Some government data suggest computer and car prices, among many other things, are falling. But when was the last time you paid less for a car? Here’s why you should be concerned.
By Bill Fleckenstein
Please join me this week in a trip to the government department responsible for fun with numbers. Those D.C. statisticians may churn out their work with a straight face, but that doesn't mean we have to fall for it. Among the skeptics are Steve Milunovich of Merrill Lynch, Jim Grant of Grant's Interest Rate Observer, and, of course, yours truly.
In a recent report, Milunovich noted that the Bureau of Economic Analysis (BEA), whose job it is to compute the Gross Domestic Product each quarter, has "stopped reporting the real computer hardware shipment figure used to calculate real GDP growth, though it is still used in GDP calculations." The BEA, which is part of the Commerce Department, made this readjustment because it is "concerned the rapid price declines for computers made the figures misleading."
Let's stop and review the bidding for a second. Remember: GDP is the measure of goods and services produced in this country. The government decided that certain of its data series involved in calculating GDP were misleading. So, what did it do? Simply stop breaking them out. Makes sense to me; how about you?
This would be a humorous window into the lunacy of government calculations, were it not so important to many statistics. Regular readers of my daily column know that the magic of "hedonics" and all its attendant distortions is something that I have railed about for a long time.
Hedonics: 'miracle' tonic for an ailing economy
For those of you who don't know, hedonics is the way the government transforms price declines into quality improvements. To wit, you buy a PC with twice as much power, so the government concludes that you really paid only half as much money for it. Hedonics is also the government's way of taking quality improvements and converting them into price declines when calculating the CPI. Sure, that brand-new Chevy you just bought cost 40% more than it used to, but it's a 40%-better car for a variety of reasons. So, the government says, the price didn't really go up. (I have oversimplified these examples, but you get the point.)
The idea behind the first case at least makes some sense, though the government carries it too far by acting as though improvements can be precisely measured. The problem with the second case is that those quality improvements are not voluntary. Since you have to pay the new price, it's sheer silliness to say that the price really didn't go up.
There are other ramifications as well. It turns out that the computer-spending component has materially warped GDP calculations in many of the last eight quarters. To put the numbers into perspective, from the second quarter of 2000 through the fourth quarter of 2003, the government estimated that real tech spending rose from $446 billion to $557 billion, when nominal spending only increased to $488 billion. That extra $72 billion represents the value the government imagines the improvement in computer quality is worth.
Now $72 billion doesn't sound like a huge amount in a $10 trillion economy, but at the margin, it makes a difference. And in fact, the contribution of this tech component to real GDP comprised about 12% of growth in the third quarter of 2003 and more than 30% of growth in the first quarter of 2003, i.e., a big chunk of the growth. Since real growth is a factor in the calculation of productivity and productivity growth, these statistics are also distorted.
Slippery CPI, iffy TIPs
Our government has admitted its Alice in Wonderland hedonic-adjustment exercise has produced numbers so distorted that it doesn't want to show them to you. Yet it continues to use the "analysis" and some of the data in calculations of real GDP, productivity growth and CPI calculations. This is one of the reasons I've never been a big fan of Treasury Inflation-Protected Securities, or TIPS. I have stated a million times that the government's calculated cost of inflation, in the form of the CPI, is a joke. So, I refuse to buy a security that's indexed to the CPI, unless the price of the security reflects my skepticism. TIPS may be better than straight bonds, but they're not as good as most people think. As you can see from these two examples, the government aims to cheat you in the calculation, and besides, TIPS don't protect you from a decline in the value of the dollar.
Grant on BEA balderdash
Further reason to be suspicious of all government data comes from the current issue of Grant's Interest Rate Observer, in which the ever-alert Jim Grant broke the story that "the Bureau of Economic Analysis is weighing a study to explore the merits of adjusting the prices of medical services for quality changes."
In other words, the BEA is considering the use of hedonics to lower the impact of rising medical costs on the CPI by subtracting the imagined value of quality improvements in medical care from the price we’re really paying. The government recognizes it has a problem with exploding health costs and is studying the use of that same quick fix which has "worked" when unwelcome rising prices have been an issue in other areas, i.e., to define the problem away. I would imagine that when the folks at AARP and organized labor find this out, they'll be up in arms. Maybe their clout can stop this nonsense before it gets even worse.
Take heed, enjoins Sir John
On a final note, I would like to share with readers a rather interesting comment that John Templeton, founder of the Templeton Funds, made to Paul Kangas during a PBS interview last Monday.
Templeton's quote will be instantly recognizable to folks who have read the longstanding header on my Web site, which echoes one of my most fervently held views: "In a social democracy with a fiat currency, all roads lead to inflation." (Readers of the Contrarian Chronicles may also refer back to my Nov. 17 column, "All roads now lead to inflation."
And now for Sir John's wisdom: "All currencies, not only the American dollar, but all currencies, always go down, mainly because of democracy. The voters will vote for a person who is going to spend too much, and so you have to expect all currencies to go down." In future columns, I'll have more to say about the dollar, the variables affecting it and why this should be of concern to you.
(Editor’s note: the Bureau of Economic Analysis published an article in 2000 about the use of hedonics in valuing the impact of computers on the economy. You can access it here.)
