Monday, 22 May 2017

Libertarianism in 2017

A friend who is involved in politics wrote to ask me if I was still a libertarian. This deserves a longer answer.


In any given dimension, history looks like this:

Things fluctuate, getting better or worse at different times. Humans being what they are, people who think that they are on a downward slope see history like this:

They project current trends backward to a golden age that never was, and forward to a horrid but unlikely dystopia. Those who believe things are improving see this –

 – denigrating the past too much, and assuming that the future will be glorious so long as we keep doing whatever we are doing.

So, the socialists of the nineteenth century saw the injustices and crimes of capitalism, the growth of municipal provision in the burgeoning cities, and the self-organization of the workers; and by extrapolation, they imagined that in future everything could be provided by the state, and run by the proletariat.

Libertarianism emerged in the 1970s. Western societies seemed on a downward slope. The state was growing but bureaucracy was inefficient. The economy was overtaxed and sluggish. Politicians seemed powerless in the face of industrial unrest. So, the libertarians projected a dystopia. The tax rate, wrote one Public Choice economist, was “tending towards unity” (i.e. 100%). And they dreamt of a utopia in which almost nothing would be provided by the Leviathan state, and there would be “markets in everything”.

Libertarianism was a relatively harmless utopia, compared to the blood-drenched history of socialism. Really Existing Libertarianism, called by its enemies neoliberalism, sometimes damaged people’s lives; but its greatest achievement, the economic transformation of China, led hundreds of millions out of poverty. (Yes, neoliberalism does deserve credit for this. Certainly Chinese policy never got close to what Milton Friedman or Friedrich Hayek would have wanted, but nor did policy anywhere else on earth. The transformation of China was part of the overall movement towards markets in the late twentieth century, and neoliberal ideas played a part.)


Still, libertarianism was a utopia, and it shared the silliness and unrealism of all utopias. Any economist can list the ways in which “markets in everything” will fail:
•    Public goods, like defence and a clean environment, will be underprovided by the market. State coercion, forcing us to purchase these goods via taxation, can benefit anyone who values them; as Rousseau might put it, he is thereby forced to be free, or at least, forced to be a lot better off.
•    The same goes for other market failures, like health and social insurance (moral hazard, adverse selection).
•    Nowadays we can add “behavioural” failures like addiction, economic short-sightedness and susceptibility to framing. Plus, with nudges, we may be able to solve these without infringing people’s liberty.
•    All of these goods could be provided by other forms than the democratic state. But the democratic state is “what’s there” and a reasonable format for providing them, so good conservatives should start with it.

On the other hand, just as socialists’ critiques of capitalism were not invalidated by the failures of socialism, so libertarians’ negative points still have plenty of force:

•    Monopoly is inefficient. State monopoly, with a soft budget constraint, is especially so.
•    Bureaucracy is often intrusive, ineffective, or both at once.
•    There are good reasons to expect the state to be too large, and to have inbuilt tendencies to keep growing.
•    The democratic electoral process discourages long-term thinking, and there is little evidence that it can keep government indebtedness in check.

This much, I think, is commonplace. Overall, then, extreme libertarians – like Peter Thiel, the tech boffin who thinks we should float around the Pacific in reconfigurable pirate yachts – are bonkers; but practical libertarianism is a sensible orientation. France and Southern Europe, for instance, still really need more of it.

Markets and cultures

But all of this is just hedge-trimming on each side of yesterday’s big ideas. The social and political upheavals we are living through need new ideas, not just at policy level but at a deeper level that informs policy. So I will try to write some down – eight, to be exact – and  see where libertarianism fits in.

Charles Murray, a social conservative, and Robert Putnam, an impeccable left-liberal, recently published two books saying basically the same thing. Both point to increasing inequality in America; both link it back, not to economics, but to patterns of family and community breakdown. For journalism on the same topic, read Dreamland by Sam Quinones on the heroin and opioid epidemic. (2015 saw more US deaths from drug overdoses than from car accidents.) Or read Case and Deaton, two economists, on ‘deaths of despair’. As the New York Times recently suggested, America doesn’t need economists;  it needs sociologists.

This is about more than subject matter. The difference between sociology and economics is that while economic man calculates the optimal means to pursue his goals, sociological man is influenced by what other people say and do: he follows “norms”, or “scripts”, or picks from “repertoires” of action.

Libertarianism started from economic man. Its point was always that people (bureaucrats, politicians, welfare claimants) were responding to incentives. The observation was correct, but the theory was wrong.

