While rational choice models of Supreme Court decisionmaking have enhanced our appreciation for the separation of powers built into the Madisonian Constitutional design, convincing empirical support lor a Separation-of-Powers (SOP) constraint on justices' behavior has been elusive. We apply a standard spatial voting model to identify circumstances in which Attitudinalist and SOP predictions about justices' behavior diverge. Our reconsideration of the theory indicates that prior efforts to test quantitatively the two models have failed to select out cases for which the two models' predictions do not differ. While our more focused test offers a fairer test of the SOP constraint, the results strongly reject the SOP model. Nonetheless, our analysis provides leverage on this issue by: (1) delineating and executing necessary research design protocols for Grafting a critical test of the SOP model; and (2) rejecting the two exogenously fixed alternative SOP models and suggesting avenues for future research.
Supreme Court justices make choices with the expectation that Court decisions will affect the interests of actors in other political institutions. This is not a oneway street, however. The Supreme Court's decisions may be overturned, modified or circumvented by the legislative or executive branches of government. justices who care about policy outcomes therefore have an incentive to take the preferences of other governmental actors into account in their own deliberations. The logic of the linkage across institutions seems clear, and models that include the Supreme Court as a strategic player on par with Congress and the president have proliferated since Marks's path-breaking work in the late 1980's (Marks 1989; Eskridge 1991; Ferejohn and Shipan 1990; Knight and Epstein 1996a,1996b; GeIy and Spiller 1990; Spiller and GeIy 1992; Spiller and Tiller 1996).
Convincing empirical support for a checks-and-balances constraint on justices' choices has been considerably more elusive, however.1 In this article, we propose a direct test of one version of the thesis of constrained choice. We contrast justices' expected vote choices under a behavioral conjecture of sophisticated spatial voting with choices that would arise under a conjecture of myopic spatial voting. The model is demanding empirically. A complete test of our predictions requires precise measures of the (assumed) fixed alternatives facing the justices, the preferences of the relevant decisionmakers, and the expected responses lo the Courts choices by Congress and the president. Each of these concepts is difficult to measure. We show how current measurement technology allows us to identify only a subset of Supreme Court cases that our theory identifies as providing leverage on our research question. Nevertheless, our analysis comes as close as currently possible to providing a "critical" test of our theory.
The article proceeds as follows. In the next section, we explicate two spatial voting games. The first embodies the non-strategic conjecture, in which Supreme Court justices are assumed to vote on exogenously-lixed case alternatives without regard for the sustainability of the Court's choices. This model of sincere voting behavior is consistent with the Attitudinal model of Supreme Court voting (e.g., Segal and Spaeth 1993). The second model applies the strategic conjecture that justices anticipate the policy consequences of their actions. In this model, justices anticipate how (if at all) legislative actors would modify policy in the near term after the court acted. This model of sophisticated voting is known in the literature as the Separation of Powers (SOP) approach (see Segal 1997). We then operationalize the key parameters of the spatial voting models that allow us to distinguish between Attitudinalist-consistent and SOP-consistent behavior. In the following section, we present and discuss an econometric analysis of justices' voting in cases involving the constitutionality of federal statutes from the 1946 through the 1999 terms of the Court. …