Wednesday 15 December 2010

What's the model?

Karl at Modeled Behaviour says that Peter Orszag's flip from OMB to Citigroup is about what we expect when high level public sector salaries are too low.
One thing the low pay of senior public officials allows for is a pump-and-jump. Even in the most noble of circumstances smart folks will notice that they can get to the front of the line pretty fast in the low competition public sector, build an impressive resume and then jump ship to the private sector to make a load of cash.

Less noble would involve actively selling the benefits of one’s position to the highest bidder. What would stop people from doing this? The fear that they would be fired and thus loose out on a lucrative salary. However, with no lucrative salary there is little incentive not to do this.

I do hope that economically oriented folks aren’t suggesting that we use moral suasion to control government corruption. People respond to incentives. If you don’t want them to sell you out then you have to pay them more.
But the only way this works is if public salaries are no less than the relevant private alternative. If the bundle of connections and institutional knowledge embedded in a high ranking official is worth more to folks in the private sector than in the public, we're going to have problems. At best, we'd delay folks jumping ship for a few years: lots of these appointees are dumped whenever there's a change in administration. Their value to the private sector then is less than it would be as a mid-term jump, but the problem isn't much changed.

The only way I can see higher pay being a solution to the problem is if much of it comes as a bond that pays out if, after some period of time, the official hasn't flipped to work for somebody he was regulating while in office; the bond would have to be big enough to outweigh the official's value to the private sector. That seems unlikely.

We can also build a pretty reasonable efficiency case for allowing these guys to jump to the private sector. Suppose that the regulatory barriers facing firms are exogenous and largely silly. A official bringing to the private sector knowledge of how to best avoid those costs brings real value. The main worry is if the barriers are endogenous or if the private sector reward is endogenous to diversions made by the official to private agents.

First best is to reduce the value of these guys to the private sector by having fewer private sector profits be contingent on intimate knowledge of political processes.

Will Wilkinson writes more eloquently than I do:
In my opinion, the seeming inevitability of Orszag-like migrations points to a potentially fatal tension within the progressive strand of liberal thought. Progressives laudably seek to oppose injustice by deploying government power as a countervailing force against the imagined opressive and exploitative tendencies of market institutions. Yet it seems that time and again market institutions find ways to use the government's regulatory and insurer-of-last-resort functions as countervailing forces against their competitors and, in the end, against the very public these functions were meant to protect.

We are constantly exploited by the tools meant to foil our exploitation. For a progressive to acknowledge as much is tantamount to abandoning progressivism. So it's no surprise that progressives would rather worry over trivialities such as campaign finance reform than dwell on the paradoxes of political power. But it really isn't the Citizens United decision that's about to make Peter Orszag a minor Midas. It's the vast power of a handful of Washington players, with whom Mr Orszag has become relatively intimate, to make or destroy great fortunes more or less at whim. Well-connected wonks can get rich on Wall Street only because Washington power is now so unconstrained. Washington is so unconstrained in no small part because progressives and New Dealers and Keynesians and neo-cons and neo-liberals for various good and bad reasons wanted it that way. So, what is to be done? Summon a self-bottling genie-bottling genie?

The classically liberal answer is to make government less powerful. The monstrous offspring of entangled markets and states can be defeated only by the most thorough possible separation. But public self-protection through market-state divorce can work only if libertarians are right that unfettered markets are not by nature unstable, that they do not lead to opressive concentrations of power, that we would do better without a central bank, and so on. Most of us don't believe that. Until more of us do, we're not going far in that direction. And maybe that's just as well. Maybe it's true that markets hum along smoothly only with relatively active government intervention and it's also true that relatively active government intervention is eventually inevitably co-opted, exacerbating rather than mitigating capitalism's injustices. Perhaps the best we can hope ever to achieve is a fleeting state of grace when fundamentally unstable forces are temporarily held in balance by an evanescent combination of complementary cultural currents. This is increasingly my fear: that there is no principled alternative to muddling through; that every ideologue's op-ed is wrong, except the ones serendipitously right. But muddle we must.
Ridiculously depressing. Two brighter notes:
  • The world as it is is still a pretty decent place
  • Seasteads may come


  1. I can't really see Wilkinson's argument. Powerful market institutions (e.g. large financial corporations) use government systems to exploit us (presumably because they're more powerful than the regulatory pressure exerted by the system). Therefore.... abandon any efforts at controlling them? Accept defeat permanently? Rely on serendipitous, fleeting states of grace?

    It seems a better solution would be a better-designed system. The relevant question here has to be a counter-factual: given the problems identified by you/Wilkinson, would we be better off without the regulatory efforts in the first place? It's not a matter of whether the system as-is works as it ought. Or am I missing something?

  2. @haitijustice: Dwight Lee had a great piece a while back: the impossibility of desirable minimalist government. The postcard: if it were possible for government to be constrained against doing bad things, we'd want more government. So minimal government wouldn't be desirable. But if we can't constrain government against doing bad things, we want it to be very small. But it's impossible to constrain it if it's bad.