Rejoinder

In my critique of ‘monetary-base monetarism’ (Congdon, 2023), I took the offending set of ideas to pivot on the claim that the change in the monetary base by itself determines inflation. I concentrated on the weaknesses of that approach to macroeconomic analysis. I did not develop my favoured alternative view, which might be called ‘broad-money monetarism’, where a key proposition is that changes in the quantity of money, broadly defined, determine inflation. But a potential debate between the two accounts of monetarism was implicit in my article.

Important issues are at stake. Consider the numbers for the four years to mid-2012, given in Table 1. In the Great Recession central banks adopted measures which had contrasting effects on the different money aggregates. Under the Basel III rules, being introduced from September/October 2008, they imposed higher capital-to-risk-asset ratios on the commercial banks. The change in capital regulation caused these banks to shrink risk assets. This would have led to a contraction in broad money if nothing else had been done.¹ Fortunately, something else was done. From late November 2008 the Federal Reserve embarked on large-scale purchases of assets from private sector non-banks (or ‘quantitative easing’), financed by the issue of cash reserves to the commercial banks. Such purchases added to bank deposits and stopped broad money from falling. But the issue of cash reserves led to an explosion of the monetary base. The Bank of England followed a few months after the Fed, with its QE programme beginning in March 2009.²

Which concept of ‘money’ was the correct one to use in macroeconomic prognosis? Should economists have worried – in the four years under discussion – about the inflation perhaps threatened by the trebling of the base in the USA and its quadrupling in the UK? Or should they instead have been concerned that broad money was barely growing at all, signalling very low inflation and possibly even deflation?

On 15 November 2010, 23 top-ranking, mostly American economists wrote an ‘Open Letter to Ben Bernanke’ (then the Fed's chairman), which was reported in the Wall Street Journal (2010). Several of them had purported monetarist affiliations. The letter contained the following sentences:

To some extent my differences from Scott Sumner may be semantic, with the two of us merely – to use his words – ‘talking past each other’. But I suspect that substantive disagreements remain. I was dismayed that in his reply to my article he said:

It was exactly this sort of reasoning which provoked my article. Let us recall what actually happened in and after the Great Recession. In the USA the monetary base trebled and in the UK it quadrupled, even more dramatic surges than Sumner's doubling, yet in both countries inflation in the early 2010s was the lowest since the 1950s. Not only were the signatories to the Open Letter to Ben Bernanke wrong, but also their forecasting error went far to discredit monetarism in the public debate.

In my view, the Open Letter to Ben Bernanke was an exercise in monetary-base monetarism. I did comment on it briefly in my article, mentioning it close to a reference to George Selgin (Congdon, 2023, p. 187). In an article by Selgin in the February 2021 issue of Economic Affairs, he dismissed “dire warnings” about the USA's inflation prospects made by myself and other people in spring 2020. Instead he took comfort from “bond speculators”, since on 21 October 2020 the ten-year breakeven inflation rate derived from relative bond pricing was a tiny 0.142 per cent (Selgin, 2021, p. 13). I am afraid Selgin's February 2021 complacency – and that of the bond speculators – has been wrong too.

I was unhappy about Selgin's remarks and the 2010 Open Letter, as well as about Sumner's previous contribution to Economic Affairs (Sumner, 2022). They all appealed, if in different ways, to the monetary base as crucial to the inflation outlook. In his reply to my article, Sumner accepts the trivial role of currency in transactions nowadays. He nevertheless tries to restore plausibility to analyses focused on the monetary base by insisting on the stability of the ratio of the public's cash to national expenditure. His position is that the stability of the ratio of the public's cash to national expenditure ensures that national expenditure must adjust – indeed, must adjust equi-proportionally – when the quantity of currency changes.

But I had a full section explaining that, in practice, the institutional arrangements in modern economies are for the quantity of currency to adjust to the public's requirements, not the other way round. Further, such a high proportion of the public's cash is held in the underground economy that the quantity of cash may be more relevant to transactions in illicit substances than to those in legitimate national expenditure.

Sumner was very kind to me in one of his blogs early in 2023 (Sumner, 2023). He noticed that – along with Robert Hetzel, senior economist at the Reserve Bank of Richmond from 1975 to 2018 – I had been largely correct in my forecasting and research on the leading economies in recent years. In fact, I was also correct in my forecasting and research on the other relatively recent example of cyclical instability, the Great Financial Crisis and the ensuing Great Recession from August 2007.

There is no secret to these successes. The essence of broad-money monetarism is, to repeat, that changes in the quantity of money, broadly defined, determine inflation. Strictly speaking, this does of course need to be elaborated in more detail. Changes in broad money relative to the trend growth of output should approximate the inflation rate over the medium term. In qualification, an allowance may need to be made for a trend change in the ratio of money to output due to such forces as the increase in financial sophistication as economies grow.

All the same, the analytical focus should be on broad money, not the monetary base. Table 2 shows that the monetary base grew very rapidly in the Covid-19 cycle, just as it did for a few years from late 2008. But the behaviour of broad money was very different in the two episodes. Whereas in the four years to mid-2012 the growth rate of M3 broad money in the USA was 1.2 per cent a year and of M4x broad money in the UK 1.6 per cent a year, the corresponding figures for the three years to mid-2022 were 12.4 per cent in the USA and 8.4 per cent in the UK. The inflation sequels to the central banks' actions were poles apart in the two episodes because of this contrast in the growth rates of broad money.