Under- and Overinvestment as a Consequence of Informational Asymmetries
Ulrike Neyer, Martin-Luther-University Halle-Wittenberg-Germany
Objectives
The aim of this paper is twofold. The first is to identify the conditions under which informational asymmetries lead to under- and overinvestment and to determine which factors influence the extent of these inefficiencies. The second is to transfer the results of the theoretical analysis to the transition in Eastern Germany.
Background
In the 1980s and 1990s a vast literature regarding the consequences of informational asymmetries on credit markets on the capital allocation has emerged. But most studies only show that informational asymmetries can lead to inefficient capital allocations. This paper adds to the debate by identifying the conditions under which inefficient capital allocations occur and by determining the factors influencing the extent of these inefficiencies, in order to analyse whether under- and overinvestment have been relevant problems during the transition in Eastern Germany. Investments have been a crucial component of the transition in Eastern Germany. But experience and anecdotal evidence indicate that the process has been plagued by various sources of inefficiencies. This study gives a theoretical foundation for the thesis that under- and overinvestment generated by informational problems on the credit markets have impeded significantly the build up of a new capital stock.
Results
Informational asymmetries are not sufficient for allocational inefficiencies to occur, but the extent of informational problems must be sufficiently high, the rate of internal finance must be sufficiently low and entrepreneurs must operate near the break even point. If these conditions are fulfilled, the extent of both forms of inefficiencies will be generally higher, the bigger the informational problems and the smaller the rate of internal finance are. But whereas the overinvestment rises with an increasing expected profit of the entrepreneurs, the underinvestment decreases in this case. Transferring these results to Eastern Germany there is a strong indication that the transition has been accompanied by substantial over- and underinvestment. Furthermore one can proceed on a specific assumption concerning the risk-return-combination of the investment projects that underinvestment has been the more relevant problem.
Method and Data
In this study two credit markets are modelled. One market is burdened with informational asymmetries, the other does not suffer from this problem. The comparison of possible credit market equilibria shows under which conditions underinvestment and overinvestment occur. A comparative static analysis indicates the variables determining the extent of these inefficiencies. In order to analyse the consequences of this theoretical analysis for the transition in Eastern Germany, statistics of the Deutsche Bundesbank and the German Federal Statistical Office are considered to receive indicators for the magnitude of the relevant variables in Eastern Germany. With this data simulations are conducted.
The set up describing a possible underinvestment builds upon the seminal article of Stiglitz and Weiss (1981), analysing that informational asymmetries can lead to credit rationing. But on contrary to Stiglitz and Weiss in this setup credit rationing cannot occur. For credit rationing to occur a specific "restriction" on loanable funds is a necessary condition. But this kind of "restriction" has not posed a major problem in the transition in Eastern Germany. The overinvestment setup builds upon the idea of de Meza and Webb (1987).
The Basic Story of this Study
There is an informational asymmetrie between banks and entrepreneurs about the specific risk of the projects to be financed. Whereas the entrepreneurs know the specific risk of the projects the banks are only able to classify the entrepreneurs into specific risk categories. The higher the extent of informational problems is, the more heterogenous are the entrepreneurs within a risk category concerning the risk of their project. Since all entrepreneurs within a risk category are charged the same interest rate, there will be a subsidy of the entrepreneurs with the relatively risky projects by those with the relatively safe projects. Depending on the risk-return-combination of the investment projects this subsidy effect can either lead to over- or underinvestment.
If there is a mean preserving spread, the marginal entrepreneurs will be those with the relatively safe projects, which implies that the marginal entrepreneurs are those who subsidise. This means that the subsidy effect will lead to underinvestment, if there are entrepreneurs who cannot bear the additional costs.
If there is no mean preserving spread, but all projects in a risk category have the same return in case of success but differ in risk, the marginal entrepreneurs will be those with the relatively risky projects, i. e. the entrepreneurs who are subsidised. In this case overinvestment will occur, if there will be entrepreneurs who only realise their projects because of the subsidy.
For the occurrence of these inefficiencies there must be a sufficiently high subsidy effect, which means that the informational problems must be sufficiently high and the rate of internal finance must be sufficiently low. In case of underinvestment the expected profit of the entrepreneurs with the relatively safe projects must be smaller than their opportunity costs, which is the risk free interest rate, but without these additional costs their expected profit from the project must be at least as high as their opportunity costs. In case of overinvestment the expected profit of the entrepreneurs with the relatively risky projects must be smaller than their opportunity costs, if it there is no subsidy. But the expected profit must be at least as high as their opportunity costs if they are subsidised.
During the transition in Eastern Germany there have been substantial informational problems between banks and firms, since most of the firms have been very young or just privatised, which means that relevant information has not existed. The rate of internal finance has been relatively low and a big share of entrepreneurs has operated narrowly under or above the break even point. These aspects indicate a strong evidence for substantial over- and underinvestment. If an equal occurrence of both project types (the "mean-preserving-spread-projects", i. e. the underinvestment-type-projects and the overinvestment-type-projects) were assumed, underinvestment might have been the more relevant problem, since due to their weak profitability, it is more likely that East German firms were not able to bear the necessary additional costs than being able to subsidise a significant number of other firms. The results for Eastern Germany are supported by simulations.
Discussion Types of Project
Crucial for the parallel occurrence of under- and overinvestment is the assumption, that both types of projects exist: the "mean-preserving-spread-projects" (underinvestment-type-projects) and the overinvestment-type-projects. Crucial for the question, which kind of inefficiency outweighs, is the share of underinvestment-type-projects compared with the share of overinvestment-type-projects. The question is, whether there is a way of identifying the project types. Generally, existing studies speak for a dominance of the underinvestment-type-projects, since empirical evidence indicate a positive relationship between the rate of internal finance and investments. But de Meza and Webb (1999) show that there might be incentive mechanisms implying overinvestment although having a positive relationship between these two variables.
Government Intervention
The transition in Eastern Germany has been accompanied by a huge number of government programs promoting investments. Considering the setups of this study these programs So the underinvestment but reinforce overinvestment. As long as one cannot identify the project type, which means as long as one cannot assess whether the over- or underinvestment problem is more severe, market failure as a consequence of informational asymmetries does not hold as an argument for those government interventions.
Furthermore even with the underinvestment-type-projects the government programs may lead to overinvestment and/or pure windfall profits (compare Gale (1991)). The same arguments hold for other areas of government programs promoting investments, for example in the financial support of start-ups.
Informational Asymmetries about Other Variables
In this study it is assumed that the informational asymmetries concern only to the risk of the projects, but that all other variables are public knowledg. If it were not for the public knowledge of specific variables, and banks and entrepreneurs had a divergent view concerning the magnitude of this variable, it would depend on the fact who is right, whether there will be an inefficient allocation. If the bank is wrong, a red lining problem, resulting in underinvestment, might occur. If the bank is right, the informational problems will not result in underinvestment (compare de Meza and Southey (1996)).
A further point to discuss is, in what respect the results of this study can be transferred to other transition economies.