Updated: Date

Published: FEBRUARY 2022

For real insights into inflation, check your bank account monthly for one year

Confused by mumble-jumble inflation jargon?

Analysis by STAN CROCK
Writing from Washington, D.C.

President Harry S. Truman is said to have yearned for a one-armed economist so he would have someone who would say something without adding, “On the other hand.” We sure could use the kind of certainty that Truman sought when it comes to inflation.

In January, the Bureau of Labor Statistics (BLS), an agency of the U.S. Department of Labor, announced that the Consumer Price Index (CPI) for urban consumers rose 7% in 2021, the highest inflation rate in four decades. Then in February, the BLS announced a worse number – 7.5% – for the 12 months ending in January. All of this spawned a cottage industry of economists and other policy wonks debating the causes: the huge recovery packages that may have provided too much fiscal stimulus, COVID-19, supply-chain disruptions, industrial concentration, and wage growth – with surprisingly little talk about the monstrous growth in the U.S. money supply (see accompanying M2 money supply chart).

Of course, some say it’s a global phenomenon (true), so U.S. policies aren’t the cause. But I have not seen an analysis of whether other countries have similar basic Keynesian fiscal policies plus monetary stimulus, so maybe the policies are being dismissed too easily. The debate over causes set off a debate over solutions, particularly about how far, fast, and frequently the Federal Reserve should raise interest rates to shut the money spigot and to stifle inflation.

What gets less attention, though, is the inflation number itself. I find it might be a little like my wife saying I never passed the bar when I never took it: absolutely true and utterly misleading. Alas, we don’t know in which direction we’re being misled. Some economists say the inflation figure is (forgive the pun) inflated, some argue it’s understated, and some insist it’s actually irrelevant to any individual. Let me explain.

Overstated

New York Times columnist Peter Coy recently cited work by Laurence Ball of Johns Hopkins University and Daniel Leigh, Prachi Mishra, and Antonio Spilimbergo of the International Monetary Fund that suggests inflation might not be as bad as it seems.¹ They noted that two Federal Reserve regional banks make adjustments that come up with lower inflation rates than do the BLS and the Bureau of Economic Analysis (BEA) of the U.S. Department of Commerce.

The Federal Reserve Bank of Cleveland tracks the price of stuff just in the middle of the distribution of price changes. Its “median” CPI measure rose 3.5% in the 12 months through November 2021, vs. 4.9% for the BLS’s CPI (excluding food and energy).

Similarly, the Federal Reserve Bank of Dallas makes adjustments to the BEA’s Personal Consumption Expenditures (PCE) price index by eliminating the items that rose and fell the most before calculating the average. The result: a 2.8% rise in the PCE in the 12 months through November 2021, vs. the BEA’s 4.7% (excluding food and energy).

These adjustments may seem arbitrary. After all, people buy products on the ends of the inflation spectrum, as well as a lot of food and gas, all of which are excluded by the adjustments.

But New York Times columnist Paul Krugman makes a point about possible overstatement that avoids those reservations. He notes that the 7% figure for 2021 may be both accurate and useful because it smooths over seasonal variations and aberrations. But the annual number may not reflect the most recent trend. A dip in the inflation rate at the end of the year can still leave the overall inflation rate higher than a year earlier while hiding the more positive trend.

Consider the fourth quarter of 2021.² The CPI rise was 0.9% in October, 0.8% in November, and just 0.5% in December. The December 0.4% rise in grocery food prices was half the 0.8% jump in November and well below the 1% in October. The price of gas at the pump dropped by 0.5% in December after rising 6.1% in each of October and November. New car price increases declined pretty steadily from 2% in June to 1% in December. Increases for medical care commodities and services also were lower in December than in October.³

In January 2022, the monthly rate ticked up slightly to 0.6%, but that’s still lower than in October and November.⁴ The BLS report contained a lot of bad news, but some glimmers of hope, too. The rise in the core inflation index (excluding food and energy) in January was flat from December at 0.6%. Gas (4% of the CPI)⁵ was down 0.8%. Shelter, one-third of the index,⁶ was down slightly from 0.4% to 0.3%. New cars (about 4% of the CPI in 2019)⁷ showed no increase at all.

The Washington Post’s astute columnist Catherine Rampell provided some additional context. Gas prices in mid-2008, adjusted for inflation, would be more than $5 a gallon vs. the current $3.48. And the 40% year-over-year rise in gas prices is a bit of an aberration because the baseline is the pandemic-depressed January 2021 number. ⁸

With this kind of trend line, the teeth-gnashing and hand-wringing that the big annual figures generated might not be so justified. Economists are predicting a decline in the inflation rate in 2022. I don’t put much stock in such predictions for the reasons I noted in my piece in September 2021 on BestStory.ca - Inaccurate economic predictions rile politics and markets. The prognosticators screwed up big time yet again with the employment numbers for January 2022, predicting everything from a loss of 400,000 jobs to a measly 150,000 gain, while the actual number was a 467,000 jump.⁹ But I do put stock in actual numbers, and some of the recent data tell a story that is different from – and better than – what is generally recognized.

