to have unpaid credit card balances—59.5 percent and 54.5 percent, respectively—than were white households with an incarcerated family member. Black and Latino households face systematic credit market obstacles, especially high mortgage application denial rates, that result in them being burdened with more costly debt.²²

Structural racism is pervasive and results in less wealth because it erects barriers to asset ownership and pushes people of color into more costly debt—even when Black and Hispanic households do not have interactions with the criminal legal system. Incarceration is only one of many barriers, albeit a substantial one, that Black and Hispanic households face when building wealth.

Households with an incarcerated family member face greater financial insecurity

Households maintain wealth in part as a buffer against unexpected shocks such as layoffs, medical emergencies, or family members’ interaction with the criminal legal system. Table 5 summarizes a number of financial security measures by interactions with the criminal legal system, asset levels, and race and ethnicity. Households with an incarcerated family member tend to be less financially secure. For instance, in 2019, 22.6 percent of households with an incarcerated family member could not pay all their bills. (see Table 5) In comparison, just 14.5 percent of households without an incarcerated family member encountered such difficulties. In part, this difference results from having fewer assets, since a smaller financial cushion increases the chances of financial insecurity. Yet even among households with less than $50,000 in assets, those with an incarcerated family member faced systematically greater financial insecurity than those without one.

Table 5

Black and Hispanic households always experience greater financial insecurity than white households, whether they have an incarcerated family member or not. (see Table 5) Reflecting the gap in assets and debt discussed in the previous two sections, Black and Hispanic households without an incarcerated family member are much more likely—28.8 percent and 22.8 percent, respectively—to have trouble paying their bills than are their white counterparts, at 19.1 percent. The data again reflect the systematic biases that Black and Hispanic households face in building financial security. The racial biases in interactions with the criminal legal system only serve to make a bad situation worse.

Economic mobility is lower among households with incarcerated family members

Households build wealth to gain opportunities for upward economic mobility. The data already show greater homeownership and higher educational attainment among those households that do not have a family member incarcerated. (see Table 2 and Table 3) Additional data capture other aspects of economic mobility, such as whether a household is doing better or worse than their parents. (see Table 6) These and other data show that households with an incarcerated family member experience less upward economic mobility than do households with no interactions with the criminal legal system. As shown in Table 6, almost one-fifth—24.7 percent—of households with an incarcerated family member reported that they did worse than their parents, while only 19.1 percent of households with no incarcerated family members said the same. Households with criminal legal system interactions were also more like to have a loan application denied, to forgo college to support family, and to rent rather than own a home because they could not afford a mortgage or down payment. The incarceration of a family member goes along with less upward economic mobility.

The data also show noticeable differences by race and ethnicity. For example, while Black and Hispanic households start off with fewer opportunities for upward economic mobility than is the case for white households, interaction with the criminal legal system further reduces those chances. In general, Black and Hispanic households are less likely to be homeowners. (see Table 3) They are also substantially more likely to be denied a loan application than are white households, regardless of their interactions with the criminal legal system. Specifically, 31.5 percent of Black households with an incarcerated family member said that they had been denied a loan application or did not apply out of fear of being denied. (see Table 6) For Hispanic households, this share was 28.3 percent, while it was 18.1 percent for white households. As with other indicators, the share of white households with a denied loan application and an incarcerated family member was lower or similar to the respective share for Black and Hispanic households without an incarcerated family member. The same is true for the shares of households where the respondent did not attend or complete college due to the need to earn money or support their family. Slightly more than 1 in 4, or 26.2 percent, of white households with an incarcerated family member were in this situation. This was a smaller share than was the case for Black and Hispanic households without an incarcerated family member—31 percent and 35.8 percent, respectively. Finally, upward economic mobility is especially low among people of color with an incarcerated family member. For example, 26.1 percent of Hispanic households with an incarcerated family member reported that they were worse off than their parents, compared with 21.2 percent of their Hispanic counterparts without an incarcerated family member and 12.5 percent of white households without an incarcerated family member. Overall, interactions with the criminal justice system are linked with reduced upward economic mobility.

Households with an incarcerated family member face greater obstacles to building wealth

Many households face a number of obstacles to building wealth, but those barriers are more significant for households with a currently or previously incarcerated family member. Take income from earnings, or the lack thereof, as an example. Overall, 16.9 percent of households without an incarcerated family member have assets greater than $50,000, while the same is true for 12.7 percent of households with an incarcerated family member. (see Figure 3) At the same time, 57.1 percent of households without an incarcerated family member and incomes above $30,000 had assets greater than $50,000, while only 41.3 percent of households with incomes above $30,000 with an incarcerated family member did. Higher incomes go along with more assets, but interaction with the criminal legal system is linked with fewer assets in all income categories. (see Figure 3) The data similarly show that incarceration goes along with fewer assets, even when people have a job, health insurance, and stable incomes; live in safe and secure neighborhoods and communities; and have responsive landlords who quickly fix problems and thus keep costs lower than they otherwise would be. That is, interactions with the criminal legal system are associated with higher costs for households, likely reducing their ability to save money and requiring them to go deeper into debt to meet necessary costs.

