Priced to Move
The CPI can't keep up with a dynamic economy.
Styling aside, would you rather have a brand-new 1972 car or the equivalent 1997 model? A new house circa 25 years ago or one built today? The medical care of 1972 or 1997? The restaurants? Winter fruits and vegetables? Disposable diapers, shampoo, TV sets, diet soda, panty hose, sneakers, bathroom cleansers, alarm clocks, contact lenses, or stereo systems?
The past few decades have witnessed extraordinary progress in the quality and variety of goods and services. Nor does that progress end with the microprocessor and its well-publicized offspring, however amazing and important they may be. It encompasses adhesives (in everything from Post-It notes to beltless sanitary napkins to easily removable price tags), packaging (remember when shattered glass shampoo bottles were a real danger, store-bought cookies were always hard, and metal toothpaste tubes crumbled with age?), fibers (wrinkle-free slacks, harder-to-run hose, stain-resistant carpet), and chemicals galore (from scrubbing bubbles that attack soap scum to Retin-A that attacks wrinkles). Cars last longer, airplane rides are smoother, long-distance calls are clearer, and even the bread tastes better.
Americans are, however, a forgetful lot. We take our good fortune for granted and concentrate on what we don't have. That restlessness can itself be a spur to progress, but it can also become perverse: GenXers who have never seen a blurry TV picture or lived without central air conditioning grouse about their hard lives–and they're often more optimistic than their elders. The answer to the question, "Are you better off than you were 25 years ago?" is widely assumed to be no. Indeed, when REASON published a cover story in December 1995 arguing that by numerous objective measures–from the size of new homes to the length of work weeks–the standard of living has been rising since the early '70s, many readers were outraged. (See "The Good Old Days Are Now," December 1995.) Conservatives believe the good old days were yesterday, while leftists just hate capitalism and, increasingly, progress as well.
This debate is at the heart of a seemingly technical report titled "Toward a More Accurate Measure of the Cost of Living," prepared for the Senate Finance Committee by an all-star team of economists led by Stanford's Michael Boskin. Released in early December, it argues that the consumer price index has been significantly overstated at least since the 1970s and should be revised. It's rare for a handful of economic technicians to make big news, but the Boskin Commission did–and will continue to over the coming year.
That's because the commission's recommendations, which include ideas for creating a truer cost-of-living index, have significant budgetary implications. Even before the final report was out, journalists were warning of "a looming stealth attack on the deficit and growing entitlement programs," in the words of Susan Dentzer of U.S. News & World Report. The CPI drives both automatic increases in federal handouts, notably Social Security, and inflation adjustments in income tax rates. But, says the commission, the CPI overestimates inflation by at least 1.1 percentage points a year. That translates into more than a trillion dollars in federal taxes and outlays between now and 2008. Not surprisingly, the entitlements lobbies have geared up big time to attack the commission's conclusions.
But there's even more at stake in this debate. By questioning the CPI, the commission is challenging not only some big items in the federal budget but the conventional wisdom about our recent economic history. The entire gloom-and-doom industry is threatened. As Boskin wrote in The Wall Street Journal, "The implications of overstating inflation for understanding economic progress are equally striking. Instead of falling by 13%, real hourly earnings have risen by 13% from 1973 to 1995. Certainly, there has been a slowdown in wage growth, but not a decline. Real median family income over the same period grew 36%, not the puny 4% in the official statistics that deflate by the CPI."
The commission's report identifies several sources of error in the CPI, all stemming from a central fact: the dynamism of the American economy. "If the American economy was quite static," says the report, "with very few new products introduced, very little quality improvement in existing products, little change in consumers' income, and very small and infrequent changes in the relative prices of goods and services, measuring changes in the cost of living would be conceptually quite easy and its implementation a matter of technical detail and appropriate execution. Fortunately for the overwhelming majority of Americans, our economy is far more dynamic and flexible than that."
As a result, pricing an essentially static "market basket" of goods–without considering how consumers switch among products in response to price changes, how people change where they shop, what new products have been introduced, or what quality improvements have taken place–tends to make the cost of living appear to rise faster than it really has.
In real life, we don't stick with a fixed market basket. When the price of beef rises, we eat more chicken; the price of maintaining the same standard of living may still go up, but not as fast as it would if we couldn't switch. We also buy new products, whose prices are often falling rapidly. (By waiting until 1987 to add VCRs and 1998 to count cellular phones, for example, the CPI has missed huge price drops.) And we all know–at least when we bother to consider the question–that most of today's products are superior to yesterday's. From laundry detergent to computers, light bulbs to shaving cream, we simply assume improving quality. Yet the CPI generally treats today's product as identical to yesterday's in every aspect except price. It also misses, or misinterprets, changes in where we shop. Although discounters such as Wal-Mart have boomed in recent decades, offering both lower prices and often equal or better service than competing stores, the government's statisticians have all but ignored the trend.
