How could social security privatisation cause increased economic efficiency ?

This is an attempt to explain the post immediately below.

Michael Kinsley argues (brilliantly as usual) that the claims Bush makes for “personal accounts” can’t possibly be true. He summarizes these claims as 1) people who invest in personal accounts will gain financially and 2) no one will suffer financially. He argues that, for this to be true, personal accounts would have to either increase national savings or increase the efficiency of allocation of capital. If the argument is translated into standard economic theory, the case is somewhat stronger, since increasing savings makes you richer in the future at the expence of less consumption in the present and doesn’t count as “gaining financially”. There is also no good reason to think personal accounts will cause people to save more.

I want to present a model in which personal accounts can improve the efficiency of allocation of capital. The point is that the CEA confidently asserts that this model is nonsence and that, therefore, Bush’s claims are nonsense. I find this amusing.

The story (see post below) is very simple. There is a mysterious equity premium – the return on stock is higher that in should be given the risk unless people are absurdly risk averse. In other words stock is cheaper than it would be if people were rational. This sends a misleading signal to industrial firms. The amount of investment which is in the short term interests of shareholders is less than the amount which is socially efficient. The key words are “short term”. If firms paid lower dividends and invested more and shareholders were forced to hold on to their shares, they would be angry, but would benefit in the long run. The solution to the underpricing of stock is to force people to buy stock. This would drive up the price of stock and make high investment (low diviidend payout) attractive even in the short run. Here the problem is the equity premium which causes inefficient allocation of capital and the solution is to get rid of the equity premium. Oddly the CEA insists that the equity premium is here to stay in any case. This seems to me to imply that personal accounts can’t help anyone without hurting someone else. That is the CEA claims seem to me (as argued by Kinsley) to be contradictory.

Now I want to present a sketch of a formal model. There are various ways do write such a model but all must involve a choice of how to allocate capital. I will assume that there are two types of firms safe firms and risky firms. The expected return on capital in risky firms is higher but it comes with an idiosyncratic disturbance term (is stochastic or partly random so is risky). The key word here is idiosyncratic, that is, firm specific. It means that the average return on all risky firms is not at all risky and is just higher than the average return on safe firms. I will assume that for each risky firm the return on capital takes two values high and zero. For each safe firm the return is low and absolutely predictable.

Everyone would be better off if the safe firms just shut down and everyone invested their wealth in a balanced portfollio of risky firms. Unfortunately people are real real dumb and don’t understand this. They look at the risk they would face if they bought the stock of one single firm and assume that this is the same risk they would face if they bought a balanced portfollio. This means that they demand an equity premium which makes no economic sense. I actually believe this is the reason there was an equity premium which might or might not have vanished allready.

Finally there are no banks. Firms are financed with stock and or short term bonds. Safe firms sell bonds promising to pay the safe low return. Thus they can raise as much money as they want and never go bankrupt. Investors rationally know the bonds are perfectly safe and accept the low return.

Risky firms can’t sell bonds. If their return on capital is zero they can’t redeem the bonds. I assume that the crazy investors invoke chapter 7 of the bankruptcy code and the firm goes into ignominius and costly bankruptcy. This makes bond financing totally impossible for risky firms.

The risky firms can finance themselves by selling stock which has a higher expected return than the low safe return on bonds. However, the crazy investors don’t want to buy the stock because they don’t know about the law of large numbers. Instead they demand an even higher expected return making it unprofitable for the risky firms to sell stock.

Thus the economy has only safe firms and a low return on capital.

If the state forces people to buy a balanced portfollio of stock however, the risky firms can sell stock and operate. This means that the economy has a higher expected return on capital. By the law of large numbers all the risk is diversified and so the reluctant investors get a higher return for no cost. The way this works is that they are forced to pay more than they want to for stock. The orignal problem was that they underestimate the value of stock.

This simple model has three key features. Investors who don’t understand diversification are required bo buy diversified portfollios, investors who do not want to buy stock are forced to buy stock, and this drives up the price of stock and drives down the return on stock. The last feature is the whole point. For the risky firms, the return on stock is the cost of capital Driving it down is the whole point of the exercise.

Now this model obscures a few key points. A balanced portfollio of stock is not a totally safe investment and investing in stock involves transactions costs which are quite high. That is, in the real world, even if personal accounts drive down the equity premium, they would probably (almost certainly) be socially undesireable anyway.

However, the claim that the equity premium will not be affected and someone will gain and no one will suffer is obviously total nonsense.