Yet some people have argued that inflation is vastly understated. If you gave it any thought, you probably would think that the CPI reflects the government looking at the same market basket of items you buy and using the same methodology every month for decades. The result would tell you how much more an item costs today than it did in, say, 1970.

In one sense the government does use the same sampling techniques. Here is the BLS explanation of how it collects the data:

Prices are collected each month in 75 urban areas across the country from about 6,000 housing units and approximately 22,000 retail establishments (department stores, supermarkets, hospitals, filling stations, and other types of stores and service establishments). All taxes directly associated with the purchase and use of items are included in the index. Prices of fuels and a few other items are obtained every month in all 75 locations. Prices of most other commodities and services are collected every month in the three largest geographic areas and every other month in other areas. Prices of most goods and services are obtained by personal visit, telephone call, or web collection by the Bureau’s trained representatives.”¹⁰

The data help determine the importance, or weight, of the item categories (food, shelter, etc.) in the CPI index.

But the BLS changes what’s in the basket, and some modifications might understate inflation. For example, the BLS eliminated car finance charges. It sometimes accounts for consumer substitutions of lower-priced products (maybe chicken for steak) within CPI categories. It also accounts for changes in quality, for example, the difference between a $300 TV set in 1970 and one today—a tough calculation.

The CPI doesn’t reflect changes in buying habits fast enough to keep up with actual purchases, however. For example, the BLS announced that starting this year, it will calculate weights in the CPI based on consumer expenditure data from 2019–2020.¹¹ It considered making changes, but decided against doing so. How much the CPI market basket weighting reflects what people actually buy today is a big question hardly anyone asks. And are any adjustments for the anomalous pandemic correct?

Moreover, the BLS makes significant changes in methodology. In 1983, the BLS changed the housing component of the CPI quite drastically.¹² It replaced housing costs – housing prices, mortgage rates, insurance, property taxes, and maintenance costs – with what a homeowner would charge someone to rent the house. The rationale was that a house purchase is an investment, not consumption, while rent is consumption. And that’s what the CPI is supposed to measure.

It sounds like a conceptual change, but it was political as well. It cut cost-of-living adjustments for Social Security and other government benefits, slowing depletion of the Social Security Trust Fund. The distinction between investment and consumption might be a nice theoretical problem for economists, but for consumers, does it matter if that big monthly check goes to a mortgage or rent? I doubt it. Does getting a smaller Social Security check matter to pensioners? Of course.

An early study showed that the change reduced monthly costs for shelter in the index by 1.1% in 1975, 6.5% in 1982, and 4.7% in 1984.¹³ The imputed housing rent accounts for about two-thirds of the shelter component or about 22% of the entire index,¹⁴ (two-thirds of one-third of the CPI), so this change has a significant impact.

This modification probably continues to keep the CPI lower than it would have been before the change. Housing prices ballooned 19.2% last year while rents soared 14.3%, according to Zillow,¹⁵  yet the CPI shelter component grew only 4.1%.¹⁶ I’m not saying that the real number should be in double digits, but it’s a huge discrepancy.

Just to make it all a bit more complicated, I must note the view of Austan Goolsbee, chairman of President Obama’s Council of Economic Advisers and an economics professor at the University of Chicago’s Booth School of Business. In a piece for The New York Times, he made the point that inflation is different for different people, depending where you live, the gender of your children, where you buy, and a host of other factors.¹⁷ He noted that boys clothes were up 8.4% in 2021, while girls apparel was down 0.4%.

An index he helped create for online purchase inflation was three percentage points lower for 2021 than the CPI, and online prices dropped 0.2% in November as the CPI rose 0.8%. The BLS says that as of 2017, 8% of the numbers in the CPI sample were from online sales. Online sales roughly doubled during the pandemic, but it’s not clear how much of that the CPI reflects.¹⁸

Goolsbee also said studies have shown that lower-income consumers face higher inflation rates than top earners. In addition, you may have noticed when I first mentioned the CPI, the full name shows it’s the CPI for urban consumers. What’s inflation like for rural residents? No idea.

The government agrees with Goolsbee. According to the BLS, “A national average reflects millions of individual price experiences; it seldom mirrors a particular consumer's experience.”¹⁹ The BLS notes that if you spend more than average on medical expenses that rise faster than other market basket items, your personal rate of inflation may exceed the CPI increase. And if you use solar energy to heat your home while fuel prices take off, you could see less inflation than everyone else.

How can the average person make sense of all this? At the end of every month, see how much is in your bank account. Jot it down. Do it for a year, so that you will even out extraordinary expenses from car repairs to vacations. See how much you have left on average each month. Compare this with your checkbook from last year. Then you’ll know how inflation affects you. If you don’t like swordfish prices, switch to cod. Switch from an expensive cable bill to some streaming services. You don’t need to worry if the government makes accurate product substitution calculations. DIY. In the end, ignore the inflation news and the government’s market basket. Just see how you’re doing. That, after all, is the only thing that matters.