In the nineties, under Clinton, the left Keynsians and the right Keynsians – the liberals and the supply siders – staged a quiet revolution. They revolutionized the standard by which the CPI – the cost of living – was measured. No longer would inflation be considered, as it was by primitive man, an increase in price. Primitive man has a broken club, and he trades 5 clams for it. 3 months before, he traded 2 clams for it. Primitive man scatches his head and makes up a big word: inflation. Today’s economists have to sniff at such foolishness.
Why? Primitive man is forgetting that the shape of the club is getting more pleasing all the time! Yes, one must calculate the relationship between price and quality. As Bill Fleckenstein at MSNBC puts it:
“… 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 liberals liked this because, in the 90s, they certainly didn’t want the Fed raising the interest rate. If they were going to have to bite the bullet and balance the budget, they at least wanted an easy money policy. Similarly, the conservatives just loved the idea of not having to pay cost of living clauses in entitlements, balancing the budget on the backs of the increased quality of the goods that the retired and the sick could buy with their reduced money. Economists liked it because it applied the rules of the neo-classical model of marginal utility that they all think founds their discipline as a science.
So the righteous circle – or circle jerk – was formed.
The result is that we are now living in a disconnect between reality and government statistics – and not for the first time. We all know that the stage has been set, with our current administration’s gross overspending, the slipping dollar in an economy that bears a 400 billion dollar yearly trade deficit, and the easy money policy of the Fed, for seventies style inflation. That is, inflation that is not driven by labor costs, but by the sinking costs of money. Yet, according to the statistics, it hasn’t arrived.
That is, according to the reconfigured statistics.
Meanwhile, the great concord of the Clinton era is starting to slip. Because the government can just about make inflation disappear at the drop of a qualitative change, it turns out that there is a lack of at least one external control on federal spending. Moreover, there eventually comes a point when reality kicks in. When consumers are paying higher prices while their economic gurus are claiming that, in actuality, these prices are an illusion, the consumers are eventually going to revolt. In a move that particularly appeals to economists, since it involves three plus variable equations that can only be done by econometricians, even when quality can’t be appealed to, you can appeal to the quality chain. The price of baseballs is going up? But isn’t this a perfect opportunity to switch to tennis balls, the prices of which aren’t going up? And so there is no inflation in the ball market, really. Or so our inflation fighting pals in the Gov. are are determined that we think. They are, in fact, about to bust the back of medical costs by waving just such magic procedures over them, and showing that medical costs are actually at a standstill.
There is a small problem with this: Americans are not inclined, generally, to contemplate appearance and reality while shelling out their money for more expensive stuff. They have a problem with the three variable plus equations. Linear, narrow minded folk, they think the cost of living is about the cost of living. They aren’t with the program.
Gold bugs, who go at economics with sawed off shot guns and usually skew to the right, are pretty incensed, right now, about the CPI shenanigans. This analysis of the CPI is provided by prominent Gold Bug, John Hathaway, at Tocqueville investments, contains three fascinating grafs about the CPI:
“Several years ago, the very important housing component of the CPI was increasing at an annual rate of 4%. Today, that number is 2.2% and heading lower. Housing is weighted at 40.85% of the total CPI. How is it falling when house prices are rising? Simple. The BLI calculates this important component on the basis of “imputed rent” rather than the capital cost of buying a new home. Imputed rent synthesizes the cost of home ownership into a rental factor putting all citizens, both renters and homeowners, on the same footing. The BLS gathers the information for imputed rent, or the “Owners’ Equivalent Rent Index” by asking “each homeowner (surveyed) for their estimate of the house’s implicit rent and what the occupants would get for their rent …. if the owner did rent their home.” (US Department of Labor Program Highlight-Fact Sheet No. BLS 96-5.)
It should be noted that in light of the Federal Reserve’s highly expansionary monetary policy, single-family owner-occupied housing has enjoyed an unprecedented new construction boom. Mr. Banerji observes that a felicitous (for the CPI) consequence of the single family housing boom has been a rise in vacancies and a decline in rental rates for apartment properties. Pressure on the rental market appears to go a long way towards explaining the mystifying decline in the housing component of the CPI. Could it be that the sagging apartment rental market also explains rising bond and equity markets?
