United States Gross Domestic Product increased 3.0 percent in 2003, a pretty satisfactory performance even allowing for the approximately 0.9 percent increase in population. Yet Census Bureau figures released Thursday showed that real median per capita income declined by $63, to $43,318. So where’s the growth going, and why aren’t we having more fun?

The detailed breakdown of the Census Bureau figures throws some light on the story, although it casts even more gloom over the figures if you’re concerned about income distribution and poverty levels. The Gini coefficient, a measure of inequality, stayed level at 46.4 in 2003, but inequality in the United States remains far higher than it has been traditionally since World War II – the Gini coefficient in 1967, for example, was only 39.9, already substantially up from its level in the late 1940s. This measure of inequality is not just higher than in Europe, it is significantly higher than in most countries with adequate economic growth, and is approaching the pathological Latin American levels of over 50. It is also, as far as one can tell, higher than in 1929, indicating that all the advances in income distribution of the New Deal have been lost, leaving us only with an immensely enlarged and distorted government.

Further, poverty increased in 2003 from 12.1 percent of the population to 12.5 percent; while this calculation is highly spurious (since the poverty level is adjusted by earnings not prices, and hence measures only relative poverty, not absolute) it is a further indication that all is not well in the American heartlands.

As proponents of educational reform point out, the losses in income are concentrated among the less educated. Those with some high school education lost no less than $751 of median earnings between 2002 and 2003, falling from $19,095 to $18,344, while the median earnings of professionals jumped by no less than $5,338, from $76,659 to $82,007.

Over the longer term, the picture is both clearer and even gloomier. The Census Bureau points out proudly that median earnings have increased by 30 percent since 1967, but a moment’s thought shows is that there’s something wrong here. Real GDP per capita has more than doubled since 1967 (up 103 percent to 2003), so why have median earnings increased so little, at well under 1 percent per annum?

If you adjust for education, the picture becomes clearer. Roughly (the Census Bureau keeps changing the basis for its figures, so that accurate tabulation becomes impossible) while median income has risen by 30 percent since 1967, those with postgraduate qualifications have seen their median income increase by 20 percent, while those with some high school education but no diploma have seen their median income decline by about 28 percent.

These figures show two things. First, both the returns to education, and income inequality in general have markedly widened since 1967. Second, median incomes for any given educational cohort have increased by much less than 30 percent, but in 2003 there were relatively far more professionally qualified people and far fewer high school dropouts than in 1967.

So if you felt that you hadn’t made much material progress over the last 35 years, you were right. Indeed, the American consumer’s distress at his lack of material progress is pretty clearly indicated by his ever escalating levels of consumer debt. The rise in house prices as a percentage of income has priced big-city housing out of the range of many middle income consumers, but it has enabled those lucky enough to have got on the housing ladder to refinance their mortgages at regular intervals and raise their living standards that way.

The puzzle of why GDP has risen so strongly while median incomes haven’t is resolved in two ways, each responsible for explaining part of the differential. First, government statistics, particularly in the prices area, have been massaged substantially in the last decade, and so don’t represent what you think they represent. Second, 1967 represents the beginning of the second great wave of immigration into the United States; while the immigrants may be beneficial to the economy as a whole, the law of supply and demand says that if you greatly increase the supply of something, you will lower its price – in this case, the sharp increase in the supply of new labor has lowered wage rates.

Consumer price statistics are particularly dubious, because the Bureau of Economic Analysis played with the Consumer Price Index in the middle 1990s, during the Bill Clinton administration, shifting it to a “hedonic” basis in respect of goods like computers that were increasing rapidly in performance and decreasing in price. “Hedonic” (a fancy bit of Greek meaning pleasure) price indexes are supposed to capture the amount of satisfaction produced by an item, as distinct from its mere physical size and characteristics.

Suspicious consumers would be wise to question whether the hedonic adjustments that are made are “real” in the sense that they reflect genuine quality improvements. On close examination, it is clear that they are not, for two reasons. First, an increase in processing power or speed of a computer product or a telephone may in some cases produce an increase in that product’s utility to its user, but in other cases, it may not. Internet-enabled computers, for example, are many times faster and more powerful than they were half a decade ago, but their ability to carry out most tasks has not significantly improved; indeed, because of the blizzard of spam and viruses that infest the Internet, it has generally declined. The exception is applications, such as video, requiring large bandwidth and rapid download speeds — but those applications are heavily used by only a minority of computer users. Spreadsheets, word processing and database management are all only infinitesimally more capable than they were in the middle 1980s, after the first generation of PC applications of those types was well established. Thus the “quality improvement” captured by continually declining prices often does not reflect anything that is real to the consumer of the product.

