Do Asset Limits in Social Programs Affect the Accumulation of Wealth? Henry Chen and Robert I

An Urban Institute Opportunity and Ownership Project Project Exploring Upward Mobility

THE URBAN INSTITUTE No. 4, August 2005

Do Asset Limits in Social Programs Affect the Accumulation of Wealth? Henry Chen and Robert I. Lerman

Public assistance programs provide bene- This brief describes current asset tests fits only to those with too few resources to and discusses their role in reducing asset support themselves. Generally, such pro- building by low-income families. We iden- grams consider both assets and income in tify the target population subject to asset determining eligibility for benefits and tests, describe the various asset tests, payment amounts. The rationale for this review background data and previous approach is that the government should research, consider strategies for meeting a not have to support people who can take mix of objectives, and point to the need for By excluding low- care of themselves by converting bank additional research and policy analysis. income people who accounts, stock holdings, vehicles, or retirement accounts into cash or collateral. As Background on Asset Tests such, assets are considered “means” that have saved enough to The share of the population potentially should allow families to avoid destitution. subject to asset tests is high, as table 1 doc- deal with minor Assistance programs typically do not count uments. Of a total of about 105 million assets below some low threshold, allowing contingencies, asset households in the United States, approxi- families some savings or a car that may be tests limit the mately 27 million were participating in necessary for work. However, families with major state and federal benefit programs in effectiveness of assets above the threshold may be ineligi- 2000. Medicaid had the most participants, ble if they fail the so-called “asset tests” programs that promote followed by food stamps, Supplemental designed to target payments to those most Security Income (SSI), Temporary As- self-sufficiency. in need. sistance for Needy Families (TANF), and Asset tests can create a disincentive to housing assistance. All of these programs save among families who might subse- test assets prior to enrollment, exposing quently qualify for benefits. Sometimes one about one in four U.S. households to the additional dollar of assets can result in the associated disincentives to accumulate loss of thousands of dollars per year in assets. public assistance benefits. This raises the Assistance-providing organizations question: Do asset tests actually discourage have taken account of assets for centuries. savings and reduce asset accumulation In the early 20th century, Charity Organi- among families who might qualify for pub- zation Society caseworkers performed lic assistance benefits? At least one influen- home visits, partly to determine whether tial paper says yes. Hubbard, Skinner, and families possessed assets that could sup- Zeldes, in their 1995 article, interpret the port their needs. Today’s primary rationale low levels of wealth accumulation among for asset tests is that the limited funds low-income households as a rational, available for social programs should be utility-maximizing response to asset- targeted to those most in need of financial based, means-tested welfare programs. assistance. Targeting the needy means

1 OPPORTUNITY AND OWNERSHIP An Urban Institute Project Exploring Upward Mobility

TABLE 1. Households in Major Federal and State Beneﬁt Programs Subject to Asset Tests, 2000

Households Receiving Beneﬁts

Program Not elderly Elderlya Total

Temporary Assistance for Needy Families (TANF) 2,828,946 247,106 3,076,052 Supplemental Security Income (SSI) 4,096,337 1,526,638 5,622,975 Food Stamps 7,631,884 1,381,235 9,013,119 Medicaid 19,219,759 5,740,734 24,960,493 Housing Assistanceb 2,971,749 1,149,163 4,120,912 Non-duplicative total 20,847,103 6,769,017 27,616,120

Source: The Urban Institute (2004) Notes: Based on March 2002 CPS data extracted from the Urban Institute’s Transfer Income Model (TRIM). The TRIM-simulated caseload numbers may deviate from the actual caseload reported in administrative sources for several reasons. First, TRIM does not include U.S. territories in its imputations. Second, TRIM does not include all portions of a beneﬁt program in its simulations. For example, TRIM simulates eligibility for full coverage under Medicaid, excluding immigrants (who receive only emergency coverage) and those for whom Medicaid only pays premiums for Medicare Part B. a ”Elderly” includes households containing persons 65 or older who are the only recipients of program beneﬁts. b TRIM only includes federal housing subsidy programs for renters.

excluding those with low incomes but suf- of assets can be dangerous. As a result, ficient assets (savings, retirement accounts, asset tests are less common in programs etc.) to finance basic needs. For example, with annual accounting periods than in many retired people have no income but programs with monthly accounting maintain their pre-retirement levels of con- periods. sumption by drawing on their wealth. Without asset tests, political support Asset Tests in Major for assistance programs could be under- Government Social Programs mined if benefits were paid to families with low incomes but enough assets to Asset tests vary widely across government take care of themselves. Yet low-income social programs, as shown in table 2. The families need cushioning for hard times. differences arise from decisions by the fed- By excluding those who have saved eral government for some programs and enough to deal with minor contingencies, by state governments for other programs. asset tests limit the effectiveness of pro- The federal government sets rules for grams that promote self-sufficiency. Low- SSI, housing assistance, the Earned Income income potential recipients face a financial Tax Credit (EITC), and the Pell Grant pro- disincentive to save, since their savings gram, while states decide on TANF, the may prevent them from drawing on public State Children’s Health Insurance Program benefits at a time of need. (SCHIP) and Medicaid rules and play a Asset tests are most appropriate for role in selecting policies for food stamps. programs that use short accounting peri- Usually, the programs ignore a family’s ods to determine income and payments. net worth in their home and in a car worth Basing payments on monthly income less than a specified figure. For liquid instead of annual income allows programs assets, families are ineligible for benefits to respond to families’ immediate needs. above a set threshold. For example, a bank But some families with low monthly account of $2,000 or more would exclude a incomes have adequate long-term incomes household from receiving food stamps, no and savings. Although it is appropriate to matter how low the household’s income. expect families with short-term disruptions Similarly, an elderly couple with $3,000 or to draw on their savings, dealing with more in liquid assets would be ineligible long-term shortfalls through the depletion for SSI.

