V. ASSUMPTIONS AND METHODS UNDERLYINGACTUARIAL ESTIMATES
The future income and cost of the OASDI program will depend on many demographic, economic, and program-specific factors. Trust fund income will depend on how these factors affect the size and composition of the working population and the level and distribution of earnings. Similarly, program cost will depend on how these factors affect the size and composition of the
beneficiary population and the general level of benefits.
Basic assumptions are developed for several of these factors based on analysis of historical trends and conditions, and on expected future conditions. These factors include fertility, mortality, immigration, marriage, divorce, productivity,
inflation, average earnings, unemployment, retirement, and disability
incidence and
termination. Other factors are projected using methods that reflect historical and expected future relationships to the basic assumptions. These include total population, life expectancy, labor force,
gross domestic product, interest rates, and many program-specific factors. It should be noted that all factors included in any consistent set of assumptions are interrelated directly or indirectly. It is also important to note that these interrelationships can and do change over time.
The assumptions and methods used in this report are reexamined each year in light of recent experience and new information about future conditions, and are revised if warranted.
Because projections of these factors and their interrelationships are inherently uncertain, a range of estimates is shown in this report on the basis of three sets of assumptions, designated as
intermediate (alternative II),
low-cost (alternative I), and
high-cost (alternative III). The intermediate set represents the Board’s best estimate of the future course of the population and the economy. In terms of the net effect on the status of the OASDI program, the low-cost is more optimistic, and the high-cost is more pessimistic. The low- and high-cost sets of assumptions reflect significant potential changes in the interrelationship among factors, as well as changes in the values for individual factors. The probability is very low that all the factors and interactions would differ in the same direction from those expected for long periods of time. Outcomes with overall long-range cost as low as (or lower than) the low-cost scenario or as high as (or higher than) the high-cost scenario also have a very low probability. This report also includes a stochastic projection that provides a probability distribution of possible future outcomes that is centered around the intermediate assumptions. These are discussed in Appendix
E.
Although these three sets of demographic and economic assumptions have been developed to provide a broad range of possible outcomes, the resulting estimates should be interpreted with care. The estimates are not intended to be specific predictions of the future financial status of the OASDI program, but rather, they are intended to be indicators of the expected trend and a reasonable range of future income and cost, under a variety of plausible demographic and economic conditions.
The values for each of the demographic, economic, and program-specific factors are assumed to move from recently experienced levels or trends, toward long-range ultimate values, generally over the next 25 years. Ultimate values or trends reached by the end of the 75‑year long-range period are generally maintained at these levels or trends for extrapolations beyond 75 years. One exception is for real-wage growth, as described in section
IV.B.5.
The assumed ultimate values, which are reached within the first 25 years (and apply thereafter through the end of the 75-year long-range period) for both the demographic and the economic factors, are intended to represent average annual experience or growth rates. Actual future values will exhibit fluctuations or cyclical patterns, as in the past.
The following sections discuss, in abbreviated form, the various assumptions and methods required to make the estimates of trust fund financial status, which are the heart of this report.
1 There are, of course, many interrelationships among these factors that make a sequential presentation potentially misleading. Nevertheless, the following sections roughly follow the order used in building the trust fund estimates presented in chapter
IV.
The principal demographic assumptions relating to fertility, mortality, and net immigration for the three alternatives are shown in table
V.A1. The rationales for selecting these assumptions are discussed in the following three sections.
Fertility assumptions are developed for women in the form of birth rates by single year of age, from 14 to 49. They are applied to the total number of women in the population at each age, for all marital statuses.
Historically, birth rates in the United States have fluctuated widely. The total fertility rate
2 decreased from 3.31 children per woman at the end of World War I (1918) to 2.15 during the Great Depression (1936). After 1936, the total fertility rate rose to 3.68 in 1957 and then fell to 1.74 by 1976. After 1976, the total fertility rate began to rise again, reaching a level of 2.07 for 1990. In the 1990s, the total fertility rate remained fairly stable, around 2.00 children per woman. Since 2000, the total fertility rate has been consistently above 2.00, and was 2.06 in 2005.
These variations in the total fertility rate resulted from changes in many factors, including social attitudes, economic conditions, and the use of birth-control methods. Future total fertility rates may be expected to remain close to recent levels. Certain population characteristics, such as the higher percentages of women who have never married, of women who are divorced, and of young women who are in the labor force, are consistent with continued lower total fertility rates than experienced during the baby-boom era (1946-65). Based on consideration of these factors, ultimate total fertility rates of 2.30, 2.00 and 1.70 children per woman are assumed for the low-cost, intermediate, and high-cost assumptions, respectively. These assumptions are unchanged from those used in last year’s report.
