Utilitarianism corrupts liberal institutions
Free societies need institutions which do one job
Utilitarianism has the civilizing aspect that it forces people to think about tradeoffs. Promised a golden future, utilitarians ask for an itemized bill, and when Jeremiah cries that all is lost, they reply “how much exactly?” Utilitarianism can be criticized for unconscionably trading away basic rights and absolute values, but more often, it pushes back against the spurious absolutes that come out of political debate, where everyone has an incentive to claim their cause is one to die for. So, utilitarianism seems well-designed for liberal politics. Indeed, that is the premise of public economics. Faced with a policy question, the public economist tots up people’s preferences, weights them in some reasonably fair way, and recommends the welfare-maximizing option.
But some recent examples show how utilitarianism can also pose a danger to liberal societies.
Here is an idea from Yuval Levin:
We now think of institutions less as formative and more as performative, less as molds of our character and behavior, and more as platforms for us to stand on and be seen. And so for one arena to another in American life, we see people using institutions as stages, as a way to raise their profile or build their brand. And those kinds of institutions become much harder to trust.
This brings to mind political grandstanders and self-promoters, and sure, plenty of politicians use the legislature as a way to be retweeted or get on TV (mentioning no names). But a subtler form of the same thing is to use an institution to advance your own moral agenda — your vision of the good.
It’s been a great week for crypto comedy, so let’s start with Sam Bankman-Fried. What’s unusual about the tech billionaire’s fall from grace is not that he lost billions of investors’ money, or that he was trading his own altcoin back and forth with himself to inflate its valuation, or that he admitted that crypto was a Ponzi in an interview earlier this year, or that he speculated with his customers’ money, or that he built a backdoor into his accounting software to let him do that, or that his exchange lost millions to a computer hack. That’s all normal for crypto! What’s unusual is SBF’s motivation. He is famously an effective altruist. Maybe that is just talk, but it wasn’t cheap talk: he channeled huge funds into effective altruism organizations, and is not just a wealthy donor but an organizer in the movement. In fact, he may have been inspired by a utilitarian philosopher to get rich so as to be an altruist:
Over lunch a decade ago while he was still in college, Mr. Bankman-Fried told Mr. MacAskill, the philosopher, that he wanted to work on animal-welfare issues. Mr. MacAskill suggested the young man could do more good earning large sums of money and donating the bulk of it to good causes instead.
The risk here is pretty obvious. If you are just greedy, you might be satisfied with a yacht, a private plane and some nice girls — small change, in the big picture. Improving the world, OTOH, takes real money. The needs are limitless, and the great good you could potentially do overrides less important concerns, like trusteeship of your customers’ assets.
From SBF to GOD: Gus O’Donnell, one of Britain’s most senior civil servants, notoriously said “I think it’s my job to maximise global welfare not national welfare.”1 This is sound philosophy! Someone in Burundi should count for as much as someone in Birmingham. But doing good around the world is not a civil servant’s job.
There is a similar flavour to some of what health professionals did during the pandemic. In the US, the CDC and Surgeon General took flak because they seemed to make public pronouncements based on judgments beyond thir expertise. For example, the CDC originally recommended against wearing masks, and this may have been to conserve masks for health professionals. That’s reasonable logic. But again, maximizing global welfare is not really the CDC’s job.
There’s more than one trust issue here. Modern organization requires people to play roles that are necessarily partial. If you think everyone else is doing their job, it is easier to get on with yours. On the other hand, if you are constantly worried that the boss might go on TV and recommend Ivermectin or bleach, then of course your responsibilities get much bigger. Surely these possibilities lay heavy on Dr Fauci’s mind. But conversely, when specialists overreach, they put their reputation at risk with the public, because their judgment calls might be based less in their expertise, and more in their own social or political views. The same issues come up when fund managers weigh ESG in their decisions, or when scientists try to shape the research agenda towards the “right” political questions.
