|
Financial
Models A number of recent financial models have been constructed as models of asset returns that are either entirely disconnected from the real side of the economy, or have only a small connection with the real side through the relationship between asset returns and aggregate consumption, a relationship emphasized in the Consumption Capital Asset Pricing Model (CCAPM). In Discussion Paper No. 1353, Research Fellow Jean-Pierre Danthine and John Donaldson argue that certain features of the real side of the economy may have an important influence on the characteristics of asset returns. The common methodological basis of the CCAPM on the one hand, and the majority of the new business cycle models on the other, make it possible to take account of the impact real market ‘frictions’ may have on the asset side of the economy. The starting point is an asset pricing model in an economy with a frictionless Walrasian production sector. The results are disappointing and the authors diagnose the source of the model’s failure at this stage to be the equality of the return on physical capital and the return on the financial asset. The authors argue that the inclusion of institutional features of the real side of the economy is necessary for breaking this link. They then enrich the model by introducing the costs of adjusting physical capital. In the next step they introduce financial leverage – that is, corporate debt – and lastly they include labour contracts. The results of these modifications are striking. There is now a significant equity premium. The variability of stock returns is multiplied more than twenty-fold. Equity and risk-free assets start to behave very differently, more in line with real world observations. The ratio of stock market value to GNP exhibits large movements comparable to what is observed. The authors show that if all three modifications are introduced simultaneously, the variability of stock returns takes a value which is almost identical to the number generated for the US: there seems to be no trace of excess volatility of stock returns in this model economy peopled with fully rational agents whose actions result in a Pareto-efficient equilibrium.
Discussion Paper No. 1353, February 1996 (FE/IM) |
|