A new monetary policy for the 21st century

For the past seventy years, policy makers have relied on fiscal and monetary policy to combat recessions. Monetary policy works by lowering real interest rates and stimulating private expenditure. Since the nominal interest rate on three month treasury bills has now reached zero in the US, the scope for further easing is limited. This has led to an intellectual tsunami of proposals for a Keynesian-style fiscal stimulus of historic proportions.

Although a fiscal stimulus is needed; I am skeptical that the $775 billion proposal of the Obama team will be large enough to have the desired effects. A proposal that is large enough to restore full employment may place an unacceptable burden on future generations by increasing debt at a time when implicit liabilities to fund Social Security and Medicare are already large and growing.

I argued in the FT Economists Forum, Dec 30 2008, that Central Banks throughout the world should intervene directly in asset markets by buying and selling a broad index of stocks. My proposal is based on new research, summarized in a working paper on the Great Depression, and developed further in a forthcoming book, Expectations, Employment and Prices. This work brings Keynesian economics into the 21st century by filling in the missing pieces of Keynes’ General Theory.

Classical economists believe that the economy fluctuates around a unique ‘natural rate’ of unemployment. In contrast, Keynes argued that there are multiple equilibrium unemployment rates.

Keynes was right. The economy fluctuates around an equilibrium that depends on business and consumer confidence and almost all of the possible equilibria are socially inefficient. If confidence is low, wealth will be low. If wealth is low, aggregate demand will be low. If aggregate demand is low, firms will hire fewer workers and unemployment will be high.

Fed policy makers are aware that there is wide variation in the unemployment rate from one recession to the next; but they attribute this variation to changes in the ‘natural rate’ of unemployment. In fact, there is nothing natural about the natural rate; it depends on the confidence of market participants.

To counteract the effect of swings in confidence the Fed should target stock market prices. Some economists have suggested that the Fed should raise domestic interest rates to prick stock market bubbles and lower the interest rate to prevent market crashes. This is not what I am advocating.

The Fed should target a stock market index in addition to its traditional role of setting the interest rate. My proposal allows the Fed to use variations in the fed funds rate to fight inflation and variations in the growth rate of a stock price index to manage confidence and select a high employment equilibrium. How would this policy operate in practice?

First: Define an index. In my Dec 30th FT piece I mentioned the S&P 500 as an example. It would be better to include every publicly traded stock, weighted by market capitalization and adjusted on a regular basis to reflect changes in the composition and size of firms.

Second: Buy shares in all publicly traded companies at current market prices in proportion to the weights by which they enter the index. To pay for these assets the Fed should issue three months bills, backed by the treasury. These securities would trade at the same price as three-month treasury bills since the two classes of assets would be perfect substitutes.

Third: Decide on the size of the Fed position in stocks. The Fed balance sheet was approximately $800 billion before the recent expansion. An initial capitalization of a further $800 billion in market securities seems like a good place to start. Since an operation of this kind will balance an asset with an offsetting liability it will leave the federal debt unchanged. In contrast, a fiscal expansion would increase government debt obligations that must be paid off by future generations.

Fourth: Establish a market in the newly created index and stand ready to buy and sell it at a price to be set at regular intervals. Since the price of the index needs to be coordinated with interest rate policy, an appropriate body to set the price would be the Fed open market committee (OMC).

Fifth: Choose a policy to adjust the price path of the index. At each meeting, the OMC would announce this path in addition to its announcement of a target for the fed funds rate. The chosen price path needs to reflect anticipated real output growth and anticipated inflation. Just as the fed funds rate is adjusted to target the inflation rate so the rate of growth of the index should be adjusted to target the unemployment rate.

In summary, Central Banks throughout the world should target stock price indices in addition to domestic interest rate targets. A fiscal expansion is necessary in the current situation, but fiscal policy alone is not enough: A stimulus package large enough to restore full employment will increase the debt burden on future generations to an unacceptable level. The time has come to implement a new monetary policy for the 21st century.


Originally published at the Financial Times and reproduced here with the author’s permission.

One Response to "A new monetary policy for the 21st century"

  1. Hardernut   January 16, 2009 at 2:37 pm

    Wow. There are a myriad of fatal flaws of such a plan, but I think the most obvious is the idea that confidence can be directly controlled through the manipulation of stock prices. How significant was the impact to confidence in the huge injections of cash by the government into the ‘healthy’ banks? When Warren Buffett put large amounts of cash into GE and Goldman with warrants for purchase at $22.25 and $115 respectively, this infused confidence in these companies for all of about five days. The market/investing population can see beyond any artificial attempt to manipulate perception. If Mr. Buffett’s 8 billion dollars in only two companies couldn’t hide the truth about these companies and the economy as a whole, why would 100 times that amount have any impact on the market as a whole. The difference is whether you think confidence leads or lags economic reality. The truth is that it does both, which is why it can be self reinforcing. However, confidence doesn’t turn south until there is good reason to turn south. This turn in confidence can accelerate the rate of decline of real economic activity, but it doesn’t precipitate it. The same goes for recovery. Confidence turns around when it is clear that we have regained our footing. If confidence improves quickly, the rate of recovery will be faster than if confidence remains subdued. However the turnaround must be triggered by actual economic health, not by a government program to inject money into the stock market to make people think the turnaround has arrived. Unless we want to convert to a fully planned economy, good analysts will be able to look at equities, project expectations of the future, and fairly determine prices. No government plan will prevent this, short of a total planned economy with restrictions on the press and every other private enterprise capable of spreading accurate information.