Assets and Well-Being

Assets are the stock of resources that people own or have, while income is the flow of resources that people receive and allocate over a period of time. Assets are important because they provide resources and security for daily living, and serve as a form of insurance by enabling people to weather crises and meet irregular expenses. Assets also enable people to invest in education, homes, small businesses, and other opportunities that support development over the long term (Sherraden, 1991, 2014).

Widespread belief and a growing body of evidence indicate that holding assets changes a person’s attitudes and behaviors in positive ways. Many studies show that financial assets and homeownership are positively associated with children’s educational attainment and emotional and behavioral well-being, probably at least in part because assets change expectations about the future (see reviews in Elliott et al., 2011; Grinstein-Weiss et al., 2014; Williams Shanks et al., 2010).

Income (as a proxy for consumption) has been the standard definition of poverty and well-being in 20th-century “welfare states.” And this definition has considerably shaped public policy. For the most part, social policy provides income support to those who do not have sufficient income from labor or other sources. As inequality in labor income increases, these policies are under increasing strain. In the early 21st century there is growing recognition that income alone may be insufficient to provide for well-being, even economic well-being. Sen (1993) and others are looking toward capabilities. Asset-based policy—broadly defined as any public policy that encourages and supports individuals to accumulate, hold, or develop assets (Emmerson & Wakefield, 2001; Sherraden, 1991)—can be seen as part of this larger discussion, as one measure of long-term capabilities. Asset accumulation and investment enable development and expand capabilities which are essential for success in today’s rapidly changing knowledge-based global economy (Paxton, 2001, 2002; Sherraden, 1991, 2014). As public policy, asset building may be a form of “social investment” (Midgley, 1999; Sherraden, 1991). From this perspective, asset-based policy could be an explicit complement to income-based policy.

Asset Inequality and Dual Asset-Based Policy

In the United States, asset inequality is much greater than income inequality, and this is especially evident when looking at race. In 2009, the median net worth of Whites was 18 times greater than the median net worth of Hispanics and 20 times greater than that of African Americans. That is, the typical minority household owned about a nickel for every dollar of net worth held by the typical White household. In the same year, African American (35%) and Hispanic (31%) households were twice as likely as White households (15%) to have zero or negative net worth (Kochhar et al., 2011).

These patterns are the result of “officially or quasi-officially sanctioned institutions and policies” (Sherraden, 2014, p. 268). As Oliver and Shapiro (1995, 2006) have documented, past and ongoing institutional arrangements, including slavery, Jim Crow laws, land confiscation, residential and mortgage discrimination, unequal educational opportunity, predatory lending, and tax laws that favor asset holding and inheritance have made it extremely difficult for minority families, especially African American families, to accumulate wealth, and this has become cumulative across generations (see also Shapiro, 2004).

Part of the larger historical pattern, emerging in the latter half of the 20th century, is what Sherraden has called “dual”-asset policy: social policy that promotes asset building for middle- and upper-income families but not for lower-income families (Sherraden, 2014, p. 269; see also Howard, 1997; Sherraden, 1991). Much of the regressive asset-based policy in the United States works through the income tax system. Examples include home ownership tax benefits; investment tax benefits; defined contribution retirement accounts with tax benefits such as 401(k)s, 403(b)s, individual retirement accounts (IRAs), and Roth IRAs. Other asset accounts with tax benefits include state college savings plans (commonly called “529 plans”), health savings accounts, and medical savings accounts. These defined contribution policies, consisting of individual accounts with benefits dependent upon the amount of assets accumulated, have all appeared since 1970 and are growing rapidly.

Unfortunately, the poor receive almost none of the benefits. Because these public subsidies operate through tax deferments and exemptions, families with higher incomes receive greater tax subsidies. Today the United States spends over $500 billion annually in tax expenditures for asset building, primarily subsidizing homeownership, investments, and retirement savings. The vast majority of these subsidies go to upper-income households (Sherraden, 1991; Steuerle et al., 2014; Woo et al., 2010). For example, in 2013, 72% of federal tax expenditures on the home mortgage interest deduction went to taxpayers in the highest income quintile and 19% to taxpayers in the second highest quintile. The remaining 60% of taxpayers received just 9% of these subsidies. In the same year, 84% of federal tax expenditures on employer-based retirement savings plans and IRAs went to taxpayers in the two highest income quintiles. Taxpayers in the lowest income quintile received less than 1% (Steuerle et al., 2014).

