July 27, 2018

The Tale of the Changing Measure of Inflation

When we hear about the current low rate of inflation, are we talking about the same inflation we might have discussed 20 or 30 years ago? You may be surprised to discover that, in fact, we are not really talking about the same thing at all. Since 1983, the government has been making major changes to the way it measures inflation, and some say the current methodology now understates true inflation.

Note: due to rounding, total may not = 100%
Source: BLS, data as of December 2017

The Consumer Price Index (CPI), tracked by the Bureau of Labor Statistics (BLS), is the most common measure of inflation, and its impact is felt in a lot of places, from Social Security and pension increases to Federal Reserve policy. Most of the country uses the CPI as a measure of the increase in the cost of living, and the decrease in purchasing power of the dollar. The CPI measures a basket of a variety of consumer goods, taking price changes for each item and calculating a weighted average. Current inflation is estimated at just above 2%. But according to economist John Williams, if inflation were calculated as it was in 1990, it would actually be closer to 6%, and if we went back to 1980 methodology, it would look closer to 10% (see the charts below).

Source: Courtesy of ShadowStats.com

Note: the CPI-U is the Consumer Price Index for all Urban Consumers, the broadest subset of the CPI that is generally noted in monthly headlines.

ShadowStats’ methodology is to apply a constant to officially reported numbers, and does not involve recalculating  the CPI from BLS data. To learn more about ShadowStats and economist John Williams, visit http://www.ShadowStats.com/

How is this possible? Doesn’t the CPI track the same things over time? Not exactly. A few major changes to the way the CPI is calculated have had significant impacts on the measure:

Housing: In 1983, the BLS removed housing prices from the CPI and used “owner’s equivalent rent” instead, in which housing price increases are measured by the amount of rent that could be paid to occupy a rental property equivalent to an owned home. Because rental prices tend to be more stable than home prices, this change reduced the volatility of the CPI.

Hedonics: In 1998, the BLS instituted “hedonic quality modeling,” whereby the CPI takes into account increases in a product’s quality along with its price. If a product’s price goes up, but it includes product improvements, the price change is adjusted to account for the quality improvements. So, even if the cost of a television may have risen from $150 to $1,500 over the last 15 years, the BLS may say manufacturers have implemented $2,000 worth of product improvements: the end result is that the CPI would actually show a price decrease.

Product substitutions: In 1999, the CPI experienced another major shift. CPI statisticians assume that consumers’ goal is to maintain the same standard of living but not necessarily by using the same products. For example, if the price of steak increases, the CPI assumes that consumers will switch to hamburger, and will not register an increase in costs. The assumption is that this better measures the true cost of living, as consumers maintain a “constant level of satisfaction” by substituting cheaper goods for the ones that increase in price.

While arguments could be (and are) made that these changes contribute to a more accurate measure of inflation, the government may not be completely unbiased in implementing them. The government benefits in myriad ways from a low CPI – from giving the appearance of managing inflation to lower Treasury borrowing costs to limiting increases in the costs of Social Security, food stamps, military pay, and pensions. With these incentives to artificially depress the CPI, it’s difficult to say whether the measurement has improved or been manipulated.

Either way, it is helpful for investors to understand that the CPI in the headlines may not be telling the whole story of inflation in the same way it used to. While it’s likely to remain the standard measure for cost of living increases and purchasing power decreases, keep in mind that your purchasing power may be decreasing faster than it indicates.