Kelly Ragan, Economy, Handelshögskolan, Stockholm

The project's objective was to evaluate how distortions in the marginal product of labor induced by tax and transfer policies influenced the labor supply decisions of Swedes along a several less well understood margins of behavior, providing a better understanding of labor supply beyond hours worked. The project aimed to understand not only the immediate impact of tax and transfer policy, but also the long run impact of policy through changes in the disutility of work, an equilibrium outcome of the labor supply behavior of others. Quantifying alternative channels through which policy can impact labor supply in the long run allows one to draw a more comprehensive view of how policy affects behavior, as long run responses are arguably the most relevant to policy makers in evaluating the fiscal sustainability of the welfare state.

The project has resulted in five published articles in refereed journals, four articles which are currently undergoing review and revision, two additional articles which are circulating and have been presented at academic conferences and seminars, as well as three ongoing collaborations with researchers at well regarded universities and research institutes in the U.S. (NYU, UCLA, NBER, Harvard University, the University of Chicago GSB) and Europe (University of Bologna and the EUI).

Three Most Important Findings and Two Most Important Publications

The most important findings of the project highlight how welfare state policy can induce behavioral responses along many margins and how behavioral responses may evolve slowly to changes in policy as social feedback mechanisms magnify the short run responses which are often the focus of policy evaluation. The most important results in this regard are highlighted in Ljunge (JHC, 2012) where a model of preference formation common in the theoretical literature is applied to the analysis of sick leave take-up in Sweden. This paper demonstrates how increasing benefit take-up by younger cohorts over time can be largely explained by the transmission of work norms across generations, work norms that are shaped by exposure to the welfare state. Changes in work norms brought about by changes in role model behavior (sick leave use several years earlier by cohorts several years older) are shown to be causal. An instrumental variables strategy uses extreme health shocks in the form of mortality to identify how increased sick leave take-up among one cohort can alter the behavior of younger cohorts. The empirical analysis is the first to quantify the intergenerational ripple effect of welfare state policy on the behavior of the young and makes a persuasive case that the long run fiscal consequences of welfare state policy dwarf estimates based on short-run policy evaluations.

Beyond the long run impact of welfare state policy, often ignored in the treatment effect literature, the project also contributed to broadening our understanding of how tax policy affects labor supply beyond hours worked. Ragan (AEJ Macro, 2013) demonstrates how tax and transfer policy not only affect how Swedes allocate time to the market but also how welfare state policies alter how Swedes spend their time in activities like family care, meal preparation, and leisure. Ragan (2012b) extends this analysis to show how the structure of public expenditures on goods like daycare and eldercare can have sizable welfare consequences, and hence points to how expenditure policy (subsidies versus workfare transfers) interacts with tax policy to influence optimal policy.

Ljunge and Ragan (2012) also has implications for the optimal design of policy, in this case income taxation. The paper makes several contributions, not the least is demonstrating that Swedes labor earnings are highly responsive to tax rates, in contrast to a literature on hours worked which had suggested that Swedes responded little to changes in compensation. The paper estimates large earnings responses to tax prices as well as substantial heterogeneity across different demographic and income groups. The high responsiveness of the earnings of the low income to changes in marginal tax rates suggests that large efficiency and welfare gains could be had from reducing the marginal tax rate at the lower end of the earnings distribution. These results are echoed in Ljunge (2012c) who shows how part of the earnings response to taxes is driven by individuals increasing their use of benefits when tax rates are high and the returns to work are low.

Taken together these papers demonstrate a dangerous dynamic for the future of fiscal policy in the Swedish welfare state. A vicious cycle of growing demands for welfare state benefits leading to higher tax rates may lead to further demand for welfare state benefits and an unsustainable fiscal position. The two sides of the government's balance sheet, revenues (taxes) and expenditures (benefits), have interacting influences on individual's behavior which can create a precarious balance for policy makers interested in ensuring the future of the welfare state.

New research questions and collaborations

New research questions that have sprung from the project are, in part exemplified by the ongoing international collaborations. The study of long term effects of the welfare state in Ljunge (2012a) has started a collaboration with Professor Alberto Bisin (NYU). We set out to estimate a structural model of work norm transmission and behavior in the Swedish welfare state. From the paper Ljunge (2012c) on the elasticity of sick leave has evolved a project on labor supply elasticities. This project is together with Professors Alberto Alesina (Harvard), Andrea Ichino (EUI and Bologna), and Loukas Karabarbounis (Chicago GSB). The project was initiated during a workshop presentation sponsored by the grant. Another workshop presentation of the paper Ljunge (2012c) sparked a collaboration on wealth dynamics together with Day Manoli (UCLA) and Lee Lockwood (NBER and Northwestern).

In addition, Ragan (2012a) has sparked a whole new research agenda on how the birth control pill has affected women's outcomes along a range of margins. Ragan (2012a) studies both current and historical determinants of demand for the birth control pill when it becomes available in Sweden. The paper provides an important first stage that can be used for in turn analyze education, carrier, fertility, and marriage choices of Swedish women. Similarly, the study of the transmission of trust in Ljunge (2011c) provides the basis for analyzing how trust influences a range of outcomes ranging from income to health that are explored in several works in progress. The study of trust complements the analysis of work norms as another important cultural belief that influences earnings and effort choices of individuals.

Publications

Published in peer reviewed journals
M. Ljunge (2012a) “The Spirit of the Welfare State? Adaptation in the Demand for Social Insurance” Journal of Human Capital, Fall 2012, vol. 6 no. 3 (forthcoming).
K. Ragan (2013) “Taxes and Time Use: Fiscal Policy in a Household Production Model.” American Economic Journal: Macroeconomics, forthcoming.
M. Ljunge (2012b) “Cultural Transmission of Civicness” Economics Letters 117 pp. 291-294.
M. Ljunge (2011a) “Increasing Demands on the Welfare State? Trends in Behavior and Attitudes” Economic Studies 57(4): 605-622.
M. Ljunge (2011b) “Do Taxes Produce Better Wine?” Journal of Agricultural & Food Industrial Organization: Vol. 9: Iss. 1, Article 12.

Under review
K. Ragan (2012a), “Sex and the Single Girl: Cultural Persistence and the Pill” under review at the Journal of Political Economy.
K. Ragan (2012b), “The Welfare Effects of Home Sector Fiscal Policies: Quantitative Evidence from a Household Production Model” under review at Economics Letters.
M. Ljunge (2012c) “Sick of Taxes?” under review at the Journal of Public Economics.
M. Ljunge (2011c) “Trust Issues: Evidence from Second Generation Immigrants” University of Copenhagen Discussion paper 11-31, under review at the Journal of Economic Behavior and Organization.

Other related projects/papers
K. Ragan (2012c), “Estimating a Collective Model of Labor Supply: How Did Households Respond to the Tax Reform of the Century?” SSE Mimeo.
M. Ljunge and K. Ragan (2012), “Who Responded to the Tax Reform of the Century?” SSE Mimeo

Ongoing International Collaborations:
M. Ljunge and Alberto Bisin (NYU) “The Welfare State and Social Norms” is a structural analysis spawned from Ljunge (2012a).
M. Ljunge, Alberto Alesina (Harvard), Andrea Ichino (EUI and Bologna) and Loukas Karabarbounis (Chicago) “Asymmetric Labor Supply Elasticities” is directly a continuation of Ljunge (2012c).
M. Ljunge, Lee Lockwood (NBER) and Day Manolo (UCLA) “Understanding the Process of Wealth Accumulation” is a collaboration initiated during a workshop where I presented Ljunge (2012c).