Meanwhile, some means-tested public assistance programs have asset tests which allow families to have only a limited amount of savings or even require them to spend down savings in order to qualify for benefits (Beverly & Clancy, 2016; Nam et al., 2008). These rules create a disincentive for asset accumulation (see Nam et al., 2008, for a review of evidence) and are misguided and unfair. Note that some asset-building policies for the non-poor have no limits—some wealthy people accumulate tens of millions of dollars in IRAs, all with public tax benefits. Thus, U.S. social policy substantially subsidizes asset building for upper-income families but penalizes asset accumulation by low-income families. Prior to the publication of Assets and the Poor (Sherraden, 1991), nearly all recipients and would-be recipients of means-tested public assistance faced asset tests. Since the early 1990s, federal and state governments have substantially relaxed asset tests. However, differences in asset tests across programs and across states are likely confusing, and families may still believe that stringent asset tests are in effect. By loosening and eliminating asset tests for those who receive means-tested benefits, social policy is moving in a positive direction, yet there is a long way to go before asset-building public policies are as generous for those with lower incomes as for those with higher incomes.

Asset Building for All: Toward Inclusive Policy

In contrast to the limited and regressive asset policy in place in the United States in the early 21st century, comprehensive asset-building policy should be universal, progressive, and lifelong (Sherraden, 1991, 2014).

Under universal policy, there is full inclusion: everyone participates. Full inclusion cannot be achieved without automatic enrollment and automatic deposits that are not contingent upon family deposits (Beverly, Kim, et al., 2015; Clancy & Sherraden, 2014; Clancy et al., 2016). If participation is voluntary, requiring people to actively enroll, a higher proportion of advantaged families will participate and benefit. If asset accumulation depends primarily on family deposits (as in the defined contribution programs described), advantaged families will receive nearly all of the subsidies (in 2021 policy, more than 90% of the subsidies go to the top 50% by income). This is true for asset-building policies for all purposes, whether for home owning, retirement, college expenses, healthcare, or other purposes. In contrast, automatic enrollment (with the ability to opt out) and automatic deposits extend the benefits of asset holding and asset subsidies to everyone, regardless of socioeconomic status. In fact, automatic features have larger impacts on disadvantaged families.

Under progressive policy, the poor would receive greater public support than the non-poor. Good governance often defines policy for the people who most need the support and services (e.g., disability benefits for the disabled, retirement support for older adults, and so on). Asset-building policy, however, does just the opposite—the policy is constructed to make the rich richer. This is an ineffective use of public resources. If U.S. policy is supporting asset building, at a minimum policy should be fair (in the sense of the same dollar amount for all). And ideally, the policy should be progressive (more for those who are most in need).

Finally, under lifelong asset-building policy, investment accounts would be opened early—ideally at birth—and would follow individuals into retirement. Opening accounts early is important for a number of reasons. First, asset accumulation is a long-term process for most people. Over time, regular deposits—even small ones—can result in significant asset accumulation. Second, opening accounts and providing subsidies early allows families to benefit from investment earnings so assets may grow substantially even if families do not contribute. Third, asset holding appears to affect attitudes and behaviors in positive ways, and it seems helpful to initiate these changes early. An early start also provides an opportunity to build financial capability in households, which establishes a foundation for positive financial functioning and asset building throughout life (Sherraden, 2013; Sherraden & Grinstein-Weiss, 2015).

In addition to supporting asset holding early, lifelong asset policy recognizes and supports different needs and goals over the life course, including education in childhood and early adulthood, home purchase in the middle years, and retirement security in the later years.