Figure 3

Higher costs associated with incarceration—even when paired with factors otherwise beneficial to wealth building such as higher incomes and employment—are only one of the barriers to building wealth associated with interactions with the criminal legal system. Households with an incarcerated family member are more likely to have lower incomes, to be without a job, to rely on public assistance, to lack health insurance, to have unstable incomes, and to live in unsafe and undesirable neighborhoods.²³ This is especially true for Black and Hispanic households, who already face greater economic obstacles that are made worse by having an incarcerated family member.²⁴ Households with an incarcerated family member have less wealth, which goes along with both a higher chance of unfavorable economic conditions and the higher costs associated with those conditions.

Failed criminal justice policies are a major driver of America’s racial wealth gap

This report highlights three important facts related to incarceration and the racial wealth gap: 1) Black and Hispanic households tend to have less wealth, fewer assets, and more costly debt than white households; 2) Black and Hispanic households have more criminal legal system interactions and are more likely to have an incarcerated family member with legal debts than white households; and 3) Black and Hispanic households face greater costs in building wealth when they have an incarcerated family member than do white households. In sum, communities of color face higher risks of costly criminal legal system interactions and higher relative costs of such interactions, which further exacerbates and compounds America’s significant racial wealth gap.

In general, families with an incarcerated family member have a lot less wealth than families without an incarcerated family member, and those differences are even more pronounced for Black and Hispanic families. As shown in Table 7, in 2019, the median wealth of Black families with an incarcerated family member was $3,970, compared with $50,400 for white families with an incarcerated family member.

Table 7

Moreover, the data suggest that incarceration likely contributes to worsening the racial wealth gap. For one, the drop in median wealth associated with incarceration is typically much larger for Black and Hispanic households than for white households. Black households with an incarcerated family member had about 2.4 percent the wealth of Black households without an incarcerated family member—$1,101 vs. $44,140. (see Table 7) The shares for Hispanic households and white households are 4.4 percent and 6.3 percent, respectively. Moreover, the differences in average wealth associated with incarceration are relatively similar by race and ethnicity, but Black and Hispanic households are much more likely to have an incarcerated family member. On average, incarceration worsens the racial wealth gap, since the associated average wealth drop is much more widespread in communities of color than among white households.

The Appendix presents more detailed statistical analyses to capture the estimated dual correlation of interactions with the criminal legal system: higher likelihood of incarceration and higher costs. The results show that differences in interactions with the criminal legal system capture 27 percent of the explained Black-white wealth gap. (see Figure A-1 in the Appendix) These interactions make up 40 percent of the statistically explained difference in the Hispanic-white wealth gap.

It is important to keep in mind that the estimates in the Appendix only show the systematic correlation between interactions with the criminal legal system and the racial wealth gap. They do not establish causality. Moreover, the underlying calculations show what would happen if Black or Hispanic households had the same characteristics as white households. The results thus account for the differences in criminal legal interactions by race and ethnicity but not the overall correlation between incarceration and wealth shown in Table 1 and the subsequent discussion.

Differences in interactions with the criminal legal system worsen an already substantial racial wealth gap. Black and Hispanic households are more likely to experience these interactions—and to have them be more costly than is the case for white households. In addition, Black and Hispanic households have less wealth to begin with because of things such as lower inheritances. The data summary in this section reflects the vicious cycle between wealth and criminal legal system interactions as one mechanism by which the racial wealth gap persists.

A large and growing body of research literature has already confirmed that considerable negative economic consequences flow from incarceration or other interaction with the criminal legal system. These consequences include dramatically reduced employment and earnings opportunities and barriers to accessing or affording basic necessities such as housing, education, and health care.²⁵ This report builds upon existing literature by adding a wealth dimension. These new data underscore the economic effects that criminal legal system involvement have on families, including overt wealth stripping, the accumulation of unaffordable debt, and significant diminishment of wealth-building potential—all of which are major drivers of poverty and inequality in the United States. These negative consequences are most profound for families of color, who have borne the brunt of America’s failed experiment with mass incarceration and overcriminalization that has, in turn, further widened the country’s racial wealth gap.  

A comprehensive set of policy recommendations to close America’s racial wealth gap and reform the nation’s broken criminal legal system is outside the scope of this report.²⁶ However, the stark findings highlighted by this study show how deeply intertwined this nation’s criminal justice and economic policies have become. Consequently, government leaders must prioritize policy solutions that intentionally target this nexus, in tandem with larger reform of the criminal legal system that ends mass incarceration and dramatically shrinks the footprint of America’s prisons and jails. Fortunately, an array of existing, proven policy reforms to remove barriers to economic security and opportunity for individuals and families affected by the criminal legal system enjoys broad voter support across the political spectrum. While far from a comprehensive list, the sections below highlight some of the leading policy reforms at this nexus that have gained bipartisan momentum at the federal and state levels in recent years.

