Policy in coalition governments (a) depends on negotiations between parties that (b) continue between elections. No extant means of predicting policy—bargaining power indices, vote shares, seat shares, polling, veto players or measures of electoral competitiveness—recognizes both of these facts. We conceptualize, estimate and validate the first dynamic measure of parties’ bargaining leverage intended to predict policy and politics. We argue that those parties with the greatest leverage in policy negotiations are those with the highest probability of participating in an alternative government, were one to form. Combining a large set of political polls and an empirical coalition formation model developed with out-of-sample testing, we estimate coalition inclusion probabilities for parties in a sample of 21 parliamentary democracies at a monthly frequency over four decades. Applications to government spending and to the stringency of environmental policy show leverage from coalition inclusion probabilities to be strongly predictive while the primary alternatives—vote shares, seat shares and polls—are not.