Individual Development Accounts

As a response to regressive policy, individual development accounts (IDAs) were proposed as a universal and progressive asset-building policy (Sherraden, 1991). As originally proposed, IDAs would include everyone, provide greater support for the poor, begin as early as birth, and be used for key development and social protection goals across the lifespan, such as education, home ownership, business capitalization, and retirement security in later life. Although proposed as a universal and lifelong concept, IDAs have been implemented so far as a targeted and short-term policy for low-income adults. (In the typical IDA program, individuals open and save in a restricted bank account and earn 1:1 or 2:1 matches if withdrawals are used for postsecondary education, homeownership, or small business development.) Why did this occur? New ideas must enter and compete in the always challenging arena of policymaking. IDAs in this version were what was possible to do at the time. Thus, IDAs as targeted and short-term asset building have been in a demonstration mode for over two decades, with many variations in the United States and other countries. A great deal has been learnt during this demonstration process (see the section “Lessons from Implementation and Research”). But this is far from a comprehensive asset-based policy.

Child Development Accounts

The next challenge was to return to the original concept of universal and lifelong asset building. Child development accounts (CDAs) are savings or investment accounts supporting asset accumulation for developmental purposes and life-course needs. Under the policy vision articulated by Sherraden (1991, 2014), the federal government would automatically open an account for every newborn and provide a substantial initial deposit (e.g., $500 to $1,000). Low- and moderate-income children would receive additional automatic deposits at certain milestones, such as entering kindergarten and graduating from high school. And public or private funds would match deposits by parents and others into the accounts of low- and moderate-income children. Accounts would eventually follow individuals throughout the life course, supporting asset accumulation for postsecondary education, home purchase, small business development, and retirement security. That is, CDAs would become universal, progressive, and lifelong. The original proposal called for subsequent deposits related to educational achievements and service to the country, altogether totaling over $14,800 (Sherraden, 1991), which would be about $28,000 in today’s dollars.

CDAs have been implemented at national, state, and local levels (see Sherraden et al., 2016, for more detail). Singapore has the oldest and most comprehensive CDA policy, with accounts, deposits, and savings matches supporting asset accumulation for education and children’s health-related expenses (Han & Chia, 2012; Loke & Sherraden, 2009; Sherraden et al., 2016). In 2016, Singapore substantially expanded its CDA policy, increasing the cash gift and providing a new automatic deposit for preschool and early childhood expenses when parents open a special account (see Made for Families). Other countries with national CDA policies include Canada (for postsecondary education; low- and moderate-income children receive subsidies), Korea (for postsecondary education, housing, small business development, medical expenses, or wedding expenses; children in the child welfare system and some children in families receiving welfare are eligible), and Israel (initially for postsecondary education, homeownership, small business development, and wedding expenses; beginning in 2017, all newborns automatically receive accounts and monthly deposits). The United Kingdom had a universal and progressive CDA from 2005 to 2010; it was eliminated as a budget-cutting measure in an “austerity” response by the newly formed UK coalition government in 2010.

In the United States, legislation to create a national CDA policy has been introduced in several sessions of Congress, notably through the America Saving for Personal Investment, Retirement, and Education (ASPIRE) Act (Cramer & Schreur, 2015), and, in 2015, through the USAccounts: Investing in America’s Future Act (H.R. 4045 2015). Also, in early 2016, four states (Connecticut, Maine, Nevada, and Rhode Island) had statewide CDAs, with more in discussion (Clancy & Beverly, 2016). The first comprehensive statewide CDA policy emerged in Maine, where every resident newborn automatically receives a $500 grant for postsecondary education and every child is eligible for up to a $300 annual match on savings deposited into a state 529 account (Clancy & Sherraden, 2014). (Named after the relevant section of the internal revenue code, 529 plans provide tax-preferred investment accounts for college savings). Almost every state has a 529 plan (see, e.g., SavingforCollege.com). In recent years, Pennsylvania, Nebraska, and several other states have initiated universal CDA policies for all newborns. A number of U.S. cities and localities also have child accounts. Perhaps best known is the Kindergarten to College (K2C) program in San Francisco, which aims to include all public school children (Phillips & Stuhldreher, 2011).