Based on preliminary data for 2006 and 2007, the total fertility rate is assumed to reach a level of 2.10 children per woman for both 2006 and 2007 and 2.08 for 2008. These levels are slightly higher than those estimated in last year’s report for the intermediate assumptions. For all three alternatives, the total fertility rate is then assumed to follow a gradual trend toward the selected ultimate level, which is reached in 2033.
For the projections in this year’s report, assumed average percentage reductions in future mortality rates were developed by age-group, sex, and cause of death. These assumptions were then used to estimate future central death rates by age-group, sex, and cause of death. From these estimated central death rates, resulting probabilities of death by single year of age and sex were calculated.
Historical death rates (for years 1900-2005) used in developing estimates for this report were calculated for ages below 65 (and for all ages for years prior to 1968) using data from the National Center for Health Statistics (NCHS).
3 For ages 65 and over,
final Medicare data on deaths and enrollments were used for years 1968 through 2005. Death rates by cause of death at all ages, for years 1979-2005, were produced by the NCHS.
The total age-sex-adjusted death rate4 declined at an average rate5 of 1.07 percent per year between 1900 and 2005. Between 1979 and 2005, the period for which death rates were analyzed by cause, the total age-sex-adjusted death rate (for all causes combined) declined at an average rate of 0.82 percent per year.
Death rates have declined substantially in the U.S. since 1900, with rapid declines over some periods and slow or no improvement over the other periods. Historical death rates generally declined more slowly for older ages than for the rest of the population. The age-sex-adjusted death rate for ages 65 and over declined at an average rate of 0.75 percent per year between 1900 and 2005. Between 1982 and 2005 the age-sex-adjusted death rate for these ages declined at an average annual rate of 0.57 percent.
Reductions in death rates resulted from many factors, including increased medical knowledge and availability of health-care services, and improvements in sanitation and nutrition. Based on consideration of the expected rate of future progress in these and other areas, three alternative sets of ultimate annual percentage reductions in central death rates by age group, sex, and cause of death are assumed for 2033 and later. The intermediate set, which is used for alternative II, is considered to be the most likely to occur. The average annual percentage reductions used for alternative I are generally smaller than those for alternative II, while those used for alternative III are generally larger. These three sets of ultimate annual percentage reductions differ slightly from those used in last year’s report. Overall, age-sex-adjusted death rates are projected to decline slightly faster than in last year’s report. The differences reflect an assessment of long-term trends based on mortality data from 1900 through 2005. The most significant changes in the ultimate rates of reduction in death rates are increases in these rates for heart disease and cancer between ages 65-84. For these ages and these causes of death, declines in death rates have consistently exceeded expectations in recent years.
After 2005, the reductions in central death rates for alternative II are assumed to change rapidly from the average annual reductions by age group, sex, and cause of death observed between 1985 and 2005, to the ultimate annual percentage reductions by age group, sex, and cause of death assumed for 2033 and later. The reductions in death rates under alternatives I and III are also assumed to change rapidly to their ultimate levels, but start from levels which are, respectively, 50 or 150 percent of the average annual reductions observed between 1985 and 2005.
Projections of age-sex-adjusted death rates are presented in table V.A1 for the total population (all ages), for under age 65, and for ages 65 and over. Under the intermediate assumptions, projected age-sex-adjusted death rates for the total population are lower than the death rates in last year’s report. However, for the age group under age 65, projected age-sex-adjusted death rates are slightly higher than in last year’s report. This change primarily results from updating starting levels of mortality to reflect the most recent data. For the age group 65 and over, projected age-sex-adjusted death rates are lower than in last year’s report. This is primarily due to the increased ultimate rates of decline in mortality that are assumed for ages 65 through 84.
After adjustment for changes in the age-sex distribution of the population, the resulting total death rates are projected to decline at ultimate average annual rates of about 0.35 percent, 0.77 percent, and 1.24 percent between 2033 and 2083 for alternatives I, II, and III, respectively. In keeping with the patterns observed in the historical data, future rates of decline are assumed to be greater for younger ages than for older ages, but to a substantially lesser degree than in the past. Accordingly, age-sex-adjusted death rates for ages 65 and over are projected to decline at average annual rates of about 0.32 percent, 0.71 percent, and 1.18 percent between 2033 and 2083 for alternatives I, II, and III, respectively.
