Wednesday, July 16, 2014

QE and the economic risks of underfunding - by Andrew Smithers

In the past governments have funded their deficits – for example, they have borrowed in the bond market rather than through treasury bills. This is despite the fact that, for the past 80 years, the rate of interest on bonds has been greater than that on Treasury bills; that is, we have had an upward sloping yield curve. 
I suggested in a recent blog that this was because governments correctly perceived that there were considerable economic risks in not funding, and that it was worth paying the additional cost to avoid these risks. Quantitative easing, which is a form of underfunding, must therefore have increased these risks. Defenders of QE need either to argue that these risks have not risen or that the benefits we have received from QE outweigh the rise in risks. To be consistent, those who hold that no additional risks have been incurred must now hold that governments should not have funded in the past and must now stop. But their silence is deafening, and such views are implausible, being held, I think, in the hope of dissuading discussion rather than from any conviction that they would survive much debate. 
The standard defence of QE is that it has benefited the economy more than it has increased the risks to it. Others doubt this. For example, in a recent speech Masaaki Shirakawa, the previous governor of the Bank of Japan, said: “The emerging consensus seems to be that, even though unconventional monetary policy affects prices of financial assets, its effect on real economic activity and hence the output gap is rather limited and uncertain.” But this view is not held by either the Bank of England or the US Federal Reserve. We are therefore in disputed territory, and there is always a risk that anxiety to defend past actions will prevent sufficient attention being given to mitigating the consequent risks. Whether or not QE has been a good thing or a bad thing, it is a risky thing; and policy should now be seeking to reduce the dangers that it has increased. 
QE or any other failure to fund is often referred to by its critics as printing money, and this is not an unreasonable description. When governments sell bonds, those who buy them have less money; and, when they don’t, they have more. The money from the sales of bonds to the Fed has been deposited with banks, and they have in turn increased their deposits with the Fed. The direct increase in the supply of money that has resulted from QE has been small. But it has greatly increased the power of US banks to expand money supply in the future. There is clearly a risk that this opportunity will at some stage be taken, and it is vital that this is prevented.
When the Fed buys bonds, its assets and liabilities rise; and, for the most part, the liabilities consist of the deposits of commercial banks and are reflected in the rise in the monetary base. Chart one shows that the ratio of the monetary base to M2 was between seven and 12 to one until recently, and has now fallen to under three. For the ratio to return to its average of nine, without a decline in the monetary base, money supply would need to triple. 
The relationship between money supply and inflation can, to put it mildly, be hotly disputed. My own view is that money matters and that too much will produce inflation, but that the relationship is complex and unstable, as I illustrate in chart two. This vague assessment is probably in line with that of most economists, though the matter is too contentious for it to be safely described as the consensus view.