Inflation Risk?

Inflation is on the rise again in the industrialised world. This has led to fears of a sustained surge in inflation. This article argues that while such fears may make sense in the US, they do not in the eurozone, where the monetary-fiscal policy mix has been much less expansionary than in the US. The fear expressed by some that the monetary overhang from the large injections of liquidity through quantitative easing might lead to inflation in the eurozone does not stand up to scrutiny either. The conclusion offers some observations on the monetary operating procedures in the ECB. It argues that in the future, when interest rates rise again, the ECB risks transferring all (and even more) of its profits to the banking system. This article proposes a way to avoid this unacceptable outcome.

Infl ation is on the rise again in the industrialised world. In the US, the infl ation rate reached 5.4% in June 2021; in the eurozone infl ation was estimated to be 1.9% in the same month. Should we worry about this surge in infl ation? In order to answer this question, one must answer a preliminary one: What are the causes of this increase in infl ation?
The short answer is that it has everything to do with the economic recovery from the pandemic that hit the world in 2020. This recovery has been made possible by two things: the release of excess savings accumulated during 2020 and strong expansionary fi scal and monetary policies, especially in the US, where the mix of fi scal and monetary policy has been extraordinarily expansionary. While US GDP growth declined by 5.8% in 2020, the budget defi cit increased by 10.7% of GDP. Thus, for every 1% decline in GDP, the US fi scal authorities allowed the budget defi cit to increase by almost 2% of GDP. This compares with a decline in eurozone GDP of 7.5% and an increase in the budget defi cit of 6.4%. In other words, for every 1% decline in GDP in the eurozone, fi scal authorities allowed the budget defi cit to increase by less than 1% of GDP. Thus, it appears that the fi scal expansion was about twice as strong in the US compared to the eurozone.
A similar story emerges from monetary policies. Both the US Fed and the ECB intensifi ed their asset purchases since the start of the pandemic, thereby fl ooding the markets with liquidity. But again, the amount of asset purchases was signifi cantly higher in the US than in the eurozone. Since January 2020, the US Fed bought approximately $2.6 trillion in government securities, about twice the amount bought by the ECB, €1.3 trillion, thereby increasing the amount of liquidity in the US signifi cantly relative to the eurozone.
It appears that since the start of the pandemic, monetary and fi scal policies have been expansionary in both the US and the eurozone, but that the intensity of this expansion has been much higher in the former. This has led some to argue (e.g. Summers, 2021;Blanchard, 2021) that the US fi scal-monetary policy mix is excessive and exceeds the capacity of the economy to absorb it without major price increases. This probably explains why the surge in infl ation is much higher in the US than in the eurozone. I conclude from this that while the risk of infl ation is real in the US, it is much less so in the eurozone, where monetary and fi scal expansion does not appear to be hitting capacity constraints in the economy.
Diehard monetarists will probably object to this conclusion. Quantitative easing (QE) performed by the ECB since 2015 has created a monetary overhang. Does this overhang not create a risk for future infl ation? Did Milton Friedman not teach us that infl ation is always and everywhere a monetary phenomenon ? The fact is that the QE programme of the ECB has led to a surge in the money Inflation stock in the eurozone. This in the end will lead to a surge in infl ation, according to this monetarist analysis.
Let us look at the numbers. This is done in Figures 1 and  2. Figure 1 shows the evolution in time of infl ation and the growth rate of M3 in the eurozone. We observe fi rst that the relation between these two variables is weak. This is also confi rmed by Figure 2, which shows the same numbers in a scatter diagram, with M3 growth on the horizontal axis and infl ation on the vertical one. A simple regression line, which is almost horizontal, shows that there is no signifi cant relation at all between money growth and infl ation. Of course, one might argue that this fi gure only measures the short-term relation between money growth and infl ation, while the monetarist theory tells us that this relation only holds in the long run. De Grauwe and Polan (2005) looked at the long-run relationship between money growth and infl ation (over a 30-year period) for about 100 countries and concluded that, except for countries experiencing hyperinfl ation, there is no signifi cant longrun relation between money growth and infl ation.
Coming back to Figure 1, we observe two waves in the expansion of the money stock in the eurozone. There is the pre-fi nancial crisis wave of monetary expansion that reached yearly growth rates of more than 10% from 2006 to 2008, and there is the post-2015 wave that led to increases in the money stock exceeding 10% in 2020. Both waves of monetary expansion that were sustained over fi ve years or more do not seem to have affected infl ation much (confi rming our previous observations). The pre-fi nancial crisis wave, however, teaches us that the risk was elsewhere, i.e. in the fi nancial markets. Underlying the strong growth in M3 there was the surge in bank credit that fuelled real estate and other bubbles. Ultimately, this led to the banking crisis of 2007-08. Will history repeat itself?
From the preceding, I conclude that the risk of a significant surge in infl ation in the eurozone appears to be limited. This contrasts with the US, where the risk appears to be more serious. However, the large expansion in liquidity both in the US and in the eurozone creates potential risks in fi nancial markets. These might affect the future conduct of monetary policies.
At some point when infl ation returns to its normal level, central banks will have to raise interest rates. These interest rate increases create two potential problems that may act as obstacles for these interest rate increases.
The fi rst one relates to the problem just identifi ed. An interest rate increase may destabilise fi nancial markets, in particular the bond markets. Interest rate increases will lower bond prices and affect the balance sheets of many fi nancial institutions (banks, pension funds, insurance companies). Can such an increase in the interest rate be done in an orderly way that minimises the risks of a fi nancial crisis? There is no easy answer to this question. The uncertainty about this may induce central banks to wait too long to raise interest rates.
The second obstacle derives from the way central banks conduct monetary policies. For about 20 years, major central banks (including the Fed and the ECB) have switched to an operating system in which they remunerate banks' reserves held at the central bank. This means that when central banks have to raise the interest rate, it  Inflation will affect their profi t and loss account. This effect will be particularly pronounced for the ECB. At the end of 2020, bank reserves held at the ECB amounted to a massive €3.5 trillion (mostly the result of the large bond purchase programmes of the ECB). A 1% increase in the interest rate means that the ECB will have to pay out €35 billion to the banks holding these reserves. Going back to normal implies that the short-term interest rate would probably increase to 2%, leading to interest payments to banks of €70 billion. This represents about three times the annual profi ts recorded by the ECB, €10-€20 billion during 1999-2017 (Chiacchio et al., 2018), and thus would lead to massive losses of that institution. This could make the ECB reluctant to raise interest rates when this becomes necessary.
There is no easy way out of this problem. One possibility would be for the ECB to massively sell government bonds again, thereby reducing bank reserves. This would not only raise long-term bond rates signifi cantly but, more importantly, it would imply that the government debt relief that is implicit when the ECB holds government bonds on its balance sheet would be reversed. This would put new pressure on member states' budgets, especially those with large outstanding debt levels. It could also trigger a new sovereign debt crisis.
My favourite solution to this conundrum would be for the central bank to return to the old operating procedure of not remunerating bank reserves. There is no good economic argument why banks should be remunerated for holding liquid assets. Holders of banknotes are not remunerated either. The current operating procedure that pays bankers for holding liquid reserves will in normal times ensure that all (if not more) of the seigniorage gains that should go to the Treasury are actually transferred to fi nancial institutions. This is an unacceptable outcome. The old operating procedure together with the use of minimum reserve requirements can be made to work again (as it did in the past) and provide for a better tool to conduct monetary policies in normal times.