This Article offers new evidence on the determinants of U.S. unwed birth rates from 1981 to 1990. We show that illegitimacy rates are positively and significantly correlated with payments under the Aid to Families with Dependent Children program over a period in which real AFDC payments declined. We attribute this result to a decline in the social sanctions for illegitimacy. Because social sanctions declined, so did the cost of deviance, as well as the price for which unwed women sold their virtue.

* We gratefully acknowledge the support of the George Mason University Law School and the helpful comments of Doug Allen, Ian Ayres, Gary Becker, Douglas Besharov, Lloyd Cohen, Roger Congleton, Richard Epstein, David Levy, Lawrence Mead, Bill Niskanen, Eric Posner, Eric Rasmusen and Mark Rom, and participants at the 1995 Canadian and 1996 American Law and Economics Association 1996 Annual Meetings.
 

The effect of welfare subsidies on illegitimacy rates is one of the most contentious issues in welfare reform. Unwed birth rates in the United States have soared in the last twenty years, and illegitimacy has come to be accepted as the root cause of other social pathologies. These concerns prompted the 1996 Personal Responsibility Act ("PRA"), which curtailed payouts under the Aid to Families with Dependent Children program ("AFDC"). The AFDC program offers cash payouts to unwed mothers, and is the most direct economic subsidy of illegitimacy in the United States.

One might therefore expect illegitimacy rates to fall after passage of the PRA. However, there are two reasons why this might not happen. First, the PRA does not mandate payout ceilings, and some states may choose to continue their high payout policies. Second, the link between the AFDC subsidy and illegitimacy rates is not clear. Prior studies have failed to detect a significant correlation between the two. These studies have noted that, during the 1980s, illegitimacy rates rose as real AFDC payments fell.

This study offers new evidence of the determinants of illegitimate birth rates. We find that AFDC payments are significantly and positively correlated with state illegitimacy rates, in a model which employs fixed state effects. Part I reviews the existing empirical literature. Part II discusses our model and variables, and Part III presents our results.

I. Prior Empirical Studies

Illegitimacy rates have substantially increased in recent years. As a percent of total white births, illegitimate white births nearly doubled, from 11 to 20 percent, between 1980 and 1990. During the same period, black illegitimacy rates increased from 55 to 65 percent. For all Americans, the increase was from 18 to 28 percent. Over a longer time horizon, the increase is even sharper. The total illegitimacy rate was 5.3% in 1960, and only 2.3% for whites.

While some academics have argued that these trends are benign (Fineman 1995), there is today a consensus that the increased illegitimacy rates are pathological. Illegitimate children are more likely to be born at a very low birth weight and to have lower cognitive scores (Hill and O'Neill: 1994). Involved natural fathers provide strong role models and a dependable source of income (National Research Council: 1993). Without these benefits, children do much less well than those from married families. For example, illegitimate boys are more aggressive (Dilalla et al.: 1988; Wallerstein and Kelly: 1982 and 1986), and are more likely to break the law (Smith and Jarjoura: 1988). Moreover, the pathologies appear to be passed on to succeeding generations, since illegitimate children are more likely to become unwed parents (An et al.: 1993; Moffit, 1992:36; McLanahan and Garfinkel, 1989: 98; Gottschalk 1990; Rosenbaum 1989; Hanson et al. 1987).

The focus of attention has therefore shifted from the consequences to the causes of illegitimacy, and particularly to welfare subsidies for illegitimacy. The most direct subsidy to illegitimacy in the United States is the AFDC program, which offers cash benefits to unmarried mothers. When the program began, its intended beneficiaries were the children of widowed, divorced and abandoned wives. Today, however, slightly more than half of AFDC children are illegitimate (1994 Green Book). This figure is far higher in the inner city. In the District of Columbia, for example, about 80% of AFDC children are illegitimate (O'Harrow and Davis: 1995b).

The claim that the AFDC program has led to higher illegitimacy rates might be thought uncontroversial. Subsidize something, and you get more of it. To date, however, most empirical studies of illegitimacy have failed to detect a significant positive AFDC or welfare coefficient (Ellwood and Bane: 1985, Garfinkel and McLanahan: 1986). Some studies report ambiguous results, such as a significant positive AFDC coefficient for some but not other measures of welfare availability (Plotnick: 1990). A longitudinal work based upon survey data reported that whether a mother received welfare prior to conception was a significant predictor of illegitimacy, but not the amount of the welfare payout (Duncan et al.: 1988). Lundberg and Plotnick (1995) reported a significant positive welfare coefficient for unwed white births, but not unwed black births. Duncan and Hoffman (1990) reported a positive but insignificant welfare predictor of black teenage illegitimacy. A review of these studies comments that "The failure to find strong benefit effects is the most notable characteristic of this literature" (Moffitt, 1992: 31).

