What Economy Matters? Contemporaneous and Refined Economic Measures, Perception of Economic Conditions, and Political Evaluation

Many previous studies have found a significant positive relationship between the electorate's perceptions of the condition of the economy and both the public's approval of the president and their division of the parties' shares of the popular vote. One of the questions political scientists have tried to answer is "What exactly are the economic outcomes to which voters respond?" (Kiewiet and Rivers 1984: 371). Measures of economic conditions by the Bureau of Economic Analysis (BEA) are constantly being improved and therefore are routinely changed. This study finds sometimes there are big differences between the contemporaneous and refined measures of economic growth by the BEA. The central question posed in this dissertation is which reading of economic conditions the electorate is responding to in forming their perceptions of the economy and their evaluations about the political leaders presiding over the economy, the contemporaneous economic reports by the BEA, or the real economy as measured by the refined measures of economic growth by the BEA? Ordinary least square regression helps answer the question, especially time series regression techniques of partial difference transformation and full difference transformation, based on data sets consisting of (1) the Consumer Sentiment of the University of Michigan from Economic Research, Federal Reserve Bank of St. Louis, (2) Approval Rates of Presidents from Public Opinion Archives, Roper Center, (3) Popular Vote of In-parties' Presidential Candidate, (4) "Advanced," "Preliminary," and "Final" Estimates of Annual Growth Rate of real GNP from the Survey of Current Business by the BEA (Bureau of Economic Analysis), and (5) Revision Estimates of Annual Growth Rate of real GNP from National Economic Accounts in the BEA. In my findings, (1) both the contemporaneous measures of economic growth and the real economy indirectly influence the electorate's political attitudes and political behavior via their economic perceptions; (2) the electorate responds not only to the economy as the contemporaneous experts at that time describe it to be, but also to the real economy as later experts were able to determine it more accurately to have been; and (3) between the two, the real economy holds a more significant influence over the electorate's economic perceptions than does the contemporaneous governmental statistical index. In implication, it should be emphasized that the limited reliability of contemporaneous measures of economic growth may cause mistakes in presidential election forecasts, and politician and campaign strategy designs.