Recent Trends in Various CPI-Based Inflation Measures
In the Bureau of Labor Statistics’ most recent release of the Consumer Price Index (CPI), the index declined in April at an annual rate of 4.3 percent. This follows a monthly decline of 2.2 percent in March. On a year-over-year basis, CPI inflation slowed from a recent peak of 2.0 percent in February to 1.1 percent in April. Taken at face value, this slowing of inflation suggests some further deceleration of inflationary pressure in the U.S. economy.
A careful assessment of underlying price trends suggested by other, “core” measures of CPI inflation provides further evidence of this deceleration. CPI inflation can be significantly affected by large, temporary movements in volatile individual components of the price basket, and core measures—such as the CPI excluding food and energy, the median CPI, and the trimmed-mean CPI—are less affected by these temporary, idiosyncratic price changes.
Inflation in the CPI excluding food and energy was 0.6 percent at an annual rate in April, down from 1.3 percent in March. Inflation in the median and trimmed mean CPIs came in at 1.8 and 1.0 percent, respectively, compared to 1.1 and 0.7 percent in March. On a year-over-year basis, inflation in the CPI-excluding-food-and-energy measure slowed from 1.9 percent in March to 1.7 percent in April. Year-over-year inflation in the trimmed-mean CPI fell from 1.7 percent in March to 1.6 percent. However, year-over-year inflation in the median CPI has continued to remain close to 2 percent, coming in at 2.1 percent in April.
While the recent discrepancies between these various measures are small in historical terms, over the past six months inflation measured by the median CPI has been consistently above inflation measured by the trimmed-mean or core CPI. The reason for this is that, among the components of the CPI, there has been a slight shifting of prices below the median. Looking at the boundary for the lower 25th percentile of price changes, it has shifted downward slightly over this time period, while the boundary for the upper 25th percentile has remained relatively constant. This basically means that there is a wider distribution of price changes below the median than there is above. While this may not impact the median CPI, it would have the impact of pulling down the other measures, which has been the primary cause of the recent differences.
One reason that median CPI inflation has been so stable is that a major component of the CPI, owner’s equivalent rent of primary residency (OER), has been increasing at a steady rate. Year-over-year inflation in the OER component has been 2.1 percent each month since last September. The weight (relative importance) of this component in the CPI is currently around 23 percent, much larger than any other single component, so OER has a large impact on all measures of CPI inflation. For example, excluding OER and other shelter components from the calculation of the core CPI in April gives a year-over-year change of 1.4 percent, compared with 1.7 percent when these shelter components are included. OER is particularly important for the median CPI because OER is often at or near the median of the distribution of CPI components.
To shed more light on the recent behavior of core inflation, it is helpful to distinguish inflation in core services (services excluding energy services) and inflation in core goods (goods excluding food and energy goods). Normally, inflation in services exceeds inflation in goods. While this pattern broke down briefly following the last recession, it has returned in the last couple of years. Year-over-year inflation in core goods prices has averaged 1.0 percent since the beginning of 2012, while inflation in core services prices has averaged 2.4 percent. Recently, goods inflation has trended down sharply, actually reaching negative territory in April (around −0.1 percent), while services inflation has remained steady. This makes clear that the recent slowing of some measures of core inflation has been driven by deceleration in goods prices, not services.
One variable that is positively correlated with the gap between services and goods inflation is the exchange rate. As the exchange rate appreciates, imports tend to become less expensive. Since it is primarily goods that are imported rather than services, the downward pressure from exchange rate appreciation falls primarily on goods prices, boosting services inflation relative to goods inflation. Recently, there has been some appreciation of the exchange rate, which could explain some of the rewidening of the gap in goods and services inflation. Consistent with these developments, inflation in imported consumer goods (excluding autos) has been moving downward recently, similarly to inflation in core goods prices.
Putting all of this together, it is clear that the CPI report for April revealed some further, modest slowing of inflation, reflecting stable inflation in services prices and additional declines in goods inflation.