The Pulse of Capitalism

Issue 99-2, July 1999

Income Inequality

In August 1998, the Federal Reserve Bank of Kansas City sponsored a symposium on "Income Inequality: Issues and Policy Options". The speakers came from a number of countries and from academic, business, and governmental backgrounds.

Two speakers summarized the evidence of rising income inequality in advanced economies. A.B. Atkinson, Oxford University, noted that the United Kingdom, the United States, and some other OECD countries have experienced rising income disparity since the 1970s, led by the U.K. Poverty rates, defined as the proportion of households with income levels below half the national average, have also increased in some OECD countries including the U.S. and the U.K. Atkinson also stressed that factors other than individual labor earnings are important in determining the distribution of disposable household income. These include income from capital, private and public transfers, taxation, and household composition.

Lawrence Katz, Harvard University, pointed out that, "based on the Luxembourg Income Study, 10 out of 13 OECD countries have experienced some increase in family income inequality since 1979. Indeed, for most of these countries, these increases represent a break from patterns of sustained reductions in inequality over most of the twentieth century. Katz also made the point that when one expands the measure of economic well-being to consider both the distributions of income and employment opportunities, a common pattern of rising "economic inequity" becomes apparent for most industrialized countries.

Causes

"The conventional analysis begins with the observation that, over the past two decades, the earnings of skilled individuals have risen relative to the earnings of unskilled individuals despite the fact that the supply of skilled labor has risen relative to the supply of unskilled labor. The inference is that the demand for skilled labor has increased faster than its supply. Three possible sources commonly suggested for such a relative demand shift are globalization, deindustrialization, and skill-based technological change. This model has come to be generally accepted...." [Quotations are from Economic Review, FRB Kansas City, Fourth Quarter 1998.]

Definition Problems

From the foregoing, it is apparent that measuring income inequality presents problems. Most measures of income focus on either family income or household income. These measures are influenced not just by money income but by their composition. Many more families today include two or more wage earners than was the case a few decades ago. Likewise, households today include more individuals than was true in the past. Comparisons based on these units are, therefore, difficult, over time and across countries.

Labor force participation rates are another complicating factor in measuring income inequality. In 1970, 23% of females participated in the labor force, whereas, by 1998 this ratio had risen to 59.8%. The corresponding rates for men were 79.7% and 74.9%. As we shall see later, average female earnings are considerably lower than average male earnings, but the median for females has risen considerably whereas for men it has risen only slightly. Therefore, average earnings for all workers give us an incomplete picture of trends in income inequality.

The median income for families in 1996 was $42,300 whereas for households it was $35,492. Likewise, the number of family units was 70.2 million compared with 101.0 million households. Nevertheless, the underlying trend is the same using either unit of measurement.

Family Income

The growth of income inequality among families shows up sharply in this table. In reviewing the changes it should be noted that one percentage point of personal income in 1996 equaled $64.3 billion.

The striking development revealed by this chart is that every fifth of families lost income share from 1970 to 1996 except the top fifth. The cumulative change was 5.9 percentage points or $379.4 billion in constant dollars. The top twenty percent of families were receiving close to half of aggregate income while the top 5 percent were receiving a fifth.

How much of this shift was due to changes in income? Lawrence Katz found "that increases in the wages of husbands and wives in the top quintile relative to those in the bottom quintile can explain 75 percent of a 36 percent increase in the mean income of the top quintile relative to the bottom quintile of U.S. married couple families....from 1979 to 1996."

Income of Persons

A more complex picture emerges if we examine the money income of persons in constant dollars and separated into male and female categories. Among males, from 1970 to 1995 the percent of individuals in the lower and middle income ranges has shrunk, while the percent in the upper range has increased. Among females the percentage of individuals in the lower income ranges has shrunk drastically while the percent in both middle and upper ranges has grown. It should also be noted that the median for males increased only $1,175 or 5.2 percent whereas the median for females rose $5,216 or 68.6 percent. But even with this gain the female median was still only a little more than half the male median. These changes may be summarized as follows:

One of the more interesting developments for males is the decline in the percentage of individuals in the $25,000-$50,000 range. This group can logically be considered the traditional "middle class". If we assume that most of the decrease in the "lower" income group represented a move into the "middle" group, then the overall 7.8 percentage point increase in the "affluent" classes came, in a sense, at the expense of the middle class. Among females, on the other hand, the largest single change was a movement from lower wage groups into the "middle" range - an increase in the "middle" range of 11.3 percentage points.

These changes have decidedly had major effects on American society, including a weakened family structure and greater concentration of economic power. The long-run consequences are impossible to assess.

Occupational Disparities

There is a popular impression that the U.S. has been trading well-paying blue collar steelworker jobs for fast-food "hamburger" jobs. In fact, job growth in goods production - agriculture etc, mining, construction, manufacturing - has been stagnant since 1970; "services producing" categories have accounted for all of the growth. But the categories shown in the table are broad, with considerable variation in each. For example, "Transportation and public utilities" includes "Pipeline except natural gas" at a 1997 wage of $60,500 and "Local and interurban passenger transit" at $21,038. Averages by category include and conceal these variations.

As shown in the chart, the categories with the largest gains from 1970 to 1997 and those with the smallest gains are as follows:

It will be noted that "goods production" was represented by one of the largest gains categories as well as one of the smallest gains categories. Mining has become highly concentrated and mechanized, requiring both skill and responsibility, whereas construction is decentralized and requires lower skill levels. These differences make for very different outcomes.

