In a recent and very important contribution, Larry Summers warned about the risks attached to the $1.9 trillion fiscal stimulus pushed by the new US administration (Summers 2021). Specifically, he is concerned that " macroeconomic stimulus on (this) scale … will set off inflationary pressures of a kind we have not seen in a generation ". His assessment was shared by Olivier Blanchard – and backed by specific calculations – in a series of tweets.1
Almost at the same time, a somewhat different approach has been taken by Paul Krugman in a new blog (Krugman 2021). Looking at the flatness of the Philips Curve – as illustrated by recent empirical work – Krugman judges that "even a very hot economy only leads to modest inflationary overheating".
These contributions relaunched a debate that is key for the conduct of fiscal and monetary policy following the Covid crisis. It underlines the important challenges that economists still face in understanding inflation and its dynamics in the present environment
What are the differences
Both Summers and Blanchard base their judgement on the size of the projected stimulus compared to the existing output gap. Depending on estimates, the Biden plan could amount to three to five times the remaining (negative) output gap in the US. Its impact could be amplified by the release of pent-up demand once the Covid restrictions are lifted. It is natural to assume that such a disproportion will create inflation pressures – unless fiscal multipliers are extremely small. Although they don't allude to the Phillips curve, Summers and Blanchard seem to assume that it is normally sloped when the economy gets close to full employment. Tellingly, they do not mention inflation expectations and their ‘anchoring’ – a concept that is central to the analytical toolkit and language deployed by central banks when looking at inflation.
In contrast, expectations are at the centre of Krugman's analysis. He pictures inflation as a succession of long-term waves (regimes) driven by expectations. These regimes are very robust and expectations well anchored. That robustness was apparent, for example, with the ‘missing disinflation’ that followed the Great Recession when, despite a huge negative output gap, inflation remained relatively stable (Belz and Wessell 2020).
But inflation is also subject to (brutal) regime shifts. Through those shifts, expectations dominate inflation dynamics, much more than any demand or supply shock. Krugman goes as far as to revisit Volcker's disinflation, wondering whether it owed less to demand contraction (and subsequent unemployment) than to Volcker's personal credibility in influencing expectations.
What does matter really?
Two distinct questions should be asked when considering the risk of inflation. First, what is the short-term impact of a shock? And second, will it affect the underlying long-run dynamics? Summers and Blanchard clearly see the short-term impact of the stimulus as very disturbing. A positive output gap of around 14% of GDP is incompatible with price stability. (They also regret that the package is biased towards income support at the expense of infrastructure development). Their reasoning seems to further assume that the stability of expectations will not resist a strong uptick in instant inflation. That is not so clear in Krugman's framework, where there is considerable persistence in inflation (and stability of expectations) as long as no regime shift occurs.
Looking at both analyses, one policy conclusion emerges. Whatever the immediate inflation impact, the objective is to avoid instant inflation triggering a regime shift and turning into a permanent, possibly accelerating, process.
It may be that an additional and crucial element has to be introduced to the debate. The main risk of a regime shift may not come from a sudden increase in inflation but from the uncertainty created the unprecedented (in peace time) level of public debt. Also, there are calls for closer and more permanent coordination between fiscal and monetary policies. In many quarters, central bank independence is considered to be somewhat obsolete. In this environment, doubts may arise on the reaction function of central banks and their ability to raise interest rates if necessary. That would be a true regime shift for inflation – the reverse of the Volcker shift postulated by Krugman.
This points to the need for a new, revitalised approach to central bank independence which would aim less at solving the time-inconsistency problem (i.e. eliminating the incentive to cheat) and more at preserving central banks’ unconstrained ability to take whatever measures are necessary and avoid fiscal dominance.
Conducting economic policy in times of crisis
There are more general lessons to be drawn. In times of crisis, policymaking must be schizophrenic. Deciders must at the same time tend to short-term risks and keep thinking about long-term imbalances. But they should not let those two parts of their brain interact too much with each other. They are vulnerable to two opposite errors. One is to project the present environment and measures into the indefinite future – for instance, jettisoning central bank independence because, right now, monetary and fiscal policy objectives converge. The other, symmetric mistake is it to let legitimate concerns over the long run limit or paralyse action in the shorter term – reducing stimulus today because of concerns about future debt and inflation. When the economy is path-dependent, too much caution in the short run may be very costly.
Performing that difficult balancing act is made easier by strong institutions. Institutions create a bridge and a link between the short and long horizons. If they are strong and trusted, they can be relied upon to preserve their capacity for action in any possible circumstances. They stabilise expectations and consolidate the existing regime. Their existence prevents short-term necessary excesses from spilling over and creating permanent instability. In the current environment, preserving and strengthening central bank independence may be as important as the size and modalities of policy actions.
Belz, S and D Wessel (2020), "What’s (Not) Up With Inflation?" The Brooking Institution, January.
Hazell, J, J Herreño, E Nakamura and J Steinsson (2020), “The Slope of the Phillips Curve; Evidence from US States”, Working Paper 28005.
Krugman, P (2021), “Stagflation revisited. Did we get the whole macro story wrong?”, 5 February.
Summers, L (2021), "The Biden stimulus is admirably ambitious. But it brings some big risks, too", 4 February.