Research Affiliates: Where's The Beef? ‘Lies, Damned Lies, And Statistics’

Key Points

1. American households, pinched by rising prices at a rate higher than headline inflation, have generally not benefited from the unrelenting stimulus of quantitative easing and zero interest rates, and instead have experienced a decade of zero growth in income and spending power.
2. The high valuations and low interest rates born of accommodative monetary policy lower forward-looking returns for both the wealthy and the middle class, but the middle class, who must invest today to prepare for retirement tomorrow, suffers relatively more.
3. When the Fed eventually steps away from overt market interventions, capital market valuations should revert to more normal (i.e., lower) levels, which would bring with them more sensible forward-looking returns.

The price of beef has been soaring over the past five years—up 80% cumulatively at the end of December 2015—but you'd never know it by looking at the official U.S. Consumer Price Index (CPI), which is up 7%, or 1.4% a year, over the same five-year span, and therein lies our beef. The developed nations of the world, are supposedly living in a low-inflation environment, at risk of tipping over into the abyss of deflation. That's what the folks at the U.S. Bureau of Labor Statistics (BLS) tell us. And they should know, right?

Well, maybe not. Surveys suggest that the average American's daily experience may be quite different. One-year consumer inflation expectations have been consistently higher than trailing and realized inflation over the last 20 years, and higher than more recent market-based inflation expectations, measured by one-year swap rates. Figure 1 shows how this divergence has grown larger since the financial crisis, suggesting the average household might have been feeling even greater pain during the recovery process than has been believed.

Since 1995, households have expected inflation to be, on average, 3.0%, whereas realized inflation has been around 2.2%, leaving an inflation "gap" of almost 0.8%. What explains this gap? The following is our hypothesis. The four "biggies" for the average American are rent, food, energy, and medical care, in approximately that order. These "four horsemen" have been galloping along at a faster rate than headline CPI. According to the BLS definition, they compose about 60% of the aggregate population's consumption basket, but for struggling middle-class Americans, it's closer to 80%. For the working poor, spending on these four categories can stretch to as much as 90% of total spending. Families have definitely been feeling the inflation gap, that difference between headline CPI and inflation in the prices of goods they most frequently consume.

Since 1995, the average year-over-year inflation rate for energy has been 3.9%, for food, 2.6%; for shelter, 2.7%; and for medical care, 3.6%. If we strip out all other items and recalculate the index based exclusively on these four components, we find the average rate has been about 2.9%, right in line with households' expectations. Let us be provocative. If inflation—as experienced by the average American—is higher than official BLS "inflation," then what exactly is the BLS statistic measuring?

It looks like the BLS thermometer is broken! Whatever temperature they show us, the actual temperature is higher, as experienced by the average American family. For instance, the reported CPI inflation over the past 10 years ending December 2015 was about 1.9% a year. If we focus on the "big four" over the last decade, the inflation that Americans experienced was about 0.5% more. Let's call this 0.5% difference a "measurement bias." Paradoxically, the inflation measures for these four categories are produced by the same people who assemble the CPI. Other sources peg the gap as being considerably larger.

These considerations have a direct bearing on our prosperity. How much real growth, for example, has occurred in the past decade? Officially, GDP has grown 1.4% a year, over and above inflation. Over the same period, the U.S. population has grown by 0.9% a year. Thus, real per capita GDP has risen by a scant 0.5% a year. Subtract the 0.5% measurement bias—probably a conservative estimate—and the average American has experienced zero growth in personal spending power over the past decade. With wealth and income concentration, if the average is flat, then median per capita spending power must be lower. Comparing 2015 with 2005, this feels about right. The official statistics do not.