Monday, September 22, 2014

WSJ debunks Malthusian UN study claiming population bomb by 2100: "Could easily be a dud"

The New Population Boom Could Easily Be a Dud

A high-profile study predicts as many as 12.3 billion people on earth by 2100. Human behavior may not cooperate.

Sept. 22, 2014        THE WALL STREET JOURNAL

The steady slowdown in world population growth in recent decades has dimmed interest in what was once called "the population question." But now concern, even alarm, about global population growth may have been reawakened by a widely publicized new study in the journal Science.

The study warns that the current expert consensus, which envisions the world's population stabilizing over the coming century, is likely wrong. Applying elegant and innovative demographic techniques for long-range population projections, the study concludes: "There is an 80% probability that world population, now 7.2 billion, will increase to between 9.6 and 12.3 billion in 2100."

If global population were to hit the upper boundary of that range, human numbers would leap by more than five billion over the rest of the century—a larger absolute gain than during the entire 20th-century population surge.

The study's 14 co-authors are largely drawn from the U.N. Population Division, a respected demographic unit separate from U.N. policy-making authorities, and co-author John Wilmoth, the division's chief, is renowned in his field. With that pedigree, the study will be taken seriously in scientific and policy circles. And it is all but certain to reignite Malthusian debates about the race between mouths and food, and to re-energize the international population-planning activists who castigate governments and aid donors for their complacency about the global demographic threat.

No doubt the study will also color the climate-change debate—after all, billions more consumers will mean even greater greenhouse-gas emissions. Count on the Science study to figure prominently at this week's climate-change summit during the U.N. General Assembly.

Yet some skepticism is in order. In technical terms, the innovation that made this study worthy of publication in a flagship scientific journal is its use of probabilisticlong-term population projections. Conventional population projections—including those by the U.N. Population Division—typically offer high-, medium-, or low-variant scenarios. But these are only meant to illustrate plausible future paths, without making claims about the likelihood of ultimate outcomes. By contrast, the new projection (using Bayesian probability, for the 18th-century mathematician Thomas Bayes ), assigns odds to future scenarios, based on past and current experience.

In the Science study, the main reason for concluding that the world's population will most likely continue to rise is sub-Saharan Africa. At an estimated five-plus births per woman, average fertility today is well over twice as high for sub-Saharan Africa as for the rest of the world. Between 1950 and the present, the region's numbers more than quintupled, jumping by almost 800 million (to nearly one billion) and accounting for about a fifth of the overall growth in human numbers. In the new probabilistic projections, sub-Sahara's population would account for practically all world-wide population growth for the rest of the century—and could end up with four billion or five billion-plus people by 2100.

The key question: Does this probabilistic approach increase the accuracy of our long-term perspective on population? The short answer: no. The basic trouble with all long-range population projections is that they are driven by assumptions about birth levels—and there is still no reliable method for predicting fertility levels a generation from now, to say nothing of a century hence. Demographers are even hard-pressed to explain historical fertility patterns.

Half a century ago, in the early 1960s, East Asia's overall fertility level was about 5.5 births per woman; today, according to the U.N. Population Division, it is about 1.6 per woman—70% lower. In scope, scale and speed, nothing like this decline had happened in human history. Bayesian projections would never have regarded that outcome as likely, just as they would have most likely missed the startling 70% drop in fertility in Iran between the early 1980s and early 2000s.

Unlike contemporary fruit flies or red deer or any other species, modern homo sapiens exhibit demographic rhythms that are new and fundamentally unfamiliar from our past experience. Around the world we see record high life expectancies, record low birth rates, and record speeds in improving health conditions or effecting family change. Under such circumstances, modeling the demographic future is more difficult than ever.

Fertility decline in sub-Saharan Africa has to date been more halting and tentative than in any other major region in the world. Using that past performance as the basis for projecting future fertility trends shouldn't inspire confidence. Currently sub-Saharan fertility levels are twice as high as what would be needed for long-term population stability, and the authors of the Science study assume that the region will remain a demographic exception for generations. Maybe that will prove correct—but maybe it won't.

Some variables to consider: In 2000 a third of Africa's women of childbearing age (15-49) had a high-school education or better; researchers at the International Institute for Applied Systems Analysis in Vienna anticipate that by 2050 the figure will be 70%. Also around 2050, the U.N. Population Division projects that life expectancy at birth will average nearly 70 for the sub-Saharan region, up almost a decade and a half from today.

If such changes come to pass—changes that have corresponded with lower birth rates elsewhere in the world—would Africa still remain resistant to fertility decline? Some would reply that cultural tradition and related factors will continue to support high fertility rates, and those voices may ultimately be right. But that same argument was made about the greater Middle East not so long ago—and we now know how mistaken the assumption of unchanging "family values" was.

Global fertility is a matter of human volition, and no computational breakthrough can alter this fundamental fact. The ability to make reasonably precise guesses about the fertility patterns of the unborn, not to mention their children and grandchildren, is one we manifestly lack.

Mr. Eberstadt is a resident scholar at the American Enterprise Institute.

No comments:

Post a Comment