A last example comes from central banks. Traditionally understated and boring, central bankers have moved into the spotlight since the financial crisis. They’ve started to use a wider variety of policy instruments, and at the same time they have almost naturally started to think about wider dimensions of policy. Controlling the money supply was fine, but shouldn’t they also be concerned with social equality, moral values and global warming?
Utilitarians are particularly at risk of this kind of overreach, because it fits easily with the structure of their ethics. As a utilitarian, you are trying to maximize global welfare. Everyone else is just doing what they do, perhaps their motives are selfish, perhaps they’re altruistic, but from your point of view they are just (a) billiard balls who react causally on other people and (b) repositories of welfare. You want to make them happy! But in the same way you want to please your dog. They aren’t really moral agents. Well, maybe they can correct you about facts, but they can’t correct your ethics.
Limited institutions with limited goals
What has this got to do with liberalism specifically?
One view of liberal societies goes like this: people are following their selfish ends, and the job of liberal institutions is to bring those ends into harmony. One classic institution for this is the market. We bring our desires to market, and the price mechanism lets us satisfy them in a way that is efficient (under well-known restrictive conditions). Elections and voting are another. This view of institutions as harmonizing different ends is embodied in economic models like the workhorse Meltzer-Richard story in political economy. The rich want lower taxes, the poor want higher taxes, the median voter wins out, it’s not perfect but it’s reasonable.
This view underestimates the difficulty of the problem that liberal societies try to solve. The history of liberalism is not that we were selfish and then we invented (or muddled our way into) institutions to harmonize that. Liberalism grows out of a reaction to the bloody religious — so, ideological — conflict of the 16th and 17th centuries. It is conflicting ideals, conflicting visions of the good society, that liberal institutions must cope with. This is why the growth of religious toleration is so central to their story.
Being a bit tendentious, you could even argue that as ways of dealing with human selfishness, Calvinist Geneva or Puritan New England are better models than liberalism! After all, if public goods and externalities are pervasive, then we can’t just let people be selfish… maybe their ends need harmonizing more forcefully, via vigorous normative pressure.
But, as the SBF story suggests, the destructive power of selfishness is nothing compared to the devastation that idealism can wreak. Norms and ideals are much more powerful coordination devices than people realize. They can work when institutions are absent, or even overturn those institutions. Selfishness makes Russian soldiers steal washing machines. The ideals of Putin and Putin’s people sent them into Ukraine in the first place.
Liberalism’s toughest challenge, then, is to get groups with deeply differing beliefs and world-views to co-exist, without any group feeling so offended that it seeks to overturn the social order.
A central requirement for this is that liberal institutions must be limited, not just in their powers, but in their goals. They must not seek to impose any one set of ideas on the whole of society. Without this requirement, different groups have no reason to accept their power, and that is a recipe for conflict. On the other hand, when institutional goals are limited in scope, and command a broad consensus, then different groups with different worldviews can agree to accept them.
Everyone who accepts to be governed democratically can agree that we need a civil service. The civil service must be bound by impartiality to serve the government of the day. Individual civil servants can hold whatever ethics or politics they prefer, but when they act for the institution, they must put those views to one side.
Everyone who wants scientific and technical progress can agree that we need scientists, whose job is to find out how the world works. Scientists and scientific institutions command acceptance and respect because they aim for that limited goal. If they use that acceptance to further broader goals which they happen to believe are important, even vital, then they will diminish it, among groups who don’t share those goals. The appropriate ethics for science are scientific ethics only.
If we want a free society, we’re going to need journalists to tell us the truth. It is a basic part of journalistic ethics, and also of being a successful hack, that when you get a scoop you have to publish it. Doesn’t matter if it doesn’t support your side! We all know that newspapers have their political biases, but we do not expect them to be run by them.
We all need sound money. There is an idea from macroeconomics that we can ensure sound money by hiring a central banker to manage the money supply. The central banker should be “conservative”, that is, biased towards tighter monetary policy. That way, he can credibly commit to keeping money sound in future. Obviously, if central bankers were to acquire a wider range of policy instruments and meddle in broader social issues, then it would not be appropriate to hire someone with that bias — or any other. Broader social issues must be judged by the whole society. Traditionally, that is politicians’ job.