The absence of evidence that AFDC payouts are associated with increased illegitimate births has provided welfare supporters with their strongest argument against AFDC cuts. They note that the recent run-up in illegitimacy rates occurred when real AFDC payouts declined. Between 1980 and 1990, while illegitimacy ratios soared, average AFDC payouts for a family of four dropped from $303 to $268 (in 1983 dollars). Supporters of the present regime attribute the increase in illegitimacy rates to a decline in exogenous social sanctions, rather than to welfare subsidies. What has happened, they say, is that the social stigma of illegitimacy, always weaker for blacks, has declined for whites, and this has made all the difference. On this view, welfare cuts will harm children without reducing illegitimacy rates.

However, some of the prior empirical studies might have failed to detect a significant welfare coefficient because the data set was too small. The most frequently cited study of the determinants of illegitimacy looked at state level data for a single year, with only 37 observations and a dependent variable of combined white and black unwed birth rates. With an R-squared of 93%, only three of the nineteen independent variables were statistically significant, suggesting a serious multicollinearity problem (Ellwood and Bane, 1985: 146). The authors also regressed panel data from the Current Population Survey over a three year period, and again employed small data sets (Ellwood and Bane, 1985: 179).

More recent studies suggest that the conclusion that illegitimacy rates are unrelated to welfare payouts is premature. A 1993 study by Douglas Allen reports that provincial variation in welfare payments is significantly and positively correlated with single parenthood and illegitimacy in Canada (Allen: 1993). Rosenzweig (1995) found that higher AFDC benefits had a small but statistically significant effect for unmarried young women, and a more substantial effect for unmarried indigent women. Other recent studies report that New Jersey's family-cap pilot program, which limits benefits for children born after July 1993 to mothers on AFDC (Feldmann: 1995), has resulted in reduced illegitimacy rates. An initial study, based on a limited sample of the AFDC population from August 1993 to July 1994, failed to detect a significant difference between those who were subject to the new law and those who were not. (Camasso: 1995). However, a subsequent study, based upon the total population of women receiving AFDC, reported a 9 percent decline (Rector: 1995). Another New Jersey study reported an 11% decline in the reported birth rate among AFDC cases, and a 17.8% decline in the high risk category of child-bearing age women who had been on welfare for at least a year and who had not had a child in the prior ten months (Bryant: 1995).

The argument that the run-up in illegitimacy rates can be attributed solely to exogenous social norms, and not to the welfare subsidy, is also unpersuasive. First, social norms might be endogenous, and shaped in part by the welfare subsidy. Social norms were certainly stronger in the past, before the subsidy for illegitimacy. Brides were expected to be chaste when wed, and women who deviated from these norms bore a heavy social cost (Brinig, 1990: 205; Akerlof et al., 1995: 28). For many men, an extramarital pregnancy meant a hasty marriage (Wadlington: 1967). But now the social sanction is far weaker for both men and women. Indeed, it becomes hard to talk of a social sanction in communities where more than half of the births are illegitimate. Arguments from endogenous preferences are enjoying a revival today (Magnet: 1993), and while they are extraordinarily difficult to evaluate, we think it foolish to dismiss them.

Second, even if social norms are exogenous, there is no reason to suppose that they explain all of the variance in illegitimacy rates. It is as short-sighted to claim that economic incentives are irrelevant as it is to claim that social sanctions are irrelevant. Indeed, one would expect more unwed women to react to the monetary subsidy for illegitimacy after social norms weaken. When sanctions were stronger, the cost of deviance was higher, and it cost more to procure a deviant act. But as the social sanction declined, so too did the price at which virtue was sold. This might explain why illegitimacy rates increased while real AFDC payouts declined during the period of our study.

Moreover, even though real AFDC payouts declined, the real value of other welfare payments increased during the period of our study. Real per capita state and local public welfare expenditures (excluding education, health and hospitals) increased from $209 in 1980 to $276 in 1989. Real payouts also increased under federal and joint federal-state programs such as food stamps, school lunches, supplemental security income, and Medicaid. These increases largely offset AFDC cuts. For every dollar cut from AFDC between 1968 and 1984, a dollar was added to food stamps and medical benefits (Moffitt: 1988).

