Jack E. Triplett (1982) "Escalation Measures: What is the Answer? What is the Question?"

Economists generally believe that a cost-of-living index is the appropriate measure for escalation purposes, a professional judgment that is documented in--to take examples covering a span of two plus decades--the 1961 report of the Price Statistics Review Committee (Stigler Committee) and the 1982 Price Measurement Review Program Consultations Feedback Report of Statistics Canada. Economists who are specialists in index number theory as well as economists who are primarily users of index numbers for research and policy analysis share this opinion. This paper challenges this view. Escalation, whether in public or private sectors, seldom implies sets of circumstances that correspond to those on which the cost-of-living index has traditionally been defined. Section I reviews the cost-of-living index concept, emphasizing the multiplicity of measures that the general concept implies. The cost-of-living index is an answer to an economic question, of the general form: "What is the minimum change in an economic variable that would be required in order to leave a specified individual consuming unit indifferent between pre- and post-inflationary states?" It has sometimes not been recognized that the literature encompasses a whole family of cost-of-living indexes, which vary with the "economic variable" on which the index is defined (e.g., “expenditure,” “pre-tax income,” “wealth”). Each of these alternative definitions can be thought of as the answer to an economically meaningful question. Thus, there are many cost-of-living index answers to many cost-of-living questions. Each question can be thought of in terms of a compensation for inflation, and each cost-of-living index is an answer that provides an appropriate measurement for some purpose. Section I of the paper discusses these alternative definitions of a cost-of-living index. Escalation of collective bargaining agreements, divorce settlements, social security payments, and so forth is also frequently interpreted as requiring a measure that will compensate for inflation, and this interpretation is sometimes correct. Section II of the paper points out, however, that even when the escalation objective is inflation compensation, the question implied by escalation seldom corresponds to the compensation question that is inherent in any of the traditional cost-of-living index formulations. That is perhaps one reason why participants in escalation arrangements seem uninterested in the economists' concept of the cost-of-living index. The paper develops one cost-of-living like price measure that is relevant to escalation, and discusses the problems of formulating and estimating an “escalation index.”