Digging holes and inertia.

Under my post last night about uncertainty and counter-cyclical policy Robert Lewis wrote

I'm not 100% convinced that the rewards don't outweigh the costs on all government intervention, but perhaps the answer is to just have people dig holes and fill them back up. This would allow some countercyclical spending with little-to-no other side effect.

The Onion responded for me much more hilariously than I ever could:

In The Know: Should The Government Stop Dumping Money Into A Giant Hole?

At the risk of killing the humor by explaining the joke, the real point of this video is not to mock senseless government waste (although it does that) but to point out how much inertia even the most senseless of policies can acquire over time. The pundit who opines that those who favor closing the national money hole should try their arguments on "the digger who's worked the graveyard shift for twenty years" perfectly captures why government programs so far outlive their useful (and even their intended) life. We still have farm subsidies nearly a century after they were instituted to save poor farmers from ruin; now they form a major part of the balance sheets of massive agribusiness corporations. Our bases in Europe and Japan remain twenty years after the Cold War ended; Social Security and Medicare persist even though the elderly are now the richest age bracket in the United States. Ad hoc policies established in an emergency become the new normal; if you don't believe me just compare spending levels in 2006 to the proposed levels viewed as the harshest of austerity today. If we enacted the Hole Digging Act of 2011, in fifty years the Corps of Hole Diggers would be a venerated American institution and tampering with it will be viewed as political suicide.

The usual answer to this worry is making countercyclical fiscal policy take the form of "automatic stabilizers," mechanisms that kick in when the economy takes a nosedive and automatically disengage as the economy recovers. The standard example of an automatic stabilizer is of course unemployment compensation- give the unemployed enough money to keep them from rioting in the streets but not so much or for so long as to create a serious incentive to loaf.

But those two qualifications mask a lot of complexity. Just how do you determine the right length of time and the right amount of money? Here again I think Keynesian radical uncertainty runs against Keynesian policy. Getting unemployment compensation right depends greatly on what the recovery is going to look like. Will the jobs lost be replaced by equivalent jobs, or are they gone forever, to be replaced by very different jobs? We learn from behavioral economics that the more heavily invested one is, the less willing one is to write off the investment. A skilled maker of buggy whips, with a lifetime invested in his skills, will wait for a recovery that for him is never coming far longer than appears rational, and this behavior is only encouraged by unemployment compensation. Further, as we are seeing today, political incentives create heavy pressure to extend the term of these benefits again and again.

I wrote the immediately preceding paragraph and while I think it's true, I don't find it satisfying intellectually or morally.  It's the standard Austro-libertarian answer to the Keynesian argument about the value of unemployment benefits in a downturn. I do think it's broadly correct, but it still doesn't truly address what we should do immediately about the real suffering caused by unemployment. "Let them die" works well for corporations, but not so well for people. For now I'll just fall back on my earlier arguments about libertarian priorities and repeat that we should go after government policies designed to comfort the comfortable before we trouble ourselves with afflicting the afflicted. But I am slowly starting to have a sneaking suspicion that some form of countercyclical policy may well have a place in Libertopia. Libertarian true believers, scourge me for my heresy in the comments.