II. Methodology and Description of Variables

A. Social Capital Theories

This study examines the determinants of state-level illegitimacy rates. A state-level focus may be useful over a time series, such as the ten-year period of this study. State level data may also be more accurate than self-reported panel data about sensitive personal matters (Feldmann: 1995, Cooksey: 1990). In addition, a state level focus may be more appropriate than an individual focus in a study of changes in social behavior. Like an individual, a society might also have a character, for better or worse. For example, whether a woman will have an illegitimate child may depend not only on her religious beliefs, but also on those of her neighbors and community.

This Article is therefore a contribution to social capital explanations of behavior. Social capital may be defined as the network effects of individual human capital. (Coleman: 1990, Becker: 1996) For individuals, human capital refers to intrinsic skills and the habits of industry and cooperation which increase expected lifetime earnings (Becker: 1993). Like human capital theories, social capital theories posit that wealth depends importantly on non-material factors. However, human capital is personal to the individual, and can be taken with him if he emigrates, while social capital consists in the external benefits of membership in a society of high human capital individuals.

Social capital theories have been advanced to explain the differing abilities of people to work together for common purposes in organizations (Coleman: 1988). In addition, social capital implicates the social sanctions a society imposes on deviant behavior (Buckley and Brinig: 1996a). The perception that America's increase in physical and human capital has been offset by a decline in social norms likely explains the popular interest in social capital theories.

B. Our Model

Our model estimates illegitimacy rates for whites and blacks in each state from 1981 to 1990. The equations we used to estimate illegitimacy rates in Tables III and IV were of the form:

(1) WHITEUNWED_it = ₀ + ₁ AFDC_it-1 + 2 CON_it-1 + 3 METRO_it-1 + ₄ ABORT_it-1+ ₅ YEAR_it-1 + ₆ WHITEUNWED_it-1 + e_i

(2) BLACKUNWED_it = ₀ + ₁ AFDC_it-1 + 2 CON_it-1 + 3 METRO_it-1 + ₄ ABORT_it-1+ ₅ BLACK_it-1 + ₆ YEAR_it-1 + ₇ BLACKUNWED_it-1 + e_i

where the variables are defined as provided in Table I, and where

₀ ... ₇ = regression coefficients

e = residual

i = 1, 2, ..., 50 index for each state

t = 1, 2, ..., 12 index for each year from 1980-91

We hypothesize that AFDC payouts affect illegitimacy rates. But causation may work the other way, with illegitimacy rates affecting AFDC payouts. This will happen if legislators react to increased illegitimacy rates by altering welfare payouts. With an increase in illegitimacy rates, legislators might either reduce AFDC payouts to respond to the social pathologies of illegitimacy and to repel welfare-motivated migrants, or increase AFDC payouts to attract welfare-motivated migrants for their votes (Brinig and Buckley 1996).

We addressed the causation problem in two ways. First, we lagged all independent variables by a year. That is, we estimated how women reacted to Year 1 AFDC payouts in their Year 2 fertility decisions. This reduces concerns about the direction of causation, inasmuch as causes tend to precede consequences. In estimating pregnancy decisions, of course, a one-year lag is only a three month lag (or possibly a half year lag, given the alternative of abortion).

The causation problem remains, if states set their Year 1 AFDC payouts in anticipation of Year 2 illegitimate births. When illegitimacy rates are timewise correlated, states might indeed be able to form a judgment about future unwed birth rates. We therefore estimated illegitimacy rates through a seemingly unrelated regression (SUR) system of equations (Judge et al. 1988: ch. 11). Our SUR equations in Table V took the form:

(1) WHITEUNWED_it = ₀ + ₁ AFDC_it-1 + 2 CON_it-1 + 3 METRO_it-1 + ₄ ABORT_it-1+ ₅ YEAR_it-1 + ₆ UNWED_it-1 + e_i

(2) BLACKUNWED_it = ₀ + ₁ AFDC_it-1 + 2 CON_it-1 + 3 METRO_it-1 + ₄ ABORT_it-1+ ₅ BLACK_it-1 + ₆ YEAR_it-1 + ₇ UNWED_it-1 e_i

(3) AFDC_it-1 = ₀ + ₁ SOUTH_it-1 + 2 DEMS_it-1 + 3 POVERTY_it-1 + ₄ AFDC_it-2 + e_i

Our reliance on time series, cross-sectional (TSCS) data heightens concerns about idiosyncratic state factors not captured by the other variables. The payout decision across states may be influenced by a variety of factors not captured by our socio-economic variables. It is therefore appropriate to employ a fixed-effects (FE) model, with a separate intercept for each state. (Becker 1993b, Judge 1988: § 11.4). Specifications 3 and 4 of Tables III and IV report on cross-sectional (CS) results, in which fixed state effects are not employed. We also conducted a Hausman test to compare FE and CS regressions.