What leads people to respond to incentives? Why does my local curry house sell me a balti for £8 (so I buy it) and not £80 (when I wouldn’t)? Profit maximization seems like a good explanation. But look more closely and this idea falls apart. For example, the really good curry house in Stoke Newington sells me delicious new dishes from the Indian South. The owner has achieved fame and fortune and has opened several new branches. My local could do the same… but it is still peddling the same old same old. Everywhere you look, instead of hunting for opportunities like sharks, people settle into well-worn routines. The sociologists are right:

1.    Humans are creatures of habit.

In fact, the economic theory of man was always held as an “as-if” explanation: smart economists always treated rationality as a useful simplification, not the true theory. Also, as it happens, the theory was not self-consistent. For, suppose that everyone around you were always questing for the optimal thing to do; should you do the same? No. You can just act like sociological man, copy whatever your friends come up with, and save yourself the effort. (This idea has long been formalized in the theory of finance, where it shows that stock markets will not in general be efficient, since people will free-ride on others’ knowledge: momentum trading is a real-world example.)

2.    A few people innovate, most people copy.

Let’s apply this to two classic arguments. The first is about welfare. The economic critique of welfare benefits was “people respond to incentives”. If you pay them only when they are unemployed, then you disincentivize work. In the 70s and 80s, when unemployment took off to new highs, this argument really started to bite. And papers from then indeed show that if you reduced the length of time people claimed benefits, they found work faster. Put more crudely, unemployment benefits made people lazy.

But it seems as if this was not such a problem in the 50s and 60s when the welfare state was being built. Perhaps politicians were blind to the issue then? Maybe, but it also seems not to be such a problem in the latest recession.

Why might people’s behaviour differ between these periods? Here is a story: in the 1950s, if you lost your job, you got on your bike and you looked for work. (For non-UK readers, this phrase is from the blessed Norman Tebbit.) That was a sensible thing to do. The coming of the welfare state changed incentives, and people responded – but not immediately. It took a generation for the culture to change and for it to become acceptable in some areas to be on the dole long term. Then, after the welfare reforms of the 80s, it took another generation for the culture to shift back.

Another example is inflation. The standard story used to be that the Keynesians assumed inflation could always be a surprise; the monetarists understood that people would eventually get used to it. Well, it turns out that Keynesianism still has a lot of empirics going for it: the inflation-unemployment tradeoff seems really to exist, for instance. Yet, the failures of 1970s Keynesianism seem real too. Again, this makes sense if people are not ultra-rational but adjust their views more slowly. Perhaps most people just accept the conventional wisdom of the day about inflation, without thinking about it too deeply; only a few do their own research, and may be faster to spot a change in “regime”. A few innovate, most copy.

What difference does this view make? Well, when choosing institutions and policies, we should expect that people’s behaviour will change to fit them only slowly. In particular, this means being skeptical about "evidence based policy" which tends to look at only the first few years of any change. (Statisticians like to look at the exact point a policy changed, so as to get “clean estimates” without all that awkward history getting in the way.) Everything we do should ex ante preserve incentives for good behaviour, because by the time we have destroyed them, it will take a long, long time to get back. Lessons here for those concerned with the welfare state, or equally with regulation of financial markets.

But let’s go deeper.

People are creatures of habit; and they copy others more than they innovate. Listen, copy, repeat. Obviously, most of the listening and copying happens when people are young. Evolution has made children quicker learners than the old, because children have their whole lives ahead to benefit from the learning.

A second point. I have talked so far as if people had goals, and learned how to achieve them. But the unlearned organism, the baby at the breast, has no goals, only needs. In fact, as well as means to ends, we learn ends themselves, or in economists’ terms, preferences. The capacity to plan itself is acquired by learning. Putting these points together:

3.    Children learn their preferences.

It is great that the human neonate brain is so flexible, but it brings some problems. The market can efficiently satisfy our preferences, whatever they are. Unfortunately, that includes the preferences that have left 25% of us obese; desires for addictive goods like tobacco, alcohol and illegal drugs; or other things that we want but do not actually make us happier or better off. I hope it is not too controversial to suggest that in modern markets, goods like this are legion.

So, market efficiency is not the same as social welfare.

4.    It matters what preferences we have, as well as whether they are satisfied.

The insights of the neoclassical welfare theorems about market efficiency are important. But they are not a complete social theory, and would not be even if there were no public goods or market failures. Markets are preference-delivery systems. Cultures are preference-creation systems.

