The Inflation Process: Where Conventional Theory Falters
指出传统经济理论忽视了通胀的传导过程，并分析了原因。作者将传导阶段定义为通胀被普遍预期和补偿之前，认为只有在这一阶段通胀才具有实际影响，但现有理论对此缺乏系统解释。
This paper suggests that conventional economic theory has tended to ignore the inflation transmission process, and sketches some of the reasons why that is so. Harvey Leibenstein's paper in this session is intended as companion piece; it outlines an unconventional theoretical approach to the inflation process.' Inflation just means rising prices. Conventionally, however, economists mean something more. Robert Solow, for example, has defined it as a substantial, sustained increase in the general level of prices (p. 31). But when inflation is sustained at steady rate for long enough to be anticipated, those affected adversely begin protecting themselves against its consequences. Moreover, general inflation implies an absence of relative price changes and their consequent allocative and distributional effects. As economists we agree that inflation has no real consequences once sufficiently long-lived and general,2 yet we definitionally preclude short-run and relative price changes from the purview of inflation theory! For present purposes, the transmission stage is defined as inflation before it is universally anticipated and compensated for. Only in its transmission stage is inflation of practical consequence to anyone, but extant theory has little systematic to say. The one aspect of the transmission process that economists do try to analyze is the macroeconomic price/output problem: how is macro shock to demand or supply divided over time between price change and output change? This problem is far from solved, and is perhaps the central bugbear of modern macroeconomics. Underlying our ignorance of the inflation transmission process are inadequacies inherent in microeconomic theory. Section I briefly discusses some of these inadequacies. Section II illustrates them by looking at conventional treatments of the price/output problem. For space reasons much of what follows consists of bald assertions, and most references have been deleted. The reader is referred for elaboration to the manuscripts listed at the end.