2 An Urban Institute Project Exploring Upward Mobility OPPORTUNITY AND OWNERSHIP

The 1996 welfare reform law gave the percentage of net assets counted as states the discretion to set low asset limits income exceed actual financial returns for TANF, or no limits at all. In fact, the from assets). For example, if a family had limits on liquid assets for cash assistance $10,000 in assets earning 3 percent and applicants range from $1,000 in some states with a (hypothetical) imputed annual rate to $10,000 in others. In a striking and un- of return of 0.35 percent, the family’s usual move, both Ohio and, more recently, countable income from assets would be the Virginia have eliminated the TANF asset larger of $300 (0.03 times $10,000) or $35 test. Although no estimates of effects on (0.0035 times $10,000). Hence, $300 would caseloads are available, a close observer of be counted as income from assets. Like the Ohio welfare program reports seeing other countable income, income from little or no impact from the decision to assets lowers net housing benefits by about eliminate asset tests. Medicaid asset limits 25 percent of each added dollar of income. vary not only by state but also by potential But the income won’t prevent those with recipient. Most states have eliminated asset modest assets from retaining eligibility for tests that determine children’s eligibility, benefits. A similar provision applies to the but half retain asset limits ranging from Pell Grant. Although potential recipients $1,000 to $20,000 for low-income adults face no strict asset threshold, the program who are neither elderly nor disabled. All imputes as income 12 percent of the par- state Medicaid programs retain asset tests ents’ assets and 35 percent of student par- for the elderly and disabled categories. ticipants’ assets. Under the EITC, there are States generally do not restrict eligibility no limits or any imputation of income from for SCHIP on asset limits, but when they assets. However, individuals earning more do, limits range from about $5,000 to than $2,650 (in 2004) from interest, divi- $10,000. dends, royalties, or rental income are ineli- The treatment of retirement accounts, gible for the EITC in that year. For those interestingly, depends on the type of with less than $2,650 in property income, account. Often, employer-provided defined EITC benefits phase out with actual income contribution retirement accounts such as from assets or other sources at about a 401(k) and Individual Retirement Accounts 21 percent rate. (IRAs) are subject to asset tests, while 1 defined benefit pensions are not. The pre- Equity, Incentives, and sumed rationale is that funds from IRAs Administrative Costs and 401(k) accounts can be more easily converted into liquid assets to deal with a One clear inequity built into asset tests short-term income loss than can earned results from ignoring the liability side of rights in defined benefit pensions. Yet the balance sheet. A family’s wealth and this policy treats equal pension assets capacity to be self-supporting depend both unequally. And subjecting any pension to on assets and liabilities. Yet assistance pro- an asset test may reduce public assistance grams count only assets in determining payments but deplete retirement savings, eligibility. As a result, program rules can which would increase participation in be more generous to wealthier families. assistance programs during retirement Consider two families with the same char- (Neuberger, Greenstein, and Sweeney acteristics except that one family has zero 2005). net worth (say, $3,000 in a bank account Low-income housing programs apply but $3,000 in debts) and would be deemed an innovative treatment of assets. Instead ineligible for benefits, while the family of specifying a threshold and excluding with net worth of $1,000 (all in a bank those with $1 more, housing programs account and no debt) would be eligible. count as income either actual financial Treating some forms of assets more returns from assets or a percentage of net generously than others is another common assets (when net assets exceed $5,000 and and arguably inequitable feature of asset

