The trouble is that despite its normative appeal, voting (as it is conducted in liberal democracies today) is simply not a very good way for translating preferences into welfare-enhancing outcomes. First, and most obviously, voting in electoral processes yields “all or nothing” results. If Al and Bob are running against each other, the voting public really faces a discrete choice between Al’s and Bob’s list of policies. Folks who prefer Al’s policies on some issues but prefer Bob’s on others are left with little or no way to act on their concern. Similarly, there is no way for voters to express their degree of preference. If 51 percent of voters weakly prefer Al to Bob, and the other 49 percent strongly prefer Bob to Al, normal electoral procedures would result in Al’s beating Bob, even though Bob’s supporters have a good argument for why it oughtn’t be Al who governs them. (As an aside, if Bob supporters could trade with Al supporters for their votes, both groups would be better off in their own estimation. Such deals are usually forbidden by electoral rules.) In addition, the appeal of voting drops off considerably as the voting population grows in size. “One man, one vote” may sound like a noble conformance with the principle of equality before the law, but in practice that equality simply translates into equal powerlessness before the law. Those examples are admittedly oversimplified, but they get to the heart of the problem with voting, and hence with voice.
Far more important than equal participation in the collective decision-making process is the ability to insulate oneself from the process’s harmful outcomes. It matters much less that my voice is heard if there is an easy way for me to get around laws or regulations that impose significant costs. This is typically called “exit” and is contrasted with voice as the fundamental quality by which collective processes ought to be scrutinized. Exit addresses many of the concerns with voting raised above. By opting out of collective outcomes that impose undue harm, it is much more likely that the arrangements I choose not to opt out of command my assent. Exit is far less romantic a concern than voice — which does a lot to explain its relative neglect by political philosophers — but far more effective a solution to the problem of translating individual preferences into welfare-enhancing outcomes at the group level.
While political processes rely heavily on voice, market processes rely heavily on exit. If Safeway stops offering me high-quality food at prices I am willing to pay, it is easy for me to switch to Trader Joe’s. That is because exercising my exit option in market processes is not very costly. In contrast, exit is a highly imperfect check on political power, because the only way for me to exit my polity is to move to another. However, other polities probably don’t look very different from my own in terms of implementing policies that I enjoy. In the current range of political institutions, governments do not face anywhere near the pressure to compete for citizens as private businesses do for customers. That is why so many of a polity’s citizens approve of only a small number of its policies at any given time.
It is possible to structure political institutions to improve the viability of exit. Federalism, the separation of powers between local and national political entities, is probably the best-known example. Unfortunately, in the United States, discussions of federalism and its manifestation in states’ rights has become mistakenly entangled with pro-Confederacy apologias, making an honest public discussion of federalism’s benefits nearly impossible.
That is a shame, because dedication to federalism has the potential to result in welfare-enhancing outcomes for a great number of people. By requiring the majority of political activity to take place at the state level, with the relegation of only a few specifically defined powers to the national government, states could experiment with different policies in an attempt to attract citizens. Progressives who enjoyed heavily subsidized government health care and strict commercial regulation could live in California; conservatives who enjoyed government support of religious institutions and strict regulation of personal conduct could live in Mississippi; and libertarians who enjoyed none of those things could live in New Hampshire. But since all of those states are also part of a larger political union, the costs of moving from state to state would be much lower than the costs of moving from country to country, ensuring that people will continue to be happy with the policies they get over time. The important thing is that, in a truly federal system, it is much more likely that individual members of these groups can all get what they want, without infringing on other individuals’ ability to achieve the same. In contrast, in a system where major political decisions are made at the national level, everybody will have to settle for a compromise with which nobody is happy.
The concern with giving members of a political community equal voice is admirable. But without a corresponding concern for the practicability of exit, collective decision-making processes will continue to be deeply flawed. It is exit, not voice, that will promote welfare-enhancing collective decisions, and hence governance founded more closely on the consent of the governed.