Because precise prediction of these various factors is impossible, estimates are 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, alternative II, 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 alternative I is the most optimistic, and the high cost alternative III is the most pessimistic of the plausible economic and demographic conditions.
The values assumed after the first 5 to 25 years for both the economic and the demographic factors are intended to represent the average experience and are not intended to be exact predictions of year-by-year values. Actual future values will likely exhibit fluctuations or cyclical patterns, as in the past.
Although these sets of economic and demographic assumptions have been developed using the best available information, the resulting estimates should be interpreted with care. In particular, the resulting estimates are not intended to be exact predictions of the future status of the OASDI program, but rather, they are intended to be indicators of the trend and range of future income and outgo, under a variety of plausible economic and demographic conditions.
1. Economic Assumptions
The principal economic assumptions for the three alternatives are
summarized in table II.D1
Alternatives I, II, and III represent a range of economic assumptions which have been designed to produce variation in Social Security's financial status that should encompass most of the possibilities that might be encountered. The intermediate set of assumptions (alternative II) represents the Trustees' consensus expectation of moderate economic growth throughout the projection period. The low cost assumptions (alternative I) represent a more optimistic outlook, with relatively strong economic growth projected during the short-range period and tapering off a little during the long-range period. The high cost assumptions (alternative III) represent a relatively pessimistic forecast in which the economy experiences generally weak economic growth and business cycles with two recessions in the short-range period. Economic cycles are not included in assumptions beyond the first 5 to 10 years of the projection period because inclusion of such cycles has little effect on the long-range estimates of financial status.
A period of sustained real economic growth began in 1982 and ended with the recession that started with the third quarter of 1990. After a total decline in real GDP of 2.0 percent through the first quarter of 1991, the economy has returned to another sustained period of real economic growth. This period is assumed to continue for alternatives I and II. For the short-range period (1997-2006), average annual real GDP growth is assumed to be about 2.6 percent for alternative I and 2.0 percent for alternative II.
For alternative III, weak growth and an increased rate of price inflation are assumed for the first three quarters of 1997. The first projected recession begins in the fourth quarter of 1997, lasts 3 quarters, and results in a total decline in real GDP of 1.4 percent. After 8 quarters of recovery, a second recession, with a total decline in real GDP of 3.0 percent, is assumed to begin in the third quarter of 2000, lasting 4 quarters. After the second recession, a moderate economic recovery is assumed through the year 2003, with continued modest economic growth thereafter.
After the year 2006, the projected rates of growth in real GDP, for all three alternatives, are determined by the assumed rates of growth in employment, average hours worked, and labor productivity. The trend toward slower growth in real GDP after 2006 results primarily from much slower growth in the working age population, as the "baby-boom" generation approaches retirement and succeeding generations reflecting lower birth rates reach working age. The slowdown in the growth rate in real GDP also reflects the assumed leveling of some labor force participation rates for women, which have risen substantially over the past 25 years, and the continuation of the historical downward trend of labor force participation rates among men in the future. The annual rate of growth in total labor force decreased from over 2 percent in the 1970s to about 1.2 percent for 1990 through 1996. After 1996 the labor force is projected to increase at about 1.0 percent per year, on average, through 2006, and to increase more slowly thereafter, reflecting the projected slowing of growth of the working-age population as compared with the experience of the 1980s and early 1990s.
The age-sex adjusted unemployment rate, for alternatives I and II, is assumed to move gradually toward ultimate average levels of 5.0 and 6.0 percent, respectively, by 2002. For alternative III, the age-sex-adjusted unemployment rate is assumed to reach its ultimate average level of 7.0 percent by 2005, after a recovery that is assumed to follow the second projected recession.
Unemployment rates through 2006 are in the most commonly cited form, the civilian rate, which describes the differences between aggregate civilian labor force and aggregate civilian employment. For years after 2006, however, total age-sex adjusted rates are presented. These include the military (which reduces the rate by about 0.1 percentage point relative to the civilian rate) and are age-sex adjusted to the 1995 labor force. Such total rates better represent the total population covered by the OASDI program and adjust for the changing age-sex distribution of the labor force, which can obscure the comparison of unemployment rates over different time periods.
As indicated in last year's report, the Bureau of Economic Analysis changed from a fixed-weighted to a chain-weighted price measure for the purpose of calculating the real-growth component of GDP. In addition, the historical values of real GDP growth over the period 1959-94 were revised. The effects of these changes were not reflected in the projected growth rates in real GDP for years after 1995 that were shown in last year's report. These changes are now reflected in the assumed real GDP growth rates shown in this report.
