Thursday, August 28, 2008

Unintended Consequences, Intent, and Philosophy

In the context of discussing the causes and effects of income inequality, Will Wilkinson has a post up (which Megan McArdle picked up in turn) regarding the willingness of partisans and, really, most people to claim or imply that the unintended consequences of an opponent's policy preference are in fact the primary goals of that opponent.

Wilkinson writes:

I feel like there is an unarticulated doing/allowing issue floating around in the background in this debate. Say the U.S. Congress cuts top tax rates. Is this politics causing higher inequality? Or is this evidence of relative indifference about allowing higher inequality? The left has the tendency to characterize every policy that might allow income inequality to rise as one intended specifically to have this result. This is a lot like the right’s characterizing, say, workplace safety regulation as a specific attempt to stymie the growth of small business. In each case, those opposed to a policy see its side-effects as more salient than the primary effects intended by those who favor it. Imputations of bad faith — “you’re really after the side-effect and your stated intention is garnish for malice” — are never far behind.

Wilkinson is of course 100% correct about this, and this is something that can be observed in just about any political debate these days (and, perhaps, since the beginning of time).

In the context of discussion of income inequality, there is a tendency amongst some of those on the "Right" of the economic policy spectrum to view income inequality as a healthy sign of a level playing field demonstrating the existence of a strong meritocracy. Meanwhile, there is a tendency on the "Left" of the economic policy spectrum to view income inequality as proof of the opposite - far from demonstrating a meritocracy, significant (and growing) income inequality is proof of a self-perpetuating cycle in which those who are provided with greater opportunities due to their parents' economic well-being succeed and those who are unfortunate enough to be born in poor areas do not participate in economic growth. Thus, the Left will perceive policies that will inevitably have the effect of allowing the already-successful to retain more of their wealth as being intended to further entrench this self-perpetuating cycle. The Right on the other hand will perceive policies that will inevitably interfere with the operation of business as being intended to undermine the meritocracy.

But isn't there a third way of looking at this, though (and one which I suspect is probably closer to the truth)? Specifically, can't income inequality be, at the same time, both a sign of a healthy meritocracy and a sign of a self-perpetuating cycle? In that case, the result is that the "playing field" is level enough for many people to have social mobility, but not for other groups of people. That then raises the issue of how one maintains the social mobility for the former group while still seeking to obtain it for the latter group. I would argue that the answer to this lies in the idea that the factors that would make the playing field "level" are different for each group.

I am willing to concede that, in some/many circumstances, wealth redistribution may be what is required to give someone a realistic opportunity of participating in the "meritocracy." However, in other cases, wealth redistribution may actually reduce one's opportunity of participating in the "meritocracy" because of the various disincentives it provides (i.e., it can and at least in some - though by no means all, and probably not even most - cases does foster laziness amongst recipients while disincentivizing work beyond a certain amount by potential upper bracket-earners).

How one would go about solving this dilemma is probably far beyond any human's capability. Unfortunately, politicians and ideologues on all sides seem to think they have the answer. Some conservatives and libertarians, for instance, often believe that the answer lies in part in school choice because they figure that equal access to education is the ultimate leveler of the playing field and that school choice would provide such equal access. Liberals/Progressives on the other hand often tend to believe that the answer lies in providing stronger safety nets such as welfare benefits, minimum wages, etc.

The problem is that any of these proposals is, by necessity, a one-size-fits-all approach that will, as is the case with any policy, have unintended consequences as a result. So although each approach will almost certainly benefit some unknown number of deserving individuals, it will also almost certainly hurt an unknown number of others who do not deserve that hurt*. Moreover, because of the complexity of the economy, there is both no way of knowing whether any of these approaches will do more good than harm and no way of definitively proving whether any approach actually succeeded once implemented.

Ultimately, whether you think a particular policy is likely to have net positive/negative effects is probably entirely a function of one's political philosophy since there is no real way to demonstrate these effects conclusively. Personally, I think this is a particularly strong argument for libertarianism since I think that, without any real proof of whether a policy will be a net good or a net bad, one should err on the side of liberty and freedom. BUT, I can also see how this could be a particularly strong argument for Burkean conservatism (one should err on the side of tradition and existing norms which are the result of millenia of accumulated knowledge) and for modern Progressivism (one should err on the side of change when large numbers of people are hurting under the existing system).

*Politicians and policy advocates will almost always claim that the number of those benefited is, in fact, known and is equal to the number of people who are expected to receive the outlays of the policy. They will also imply that the policy will have no unintended negative consequences. This is poppycock, since, as I said, these policies are one-size-fits-all approaches to problems that have myriad causes (and in some cases may not even be problems at all).

(cross-posted as a comment at Megan McArdle's post)