WORLD POPULATION stands at 5.3 billion, with an additional billion due to arrive every 11 or 12 years. What ultimate size can be expected? When will the growth stop -- if it ever does? Following the population-control progress of the 1970s, these questions lost much of their urgency. But recent developments suggest that the optimism was premature, and the '90s may be decisive in determining the final size of future world population.

Even in the most optimistic scenarios, the planet will gain a record 95 million more people each year during this decade. The conventional wisdom is that the number will peak at 10 to 11 billion in the latter half of the next century. (Even then, we will be looking at an India of 1.8 billion -- a billion more than today. Nigeria will have 500 million, five times its present size and twice the current U.S. population.) This expected stabilization level derives from frequently cited projections by the United Nations and World Bank.

But projections are only as good as the assumptions on which they are based. And as we approach World Population Day on Wednesday (designated by the U.N. on the anniversary of the day that global population first reached 5 billion -- July 11, 1987), many of our familiar assumptions have been cast into doubt.

One is that all countries will see their fertility someday arrive at the replacement level. "Replacement-level" fertility is achieved when each woman in a developed country has two children during her lifetime. At that rate, couples simply "replace" themselves without increasing the size of successive generations. Mathematically, replacement-level fertility is actually about 2.1 children per woman since not all women survive to or through their childbearing years.

The current U.N. projections assume that world population will reach replacement somewhere between 2010 and 2065, with totals ranging from 7.5 to 14.2 billion, with the middle series at 10.2. The spread between low and high prognoses is enormous: South Asia would plateau at anywhere from 2.7 to 4.2 billion; Africa from 1.4 to 4.4 billion. Two recent trends, however, suggest that the low figures are impossibly low and even the middle may not be high enough.

Deferred-Birth Syndrome One trend is a surprisingly sharp increase in the fertility rates of some industrialized countries -- in many of which only recently the paramount demographic worry was population decline caused by very low birth rates.

Of course, fertility is still quite low in some devel- oped countries and dropping in others. (Italy now has the world's lowest birth rate at 1.3 children per woman.) But in Sweden, the total fertility rate or TFR -- the average number of children a woman will have during her lifetime if the age-specific birth rates of a given year were to remain unchanged -- has risen to 2.0 after sinking to 1.6 in the early 1980s. West Germany, which set a world record in 1985 with a very low TFR of 1.28, is back up to 1.42. In the United States, the rate has risen from a low of 1.7 in 1976 to 2.0 today -- and is still climbing. Are these increases prophetic of the future of less developed countries?

These "mini baby booms" are probably due to delayed childbearing in the latter half of the 1970s and early '80s, giving a false impression of extremely low lifetime fertility. Now that women are beginning to have the children whose birth they deferred, fertility is rising.

Do these seemingly tiny shifts in fertility really make much difference? Indeed they do. Consider the effect on the United States: The Census Bureau publishes an excellent set of projections offering alternative population scenarios based on different fertility rates. The middle (and most commonly used) projection foresees a total population of 292 million in 2080. Under that scenario, the population would peak in size in 2038 and would be slowly declining by 2080. This assumes a below-replacement TFR of 1.8 -- a reasonable assumption when the projection was prepared in 1988. Suppose, however, that we substitute a TFR of 2.2, as the bureau does in its "High Fertility" series. The result is a U.S. population of 421 million in the year 2080: 129 million higher than the middle series -- and still growing at 20 million per decade.

The higher (2.2) TFR could easily occur in contemporary U.S. society, particularly if couples had more access to child-care. Yet just as a ship that veers one degree off course will miss its destination by hundreds of miles, this small shift in TFR could result in a significant population difference. Reversing the Tide A second and more troubling trend is the retreat of many developing nations from population progress of the past decade.

In the 1960s, the specter of never-ending growth was quite real. The world rate hit an all time high -- 2 percent per annum, enough to double the total every 35 years -- as that decade ended, with little prospect of a slowdown. A 1963 cover story in U.S. News & World Report cited projections that population would rise from about 3 billion to 22 billion in 2050 and keep growing. Whereas famine and disease had once kept growth in check, better food distribution and medical services were producing a population boom, with the fastest growth rates in the poorer nations. Women in developing countries were averaging about six children each during their lifetime.

Then in the '70s, things began to change. Family planning programs in some countries began to receive greater acceptance, with notable successes in Indonesia, Costa Rica, South Korea and Taiwan. By the end of the decade, women in Third World countries were averaging about 4.5 children each. (China's population-control program was a major contributor: Fertility dropped from a late-'60s level of six children per woman to about 2.3 by 1980.) There was a flurry of speculation that the population explosion had been defused, and that optimism has persisted -- dangerously -- to this day.

True, during the '80s several countries continued dramatic birth-rate declines -- Thailand and South Korea are notable examples. Others were not so successful: In India, Tunisia, the Philippines, Egypt and Indonesia the momentum seemed to drain from family-planning programs; in China, the birth rate actually rose again. Perhaps, demographers speculated, it is comparatively simple to bring fertility down to middle levels -- say, from six to four children per woman -- but much harder to reduce it to a two-child family.

High fertility is normally seen as a response to the perceived need for a large number of children to ensure old-age support for parents, to provide farm labor or to offset high infant mortality. It is hypothesized that as nations industrialize and urbanize, and mortality declines, preferred family size falls. This pattern is usually referred to as the "theory of demographic transition." While such theories often oversimplify the situation, the demographic transition describes the historical experience of developed countries fairly well.

The demographic transition took over 100 years in the developed countries. Today's developing countries do not have that time luxury, and family planning has become a critical factor in the transition process.

A primary goal of population policies today is to brake population growth, thereby buying breathing room for countries to deal with development issues such as rapid urbanization, environmental problems and demands on the educational system. The goal is not to cause population decline, but simply to slow growth so that developing countries will be able to make a "soft landing."

Those possibilities, along with larger concerns for a stable world population, are now engaging the interest of many developing countries. Even in sub-Saharan Africa, 29 out of 47 governments now feel that their birth rates are too high. This represents quite a change from even 10 years ago. But results often lag years behind policy declarations. That is because population growth does not exhibit the characteristics of an explosion as much as it does a glacier. Relatively large numbers of women of childbearing age in a developing country population ensure that growth will continue long after birth rates fall, and even after they reach replacement.

The challenge of the '90s will be to understand these forces. Recently Bangladesh's fertility fell to about five children per woman, suggesting that family planning can work even in the face of poverty and overcrowding. But scores of uncertainties remain. Will India's family planning program regain its momentum? When will Pakistan's and Nigeria's fertility decline begin? Will Iran continue its policy of high birth rates? The answers to these questions will dramatically affect the quality of life on Earth.

Demographer Carl Haub is director of information and education at the Population Reference Bureau Inc.