With real house price inflation based on market data, the headline CPI would have jumped 3.7%. Lifting the veil on the trickery to keep the CPI low.

By Wolf Richter for WOLF STREET.

For most Americans, housing costs are the largest item in their budget, ranging from 30% to 60% of their total monthly expenses. In its consumer price index (CPI) for February, released yesterday, the Bureau of Labor Statistics reported that homeownership costs (which the BLS calls “the equivalent owner’s residence rent”) rose only 2.0% from a year ago and that rents ( “rent of the principal residence”) increased by 2.0%. These are the largest items among the 211 items in the CPI basket and together they make up about one-third of the overall CPI. They play a huge role in the CPI. So…

US rent inflation averaging 2.0% year-over-year could be about on target, from what i can see in other rental data. But homeowners inflation of only 2.0%, given soaring house prices? Ridiculous. In its latest release, the national Case-Shiller home price index jumped 10.4%.

This discrepancy between house price increases and the CPI for homeowners – which has for years contributed to underestimating the overall CPI – is illustrated in the chart of the Case-Shiller National House Price Index ( red line) and the CPI for “equivalent homeowner’s rent” (green line). I set the owners’ CPI at 100 for January 2000 to match the Case-Shiller index, which defaults to 100 for January 2000. This allows you to see the progress of both indexes on the same axis.

The CPI thus corrected increases by 3.7%.

The “equivalent rent for the owner’s residence” represents 24.2% of the CPI. If it had increased by 10.4%, in line with the Case-Shiller index, instead of 2.0%, the overall CPI would have increased by 2.03 percentage points. Continued. So add the 2.03 percentage points to the overall CPI increase of 1.7%. And the overall CPI thus corrected would have jumped by 3.7%!

During the housing crisis, after five years of decline, the Case-Shiller index briefly joined the CPI for homeowners before rising again – and there’s a reason for that which we’ll get to in a moment.

The S&P Case-Shiller Home Price Index is a good measure of home price inflation because it is based on the “sales pairs” method, comparing the price of a home when it was sold during the current month, at the price of same house in previous transactions years ago. It also accounts for improvements and removes outliers. In other words, it measures how many dollars it takes to buy the same house over time – and thus it measures house price inflation.

This discrepancy – in effect, a form of deliberate deception – between actual house price increases and the CPI for homeownership has been lamented before and is no secret. But it’s not widely discussed in the media, so everyone knows how much the homeownership component in the CPI underestimates the real homeownership inflation at which Americans face.

To its credit, the BLS includes ownership costs in the CPI basket. In contrast, the EU does not include home ownership costs at all in its basket underlying its Harmonized Index of Consumer Prices (HICP). It only includes the rent. And as a result, housing costs are woefully underrepresented in EU inflation data.

The rationalization put forward by the BLS for this discrepancy is that it does not actually consider house prices relevant to inflation. He considers houses an “investment”, and investments do not enter the CPI. What he is trying to put a price on is the cost of “housing”, which is a service. Makes sense with the rent. A tenant pays for a service. But the BLS extrapolates that rent as a service to homeownership. And this is where the logic croaks.

BLS housing inflation data is based on the “Consumer Expenditure Survey” sent to consumers, rather than market data for prices and rents. It’s up to the landlord to decide what his primary residence would rent for the one he tried to rent.

That’s the survey question: “If someone were to rent your house today, how much do you think they would rent monthly, unfurnished and without utilities?” Owners therefore need to come up with a number.

For the CPI, according to the BLS, the housing cost of the owner-occupied house “is the implicit rent that owner-occupiers would have to pay if they rented their house.”

This explains the black line in the graph above which is completely out of step with the reality of house prices: it tracks rent inflation, not house price inflation, and it basically represents what landlords think that the rents would be. Thus, the CPI for rents, which represents 7.8% of the total CPI, and the CPI for owners, which represents 24.2% of the CPI, measure roughly the same thing: rents. And this thus cleanly and deliberately excludes runaway house price inflation from the index. And it also turns CPI into the sad joke that it is.

Durable goods inflation +3.3%. Food inflation +3.4%. Services inflation is rising, but remains contained by airfares, accommodation, event tickets, etc., until people start traveling and attending events again. Read… The purchasing power of the dollar is falling to an all-time high. The Fed gets its wishes

Do you like to read WOLF STREET and want to support it? You use ad blockers – I completely understand why – but you want to support the site? You can donate. I greatly appreciate it. Click on the mug of beer and iced tea to find out how:

Would you like to be notified by e-mail when WOLF STREET publishes a new article? Register here.