3 OPPORTUNITY AND OWNERSHIP An Urban Institute Project Exploring Upward Mobility

tests. In particular, one applicant might couple moved from less than $1,000 in own a home with tens of thousands of dol- liquid assets to between $1,000 and lars in equity but remain eligible for bene- $2,000 in liquid assets; and benefits fell fits, while another is ineligible because of a from $15,669 to $10,017 (over $5,600) if savings account with as little as $1,100. The they moved from between $1,000 and usual argument is that homes are illiquid $2,000 in liquid assets to between $2,000 and those in need should not be required and $3,000 in liquid assets. to leave their homes to qualify for benefits. Ⅵ TANF and food stamp asset tests drive The ability of homeowners to obtain most of the decline in benefits as assets home equity loans reduces the force of increase up to $3,000. this argument, but again raises the prob- Ⅵ At low asset levels, asset tests based on lem of ignoring debt. The homeowner income from assets, as used by the EITC, with $100,000 in equity would be treated have virtually no impact on benefits. the same as the homeowner with no equity. Embedded in these calculations is differentially A few programs have looser asset tests counting various forms of assets and ignoring for some demographic groups than others. debt. For example, assets in the form of a fam- Medicaid generally has no asset tests to ily’s first car are only counted in some pro- determine the eligibility of children, but grams if the value exceeds a threshold (about half the states retain asset limits on eligi- $5,000). Again, by taking no account of the bility of adults in families with children. loans required to finance the car, programs SCHIP, targeted toward children, rarely may well treat people with equal net worth uses an asset test. Food stamps provide a unequally. slightly higher asset limit for households These calculations may overstate or with a disabled or elderly person than for understate the impact of assistance pro- other households. However, SSI, the wel- grams on the assets of low-income families. fare program that covers disabled people Families concerned about losing eligibility (and the elderly), uses a strict asset test of for benefits can potentially shift their asset only $2,000 for individuals and $3,000 for holdings into those not counted in deter- couples. mining benefits. In principle, those with These differences in the treatment of relatively high gross assets but low net assets by program and by type of asset assets can become eligible for benefits complicate the savings disincentives asso- while leaving their net worth constant by ciated with assistance programs and their paying off debt. asset tests. Certainly, for families with few Dealing with the inequities built into savings outlets other than liquid assets, current asset tests may improve fairness asset tests reduce the potential gains from but may raise administrative costs substan- accumulating assets because slightly tially. A liquid asset rule is relatively easy higher asset levels could exclude the fam- and less costly to administer, while one ily from benefits. We can illustrate these that requires valuing all assets and liabili- savings disincentives by examining how ties is complex. The approach used in asset tests affect hypothetical low-income housing programs—valuing all assets families. Rules from a state typical in then counting a percentage as income— benefit levels (Pennsylvania) illustrate is attractive from an equity viewpoint, how asset tests can cause sharp reductions but is potentially costly to administer. in benefits as changes in assets affect program eligibility. The administrative effort may be especially cumbersome if valuations were Ⅵ For a married couple with two children frequent. Such calculations might be dif- in tax year 2003, benefits fell from ficult for caseworkers and lower-level $19,118 to $15,669 (almost $3,500) if the administrators.

4 An Urban Institute Project Exploring Upward Mobility OPPORTUNITY AND OWNERSHIP

Empirical Analysis of the (2004) found that eliminating an asset test Impacts of Asset Tests increases participation by 17 percent. Few have examined the impact of asset tests on Researchers have focused on two impacts Aid to Families with Dependent Children of asset tests: how asset tests affect assis- (AFDC) or TANF participation. One work- tance programs’ participation and costs, ing paper by Mach (1999) finds that liberal- and how they affect recipients’ savings and izing asset limits reduced the TANF exit asset holding. Whether tests affect partici- rate. pation and costs is not obvious since eligi- Most of the empirical studies focus on ble participants have few assets and since how asset tests in benefit programs influ- some eligible families might have chosen ence savings and asset accumulation. not to participate, even without asset tests. Variation in asset tests across states and On savings and asset holding, Hubbard, over time has enabled researchers to Skinner, and Zeldes (1995) point out that empirically measure the effects of asset means-tested social assistance programs tests on asset holdings. Table 3 sum- can discourage saving by low-income marizes several of the effects reported in households for two distinct reasons: the literature. households see less need for precautionary In general, the studies find that asset savings if they know public assistance is limits lower the net worth of potentially available, and asset tests can cause people eligible low-income individuals and fami- who save to lose benefits. lies. However, the size and significance of Although income and asset data can the effects varies across studies and pro- tell us by how much asset tests reduce grams. The SSI and Medicaid programs eligibility for benefits, estimating impacts exert clear and relatively large impacts. on participation is more complicated. One Gruber and Yelowitz (1999) find that must project, for example, how many more Medicaid lowered wealth holding by people would become eligible for a pro- 16 percent among eligible families headed gram if asset tests were eliminated, and by 18- to 64-year-olds. Researchers study- how many of those would enroll. In addi- ing the AFDC/TANF program have tion, one must also estimate whether cur- concluded that asset tests have reduced rently eligible people would nevertheless vehicle ownership (Hurst and Ziliak forth- be more likely to enroll if they did not have coming; Sullivan 2004), but the effects on to report assets. liquid assets and net worth are insignifi- Some existing research is relevant. cant in the most recent study (Hurst and Tabulations on the Food Stamp program Ziliak forthcoming). Ziliak (2003) also presented by Russo (2003) show that as of finds that the ready availability of asset- 1999, participation was about 40 percent and income-tested programs reduces among households eligible by income liquid assets and total assets modestly and (whether or not they qualified when assets by more than the availability of benefit were taken into account) and 54 percent programs not subject to asset tests. among those eligible by both income and Overall, researchers have uncovered assets. Thus, about 26 percent of house- evidence documenting both the intended holds without elderly members and eligi- effect of asset tests (in limiting participa- ble by income were not fully eligible but tion) and the unintended effect of asset would become so if asset tests were elimi- tests (lowering savings and asset accumu- nated. If 25 percent of this newly eligible lation by low-income families). However, group actually participated, the food we find no studies that provide a com- stamp caseload would rise about 16 per- prehensive picture that compares the gov- cent. The percentage growth in elderly ernment cost savings from asset tests households receiving food stamps would (including lower benefit costs but added be even higher. In a study of Medicaid and administrative costs) with the magnitude SCHIP enrollment, Bansak and Raphael of the reductions in asset holdings. Yet it is

