1/ I asked this question -- how do lockdowns stop? -- almost 3 weeks ago and I don't think anyone has a really good answer.

Today I want to leave aside the efficacy of lockdowns and talk about ending lockdowns in the context of the agency problem https://twitter.com/Coyoteblog/status/1242176183447384064

2/ When I first studied the agency problem, it was in the context of corporate governance. Shareholders hire managers to run the company for the owners, but managers have their own goals and incentives that can diverge from those of owners

3/ This problem can be made worse or better based on the crafting of compensation and incentive systems, but never ever goes away entirely. Consider an example from my business. A large corporation has a tiny division that operates contracts in my industry

4/ This company is constantly on the verge of divesting this division, a fact known to the managers who run it, who are fairly sure they will lose their jobs if the company gets out of this business. Maybe if they grow the business it will get too big to kill?

5/ So they make crazy aggressive bids for every contract that comes along to grow the business at any cost. It is pretty clear that getting new contracts on bad terms in a tangential business is not helping shareholders, but it makes sense for the managers

6/ We see this agency problem all over business. It can, for example, create a bias towards acquisition programs (managers in larger companies get more pay & prestige) and a bias towards stock buy-back programs rather than dividends as a way to return capital to shareholders.

7/ This exact same problem exists in government. In fact, I would argue that it is far worse in government as private entities at least have clear metrics for success that manager actions can be judged against -- no such thing exists in government for any government program

8/ This agency problem in government is one that used to be ignored -- it was assumed in political theory that government officials were dispassionate beings who tried to optimize the public good. The public choice theory of economics changed all that.

9/ It is amazing how many political scientists and observers STILL ignore public choice theory and the agency problem in politics, but at least you can find some discussion of it out there, somewhere.

10/ So, returning to COVID19. Somewhere along the way in this crisis the political rules of engagement were defined in such a way that politicians could be blamed for COVID19 deaths.

11/ This may seem natural now, but it's actually not how we normally operate. Two years ago, 60,000 Americans died of the flu but no one even considered blaming a politician for that. Ditto the 36,000 annual automotive deaths.

12/ Well, you might say, politicians couldn't affect those things, but that's not true. The same logic behind current lockdowns potentially preventing COVID19 deaths applies equally well to flu or driving fatalities.

13/ However we got here, we have created a massive agency problem. The actions on the table for government officials might save lives, but they also have massive costs. Idealized political theory would say that politicians would then balance these costs and benefits.

14/ But public choice theory teaches us this is BS. Politicians are rational individuals who (under the current rules of engagement) know that they may be blamed for COVID19 deaths but likely will not be blamed for the lockdown costs.

15/ From a personal incentive issue, there is no balancing to be done -- they are going to go all the way, no matter the cost, to leave absolutely no possibility to argue that any deaths could be blamed on them.

16/ The incentives were so unbalanced, particularly in blue states where many see government economic control not as a cost but as a wonderful precedent for future interventions, that nearly every politician went all the way to the most extreme containment actions possible

17/ Which brings us to the original question: How do we end the lockdown? Note that the incentives on politicians have not changed. Few people seem to give a shit about cost (particularly given the bipartisan effort in DC to helicopter drop trillions of dollars).

18/ And the fear of getting blamed for deaths is if anything even greater. Case loads and deaths will almost certainly have a second bulge after removing lock-downs, and politicians fear that every death in that bulge will be blamed on them.

19/ Just look all over my Twitter responses. Every time I argue that we entered these lockdowns without good science, good data, clear goals, or any threshold metrics, ten people write that its wrong to balance death against mere economic output.

20/ But for those who are committed to the latter position, the lockdown literally cannot ever end. A number of prominent politicians have put themselves out on exactly this limb. It will be interesting to see who has the cajones to make the first move.

@ElonBachman is getting at something similar here https://twitter.com/ElonBachman/status/1249033757736271879

A simpler way to put this whole series of tweets is, "who will be the first person willing to stop clapping at the end of Stalin's speech?" http://www.disappearingman.com/communism/men-wouldnt-stop-clapping/