Experts express a wide range of views on the likely rate of future decline in death rates. For example, the 2007 Technical Panel on Assumptions and Methods appointed by the Social Security Advisory Board believed that ultimate rates of decline in mortality will be higher than the rates of decline assumed for the intermediate projections in this report. Others believe that biological and social factors may slow future rates of decline in mortality. Evolving mortality trends and developments in health care and lifestyle will be closely monitored to determine what further modifications to the assumed ultimate rates of decline in mortality may be warranted for future reports.
In order to develop projections of the total Social Security area population, assumptions are made for annual legal immigration, legal emigration, other immigration, and other emigration. Legal immigration consists of persons who are granted legal permanent resident (LPR) status. Legal emigration consists of those legal immigrants and native-born citizens who leave the Social Security area population. Net legal immigration is then calculated as the difference between legal immigration and legal emigration. Other immigration consists of immigrants who enter the Social Security area in a given year and stay to the end of that year without having LPR status, such as undocumented immigrants and temporary foreign workers and students. Other emigration consists of other immigrants who leave the Social Security area population or who adjust their status to LPR. Net other immigration is then calculated as the difference between other immigration and other emigration. Net immigration refers to the sum of net legal immigration and net other immigration.
Separate assumptions are developed for the low-cost, intermediate, and high-cost scenarios. The low-cost scenario includes higher annual net immigration and the high-cost scenario includes lower annual net immigration.
Legal immigration increased after World War II to around 300,000 persons per year and remained around that level until shortly after 1960. With the Immigration Act of 1965 and other related changes, annual legal immigration increased to about 400,000 and remained fairly stable until 1977. Between 1977 and 1990, legal immigration once again increased, averaging about 580,000
6 per year. The Immigration Act of 1990, which took effect in fiscal year 1992, restructured the immigration categories and increased significantly the number of immigrants who may legally enter the United States.
Legal immigration averaged about 790,0006 persons per year during the period 1992 through 2000. Legal immigration increased to about 900,000 in 2000 and about 1,000,000 in 2001 reflecting primarily an increase in the number of persons granted LPR status as immediate relatives of U.S. citizens, the only category of legal immigration that is not numerically limited. However, legal immigration declined to less than 800,000 by 2003 as the number of pending applications increased. From 2003 to 2006, legal immigration increased, reaching about 1,200,000 for 2005 and 2006. For 2007, legal immigration decreased to about 1,100,000. Legal immigration in excess of 1,000,000 reflects the concerted effort in recent years to reduce the backlog of pending applications for LPR status.
For the intermediate alternative, the remaining backlog of pending applications is assumed to be reduced by the end of 2009, and thereafter legal immigration is assumed to average approximately 1,000,000 persons per year. For alternatives I and III, annual legal immigration is ultimately assumed to be 1,200,000 persons and 800,000 persons, respectively. These are the same assumptions used in the 2008 report.
The ratios of annual legal emigration to legal immigration are assumed to be 20, 25, and 30 percent for alternatives I, II, and III, respectively. This range is consistent with the limited historical data for legal emigration from the Social Security area. These are the same ratios used in the 2008 Trustees Report. Combining the annual legal immigration and emigration assumptions results in ultimate net legal immigration of 750,000 persons per year under the intermediate alternative. For the low-cost and high-cost scenarios, ultimate annual net legal immigration is 960,000 persons and 560,000 persons, respectively.
The number of other immigrants residing in the Social Security area population is estimated to have been about 9.7 million persons as of January 1, 2000, increasing to about 12.8 million persons as of January 1, 2006. This other-immigrant population is highly mobile and far more likely to leave the Social Security area than is the native-born or legal-immigrant population. The average number of persons entering the other-immigrant population in the period 2000 through 2006 is estimated to have been about 1.5 million per year. During the same period, the number of other immigrants who left the Social Security area or adjusted status to become LPRs is estimated to have averaged about 960,000 per year. Thus, annual net other immigration during this time period is estimated to have averaged approximately 540,000 persons.
For the intermediate assumptions, annual other immigration is assumed to continue at the level of 1.5 million persons throughout the projection period. For the low- and high-cost scenarios, future annual other immigration is assumed to average 1.8 million persons and 1.2 million persons, respectively.