Several of our specifications feature a Koyck distributed lag model, in which the lagged value of the dependent variable is employed as an a predictor. The crucial pregancy decision might be made not nine months before birth, but several years before then, when a women chooses to be sexually active. Socio-economic factors which affect the pregnancy decision are likely strongest in periods closest to the pregnancy decision, and gradually decline over time. Because of this, we employed the Koyck model, which assumes that the influence of predictors declines geometrically over time (Gujarati, 1988: 513-15).

C. The Variables

Our dependent variable is the number of births to unwed mothers divided by the total number of births. Since white and black illegitimacy rates are strikingly different, we estimated them in separate sets of equations.

We regressed illegitimacy rates on welfare, economic and social predictors, as well as a time trend. Our welfare predictor is AFDC, the average monthly payout per family under the Aid for Families with Dependent Children program, adjusted for inflation. The range in AFDC payouts across states is striking. In 1989, for example, real monthly payouts per family in 1983 dollars ranged from $92 to $556. As noted, the AFDC program is the most direct economic subsidy of illegitimacy. It is also an important welfare program in its own right. At $22 Billion a year (1992 figures), it is the second largest welfare program, after Medicaid. Moreover, AFDC recipients are almost automatically eligible for Food Stamps and Medicaid. As such, "AFDC is in many ways the key to the welfare system for the single parent." (Butler and Kondratas, 1987: 138).

We employed an economic predictor because we expected couples to be more ready to make long-term commitments in expanding economies which offer more secure employment prospects (Testa et al.: 1989). Our measure of economic growth, CON, is the total dollar value of commercial and residential construction contracts for projects completed within the year, adjusted by inflation, and divided by the adult population. The CON variable is particularly sensitive to business cycles.

Because the increase in illegitimacy rates has been attributed to weaker social sanctions for deviant behavior, we employed several social predictors. These were METRO, the percentage of the population living in urban areas; ABORT, the ratio of abortions to live births; and a time trend, YEAR. We would expect higher unwed rates in metropolitan areas. Social support groups, which enforce social norms, are weaker in more anonymous cities, and social norms themselves appear to be laxer in cities (McLanahan and Garfinkel: 1989). On social capital theories, we might also expect illegitimacy and abortion rates to be positively correlated. However, abortion is a substitute for births, and this might suggest a negative correlation. Because of this, it has been feared that AFDC cuts might increase the demand for abortion. Finally, the YEAR predictor might plausibly be regarded as a proxy for changes in social norms during the period of our study. With the sharp increase in unwed rates, it is reasonable to suppose that the social sanction for illegitimacy weakened.

For Table IV's regressions of black illegitimacy rates, we also employed a BLACK predictor, representing the percent black population of the state. A BLACK predictor was employed in Ellsworth and Bane (1985), Lundberg and Plotnick (1995), and Duncan and Hoffman (1990a). We expected the BLACK coefficient to be positive, since illegitimacy rates are higher in black inner cities (McLanahan and Garfinkel: 1989). In part, this might reflect lower social sanctions for illegitimacy. In addition, the BLACK predictor might serve as a proxy for reduced economic opportunities, which in turn reduce the opportunities for marriage.

An interesting literature, most closely associated with the work of William Julius Wilson, attributes the high black illegitimacy rates to a decline in the number of eligible black husbands. (Wacquant and Wilson: 1989, Wilson: 1987). Black men have much higher unemployment and imprisonment rates than white men. Almost 37% of black men aged 18-34 are in jail or prison or on parole (Freeman: 1995, Thomas: 1995). If one adds to this those anticipating arrest and awaiting trial, the loss of marriageable men is striking (Rasmusen: 1995). Because the pool of available men is smaller, there are far fewer black women who marry (Cooksey: 1990). We were unable to test this theory, since data on black imprisonment and unemployment rates during the period of our study are sketchy. However, the theory is obviously not inconsistent with social capital theories which attribute higher illegitimacy rates to a breakdown in social norms.

