Four years ago, at the worst point in the Great Recession, several leading American economic gurus said that the Fed’s easy monetary policy risked higher inflation which might imperil the recovery. In some of the early commentaries from International Monetary Research (in the spring and summer of 2009), I argued that these gurus – who included the revered Alan Greenspan and Martin Feldstein, a prominent adviser to President Reagan – were seeing ‘ghosts’. Also at that time Liam Halligan of Prosperity Capital Management used his column in The Sunday Telegraph to attack ‘quantitative easing’ (i.e., the large-scale creation of money by the state) as aggressive ‘money printing’ which would lead to a sharp rise in inflation.
In a letter to the Financial Times published on 19th June, I showed that the inflation jeremiahs had been wrong. The last five complete years (i.e., the five years to 2012) saw the lowest increases in nominal GDP in all the G7 economies since the deflationary 1930s. The jeremiahs of 2009 made a serious analytical mistake. They thought that inflation was caused by excessive growth of the monetary base, not of the quantity of money broadly-defined to include all bank deposits. In future economic commentators should pay more attention to the quantity of money as such and de-emphasize the monetary base by itself.
In the current cyclical episode large-scale official asset purchases were inaugurated by the Federal Reserve in November 2008. The Bank of England’s announcement of an enormous gilt purchase programme, which became known as ‘quantitative easing’ or QE, in March 2009 was a further development of the theme. In short order several commentators – not just in the UK but around the world – lamented that QE amounted to ‘money printing’. They predicted that an inevitable result would be a big surge in inflation. (Liam Halligan in the UK was particularly shrill with warnings along these lines in his Sunday Telegraph column, while in the USA the inflation jeremiahs included such distinguished figures as Martin Feldstein, Alan Greenspan, James Grant [of the Grant Interest Rate Observer] and Allan Meltzer. For my views on their inflation pessimism, see the attempt at a ‘debate’ with Halligan in my weekly e-mail of 26th October 2012 ‘Halligan wrong on inflation’, and on the Americans 21st May 2009 ‘Inflation jeremiahs are wrong’ and 2nd July 2009 ‘Greenspan sees inflation ghosts’.)
Assuming that the trend growth rate of output is constant, a big surge in inflation must mean acceleration in the rate of increase in nominal gross domestic product. (The increase in nominal GDP is equal to the increase in real GDP plus the inflation rate.) It follows that Halligan in the UK and the Feldstein-Greenspan-Grant-Meltzer group in the USA were expecting in 2009 that the next few years would see higher growth of nominal GDP than before. (I cannot otherwise make much sense of what they were saying.)
What, in fact, has happened to the rate of increase in nominal GDP in the G7 advanced nations since the start of the Great Recession? To prepare the chart above I calculated the annual increases in nominal GDP since 1980 for the G7, using the IMF’s database. I then estimated five-year averages of these annual increases, with the first such value being of course the five years to 1985, which is when the chart begins. The chart has a compelling message. With one (very minor) exception (discussed at the end), the five years to 2012 – the period, in other words, in which QE operations had been adopted by the USA and the UK, and openly discussed in the context of the Eurozone and Japan – saw the lowest increases in nominal GDP of any five-year period since the 1980s. It is undoubtedly true that increases in nominal GDP were higher in the four decades between the 1930s and the 1980s, since the 1940s included the Second World War, and the 1950s, 1960s and 1970s were an era of inflationary growth. The conclusion can therefore be drawn, ‘the period in which QE has been a prominent method of monetary policy has been accompanied by the lowest increases in nominal GDP since the deflationary 1930s’.
Evidently, Halligan and the Feldstein-Greenspan-Grant-Meltzer group were wrong. As it happens, they were less wrong on inflation than in their more general analysis. The supply- side performance of the leading economies in recent years has been poor, with most estimates of the annual trend output growth rate being scaled back by between 1% and 2% relative to the average growth rate in the three closing decades of the 20th century. Consequently, bad supply-side performance has affected inflation. Inflation has been higher than would otherwise have been expected from any particular stance of monetary policy.
Why were the inflation jeremiahs mistaken? After all, QE does involve large-scale money creation by the state, even if the characterisation of such money creation as ‘printing’ is naïve. The answer is that the relevant money aggregate in any organized account of the determination of asset prices and national income has to be one that is broadly-defined and includes all bank deposits. QE has been accompanied by huge surges in ‘the monetary base’ (i.e., notes and coin in circulation, and banks’ cash reserves with the central bank). But it is the quantity of money, broadly-defined, that should be the commentariat’s focus, not the monetary base.
In the last few years the growth rate of the quantity of money has been held back across all the leading economies – and not just in the USA and the UK, which are in fact the only two monetary jurisdictions that had uninhibited QE in the 2008 – 12 period – by official demands for more bank capital, less inter-bank funding, higher ratios of liquid assets to banks’ total assets, and so on. The growth rate of the quantity of money has been the lowest since the 1930s and so also has been the growth rate of nominal GDP. Am I alone in finding a familiar pattern here? I finish this note by republishing my letter to the Financial Times on the next page and then reviewing each of the G7 countries in turn.
(What about the exception? In Germany the increase in nominal GDP in the five years to 2009 was 1.6%, compared with 1.8% in the five years to 2012. Otherwise my statement in the letter to the Financial Times was correct. To repeat, the last five complete years have seen the lowest increases in nominal GDP in the G7 countries since the 1930s.)
My letter to the Financial Times (published on Wednesday, 19th June)
Letter written on 14th June, 2013
Martin Wolf (“The overstated inflation danger”, June 12) correctly identifies a gross misjudgement made by some economic commentators three or four years ago. This misjudgement was to propose that the then large increases in the monetary base in several leading countries would result, within an operationally relevant period (for example, by 2013 or 2014), in sharply rising inflation or even hyperinflation. But he understates the scale of the mistake those commentators made.
According to standard theory, the relationship between changes in money and nominal national income ought to be tighter than that between changes in money and prices. A remarkable feature of the last few years brings out the extent of the error made by the inflation doomsters of early 2009. This is that the increases in nominal GDP in every one of the G7 economies were lower in the five years to end-2012 than in any previous five-year period since the 1930s. Economists who claim a direct link between the monetary base and nominal expenditure have made fools of themselves.
However, the massive divergence between the monetary base and nominal GDP does not mean that money is irrelevant to inflation. The relevant money aggregate in any plausible account of national income determination has to be one which includes all relevant money balances. In practice this means that broad money – dominated by bank deposits in the hands of genuine non-bank private sector agents – is the measure that matters. (Note that such deposits are not included in the monetary base, as usually defined.)
The Great Recession fully confirms the power of money when understood in this broader sense. Across the G7 the five years to end-2012 in fact saw the lowest increases since the 1930s in the quantity of money, broadly-defined, as well as the lowest increases in nominal GDP. The traditional medium-term relationship between the quantity of money and income has proved resilient in the Great Recession, just as it has in other episodes of macroeconomic instability and upheaval. The central point here cannot be emphasized too strongly. The quantity of money is the relevant variable in national income determination, not the monetary base by itself. The quantity of money and the monetary base are very different things.