Academic journal article Contemporary Economic Policy

Isolating the Family Cap Effect on Fertility Behavior: Evidence from New Jersey's Family Development Program Experiment

Academic journal article Contemporary Economic Policy

Isolating the Family Cap Effect on Fertility Behavior: Evidence from New Jersey's Family Development Program Experiment

Article excerpt

I. INTRODUCTION

Welfare reforms since the 1960s have typically comprised packages or bundles of financial incentives, rule changes, and services designed to improve the Aid to Families with Dependent Children (AFDC) program. As Moffitt and Ver Ploeg (1999) note, the motivation behind this bundling has been the presumption by policy makers that the collective effect of reform components is likely greater than the effects of each component implemented separately. Bundling, moreover, has been viewed as the best way to change a pervasive welfare culture (Moffitt and Ver Ploeg, 1999), deliver reform coherently and realistically (Greenberg and Schroder, 1997), and produce systemwide, synergistic impacts (Harvey et al., 2000; Moffitt and Ver Ploeg, 2001).

The practice of bundling welfare reform components has continued through to the passage of the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) and its impending reauthorization. Bundling, however, has presented policy analysts with considerable challenges in their efforts to determine which specific components of reform are responsible for changes in recipient behavior. Fein (1994) has identified four research strategies that have been used to understand the component effects of a reform that has been implemented as a package: viz., (1) document client exposure levels to each provision, (2) identify the unique goals that a component may have for recipients, (3) conduct crossstate comparisons if components vary among states, and (4) perform experiments with multiple treatments. There is a rather broad consensus, however, that classical experiments with random assignment into treatment and control groups are less likely to face the selection and endogeneity issues of other research approaches (Burtless, 1995; Friedlander and Robins, 1995; Lalonde, 1986; Orr, 1999).

Yet experiments, especially those conducted under Section 1115 of the Social Security Act and under PR WORA, have not been very successful in isolating which broad components of reform--work requirements, time limits, family caps, income disregards, and so on--influence recipient outcomes (Moffitt and Ver Ploeg, 2001; U.S. General Accounting Office, 2001). Unlike earlier welfare experiments--such as the New Jersey Income Maintenance Experiment and the Housing Allowance Demand Experiment, where recipients received varying levels of one or more reform components, such as tax rates, and benefit amounts in what amounted to factorial research designs--more recent experiments have had to contend with the same set of components being made available at the same levels, relatively simultaneously, to all recipients in the experimental group. This has resulted in what has been termed black box evaluations (Greenberg and Mandell, 1991; Greenberg and Shroder, 1997; Wiseman, 1993), where overall impact of reform taken as a composite of components is the principal research question that can be examined.

This article attempts to isolate the marginal impact of one of welfare reform's most controversial provisions, the family cap, on recipient fertility behavior. (1) Our research was undertaken in New Jersey where the nation's first family cap was implemented in early 1992 as part of a package of waivers under the Section 1115 program. This study differs from the previous work on family cap effects, including work the author and colleagues have published on the subject (Camasso et al., 2003a; Jagannathan et al., 2003; Horvath-Rose and Peters, 2001; Schettini-Kearney, 2002; Mach, 2001; Kaushal and Kaestner, 2001) (2) by employing two sources of exogenous treatment variation, that is, random assignment of recipients into reform and control conditions and a phased-in implementation of reform components at the county level. Together these sources create a multiple treatment context that helps unbundle the family cap effect from the influence of an enhanced JOBS program--a second reform component that has been hypothesized to influence recipient fertility behavior. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.