We have two different inflation measures in the US.
The Federal Reserve prefers to look at the Commerce Department’s Personal Consumption Expenditures (PCE) Price Index. They believe it is more accurate than the Labor Department’s Consumer Price Index (CPI).
Are they right?
Despite the differences between the two measures, I think the Fed is exactly backwards here. Neither measure is foolproof, but the flexibility and adjustments of PCE also take the index further from reflecting the average citizen’s economic condition.
This bias shows up in Fed policy, and not in a good way.
CPI is not brilliant, either.
Unlike some, I don’t believe it is intentionally manipulated. I think the wonks (and I say that in a complimentary way, as a fellow wonk) who compile price data do a nearly impossible job as well as anyone can.
Browse through the methodological explanations on the CPI home page and you’ll quickly see how much effort goes into that work. They have a whole “data available” shopping list:
• Price indexes are available for the US, the four census regions, nine census divisions, two size of city classes, eight cross-classifications of regions and size-classes, and for 23 local areas. Indexes are available for major groups of consumer expenditures (food and beverages, housing, apparel, transportation, medical care, recreation, education and communications, and other goods and services), for items within each group, and for special categories, such as services.
• Monthly indexes are available for the US, the four census regions, and some local areas. [You can see those indexes here.]
• More detailed item indexes are available for the US than for regions and local areas.
• Indexes are available for two population groups: a CPI for All Urban Consumers (CPI-U) which covers approximately 94% of the total population and a CPI for Urban Wage Earners and Clerical Workers (CPI-W) which covers 28% of the population.
• Some series, such as the US City Average All Items index, began as early as 1913.
All that data gets worked into “baskets” that try to match the spending habits of typical consumers. That’s where the effort starts going wrong. The problem is quite simple and beyond anyone’s control: None of us are average.
We all spend our money differently, for an endless variety of reasons that change all the time. When you say inflation is higher than CPI shows while your neighbor says inflation is no big deal, you can both be right.
Worse yet, even someone with spending patterns identical to yours can experience an entirely different inflation rate simply because they live in a different city or state. Or they choose to send their kids to a more expensive school. Or they spend a larger amount on health care and less on goods but more on services. It can get quite nuanced.
Reducing this complexity to one number and then using that number to guide monetary policy is asking for trouble. And trouble is what we get.
CPI isn’t entirely useless.
It can show us broad price trends over long periods.
Much to our surprise, the highest inflation is in necessity goods and services. This puts more burden on lower-income Americans.
The disinflation that so vexes the Fed impacts more optional purchases. Here’s how my friend Barry Ritholtz describes the pattern:
It is notable that the two big outliers to the upside are health care (hospital, medical care, prescription drugs) and college (tuition, textbooks, etc.).
Clothes, cars, TVs, cell phones, software—technology in general—showed disinflation or outright deflation in prices. (Housing and food & beverage have been right in the middle of inflation levels.)
Wages have barely ticked over the median inflation measure, but that did not stop some people from blaming the correction on rising wages.
Reading the pundits, I cannot tell which fate awaits us: the robot-driven apocalypse where we are all out of work, or the inevitable spike in wages that sends rates much higher and kills the market. Perhaps both—higher wages send employers into the waiting arms of our automated future.
You can quibble with this data. Have TV prices really fallen 99%? No, unless you hedonically adjust, because today we can buy TVs of a quality that didn’t exist in 1997. If you use hedonic prices that adjust for quality and technological sophistication, then you can argue that the price of TVs is down 99%.
But we all know that we are paying less for TVs. Same for other technology goods. But you simply cannot argue that we are currently paying the same price for new vehicles that we did 20 years ago, even though the cars we buy today are technologically vastly superior to what we could buy back then. These hedonic price adjustments are guesswork.
Still, the broader point seems right. Inflation is a real problem for some people and no big deal for others. But the Fed uses inflation measures to impose a single monetary policy on everyone.
Get one of the world’s most widely read investment newsletters… free
Sharp macroeconomic analysis, big market calls, and shrewd predictions are all in a week’s work for visionary thinker and acclaimed financial expert John Mauldin. Since 2001, investors have turned to his Thoughts from the Frontline to be informed about what’s really going on in the economy. Join hundreds of thousands of readers, and get it free in your inbox every week.