Vol. 7, No. 2
Uri M. Possen and Mikko Puhakka:
Rationality of limited rationality: some aggregate implications (pp. 83–93)
In this paper we let economic agents choose whether to become fully rational or stay boundedly rational. Boundedly rational agents are less sophisticated in their information processing abilities. It is costly to acquire information needed to become fully rational, and thus not all agents are willing to incur those costs. We then explore the aggregate effects of endogenizing the decision whether the agent should or should not become fully rational in handling information. Since fully and boundedly rational individuals make different saving and consumption decisions, their interaction has a direct impact on the size of capital stock, aggregate savings and the volatility of output. E.g. we are able to show that in many circumstances our model will imply a smoother consumption than would be observed in a model with only fully rational Consumers.
(JEL: D81, E21, E22)
Risk aversion, multivariate proxies and the behavior of asset returns (pp. 94–107)
An asset pricing model with constant relative risk aversion (CRRA) is tested with data from Sweden for the period 1977-1990. As a proxy for consumption growth, we employ a return based mimicking portfolio. We find that significant structural shifts in the model parameters occur between 1977-83 and 1984-90. The results indicate that the goodness-of-fit for the model in the two subperiods is fairly good and the estimates of CRRA seem reasonable. However, contrary to the cross-sectional implications of the model, CRRAs implied from individual assets differ a lot from each other.
Testing weak exogeneity by multivariate cointegration techniques: the demand for labour in Finnish manufacturing (pp. 108–119)
The study examines the efficiency of the single-equation approach in modelling employment determination in the context of Finnish labour market data. The task is accomplished by using the multivariate cointegration techniques developed in Johansen (1988) and Johansen and Juselius (1990). The results suggest that the weak exogeneity of the conventional conditioning variables, including output, capital stock, wages, product prices and material prices cannot be taken for granted. The possibility of the loss of information involved in contemporaneous conditioning implies that care should be taken in applying single-equation methods to Finnish labour market data.
(JEL: C10, J23)
Time-dependent effects of unemployment benefits (pp. 120–129)
This paper presents methods to estimate the effects of time-dependent covariates in parametric duration models using microeconomic data. In the Finnish system persons who are eligible for the benefits have a risk to lose them after the first three months. In addition the earnings-related unemployment allowances decrease 20 per cent after the first 100 days unemployment. Using Finnish data it is shown that unemployment benefits have a negative effect on the probability of becoming employed during the first months, but after that period the effect vanishes.
An n-person bargaining game (pp. 130–132)
Rubinstein type bargaining is extended to n-person case with linear utilities. The game possesses a unique subgame perfect equilibrium in which evelybody’s share is equal as impatience goes to zero.