Money velocity: the macro driver of GDP

I've read lots of commentary about what caused the larger-than-expected increase in GDP in the fourth quarter. Most observers note that the main driver of the 5.7% annualized growth rate was a decline in the pace of inventory liquidation, which contributed 3.4%. From this they conclude that the economy's strength was not that impressive, especially considering that the growth in personal consumption expenditures only contributed 1.4%. This may explain the market's weakness today.

I'm going to step above the fray and look at one of the macro sources of the growth in GDP. The one that immediately jumps out is the velocity of M2. (The inverse of velocity, i.e., the demand for money, is shown in the second chart above.) Money velocity rose in the fourth quarter at an annualized rate of 2.9% (and the demand for money fell by a similar amount). Some of the money that people had hoarded in the previous year (under the mattress, in money market accounts, or in bank deposits) was taken out and spent in the fourth quarter. This is the physical manifestation of a general rise in confidence; people came to realize that the precautionary money balances that they had built up were more than enough given all the signs of improvement in the economy. If this process were to continue until money demand returned to where it was pre-crisis, that would add a whopping 13% to GDP (assuming M2 doesn't decline in the interim).

One thing helping this process is the Fed. By keeping short-term interest rates close to zero, the Fed is trying desperately to convince people that they don't need all the money that they have been holding onto. It was the huge increase in the demand for money that prompted the Fed to expand its balance by purchasing $1 trillion of Treasuries and MBS; the Fed was simply reacting, as it should, to an unprecedented increase in the demand for money. As the demand for money declines further, the Fed will sooner or later be able to withdraw its injections of reserves into the banking system, and that needn't result in any major economic dislocations. What might happen, of course, is that the Fed may well decide to wait too long to reverse course, in which case we would end up with higher-than-expected inflation.

The other thing helping this process is the recovery itself. Note in the second chart that money demand usually tends to decline after a recession. Money demand rises in advance of and in the early stages of recession, reflecting rising fears and the demand for precautionary balances. It then declines as animal spirits return. This process was particularly intense in the recent recession, since it was mainly a financial-panic-induced recession. As the recovery progresses and confidence returns, the demand for money has plenty of room to decline (i.e., money velocity has plenty of room to rise), and this will be an important source of growth in nominal GDP in the years to come.

A final note: rising M2 velocity can actually do two things to GDP: it can increase nominal GDP and/or real GDP. So far it has mainly contributed to propel real GDP higher (since the GDP deflator has been less than 1% in the third and fourth quarters), but rising velocity could certainly contribute to a rise in the general price level at some point in the future.