Our SUR equations are reported in Table V. The instrumental variables we used to predict AFDC payouts were SOUTH, DEMS, POVERTY, and YEAR. AFDC payouts are lower in the south, and the SOUTH dummy variable took the value of 1 if the state had been a member of the Confederacy, and 0 otherwise. Outside of the south, welfare payments are higher when Democrats control the state house. (Buckley and Brinig: 1996b). Our DEMS instrumental variable took the value of 0 if the state had been a member of the Confederacy, and of the per cent of the lower house which was Democratic otherwise. We would also expect lower payouts in poor states, which there are fewer resources and more welfare demanders. Our POVERTY instrumental variable is the per cent of the population living below federal standards for meeting basic needs.

III. Results

The results of our regressions are reported in Tables III-VI. White illegitimacy rates are estimated in Table III and black illegitimacy rates in Table IV. Table V reports on our SUR estimation of white and black illegitimacy rates, and Table VI reports on sensitivity tests. Elasticities are presented in Table VII.

Our most noteworthy result is that the AFDC coefficient is significantly and positively correlated with increased illegitimacy rates for both whites and blacks in our FE specifications. In Table IV's CS specifications 3 and 4, the AFDC coefficient (for black births) is negative. However, it is positive and significant in FE specifications 1 and 2, and a Hausman test for omitted variables casts considerable doubt on the CS model For whites, the coefficients for AFDC are positive and significant in all specifications.

Our use of a fixed effects model distinguishes this from prior studies, and in part may explain why we, more than others, found that illegitimacy rates are significantly correlated with AFDC payouts.
 

Table III. The Determinants of Illegitimate White Births

Akerlof, G. A., Yellen, J. L., and Katz, M. L. (1995). An Analysis of Out-of-Wedlock Childbearing in the United States. Preliminary Draft, Department of Economics, University of California, Berkeley.

Allen, D. W. (1993). Welfare and the Family: The Canadian Experience. Journal of Labor Economics 11: S201-203.

An, C., Haveman, R. and Wolfe, B. (1993). Teen Out-of-Wedlock Births and Welfare Receipt: The Role of Childhood Events and Economic Circumstances. Review of Economics and Statistics 75(2): 195-208.

Anderson, E. (1989). Sex Codes and Family Life among Poor Inner-City Youths. (January) Annals AAPSS 501: 59-78.

Becker, G.S. (1996). Accounting for Tastes. Cambridge: Harvard University Press.

Becker, G. S. (1991). A Treatise on the Family. New York: Cambridge University Press 2d ed.

Becker, G. S. (1993). Human Capital: A Theoretical and Empirical Analysis, With Special Reference to Education. Chicago: University of Chicago Press 3d ed.

Becker, G. S. (1993b). Comments on Danzon, Maki, Murray, and Allen, Journal of Labor Economics 11: S326-329.

Becker, G. S., Landes, E. M. and Michael, R. T. (1977). An Economic Analysis of Marital Instability. Journal of Political Economy 85: 1141-1187.

Brinig, M. F. (1990). Rings and Promises, Journal of Law, Economics and Organization 6: 203-215.

Brinig, M. F. and Buckley, F.H. (1996). The Market for Deadbeats. Journal of Legal Studies 25: 201-232.

Bryant, W. R. (1995). New Jersey's Welfare Overhaul. Washington Post, Oct. 1, 1995, at C1, col.2.

Buckley, F.H. and Brinig, M. F. (1996a). The Bankruptcy Puzzle. George Mason University School of Law Working Paper, Arlington, Virginia.

Buckley, F.H. and Brinig, M. F. (1996b). Welfare Magnets: The Race for the Top. George Mason University School of Law Working Paper, Arlington, Virginia.

Butler, S. and Kondratas, A. (1987). Out of the Poverty Trap: A Conservative Strategy for Welfare Reform.

Camasso, M. (1995). Preliminary Report on New Jersey Family Cap Program. Office of Continuing Education, Department of Social Work, Rutgers University, New Brunswick.

Coleman, J.S. (1990). Foundations of Social Theory. Cambridge: Harvard University Press.

Coleman, J. S. (1988). Social Capital in the Creation of Human Capital. American Journal of Sociology 94: S95-S120.

