Volkswirtschaft - Open Access LMU - Teil 01/03

The standard contest model in which participants compete in a single dimension is well understood and documented. Multi-dimension extensions are possible but are liable to increase the complexity of the contest structure, mitigating one of its main advantages: simplicity. In this paper we propose an extension in which competition ensues in several dimensions and a competitor that wins a certain number of these is awarded a prize. The amount of information needed to run the contest is hence limited to the number of dimensions won by each player. We look at the design of this contest from the point of view of maximizing effort in the contest (per dimension and totally), and show that there will be a tendency to run small contests with few dimensions. The standard Tullock model and its results are encompassed by our framework.

We set up a simple political economy model where economic integration raises the profitability of multinational firms. In this setting redistributive taxation may rise following economic integration, if the effects of the widened income gap dominate the higher excess burden of the tax.

While most market transactions are subject to strong incentives, transactions within firms are often not incentivized. We offer an explanation for this observation based on envy among agents in an otherwise standard moral hazard model with multiple agents. Envious agents suffer if other agents receive a higher wage due to random shocks to their performance measures. The necessary compensation for expected envy renders incentive provision more expensive, which generates a tendency towards flat-wage contracts. Moreover, empirical evidence suggests that social comparisons like envy are more pronounced among employees within firms than among individuals who interact only in the market. Flat-wage contracts are thus more likely to be optimal in firms than in markets.

For the estimation of many econometric models, integrals without analytical solutions have to be evaluated. Examples include limited dependent variables and nonlinear panel data models. In the case of one-dimensional integrals, Gaussian quadrature is known to work efficiently for a large class of problems. In higher dimensions, similar approaches discussed in the literature are either very specific and hard to implement or suffer from exponentially rising computational costs in the number of dimensions - a problem known as the "curse of dimensionality" of numerical integration. We propose a strategy that shares the advantages of Gaussian quadrature methods, is very general and easily implemented, and does not suffer from the curse of dimensionality. Monte Carlo experiments for the random parameters logit model indicate the superior performance of the proposed method over simulation techniques.