By chance I today found an ECB working paper from 2004 – “The Great Depression and the Friedman-Schwartz hypothesis” by Christiano, Motto and Rostagno.

Here is the abstract:

“We evaluate the Friedman-Schwartz hypothesis that a more accommodative monetary policy could have greatly reduced the severity of the Great Depression. To do this, we first estimate a dynamic, general equilibrium model using data from the 1920s and 1930s. Although the model includes eight shocks, the story it tells about the Great Depression turns out to be a simple and familiar one. The contraction phase was primarily a consequence of a shock that induced a shift away from privately intermediated liabilities, such as demand deposits and liabilities that resemble equity, and towards currency. The slowness of the recovery from the Depression was due to a shock that increased the market power of workers. We identify a monetary base rule which responds only to the money demand shocks in the model. We solve the model with this counterfactual monetary policy rule. We then simulate the dynamic response of this model to all the estimated shocks. Based on the model analysis, we conclude that if the counterfactual policy rule had been in place in the 1930s, the Great Depression would have been relatively mild.”

It is interesting stuff and you would imagine that the model developed in the paper could also shed light on the causes and possible cures for the Great Recession as well.

Is it only me who is reminded about 2008-9 when you read this:

“The empirical exercise conducted on the basis of the model ascribes the sharp contraction of 1929-1933 mainly to a sudden shift in investors’ portfolio preferences from risky instruments used to finance business activity to currency (a flight-to-safety explanation). One interpretation of this finding could be that households . in strict analogy with commercial banks holding larger cash reserves against their less liquid assets . might add to their cash holdings when they feel to be overexposed to risk. This explanation seems plausible in the wake of the rush to stocks that occurred in the second half of the 1920.s. The paper also documents how the failure of the Fed in 1929-1933 to provide highpowered money needed to meet the increased demand for a safe asset led to a credit crunch which in turn produced deflation and economic contraction.”

The authors also answer the question about the appropriate policy response.

“We finally conduct a counterfactual policy experiment designed to answer the following questions: could a different monetary policy have avoided the economic collapse of the 1930s? More generally: To what extent does the impossibility for central banks to cut the nominal interest rate to levels below zero stand in the way of a potent counter-deflationary monetary policy? Our answers are that indeed a different monetary policy could have turned the economic collapse of the 1930s into a far more moderate recession and that the central bank can resort to an appropriate management of expectations to circumvent – or at least loosen – the lower bound constraint.

The counterfactual monetary policy that we study temporarily expands the growth rate in the monetary base in the wake of the money demand shocks that we identify. To ensure that this policy does not violate the zero lower-bound constraint on the interest rate, we consider quantitative policies which expand the monetary base in the periods a shock. By injecting an anticipated inflation effect into the interest rate, this delayed-response feature of our policy prevents the zero bound constraint from binding along the equilibrium paths that we consider. At the same time, by activating this channel, the central bank can secure control of the short-term real interest rate and, hence, aggregate spending.

The conclusion that an appropriately designed quantitative policy could have largely insulated the economy from the effects of the major money demand shocks that had manifested themselves in the   late 1920s is in line with the famous conjecture of Milton Friedman and Anna Schwartz (1963).”

So what policy rule are the authors talking about? Well, it is basically a feedback rule where the money base is expanded to counteract negative shocks to money-velocity in the previous period. This is not completely a NGDP targeting rule, but it is close – at least in spirit. Under NGDP level targeting the central bank will increase in the money base to counteract shocks to velocity to keep NGDP on a stable growth path. The rule suggested in the paper is a soft version of this. It could obviously be very interesting to see how a real NGDP rule would have done in the model.

Anyway, I can highly recommend the paper – especially to the members of the ECB’s The Governing Council and I see no reason that they should not implement a rule for the euro zone similar to the one suggested in the paper. If the ECB had such a rule in place then Spanish 10-year bond yields probably would not have been above 7% today.

PS Scott Sumner has been looking for a model for some time. Maybe the Christiano-Motto-Rostagno model would be something for Scott…