Target the S&P 500?

Posted on by

Lifted from the comments:

EMH has me puzzled. Since stocks are linked to the economy you must also conclude that long-range predictions of the economy are no better than throwing darts. Yet many economists seem to believe that the Fed could in fact target NGDP and therefore create the economy they want in some respect (obvisously there are plenty of variables they can’t control). Is a variable really efficient if someone can target it?

To put it more simply perhaps, the Fed COULD target the S&P 500 if they wanted to (essentially pick a value). If an entity exists that can control a variable then isn’t it impossible for that variable to be completely unknowable?

1. Long-range predictions of the economy are not much better than throwing darts. Even projections of a year ahead are not much better than just guessing that the real GDP will grow by 2.5 percent.

2. I think that Scott Sumner would say that the Fed could target the level of the S&P 500. That is a nominal variable.

3. However, the Fed cannot target the real return on stocks. If the Fed targets an S&P 500 of 2200 for one year from now, and this is credible, then the S&P 500 has to rise today to the point where the expected real return is comparable to that on other assets.

4. Part of the EMH is that markets anticipate what the Fed will do. So the Fed cannot suddenly surprise markets by targeting an S&P 500 of 2200. If you extend that, you would say that the Fed cannot suddenly surprise markets by targeting a particular level of NGDP.

5. I think there is a bit of tension between believing in the EMH and believing that the Fed can choose any NGDP value it wants. I think Scott is aware of the tension, and I forget how he resolves it.

6. I think that bringing up the stock market is a very good way to raise the issue of whether the Fed can target any nominal variable.

7. Of course, I am not the one who has to defend the proposition that the Fed can target nominal variables. I believe that financial markets can do what they want with asset prices, and that money and prices are consensual hallucinations.

7 thoughts on “Target the S&P 500?”

  1. ” I think there is a bit of tension between believing in the EMH and believing that the Fed can choose any NGDP value it wants. I think Scott is aware of the tension, and I forget how he resolves it.”

    He resolves it by saying that the Fed always has unlimited ammunition through fiat money creation to adjust the overall price level, even if it has to resort to price-control-like purchasing of unlimited quantities of one particular asset (or some basket) until the nominal price reaches the target.

    Yes, this might cause all kinds of distortions and relative real adjustments of every other asset in the economy, but any entity that can print money can control the price of one particular asset. I don’t think this should be controversial, because otherwise you have to believe the supply curve for that asset goes completely horizontal as some nominalprice in dollars, and the market will continue to indefinitely provide an infinite quantity of that asset at that particular nominal price, no matter how much is bought. That doesn’t seem right to me.

    Sumner says that it’s important for the Fed not just to control the current price of any asset or target, but to keep the entire path of future expected prices on track.

    But how does the Fed know about future expected prices? How does it know that it’s current policy is achieving the goal?

    Sumner says the EMH tells us that there is no better way to guess or model those future prices except to actually gather the information from market data in a futures market, subsidized to whatever (market-determined!) extent necessary to generate sufficient depth and liquidity.

    It is a little confusing when you think that if the Fed policy is such that it will print money to guarantee the prices of a particular asset, how can the market expectations for the future prices of that asset ever deviate from that guarantee?

    I think he resolves this by saying that the way the Fed guarantees the price is to intervene in that particular market, and also, that the market is tied to a real quantity of nominal aggregate demand, and that arbitrageurs can make money if they notice that expectations are not matching reality, and their other market activity will close whatever gaps pop up.

  2. The central bank cannot target the actual level of NGDP precisely. It can only operate indirectly by increasing or decreasing the monetary base. Hence, NGDP targetting is about increasing or decreasing the monetary base to keep the expected level of NGDP on target. The NGDP targets are set on the basis of a level-targeting system; so, if actual NGDP is below or above the target, then targets will be raised so as to increase NGDP more in the future. A feedback mechanism for the system in case forecasts are wrong.

    The key difference between the S&P 500 and NGDP is that the S&P 500 is tradeable. I can buy and sell quantities of the underlying stocks. If I had enough money I could put in place targets for each stock and defend them.

  3. “However, the Fed cannot target the real return on stocks. If the Fed targets an S&P 500 of 2200 for one year from now, and this is credible, then the S&P 500 has to rise today to the point where the expected real return is comparable to that on other assets.”

    This is interesting, as a thought exercise. I can see how this would fairly easily affect the value of other financial assets (portfolio rebalancing). However, it is much less clear that this could affect the overall price level (as captured by the CPI). The closest mechanism I could think of is that increased capital gains tax receipts then lead to greater government expenditures.

    Given that very leaky transmission mechanism it would seem to me that, on the surface, the Fed could create real returns. If it targeted SPX 2200 and a 10% annual increase I don’t think the CPI would rise sufficiently to wipe out the added real returns. So perhaps wealth, in addition to money, is a hallucination.

  6. It is not part of the EMH that the market correctly anticipates Fed actions; just that market prices embed the best expectations of Fed actions. The Fed can always surprise the market (and has many times).

Comments are closed.