This idea is political philosophy, not moral philosophy. If you want to have a free society, your institutions had better be run according to these principles, because in a free society people have deep disagreements. I don’t claim these are the best institutions. Presumably if you are a utilitarian, the best institutions would all try to maximize utility. If you are a Thomist, they would implement Catholic integralism or whatever. It’s precisely because they aren’t the best, according to any one world view, that they can survive and preserve the precious form of a free society.
There is a broader vista opening up, which is that “liberals” of the recent past have ignored this basic principle, and instead tried to use all society’s institutions to implement deepseated changes which they believed were necessary to fulfil liberalism’s promise. You might say that post-60s liberalism murdered liberalism and has been walking around in its skin (to steal a metaphor), and that this is why (a) liberalism is very popular, (b) liberalism makes many people angry, and (c) liberals don’t understand why people are mad at them. But for now let’s focus on utilitarianism.
It would be bad if society were taken over by crazed Thomists. But utilitarianism is a more present danger. It’s more plausible to, oh say, clever but ethically-challenged young billionaires, and to elites more generally. It is also ensconced as the kind-of-official ethical theory of the most influential social science, which has an important line in institutional analysis. But those models don’t take into account the long-term risks of rejigging social institutions to suit one particular version of the good.
Not a benevolent dictator, he’s a very naughty boy
A utilitarian might reply that utilitarianism weighs everyone’s welfare in the most scientifically valid way possible. Sure, you don’t get your ideal vision of society, but the calculation has taken what you want into account, and why should you get more than anyone else? This would be great if it were true. In fact it very much underestimates two things: the frailty of our models, and the human weaknesses of the people who implement them.
First, we simply aren’t very good at totting up people’s welfare. We can count money and production. We’re much worse at estimating the value of, say, health. And for subtler things, like say the value of natural beauty, our best approaches come down to either asking people; or estimating them from the market prices of other stuff. Neither method works. People notoriously don’t know what they want in surveys; on the other hand, if you ask the market how much to value, say, the future, it might discount it at 5% per year, which is nuts.
Second, people with a highly rational theory do not become highly rational people. The online rationalism community is kind of an extreme test case for this. They are all very smart and critical! They write articles like “make sure rationalism doesn’t become a cult” and “test everything, even rationalism!” And of course they are absolutely a cult, complete with emotional drama, gurus, turgid religious documents (“if you still don't like it after Chapter 10, give up”), and visions of the apocalypse. Nobody is madder than somebody who has really thought things through.
This means that really existing utilitarians, who try to implement real utilitarian plans, are not especially likely to maximize utility. They may do some good, even a lot, but that is because they are an idealistic movement, not because of their theory. (The Knights Hospitallers, the Boy Scouts, and Greenpeace have also all done some good in their time.) They are also capable of doing much bad, because, as stated above, that is what ideals and norms can motivate people to do.
I started off by saying utilitarianism is good at thinking through trade-offs. That is still true. The cost-benefit calculations for, say, the High Speed 2 railway almost certainly miss more than they include, but it’s still better to try to do them than not to bother! Politicians should think of this work as a good servant, but a terrible master. There is also a variety of utilitarianism called Rule Utilitarianism, which relates to some of my points. Rule Utilitarians don’t try to maximize utility with every act, instead they follow the rules that would maximize utility if they always follow them. For example, they might have a rule of not paying ransoms, because that will prevent ransom demands being made. In the same way, utilitarians ought to take into account the long-run value of liberal institutions, and keep their beliefs for their private life. Meanwhile, the personnel of the great social institutions should understand the unique values that animate them, and defend those values against people who want to reshape them to serve their own, however noble, purposes.
If you enjoyed this, you might like my book Wyclif’s Dust: Western Cultures from the Printing Press to the Present. It’s available from Amazon as a paperback/hardback/ebook, and you can read more about it here.
You can also subscribe to this newsletter (it’s free):
And, if you liked it, share this post on social media:
Quoted by David Goodhart in The British Dream.