5 OPPORTUNITY AND OWNERSHIP An Urban Institute Project Exploring Upward Mobility

the relative magnitudes of these two couraging low-income families from sav- impacts that are most relevant for policy. ing. Asset tests are often inequitable because they completely exclude some Implications for Policy families, although they have only slightly more economic resources than families eli- Should asset tests be restructured? Should gible for benefits. they be liberalized or eliminated? One Our agenda for the future should be as restructuring approach would be to impute follows: income from assets when determining benefits, preventing the “cliff” problem of one 1. Improve the equity of existing asset additional dollar of assets excluding low- tests, perhaps by turning to imputation income families from thousands of dollars of income from assets, by taking account of potential benefits. Federal housing pro- of liabilities, and by treating all types of grams already use the imputation strategy. assets similarly. While few disagree with the approach’s equitability, some oppose imputation 2. Reduce the administrative complexity because of administrative complexity. and administrative costs of asset tests. Another sound policy change may be to 3. Take account of inflation by periodically make asset limits similar across programs. adjusting asset limits. Such a change could reduce administrative costs and make participation less confusing. 4. Quantify the magnitudes most impor- Good arguments can be made for liber- tant to judging the size and overall alizing existing asset tests, especially since desirability of asset tests. Using esti- they have not been indexed for inflation or mates based on solid research, policy raised for several years. For example, the analysts should determine the size of asset limit for SSI has remained constant the cost savings for government and the since 1989; the asset limit for the Food amount of reduced savings and asset Stamp program, since 1985. Asset limits holding by potentially eligible families. are often extremely low. The SSI asset limit Armed with this information, policy- of $2,000 for singles and $3,000 for couples, makers will be able to make more revised in 1989,2 now has the buying informed decisions about the level power of approximately $3,135 and $4,700.3 and structure of asset tests. If the SSI asset limit appropriately re- stricted eligibility in 1989, it no longer does Notes today. In the absence of inflation adjust- The authors are grateful for comments from ments, even moderate inflation will lower Signe-Mary McKernan, Eugene Steuerle, the real value of asset limits and could Bob Greenstein, Zoe Neuberger, Dottie Rosenbaum, reduce eligibility substantially in Eileen Sweeney, and Barbara Sard. Elizabeth Bell the future. provided excellent research assistance.

1. The Food Stamp program is an exception. Neither Conclusion 401(k) nor defined benefit accumulations are subject to an asset test, but IRAs are. Asset tests serve some role in targeting government safety-net benefits to those 2. When SSI was created in 1974, resource limits were with few resources as well as low incomes. $1,500 for an individual and $2,250 for a couple. Between 1984 and 1989, Congress increased the SSI This approach can be equitable in treating asset limits in five annual steps. Each year, the indi- those with more economic resources less vidual limit increased by $100, reaching $2,000 in generously than those with the same 1989. The limit for couples was increased by $150 incomes but fewer assets. Yet many prob- each year and reached $3,000 in 1989. (Section 2611, lems arise when implementing asset tests. Public Law 98-369, July 18, 1984) They raise administrative costs and the 3. Calculations using the Bureau of Labor Statistics stigma of receiving assistance, while dis- inflation calculator: http://www.bls.gov.

6 An Urban Institute Project Exploring Upward Mobility OPPORTUNITY AND OWNERSHIP

References Russo, Randy. 2003. “Tables Describing Asset and Vehicle Holdings of Low-Income Households in Bansak, Cynthia, and Steven Raphael. 2004. “The 1999.” Mathematica Policy Research Report No. Effects of State Policy Design Features on Take Up 8659-318. Washington, DC: Mathematica Policy and Crowd Out Rates for the State Children’s Research. Health Insurance Program.” Discussion Paper 05- 02. San Diego: San Diego State University. Sullivan, James X. 2004. “Welfare Reform, Savings, and Vehicle Ownership: Do Asset Limits and Gruber, Jonathan, and Aaron Yelowitz. 1999. “Public Vehicle Exemptions Matter?” Working Paper. Health Insurance and Private Savings.” Journal of Notre Dame, IN: University of Notre Dame. Political Economy 107(6): 1149–1274. Ziliak, James P. 2003. “Income Transfers and Assets of Hubbard, R. Glenn, Jonathan Skinner, and Stephen P. the Poor.” Review of Economics and Statistics 85(1): Zeldes. 1995. “Precautionary Saving and Social 63–76. Insurance.” Journal of Political Economy 103(2): 360–99. About the Authors Hurst, Erik, and James P. Ziliak. Forthcoming. “Do Henry Chen is a research Welfare Asset Limits Affect Household Saving? Evidence from Welfare Reform.” Journal of Human assistant at the Urban Resources. Institute. His research deals with employment and Mach, Traci. 1999. “Determinants of AFDC Caseloads: training, welfare programs, How Have Exits and Take-up Rates Been Affected fatherhood and family by Welfare Reform?” Paper presented at the National Association of Welfare Research and structure, and the causes Statistics annual workshop, Cleveland, OH, and effects of asset building for low- August 8–11. income households.

