Current US economic weakness was predicted by a fall in real money supply expansion in late 2011 / early 2012 – see previous post. Real money growth has stabilised recently at a respectable historical level but pessimists argue that this will not prevent a recession because of a decline in the velocity of circulation.

Velocity can be thought of as the flow of GDP income generated by each dollar of the money supply. The pessimists cite the ratio of current nominal GDP to the current M2 money supply in support of their claim; this ratio has plunged since early 2011, reaching its lowest level since 1950 – see chart.

There are, however, two problems with the pessimists’ argument. First, the money supply affects nominal GDP with a lag – of about six months for activity and two years for prices, according to the Friedmanite rule. Velocity, therefore, should be measured by comparing current nominal GDP with the money supply between two and eight quarters earlier. As the chart shows, the ratio calculated using a five-quarter lag – a reasonable compromise – is still above its 2010 low, though may fall as a bulge in M2 in mid 2011 enters the calculation (unless nominal GDP accelerates).

Secondly, velocity should be measured with reference to a broad monetary aggregate that is unaffected by voluntary shifts between different types of account. The M2 surge in mid 2011 was caused by an inflow from large time deposits and institutional money market funds – components of the old M3 definition. The velocity of a broader “M2+” measure incorporating these “wholesale” forms of money – and calculated using a five-quarter lag – has recovered significantly from a 2010 low and may continue to trend higher (since M2+ has been growing much more slowly than M2).

The view here remains that the probability of a US recession in 2012 is higher than in 2010 and 2011 but still well below 50%, based on recent continued expansion of real narrow money.