Increase access to criminal record-clearing and make it automatic (‘clean slate’)

Nearly all states now allow people to petition to have certain records sealed or expunged. However, several recent studies indicate that an estimated 90 percent of eligible individuals are effectively barred from petition-based record-clearing systems due to cost, complexity, and other barriers.²⁷ On the other hand, a large body of research now demonstrates the significant economic benefits of record-clearing for individuals, families, and communities. In addition, there is a strong public safety case for expanding access to record-clearing, given that the risk of recidivism declines sharply when an individual secures employment. ​​In a recent bipartisan trend, a growing array of states have enacted “clean slate” laws that make record-clearing automatic for qualifying individuals; expand access to the benefits of record-clearing; and reduce the barriers that make it hard for people to move on with their lives after they’ve served their time and remained crime free.²⁸ With a cleared or sealed record, people get fair shots at employment, housing, education, and more. Without such opportunities, people have few chances to support themselves and their families, let alone build wealth. Measures such as clean slate are essential for justice-impacted individuals of all backgrounds and are vital to addressing the racial wealth gap.

Provide federal funding to help states implement automatic record-clearing

On the heels of states that have already enacted clean slate policies, more than a dozen states are currently advancing clean slate legislation to give more of their residents a fair shot at a second chance. Federal policymakers should support these efforts by providing funds to help states shoulder the upfront costs of clean slate implementation. While clean slate laws are expected to save states money in the long term by streamlining costly and burdensome workloads, the initial costs of implementing a new automated process may be a barrier to take-up for states facing budget gaps in the wake of the COVID-19 pandemic. The bipartisan Fresh Start Act championed by Reps. David Trone (D-MD), Guy Reschenthaler (R-PA), Sens. Chris Van Hollen (D-MD) and Joni Ernst (R-IA) would provide $50 million in federal funds for this purpose. Under the proposed legislation, states can apply for up to $5 million to cover the cost of planning, researching, and implementing a record clearance system.²⁹ States that condition access to record-clearing on ability to pay fines and fees should not be eligible for federal grants. An interim step would be to increase federal funding through the National Criminal History Improvement Program (NCHIP)—a source of federal funding that congressional appropriators have clarified can be used to support planning for as well as implementation of states’ automatic record clearance programs.³⁰

Create a federal record-clearing system

Currently, federal record-clearing is unavailable—even for arrests on federal charges that do not lead to a conviction. However, inspired by the groundswell of support for expanding and automating record-clearing in the states—such as Pennsylvania, Utah, Michigan, Connecticut, Oklahoma, Delaware, and many more—both the House and the Senate have introduced bipartisan legislation to create the first federal record-clearing remedy and establish a platform for automatic federal sealing. A number of federal lawmakers have publicly stated their intention to build on the success of clean slate in the states. The Clean Slate Act, championed by Reps. Lisa Blunt Rochester (D-DE) and Guy Reschenthaler (R-PA) in the House of Representatives and Sens. Bob Casey (D-PA) and Joni Ernst (R-IA) in the Senate, would create a process to enable people with federal arrests and a broad set of low-level conviction types to petition to have their records sealed.³¹ The proposed legislation would also establish a system for the automatic sealing of qualifying federal records, starting with nonconvictions, nonviolent marijuana crimes, and federal drug possession records. Although it is essential to enable a much broader swath of people with federal records to access the benefits of record-clearing, this legislation would create a starting point and infrastructure on which to build. Meanwhile, President Joe Biden’s recent announcement of pardons for people with marijuana convictions³² offers another reason for the swift passage of the Clean Slate Act, as pardons still will not result in record clearance for the vast majority of individuals who receive them.

Reform punitive criminal legal debts

For system-impacted individuals, the increased use of fines and fees by courts across the country can be a significant barrier to financial stability and growth. Judges impose fines as punitive or deterrent measures in a sentence, while fees are usually automatically levied to gain revenue for the courts and other government operations. Monetary sanctions are often imposed without concern for an individual’s ability to pay and can easily result in hundreds or thousands of dollars³³ of court debts that a person may never be able to pay. Failure to pay can result in an individual being sent back to jail, which ultimately devastates rehabilitation efforts and pushes system-impacted individuals into a cycle of poverty. This is especially true for people of color and other marginalized groups.

Fines and fees have also become a major barrier to record-clearing. Courts often deny requests to pardon a record if the individual has outstanding monetary obligations. This leaves system-impacted individuals to deal with the collateral consequences of their records, which makes it extremely difficult to pursue financial stability and wealth-building opportunities for themselves and their families.

Court-imposed monetary sanctions should not inhibit the ability to achieve financial stability or generate wealth. State and local governments must establish measures to either eliminate or reduce the use of fines as punitive measures, as well as abolish court fees. There is an estimated $27 billion in national court debt, trapping millions of people in a cycle of poverty and justice system involvement that has proved to be harmful.³⁴ Additionally, courts should not deny petitions to expunge eligible records because of outstanding monetary obligations. It is argued that fines and fees are meant to ensure public safety and increase court revenue,³⁵ but there is little evidence to support these claims. In fact, many jurisdictions use so much of their resources from fines and fees to collect these massive debts from the public that little remains to put toward actual public safety measures.

