What Works in Work-First Welfare: Designing and Managing Employment Programs in New York City

1. Introduction

1 1 Introduction This book is a case study of how New York City’s welfare-to-work programs were managed and implemented in the mid 2000s. New York City’s welfare system is unique in many ways, so the results may or may not be generalizable to other cities. Even so, the case study is intended to be a rich source for the generation of hypotheses and a compelling and interesting story in itself. What makes New York City’s welfare system unique? To start, it is the largest urban welfare system in the nation, with about 350,000 individuals receiving cash assistance in early 2010 at a cost of more than $10 million per month.1 About one in 25 individuals in the national welfare caseload resides in New York City.2 Another relatively unique feature is the use of private contractors (nonprofit and for-profit organizations) to provide all of the employment services for welfare recipients. A few other cities also use private contactors, including San Diego, Milwaukee , and Houston, but most cities use government agencies to provide welfare-to-work services (Sanger 2003). New York City also stands out because of its use of performance-based contracts, with the providers’ compensation tied to job placement and employment retention outcomes of participants, not simply to the number of people served. And finally, the scope of the New York City’s welfare benefits is unique. In terms of eligibility, for example, the city (and, in fact, all of New York State) provides benefits to qualifying noncustodial individuals, in addition to qualifying custodial parents and their children. As a result, the city’s caseload has a much larger percentage of noncustodial adult men than most other cities. And in terms of time limits, New York City (and State) does not have the typical five-year limit on lifetime benefits. Instead, once custodial recipients exceed five years of cash benefits (or noncustodial individuals exceed two years), they are eligible for a safety-net program without a time limit, paid for with state and local funds.3 Two additional unique features of New York City’s welfare system are, coincidently, useful from a research perspective. First, 26 welfare- 2 Feldman to-work programs operate within the five boroughs in New York City, and the city randomly assigns welfare recipients to different programs within their boroughs.4 Recipients who live in Brooklyn, for instance, are randomly assigned to one of the eight programs within that borough . The city uses random assignment to be fair to programs, aiming to create an even distribution in terms of participant characteristics. But from a research perspective, this form of assignment creates a natural experiment that reduces selection bias when comparing programs’ results. Second, the city gives programs latitude to design their own service strategies, as long as those strategies emphasize a relatively quick entry into jobs. The resulting differences in program practices create useful variation for investigating which practices are more effective than others. Despite its unique features, New York City’s welfare system shares a fundamental similarity—its work-first approach—with almost all other current U.S. welfare systems. Work-first programs use immediate job search, or short-term training followed by job search, rather than longer term education and training. Their goal is to move individuals quickly into unsubsidized employment. The shift to work first occurred across the nation in the mid 1990s catalyzed by federal welfare reform. In the decade preceding that reform, rigorous evaluations of welfare-to-work programs were interpreted as documenting that work-first (or “labor-force-attachment”) programs produced better results than skill-building (or “humancapital -development”) programs, including higher employment rates, less welfare usage, and higher incomes (Bloom and Michalopoulos 2001). Influenced by those evaluations and, most importantly, by the new mandates of 1996 Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA), states and localities across the nation adopted a work-first approach. Today, almost a decade and a half after that sea change in policy and practice, we still know surprisingly little about which frontline practices are most effective within the work-first framework. In other words, why are some work-first programs better able to help welfare recipients become and stay employed? This book aims to provide new insights into that question. Introduction 3 TANF AND OTHER U.S. EMPLOYMENT AND TRAINING PROGRAMS PRWORA created the Temporary Assistance for Needy Families (TANF) program...