What kind of learners are children, or indeed people in general? Would it surprise you if I said “undiscriminating”? The point of copy-and-repeat is to save the effort of thinking, so it would not be surprising if the copying process itself was pretty thoughtless. That turns out to be true. Even adults basically believe what they are told, in most circumstances. Children still more so. This means that

5.    The human brain is hackable.

In the societies almost everyone lived in until the last century, there was not much potential for hacking. You learnt what you needed from those around you – other people in the band, tribe or village. But that is no longer true today. People have many more ways to learn than they ever did, they live in more complex societies, and there is a lot of profit in the business of persuasion. This means there are many more people we can learn from, and who have an interest in teaching us, than there are people who will have our interests at heart as teachers. In fact, usually, just two people will be in the latter group with reasonable certainty:

6.    Parents are white hat hackers.

What distinguished Western societies from the rest of the world over the past centuries? Feng Guifen, the 19th-century Chinese scholar-bureaucrat, considered this as he watched the British achieve dominance in his country. He came up with four answers: better use of labour power, superior agriculture, control over their rulers, and “the necessary accord of word with deed”. The first two are economics, the third politics. The fourth is a difference in culture. This difference did not come about by chance; it was the product of an intensive process of cultivation.

Ever since the printing press, Westerners have been cultivating their children’s preferences in a way that nowhere without print could match, in the cultural equivalent of a hothouse. It started off in circumstances that are closer to a contemporary religious cult than to today’s mainstream churches. Later on, it cooled down a bit, but the effort of the bourgeoisie to educate its children – and its crusade to remake society in its own image – remained very intense. It encompassed both families and schools, which focused on character before even the most basic technical achievements.

Over the past fifty years, we’ve scrapped all that. We put our kids in front of the television; our schools focused on technical achievements; and without meaning to, we messed our family structure up spectacularly. What happened next? I think Charles Murray, Robert Putnam and Dreamland are telling us the answer.

7.    Culture abhors a vacuum.

If society does not take care of the job of preference formation, it is not that the next generation ends up unsocialized. That would be unnatural. Instead, they socialize themselves: they copy the successful. That’s fine, so long as they don’t learn to succeed at the expense of society. Unfortunately, no society and no institutions can prevent this possibility. People, from the local drug dealer to Philip Green, will find socially harmful ways of getting rich or famous.

Complaints about culture are perennial, and did I mention about looking back to golden ages? However, the evidence that culture matters, and that our societies have problems in that regard, I think is quite strong. Complaining about consumerism used to be done by Marxist theorists. Now we have the Nobel prize winner  George Akerlof and Robert Shiller making the same point in Phishing for Phools.

By now, we are far from libertarianism. Where might libertarian ideas might fit in to this world view? Consider that many modern calls for regulation come from “behavioural” failures like obesity or failure to save for retirement. These have been psychologized as universal tendencies, but I don’t think we know that. They may be failures of culture. Let’s put it like this:

8.    A free economy is the outcome of a free society.

In other words, if we get the preferences right, then we can often trust the free market to deliver. If we don’t, then we can’t, and we will face calls for governmental regulation, and in fact a whole new, behaviourally-based regulatory agenda.

Now the government does have a role to play. But I do not think it should be a leading role. Instead, I’d go back to Thatcher and say “there is no such thing as society; there are only individuals and their families”. Culture, of the kind we need to focus on, starts with families and grows out from there.

In this respect I think it is worrying that Britain, unlike the US, has a strong Conservative party but no conservative movement. A lot of what we need to do now is not fundamentally about changing policies or institutions; instead, it is about changing ourselves: people, families, groups, communities. The Americans have a large, diverse movement – with its share of charlatans and nutters, I admit – who get that. We don’t.

Political afterthought

These are the ideas that I think should inform the political agenda of today. You may disagree. It is still not hard to make a case that all that matters is “peace, easy taxes and a tolerable administration of justice”, as Adam Smith said.

But certain realities will inform the political agenda. The anger of poor people in rich countries is likely to be long-lasting; they face genuine economic threats from trade and migration; redistribution and protectionism will grow their appeal in coming years. Since redistribution and protectionism harm the economy, we face a nasty choice: countries which continue to “come apart” into comfortable rich and angry poor; or populism and economic stagnation. A standard response – the Blairite response, for example – is to call for investment in education, so as to create skilled workers “able to compete”. I agree, but the question is, what education and what skills are needed?

Humans clearly differ in skills. Genetics tells us that quite a lot of these differences are inborn. The optimistic story, in which everyone is capable of learning advanced skills for the modern economy, may be a fairy tale. There are some basic skills, though, that seem to make a lot of difference to employment. These are what the great Chicago economist James Heckman calls “soft skills”: being reliable, working well with others, turning up on time. There is even suggestive evidence that these skills lead to big differences in productivity of seemingly very similar industries in different countries. The word “skill” is a bit misleading, though; Heckman is really talking about character, and character is not taught in evening classes but is learnt in schools, churches and above all, families.

So, perhaps cultural change is not just a luxury, but is part of a conservative program for social justice, an alternative to populism in its Left and Right varieties. If so, it will be a way to help liberty survive.