Neuberger, Zoe, Robert Greenstein, and Eileen Robert I. Lerman is a senior Sweeney. 2005. “Protecting Low-Income Families’ Savings.” Retirement Security Project Policy Brief fellow in the Labor and No. 2005-6. Washington, DC: The Retirement Social Policy Center at Security Project. the Urban Institute and a professor of economics at Neumark, David, and Elizabeth Powers. 1998. “The Effect of Means-Tested Income Support for the American University. Dr. Elderly on Pre-Retirement Saving: Evidence from Lerman’s research deals the SSI Program in the U.S.” Journal of Public with youth employment and training, Economics 68: 181–206. welfare programs, fatherhood and family structure, and economic inequality. His Powers, Elizabeth. 1998. “Does Means-Testing Welfare Discourage Saving? Evidence from a Change in recent work deals with the impact of mar- AFDC Policy in the United States.” Journal of riage on the economic well-being of fam- Public Economics 68(1): 33–53. ilies and the earnings of men.

7 OPPORTUNITY AND OWNERSHIP An Urban Institute Project Exploring Upward Mobility r - s - d - d g s m o c e s e ; e e e d n / y l ? r n ) l k A a o r l - i u s i ; d i s l F e c r f s o l c t ; ; m t i n r e f D d d e o a h n ( u d i I l i n o N s l i r y y fi v h o g l o i c n n o h s s s t t s c a f s i e c a e s e A e a i l t n r r ; a e n d t A l i A A A c e e d t x I d T v n e y e e i p , ; p p o g m o e c d e e n i r F D D D n s y e s f ; p p t d e s r t m I I I t n t o t p u a ’ e m s a d i r b A e l e e r fi a o o u d e o s s e e t a d t r , e r r l d c o A A A e e e l r s I h m o x s s I I I a ; s s a t M a t d p p c u n p t p n s t e o o y s ; e S l F F F i r e ; t p r u l s n e e e e a o c h e o c l d d a e S d s a F A A A l r s r c r a o s b e e c i x u m y e n n , d s r r a u h l p r r r s p N r t s t x e t e n h e F s m a a b i e u t o l a r d a o o o , f a u e A e x a e u m s a s s v N t e b n o s e e e r s T t t r o i i v e e F F F q e n h n s e n e r i p i A e - g s e e e e i h i ; l m m m I f N N N b s e n fi r d T r m p s l W s s o h e c g e o o o S r t u n n e i a o n s s A A A e f e u o h o i i n n o a T t p T V a i i T H b l t H t H c d s S s o h a o s t e f ? s k o t t n r e r s s t i o a n e e i n e o a e i t g o t c d r t b m i a h p t s n - t e s n m t i i r e r d a p e i d f l m r s d r h o e a e u s o t l u w u e t e m f l p x s e f t o o h e b e c , p l e t e l t e l l a a c y h , x x l i a a l e m t l s s x o e e e l u d h h I e e f s e a r e c u e t 0 i d d s s r l w t x s i i y F h 5 u e e a u u , e n e l l e n 6 e e t v u o n N e u , c c l v a e a n n e b n s 4 x x a r A o t t m i a s T O c O e G $ o v e A - - r f e u l o , m e s ; s a s o i 0 a r - s t h 0 u l h f o d l , t i 0 l e t w 6 ; t a r s 0 r 0 t h o b t , s r i 0 s u e o 0 e e ? c a h t f 6 o o 0 e s m d h 0 h l r e i m a , t s t $ t 0 i e i e t s , l p d o v m 1 e o i l e f s t n o u 1 e m u $ r m t a l i d o a r o 0 $ l o a e o t b n n a e n 0 0 t n i c h t n s m i a r h , e 0 0 a m t o g r r r o t s , , d s s i o 0 h r s o n n o o o 2 , o s d f n i e r f f f d o t a 0 f $ 1 W n t e y n n l i a s 0 0 0 0 0 $ i a ; e l i s m t r a r 0 0 0 0 0 s a ; s e t e , n i e o 0 0 0 0 h s m u n , , 0 , , r t y e e d w h i l e s o l r a 3 1 2 3 3 l o i h r o l l c w w o f $ U i $ $ d t V $ $ a , , m m - - e e ? d d s s o o s v v t t l e e r r m m a a f f e e e c c o o s s v s s u u e e c c t t e s s s l s s d d n n s s t i i a a i i t e e r r a a e r r r r fi s a a d d p p e e s e e e e e y y a h h n n n b b n n t t a a e o o r r o o o b n n a a t t m m D r r t a a s s e e c o o h h c c e e n n e r r c c , , f s s u u i i f e e s s s g t g t h h o a y y e e e n n n n i i N Y i Y Y i b w b w s m a r y t g - e i a ? I l i o h t i v y r n m e g s S t r a i b i P l i s e l e u s l S o h i g i t s t g l f r o a i i b i m n fi e a e s l i t h e s s e v t e V e g m p i d s i r d t n s l l , a e e a e e y a d d e f y e o l t t l c c B r i b n o t r s h a n x e l h o a c D r s e o n a u e e c i e f g r s e f o , , s i e f e i t u r d r s s s m m t d a s i h e o a l e e e n a e o e a O Y Y V h Y t e b s c F d n a t s - e ? t r n e y t a e o c t t a i g l t m S r e o e s n s r o t y r p t t o s a i e j t y l u a c a i v f e i b o b s e a i d M , t i s g l l g o a v a r i a a h l o r r n s f c s i m e e e i t e e t t t i s l h e d d m t n a a i s a s t t l a e e m W e c i l c F S e S F i T t e s h s e s c d A i a n s a . c m a p 2 c a t i F r s d i m E d g I N o s a e L o S t s A r o B M S F S a P T A T

