With the presidential election little more than a year away, the worst statistics are already being viewed as the best political weapons.
A front-page "news" report by The Washington Post's writer Ceci Connolly hoped to raise undue alarm because "the number of Americans who lack health insurance climbed by 5.7 percent in 2002, to 43.6 million." Yet the author could not resist partisan editorializing about a mélange of other phony facts: "Coupled with a report last week showing a similar rise in poverty," she wrote, "the health insurance data help illuminate the human toll of the nation's stalled economy. Since President Bush took office, the United States has lost 2.7 million jobs and household incomes have fallen for three years in a row. ... Democrats have seized on economic issues in the quest to defeat Bush in next year's presidential election."
Well, at least the last sentence was true. But within just three or four sentences, innocent readers were told a half-dozen statistical lies. This may be a new record.
Consider first the headline estimate that 43.6 million are said to be without health insurance. The trouble with such estimates -- as I first explained in a June 22, 1992, Forbes column -- is that the March Current Population Survey (CPS) shows us only an instant snapshot of what is actually a moving picture. People often have short spells without health insurance, just as they have short spells being statistically poor or unemployed. Many of those without health insurance in November were different people than those without insurance when the March survey was taken.
To fix such problems, a superior Survey of Income and Program Participation (SIPP) is less frequent but more intense. That survey is conducted four times a year but not every year. In 1998, the latest available, the SIPP survey found only 21 million without health insurance for the entire year -- half the number in the headlines. The SIPP survey confirmed, however, the CPS finding that about 40 million were indeed without insurance for part of the year. In May, a widely ignored Congressional Budget Office report noted that "between half and two-thirds of the people who experienced a period of time without insurance... had coverage for other portions of that year." Although reporters and politicians will keep shouting that more than 40 million people are continuously without health insurance, that is just another Big Lie.
One reason many young people lack health insurance is that they rightly expect paying their own medical bills will be cheaper in most years than paying insurance premiums. One reason a disproportionate number of Hispanics lack health insurance is that many are here illegally, and therefore avoiding a conspicuous paper trail. Ms. Connolly, however, tried to make it appear as though the only conceivable reason people don't buy even sensible ("catastrophic") health policies is the average American family is broke.
She contrasted someone's $9,000 estimate of the cost of a so-called "average" family health plan with "the average wage in this country being only $27,000." But the cost of any insurance, even home and auto, varies with the size of the deductible. And typical family income comes from one or two salaries, not from any "average wage." Median income among intact families last year was $61,254.
Median income among "households" is lower and more cyclical because it includes young singles with frequent, short spells of unemployment. Median household income always falls in recessions, not because "President Bush took office." The 3.4 percent drop in median household income from 1999-2002 was less than the 4.7 percent to 5.7 percent declines during the previous three recessions.
But the relatively minor drop in median household income over the last few years is not at all the same as saying total incomes are falling this year. Over the past 12 months, after-tax personal income rose 5.7 percent, thanks in part to Bush tax cuts.
Ms. Connolly's allusion to a "similar rise in poverty" is at least as misleading as the number of uninsured, partly for similar reasons. The new report on poverty uses SIPP data to show that more than half of those counted as poor over a four-year period had spells of poverty that lasted only two to four months. Only 2 percent remained in poverty for four years. Although this writer for The Washington Post went out of her way to assign political significance of the poverty rate rising to 12.1 percent last year, she neglected to notice last year's poverty rate was nonetheless lower than in any year from 1980 through 1998. The poverty rate was 15.1 percent in 1993 and 12.7 percent as recently as 1998. Why is a lower poverty rate now so damning simply because George Bush is in office?
What about that remark about "the nation's stalled economy"? The economy grew at 3.2 percent in the second and third quarters -- slightly quicker than the average growth rate during President Clinton's first three years in office (1993-95). Why is a similar pace of expansion now described as "stalled"?
And what about that endlessly repeated chant that "the United States has lost 2.7 million jobs." The Bureau of Labor statistics asks people if they are employed every month, and their answer has been they have 1.2 million more jobs than they had at the start of this year. They ought to know.
Once upon a time, when true editors shunned false facts, we used to rely on obese newspapers like the New York Times and The Washington Post for accurate information. Until next year's election is over, however, it looks as though we will have to be even warier than usual when it comes to reading front-page economic illiteracy disguised as factual news.