Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We have estimated two models explaining political support for the president as a function of economic performance. One model reflects the assumption characterizing most of the literature, that voters indiscriminately punish inflation and unemployment without regard to the tradeoffs between them. The other represents a more demanding conception of voter sophistication about macroeconomic possibilities and about control of the money supply. While the data do not unambiguously designate a ‘best’ model of behavior, the sophisticated voter hypothesis performs as well as the naive voter hypothesis in the context of our study. Clearly this result is subject to the usual econometric caveats including possible misspecification of the model, multicollinearity among independent variables, etc.. However, there is no particular reason to believe that these problems should favor one model as opposed to the other. Thus, while we cannot conclude that voters really think in terms of rates of growth of velocity and natural output, we can question the conventional wisdom that consistent rule-based monetary policies may not be politically viable. Our findings are somewhat reassuring in suggesting that the electoral process does not inevitably involve adverse incentives, as the political business cycle literature suggests that it does. However, the fact that the conventional naive voter model performs comparably well does illustrate that questions about the nature of electoral incentives in economic policy are far from resolved. Our claim is not to have provided a better explanation for patterns of political support, but rather to have conceptualized and tested an alternative to models which suggest that voters are vulnerable to cynical manipulation. Copyright Martinus Nijhoff Publishers 1985