8 An Urban Institute Project Exploring Upward Mobility OPPORTUNITY AND OWNERSHIP d n a g s h - n i s s g t s d t e u u n e p d m n f o e o i t e a s r H m r s h p d m t f e e e u g t u o , l x h d m b s o t i t e c r . e e n e t s t o k d e x p x v a e o s d e n t e t m o n b e t n i s m a r fi t B e e a a s o e r t s y n d t b r p t s e a f n e e b n e e n e o s e r s a l D s s t s u s t t s G b c e k e i a o e e a i d l h 4 t n r h c s s 0 l a n ; o l 0 a a r e s s c y e 2 t a v e a V t N a A b r e e h v t o d P t n ’ a n i n ; n o n e o d i r t d a s s l r i t t e h s s t i t C e a n t s i t r o e s t m f t e t d r t s s e A t e s o n y s a t e i s m r C e a u l ; v c s a e o o o a n e S o N h N N Y i l t a i a c N o f S e h d e o t s t t n s - i h i d e i m s ; t d r o e a 3 f t n ; n u o ; e i g u ) 0 g r r s e s 0 t g f e n 0 t t d a a m r e i 2 e t r c n s 0 e o t n e r e . d s d e n t x u s 0 e o s m a t o s , t e e t e e b e s n o m c o w n a 5 e s c m d a d a e c m c o s o e f m r c $ t l e t c u e s f t t c h n h l e d t o x r v s t i e t n s o l u n c e o i c e p u e n i r o m s a u s t t o n h e n N a p u s t i o w s t i a s a n e h m t r ” a e m o u , , e t f 5 s e ; ; s s t e s d i c t t 0 b n a , c s a a 3 e e s i d t s e t e r m 0 0 , s s s s h t t s s e a . e t t r n n e m m 0 0 e e r a r p n n , e e w e e e e g 0 p o o t n n e t d u u , 0 c r s s o h w t c c g a o o r t e 5 1 2 x a s n s o o t n n n n w P c 1 P a $ a i i e N s I n $ c a i N o / t / fi : e p n t t e h h B ( l g s a t u r c s e a o d f a r e s h h F d t e t l t e n a i t a y e l r s h n t n e s e u l e t t o e o a T . t u t t c c s e e o u c s n s e s a b t o m i s , , A t e o a a s s s s d c o d h e e e s n n n a p i Y N Y Y a u s o t F n u r e o s c i t - c i d a s s - A i t e K o t m e s d s i e n l t m h s e h e d m e i t t t e e e y h a ; l m m s t e s s h m o e e g t s t n e s t r e o d g o i c v a a u r e o a t i s f c u r y a e s h v o s t l d d t r n h r o R o s h e e a i e u n , e e t l a g h h s S s t t l s t b o t t d n m m a e u a n n l d e , , , , e i e e u c o o t m o n i t u u t s s s s a t s r c c a c h o a o x e e e e c s o o i t i h r e n n d c t a Y e f Y Y Y s t a i i n c V e , s M A d l D n I a a “ r y , e l t r e n i i a a l d i c e r d i t e b e s d f n i n e d a x e — e e M e ) f r e 8 r 4 t fl G o 0 l l a f — n 0 b t y a a s 2 t E o o s ( r r r i i l B e e t e e e t t M t h a c n d d t a m u c i e O t t e o e e i o r C t F F l S w H S f s ; t n a I t n a e n d a m e t b n p c r n o o l n g U a e d a r n e e v t i P s h e I s G s a i T D C l H u B s : l e : T n o s C I s c e a r e t S H A P E b u o r o S N U

