Astronomical Incomes

When my son was in the sixth grade, his science project was a scale model of the solar system. His "sun" was a basketball in his bedroom. Mercury, Earth and Venus were in our house and front yard, and Mars was in his friend's house across the street.

The more interesting pieces of the model were the outer planets. Jupiter, Saturn, Uranus and Neptune -- represented by tennis balls, ping-pong balls, etc. -- were placed at increasingly greater distances around our neighborhood and beyond, while Pluto was a quarter-inch ball of clay on the university campus a mile from our house.

One purpose of the exercise was to illustrate for students what statisticians call a "skewed distribution," in which numbers pile up at one end of a scale and stretch out in a long "tail" at the other. A picture or model illustrates "skewness" much better than a list of numbers does.

Household income in the United States has a skewed distribution. The imbalance is big, too -- so big that statements like, "The benefits of the Bush tax cuts will go mainly to households in the top 1 percent tax bracket," tend not to tell the whole story.

I'm proposing that the Internal Revenue Service, with the help of the National Park Service, construct a scale model of household income in America (see above). Let's imagine that they start by designing a big bronze plaque that reads, "U.S. median income, 2002" and place it near the center of the country -- say, in Salina, Kansas, where I live. (We're an obvious choice, being only about a 90-minute drive from the geographical center of the continental United States.)

The median income was about $43,000 last year. Half of American households received less than that amount, and half received more. Suppose the IRS were to locate the median-income plaque at one of Salina's exits on Interstate 70 and then place plaques symbolizing other incomes along I-70 to the east and west, using a scale of, say, $1,000 of income per mile of road.

The plaque showing the federal poverty level for a family of four in 2002 -- about $18,000 -- would be located about 25 miles west of Salina, at the exit for Ellsworth, Kansas. Eighteen miles beyond, in the town of Wilson, would be the spot for the "zero income" plaque (assuming the good people of Wilson would accept that dubious distinction).

That's it for the incomes of half of all U.S. households. They would fit within a 43-mile stretch, across only two Kansas counties. (See this map to find Salina, Wilson and the other landmarks in the model.)

The top half of the income scale would be illustrated by plaques to the east of Salina. About 150 miles from the zero point, in the western suburbs of Topeka, the capital city of Kansas, would be the spot for a plaque reading, "Income: 95th percentile," because 95 percent of households take in less than $150,000 a year. Ninety-nine percent of households make less than $374,000, so the plaque reading "Income: 99th percentile" would go in Williamsburg, Missouri, about 75 miles west of St. Louis.

That takes care of 99 percent of us. But it's in the top 1 percent of incomes -- those east of Missouri -- where we see what "a skewed distribution" really means. The plaque for "Average income of the top 1%" ($1,082,000) would lie a few miles southeast of Pittsburgh -- a 20-hour drive on I-70 from median-income Salina. But averages don't mean much when it comes to the top 1%.

For example, one household that belongs to the top 1%, the George W. and Laura Bush family, reported a paltry $856,000 in 2002 income, which puts their plaque at an exit ramp a little east of Columbus, Ohio. Meanwhile, the average income of major-corporation CEOs in 2002 was $7.4 million. That's an amount that falls well beyond Baltimore, where Interstate 70 ends. To keep to scale, the IRS would have to put that plaque in ... well ... they'd have to put it in Kabul, Afghanistan. At this point, they might have to bring the Defense Department in on the project.