Not surprisingly, FIRE includes "Security and commodity brokers" at an annual 1997 income of $120,349, and "Holding and other investment offices" at $75,615. Finance today has become the "dominant revenue" and therefore commands the highest returns. "Services" includes many low-paying industries, but it also benefits from the inclusion of "Legal services" at $57,019. Retail trade has been a fast growing category but has made the lowest gains in the table. As noted "Transportation" includes the low income "Local passenger transit" industry.

Consequences

At the symposium sponsored by FRB-Kansas City, the economic consequences of rising income inequality were discussed by Joseph Stiglitz, the World Bank, and Jason Furmen, Harvard University. They concluded that "Because low-income individuals bear a disproportionate share of the burden of cyclical fluctuations, both theory and policy must be sensitive to the fact that there are social tradeoffs in macro economic management. Moreover, lower levels of inequality may enhance economic growth and reduce economic instability. Thus, the potential benefits of reducing inequality merit active government policies. These policies may include macro economic measures that reduce unemployment as well as redistributive measures that support education for disadvantaged groups." Some other speakers disagreed with these conclusions.

By focusing on economic consequences, the symposium avoided raising what may be far more important consequences of rising inequality. One of the most important of these consequences is the impact of rising affluence on the consumption of resources. The higher incomes go, the more resources are consumed. It is affluent societies that are consuming the world's energy reserves at a shameful rate, destroying world fisheries, cutting down virgin forests and wild life habitats, and polluting land, water, and air.

We must go beyond economics and ask the simple question of how much is enough? Corporate America has already decided that there can be no "enough". Through competitive upmanship, annual incomes of top executives have been boosted to over $3 million plus even more in long-term plans. If these individuals invest just $2 million of their annual income in safe investments at 6 percent, they and their descendants will receive an effortless annual income of $120,000-equivalent to the highest average income of any industry in the U.S. Every succeeding year at the same level would add another $120,000 per year. Is that level "enough", or is there indeed no level that is enough. At some point, America must ask and deal with such questions.

GDP growth at annual rates was lower in the first quarter for all countries except Japan, where it became positive after two negative years. Industrial production slowed slightly except in Canada and the U.S., while retail sales volume picked up a bit in all five economies.

Consumer prices fell in Germany and Japan and rose moderately in the other three countries. Unemployment rates also fell moderately except in Japan where the rate is at a postwar high.

Interest rates fell sharply in Britain and moderately in Canada, Japan and the U.S. The U.S. stock index rose 12.7 percent in the first quarter over one year earlier; the British index rose 5.6 percent; the Canadian index fell 13.0 percent; the German index fell 4.8 percent, and the Japanese index fell 4.0 percent. These divergences show how volatile the world economy now is.

Like the stock market, the U.S. current account deficit seems to grow ever larger, while the Japanese surplus again exceeded $100 Bn. In foreign exchange markets, the yen has been stable against the U.S. dollar while the Euro has declined steadily. The U.S. major currency index has also been stable.

On a quarterly basis, real GDP increased 4.1 percent in the first quarter of 1999 compared with a 6.0 percent increase in the fourth quarter of 1998. The deceleration was due largely to a decline in exports, which had been strong in the fourth quarter.

Industrial production rose at about the same rate as in 1998, led by output of business equipment, but capacity utilization fell slightly. Manufacturers' new orders kept pace with the expansion, and new construction expenditures also kept pace, although the contracts index has fallen since January.

Real gross investment, one of the pillars of current expansion, rose strongly again in the first quarter with increases in both nonresidential and residential fixed investment. Business and retail sales maintained their uptrends and inventories also continued to rise. Underlying these trends was a further $271 increase in per capita consumption expenditures. Strong employment growth in services contributed to this increase, but employment in goods producing industries has not risen since December.

National income rose $125 billion in the first quarter, the largest quarterly gain since first quarter 1997. Compensation of employees and corporate profits registered the largest gains. Real per capita income increased 3.3 percent, slightly more than the high rate of the fourth quarter. Average weekly earnings rose $2.92.

The increase in gross saving was due to increased government saving, as gross private saving fell; personal saving, for the first time, was negative.

Commodity prices, on March 30, were only 83.6 percent of their level in 1990, reflecting weak worldwide demand; metals were the weakest sector at an index of 63.2. After two years with little change, producer prices rose in April and May to an index of 132.4. Nondurables led the upturn while capital equipment prices were unchanged. Corporate profits rose $27.3 bn from the preceding quarter, the largest increase since the first quarter of 1997.

After dipping below 5 percent, the ten year Treasury rate rose above that level in March to 5.89 percent on June 12. The M-3 money supply growth decelerated sharply, reflecting a cutback in commercial bank credit growth in securities and in loans. Consumer credit, however, rose almost half as much in the first quarter as in all of 1998, helping to explain the negative personal savings rate noted earlier. Credit market debt expanded at the same high rate as in 1998.

The negative trade balance on goods soared over $10 billion in the first quarter as real exports fell and real imports continued to rise. The balance on services was little changed.

Capital flows into and out of the U.S. appear to have slowed drastically during the first quarter. Details on this change are not available as this is written.

Acquisition of Treasury securities by foreigners again slowed, but purchases of other U.S. securities continued at the high rate of the past two years.