HE’S NUMBERS ONE

From my perch in the White House’s little Office of Planning and Evaluation, it was a maddening prospect. How could you ever counteract so many bad studies and reports? John Cogan, then assistant budget director, and I talked to the Urban Institute folks and vented our annoyance on them, to little effect. But it seemed to me then that someone in a prominent position outside the administration was needed to call studies like that to public account, and on a regular basis. When bad statistics were reported, they should be pounced on like a bad play or book, and their false premises and ideological motivations unmasked. What was needed was a statistics critic, someone who would cover the world of numbers the way other critics cover drama or stock market trends.

Now, at last, I have a candidate for national statistics critic: Nicholas Eberstadt. A visiting scholar at the American Enterprise Institute and the Center for Population Studies, Eberstadt is author of The Tyranny of Numbers (AEI Press, 303 pages, $ 16.95), one of the best lay guides to the misuse of data since that splendid little 1954 classic by Darrell Huff, How to Lie with Statistics. Eberstadt’s great gift is to show our generation, in readable English, how bad numbers make bad policy and why good numbers must be interpreted in the light of reason as well as science.

“The modern state,” he writes, “is an edifice built on numbers. Modern governments, unlike the diverse governments of earlier times, require statistical information simply to function as modern governments: to perform the tasks now conventionally assigned to and expected of them.” Maybe being able to measure so many kinds of activities improves public decision-making, but “‘quantophrenia” — an idolatry of numbers,” Eberstadt says, represents the potential misuse of statistical tools. “Where unshakable traditional beliefs or passing superstitions played offcial roles in the past,” he writes, “we now witness overconfidence based on a false precision.”

He does not get into the philosophical origins of quantophrenia. But surely it is based in the modern intellectual’s mania for materialism — the constant effort to quantify everything and to deny that there are principles and experience that cannot be quantified. This materialism is a philosophical construct that ought to crumble, and Eberstadt is helping to knock down that corner of the structure where the over-reliance on, and the misuse of, statistics can be found.

The belief that a state can be governed scientifically, through the insights of statistics, he says, is often dangerously wrong. A strictly demographic understanding of man, for example, provided much of the rationale for establishing the procrustean categories of race and nationalism that have caused far more suffering in this century than they have alleviated.

But even in activities where reliance on statistics is justified, great damage can be done when statistical series are incompetently designed, or worse, when politics or ideology misshapes them. If, on one hand, the political system demands numbers as a scientific means of determining truth, and then, on the other, is prepared to ignore questions or data that interfere with the prevailing state ideology, what you get is truly a ” tyranny of numbers.”

Eberstadt introduces this theme early on and then elaborates it in the 11 short essays which comprise his book. Most striking are his examinations of domestic poverty statistics and the national-security implications of demographic trends. What he contributes to both subjects ought to inform the policies of the next national administration that honestly wants to know what is going on.

Take poverty-rate numbers, whose annual release creates a media frenzy. The usual purpose of the media handwringing is to prove that the rising rate of poverty — or, if it is not rising, the “continuing gap between whites and blacks” — indicates government is not spending enough on health care, welfare, and food programs. In the Reagan years, we tried to point out that the offcial poverty rate did not include the value of such welfare; if it did, the rate would be substantially lower. But Eberstadt makes an additional and useful point: “The famous poverty rate may not actually provide a good measure of material deprivation for contemporary America.”

An index of spending by Americans would measure this “material deprivation” and suddenly show a clearer and more positive picture of the “circumstances of our country’s vulnerable groups.” Had such an index been around in recent years,

a rather different policy debate might have been framed in the United States. . . . For while absolute deprivation, in a material sense, has been on the wane, dependence on government largesse among the public at large has been steadily on the rise. By 1990, more than 50 million Americans — more than a fifth of the nation — lived in households that accepted public means- tested benefits. Progress against material deprivation, it would seem, was indeed purchased over that generation — but at the price of the economic independence for large numbers of previously self-reliant Americans.

Eberstadt shows, likewise, that infant mortality figures lead erroneously to conclusions that encourage further government growth. For some decades, the infant mortality rate has been falling, but not as fast in the U. S. as in some other countries and not as fast among blacks as whites. The liberal assumption is generally that the numbers show a failure to deliver enough health care to the poor, especially children. There are other explanations, however, including statistics-gathering inadequacies in some other industrial countries that make us look worse than we are and the very success of the U. S. at bringing low-weight fetuses to term. Infant mortality rates do not measure miscarriages.

