But there is a downside to all of this. Whenever I argue that some policy X was a good idea, which has just been badly implemented, it never takes long before somebody retorts: “Oh, so you’re saying, that wasn’t real X, real X has never been tried? Funny how you sound just like the socialists you enjoy mocking!”
I suppose I’m not in a position to complain about that. You reap what you sow, and I have been sowing generously. But I’ll do it anyway. Here goes:
When an idea that we cherish does not pan out as well in practice as we thought it would, there is a temptation to look for excuses, in order to protect the idea. Blaming the details of implementation can be an easy way to do so, especially when this claim is hard to test empirically.
But it would be nonsense to jump to the opposite extreme, that is, to claim that ideas cannot be badly implemented, and that any implementation is as good as any other. There are plenty of political ideas that work in some settings, but fail in others, because some get the details right, and others do not.
The Weimar Republic failed. Does that mean that the republic, as a form of government, is a bad idea? Of course not. The idea was good. But it wasn’t implemented properly.
Since 1990, some Eastern European countries have successfully transitioned to a market economy, while others messed up the process. Slovakia, Slovenia, the Czech Republic, Poland and the Baltic nations are now high-income countries, which tend do well on social and human development indicators as well. But Albania, Ukraine and some of the former Soviet Republics remain poor.
Soviet apologists, naturally, pick out the worst examples, and claim that capitalism has failed. Defenders of capitalism retort that it has not ‘failed’ – it has just not been properly done. And they are right.
In the 1990s, several South American countries partially replaced their state pension system with personal retirement savings accounts. It worked in some places (if, by ‘worked’, we mean providing decent rates of return and security of investment), but not others. Because some did it right, and some did not.
I could go on, but you get the point: just because the good-idea-badly-done claim can be used as an excuse, that does not mean that it is always and everywhere an excuse.
So how can we spot the difference? How can we tell whether an opponent is looking for excuses, or perhaps more importantly, when we are looking for excuses? How can we more effectively police ourselves? How can we tell when we are not being honest with ourselves, because we are looking for a way to protect a cherished belief?
There are, of course, no hard-and-fast rules. We are talking about claims that cannot be easily empirically tested. But I’d suggest the following ad-hoc criteria as a place to start.
- Can you name at least one example of X that you consider at least OK-ish? It is always easy to find reasons for distancing ourselves from a particular example of X in action. Political and economic theories are almost never implemented in pure form, so if we look for it, we can always find a reason for claiming that this particular version of X is not ‘real’ X. Such is the psychology of motivated reasoning. But where did it work? One positive example is worth more than ten reasons to dismiss a negative one.
- When you say that X was ‘badly implemented’, what exactly do you mean by that? Can you spell out what, specifically, you would do differently? Which tangible steps would you take to improve it? The emphasis here is on ‘exactly’, ‘specifically’ and ‘tangible’. Don’t waffle. Don’t escape into abstraction, soundbites and platitudes. Don’t talk about the outcomes you would like to see, describe a mechanism that would achieve them.
- When your overall claim is too broad-brush to be empirically testable, can you break it down into smaller claims, which can be tested separately? Even if your preferred policy package XYZ has never been successfully tried as a package, maybe X has been tried successfully in one place, Y in another, and Z in yet another. That still does not show that they would work as a package, but it is better than nothing.
- Can you demonstrate ‘functional equivalence’ with something that does work? Even if X has never been successfully tried, maybe there is something which works quite like X, but which just isn’t called X, and which we wouldn’t usually think of as X. In such a case, you could argue that if ersatz-X works, then by extension, so should X.
Not all of these questions will be applicable to each case, but if you can’t answer at least one of them satisfactorily, I’d say, that’s a good sign that you’re fooling yourself. On this four-point checklist, socialists do not perform well.
- Can they name a successful example of a socialist society? They can easily fill a podcast talking about why neither the Soviet Union, nor the other Warsaw Pact countries, nor Maoist China, nor Venezuela, nor any other example, was socialist, but they cannot name a single positive example.
- Can they explain what they would do differently? They waffle, and they escape into soundbites and abstraction, but they cannot provide even a rough outline of how ‘their’ version of socialism would work.
- Can they find partial examples? Not really. When discussing socialist policies within market economies, socialists often resort to a small-scale version of the not-real-socialism argument. The nationalised industries of the past? That wasn’t real nationalisation. Real nationalisation has never been tried. Rent controls that have led to shortages of rental properties? Not real rent controls. Real rent controls have never been tried.
- Can they demonstrate ‘functional equivalence’? Again, not really. Some socialists have tried, by arguing that prehistoric hunter-gatherer societies were effectively ‘socialist’, and that therefore, socialism could work in the modern age as well. But the whole point of the Hayekian case against socialism is that what works in small groups with very simple economies is not scalable.
Contrast this to those who claim that the mixed record of private pension systems in Latin America is a matter of implementation.
- Can they mention a positive example? Of course: The Chilean system is a qualified success, and Colombia, Peru and Uruguay are not bad either.
- Can they spell out what they would do differently? Of course. For example, don’t force pension funds to put most of people’s money into government debt and domestic corporate bonds. Allow them to diversify internationally and institutionally.
- Can they find examples of partial success? Of course. Some systems get some things right, others get other things right. In some places, rates of return are decent, but the pension fund industry is uncompetitive, leading to high user charges. In other places, user charges are competitive, but coverage rates are too low. And so on. Pull all the good bits from a number of places together, and you have a great system.
- Can they demonstrate ‘functional equivalence’ with something which does work? Of course. There are successful investment funds in countries with dysfunctional pension systems. They are just not called ‘pension funds’. Thus, the relevant question becomes: what stops the pension funds from performing as well as the non-pension funds?
You could counter that this is all just an elaborate way of saying “It’s fine when I argue that real X has never been tried, but not when my opponents do it.” And maybe so. But that would only show that the above is not a real checklist to prevent myself from engaging in motivated reasoning.
Real motivated-reasoning-prevention checklists have never been tried.