Are you better off now than you were four years ago, or 10, or 20? The answer is to be found not in tables of economic statistics but in your kitchen, your family room, and your medicine chest. Would you want to return to a world without microwave ovens, camcorders, automatic coffee makers, VCRs, answering machines, compact discs, and disposable contact lenses? How many people do you know who are living better through Prozac or Viagra, or living longer through cholesterol-reducing drugs such as Pravachol and Lipitor? A decade ago I suffered from incapacitating hay fever; since the drug Flonase became available, I am symptom-free.
You could devote a lifetime to parsing the data on median family income and still completely miss the story of what has happened to the median family's actual lifestyle. Quality-of-life improvements aren't easily expressed in numbers.
But applied economists love numbers, for more than one reason. You need numbers to test the predictions of economic theories. More prosaically, you need numbers to calculate things like cost of living adjustments to Social Security. If the average microwave doubles in price with no change in quality, that's a cost of living increase. If the average microwave doubles in price because people are choosing better microwaves, that's something else.
Here's where Pete Klenow and Mark Bils come in. Klenow is an economist at the University of Chicago. Bils is a colleague of mine at the University of Rochester. The two of them have given a lot of thought to how you'd measure quality changes over time.
Here's their idea: For starters, forget about changes over time, because changes over time are hard to think about. (We'll get back to them later though.) Instead, look at the gap between the rich and the poor. Rich people tend to own more microwaves and they also tend to own better microwaves. More interestingly, there is some statistical regularity to the way the rich divide their wealth between quantity and quality. If you're rich enough to own two microwaves instead of one, they'll cost, on average, about 25 percent more than your poorer neighbor's single unit. The poor man pays $200 for one microwave. The rich man pays $250 apiece for two (presumably better) microwaves.
For other goods, the 25 percent rule doesn't hold, but some other rule holds instead. Take living room tables, for example. If you have two tables and your neighbor has one, yours are probably about twice as expensive as his. If he pays $500 for one table, you'll pay $1,000 apiece for two. So, the 25 percent rule for microwaves gives way to a 100 percent rule for living room tables. Klenow and Bils have estimated the rules for 50 different goods, ranging from vacuum cleaners (a household with twice as many vacuums pays about 22 percent more per vacuum) to trucks (a household with twice as many trucks pays about 140 percent more per truck).
Now, when prices rise over time, it's hard to tell whether you're seeing inflation or a reflection of quality improvement or some combination of both. But when different people pay different prices at the same time, we know inflation has nothing to do with it. So we can infer that when rich people pay more than poor people, they're paying for quality.
In other words, when two microwaves are bought on the same day, the one that costs 25 percent more is presumably 25 percent better. That gives us a measuring rod. We know that when the number of microwaves doubles, their average quality goes up by 25 percent. When the number of living room tables doubles, their average quality goes up by 100 percent, and so on.
K lenow and Bils assume these ratios haven't changed over the years (and they offer evidence to support that assumption). That's the key to computing quality changes over time.
If in 1980 the average household had, say, .5 microwaves and in 1995 the average household had one, then the twofold increase in quantity should be associated with a 25 percent increase in quality. Or, if in 1980 the average household had .5 microwaves and in 1995 the average household had 1.5, the 200 percent increase in quantity should be associated with a 50 percent (that is, 200 percent times 25 percent) increase in quality.
Averaging over all goods, Klenow and Bils estimate that quality has been increasing at about 1.5 percent per year for the past 20 years or so. (This is still probably an underestimate, because it fails to account for the introduction of entirely new products.) By contrast, the U.S. Bureau of Labor Statistics adjusts its inflation figures by assuming roughly a .5 percent annual rate of quality growth. That means the BLS systematically overestimates inflation by about 1 percent per year.
How does that affect your grandmother's Social Security payments? Arguably not at all. The naive view is that if the BLS overestimates inflation by 1 percent, cost of living adjustments will be 1 percent too high. A more plausible view is that Social Security payments at any given moment depend only on the political clout of groups such as the American Association of Retired People and not at all on how anything is measured.
If the AARP is powerful enough to demand a 5 percent increase and inflation is measured at 3 percent, they'll get a 3 percent cost of living adjustment and an additional 2 percent on some other pretense. If inflation is measured at 4 percent, they'll get 4 percent plus 1 percent. And if inflation is measured at 6 percent while the AARP is in a position to demand only 5 percent, they'll get a 6 percent cost of living increase coupled with a 1 percent cut.
I don't know how to prove that theory, but it strikes me as self-evident. The alternative is to suppose that the entire political system, with all its checks and balances, is in thrall to the way some economist happens to calculate a number. I don't believe we're that powerful.