The conventional, simple analysis of the human impact on the global climate is modeled by the equation I = PAT, where environmental impact is a product of population size, affluence and technological efficiency. To what extent do population size and the intervening population growth rate influence the magnitude of global warming? Will the contribution of population growth be sufficient to induce economic, environmental, or human catastrophe as a result of atmospheric warming? A review of four (somewhat outdated) studies finds disagreement regarding the magnitude of the impact of population growth on global carbon dioxide emissions, but more or less agreement that it is an important and significant contributor to a potential climate crisis. Whether "catastrophe" will result from population-induced greenhouse warming remains up for debate, as the ecological and economic consequences of various climate change projections are highly uncertain.
Overview
Four studies from the 1990s use projections of population growth, fertility
and emissions rates to address the question of the relative contribution
of population growth to future climate change. A range of conclusions are
made regarding the importance of population growth rates and the utility
of addressing fertility as a means to curb greenhouse gas (GHG) emissions
that result from fossil fuel consumption and deforestation. While one study
finds that population growth will account for 50% of global emissions increases
from 1985-2025 (Bongaarts, 1992), a second questions the utility of population
as an indicator of demographic impact, instead suggesting the use of average
household size (MacKellar, et al., 1995). Two additional studies provide
middle ground arguments for the importance of population growth in contributing
to global warming (Birdsall, 1994; Meyerson, 1998).
Population Growth is a Key Factor
A first set of data finds that projected population growth, especially in the least developed countries (LDCs), is a major contributor to the projected growth in anthropogenic emissions of carbon dioxide, accounting for 35% percent of world and 48% of LDC emissions increases between 1985 and 2100 (Bongaarts, 1992: 309). The US EPA's population projections of 9.0 billion LDC and 1.5 MDC inhabitants in 2100 are used in the underlying analysis. Population size (not the rate of growth) is considered as one of five factors contributing to annual CO2 emissions. The other four factors included in the model are GDP per capita, energy intensity of GDP, carbon intensity of energy consumption, and tropical deforestation. A simple arithmetic model is presented to relate these five factors to total emissions: T = P x G x E x C + D (Bongaarts, 1992: 304). Population growth (taken simply as the increase over the 1985 level) and economic development are found to be the most significant contributors to emissions increases by 2100 (Bongaarts, 1992: 307), although the possibility of interaction between population growth and per capita emissions is ignored for purposes of simplicity (309). The model indicates that population is a major player in projected emissions increases and suggests that efforts to reduce population growth may significantly reduce future emissions. However, population momentum will reduce the short term emission reduction benefits of population policy, as some future increase in population is inevitable in the near term (Bongaarts, 1992: 309). The study concludes that the significant influence of population growth on emissions outcomes necessitates the inclusion of the populous and growing LDCs in the formulation and implementation of emissions stabilization policies (Bongaarts, 1992: 312) that make use of population policy (316). However, even if population growth rates were to rapidly fall to zero in the LDCs, their rate of growth of emissions of fossil fuels would decline by less than one third, as compared to baseline population growth scenarios, due to increasing LDC incomes (Bongaarts, 1992: 314). Additionally, even in the impossible scenario of atmospheric GHG concentration stabilization tomorrow, global temperatures would continue to increase for decades to come due to climatic feedback effects (Bongaarts, 1992: 311). Thus, future population growth may be a significant contributor towards potential climatic catastrophe, but the present human population has already set in motion sufficient system disturbance that some further increase in global average temperature is unavoidable. Further, population may be a significant factor in determining GHG emissions, but it is most certainly not the only one.
Rapid TFR Reduction Provides Some Cost-Effective Help
Taking issue with the previous study and finding a less significant role
for population growth in leading to climate change, a second study examines
the effect of rapid versus standard decline in total fertility rates (TFRs)
in the LDCs (Birdsall, 1994). Utilizing the concept of "feasible reductions
in projected rates of population growth" (Birdsall, 1994: 39) in the
LDCs, that study finds that a rapid decline in TFR would lead to emissions
reductions, but that the reductions would not be as substantial as those
claimed by the previous study. This discrepancy is largely attributed to
the differing analytical emphasis on "feasible" TFR reduction
as opposed to complete reduction (growth rate of zero after 1985) (Birdsall,
1994: 39). The baseline TFR path is defined by extrapolating recent TFR
declines of 0.6 every 5 years. "Feasible" rapid TFR decline is
defined as beginning in 1990 and reaching a TFR of 0.2 per year (based on
recent experience in Korea, Thailand, and others) in all the LDCs (Birdsall,
1994: 43). A scenario of rapid TFR decline could lead to a 10% reduction
in fossil fuel-based emissions and a 6% decline in deforestation-driven
emissions from the LDCs by 2030. "Feasible" population growth
rate reductions are thus not indicated to have substantial emissions implications,
however, potential emissions reductions via population control are significant
and their impact will increase over time as population growth rate reductions
accumulate (Birdsall, 1994: 39). While rapid TFR reduction may reduce LDC
emissions by 15% from EPA projections, population momentum and the reality
that per capita emissions are low where TFR is highest are reasons why changes
in fertility trends will not provide larger emissions reduction returns
(Birdsall, 1994: 45-6). That said, MDC investments in LDC family planning
and/or female education are found to be a highly cost-effective means by
which emissions may be reduced. An average cost of $4-11/ ton CO2e for TFR
reduction strategies is compared to a carbon tax (equated with marginal
cost of emissions reductions) of $20/ton necessary to reduce CO2e emissions
by 10%. However, with a steeply increasing marginal cost curve for technological
(tax incentive-based) fixes, the return on family and educational investments
becomes even more attractive as reductions proceed (Birdsall, 1994: 48-9).