Hedonic pricing is currently used for around 20 percent of the Consumer Price Index, which immediately begs the question: what about the other 80 percent? Here is hedonic pricing’s second guilty secret: it is used only for products where performance metrics are increasing rapidly, primarily in the tech area. If the price/performance tradeoff for other products was constant, this would not matter, but it is not. For example, automobiles from 1970 onwards have been subject to a series of increasingly intrusive government regulations, each designed to benefit some social or environmental agenda, almost none of them benefiting the car buyer or user. Hedonic pricing in automobiles, if done honestly, would see a decline in the utility of the automobiles to the user, even as additional features were added to the car, thereby increasing its price. The same problem applies to domestic appliances, in which cost squeezes, which have led to their declining reliability and durability, have damaged the interests of consumers, but have not been reflected in price indexes.

The decline in standards of customer service produced by cost-cutting and call center technology also goes un-reflected in hedonic price indexes. Consumers are compelled to spend hours of their time dealing with shoddy customer service or, even worse, are bullied into paying for services they do not want by the scams that infest the consumer market today. Here, the consumer’s costs have increased, without any corresponding increase in benefits. Hedonic pricing, so quick to take account of the imaginary improvements in consumer quality resulting from technology, fails to reflect the deterioration in consumer quality caused by technology-assisted rip-offs.

Another area in which hedonic pricing would produce a negative result is housing. As population increases, and suburban sprawl grows, the consumer is forced to live further and further from his job, and to spend more and more hours per day commuting. The hedonic value of a house in an outer suburb is thus considerably less than that of a house in a close-in suburb 30 years ago. True hedonic pricing would reflect this increased hedonic cost of home ownership; needless to say, the current CPI doesn’t.

In the last few years, there’s been a further problem with the CPI’s treatment of housing. As is well known, the CPI calculates a price equivalent of the rental cost of housing, rather than looking at house prices directly. The idea of this is to avoid distorting the CPI by a period of house price inflation, unaccompanied by rental inflation. However, house price inflation is an increase in a real price that over 70 percent of consumers, those owning homes, have to pay (and is reflected in their property taxes.) In a period of very low interest rates, such as in the U.S. in 2001-04, rentals may not rise while house prices soar, as the equilibrium yield on housing investment declines, but so what? For most people, the price of one of their most basic needs is rapidly rising.

From the hedonic pricing effect and the housing effect, the CPI in since 1995 has undercounted the true level of price rises by at least 1 percent per annum, and by more in 2001-04, thereby inflating growth rates in real GDP. There never was a danger of price deflation, there was and remains a severe danger of prolonged decline in per capita GDP.

On immigration, I’ve made the argument before, as have others. If you greatly increase the supply of labor, you may increase GDP, and even GDP per capita, but you will reduce the returns available to labor, particularly at the unskilled end where labor is readily substitutable. There may well in a high-tech society be an increasing return to the possession of a college or post-graduate degree; what we can also say however is that those without such a qualification have been cruelly squeezed by the immigration of the last 30 years.

Add to this problem the increased workforce turnover, for which the costs are borne primarily by the workforce while the benefits are felt by company shareholders (or rather, by top management that rewards itself with stock options) and you have a recipe for a society of increasing inequality, with increasing insecurity and immiseration at the bottom end. It’s not a pretty picture.

The economic solution to all this is clear: we need a period of rigorously honest government statistics, high real interest rates, low stock prices, a balanced budget produced by draconian public spending economies, and highly restrictive immigration policies. It would probably take a decade, but at the end of that time we might have a society in which genuine innovation produced genuine economic growth, the benefits of which were spread to all Americans, whose living standards would begin once again the steady increase that they enjoyed in the 1950s and 1960s.

Needless to say, such a solution, which would cause considerable economic pain and immense political pain in the short and medium term, is not on offer. Both political parties favor high immigration – John Kerry and the Democrats are more choosy than the Republicans about what goods they will let into the country, but they seem to have no problem with filling it with more people, most of whom in the nature of things will vote Democrat. Both political parties favor more government spending, which will eventually have to be financed by higher taxes – though George W. Bush’s Republicans use smoke and mirrors to obscure this fact from their supporters.

Bill Clinton’s administration invented hedonic pricing, which puts a spurious glow on economic statistics that is not felt by the populace at large, but George W. Bush’s administration and the Federal Reserve produced the period of rapidly rising government spending and very low interest rates, both of which indulgences will have to be paid for through yet more middle class pain in the years to come.

Now looks a good moment to hop into a time machine and return 50 years to the Eisenhower administration. I’ll take the lack of video games, for the joys of living in a sound economy in which my income has some chance of a steady increase.

-0-

(The Bear’s Lair is a weekly column that is intended to appear each Monday, an appropriately gloomy day of the week. Its rationale is that the proportion of “sell” recommendations put out by Wall Street houses remains far below that of “buy” recommendations. Accordingly, investors have an excess of positive information and very little negative information. The column thus takes the ursine view of life and the market, in the hope that it may be usefully different from what investors see elsewhere.)

This article originally appeared on United Press International.