Remove barriers to employment for people with records through fair-chance hiring and licensing

When individuals have limited access to well-paying job opportunities, obtaining financial security can be especially difficult. Unfortunately, many justice-impacted people struggle to enter a job market in which 9 in 10 employers use background checks to make hiring determinations.³⁶ As one study from the University of Michigan Law School has shown, job applicants with a criminal conviction are 60 percent less likely to receive a callback than individuals with no conviction history.³⁷ When the odds are so heavily stacked against returning citizens, it is nearly impossible for them to reach any state of economic stability and build financial wealth.

Fair-chance hiring measures can be utilized to ensure that job applicants are given a reasonable opportunity to prove that they are well-qualified for a position, regardless of their prior involvement with the legal system. In addition, fair-chance measures prohibit record-based discriminatory practices and ensure criminal records do not factor into hiring decisions. Ban-the-box policies, in particular, prohibit employers from inquiring about past criminal history before they make an offer of employment. These measures can help establish standards that will ensure fair hiring practices free of the stigma of a record, bringing more job opportunities to justice-impacted individuals.

Unfortunately, the hiring process is not the only area where a job seeker’s record can be detrimental to their career pursuits. Today, 1 in 4 jobs in the U.S. economy require an occupational license,³⁸ but many state laws and licensing regulations unfairly allow licensors to use a person’s record as reason to deny them the licenses or certifications needed for their intended profession. As a result, a well-paying job can be out of reach for people returning to society.

Fair-chance licensing measures help ensure access to good jobs for all workers by prohibiting laws and regulations that allow the denial of occupational licenses solely on the basis of a person’s criminal record. As the number of professions requiring licensure or certification grows, so does the need to ensure people with a record are given equitable access to these jobs and the opportunity to meet the requirements for their intended profession.

Remove barriers to housing and home ownership for families affected by a record

For individuals with a criminal record, as well as for families with a member who has a record, finding safe, stable, and decent housing—let alone attaining the goal of homeownership—can be particularly difficult. An estimated 4 in 5 landlords use background checks in the housing application process, and housing loans and other financial support can be hard to come by when many lenders are free to deny applicants based on their record alone.³⁹ For individuals with a record, widespread housing discrimination translates into higher rates of housing insecurity and unstable living situations. In fact, according to the Prison Policy Initiative, formerly incarcerated individuals are almost 10 times more likely to experience homelessness than the general public.⁴⁰ At a time when affordable housing is increasingly difficult to come by, these barriers only exacerbate existing housing challenges for people and families of color and make it very difficult for families to build financial wealth and stability.

Fair-chance housing measures can help prevent discriminatory housing practices. For example, if passed, New York City’s Fair Chance for Housing Act would prohibit housing providers from conducting criminal background checks in any part of the housing application process or asking about an applicant’s record. The city estimates that this would help the nearly 750,000 residents who have a conviction record—or 11 percent of its adult population.⁴¹ Similar measures have also been introduced at the federal level. With the COVID-19 pandemic exacerbating the nation’s housing crisis, there is a growing need for accessible and affordable housing investment. In 2021, Congress recognized this need when it passed the American Rescue Plan Act, which included a number of provisions to help individuals struggling to find or maintain stable housing. These provisions include funding for emergency rental assistance, homeowner mortgage and utility payment assistance, and housing and support services for people facing housing insecurity and homelessness. Supports such as these can help individuals with a record establish a foundation for themselves and their families to ultimately achieve financial stability.

Remove barriers to education and entrepreneurship for people with records

Efforts are needed to remove barriers for individuals with a record who wish to pursue higher education. Today, roughly 66 percent of colleges and universities use background checks and record history in determining admissions, financial support, and student housing decisions.⁴² State financial assistance can also be challenging for justice-impacted students to obtain, as many financial aid programs are unavailable for individuals with certain conviction records. And although some efforts to remove barriers to higher education recently have been made at the federal level, including the restoration of Pell Grant eligibility for incarcerated students and the removal of questions around drug conviction history on the Free Application for Federal Student Aid (FAFSA), more can be done.⁴³

The Department of Education can lead national efforts to remove barriers to education by recommitting to the Obama administration’s Fair Chance Higher Education Pledge. The pledge encouraged colleges and universities to adopt institutional practices that “eliminat[e] unnecessary barriers for those with a criminal record and creat[e] a pathway for a second chance.”⁴⁴ Such practices could include the implementation of fair-chance admissions practices, support for systems to ensure successful completion of the college program, and the provision of supportive resources to help students obtain future career opportunities. Additional actions should be taken at the federal level to fund and expand education grant programs that aim to provide support for justice-impacted individuals. One such grant program is Improving Reentry Education and Employment Outcomes, which aims to support state and local actions “focused on strengthening education and employment outcomes” for returning citizens.⁴⁵