But the greatest factor in worsening these numbers might be the attitude of the mothers who typically produce at-risk babies. It is possible that such mothers do not get care because they do not care enough to seek it out. For example, there would seem to be a high correlation between low birth weights and the drinking, smoking and drug-taking habits of the babies’ mothers. This introduces an attitudinal dimension that federal health surveys do not even attempt to explore, leaving a big hole in our understanding of the problem and quite possibly biasing public policy in the wrong directions.

In short, if social de-moralization is an important contributing factor to a wide array of negative social pathologies that do not yield to improved economic conditions, and no statistical series measures such a thing, how can a statistics-driven polity make correct decisions?

De-moralization is also a theme of Eberstadt’s tour of the inadequate statistics that charted — and failed to chart — the decline of the Soviet Union. Throughout the 1980s, the CIA was responsible for keeping track of the economy and demography of the Soviet Union. Eberstadt describes it as ” probably the largest research project ever undertaken in the social sciences.” Unfortunately, the corridors of faceless, but influential, analysts at the CIA often could not match one Murray Feshbach at the Census Bureau (now at Georgetown University).

I remember Feshbach as a rumpled numbers sleuth who could shake a Soviet table of health statistics, hold it against a mirror, compare it to student enrollments in Tashkent and blanket sales in Minsk, and then suddenly make it tell us that mortality rates were rising, not falling, in the USSR and that ethnic Russians were about to become a minority population in the nation they had long dominated. Feshbach wasn’t just a statistician. He knew Russia, Russians, and human nature. He was well liked and admired at the CIA, I’m sure, but the spooks went right on giving top U. S. officeholders an overly generous accounting of Soviet life.

Failure by the CIA to assess the deterioration of the Soviet Union in its last decade may well have contributed to the surprise with which the ultimate Soviet collapse was greeted in the West. (In a foreword to The Tyranny of Numbers, Sen. Daniel Patrick Moynihan reminds us that the record shows that he was not surprised. Neither, let it show as well, was Ronald Reagan.) With pointed understatement typical of his writing, Eberstadt wonders ” whether a better analysis of available data on the Soviet Bloc might not have helped hasten the end of the cold war and contributed to an earlier release of those captive peoples from their bondage.”

The Soviet Union is gone, of course, but the failures of government statistics to capture the reality of international aid (measured for inputs, not outputs, Eberstadt explains) or food production (the raw data often are just not reliable) continue to cloud the minds ofpob icy-makers.

In his final and most brilliant chapter, Eberstadt urges us to take world demographic trends seriously. He says we must demand statistics that measure the real causes of population growth and population shifts, and demand public policies that speak to those causes.

Demographic trends overcome even wars and disasters, and fertility is the most determinative demographic trend of all. Looking at the world today, we can see that the relatively stable, democratic West and Japan are growing very slowly, while less stable countries are expanding fast. This, Eberstadt notes, “may be viewed as a demographic problem, but it may just as accurately be described as a moral and intellectual problem.”

And it is a security problem. Eberstadt cites the examples of Israel, where Arabs may outnumber Jews in a generation or so, and Lebanon, where the long political power-sharing arrangement of Christians and Muslims broke down under the disproportionate population gains of the Muslims and helped precipitate a civil war.

In 1930, he reminds us, “areas of European population” represented about one-third of the world’s number. “At their zenith European powers not only bestrode the earth, they also peopled it,” Eberstadt notes. The largest countries at that time were also the most powerful. By 2025, however, it is expected that all the Western countries together will have fewer people than India alone. And many fewer than China.

Big populations may mean big problems, but they also probably will mean big economies and big armies. “The Western countries’ share of global economic output could be anticipated to decline. . . . One can easily envision a world more unreceptive, and indeed more threatening, to the interests of the United States and its allies than the one we know today.”

Eberstadt recognizes that the “interests” of the United States include preservation of traditions of liberty and democracy, and he acknowledges that statisticians cannot strategize the best ways to meet this threat, or even determine whether or how it is a threat.

But many policy-makers in this country, who are so uncritical in their employment of numbers, are equally careless in seeing to it that the right numbers are collected, and in the right way, and that good sense is used in making use of them. More Nicholas Eberstadts would help. ,

Bruce Chapman, a former director of the U. S. Cencus Bureau and deputy assistant to President Reagan, is president of Discovery Institute, the Seattle-based public policy center.