Economics 3 4 Notes

[Suk Harian]

Labormarket policies succeed or fail at least in part depending on how well they reflect or account for behavioral responses. Insights from behavioral economics, which allow for realistic deviations from standard economic assumptions about behavior, have consequences for the design and functioning of labor market policies. We review key implications of behavioral economics related to procrastination, difficulties in dealing with complexity, and potentially biased labor market expectations for the design of selected labor market policies including unemployment compensation, employment services and job search assistance, and job training.

The Great Recession of 2007 to 2009 and its aftermath have been a trying period for American workers. The U.S. unemployment rate reached double digits in late 2009 for the first time in over a quarter of a century and has remained over 8 percent through mid-2012. Real compensation growth has all but stalled. The human costs of labor market turbulence have rarely been clearer, and the value of public policies, such as unemployment insurance and job training programs, that assist workers in managing that turbulence, gaining new skills, and navigating the labor market have rarely been more apparent. Similar problems of persistent joblessness have been apparent in most major economies in recent years.

Labor market policies succeed in meeting their objectives, however, only to the extent that they accurately account for how individuals actually make decisions about work and leisure, job search, and education and training. To a substantial extent such policies are built around standard economic assumptions of behavior that individuals are perfectly rational, time consistent, and entirely self-interested. The design of unemployment insurance with job search requirements intended to minimize distortions to incentives to return to work, the use of complicated eligibility criteria and administrative hassle factors to discourage social program participation except for the presumed most needy, and the shift to vouchers for training services have all been justified by these assumptions.

In these notes, we briefly review selected topics in labor market policy through the lens of behavioral economics. We identify aspects of U.S. policy design that appear at odds with behavioral findings, as well as unrealized policy opportunities those findings suggest. The results of this review are prescriptions for policy design and innovation that reflect a synthesis of traditional and behavioral economic insights. We consider the implications of behavioral findings in three areas: unemployment insurance, job search assistance, and job training. Although we focus on the specifics of U.S. labor market policies, the lessons we draw potentially have broader applicability.

Providing income support for the unemployed while encouraging their speedy return to work is a primary goal for labor policy. The ability of individuals (and even of many extended families) to self-insure, to smooth income and consumption out of savings and transfers from relatives and friends, is limited in the face of job losses and potentially extended spells of unemployment. Furthermore, private markets may fail to provide adequate mechanisms, such as loans, because of asymmetries of information. Our main policy instrument to maintain the consumption of job losers as they seek to gain reemployment is unemployment insurance (UI).

From the standard economic viewpoint, the main design challenge for UI is moral hazard. UI policies must balance the provision of liquidity and support for consumption smoothing during unemployment against the tendency of such benefits to distort incentives to search for and take new employment (Baily [1978]; Chetty [2008]). Increases in the generosity of benefits, either through increases in benefit levels or the duration of benefits, appear to lengthen the unemployment spells of unemployment insurance recipients (Meyer [1990]; Katz and Meyer [1990]).

U.S. unemployment benefits are contingent on active job search efforts (job search requirements) and benefits are strictly time limited in part to address the potential for moral hazard. Time-limited UI benefits can be interpreted as a crude version of the key implication of the Shavell and Weiss ([1979]) standard model of optimal unemployment insurance that benefits should decrease with duration of unemployment to provide appropriate search incentives. There has been some experimentation with the terms of unemployment compensation to further promote reemployment including the use of reemployment bonuses, variation in the intensity of monitoring job search, and the provision of job search assistance and self-employment assistance. But such efforts have to date generated only mixed results (Meyer [1995]; Black et al. [2003]).

These behavioral factors change the central challenge associated with providing efficient unemployment insurance in at least two ways. First, in addition to balancing the insurance value of unemployment compensation against the social costs of any associated moral hazard, optimal unemployment insurance with behavioral agents must also consider the welfare implications of decision-making errors or biases (Spinnewijn [2010]). Second, incentives in UI programs may need to account for behavioral tendencies, such as loss aversion or time inconsistency, to operate effectively.