Likewise, entrepreneurship support is vital for people with a record. Many justice-impacted people pursue entrepreneurship opportunities because of discriminatory hiring practices. However, establishing and running a business can be especially challenging when the collateral consequences of a record often prevent people from accessing capital to start or maintain a business. These challenges were clearly demonstrated during the pandemic when, in 2020, Congress passed the Coronavirus Aid, Relief, and Economic Security (CARES) Act, which was meant to provide support to small-business owners to cover operational and payroll expenses through the Paycheck Protection Program (PPP). Unfortunately, the Small Business Administration (SBA) adopted exclusionary provisions that denied PPP loans to the nearly 4 percent of business owners with a criminal record, leaving many small businesses in jeopardy of failing.⁴⁶ Many of the PPP loan restrictions against business owners with a record were eventually lifted, but the SBA continues to place broad restrictions related to an owner’s criminal history on loans it administers. The SBA should reevaluate these restrictions, which have racially disparate impacts, and advance equitable policies that assist business owners with a record and reduce the barriers they often face when seeking resources.

Remove barriers to savings and ownership in government assistance programs

Recently, state and federal policymakers have increasingly taken steps to reform outdated asset limits in income assistance programs. A growing body of research finds that such policies are counterproductive and stand in the way of wealth building and ownership, and perversely put economic stability even further out of reach for struggling individuals and families. Supplemental Security Income (SSI), which provides critical if meager income assistance to roughly 8 million disabled people and older adults, is an example of a lifeline program with archaic asset limits that have not budged in nearly four decades, even for inflation. Individuals are prohibited from saving more than $2,000 as a condition of receiving SSI benefits and the accompanying health insurance that SSI eligibility generally confers through Medicaid; couples are limited to $3,000. The recent bipartisan SSI Savings Penalty Elimination Act—championed by Sens. Sherrod Brown (D-OH), Rob Portman (R-OH), Ron Wyden (D-OR), Bill Cassidy (R-LA), Bob Casey (D-PA), Tim Scott (R-SC), and Susan Collins (R-ME), and backed by stakeholders as diverse as CAP, the American Enterprise Institute, JPMorgan Chase, AARP, and the Niskanen Center—would adjust SSI’s asset limits for inflation for the first time in nearly 40 years. The proposed asset limits would increase to $10,000 for individuals and $20,000 for couples—enabling the program’s beneficiaries, disproportionately disabled and older people of color, to finally save and plan for the future.⁴⁷

Although it is well-known that America’s broken criminal legal system is a major driver of poverty and racial inequality, this report further exposes the economic devastation caused by mass incarceration and overcriminalization. Families of all backgrounds and especially communities of color are particularly harmed by these failed criminal justice policies, which have taken a profound economic toll on tens of millions of Americans and further widen the country’s racial wealth gap. Local, state, and federal leaders, therefore, must redouble efforts to advance policies that remove barriers to economic security and upward mobility and ensure that the U.S. economy no longer leaves system-impacted individuals and families behind.

Prior literature on the link between criminal legal system interactions and household wealth

In recent years, a handful of studies have considered the correlation between incarceration or criminal records and household wealth.⁴⁸ A few findings stand out from these analyses. First, people with a criminal record or households with an incarcerated or formerly incarcerated family member tend to have fewer assets. For instance, researchers Bryan Sykes and Michelle Maroto find that incarceration is correlated with 64.3 percent fewer assets and 85.1 percent less debt.⁴⁹ Because of this, the correlation with wealth remains unclear. Moreover, Maroto concludes that homeownership rates are 28 percentage points lower after somebody has been incarcerated. She also finds that an ex-offender’s net worth is, on average, $42,000 lower after incarceration.⁵⁰ But the overall household wealth in the data set used in this study is only a fraction of that reported for the respective age group in another, nationally representative wealth-specific data set.⁵¹ Thus, the resulting wealth gap between households with a family member with an incarceration history and households with no incarceration history could be heavily understated. The analysis in report uses a nationally representative data set specifically designed to capture all U.S. households’ wealth, discussed below, and hence can provide a complementary estimate for the correlation between incarceration and wealth that may better capture a complete value of households’ wealth.

Similarly, researchers Daniel Schneider and Kristin Turney find that higher average incarceration rates at the state level lead to lower homeownership rates among African Americans.⁵² In particular, they conclude that “the Black homeownership rate is 62% of the White rate when the incarceration rate is 1/1,000 (the 10th percentile of the distribution of incarceration rates), but falls to 51% of the White rate when the incarceration rate is 6/1,000 (the 90th percentile of the distribution of incarceration rates).”⁵³ This particular paper uses a methodology that allows for some causal inference rather than just a correlation, but the underlying data only allow for an analysis of homeownership rates. The CAP report focuses on the correlation between incarceration and additional assets as well as the value of assets and total wealth. The bottom line is that a number of high-quality studies have looked at the link between incarceration and household assets and to a lesser degree at the link between incarceration and debt and wealth. The CAP report is the first, to the authors’ knowledge, that uses nationally representative data on wealth, assets, debt, and financial security in connection with incarceration.

