Walter Bauernfeind’s research while affiliated with Klinikum Nürnberg and other places

We analyse a newly constructed data set of monthly prices for wheat and rye in Nuremberg (1490 1855) to address the issue whether models of rational investment behaviour provide an adequate description of grain markets and grain storage in early modern Europe. Based on the empirical evidence presented here we conclude that the case for well functioning arbitrage in the market for grain in Nuremberg is quite weak. This can only partly be explained by the institutional background.

In our paper we analyze the influence of climatic changes on annual price fluctuations (harvest year) of agricultural products in Germany during the 16th century price revolution. The price series under analysis are from Nuremberg, Cologne, Augsburg, and Munich. The prices are compared with quarterly climatic indices for Germany, which cover the observation period 1500-1599. The main finding is that the length of the vegetation period is an important factor in determining grain price fluctuations. During the climatic deterioration in the 16th century, there is some evidence that the impact of climate (and therefore of supply fluctuations) on grain price fluctuations increases.

Price series of agricultural products are one of the few types of time series available for the description of economic activity in the preindustrial world. Therefore, they have always attracted the interest of economic historians dealing with this period. In addition to newly constructed time series of rye prices in an important economic region in preindustrial Europe, we present corresponding data on grain output based on tithe statistics. The data are analyzed using new univariate and multivariate spectral techniques. The main finding is that the price series exhibit genuine cyclical movements, which can be explained to a large extent as harvest cycles. The cycles vary over the period under consideration. The existence of similar patterns in grain price data developed by others as well as the correlation of prices in various regions of Germany is demonstrated.

In this paper we present a method of describing the cyclical characteristics of a time series in the frequency domain by computing the spectrum of the series. Classical spectral estimation has serious deficiencies if used with economic data. For such short series, Maximum Entropy spectral estimation is more suitable. To demonstrate the method, we analyze the cyclical structure of tithe series from Germany and Switzerland in the period 1339-1708. These series are indicators for grain output. Grains were the most important agricultural products in the pre-industrial economy, providing no less than 70 per cent of human caloric requirements. Hence, fluctuations in output had implications for life and death. We find a robust cyclical structure in the tithe series with a cycle length of about 4 years. These cycles can be found in both the Mid Franconian data and the Swiss data. The coherence between cycles in different regions declines with geographical distance.