There is still more to the tale. Gertrude Stein’s famous dictum: “Rose is a rose is a rose” speaks to the mutation of a word’s meaning over decades or centuries of usage. We can surmise that Big Brother is alive and well at the BLS where a computer is not a computer is not a computer. In other words, added features, memory capacity, and random bells and whistles are not captured in the straightforward list price of a computer. To expunge all continuity of meaning, the BLS brought forth “hedonics”, the science of measuring the value of a product or a service after allowing for qualitative improvements. A laptop with twice the memory as last year’s model sold at the same price this year is counted as a 50% price reduction. This sort of analysis was applied initially to computers and IT equipment. More recently, a broad range of consumer goods including electronics and automobiles has been subjected to hedonic measurement. Health care has been a particularly ill behaved sector of the CPI. Hospital services, nursing homes and adult day care, for example, increased 141.4% over the period 1990 to 2003, versus an average of 46% for all items measured. It should come as no surprise, then, that the Bureau of Economic Analysis is considering adjusting prices of medical services for quality changes (Grant’s Interest Rate Observer-1/30/04.)”
LI agrees with the anger of the gold people – in spite of their odd politics. In fact, we have been expecting inflation as the traditional accompaniment of war for some time now. In WWII, to pre-empt rising prices, FDR instituted rigid price controls and rationing. In the Greenspan era, however, inflation has been spun into nothingness, just as Hathaway describes. But you can only fool investors for so long with these kinds of tricks. Especially when Japan and China, with their trillion dollars of dollars, are going to be holding the bag as the real purchasing power of those dollars makes them worth less and less. For further righteous rightwinger anger at Greenspan, see links here and here.
A less biased account of hedonics, including abstruse equations that make LI’s eyes cross, has been put out by Charles Hulton, with the imprimatur of the NY branch of the Fed. One of Hulton’s examples of a qualitative change that the economists at the BEA take as a deflationary offset to a surface inflationary price change is this: the production of aircraft with a larger capacity to carry customers. It is here that economics becomes, as one hedonics critic puts it, poetry. Or more like the literary criticism of poetry. I have never encountered anyone who believes that the lessening of seat space on airplanes is a positive qualitative change. Yet a hedonics calculus could well posit that this is a positive change in quality – it leads to more customers being able to use airplanes – and thus should accrue some variable value that can be embedded in the equations to give us an inflation rate. As the Gold bugs constantly and correctly re-iterate, the bias in the BEA procedure is to posit qualitative changes as inevitably positive. That there is now a much longer wait at the airports then there was pre-9/11 simply doesn’t get into the BEA computer – but if some mechanical way was found to shorten airport wait time, that change would get into the BEA computer. It is a gamed system.
Most interesting part of Hulton’s article deals exactly with problems of the airplane seat kind: the problem of market power, of interpreting quality from the producer’s point of view as mapping non-controversially over the consumer’s point of view, and the problem of substitution. He quotes an interesting neo-Galbraithian guy named Pakes, who wants to include a variable for market power in the hedonic equations; that complexity – that is, non-linear shifts – should not be considered bad data, to be smoothed out to produce equilibriums, but should rather advance us to another level of equations to gain a finer grained matching to real markets. However, Pakes is not advocating getting rid of hedonics. By no means.
Hulton ends the paper with one of those irritating econometricians gestures, saying that the objection to hedonics is simply that it is new. This implies that the revolt against a method that stands in stark contrast to the whole purpose of figuring out a consumer price index is simply the result of superstition and inertia. This simply isn’t so – it is the result of objecting to the hijacking of a perfectly good index by methods that properly generate a whole other index – call it the Consumer Quality Price Index. Hulton's own superstitions, of course, remains consistent with the general bias of his discipline, the apologetic agenda for capitalism and an upper class p.o.v. that econometrics simply encodes.