US inflation: four phenomena

Base effects

In thinking about inflation, it is important to distinguish between one off increases in the price level that have a transient impact and more permanent increases in the inflation trend. As a direct consequence of the Covid-19 pandemic, inflation fell sharply last year as many parts of the economy experienced extreme stress. If economic activity continues to normalise as the year progresses, comparisons with the lows of last year become easier and inflation will rise. 

However, this should not be confused with an increase in the inflation trend. Some of the increase in inflation due to the base effect may well be attributed incorrectly to a more permanent rise in inflation and this is likely to be reflected in an increase in market sensitivity. One could, for example, imagine in such a scenario an increase in equity and bond market volatility. However, this should prove short lived as inflation subsides.

Distributional impact

A separate feature of the past year is that different parts of the economy have been impacted in very different ways by the coronavirus crisis. A corollary is that there has been an unusual widening in the range of inflation rates across the underlying components that feed into the aggregate.

It is interesting to note that the decline in inflation last year was driven by a collapse in prices in some parts of the economy, as evidenced by the sudden sharp drop in the lower end of the range, rather than a decline in all measures of inflation - the upper end of the range increased marginally. This is also captured in the difference between the median and the simple unweighted average inflation measure derived from the Cleveland Fed components and shown in Chart 1.

Chart 1. Different measures of inflation

Unsurprisingly the simple average tracks closely the weighted average or total measure of CPI inflation. As and when the lower end of the inflation range rises one would expect the simple and weighted average measures to increase toward the median.

This is one way in which we will be able to track whether any rise in inflation is simply a part of the normalisation of the economy or if there is a genuine increase in underlying inflationary forces. If the lower end of the range rises – signifying less deflation – whilst the upper end stays largely unchanged that would be consistent with normalisation. If, however, the upper end of the range rises at the same time as the gap narrows that would be consistent with a shift upward in the distribution of inflation and a higher trend.

CPI v. PCE

Most countries in the world use a Consumer Price Index to estimate the price level and the US is no different in this regard. However, a popular alternative in the US is the Personal Consumption Expenditures (PCE) deflator. Indeed, the Federal Reserve’s preferred measure of inflation is based on the core PCE deflator, which excludes food and energy. The differences can be important. Whilst the trends in the two measures are usually similar there can be meaningful differences in the level.

Chart 2 shows measures of inflation based on both indices. For much of the time the CPI measure exceeds the PCE measure although the gap between the two has been small over the period of time since the coronavirus crisis started. 

Chart 2. CPI and PCE Inflation

There are numerous academic articles exploring the differences between these two price indices. These differences fall into four categories:
 

• Weight: the same items are weighted differently in the aggregate
• Scope: some items are included in one measure but not the other
• Formula: there are technical differences in how the indices are constructed and the methodologies used 
• Other: anything else including factors such as seasonal adjustments

Of these four categories the most important is usually considered to be Weight followed by Scope. 

Take for example the energy weight. Whilst the direct weight in both price indices is small, energy goods are prone to occasional large movements so there are times when it influences the aggregate. In such situations it is clear the CPI measure will be more impacted since its exposure to energy goods is two-thirds greater than for the PCE deflator. 

Capacity constraints

The coronavirus crisis was associated with a sudden stop in many business activities and a sharp rise in unemployment, despite mitigating government programs. While it is possible for inflation to rise even if growth is weak, such as happened in the 1970s, it is unusual. A more typical process is one in which the economy experiences capacity constraints that then translate into higher wages and prices. Companies are able to charge higher prices because their goods and services are in short supply while workers are able to demand higher wages because companies are producing at high capacity.

Conclusions

The combination of prolonged easy monetary accommodation and a post-covid rebound in activity has led to concerns about inflation spiralling out of control. For the time being these concerns appear unwarranted. Capacity utilisation is low across much of the US economy, unemployment rates remain elevated and a high degree of uncertainty persists.

It is true that there may well be an increase in inflation as measured over the next few months but this will be largely due to easy comparisons with last year that will wash out in subsequent months. Nonetheless it is important to remain vigilant to the possibility of a more permanent increase in trend inflation.

Download the full edition of our Infocus publication here.