Wednesday, November 17, 2010

Why the CPI is understating inflation

The big news on the inflation front today is that the 12-month change in the October Core CPI fell to 0.6%, the lowest rate of inflation in the history of the index, which began in 1957. Even the headline CPI was up only 1.2%, which ranks as among the lowest rates of inflation since the low-inflation, early 1960s. To add to the "inflation-is-very-low" news, the CPI index today is still lower than it was in July '08! Yikes, that means we've had over two years of deflation! This presumably makes the Fed governors feel more comfortable with their view that inflation is now "too low" and needs to be pumped up a bit. But with all due respect (a scarcer commodity these days among many reputable economists), I think they are mistaken.

This next chart shows the CPI index less energy, which represents a little less than 10% of the CPI. Since oil prices peaked in July '08, non-energy prices have increased 2.3%, and they have increased despite a wrenching recession and almost a year and a half of 9.5% unemployment. Thus, one big reason the overall CPI is so low is that, since Aug. '08, we've seen oil prices fall by almost half; natural gas prices fall by more than 80%, and gasoline prices fall by 30%. 

The other big reason for low CPI readings is the housing market. As Brian Wesbury has noted repeatedly, one of the key factors depressing the CPI over the past year or so has been the government's estimate of how much rent homeowners would be paying if instead of owning their house they were renting it. If you subtract owner's equivalent rent from the CPI and look just at things that people are actually spending money on, then the CPI would be up 1.5% in the past year. Since owners' equivalent rent is now beginning to increase, after falling from mid-2009 through mid-2010, an important source of lower inflation in the past year or two will begin to add to CPI inflation in the months and years to come. Consider: owners' equivalent rent makes up about 25% of the CPI, is currently unchanged over the past 12 months, and averaged about 3% per year for the 10 years prior to the housing market bubble. If it were to return to 3% a year, owners' equivalent rent could add 0.75% annually to the CPI. 

This next chart comes from a relatively new and very intriguing source, the Billion Prices Project @ MIT (HT: Mark Perry). The chart compares the year over year change in the CPI (blue line) with the year over year change in the price of a basket of millions of consumer-oriented items as priced daily by MIT via online sources. There are some key prices, however, that aren't included, such as energy, cars, transportation, and services such as education, health, and tourism, but those only account for about 40% of the CPI. Despite this, MIT reports a high correlation between their data and the CPI. In any event, the MIT project is reporting inflation of about 1.8% over the past 12 months, compared to the 1.2% gain in the CPI and the 0.6% gain in the core CPI. One more reason to think that the CPI may be under-reporting inflation.

If the MIT index were to include cars, it would be registering even higher inflation, as this next chart suggests. According to the folks at Manheim Consulting, used car prices rose 4.7% in the past year. Since the end of 2009, used car prices have rebounded over 25% to a new all-time high, despite the fact that the economy has not fully recovered, and the labor market is still struggling with 9.6% unemployment.

If there is any deflation out there, it can be found mainly in the energy and housing sectors, both of which experienced a huge runup in price in the years prior to 2008. In my book, that's not deflation, it's pay-back. Almost anywhere else you look, prices are rising.


Jake said...

Energy is up significantly year over year, yet overall inflation is only 1.2%

Benjamin said...

Inflation is dead, and getting deader.

Also, please note: Serious research on this topic suggests the CPI overstates inflation.

"How big is the CPI’s bias? Well, in 1996, the Social Security Administration commissioned a study on the accuracy of the CPI as a measure of the cost of living. This so-called “Boskin Commission Report” said the CPI was overstated by about 1.1 percentage points per year. The commission identified several sources of potential bias, but about half of the 1.1 percentage points resulted from new products and quality changes that were slow or otherwise imperfectly introduced into the price statistic.

Since that time, the Bureau of Labor Statistics has initiated a number of methodological changes that have reduced the CPI’s mismeasurement. In a 2001 paper, Federal Reserve Board economists David Lebow and Jeremy Rudd put the CPI bias at only about 0.6 percentage points. And again, of this amount, the big share of the bias (about 0.4 percentage points) resulted from the imperfect accounting of new and improved goods.

Now, in an article (available to all in its working paper version) appearing in the latest issue of the American Economic Review, Christian Broda and David Weinstein say the earlier estimates of the new goods/quality bias may be a bit understated. The authors examine prices from the AC Nielsen Homescan database and conclude that between 1996 and 2003, new and improved goods biased the CPI, on average, by about 0.8 percentage points per year. If this estimate is accurate, consumer price increases since last October would actually be around zero, or even slightly negative, once we account for the mismeasurement of the CPI caused by new and improved goods."

So, the CPI probably overstates inflation by perhaps one percent.

But overall, sheesh--we are talking flatsville here. Kansas looks hilly next to this.

Benjamin said...

Also, please note: The MIT stuff is global prices. China and India etc

Benjamin said...

