A Fractal Case for Big Government

Brad DeLong made a case for big government and higher deficits. Paul Krugman wrote an exagerated pun of praise.

The part he liked was the appeal to myopia and paternalistic proposal “The problem is that as we move into the twenty-first century, the commodities we will be producing are becoming … more subject to myopia and other behavioral-psychological market failures.” I think that any hint of paternalism is rhetorical poison — even Brad doesn’t quite say that he thinks the government knows best whether, say, kids should stay in school. School attendence in mandatory, but we don’t talk about it much. Here my only contribution is to note a paradox of freedom. if people have time inconsistent preferences, they will choose to precommit. This means that people freely choose to reduce their future freedom. One example would be to introduce a payroll tax used to fund pensions. Notably that is a very very popular policy.

But I focused on something else.

1890. Since then, over any extended time period for major North Atlantic industrial economies, g > r without fail. That is now 125 years.


That fact strongly suggests that North Atlantic economies throughout the entire 20th Century suffered from a form of dynamic inefficiency, in that there was excessive accumulation of societal wealth in the form of net government capital—in other words, government debt was too low. Given the debt secured by government-held social wealth ought to be a close substitute in investors portfolios with debt secured by private capital formation, it is very difficult to understand how economies can be dynamically efficient with respect to private capital and dynamically inefficient with respect to government-held societal wealth in the absence of truly mammoth financial market failures.

Like Brad I follow John Quiggin in believing that the low return on public debt compared to corporate equities is a truly mamoth financial market failure. In other contexts, this view has caused Brad to advocate bond financed public purchases of stock, that is a sovereign wealth fund. This would be a huge expansion of the government and the public debt (public assets aren’t counted). I think risky assets are worth more to the US Federal Government than to any other entity, because it has the deepest pockets and is so big that if it fails, the world economy fails with it. The gigantic profits the US government made while saving the US financial sector are strong evidence of this.

Nothing new yet, but now I want to argue that the advantages of gigantic risk bearing capacity affects optimal public sector employment and public provision of private goods. The Medicare advantage experiment provides evidence that the public sector CMS is more efficient than health insurance companies. Part of this is the high required return on equity, some of the extra cost of Medicare advantage goes to health insurance company profits. Importantly profits are not theft. Insurance companies must have equity and investors demand high returns on equity. However, it is also a fact that public sector bureaucrats are paid much less than private sector managers (even counting the more generous benefits). It is also true, as passionate bureaucracy hater James Q Wilson taught me long ago, that they tend to be, if anything, more qualified on paper. Here I think people are willing to accept much lower incomes for much higher job security.

Now, it is generally believed that the low salaries correspond to extremely low work effort and that the job security leads to extreme moral hazard due to shirking. I am sure this can be a gigantic issue (I work in the Italian public sector — but not very hard). But I’m not convinced that it outweighs the advantages of public employment in many other countries including the USA.

So if offering job security is efficient, why can’t a profit seeking corporation achieve higher profits by offering job security ? One issue is, again, deep pockets. A corporation just can’t guarantee lifetime employment — if demand for its product falls it will go bankrupt. GM offered only lifetime health insurance and ended up bankrupt. Civil service job security corresponds to the US Treasury bearing risk. This can be efficient given its unique risk bearing capacity.

But another issue is dynamic consistency again. As noted by Schliefer and Summers, the possibility of corporate takeovers can be costly to shareholders, because it makes it impossible for shareholders to precommit to an implicit contract with workers by hiring a manager who really wants to keep her word. Managers can’t promise workers that they will stick to an agreement even if, in the event, it would be better for shareholders for the firm to renege. The shareholders can and do toss the managers.

Interestingly Summers (and DeLong) worked for the Clinton administration which had a project to reinvent government by replacing public provision of services with vouchers so they would be provided by the efficient private sector. I don’t think this has worked out very well (say Blackwater 10 times before commenting if you disagree). I think part of the problem is the risk bearing issue I discussed above.

But I think a much larger part of the problem is the nature of typical government failure. I think that the largest losses for citizens occur at the edge of the state, where lowly paid civil servants negotiate with highly paid managers as both stand near the revolving door. If so, this would mean that the cost is proportional not the volume of government but to the surface area of government. Outsourcing tasks reduces government employment, but it increases the number and dollar value of contracts negotiated by agents with extremely asymmetric incentives. I honestly think that the least efficient approach is the mixed approach where money collected with taxes is used to pay profit seeking firms to provide services.