Second, the limited literature finds that incarceration differences by race and ethnicity play an important role in the racial wealth gap. Among households with a criminal record or with an incarcerated family member, households of color tend to have substantially less wealth than white households. Moreover, researchers such as Khaing Zaw, Darrick Hamilton, and William Darity Jr. have used panel data such as the National Longitudinal Survey of Youth (NLSY) to explore the possible reverse links between wealth and incarceration.⁵⁴

Third, all data used for empirical analyses of the link between incarceration and wealth have some shortcomings. The NLSY contains information on people’s incarceration in their youth and their asset ownership and net worth at some points later in life. The NLSY is a critical data set for understanding the link between incarceration and wealth,⁵⁵ but its information on assets and wealth is limited and does not match data from other sources. Moreover, the same data set typically does not contain information on incarceration and wealth. Consequently, researchers combine the information from this data set with that from other data sources. For instance, Sykes and Maroto adjust the share of the population that is institutionalized in the Survey of Income and Program Participation (SIPP) using other information from separate data sources on incarceration rates by education, age, and race and ethnicity.⁵⁶ Schneider and Turney created a state-level panel data set with average incarceration rates and homeownership rates in each state.⁵⁷ Such data set matching methods are common but limit the analyses that are possible with the data. Schneider and Turney, for example, only focus on homeownership, not the value of assets and wealth.

In contrast, the CAP report uses information from the Federal Reserve’s Survey of Household Economics and Decisionmaking for 2019, which includes information on incarceration as well as wealth and financial security. The report also uses another unique matching procedure to combine information from the SHED with another Federal Reserve data set, the Survey of Consumer Finances (SCF). The SCF is the most widely used nationally representative data set on household wealth. It contains detailed information on wealth, assets, and debt, as discussed further below.

This report’s analysis of data from the SHED and SCF thus supplements previous analyses. It directly links incarceration likelihoods with household wealth, assets, and debt, as well as a variety of measures of financial security and economic mobility from the same data source. It shows some financial security measures for households during the pandemic, broken down by 2019 incarceration status. It also provides estimates for the differences in the total value of wealth, assets, and debt by incarceration status overall and in addition to race and ethnicity.

Data sources, variables, and methodology

This report uses data from two separate, publicly available, and nationally representative data sets and relies heavily on the Federal Reserve’s SHED for 2019. The SHED released in 2019 is nationally representative. It has a sample size of 12,238 households with interviews conducted online in October 2019.⁵⁸ This report combines the information from the SHED with data from the SCF.

The Survey of Household Economics and Decisionmaking

The Federal Reserve designed the SHED to capture a wide array of aspects that determine households’ economic and financial security.

Most importantly for this report, the 2019 iteration of the SHED included two questions related to households’ interactions with the criminal justice system. One question asked, “Have any members of your immediate family ever been in prison or jail for one night or longer?” Throughout this report, the authors refer to families that answered “yes” to this question as having been incarcerated to facilitate the discussion. The survey then asked a related question: “Do you or someone in your immediate family currently have any unpaid legal expenses, fines, fees, or court costs?” This report defines families that answer “yes” as having any legal debt.

The SHED also includes information on families’ assets and debt and thus implicitly on their wealth, which this report uses to show the correlation between incarceration, assets, and debt. Wealth is the difference between what families own, such as their houses and retirement accounts, and what they owe, such as mortgages, student loans, and credit cards. “Assets” is the broad category of things that families own, while debt is the respective overarching category of everything that they owe. The SHED includes information on total assets, broken down into six categories ranging from “less than $50,000” to “$1,000,000 or more.” This report shows the share of families that indicated they had less than $50,000 in assets. In addition, the summary of the SHED data shows, as indicators of longer-term financial stability, the shares of households that own their own house, that have a 401(k) retirement account through their employers, that have 401(k) balances of less than $25,000, and that have a defined benefit pension. Moreover, the report shows the shares of households that can cover $400 in an emergency with cash or a cash equivalent; that have three months’ worth of income in emergency savings; and that have a checking, savings, or money market account; all three of these measures indicate short-term financial security. To capture different forms of debt, this report shows the share of household that have a mortgage, that have monthly mortgage payments greater than $1,500, that owe student loans, and that have any legal debt. The SHED data analyzed allow for an approximation of the correlation between incarceration and wealth. The report summarizes information on assets and debt mainly by incarceration but also by race and ethnicity as a secondary, yet no less important, focus.

Households own wealth as a financial buffer for short-term financial emergencies and as capital with which to make longer-term investments. A number of variables from the SHED capture short-term financial security. These include the shares of households that cannot pay all their bills, that find it difficult to get by or are just getting by, that skip health care, and that have medical debt following unexpectedly large medical bills. Measures of longer-term financial investments include being somewhat or much worse off than 12 months ago, being somewhat or much worse off than parents were, having had a loan application denied or not even applying for a loan out of fear of being denied, not completing college in favor of earning money or supporting a family, and renting due to an inability to get a mortgage or afford a down payment. These longer-term measures of economic opportunity and mobility are in addition to owning one’s own home and having a college degree. The report summarizes information on short-term and long-term finances by incarceration.

