The well-known tradeoff between participation and decisiveness extends into the realm of representation. The more voters can hold their representatives accountable for policy outcomes, the more difficult policy making—particularly policy making requiring difficult decisions—becomes. The result is the degradation of government’s ability to serve the general welfare by, for instance, regulating behavior that leads to a tragedy of the commons (Hardin, 1968). Zakaria (2007) links this failure of elected government to exercise policy leadership to technological advances that allow politicians to receive near-instant feedback from public opinion. In essence, voters who can easily hold politicians qua central agents accountable will punish those agents for prioritizing the collective welfare over their individual interests. The problem is particularly acute for medium- and long-term policy initiatives, as the less immediately visible are positive policy outcomes, the less chance that voters will look kindly on the sacrifices they have to make to achieve them. The basic problem is that democratic accountability makes policy makers less willing to take the kinds of policy risks involved in innovation, experimentation, or even long-term investments or other far-sighted policies. Weakening central agency thus amplifies long-term societal welfare loss, transforming policy making into a common-pool resource problem. The problem arises because parties can’t take full credit for policies whose benefits extend to other parties’ voters, and because they can share with their coalition partners any blame for the costs of targeted policies that benefit only their own supporters. I argue that multiparty coalitions counter the common-pool resource problem of public-goods provision. The key, echoing Demsetz (1967) and Shepsle (1979), is that parties can expect to receive all the credit (or blame) they are due for policy in any issue areas that are seen as their responsibility. Blame shifting and credit claiming are feasible only when coalition partners’ responsibilities overlap, something that is more likely the more policy areas coalition parties control. Since coalition parties will as a rule control fewer ministries the more parties there are in government, all else equal, it follows that multiparty coalitions should be better equipped to make policy for the long term the more parties they include. Two-party coalitions, on the other hand, are qualitatively different from their multiparty counterparts and should perform particularly badly when faced with hard decisions.
