Must democracies hold elections?

We tend to think of elections as a defining feature of democracy. But if the point of elections is simply to find out what people want, why not simply predict what people want? Would our societies be any less democratic if we were to do so?

For the past couple of decades, University of Iowa has been running the Iowa Electronic Markets(IEM), an online platform for information markets (IMs). IMs are markets for trading contracts regarding future or otherwise uncertain events, such as the outcome of political elections. As it happens, the predictions generated by IEM are remarkably reliable (and the same goes for IMs generally). If the price signal on IEM suggests that candidate A will win the US presidential election, for example, A most likely wins.

Since people typically vote for the person they prefer to win, that means that the IEM reliably predicts political preferences (or at least: IEM is no less reliable than are elections on this score). This raises a question: Could we replace elections with IMs? That is, instead of asking people what they want, could we simply predict what they want?

We might worry that this would do away with democracy. David Estlund suggests that democracy is not simply rule for the people—i.e., in accordance with the popular will—but also by the people. Thomas Christiano expresses a similar point when suggesting that democracies require that ‘citizens be active participants in the process of collective decision making’. The worry is that, in the absence of elections, people would no longer be sufficiently involved in setting up government.

But this worry seems misplaced. We value public involvement in the political process because we believe it makes the process sensitive to what people want. That is, we value rule by the people because we think it promotes rule for the people. So the relevant worry is really one about the reliability of predicting what people want without asking them. Let’s consider that worry.

Political IMs are typically settled on the basis of election outcomes. If there is no election, how are the bets placed on the market to be settled? By implementing a self-resolving IM. In such a market, the contracts traded are settled, not on the basis of some external event (such as an election outcome), but by some factor internal to the market. For example, a market could settle its contracts on the basis of the market value at some pre-specified time, unknown to the traders.

If self-resolving markets could be shown to be as reliable as traditional IMs—and in the absence of any evidence either way, that’s still a big ‘if’—IMs could replace elections. But how could we trust that such self-resolving markets are giving us accurate verdicts about the popular will, in the absence of actual elections?

Arguably, on the same grounds that we trust any process when we have good reason to believe it reliable. After all, we typically don’t hold a second election to confirm the outcome of the first—unless the first one registers an outcome radically at odds with what we expect. Similarly, were we to find reason to believe self-resolving markets to be as reliable as traditional ones, we can start to trust such markets to reveal the popular will in the absence of reasons to withdraw that trust.

But maybe the most pressing worry about replacing elections with IMs is not one about technological feasibility but about political legitimacy. On one influential notion of political legitimacy, political arrangements are legitimate only if reasonable people—as in: people sensitive to reason—couldn’t object to them. So might reasonable people object to replacing elections with (self-resolving) IMs?

At present, yes. Again, we don’t have any reason to believe that self-resolving markets are as reliable as traditional IMs. But could it ever be politically legitimate to replace elections with IMs? Say that research a couple of years from now shows that self-resolving IMs are as reliable as traditional IMs. That would give us good reason to believe that self-resolving markets can reveal the popular will. Could reasonable people still object?

No. Consider what they would have to say: while we have good reason to believe that self-resolving markets reveal the popular will, it’s still reasonable to object. In saying that, they would in effect be putting the bar for evidence required in political reform above that of good reason. That’s to require certainty. But if that’s where we’re to put the bar, reasonable people are going to have to object to a lot more than the arrangement currently under consideration. Here’s why:

Consider any given electoral system, E. If of the kind implemented in most developed countries, we have good reason to believe E to reveal the popular will. But can we be certain of it? No. Electoral systems might fail to track the popular will for any number of reasons, including technical problems, low voter turn-out or voter suppression. In fact, if we can be certain of anything, it’s that there will be some discrepancy between E’s output and the popular will, for aforementioned reasons or otherwise.

This means that, if we take the level of evidential support necessary for political reform to be that of certainty, we can run the following argument: if there are reasonable objections to IMs revealing the popular will, then the same goes for any given electoral system.

Denying that there are any legitimate electoral systems seems a high price to pay for being able to reject the idea of replacing elections with IMs even in principle. We might certainly worry that IMs won’t reliably track the popular will—as we might do for any given electoral system. What’s unreasonable is insisting that we must be certain that any political arrangement will track the general will for that arrangement to stand a chance of being legitimate. That’s to set the bar too high.

That means that Estlund and Christiano are mistaken: democracies do not need to hold elections. The type of public involvement characteristic of elections (rule by the people) is good if it means that government reflects the popular will (rule for the people). Moreover, denying that replacing elections with IMs can be politically legitimate would entail that there are no legitimate electoral systems—a Pyrrhic victory if there ever were one. Better, then, to say that, while there might be technological obstacles to IMs currently replacing elections, there are no reasons to believe that they can’t do so in the future.