Key points are not available for this paper at this time.
Most economic models falsely predicted a landslide victory for former President Bush in 1992. Rather than winning by the predicted margin of eight to 12 percentage points, Bush lost by a margin of about four. Exit polls tend to eliminate the unique candidacy of Ross Perot as an explanation for the turnabout. This paper examines three factors: (i) changes in the proportion of the population in the military (a proxy for national security concerns), (ii) the number of consecutive terms of incumbency, and (iii) nonlinear interactions between the military variable and standard economic variables. Including these factors in an economic model of the 1992 presidential race avoids predictions of a landslide victory for Bush and also improves the model's overall power and stability.
Haynes et al. (Fri,) studied this question.