Cooksey, E. C. (1990). Factors in the Resolution of Adolescent Premarital Pregnancies, Demography 27(2): 207-218.

Davis, P. and O'Harrow, R. (1995a). Even Best Institutions Stymied by Complex Barriers, Setbacks, Washington Post, June 29, 1995 A1.

Davis, P.D. and O'Harrow, R. (1995b). The Region Grapples with Welfare Reform, Washington Post, June 29, 1995, at A1, col.2.

Dilalla, L. et al., (1988). Aggression and Delinquency: Family and Environmental Factors, Journal of Youth and Adolescence 17(3): 233-46.

Duncan, G. and Hoffman, S. D. (1990a). Welfare Benefits, Economic Opportunities and Out-of-Wedlock Births Among Black Teenage Girls, Demography 27(4): 519-535.

Duncan, G., Hill, M. S. and Hoffman, S. D. (1988). Welfare Dependence within and across Generations, Science 239 (January): 467-471.

Duncan, G.J. and Hoffman, S. D. (1990b). Teenage Welfare Receipt and Subsequent Dependence Among Black Adolescent Mothers, Demography 22(1): 16-20.

Ellwood, D. T. and Bane, M. J. (1985). The Impact of AFDC on Family Structure and Living Arrangements, 7 Research in Lab. Econ. 7: 137-207.

Ellwood, D. T. and Crane, J. (1990). Family Change Among Black Americans: What Do We Know? Journal of Economic Perspectives 4 (Fall): 65-84.

Fagan, P. F. (1995). The Real Root Causes of Violent Crime: The Breakdown of Marriage, Family, and Community. Heritage Foundation Backgrounder, No. 1026.

Feldmann, L. (1995). Senate Welfare Reformers Debate How to Curb Growing Problem of Out-of-Wedlock Births, Christian Science Monitor, August 8, 1995.

Fineman, M. (1995). The Neutered Mother, The Sexual Family, and Other Twentieth Century Tragedies. New York: Routledge.

Freeman, R. (1995). The Labor Market, in James Q. Wilson and Joan Petersilia (ed.), Crime. San Francisco: Institute for Contemporary Studies Press.

Garfinkel, I. and McLanahan, S.S. (1986). Single Mothers and Their Children. Washington: Urban Institute Press.

Gottschalk, P. (1990). AFDC Participation Across Generations. American Economic Review 80(21): 267-71.

Gottschalk, P. (1988). A Proposal to Study Intergenerational Correlation of Welfare Dependence, mimeo, Madison: University of Wisconsin, Institute for Research on Poverty, May.

Green Book, (1996). Background material and data on programs within the jurisdiction of the Committee on Ways and Means, Washington: Government Printing Office.

Gujarati, D. N. (1988). Basic Econometrics New York: McGraw-Hill 2d ed.

Hanson, S. L., Myers, D. E. and Ginsburg, Alan L. (1987). The Role of Responsibility and Knowledge in Reducing Teenage Out-of-Wedlock Childbearing, Journal of Marriage and the Family 49 (May): 241-56.

Harvey, A.C. (1990). The Econometric Analysis of Time Series, 2d ed. Boston: M.I.T. Press.

Hetherington, E. M., Cox, M. and Cox, R. (1976). Divorced Fathers. Family Coordinator 25: 417-428.

Hill, M. A. and O'Neill, J., (1994). Family Endowments and the Achievement of Young Children with Special Reference to the Underclass, Journal of Human Resources 29: 1064-1100.

Hill, M. A. and O'Neill, J. (1990). Underclass Behaviors in the United States: Measurement and Analysis of Determinants. New York: City University of New York.

Horowitz, R. (1995). Teen Mothers: Citizens or Dependents. Chicago: University of Chicago Press.

Judge, G. C. et al. (1987). Introduction to the Theory and Practice of Econometrics. Second ed., New York: John Wiley and Sons.

Kandoian, E., (1987). Cohabitation, Common Law Marriage, and the Possibility of a Shared Moral Life, Georgetown Law Journal 75: 1829-1872.

Kmenta, J. (1986). Elements of Econometrics, 2d ed. New York: Macmillan.

Lerman, R. I. (1989). Employment Opportunities of Young Men and Family Formation, 79 American Economic Review 79: 62-66.

Lundberg, S. and Plotnick, R. D. (1995). Adolescent Premarital Childbearing: Do Economic Incentives Matter? Journal Labor Economics 13(2): 177-200.