In addition to providing further justifications for existing wage-loss proposals, behavioral factors also generate specific design recommendations. For example, the effects of biased wage expectations or reference dependence with respect to those expectations call into question the likely efficacy of partial wage insurance at speeding up returns to work for displaced workers. These behavioral concerns are consistent with the limited impacts on reemployment rates and job search efforts found in a Canadian demonstration project testing partial-replacement wage insurance for displaced workers (Bloom et al. [2001]). An alternative might be to structure wage-insurance as full or nearly full insurance upon reemployment, and declining over time possibly in a manner linked to typical wage growth patterns on new jobs. Of course, the benefits of such a redesign in mitigating the effects of biased wage expectations and loss aversion must be weighed against the possible costs it might impose on targeting and allocative efficiency.

Limits to self-control represent another likely behavioral barrier to creating incentives to return to work in unemployment compensation programs. A key design challenge for UI is to provide temporary income support while maintaining the incentives of the unemployed to search for and return to work. Procrastination and other expressions of bounded self-control complicate that problem substantially. Findings that the unemployed spend only a modest amount of time per week searching for work are consistent with such procrastination (Krueger and Mueller [2010]). Much suggestive evidence indicates that many job seekers may have time inconsistent preferences and tendencies to procrastinate in job search efforts (Della Vigna and Paserman [2005]; Paserman [2008]). The effects of benefit design on search intensity are thus not simply a product of financial payments from continued unemployment, but instead a more subtle interaction between benefits, incentives, and willpower. Particularly problematic for policy, these tendencies may serve to blunt the force of design features intended to align incentives. The result is that work incentives in programs like UI must address both moral hazard as well as procrastination.

That the work incentive problem in UI may take on this different character when allowing for behavioral tendencies such as procrastination suggests an agenda of experimentation with the structure of unemployment benefits. Demonstration projects experimenting with changes to the UI compensation structure have a long history. Past efforts have included reemployment bonus experiments and a demonstration of Personal Reemployment Accounts (PRAs) (Meyer [1995]; Kirby et al. [2008]). However, evaluations of these efforts have yielded somewhat mixed results. This may be in part due to the failure of these projects to consider factors such as bounded self-control. For example, reemployment bonuses, contingent as they are on possibly distant outcomes such as gaining reemployment and holding the new job for at least several months, may provide little in the way of an effective incentive to individuals who choose levels of search effort day to day.

Labor market policies can play an important role in assisting individuals with the job search process. Basic employment services, such as the coordination of job listings by the public employment services in many nations, seek to provide a public good both to workers and to employers. More active job search assistance policies essentially provide a form of human capital to workers in improving their job search skills and labor market information. The overarching goal of such programs is to help individuals return to work quickly and to improve the quality of matches between workers and jobs.

In the United States a portfolio of interconnected programs supports this goal, and includes informational services as well as active job search assistance. Public employment services -- The Employment Service (ES) -- provide placement assistance to both workers and employers, maintains labor exchange listings, and performs outreach to employers. Services provided under the U.S. Workforce Investment Act (WIA) include counseling and assistance for job seekers. Workers obtain access to these services through multiple points of entry, one of the most important of which are worker profiling referrals from the UI system. These services are supported by employment data and projections complied by the U.S. Department of Labor and its state partners.

In addition, there are behavioral barriers to job search that arise simply due to the fact that it is an intrinsically difficult problem. Optimal job search requires considering information about job market conditions and how they match with personal characteristics in a way that is likely to be difficult for imperfectly rational individuals. Behavioral economics stresses that individuals are limited in the attention and the computational capacity they can bring to multifaceted and complex problems (Tversky and Shafir [1992]; Tversky and Kahneman [1974]; Iyengar and Lepper [2000]). As a result, the speed and quality of employment matches may both suffer due to the tendency of fallible individuals to manage the complex tasks of job search. And programs that assist individuals with managing that complexity can help them obtain work.

3a8082e126