NYT defends CPI

Do you think the consumer price index understates the real cost of living?  Well then, the New York Times says you buy into "conspiracy theories."

I'm sure it just breaks your heart that the Gray Lady thinks so ill of you.  But it is what it is: The newspaper that tried to convince you Saddam Hussein was on the verge of unleashing mushroom clouds is now trying to make the case that CPI actually overstates the cost of living.

Granted, David Leonhardt can't really begin his column with such a flat assertion.  So he adopts a sympathetic tone:

Next week, the Bureau of Labor Statistics
will release its monthly report on inflation, and it sure is going to
sound strange. Wall Street is expecting the bureau to announce that the
Consumer Price Index rose just three-tenths of a percentage point in
April. Over the last year, the index has risen only about 4 percent.

I’m guessing that doesn’t square with your sense of reality.

He invites the reader in so warmly.  And he goes on to point out you're not alone in your perceptions.

"Last month, when I did an online Q. and A. with Times readers, I got
three separate, thoughtful questions about — of all things — how the
inflation rate is calculated. The current cover story in Harper’s,
called “Numbers Racket: Why the Economy Is Worse Than We Know,” deals
with the same subject. Written by Kevin Phillips, the Nixon aide turned
left-leaning commentator, it concludes that the real inflation rate “is
as high as 7 or even 10 percent.”

But it's all illusion, Leonhardt proceeds to explain.  See, many prices rose moderately or even fell during the 1980s and 90s.  Food was getting cheaper relative to everything else.  We tend not to notice that.  "Part of this comes from the notion of loss aversion," Leonhardt avers: "Human beings dislike a loss more than they like a gain of equivalent size."  And when things actually fall in price, we tend not to notice that either — especially if the things falling in price are things we don't buy very often.  We buy food and gas more frequently than just about anything else.

All of which is true enough, and Leonhardt apparently thinks that's all he needs to prove his case because it's at this point he hits you with the conspiracy zinger:

The conspiracy theories about inflation play off these human
instincts, but they also depend on two other oddities. The first is the
amount of attention given to the so-called core inflation rate. This is
a version of inflation that excludes food and energy, which makes it a
little like a grade point average that excludes math and French.

The core inflation rate does have a purpose. Its movements help Federal Reserve
officials base interest rates on underlying price trends, instead of
being overly influenced by food or gas prices, both of which can be
volatile. But when Ben S. Bernanke,
the Fed chairman, talks publicly about core inflation, he can leave the
impression that the government is cooking the books. In fact, all the
important economic indicators, including real wages, are based on
overall inflation, as are Social Security checks and cost-of-living
raises.

Leonhardt makes it sound as if Bernanke is a sort of absent-minded professor here, focusing needlessly on an obscure and useless metric whose repetition only feeds the tinfoil-hat crowd.  But in fact, the ceaseless harping on "core inflation" is a critical part of the whole scam:  Fedheads believe they can create as much money and credit as they damn well please and face relatively limited consequences as long as "inflation expectations" stay under control.  If people actually start to think prices are rising at a significant clip, well, then the Fed has a problem.

So much for Leonhardt's first "oddity."  Here's the other.

The final piece of the puzzle — and the focus of the Harper’s
article — is the way that the Bureau of Labor Statistics has changed
the price index recently. Back in the mid-1990s, a committee of academic economists concluded that the Consumer Price Index overstated
inflation. To take just one example, years would often pass before the
index included new products — like cellphones — and therefore it missed the enormous price declines that occurred shortly after those products entered the mainstream.

In
response, the bureau tweaked the index. But economists who have studied
the changes say they have had only a modest effect on the inflation
rate, lowering it by perhaps a half point a year. More to the point,
the changes seem to have made the index more accurate than it used to
be.

Having already dismissed the idea that CPI understates the cost of living, Leonhardt feels no obligation to actually, you know, address any of the issues raised by the Kevin Phillips piece in Harper's — which draws in part on the work of the fearless John Williams of Shadow Government Statistics and deals in fact with decades of statistical shenanigans, not just the Boskin Commission flim-flams of the 90s.  Nothing about hedonics, nothing about substitution, nothing about geometric weighting.

Consider Leonhardt's column an early warning.  It's not just the voice of the power elite assuring the plebes that all is well and they can go back to watching American Idol.  It is also the beginning of an establishment push-back against Phillips, Williams, the DR, and everyone else who dares to call BS on government economic figures.  It is heartening that this initial foray is so categorically, irretrievably lame.