9 OPPORTUNITY AND OWNERSHIP An Urban Institute Project Exploring Upward Mobility s s : e e y y l : : t y l y y ? e a — l l t l l t — s c s c s i s a a s t e e e m c c e f a i i e f y t y a r r r r E i i t C o — — g p p e t I D e s o s F S m m h r s e E t a S p E A T g g e ) o s s n n n - - C i i s e - - r h g a d h t s t v t c v i D b p u e c a e n a s a w t i n a F r f n d a u a e h e p s v s n e t c i a . t s o y e A h s r d a t s c s e n t , t i e w i r i n x s e l u e n s e t s h i 4 e i c i o r t o a r s v s e n 0 i s 8 f t y b u d t h a p a n i a e r s e 0 9 t i t d w s s l t p i s e g a a s 1 p e t l 1 s i e r a e c a l s n f s g $ x a i g r n b p c s i l e e e d e d o o e n n e u r e s l i c a s i t i p I d n t l w t c l c a d a d o e o n ; a S d a t u fi r m u n n s c h n 5 i i s i c h s n a S e a 8 m i o t , . t n e 2 o F i c e t n e 7 a y n 1 d s 1 fi e g l . t w s r e y l e 9 a $ s w e 8 e u l o s r n n n t o g a i t t 1 b l 2 b e n o i p a y s e c c m o t l h a I a $ i e y m h c r f n e a l t c e p m n f r S m e b i e p o y f n e e a l n g g o e a i I S i g m r e i m e t k b r i o a t n r e i t i p S o r i l r n n s x g h e w a s o h s t l i t e p d d u t S 0 e e s o o g e g c i e h h r r r 0 v v p r m n s d o t n n n e a a a i H i i 1 o a p T c a B e j a ( p e y - r r 8 o - a t 7 s e , h a l 9 e I s c n l n p t i 1 s S i t b a p l 3 c s s a S n i i s p m u 8 t e e i l r t e x l t s 9 s a g a fi i e e t t t 1 v r e n e e e t y w n e a t n d t s s e a e h t e e n c s s K b a a C S b a a m s l - a e d t 8 t v e u , n 7 e o l a o t n g 9 c i m t 1 w n g h x n n o i t a i i ) 3 l n e l c s n n i s t 8 a , u e e u e e 7 d e u 9 e h l g g o e e m u t 1 O d c l l n n w h u i w w c a n c a a t t t d h c x g a e h h e e e e n e n A 4 C b i n w C ( v b a t e s . s n y s h l t e t A - c e c e - n d n a k r e i m e o s l w f n o e a f t r f d n s a e r m f e e e o i p g n o - d h e g t s m i o b l s t d n a p e o n l a o s i o i d y p I r l a r h h e - l s v e — h t t t g a p m S t p e e d e a e e h e o a c u r o s t k c S s i r o c c c i n r h t s i l g M e n f t P n n e n e n e t m e n i fi e I i l e e r i e m r v s r o r r t a u i S g o a i e e e — n w e e e n t c I S f e S n f f t n c d f c i o e f f a L i S h y i e o i f n n t e S d r a O b e T S d d m D d s I d o d e t s e l y T d 4 - a d n s 6 i r e n u d a – n e i c 8 l n s t o l b l t a 3 e 0 f s o 7 o o l s e g i a / 8 n m 6 s r 9 o i t a h e n e t t 9 e a l l i l e a 1 M i s e r l n 1 a m s b p h f y s p d i u g l h t e i o u I t u t a d 5 9 p m g t e o c o i s S o r e o a 7 2 l o o n c k t e i m a b S 2 e p l h h 6 S P a r p c e f f l E a d n e f i P t l o r d P a I t a o n u y , a t p S n e g i e e h D o t n g 4 v i R m t r u u n 8 f o a u 9 o o o o Y y 1 w S N L n o i t - ) ) - c y f ) e r d s e 8 8 d - l o r 5 d a - e y 9 9 e n d n i p t t 9 P e e n e t m c d a 9 9 S n ” n a t v u s i c i 9 i l g o h n n , o e n 1 1 a e n E e i i t S e k e ( ( 1 r ” A n a t e o a c d , f T ( r . o r : ? c r - F f r M m I r e e u P e s s ” U n . a g S m s u e g g g r l s n y a r r a o E a e a . S s g 3 s a l e f r n e a a e e e c n n C n n n o m e e r r b c m i e n i e c S U i i i f r r e t g t t l i E a c h e d n d o e u b u m D c a r v v v t o l n w w i i u t h s r a s o e L e e e s c d e i u F o h o k a t a a e v o e e l o r o o e T D P h h r B n n n S H E f C A S i S D S T W “ S P P t t d “ “ p R E S I M P N I S Z A T