Lessons From Implementation and Research

Given the recent emergence of this policy concept, applied research has been fundamental for learning how policy design can be inclusive, whether policy and product features spur asset accumulation, and what policy design can be sustainable—then using this information to inform policy development. By observing a wide range of CDA, IDA, and other asset-building programs, a great deal has been learnt about how to implement effective, inclusive, and efficient asset-building policies and about impacts of these policies.

IDAs were rigorously examined in the American Dream Demonstration (ADD), an experimental, longitudinal, and multimethod study (see, e.g., Mills et al., 2006; Schreiner & Sherraden, 2007; Sherraden & McBride, 2010; Sherraden et al., 1999, 2005). Analysis of account data shows clearly that some low-income people will save in a structured and subsidized savings program. Over 36 months, IDA participants saved an average of $17 a month or about $200 per year. The average participant made a deposit in about half of the months that their IDA was open and deposited about 42 cents for every dollar that could have earned a match. A diverse group of low-income people saved in IDAs. Participants’ saving was influenced by match rates, match caps, time caps, and other program rules (Schreiner & Sherraden, 2007). At the end of the experiment, the IDA program had a positive impact on homeownership rates. This relationship was mediated through debt reduction (Grinstein-Weiss et al., 2008).

Six years after the IDA programs ended, researchers followed up again with participants in the Tulsa experiment and compared outcomes for those in the treatment group to those in the control group (Grinstein-Weiss et al., 2012, 2013a, 2013b, 2015). Because these two groups were formed through random assignment, differences in outcomes can be attributed to the IDA program. The programs had positive impacts on two of five IDA uses: first, rates of enrollment in educational programs were higher in the treatment group than in the control group. Moreover, this difference in educational participation was larger for males than for females (Grinstein-Weiss et al., 2013a). Second, the program had a positive impact on home repair (Grinstein-Weiss et al., 2012). In addition, although the impact of homeownership was no longer significant at the six-year follow-up for the full sample, for participants with above-median income, there was a positive impact on homeownership rate and duration (Grinstein-Weiss et al., 2013b).

In another randomized experiment in Canada, IDAs for education and small business development had positive impacts on financial management, self-reported saving behavior, attitudes about education, enrollment in postsecondary education, and microenterprise start-ups (Leckie et al., 2010).

In-depth face-to-face interviews with IDA participants shed light on people’s perceptions of IDAs and saving and enable quantitative findings about participation and impact to be interpreted. Rigorous analysis of data from interviews with ADD participants shows that, while saving was very difficult, participants wanted to save and appreciated having a structured savings program with incentives, financial education, and other support. And account holders described noteworthy cognitive and psychological effects. For example, some of the participants who successfully saved reported feeling that they had more control over their lives and more confidence in their ability to make decisions and plans for themselves and their children (Sherraden & McBride, 2010; Sherraden et al., 2005).

Looking back after several years, and assessing all the evidence and experience, Sherraden (2014, pp. 270–271) suggested the following as important lessons from ADD and other studies of IDAs: (a) the poor can save when they have structures and incentives to do so; (b) saving is explained mostly by institutional arrangements, as in a 401(k) plan; (c) individual behavior is not enough: there has to be a structured platform and plumbing; (d) it is much easier to build on an existing policy platform rather than try to create a new one (in retrospect, it was naive not to see this at the outset); and (e) as theorized in Assets and the Poor (Sherraden, 1991), it is asset accumulation that matters most for outcomes in well-being. In sum, asset-based policy is not solely about improving choices, behaviors, or other individual characteristics or actions.

Another important contribution of IDA demonstrations and research is that saving and asset accumulation by the poor, which was seldom discussed in the 1990s, has become a mainstream, bipartisan idea in the United States (Sherraden, 2014). Both Republicans and Democrats use the language of “asset building,” “asset-based policy,” “stakeholding,” and “ownership society.” During a period of extremely partisan politics and a generally dysfunctional U.S. Congress, this is perhaps a noteworthy achievement. As always, diversity of political support presents both opportunities and risks. In this case, one political risk is that asset-building policies might be viewed as a substitute for—rather than a complement to—income support policies. Both asset-building and income support policies are important.

