Designing policies to improve the economic well-being of unskilled workers involves the understanding and quantification of all the margins affected by those policies (behavioral responses), as well as of the interactions that may arise from the combination of different policies. Disregarding quantitatively important behavioral responses would yield suboptimal policy recommendations. Also, if, for instance, two policies are complementary the effects of combining them may be very different from the sum of their individual effects. Disregarding policy interactions would also lead to biased recomendations about the policy mix. Our aim is to shed light on the quantitative importance of new margins, on the existence and nature of policy interactions, and on the general equilibrium effects of these policies.

New margins of response. Most models in the literature have focused on understanding the labor supply response of individual workers to changes in policy. In those models, workers are modeled as one-individual households, and are assumed to respond to policy along the intensive margin (i.e. hours worked), as well as along the extensive margin (i.e. labor market participation). In our research we will model workers as two-individual households, where two adults, a male and a female, make joint decisions on labor supply, consumption and savings. Unlike in the bachelor household model, in two-adult households couples engage in intra-household risk sharing, and a new set of margins of response to policy emerges. An important aim of our project is to assess this new and complex array of responses, and to understand how they shape optimal policy to assist unskilled workers. An example of a new margin that arises in the two-adult household setup stems from the fact that most existing income programs to assist low-income married couples contain phase-in and phase-out earnings regions. This implies that the household’s secondary earner (typically the female) may reduce her hours of work, or even to withdraw from the labor market, in order to maximize family income (the sum of earnings and assistance income). This is the so-called added-worker effect, which up to date has been largely overlooked in the literature on the optimal design of income assistance programs. Another example of a new margin we plan to assess is savings. While income assistance programs contain an asset limit for eligibility, most existing models abstract from savings. The extent to which modeling an endogenous savings decision changes the design of optimal tax-transfer policy is still an open question. Finally, in two-adult households with children the trade-off between guaranteed income and earnings subsidies is different than in one-adult households. When workers need to pay childcare costs while working, couples with children enjoy economies of scale relative to one-adult households. These economies of scale shape the labor supply response of couples to changes in the policy mix of guaranteed income and working subsidies.

Policy interactions. A minimum wage and subsidies to low wages are likely to interact with each other. Hence, any analysis on the optimal policy to assist low-income workers cannot consider minimum wages and working subsidies as alternative policies. Rather, the two must be considered jointly, accounting for the interactions between them. An important argument for the existence of complementaries between the minimum wage and work subsidies is that some of the latter may be captured by employers through a reduction in market wages. That is, work subsidies expand labor supply, which results in lower wages in equilibrium. With a legal minimum wage this reduction can be curtailed, thus preventing employers from capturing part of the work subsidies. Answering the question about the optimal mix of a minimum wage and work subsidies requires a quantitative model that can be calibrated or estimated. This is one of the aims of our research project.

Low-wage employment subsidies versus earnings subsidies in general equilibrium. Subsidizing firms for the employment of low-wage workers has been advocated by a number of authors (e.g. Phelps 1994). This is a demand-side policy intended to increase the demand for unskilled labor, and hence its wage. The merits of such a policy, relative to supply-side policies like earnings subsidies, are still far from being understood. We aim to advance towards this understanding with the help of a structural general equilibrium model of optimizing firms and households. The aim of this project is to quantify the effects from these demand- and supply-side policies, and to understand how they contribute to increase wages and employment of unskilled workers. Potential interactions between them will also be explored.