CPI Up 0.1 Percent: How Much is the CPI Understated?

The BLS says the CPI is up 0.1% for the month and 2.1% from a year ago. What's the real story?

The Consumer Price Index for All Urban Consumers (CPI-U) increased 0.1 percent in December on a seasonally adjusted basis, the U.S. Bureau of Labor Statistics reported today. Over the last 12 months, the all items index rose 2.1 percent before seasonal adjustment.

An increase of 0.4 percent in the shelter index accounted for almost 80 percent of the 1-month all items increase. The food index rose in December, with the indexes for food at home and food away from home both increasing. The energy index, which rose sharply in November, declined in December as the gasoline index decreased.

Percent Change CPI Items

Despite energy commodities rising 10.8% the CPI is only up 2.1% year-over-year, assuming of course you believe the numbers. I find most of them believable.

What's Believable?

Some readers may dispute this, but the price of food at home has been stable.

I have done nearly all our grocery shopping for my entire life and I worked on grocery stores from 1968-1972. I know what food prices were on sale then and I know what they are now. My judgment is based mainly on meat and cheese prices which make up the bulk of our food bill.

I buy few snacks and few prepared or frozen foods. I do not buy milk or cereal. We tend to have meat or fish with a salad. Those buying packaged goods or organic vegetables may see things differently.

Energy commodities are easily measured so those prices rate to be accurate.

New and used cars down and apparel down are all believable.

What's Not Believable?

​Given reported health care premium increases, I find medical care expenses mostly a joke.

Shelter at plus 3.2% is a proven joke. The Case-Shiller National Home Price Index is up 6.2% in the last year.

At one time, the CPI incorporated actual home prices but now it doesn't. Homes are considered capital investment.

The Fed made a huge mistake in the housing bubble years by ignoring home price inflation and they are making the same mistake again now.

Calculating OER

OER, Owners' Equivalent Rent, accounts for 24.583 percentage points of the CPI. The BLS says OER rose 3.2% year-over-year.

The BLS calculates OER by asking this exact question: “If someone were to rent your home today, how much do you think it would rent for monthly, unfurnished and without utilities?

If you think that's ridiculous, you're not the only one.

Understated CPI

If home prices were in the CPI, then we could add another 0.74 percentage points ((6.2-3.2) * .24583) to the year-over-year CPI.

That would make the year-over-year CPI 2.84% not 2.10%. And that's without a realistic assessment of medical expenses.

John Williams at ShadowStats, an economist perpetually predicting hyperinflation, goes off the deep end in the other direction. Williams believes the CPI is at 6 percent.

Flawed Methodology

Of course, there is no such thing as a representative basket of goods and services in the first place.

Moreover, it makes little sense to average all the components the way the BLS does.

To top it off, the CPI fails to factor in clear bubbles in financial assets. Those financial bubbles are a direct representation of unreported inflation.

Mike "Mish" Shedlock

No. 1-25

The GDP deflator does not use a market basket, but rather the sum total of all goods and services, so by it's nature it automatically adjusts to what we buy. I am glad you agree that substitution based on technological obsolescence is fine. In my opinion, all other substitutions are as well. If you looked at a market basket from 50 years ago, you'd find much more fabric, and much less factory made garments. You'd find produce and meat, and canned goods, but not much pre-prepared food. You'd find very little in the basket for restaurants as people rarely ate out. You wouldn't find a Cable bill at all. If you still used the 1960 market basket, it would be a very inaccurate measure of inflation today as it wouldn't reflect the lifestyle that any of us have. That is the whole point. People point to the chicken-beef argument as if substitutions are something forced on consumers. Rather it's a case where consumers make lifestyle changes, and what they purchase changes with time, and the index, if it is to remain relevant needs to change along with consumers. Substitutions have always been made. They used to only be made annually. Now they are made on an ongoing basis. The faster the substitutions are made, the closer the CPI becomes to the GDP Deflator.


“Do you think our standard of living dropped when we substituted smart phones for dumb phones?”
This is an example of new technology supplanting old technology. Again I will state that substitution due to technological obsolescence makes sense. I am not debating that point. Since you mentioned smart phones specifically, I will point out that some people think that particular device had deleterious effect on daily quality of life.