McLanahan, S. S. and Garfinkel, I. (1989). Single Mothers, the Underclass, and Social Policy, (January) Annals, AAPSS 501: 92-104.

Magnet, M. 1993. The Dream and the Nightmare: The Sixties' Legacy to the Underclass (William Morrow).

Moffitt, R. 1988. Has State Redistribution Policy Grown More Conservative? AFDC, Food Stamps and Medicaid 1960-84, Discussion Paper 851-88, Madison: University of Wisconsin, Institute for Research on Poverty.

Moffitt, R. (1992). Incentive Effects of the U.S. Welfare System: A Review, Journal of Economic Literature 30 (March): 1-61.

National Research Council. (1993). Losing Generations. National Academy Press.

O'Harrow, R. and Davis, P. (1995). Single Women, Fatherless Children.Washington Post, June 29, 1995, A14, col 1.

O'Harrow, R. (1995). Building a Life, With Welfare's Aid, Washington Post, June 29, 1995, at A14, col. 4.
 
Pearce, D. M. (1993). "Children Having Children":Teenage Pregnancy and Public Policy from the Woman's Perspective, in Annette Lawson and Deborah Rhode The Politics of Pregnancy: Adolescent Sexuality and Public Policy 46-58. New Haven: Yale University Press.

Plotnick, R. D. (1990). Welfare and Out-of-Wedlock Childbearing: Evidence from the 1980s, Journal of Marriage and the Family 52: 735-46.

Rank, M.R. and Hirschl, T. A. (1988). A Rural-Urban Comparison of Welfare Exits: The Importance of Population Density, Rural Sociology 53: 190-206.

Rasmusen, E. (1992). Observed Choice and Optimism in Estimating the Effects of Government Policies. Indiana University Working Paper, Bloomington.

Rasmusen, E. (1995). Stigma and Self-Fulfilling Expectations of Criminality. Indiana University Working Papers in Economics, Bloomington.

Rector, R. (1995). Out of Wedlock Births: The Impact of the New Jersey Family Cap. F.Y.I. 50, August 16. Washington: The Heritage Foundation.

Rosenbaum, J.L. (1989). Family Dysfunction and Female Delinquency, Crime and Delinquency 35(1): 31-44.

Rosenzweig, Mark R. (1995). Welfare, Marital Prospects and Nonmarital Childbearing. Department of Economics, University of Pennsylvania, Philadelphia.

Smith, D. and Jarjoura, R. G. (1988). Social Structure and Criminal Victimization, Journal of Research In Crime and Delinquency 25(1): 27-52.

Smith, H. L. and Cutright, P. (1988). Thinking About Change in Illegitimacy Ratios: United States, 1963-83, Demography 25: 235-247.

Tanner, M., Moore, S. and Hartman, D. (1995). The Work vs. Welfare Trade-off: An Analysis of the Total Level of Welfare benefits by State. CATO Policy Analysis No. 240.

Testa, M. et al. (1989). Employment and Marriage among Inner-City Fathers, (January) 501: Annals 79-91.

Thomas, A. P. (1994). Crime and the Sacking of America: The Roots of Chaos. Washington: Brassey.

United States Department of Commerce, Census Bureau. (Various Years). Statistical Abstracts. Washington, D.C.

United States Department of Health and Human Services, National Center for Health Statistics. (Various Years). Vital Statistics of the United States.

United States Department of Justice, Bureau of Justice Statistics. (1985-91). Correctional Populations in the United States.

United States Department of Labor, Bureau of Labor Statistics. (Various Years). Unemployment Statistics.

Wacquant, Loïc Journal, D. and Wilson, W. J. (1989). Poverty, Joblessness, and the Social Transformation of the Inner City, in Welfare Policy for the 1990s Ed. Phoebe H. Cottingham and David T. Ellwood, Cambridge (MA): Harvard University Press, 70-102.

Wadlington, W. (1967). Shotgun Marriage by Operation of Law, Georgia Law Review 1: 183-213.

Wallerstein, J. and Kelly, J. B. (1982). Fathers and Children. In S. Cath et al. (eds.) Anthology on Fatherhood. Boston: Little, Brown.

Wallerstein, J. and Kelly, J. B. (1980). Surviving the Breakup: How Children and Parents Cope with Divorce. New York: Basic Books.

White, K. (1993). SHAZAM User's Reference Manual Version 7.0, Toronto: McGraw-Hill.