How COVID-19 Kills Inflation

Yet again, everyone sees nothing but inflation following a sharp downturn. Just like after the 2008 financial crisis, policymaker, analyst and investor consensus looks for inflation to pick up in the years ahead. The argument goes that massive central bank intervention coupled with large fiscal deficits will finally see the printing of money reach the real economy. Throw in populist governments that will ensure workers’ bargaining power and you have the perfect recipe for meaningful inflation.

But these arguments confuse the symptoms for the underlying cause. The reason for the policy interventions is an incredibly weak economy – surely something that would not be inflationary. Moreover, the structural forces COVID has brought about will, if anything, dampen the one area where inflation has been present: services.

Labour Market

The US unemployment rate is two to three times higher than it was at the start of the year. A broader measure of labour market health, the employment ratio (people employed as share of working population), is around 75% compared with 80% at the start of the year (Chart 2). These are levels last seen in the years after the 2008 financial crisis, when inflation struggled to stay above 2%.

But more than that, there is also the question of whether the labour market really matters for inflation. After all, central banks including the Fed appear to have given up on the Phillips curve – the supposed relationship between unemployment or output gaps and inflation. We can see this in the divergence between wage growth and inflation (Chart 3). We will only learn in coming years if this is due to productivity gains. But at the very least, the relationship between wages and inflation is unstable and unreliable.

This also begs the question of whether central banks actually want inflation to rise. They care primarily about wages. As long as those are rising, they could believe they have met their objectives. Indeed, central banks including the Fed have increasingly focused on whether economic growth is inclusive of all segments of the labour force.

Fiscal and Debt

Outside of the central bank, the fashionable meme among economists and investors is that loose fiscal policy will lead to inflation. Larger deficits lead to more debt, which will require inflation to erode their value over time – better inflation than defaulting or having to overtly raise taxes. The problem here is that there appears to be little evidence of higher debt levels being associated with inflation. Since 2008, there has been no connection between levels of debt and inflation either in developed (DM) or emerging markets (EM).

For example, over the past 10 years, in DM, Norway and Australia have averaged the highest inflation rates, yet they have the lowest debt levels. Meanwhile, Japan and Switzerland have had the lowest inflation – the former has low debt, while the latter has the highest debt levels in DM. The same paradox holds in the 10 years before the 2008 financial crisis (Charts 4 and 5). Therefore, debt often reflects a weak economy, which in turn is disinflationary.

One would expect the debt-to-inflation relationship to hold in EM, though. However, even there the relationship is weak. The only pattern is that Turkey and Russia seem always to have high inflation (Charts 6 and 7).

Services

The larger issue is the nature of the current economic shock. The pandemic has directly impaired peoples’ ability to work in close proximity and so has impacted the services sector. This defies the standard way of analysing business cycles to which most are accustomed. Unlike any other modern slowdown, it is the services sector that has seen the biggest declines (Chart 8). This is important because the only source of inflation in recent decades has been in services – it has averaged around 2.5%, while goods inflation has been zero (Chart 9).

Though difficult to measure right now, the likely consequence is an acceleration of technology solutions to replace services sector personnel. Software and hardware are immune to viruses. We can see this acceleration already with online meal delivery replacing restaurants and the continued shift to online retail away from shops.

But other sectors such as healthcare and education, which comprise over one tenth of core inflation, will likely see their costs driven down as scalable online services proliferate (Chart 10). Currently, medical costs have spiked around COVID-related issues, but this is already passing.

The biggest component of US inflation is shelter, which makes up 40% of core inflation. It is usually proxied by rental costs. Shelter has generally hovered around 3% (Chart 11). It has already started to fall this year, but the larger question is, what will be the permanent adoption of remote working? If it is significant, as is likely, this should depress shelter costs because people would have greater choice over where they live. So they can shop around for lower rents. This could end up being the most profound effect of COVID-19 on working practices.

Together, all of this points to services inflation coming under sustained downward pressure in years to come. This in turn will likely keep a lid on overall inflation.

Bottom Line

The main consequence of COVID-19 is not that central banks are printing money. Nor is it that government deficits have exploded. Rather, it is the direct impact on the services sector. This is the largest part of the economy and has been the main source of inflation in recent decades. However, a collapse in demand here coupled with increased adoption of technology to replace in-person activities should exert significant disinflationary forces on the economy. As a result, the consensus view that inflation will rise is probably misplaced.