Thus, a second study finds that the affect of population growth on climate
change is not a severe as suggested by the first study, but argues that
fertility reduction nevertheless remains a highly cost-effective means to
reduce GHG emissions in the LDCs.
Population Stabilization and Global GHG Policy
A third study examines the role of population growth within the global GHG
emissions policy framework, suggesting that the role of population growth
in determining national emissions is sufficient enough that any effective
global policy must take population growth rates into account (Meyerson,
1998). Specifically, the 1997 Kyoto Protocol is criticized for excluding
the LDC nations from emissions reduction requirements and for not allowing
growing MDCs, like the US, more leeway in meeting Protocol targets (as their
challenge, in per capita terms, is more difficult than that facing other
MDCs having a more stable population size). Given a plateau of global average
per capita emissions in recent years, the increasing correlation between
global population growth rates and realized increases in per capita emissions
becomes an important indicator of the role of population growth in the future
emissions equation (Meyerson, 1998: 117). There has been wide variation
in per capita emissions trends recently, even among the MDCs, due to differing
economic conditions, fuel efficiency gains (resulting from technological
innovation or carbon taxation) and reduced reliance on carbon-intensive
fuel sources (Meyerson, 1998: 118). Thus, the importance of non-population
factors in determining emissions is again stressed. Fundamentally, that
study points out that in order to stabilize emissions levels, as will be
necessary to avoid climate catastrophe, until population is stabilized a
reduction in global average per capita emissions will be necessary (Meyerson,
1998: 126). However, this third study also suggests that population growth
itself is important enough in the emissions equation that any effective
policy must account for population growth, both by including LDCs and by
distinguishing among MDCs by demographics in setting their requirements
(Meyerson, 1998).
A Question of Demographic Measures
Taking issue with the utility of population, a measure of individual-level demographic impact, as opposed to the average number of households, a final study proposes not only the importance of population, but of age structure and social organization trends in determining emissions (MacKellar, et al., 1995). The I=PAT framework is challenged through proposing an alternative I=HAT equation, where the average number of households replaces population size as the demographic unit of account deemed to influence environmental impact. Justification for such a switch lies in the large emissions contribution of household-level goods, such as refrigerators, cars and home heating, which exhibit economies of scale. However, simple population growth and an increase in the average number of households are found to have substantially different consequences when used in demographic-economic impact prediction equations examining energy demand (MacKellar, et al., 1995: 850). Changes in average household size, defined as the total population divided by the number of household heads, result from the evolution of age-specific household headship rates and the age structure of the population (MacKellar, et al., 1995). Data covering average household size from 1950-1990 show that the MDCs have undergone a decrease in H from 3.6 to 2.7 people, largely as a result of population aging in recent years (MacKellar, et al., 1995: 851). This sociocultural shift towards "more atomized living arrangements" in the MDCs is expected to continue (MacKellar, et al., 1995: 853). In contrast, H has increased among LDCs (excluding China) over the same period, due largely to an increasingly youthful population structure (MacKellar, et al., 1995: 851). However, continued increases in the standard of living among LDC populations is expected to reverse this trend, leading to declining average household size worldwide (MacKellar, et al., 1995: 853) and a convergence of average household size among the MDCs and LDCs by 2100 (855). That study concludes that 1970-1990 MDC growth in the number of households accounted for twice the fraction of growth in energy consumption as did simple population increases (MacKellar, et al., 1995: 859). Extrapolating population and household trends to 2100, global emissions according the I=HAT model are predicted to be twice those found using I=PAT. This suggests that not just population trends, but trends in family organization are important in determining emissions and climate change potential (MacKellar, et al., 1995: 862). Thus, while other studies have examined the effect of fertility from the perspective that TFR reductions lead to less population growth, which reduces emissions, they overlook the competing effect that lower fertility may also be associated with smaller average household size (and thus less economies of scale), leading to increased emissions.
Conclusions
While catastrophic affects of climate change are uncertain, population growth rates will play a significant role in determining the extent of anthropogenic global temperature change over the present century. Estimates of the contribution of population to emissions vary, with a high estimate of 35% of future global emissions increases attributable to population growth from a 1985 baseline (Bongaarts, 1992). Technology and per capita consumption, related to income and demographics, will also be important factors in determining total human impact. While the I=PAT equation is a useful starting point, the studies reviewed herein suggest that it is only a rudimentary framework given that population's contribution will depend on interactions among fertility rate, population momentum, income growth and inequality, social choices and inclusion or exclusion from global emissions policies.
Bongaarts, John (1992), "Population Growth and Global Warming" in Population and Development Review, Vol. 18, No. 2, June 1992, pp. 299-319.
Birdsall, Nancy (1994) "Another Look at Population and Global Warming",
in United Nations, Population, Environment and Development: Report of
an Expert Group Meeting on Population, Environment and Development, 20-24
January 1992, United Nations, New York, 1994. pp. 39-54.
MacKellar, F. Landis, W. Lutz, C. Prinz and A. Goujon, (1995) "Population
and Households, and CO2 Emissions", Population and Development Review,
Vol. 21, 1995, pp. 849-865.
Meyerson, Frederick A. B. (1998), "Population, Carbon Emissions, and
Global Warming: The Forgotten Relationship at Kyoto", in Population
and Development Review, Vol. 24, No. 1, March, pp. 115-130.