Households can face varying obstacles to saving for retirement, depending on whether a close family member has been or is in jail or prison. Broadly speaking, these obstacles are low incomes and high costs. High costs include health care costs, due to a lack of health insurance; income instability; and costs related to housing. The SHED provides information on household income, employment, employer-sponsored benefits such as health insurance, and reliance on public assistance as examples of how low incomes and high costs can impede building wealth. The data set also includes several variables that capture households’ income stability such as whether earnings or hours fluctuate and whether a household relies on gig work. The report also uses the SHED data to capture other external costs such as people’s dissatisfaction with neighborhood amenities and safety, housing quality, and landlords unresponsive to making repairs, as well as connections available for people with an opioid addiction, who may need help and support. This report thus summarizes data comprehensively to capture the barriers that households face to building wealth by incarceration status and race and ethnicity.

The SHED contains a wide range of demographic variables for each household. These include indicators for the race or ethnicity, age, educational attainment, marital status, gender, and physical health of the respondent. They also include information on household composition, specifically whether minor children or parents and grandparents live with the respondent. This report focuses specifically on race and ethnicity of the respondent, broken down into white, non-Hispanic; Black, non-Hispanic; and Hispanic. These three categories are mutually exclusive. The report does not provide detailed information on “other races and ethnicities” since that category includes a large and very diverse population, which makes the interpretation of the data even harder than is the case for other large racial and ethnic categories. The data also include a category for multiple races and ethnicities. The report does not contain detailed information on this population since sample sizes are often unreliably small. In all cases, the report summarizes data by incarceration and race and ethnicity.

The Survey of Consumer Finances

This report combines some information from the SHED with data from the SCF. The SCF is a triennial survey of households’ total wealth. It is nationally representative, and the weighted data for wealth, assets, and debt add to the national aggregates of these variables reported elsewhere. The Federal Reserve collected these data most recently in the second half of 2019, which covers approximately the same time period as the SHED data for 2019 released in 2020. The SCF includes a total of 5,783 households. To improve the usability of the SCF, the Federal Reserve employs a statistical replication technique to create five replicates of each respondent household with varying weights that are all nationally representative and add to national aggregate totals.⁵⁹

The SCF contains more detailed information on assets, debt, and wealth than does the SHED. In particular, it has information on the total value of all assets, debt, and wealth. It also contains sufficient information on people’s DB pensions, so that researchers can impute the implicit wealth that people own in these retirement benefits. DB pensions are comparatively large assets for households that have such benefits.⁶⁰ This report then summarizes total wealth with and without DB pensions, assets with and without DB pensions, and total debt by incarceration status and race and ethnicity.

Matching the SHED and SCF

It is possible to combine the information from both data sets to gain a clearer picture of wealth by incarceration status. The SCF contains detailed wealth information, but it does not include an indication of whether a close family member has been or is in jail or prison. The SHED includes this information, but it does not have information on a household’s total wealth. Yet the SHED and the SCF are two nationally representative data sets collected by the Federal Reserve in late 2019. Both data sets include key and comparable demographic information on respondents’ age, race and ethnicity, marital status, household income, and education level. As key data in both sets are highly comparable with each other, it is possible to match them. This allows for an estimation of the likelihood of a household member having been incarcerated for each household in the SCF and thus for a summary of wealth by incarceration status.

The process of matching the two data sets proceeds in three steps.⁶¹ First, it is possible to estimate a regression model of the correlates of incarceration such as age, race and ethnicity, education, gender, and marital status using the SHED. Inversely, observations in the SCF can be used to estimate correlates for wealth. The model specifically estimates the relative importance of age, education, and income level by race and ethnicity for determining the inverse hyperbolic sine of real household wealth. Wealth is highly unevenly distributed toward very wealthy households, which makes unadjusted wealth data inappropriate to use with standard regression models. Transforming wealth by applying the inverse hyperbolic sine corrects for the skewness of wealth as is defined for all values of wealth, including zeros and negative numbers. Second, the resulting parameter estimates from this regression model of wealth can be used to predict a continuous likelihood of incarceration for each household in the SCF and the value of wealth in the SHED.⁶² The prediction assumes a correlation between wealth and incarceration—specifically, a negative correlation of -0.175.⁶³ To show the robustness of the conclusions, this report also shows some key results based on an assumed correlation of -0.15 and -0.2. (see Table A-1) Third, observations in the SCF are matched to their closest neighbors in the SHED, based on incarceration likelihood; an assumed correlation between wealth and incarceration likelihood; and other criteria such as age, race and ethnicity, marital status, and educational attainment. The result is a predicted binary variable of incarceration in the SCF. CAP’s report then summarizes wealth, assets, and debt by incarceration based on the SCF.