For the intermediate projection, each of the other economic parameters is selected reflecting what the Trustees believe to be the most likely future course of the economy at the time of preparation of this report, consistent with the assumed pattern of real GDP growth.
The annual rate of increase in the Consumer Price Index for Urban Wage Earners and Clerical Workers (CPI-W) was 2.9 percent in 1996. For alternative II, the CPI-W (hereinafter denoted as "CPI") is assumed to increase 3.2 percent in both 1997 and 1998, moving toward the assumed ultimate rate of 3.5 percent by 2001. For alternative I, the CPI is projected to increase 2.9 percent in 1997 and 2.6 percent in 1998, moving toward the assumed ultimate rate of 2.5 percent by 1999. For alternative III, the CPI is projected to increase from a relatively low 2.9 percent in 1996 to a relatively high 5.4 percent in 1999 and 6.0 percent in 2000, eventually stabilizing at the assumed ultimate rate of 4.5 percent in 2002. Recent changes by the Bureau of Labor Statistics have been reflected in assumptions for future price inflation in this report. These improvements are expected to result in measured CPI growth that is about 0.2 percentage point lower per year. In addition, the assumed future rate of growth in the CPI has been reduced by 0.3 percentage point based on historical analysis and expected future trends.
Changes in the assumed actual and measured growth rates in the CPI resulted in a reevaluation of several other assumptions. The GDP (chain-weighted) price index was assumed to grow at the same ultimate rate as the CPI for the 1996 report. The 0.2 percentage point reduction in CPI growth will result in a direct reduction in the measured growth of the GDP price index of about 0.1 percentage point. These changes, alone, tend to have no effect on the ultimate measured growth in nominal GDP, wages, and interest rates. The improvements in CPI measurement, thus, tend to increase expected real growth in average wages and productivity by 0.2 percentage point and 0.1 percentage point, respectively. In addition, the improvements tend to increase the level of the real interest rate by 0.2 percentage point.
However, these changes also produce an assumed ultimate GDP price inflation rate that is higher than that in the CPI. Imposing an assumption that the GDP price inflation rate will be 0.1 percentage point lower than that in the CPI, consistent with recent analysis of the bias in these price measures, requires a further reduction in the assumed GDP price inflation rate of 0.2 percentage point. This lower assumed GDP price inflation rate lowers assumed nominal growth in both GDP and wages, resulting in a 0.2 percentage point reduction in the ultimate real-wage differential. The ultimate assumed rate of decline in average hours worked per week is also changed for this report, from an annual decline of 0.2 percent to 0.1 percent for the intermediate assumptions, based on a reevaluation of historical data. This change in hours tends to increase the assumed real-wage differential by 0.1 percentage point.
The changes described above would raise the productivity and real wage growth assumptions by 0.1 percentage point above last year's assumptions -- from 1.36 to 1.46 percent and from 1.0 to 1.1 percent, respectively. However, we are assuming a productivity growth rate of 1.26 percent and a corresponding real wage growth of 0.9 percent in this year's report. This is the consequence of lowering the productivity growth assumption by 0.2 percentage point based on analysis of historical data after changing to the chain-weighted measure of real growth in the GDP.
The annual rate of change in the average wage in covered employment is assumed to drop, initially, from the estimated 4.2 percent increase for 1996, to a projected rate of 4.0 percent for 1997, and then to rise, generally, averaging about 4.2 percent for the period 1997 through 2006. Growth in the average wage (which is equal to price inflation plus the real-wage differential) through 2006 averages somewhat less than the ultimate assumed rate of 4.4 percent because price inflation and real wage growth rates are assumed to average less than their ultimate levels through this period. Between 2005 and 2015, growth in the average covered wage is slightly higher than the assumed ultimate rate of 4.4 percent, reflecting the gradual movement toward complete inclusion of Federal civilian employees. After 2015, the average covered wage growth rate remains at the ultimate assumed rate of 4.4 percent.
