Studies in empirical microeconomics

Sammanfattning: Essay I (with H. Sacklén) seeks to uncover the driving forces behind the seemingly irreconcilable results reported by two very influential labor supply studies, Hausman (1981) and MaCurdy, Green & Paarsch (1990). The studies represent sharply different views on the size of the disincentives effects on labor supply and the dead weight loss caused by a progressive tax system. Our findings suggest that the primary differences can be attributed to the use of differing nonlabor income and wage measures in the studies. Monte Carlo experiments suggest that the wage measure adopted by MaCurdy et al. may cause a severely downward biased wage effect such that data falsely refute the basic notion of utility maximization.Essay II analyzes, by means of Monte Carlo experiments, the effects of error in variables and alternative data generating processes on a nonparametric labor supply estimator. The results indicate that the nonparametric method is more robust to altemative data generating processes and that the predictions of tax reform effects are superior to the parametric counterparts. The sensitivity to error in variables, in terms of estimated elasticities, are similar to the effects found in parametric methods.Essay III (with S. Blomquist end W. Newey) evaluates the tax reform carried out in Sweden between 1980 and 1991. A nonparametrically estimated labor supply model is applied to account for tax incentives on hours of work. Some distinct components of the tax reform are identified and analyzed in terms of effects on hours of work, tax revenues and income distribution. The results indicate that the positive effects on labor supply, caused by a cut in marginal tax rates, were counteracted by the increased indirect taxes and redesigned transfer system. An complementary analysis, based on a parametric labor supply model, agrees in general although the predicted behavioral responses are about twice as strong. Essay IV (with A. Lunander) considers an English auction with a secret reservation price within the independent private-values paradigm. A theoretical model is mapped into an econometric specification and the distributions of private values and reservation prices are estimated. The objective is to quantity the seller's expected profit from moving to a mechanism where the reservation price is announced. Simulations indicate that the seller's revenue is slightly higher if a reservation price is announced. However, the average sales price is higher if the reservation price is kept secret.Essay V (with A. Lunander) analyzes the Swedish Road Administration's (SRA) recurrent auctions of maintenance of road markings. The objective is to find an econometric specification that may explain the observed outcomes. In comparison with several tractable specifications we find that the common value paradigm coupled with Weibull distributed signals and non-informative priors of true costs is superior. The result then implies that the SRA may lower their maintenance cost by shifting from the first-price sealed-bid auction to an auction mechanism that exploits the common value property.