Obama recently announced that he would stop deporting illegal immigrants, thus effectively nullifying parts of immigration law through non-enforcement. The injustice of immigration law today is so gross that we may well applaud this particular move. However, in response, B. Caplan writes:
I say the laws on the books are so overwhelmingly wrong that even random Presidential nullification would be a huge expected improvement.
In the year 2012, Caplan argues that an even more imperial Presidency would be a good thing, not in this one instance, but as a general rule. He has surveyed the landscape of the American government and decided that the big problem is that Presidents don't have enough unaccountable power. Yes, let's bring back the days of Worcester v. Georgia, when the Supreme Court failed to order federal marshals to enforce the ruling, for fear that Andrew Jackson would not comply!
Furthermore, the statement has a ridiculous premise. Laws will not, of course, be nullified at random. They will be nullified in the exact ways that Presidents find most politically advantageous or ideologically compelling. I suggest you ponder on your own whether you think that will be a good thing in general. At best, I think it is a mixed bag. Certainly, cravenly seeking political advantage can sometimes (as with Obama's strategic courting of the Latino population here) lead to policy outcomes that improve the welfare of many people, simply because politics is the art of getting many people to vote for you. And the dominant ideologies of American politics are not entirely malign (although a libertarian like Caplan should perhaps feel some unease at the thought that Presidents from the very parties which, allegedly, have been passing so many terrible, liberty-destroying laws will wield nullification as a liberty-enhancing tool).
It is remarkable how ideology can impair the thinking of extremely intelligent people. The points I make above are not the product of any profound thinking. Had Caplan focused his mind on his own statement in a similarly critical way, he might well have ended up feeling uneasy as well. But he did not, or he overcame that unease anyway.
I blame his ideology because I see this error as one common to a lot of libertarian thinking. Libertarians often embrace the wrong measures of the intrusiveness of government. Caplan thinks that less law will lead to greater liberty. (This is a cousin of the fallacious notions that lower tax rates lead to less government interference in the economy, or that private governments whose edicts are backed by the force of property law are not governments.) But of course that is an error. We should seek to have the right laws, and, perhaps equally importantly, just mechanisms for establishing the right laws. To focus on whether there are more of them, or fewer, is rather akin to King Philip in Amadeus criticizing Mozart's compositions for having "too many notes — just cut a few".
Now, our laws are hardly Mozart operas, but the point is it's a silly way of measuring things. Just so with a typical libertarian's ways of measuring government.
 This is an error in at least two ways. First, lower taxes without lower spending merely time-shift interference into the future. Second, a simple, universal tax injects less entropy into the market mechanism than a large number of narrowly targeted taxes applied inconsistently; if we view markets as information-gathering mechanisms then this information-theoretic view of taxation may be a better measure of government interference than tax rates.