I’ve recently written about some ideas inspired by Daniel Okrent’s book on Prohibition. The policy, which banned the production, sales, and consumption of alcohol in the US between 1920 and 1933, didn’t eliminate alcohol consumption as intended. Not only did it fail in this respect, but it also created negative unintended consequences. In short, it seems like the perfect example of a failed policy.
Bad policies can be surprisingly popular. More alarmingly, they frequently remain popular even after troves of empirical data consistently show their harmful effects. The mid-1930s attitude towards Prohibition nicely fits this pattern: 33% of Americans wanted Prohibition to come back—a mere 4 years after it ended.
One third of society is too large a constituency to be written off as a bunch of dogmatic and misinformed people. More likely, this percentage includes groups with diverse interests, views, and knowledge. If we want to convince people that a policy is bad, we need to understand why they support it in the first place. This post is my attempt to understand the mix of supporters better.
The dogmatist
Dogmatists are people who don’t care about empirical evidence. To them, the question of whether alcohol should be banned is a moral issue. Some of them are like the proverbial family matriarch who opposes divorce regardless of the amount of suffering staying together imposes on the unhappy couple. Their logic is that people’s choices need to be restricted for the greater good. On the other end of the dogmatist spectrum, we find people who believe that everyone should be free to consume whatever they want because their behavior creates no negative externality for others. For obvious reasons, proponents of a ban are much more likely to be members of the first group.
The off-labelist
Unlike dogmatists, off-labelists care about the effectiveness of the policy—but with respect to a different goal than intended. For instance, they might say that Prohibition hasn’t eliminated alcohol consumption, but it was still great because it made, say, citizens more responsible, it improved communities, or it decreased domestic violence. Sometimes, a dogmatist might masquerade as an off-labelist. To pull this off, he doesn’t define in advance how he will evaluate the success of the policy. When the time comes, he looks at many outcomes and picks one that has improved. He then claims that the policy caused this positive outcome, which is why it should be kept or reintroduced. In other words, he doesn’t face policy evaluation with an open mind, but with the preconception that the policy is good and that the task is merely to collect supporting evidence.
The self-interested
The fact that a policy is bad on average doesn’t mean that it’s bad for everyone. Prohibition was great for smugglers, bootleggers, speakeasy owners, criminal gangs, and corrupt officials. On a less cynical note, it was also good for groups we have much more sympathy for. These are people who would have suffered some harm in a counterfactual (non-Prohibition) world. Imagine a housewife who would have been battered by her alcoholic husband; the worker who would have lost his job due to alcohol-related problems; or families that would have been torn apart by an alcohol-fueled cheating incident. We may despise the former group, and empathize with the latter, but both groups would vote for Prohibition anew out of self-interest. It’s important to keep in mind that even the worst policy creates some winners who have a strong vested interest in perpetuating the policy.
The lobbyist
Lobbyists care about the effect of the policy on a particular subgroup, rather than about the policy’s average effect. An extreme case of this is the aforementioned self-interested proponent of Prohibition. In that case, the subgroup the lobbyist cares about is a single person. As the population can be split into groups based on infinite factors, creative lobbyists can always find a subgroup for whom Prohibition seems to have had a positive effect. Wouldn’t it be bad, they would argue, if we exposed twice-divorced, septuagenarian ex-Navy grandmothers of seven to the horrors of alcohol consumption?
The fact that lobbyists like using tiny constituents as a rhetorical device shouldn’t make us cynical about the importance of distributional effects. They matter, so we should always strive to understand how different groups are affected by a policy. It’s not self-evident that every policy should be evaluated based on its average effect. Sometimes, there are legitimate reasons to focus on other summary statistics.
The eternal optimist
Eternal optimists also care about the evidence, but unlike some of the groups above, they don’t argue that the policy was a success. They recognize it wasn’t. Their argument, instead, is that the policy will be successful next time it’s tried. They would say things like “the idea was great, but the implementation was bad” or that “it wasn’t real Prohibition.” Their hope is part social, part technological. In their utopia, the corrupt official and the greedy smuggler will be superseded by the incorruptible public servant and the altruistic entrepreneur. Alternatively, should their social reengineering fail, they can still hope that technological progress will help them rein in recalcitrant opponents of their utopia.
The revisionist
Like their brethren, the eternal optimists, revisionists consider Prohibition a failure. Initially, at least. Over time, however, their memories fade, selectively forgetting the negative aspects of Prohibition, the same way a surviving spouse may forget or downplay the negative qualities of her deceased partner. On a more positive note, the updated belief could also be the result of incorporating new evidence. Perhaps some aspects of Prohibition are unknown to the revisionist at the time, and he duly updates his worldview when he gathers new evidence.
The naïve
The naïve are wrong about the evidence. They mistakenly think that the policy is good. They may be intentionally blind to contradictory evidence, akin to new lovers ignoring even the most obvious signs of their partner’s failings. Alternatively, they may suffer from a focusing illusion, concentrating on the seen at the expense of the unseen. They would point to official statistics, arguing they show no alcohol consumption during the Prohibition years. Busy with this intellectual exercise, they would fail to notice the three speakeasies that popped up around the corner, the visibly tipsy uncle at the dinner party, or the unmistakable smell of a home brewery coming from the neighbor’s plot.
The uninformed
The uninformed are also wrong about the evidence. But unlike the naïve, they may not even realize there is evidence to look at. As such, they evaluate the policy based on how it sounds. Unfortunately for those who care about the evidence, and to the great delight of populists, many bad policies sound good. Populists are quick to capitalize on every new bout of historical amnesia.
Conclusion
People support bad policies for many different reasons. My guess is that the dogmatists, the uninformed, and the eternal optimists comprise the bulk of supporters of most bad policies. That said, proportions shift depending on the context. Prohibition may have been supported primarily by dogmatists, while another failed policy might draw its supporters mainly from another group. One-size-fits-all persuasion won’t cut it when faced with a mix of different supporter groups. We need tailored messaging, which in turn requires that we understand perspectives we may find bizarre, stupid, or unpalatable.