Dr. Perry just published this:

"According to the Cleveland Fed's report today, its median CPI measure of prices increased by only 0.50% in October over the same month last year, the same as the year-over-year rate for each of the last three months (August - October). October marked the 26th consecutive month that the median CPI annual inflation rate dropped or stayed the same, and the 0.50% inflation rate in six out of the last seven months is the lowest year-to-year inflation rate in the history of the Cleveland Fed's series back to 1984 (see chart above). In contrast, the regular CPI from the BLS increased by 1.2% over the last year (October 2009 to October 2010), and has held steady at between 1.1 and 1.2% for the last five months.

Historically, the median CPI has been 50% more accurate at gauging future inflation than the traditional CPI (based on the Cleveland Fed's research), and neither the median CPI from the Cleveland Fed nor the CPI from the BLS is showing any signs of inflationary pressures."

Boy, it ain't inflation that I worry about.

John said...

A valiant effort, sir, but today's data suggests you're wrong about inflation and Bernanke's right to worry about deflation. We are likely to tend toward deflation as long as unemployment is high.

brodero said...

Owner's equivalent of rent definitely subdues CPI but can you imagine if they substituted Case Shiller??

Case Shiller National Index
2q 2006 189.93
2q 2010 138.03

Owners equivalent of rent of residence

2q 2006 236.93
2q 2010 256.209

The deflation of the asset has been dramatic...

brodero said...

Not to get too esoteric but if
take the median new home price and
divide it by the owners equivalent of rent see a gradual
rise in this ratio over the last 27 years...There were 2 spikes that the late 1980's and
the 2005 to 2007 time frame. We have since returned to the levels
we saw in 2000. A gradual rise in the owners equivalent of rent index
will buttress home prices ( especially when the inventory overhang is lessened)....

marmico said...

A valiant effort. I beg to differ.

Grannis doesn't understand the difference between disinflation and deflation. I guess extreme pre-1985 supply-siders have finally crawled out of their caves.

Now if Grannis had a consistent belief system he would compare core PCE with core CPI due to differentials in scope, weight and index calculation. He doesn't, so let me elucidate.

Core PCE is running higher than core CPI. Why?

It's all about weights. Medical care always runs hot. PCE weights it 2x CPI. Owners equivalent rent is running cold. CPI weights it 2x PCE.

The only thing understated is Grannis grasping at straws (data mining) in support of his horrendously wrong call on inflation and the 10-T yields.

marmico said...

brodero, a picture is worth a thousand words. Granted it's an outdated chart as the red line has popped above 0%. The red line will once again be below 0% by January 2011, however.

Learn to link. At least you offer some data in opposition to "Stockey-Jockey" Johnnie who's raison d'ĂȘtre is buy financials. Gurgle, gurgle...

brodero said...

Give me the hard data every time
that is the way to truly judge a situation....percentage changes
always should be looked in context...valuable but with a
discerning eye....

brodero said...

Marmico, as opposed to finding things that justify your belief
study the stress test ,read some 10Q's look at outside sources and challenge your beliefs. I try to do
it every day.

Dr William J McKibbin said...

The future will be inflation, and history is my proof -- eventually, both fiscal and monetary policy-makers will yield to the horrors of deflation, which means monetary expansion -- my advice is to prepare for the coming inflation by holding high quality equities that pay regular dividends...

John said...

Financials will be big winners over the next few years. Money Market Marmico...don't forget to declare the interest on your account on your tax return. You DO file one....don't you?

John said...

Doc McKib..

I agree. In fact I suggested same in late August here on Scott's board. The S&P 500 was between 1040 and 1060. This morning its a bit over 1200. I was castigated by our resident snark at the time I recall.

Benjamin said...

Dr. McKibben:

How do we get to inflation from 0.5 percent increases in the CPI?

And remember, when the great "inflation fighter" Ronald Reagan left office, inflation was running at 6 percent.

We survived 6 percent inflation easily. We survived 4 percent inflation easily.

Now, people lose their bowels when we might get to 1 percent inflation.

When did ultra-low inflation become a fetish among right-wingers?

marmico said...
This comment has been removed by a blog administrator.
Bill said...


You seem to be suffering from a bad case of Roubini envy. I bet it just kills you that nobody knows who you are and you can't get a spot on CNBC to spout your negativity about the economy.

Steve Fulton said...

The CRB raw industrial materials index consisting of the following
components has just hit a new all time high.
01 = Burlap
02 = Butter
CC = Cocoa
HG = Copper Scrap
C2 = Corn
06 = Cotton
30 = Hides
LH = Hogs 09 = Lard
10 = Lead Scrap
11 = Print Cloth
12 = Rosin
13 = Rubber
BO = Soybean Oil
57 = Steel Scrap
LC = Steers 17 = Sugar
18 = Tallow
54 = Tin
MW = Minneapolis Wheat
KW = Kansas City Wheat
22 = Wool
23 = Zinc

I smell inflation coming.

Scott Grannis said...

Actually, the CRB Raw Industrials hit a new all-time high about 11 days ago. It has since dropped from 568 to 551. Still high enough, nevertheless, to suggest higher inflation in the pipeline.