10 An Urban Institute Project Exploring Upward Mobility OPPORTUNITY AND OWNERSHIP t - s s r e o e e — — e s n y y s s : : : s n F F t t — y y y a — — e e l l l p — N N w l l l s t t i s s s d d o s s a a a e A A h i i s s e c c c e e T T y s a a a a i i i e e y t t / / r l l r r r c c i i i t t i i e C C d d c c — i i i i — p p p e e d d n t p D D u u h h s i s s e e s F F m m m w o o e e e s s q q h e i i v E s a n E A o a E A l n v M t l y M t - - t t p s e e i u l s fi t a o 0 n e - s n i e t t - h l c b h b y n s 0 n e a e n i 0 n e o s t a s n c I i e l s i b o c 0 i t - s e v s j r 0 t g g e i r , a . r t a n i i r a c – t t s l h i u n e l t 0 a b d 1 l o e s A a h l e , c - o m a e o p s e y e b n o u d i e $ u m c . 1 p h u s n m i n v t y v e a n f w c b , i i r t i e l S s d s t u l w $ m d u l a . t i s r 2 A t o m r s . 0 e a e t s h e t n t e p o y s a f q h n u m 1 i o . t h t m n s i a r a u 0 v e x i e h i e c o e l b e r t a e a q c , 0 i h t e c c i e h r a s 5 r o s i s r e e e r t e , l v s a s w 4 c c f m r e n s h r s s i u h a d t i c a s r i w g m r y t 1 e t t i e E $ l a a a a o a h a l i t i n l . e u s r e e e o o s o i t s $ i . t s t t f n l s a l c r d a a p g m s n n e n i f e y n i p n g n a i i f e n e e l t e a o e d l l p n a t s w e r a o b d b h n o o s s d c h c g w o i i i s 3 i e p i s e i i e h i u w o t . o c h e s s s e . . t e d n e t a a t o s g h v d l t u t t h p a r ’ f i 2 a v e a t p i o a w c g e c a v e l f a a p o i c i r e s c e i x e i n n q t e i r t fl n r i s t h e e e a s w a r n n d n o m l e v s e d e e t d n w c w i i h r i m l t o c ’ a c y u e o n s e e c c b e r d d l c n e n e o e r s n n t w h o g r r i i i i a s e e t t e x d y t o d t - e v L d t n x e l i m e i r a a o o h e e e o s s s c M i a l e c n t s s k e 0 c , . e e t g c c i a a e b t t t t n p p , t r a e t i i l w c o o l a n p s s s s 0 e s j n s e e c c e j t t t 3 t a c o e e a d d y u a n i i i p u 0 3 6 i l r r s b s u o n e e d a c 9 , . . b m t n s n s s e a e e e f f c c c d r s s u y i o s o 9 6 1 4 f f n s a r m m m t u u e e i i n i n i e n n i n s e h A e 1 1 i M c t a s M $ b a l l s m h t s i i c 1 i l r S i l i m r - r ; d e i t a r s l d d a r e l a e e l a i , l t e e t t t c a e r t l a l o t a i l l i s t a a t d m e e l r c o i o c c t t d c d i 1 e t f s i t n o h i o o i s s n a m l d t t e s f 0 s i h i a d r r h r d l a l r r s f f a 0 e m e t e e o o y M e a i e e e a t t e o o 2 e v l s n r v l , , l n g e h h M r a – n e m t r e , s , s i t t d c e e b a a r n a c i i t t t t 4 i r e l i i i i i e e h t u u r a u l m d s s 9 u h l l t g s t s d h i h i u h o u a a 9 l e a a n s m m m m q e u n i i i i i w d C f d h v l I l l w v t w a e l v a C h l 1 d - i n e u - 1 e s q e 0 f n e i u r l v 0 o o d o o w i i 2 s f t t t d g h n l u i e e n p r d o o r q l n b o i n e w e h l m t , f a s s g o e h a u ; h t t o t t n s s c 4 e e e i d r e l a u l g n 9 s s d c o h h i o o 9 s s o o n w c L h I C w h a h a 1 , r , s e r n o e i , f t l a g o s p - i ; a i i a a e y s y n t s s n r t , l n h , y u i u s s o r e a a l o n s i o l d u g , d e i e x r v i s d n s p o i r b s n q s i e e l v d s a r e , t e i a g s u e x o a i n t n e s a x a t t d e r e n r fi a e n p fi c e a r r w m t n a n l o t g i g i o u e g r s a e o v c i s e m t S , i s t a d e , , r l e l a L o s s r e h m c r s s e e t l s t o a c t t e i p e a t r u r y O s , u s c e i r c c t r i r t u v b r e r e e h t g e e e l d d d n h t p o o h o a r c f f s t i e t r r f f e b n n n o a a o n n a c f R I c r e a a P m a i v g e o m w - , ) - - e h s r n s t s l s e m i n a e 0 4 4 l s r n u o r e e p - d o p 0 k h 6 6 w g l i r d o o o o t d u e i s t 0 – i l n r m i a , o o h i s m h s i s s o a 8 r r e e 0 h w r c s - h s o u e v o e 1 m t e v h l 4 s e r c c g e e o s n a 1 h r o s a r e d e t h s d d n h 2 h n s g s l u l r g a v d 1 t l o a m u g 2 i s o l e e b l e e a o o o i 8 , n h r e e s a ( b h A h o h h m l t w 2 5 f h d i h p , 3 3 P X 9 9 P E 9 9 I 1 1 C . S ) , , ; 5 4 2 6 0 P D I 0 8 9 9 P 2 I S ( 9 9 9 e 1 1 1 S P t u t i t ) s d h n g t I e d n l s r : n s e n t a a n i d r i a p a d n a e f i d l a n b l e d t m o f r e ” , m h n o i l a H ” c e , e i ” z e n o t ? s s c s n U a r t ? e h r L n s i a e e r c c r p e g g g n a ) ) e a l l m m W i t e s e a t s t l f e t h r r e h W e v n n n 4 9 c c w t l r s k c t t a i i i i i i b m i e s A b n T o o l t r 0 9 d e a o u e u r m l v s f f v v v : o e i h h u i l f u i a e 0 9 l s o m w r o u o s f o u a a a e e x v i r e e c e i W D P 2 1 F r r n V E L D R A M V H “ ( “ S O S Z A ( S R H E f S I “ P Y ( G u o S

11 THE URBAN INSTITUTE Nonproﬁt Org. 2100 M Street, NW U.S. Postage Washington, DC 20037 PAID Permit No. 8098 Ridgely, MD

Address Service Requested

For more information, call Public Affairs: Given the chance, many low-income families can acquire assets and become more finan- 202-261-5709 cially secure. Conservatives and liberals increasingly agree that government’s role in this or visit our web site, transition requires going beyond traditional antipoverty programs to encourage savings, http://www.urban.org. homeownership, private pensions, and microenterprise. The Urban Institute’s Opportunity To order additional copies and Ownership Project held five roundtables in 2004 to explore these options. This policy of this publication, call brief series presents some of our findings, analyses, and recommendations. The authors 202-261-5687 are grateful to the Annie E. Casey Foundation for funding the roundtables and policy briefs or visit our online bookstore, and to the roundtable participants for their insightful comments. http://www.uipress.org.

THE URBAN INSTITUTE 2100 M Street, NW Washington, DC 20037 The views expressed are those of the authors and do not necessarily reﬂect those of Copyright © 2005 the Urban Institute, its board, its sponsors, or other authors in the series. Phone: 202-833-7200 Fax: 202-293-1918 Permission is granted for reproduction of this document, with attribution to the Urban Institute.