Behind Recent Disinflation: 2010 Redux?
Inflation rates have been trending lower since the start of 2012. According to the primary inflation indicators used by the Federal Open Market Committee (FOMC), the year-over-year percent changes in the price index for personal consumption expenditures (PCE) and the index excluding food and energy prices (core PCE) were 0.7 percent and 1.05 percent, respectively, in April. Both inflation rates are well below the FOMC's longer-run inflation goal of 2 percent. In addition, the April core PCE inflation reading is currently the lowest on record.
Both inflation rates are also lower than they were in 2010, during the country's last episode of disinflation. Back then, PCE and core PCE inflation reached lows of 1.4 percent and 1.08 percent, respectively. The FOMC's concerns about low inflation during that time were part of the rationale for enacting the Federal Reserve's second large-scale asset purchase program.
While the disinflationary trend is evident across a range of inflation measures, the strength of that trend differs depending on which one you're looking at. Inflation readings based on the Consumer Price Index (CPI), for example, have softened like those based on the PCE. But while the decline in CPI inflation since January 2012 has been comparable to the decline in PCE inflation, the decline in core CPI inflation has been smaller than the decline in core PCE inflation. Meanwhile, median CPI inflation - which provides an alternative measure of inflationary pressure - has been relatively stable; in May, it registered 2.1 percent for the third consecutive month.
The CPI-based measures offer a number of contrasts with the 2010 disinflation. First, core and median CPI inflation are currently running above headline CPI inflation; in 2010, the ordering was reversed. Second, the CPI-based inflation measures have recently been above their PCE equivalents. For most of 2010, both core CPI and median CPI inflation were less than 1 percent, while PCE and core PCE inflation were in the 1-2 percent range. The differing patterns among these various measures suggest that comparisons between the 2010 disinflation and today's disinflation might benefit from digging into the details.
Analyzing goods and services prices separately turns out to provide a fair amount of clarity in understanding both recent inflation trends and how they relate to 2010. These two types of prices are impacted by different factors, and as such they have behaved very differently over time.
Declines in goods inflation played a key role in the downward movements in overall inflation in 2010 and today. Core goods inflation has recently fallen by about 2 percentage points, according to both the CPI and the PCE price index, and both are now showing modest deflation. A roughly similar decline in core goods inflation occurred in 2010 as well, helping to pull down inflation at that point, too.
While the recent decline in core goods inflation is similar to what occurred in 2010, a look at the components reveals notable discrepancies. In 2010, a weak dollar supported the prices of imported goods, and the cyclical recovery in motor vehicle sales helped to push up the prices of those vehicles. At the same time, weakness in the housing market was associated with relatively strong deflation in the prices of housing-related goods, like furnishings and household equipment and recreational goods and vehicles. More recently, the recovering housing market has pulled housing-related inflation up. At the same time, slowing growth abroad and strength in the dollar have weighed on import prices, and moderation in the motor vehicle recovery has pulled motor vehicle price inflation down toward the general trend in goods prices.
Given the similar trends in both PCE and CPI core goods inflation, both in today's disinflationary period and in 2010, discrepancies between PCE- and CPI-based inflation measures are primarily explained by their services components. The CPI measure of core services inflation was below 1 percent for much of 2010, but it recently has been roughly stable near 2.5 percent. By contrast, the PCE measure of core services inflation remained well above its CPI counterpart in 2010. More recently, it has shown a similar downward drift as occurred in 2010. This drift has amplified the goods disinflation and explains why PCE measures of inflation are running below CPI measures.
A key factor behind the differences in services inflation is how the indexes weight shelter costs. Shelter comprises a larger share of the CPI than the PCE price index. With the housing market and the labor market both weak in 2010, inflation in the shelter component of the PCE price index was also subdued. More recently, as the labor market recovery has slowly progressed and the housing market has improved, rents have been rising and shelter inflation has increased, thereby helping to anchor services inflation in the CPI.
A number of other methodological differences between the indexes are also contributing to more disinflation in core PCE services than in core CPI services. The PCE price index includes broader measures of financial services and insurance, transportation services, and medical care than the CPI. For various reasons, all three of these components have been experiencing low inflation recently, whereas in 2010 they helped to lift core PCE services inflation.
Over a longer time horizon, discrepancies between the behavior of goods and services prices and their impact on aggregate inflation measures are not abnormal. Core services inflation remains historically low, most likely reflecting the weak-but-improving labor market. This would be consistent with subdued demand for consumer services, limited pricing power by businesses, and limited cost pressures coming from labor, as measured by the Employment Cost Index (ECI), for example.
Interpreting recent movements in core goods prices is more difficult. Ongoing deflation among core goods prices was the norm prior to the financial crisis. One possibility is that the surges in goods inflation in 2009 and 2011 were due to transitory factors that have run their course. This possibility suggests that the disinflations in 2010 and 2013 - while differing in the details - partly reflect goods inflation returning to its longer-term trend. Alternatively, core goods inflation may have entered a new phase in which it is volatile but positive on average, thereby putting some upward pressure on inflation.
With services comprising about two-thirds of the market basket, an upward move in core services inflation in line with an improving economy and rising labor costs will be a key feature in bringing core inflation back toward 2 percent. But this return to 2 percent inflation will take longer if core goods price trends stabilize in deflationary territory.