Grain storage in early modern Europe

• 成果类型: Article
• 署名作者: University of Copenhagen
• 刊物名称: JOURNAL OF ECONOMIC HISTORY
• ISSN/ISSBN: 0022-0507
• DOI: 10.1017/S0022050700023561
• 发表日期: 1999
• 页码: 762-772
• 关键词: INTERVENTION; MARKETS

摘要

Scholars have long held strong views about the nature and extent of grain storage in early modem Europe. Direct evidence on the issue is quite poor and inconclusive.(1) Randall Nielsen's ambitious attempt at solving the problem in a recent issue of this JOURNAL therefore deserves serious attention.? Like others before him, Nielsen uses inferences from price behavior in assessing the nature of grain storage. Nielsen identifies three distinguishing implications of profit-maximizing storage for grain price structure. First, given the fact-which we shall not dispute-that the underlying harvest shocks are uncorrelated, the existence of autocorrelation in prices is due to profit-maximizing carryover and inventory changes. Second, since storage cannot be negative, the price-stabilizing property of storage is most pronounced when prices are low and granaries are full, Rational storage therefore implies that the distribution of prices will be positively skewed. Third high prices are generated by poor harvests and indicate that granaries are depleted, which will increase society's exposure to harvest shocks. Consequently, variance in future price will depend on present price. High (low) price this year implies higher (lower) variance in price next year. Nielsen tests his hypotheses on three annual price series: English wheat (1356 to 1428), and English wheat and butter (1540 to 1699, 1575 to 1649).(3) However, Nielsen does not use the original raw data but transforms the series so that price is expressed as a ratio to its 31-year moving average for wheat, and a 15-year moving average for butter. We confine ourselves to the early modern data(1540 to 1699) to ease the exposition of our arguments, but our conclusions hold for the medieval series as well.(4) We will concentrate on substantiating the following arguments: Autocorrelation in prices is not unique to storable goods, but also characterizes perishables (that is, goods which cannot be carried over from one harvest to the next); Positive skewness is not a general property in the price series investigated by Nielsen, but appears only in particular transformations of the original series; and Autocorrelation in the price of a storable good is consistent with storage strategies other than profit maximization.