As those of you who listened to our Brexit podcast know, I’m generally opposed to referenda. Allowing the masses to vote on often quite technical matters is a recipe for disaster. Voters rarely even know how policies might impact their own interests, so leaving deeply important and technical decisions in their hands is simply absurd. But this is a fairly obvious criticism and is not the main reason I disapprove of referenda. The real reason I hate things like the Brexit vote is that I believe them to be deeply undemocratic. This might seem a tad counterintuitive to some – referenda allow for direct democracy, after all – but I’m convinced its true.
The biggest issue with referenda is that they are usually a one-off, which is completely different from the system of regular elections that exists in a representative democracy. In the United States, for example, there are legislative elections every two years and presidential elections every four. This means that Americans are never stuck with the current government, as they can always install new politicians in the next electoral cycle. Indeed, that is exactly what they did last November. This is a very important feature of any good democracy because it allows for a peaceful change in power (and policy). If a Democrat is upset with Trump, he or she can feel secure in the knowledge that President Trump cannot cling to power indefinitely; he is constitutionally obligated to face off against his challenger(s) in 2020. Referenda do not work this way. There is one vote, and its outcome is (normally) binding. While one can theoretically hold a future referendum to overturn the first, this is not in any way guaranteed. It’s also difficult to overturn a referendum decision once it has been made. In Britain, for example, those affiliated with Leave lambast those who want a second referendum as “elites” who want to ignore the “people’s” decision (notice the populist framing). If Trump said that about a 2020 election, there would be riots, yet many seem to find this a perfectly acceptable argument for referenda.
Of course, the converse is also true: it is frequently quite easy to hold multiple referenda until one gets a positive outcome (one that results in change). The reason for this is path dependency. Were Scotland to leave the U.K. and develop its own institutions, rules, and standards, it would be extremely difficult to reintegrate. Consequently, when secessionists win a referendum, that outcome is locked in (more or less). Choosing to remain a part of the U.K. doesn’t really commit Scotland to anything, however, except the status quo. The same is true with Brexit or Catalonian secession. This means that referenda are rigged in a way that advantages those who want a positive outcome that creates a sharp break in policy that is difficult to reverse. Their wins are locked in and their losses are just the status quo. In other words, they only stand to gain and, even if they lose, they can just hold another referendum to get the result they want. Given that those seeking a positive result only stand to gain, it is also true that those seeking a negative result only stand to lose. This does not seem remotely just.
This leads to my final criticism, which is that people are asymmetrically motivated about certain issues, giving minority groups influence far in excess of what they deserve. We see this all the time with trade policy. Most people benefit from free trade because it lowers the costs paid by consumers. However, because the greater competition it engenders hurts certain domestic industries, there is a clear constituency that is adversely impacted. We would normally expect the large group of consumers to outvote the small group of producers, but we instead frequently witness large numbers of protectionist measures designed to prop up producers at the expense of consumers. The reason for this is that the losses for producers are highly concentrated while the gains to consumers are much more dispersed and difficult to observe. Thus, producers are far more easily motivated and organized in opposition than consumers are in support. This phenomenon is also frequently the case with referenda, as those that agitate for a referendum usually want it to yield a positive outcome. Consequently, those that want to vote yes tend to be far more motivated than everyone else, and this skews the results toward a particular outcome.
There is a reason there are no direct democracies in the world. While certain countries, like Switzerland, frequently rely on referenda to decide particular issues, even they have a regular legislative process that mediates between voters and policymaking. Referenda sound appealing because there is a romanticized view of democracy in the West: every time an issue emerges, the answer is always more democracy (just look at the brouhaha over the DNC’s superdelegates). But our blind faith in pursuing ever more democracy is completely unjustified and, ironically, actively detrimental to the smooth running of our existing democracies.