They've also crammed fluid "products" into rigid categories. For CPI purposes, for instance, "full-fare" airline tickets are one thing and "discount" tickets quite another. To travelers, a ticket is a ticket, prices fluctuate constantly, and you shop around for the best deal. The CPI may claim that prices doubled between 1980 and 1990, but that's not what we air travelers saw. "Many Americans," writes Peter Grier of The Christian Science Monitor, "might judge that their actual air transportation costs fell on a per-mile basis during that period, as super-saver fares swept through the industry." I'd guess that mine rose maybe 10 percent–an increase, thanks to the industry shakeout in the mid-'80s, but hardly a doubling.
None of this means that the people at the Bureau of Labor Statistics are incompetent. But what they're trying to do may be impossible. To analyze quality fluctuations, the Boskin Commission draws on several decades of very specialized and technical work on price changes in specific industries such as pharmaceuticals, computers, and apparel. Reading its report and some of the papers it footnotes is a great way to get a sense of how complex the economy is–and how incredibly difficult it is to measure something as subtle as the cost of living. What, indeed, is an "airfare"? How do we account for the declining number of defects in new cars? For video rentals replacing movie-going? The entertainment value of restaurants? Better stereo sound and nearly perfect TV pictures? Is a Granny Smith apple really a different product from a Golden Delicious? How different? How exactly do you measure the quality difference between Wal-Mart and Penney's? (And all this is ignoring serious mathematical and statistical issues.)
The more you know about an industry, the more questions you see. In a conservative moment, for instance, the report assumes no significant quality improvements, and hence no CPI bias, in newspapers and magazines–thereby missing the phenomenal growth of color printing and the increased timeliness of news publications, both effects of technological advances and neither reflected in higher prices. One question leads to another, eventually calling the whole enterprise into doubt, especially if it's more than an intellectual exercise. "Peer behind the statistical curtain," writes Kim Clark of Fortune, "and you'll wonder whether economists can–or should–try to measure the true cost of living." Clearly there is such a thing as an overall rise in the price level, distinct from individual product variations, but calculating it in a rapidly changing economy seems nearly impossible.
Part of the problem is that the nature of economic progress has changed. Once, in the not-so-distant past, it was mostly a matter of getting more stuff. Everyone wanted three meals a day, indoor plumbing, a bedroom for each of the kids, air conditioning, color TV. When progress is just a matter of getting more, it's relatively easy to recognize–and to measure. Refrigerators, elevators, and smooth, modern highways were exciting in the late 1930s, writes David Gelernter in 1939: The Lost World of the Fair, nostalgic for the period's optimism: "Technology was a less capable yet far more wonderful thing in 1939 than it is today." We tend to recognize dramatic changes, not subtle improvements.
The desire for more hasn't gone away: We want Net-surfing computers, cellular phones, Nintendo 64, and Tickle Me Elmo. But progress isn't as obvious as it was in the earlier, poorer eras when the CPI was developed. The "extensive" phase of economic progress–where the major challenge for producers is to distribute the same basic goods to as many people as possible–is over for most products in the United States. We've reached the "intensive" stage, where incremental improvements and niche products–increases in quality–become important. These improvements are real, but they are much harder to count. So when they show up in higher prices, it's easy to attribute the increase to inflation. Techniques that once worked reasonably well for measuring the cost of living don't work as well today. They not only distort federal spending patterns (and give us regular tax cuts), they teach us to feel sorry for ourselves.
"The current CPI seems to make sense of our recent economic history," writes a leading critic of the Boskin Commission, economist Dean Baker of the left-leaning Economic Policy Institute. "Adjusting for a large overstatement in the CPI creates measures that do not fit the world we know." The past was not, he says, as bad as the revisions would suggest and the present and future can't be as good. "If the Boskin Commission is right, our children and grandchildren can look forward to an incredibly prosperous future," and that, he implies, surely cannot be.
Baker and his allies have invested an enormous amount of emotional, intellectual, and political capital in the notion that a dynamic economy is destructive, that our recent economic lives have been not merely turbulent but terrible. By parsing the statistics behind the politics, and finding them wanting, the Boskin Commission has called that message into doubt. "The world we know" is not in fact gloomy, nor should our children expect an impoverished future. We don't really need a commission of Ph.D.s to tell us this–with some memory and imagination, a trip to the supermarket should suffice.