“As one who prefers chicken, I disagree”
Again I will point out that personal preferences of consumers should not have anything to do with measuring monetary effects on prices.

“How would you compare the standard of living of Seniors today to Seniors in 1995?...1965?”
Since the cost of medical care has become extremely distorted during that time and since medical care is extremely important to seniors, that is a difficult question. In my experience much of US medical care today is a poor value compared to what it was in the 80’s. Due to Medicare seniors may not be worse off, but many younger people without Medicare are worse off. The trend in costs and benefits of Medicare and Social Security are also not sustainable.

“What I see in the real world matches exactly with my expectations based on theory.”
Since I am not trying to match what I see with a theory, I suspect my observations are less bias than if I were trying to match observation and theory. Regarding whether or not standard of living has been cut in half in the last 20 years I would say not, but I would also say that gains in productivity deserve the credit for blunting the impact of corrosive monetary effects and monetary inflation has had the effect of transferring those gains (and more) to those who issue the currency. Due to productivity gains, I think a basic standard of living today would be easier than it was 20 years ago were it not for that effect.

“...today we eat much more frozen food....Should the index be adjusted to include frozen food?”
Since you and I have been back and forth on CPI and GDP deflator, I am really not sure which “index” you are referring to. I will simply restate that the point of CPI (and probably the GDP price deflator too) should be to measure monetary effects on prices and not consumer preference, regardless of the reason for the consumer preference. Substitution due to technological obsolescence stands on its own merits.


As another example of substitution, sticking in the food realm, today we eat much more frozen food than 40 years ago, and much less fresh and canned goods. Should the index be adjusted to include frozen foods? I believe it should reflect whatever we are currently eating - whether that is beef, chicken, frozen foods, or whatever.


Regarding the CPI, I'm saying that it is important that the market basket is constantly adjust to reflect what each person is buying at the present time. Each of us has a utility function, and we spend in a way that maximizes out personal utility. We might shift from one item to another because of price, or because we like it better. Thus, we shifted from dumb phones to smart phones not because they were cheaper, but because we liked them better. The CPI needs to shift as well. Those that oppose substitution always pick out the beef-chicken argument, but that is only one of many reasons why substitutions take place. You believe our standard of living dropped as we chose to substitute more chicken for beef. As one who prefers chicken, I disagree, but that's my personal preference. Do you think our standard of living dropped when we substituted smart phones for dumb ones? You can't have it both ways. You either have substitutions, or you don't. Even the pre-1990s CPI had substitutions. They just were done much slower, and well after the fact. I prefer the GDP deflator because it's not tied to a specific market basket, but just tracks the economy as a whole.
To me the GDP deflator is the actual answer, whereas the CPI is an estimator, and the limit of the CPI index, as the speed of substitutions approaches the speed at which we, as consumers substitute, is the GDP deflator.
Getting away from theory to the real world is important, too. How would you compare the standard of living of Seniors today to Seniors in 1995? How about to Seniors in 1965? We haven't legislated any changes to increase Social Security that I am aware of during that time, and I'm not convinced that people today are thriftier during their working years, and save more for retirement (if anything, it's the opposite). Therefore, the major reason for any change in their standard of living would be the accuracy of the CPI. My personal observation is that the standard of living rose considerably during the 70's and 80's, but has been relatively flat ever since. That matches what I would have expected to see. Those that believe the CPI is significantly too low should expect to see a significant drop in their standard of living. John Williams, for example, would expect to find the standard of living has been cut in half in the last 20 years. Is that what you see? Is AARP leading a march on Washington to protest against constantly falling standards of living for the elderly?
What I see in the real world matches exactly with my expectations based on theory. If it didn't, I'd consider changing my theoretical view. The proof is in the pudding, as they say.


I just realized that you were not focusing on CPI so much as the GDP deflator, making the point that the GDP deflator is normalized with consumption and investment patterns more often than CPI and that it should be the preferred method for that reason. Meanwhile, I was writing about CPI which is not the same thing and is not adjusted as frequently. Since GDP also includes government spending, investment and exports, I would question if the GDP price deflator is a reasonable way to measure monetary inflation effects on the consumer. I have to stop there since that’s not really my area.