Table A-1

Regression and decomposition analyses to estimate the contribution of incarceration to the racial wealth gap

This report estimates the relative importance of incarceration among correlates for household wealth. The available data cannot establish causality since it relies on cross-sectional data for one year. It can, however, show the magnitude of the correlation between incarceration and household wealth. The report first estimates a model of the correlates of household wealth—specifically, of the inverse hyperbolic sine of real wealth—that includes the predicted incarceration indicator in the SCF. The parameter estimate can thus be interpreted as an elasticity. That is, the parameter estimate for incarceration shows the relative percentage change associated with having or having had a family member in jail or prison.

The regression model uses a range of standard correlates for household wealth from the SCF. These include predicted incarceration status as well as age, marital status, education, gender, financial risk tolerance, income, firm size, health status, length of time with current employer, industry, occupation, and whether the household has received or expects to receive a large financial gift or inheritance.

Interpreting the regression results requires some transformation.⁶⁴ The dependent variable is the inverse hyperbolic sine of total wealth with or without DB pension wealth. The correlate of interest is an indicator of whether a household has had a family member incarcerated. With some minor transformation, the parameter estimate for this indicator shows the relevant elasticity. That is, it shows by how many percentages wealth changes in response to a 1 percent increase in the chance of having an incarcerated family member. This is the elasticity at the sample average. For the entire matched sample in the SCF, the average estimated chance of having an incarcerated family member varies from 25.5 percent to 29.4 percent, depending on the assumed correlation between incarceration risk and wealth. For the preferred correlation of -0.175, the average incarceration chance is 25.5 percent. The reduction of wealth is then, on average, equal to the average chance of incarceration—for instance, 25.5 percent times the estimated elasticity for the sample average.

Moreover, this report shows the relative importance of incarceration for the gap in wealth between Black and white households as well as between Hispanic and white households. The report specifically uses a Blinder-Oaxaca regression decomposition. The correlates for this model are the same as for the overall regression model estimating the parameter for incarceration status. The resulting estimates show the share of the average wealth differences between Black and white households in one case and between Hispanic and white households in the other case that the correlates capture or explain. That is, the term explained here refers to systematic statistical correlations, not causal identifications. The report then shows the share of the explained difference that is due to systematic variations of wealth by differences in incarceration status. These results specifically summarize the combined effects of higher likelihoods of incarceration among Black and Hispanic households than among white ones and of any differential consequences of incarceration on wealth—less income, greater income instability, and higher costs, as discussed in other sections of the report—by race and ethnicity.

The estimates first identify how much of the observed difference in wealth between Black and Hispanic households, on one hand, and white households, on the other hand, can be attributed to observed variables such as age, education, family status, and incarceration, among others. The calculations in Figure A-1, below, show the share of that explained racial wealth gap that incarceration captures, depending on the assumed correlation between wealth and having a family member incarcerated.

The underlying estimation models capture most of the observed difference between Black and white households’ wealth, on one hand, and almost all of the wealth gap between Hispanic and white households, on the other hand. Criminal justice interactions explain from 0 percent to 27 percent of these captured or explained gaps between Black and white households. (see Figure A-1)⁶⁵ Using the preferred correlation of -0.175 between wealth and having an incarcerated family member, differences in criminal justice system interactions capture 27 percent of the explained Black-white wealth gap. This is similar to the effect of income differences. (see Table A-1) At the same time, the model only captures 54 percent of the Black-white wealth gap, leaving 46 percent unexplained.⁶⁶ This gap reflects systematic differences in wealth between the two groups that are not captured by the various variables included in the model. This is another way of saying that Black households face systematic obstacles in building wealth beyond income, age, education, inheritances, criminal justice system experiences, and others.

Figure A-1

The results for the Hispanic-white wealth gap differ somewhat. The main model explains all of the observed differences in wealth between Hispanic and white households.⁶⁷ Interactions with the criminal justice system make up 40 percent of that explanation in this case. (see Figure A-1) Even with different statistical assumptions, interaction with the criminal justice system contributes 17 percent—or 26 percent of the explained wealth gap between Hispanic and white households. This is similar in size to the wealth difference explained by income differences—19 percent to 25 percent.

Limitations

The analysis in this report has a few limitations. First, the data can only show correlations between incarceration and wealth, not causality. Second, the survey question used to define incarceration does not account for the time since incarceration. On one hand, a past incarceration can have long-lasting effects on households’ ability to build wealth. On the other hand, households may be able to overcome those obstacles, especially if the incarceration happened in the more distant past rather than more recently. In other words, this report’s correlations between incarceration and wealth are averages over all possible lengths of time since incarceration. Third, households with a family member who has a criminal record but was not incarcerated are included in the group of households that have no family member incarcerated. Thus, the report may understate the magnitude of the negative correlation between interactions with the criminal justice system and household wealth. Fourth, in contrast, incarceration does not always mean having a criminal record.