Because CDAs are a comprehensive asset-building policy, lessons about their implementation and impact are especially valuable. The SEED for Oklahoma Kids (SEED OK) experiment, which began in 2007, is the most rigorous study of CDAs to date. Research methods include probability sampling from a full state population, random assignment to treatment or control group, and multiple data sources including surveys, extended in-person interviews, and account information provided by the account manager, not self-reported by participants (Nam et al., 2013; Zager et al., 2010). The CDA in SEED OK has characteristics recommended by CDA proponents, including automatic opening of accounts and automatic initial deposits for all and progressive subsidies to support asset accumulation by low- and moderate-income families.

The CDA in SEED OK was the first fully inclusive CDA in the United States, that is, the first to provide accounts and assets to all children (in this research case, all children in a randomly selected sample and randomly assigned treatment group). SEED OK achieved full inclusion through automatic account opening and automatic initial deposits. One of the strongest findings from CDA research thus far is that full inclusion cannot be reached without these automatic features (Beverly, Kim, et al., 2015; Clancy et al., 2016; Nam et al., 2013; see also Clancy & Sherraden, 2014). The fact that parents and children did not have to “do” anything to receive accounts and initial deposits does not make achieving full inclusion any less meaningful. CDAs are a population intervention, like an urban plumbing system providing water to all residents (Clancy et al., 2016). And, as Beverly et al. (2016, p. 8) note, “Demonstrating full inclusion paves the way for widespread participation in asset building and more equitable distribution of public resources.”

Another early lesson from SEED OK is that creating an asset-building program on an existing centralized platform, such as a 529 plan, has many benefits (Clancy et al., 2015). Centralized recordkeeping and investment create efficiencies and allow for all to be included. (It would be very difficult for multiple local programs to achieve full inclusion.) In addition, including the full population in a single platform allows larger, more profitable accounts to subsidize smaller, less profitable accounts, and this helps make a fully inclusive policy financially viable for asset managers. IDA demonstrations reveal the value of using an existing platform rather than creating a new one, so it is important to emphasize that the 529 platform exists, and experience shows that it can be adapted for CDAs. Also, unlike basic bank savings accounts, 529 investment accounts have the potential for market growth (and the risk of market losses), and SEED OK research demonstrates the common-sense notion that all households can benefit from asset growth (Beverly, Clancy, et al., 2015). In short, the 529 platform can be viewed as public resource that, with some adaptations, can benefit everyone.

Turning to findings about the impact of CDAs over time, because so few children have college accounts and college savings without a CDA, the CDA in SEED OK has very large impacts on account holding and asset holding, especially for disadvantaged children. Also, as expected, the CDA eliminates or greatly reduces variation in account holding and CDA asset holding by socioeconomic status; that is, it reduces asset inequality early in life (Beverly, Kim, et al., 2015; Sherraden et al., in press). The CDA also increases the likelihood that parents themselves save for their children’s future college expenses, and this is true in advantaged and disadvantaged families (Beverly, Kim, et al., 2015). The CDA in SEED OK also has positive impacts on mothers’ expectations for their children’s education, mothers’ mental health, and child social-emotional development. Again, the effects of the CDA are often greater for disadvantaged children, which seems to be largely due to the automatic features of the CDA and not to parental saving behavior (Huang, Sherraden, Kim, et al., 2014; Huang, Sherraden, & Purnell, 2014; Kim et al., 2015).

In sum, research on CDAs provides more evidence that asset holding changes attitudes and behaviors, even if people do not “do” anything to receive accounts and deposits. This broad finding makes the observations about achieving full inclusion through automatic features and creating efficiencies by using a centralized platform all the more valuable. There is a substantial body of evidence that asset holding is positive, and research on CDAs shows how to extend the benefits of asset holding to all, regardless of socioeconomic status.

It is not possible to predict where this will lead. Some of the policy advantages of asset building include the following: it is simple and clear, is flexible and adaptable, has multiple positive outcomes, and often enjoys widespread political appeal and acceptance. A considerable disadvantage of early 21st-century asset-based policy is that it is very regressive. The goal should be a universal, progressive, and lifelong asset-based policy. If every person and household has assets to provide for social protections and invest in future development, this would contribute to improved life chances and reduced inequality, both of which are core values in social work.