Emigration from the other-immigrant population includes those who leave the Social Security area and those who adjust status to become LPRs. The annual number of other immigrants who leave the Social Security area is estimated based on modeled departures, disaggregated into two groups, for the period 2000-06. The first departing group is set at fixed annual numbers of departures, by age and sex, which remain constant throughout the projection period. This first group is directly related to the number of other immigrants that are assumed to have recently entered the Social Security area. The second departing group is calculated by applying a set of annual departure rates, by age and sex, to the other-immigrant population in the Social Security area. In addition, the annual number of other immigrants who adjust status to become LPRs is assumed to ultimately be 500,000 for the intermediate assumptions. This level is one third of the annual number of other immigrants assumed to enter the Social Security area. For the low- and high-cost scenarios, ultimate annual numbers adjusting status to LPR are assumed to average 600,000 persons and 400,000 persons, respectively.
Under the assumptions and methods described above, the size of the other-immigrant population is projected to grow substantially. This growth reflects the excess of annual other immigration over the combined annual numbers of emigrants and deaths that occur within the other-immigrant population.
Net other immigration decreased from a level averaging over 590,000 per year in the period 2000 through 2003, to about 465,000 in 2006, reflecting an increase in the number of other immigrants adjusting to LPR status as a result of the effort to reduce the backlog of applications for LPR status. By 2010, when the backlog of applications is expected to be eliminated, net other immigration is projected to be about 440,000 persons per year. After 2010, net other immigration is projected to decline steadily to about 275,000 in 2063 and to remain fairly stable thereafter. The decline in net other immigration is attributable to the increasing number of other immigrants residing in the Social Security area. This results in an increase in the numbers who emigrate out of the area based on the rates of departure described above. All other components of other immigration and emigration are set at fixed levels after 2010, and thus do not contribute toward any change in net other immigration. The average annual level of net other immigration over the 75-year projection period is about 315,000 persons. Net other immigration is estimated to average about 410,000 persons per year under the low-cost assumptions and 220,000 persons per year under the high-cost assumptions.
The total level of net immigration (legal and other combined) is estimated to average 1,065,000 persons per year during the 75-year projection period under the intermediate assumptions. For the low-cost assumptions, total net immigration is estimated to average 1,370,000 persons per year. Under the high-cost assumptions, total net immigration is estimated to average 785,000 persons per year.
Demographers express a wide range of views about the future course of immigration for the United States. Some, like the 2007 Technical Panel mentioned in the previous section, believe that immigration will increase substantially in the future. Others believe that potential immigrants may be attracted to other countries or that the U.S. borders could be tightened in the future.
4.
Total Population Estimates
Combining the above assumptions for future fertility, mortality, and net immigration with assumptions on marriage and divorce based on data from the NCHS, projections were made of the population in the Social Security area by age, sex, and marital status as of January 1 of each year 2008 through 2085. The starting Social Security area population for January 1, 2007, is based on the Census Bureau’s estimate of the residents of the 50 States and D.C., and U.S. Armed Forces overseas. The base estimate is adjusted for net census undercount and increased for other U.S. citizens living abroad (including residents of U.S. territories) and for non-citizens living abroad who are insured for Social Security benefits. This starting population was then projected using assumed rates of birth, death, marriage and divorce, and assumed levels of net immigration.
Table V.A2 shows the historical and projected population as of July 1 by broad age group, for the three alternatives. Also shown are aged and total dependency ratios (see table footnotes for definitions).
Life expectancy, or average remaining number of years expected prior to death, is a useful analytical concept. Life expectancy is calculated in two different forms, for two separate purposes.
Period life expectancy is calculated for a given year using the actual or expected death rates at each age for that year. It is a useful summary statistic for illustrating the overall level of the death rates experienced in a single year. It is thus closely related to the age-sex-adjusted death rate that is discussed in section
V.A.2. Period life expectancy for a particular year may be viewed as the expected remaining life at a selected age
only if it is assumed that there is no change in death rates after that year.
Cohort life expectancy truly answers the question “What is the expected average remaining lifetime for an individual at a selected age in a given year?” Cohort life expectancy is calculated using death rates not from a single year, but from the series of years in which the individual will actually reach each succeeding age if he or she survives. Cohort life expectancy is shown in table
V.A4 for those born on January 1 of each calendar year, and for those attaining age 65 on January 1 of each calendar year.
Tables V.A3 and
V.A4 present historical and projected life expectancy calculated on both period and cohort bases. Cohort life expectancy is somewhat greater than period life expectancy for the same year. This is because death rates for any given age tend to decline as time passes and the cohort grows older.