The real-wage differential (i.e., the difference between the annual rates of change in the average wage in covered employment and in the CPI) was 1.4 percent in 1996, based on preliminary data. After 1996, under the intermediate alternative, the real-wage differential is projected to be between 0.0 and 1.0 percent for the years 1997 through the year 2015, thereafter remaining at the ultimate assumed differential of 0.9 percent. For the low cost alternative I, the real-wage differential is assumed to be in the range of 0.2 percent to 1.5 percent between 1997 and 2015, thereafter remaining at the ultimate assumed real-wage differential of 1.4 percent. For the high cost alternative III, a more pessimistic real-wage differential is assumed for the short-range period, averaging 0.2 percent per year. After 2015, the real-wage differential is assumed to be 0.4 percent per year for alternative III.
The average annual nominal interest rate for securities newly issued to the trust funds decreased from 6.9 percent in 1995 to 6.6 percent for 1996, and is projected to move gradually toward the ultimate assumed level through the short-range period. After 2006, the average annual nominal interest rates are assumed to be 5.9, 6.2, and 6.4 percent for alternatives I, II, and III, respectively.
For alternatives I and III, respectively, values for each of the economic
parameters are selected to generally reflect a more optimistic and a
more pessimistic future financial status of the program. Some of the
parameters would normally be expected to deviate in opposite directions
from the values assumed for the intermediate alternative. Thus,
alternatives I and III also assume structural economic shifts in the
relationships among parameters which tend toward low cost and high
Table II.D1. Selected Economic Assumptions by
Average annual percentage
rate 4/ (percent)
rate 5/ (percent)
|1996||7/ 2.5||7/ 4.2||2.9||1.4||6.6||5.4||1.2|
The real GDP (gross domestic product) is the value of total output of goods
and services, expressed in 1992 dollars.|
3 The real-wage differential is the difference between the percentage increases, before rounding, in (1) the average annual wage in covered employment, and (2) the average annual Consumer Price Index.
4 The average annual interest rate is the average of the nominal interest rates, which, in practice, are compounded semiannually, for special public-debt obligations issuable to the trust funds in each of the 12 months of the year.
5 Through 2006, the rates shown are unadjusted civilian unemployment rates. After 2006, the rates are total rates (including military), adjusted by age and sex based on the average labor force for 1995.
7 Preliminary. Wages in covered employment are considered preliminary for several years primarily due to uncertainty associated with estimates of amounts above the benefit and contribution base. Estimated real GDP for 1996 is subject to revision; the value for the intermediate assumption is shown.
2. Demographic Assumptions
The principal demographic assumptions for the three alternatives are
shown in table II.D2.
For the intermediate projection, the assumed ultimate total fertility rate of 1.9 children per woman is attained in 2021 after a gradual decline from the preliminary estimate for 1996 of 2.01 children per woman. The age-sex-adjusted death rate is assumed to decrease steadily during the entire projection period, with a total reduction of 35 percent from the 1996 level by 2071. Life expectancies at birth in 2075 are 78.9 years for men and 84.3 years for women, compared to 72.6 and 79.3 years, respectively, in 1996. Life expectancies at age 65 in 2075 are projected to be 18.8 years for men and 22.3 years for women, compared to 15.5 and 19.2 years, respectively, in 1996. Total net immigration is assumed to rise over the next several years reaching an ultimate level of 900,000 persons per year by the year 2000. The ultimate assumed level of net annual immigration is the combination of 600,000 net legal immigrants per year and 300,000 net other-than-legal immigrants per year.
For the low cost alternative I, the total fertility rate is assumed to rise to an ultimate average level of 2.2 children per woman by 2021. The age-sex-adjusted death rate is assumed to decrease more slowly than for the intermediate alternative II, with the total reduction from the 1996 level being 16 percent by 2071. Life expectancies at birth in 2075 are 76.0 years for men and 81.2 years for women, while at age 65 they are 16.5 and 19.7 years, respectively. Total net immigration is ultimately assumed to be 1,150,000 persons per year. The assumed level of net annual immigration is the combination of 700,000 net legal immigrants per year and 450,000 net other-than-legal immigrants per year.
For the high cost alternative III, the total fertility rate is assumed to
decrease to an ultimate level of 1.6 by 2021. The age-sex-adjusted
death rate is assumed to decrease more rapidly than for alternative
II, with the total reduction from the 1996 level being 54 percent by
2071. Life expectancies at birth in 2075 are 82.9 years for men and
88.3 years for women, while at age 65 they are 21.9 and 25.6 years,
respectively. Total net immigration is ultimately assumed to be
750,000 persons per year, the combination of 550,000 net legal
immigrants per year and 200,000 net other-than-legal immigrants per
Selected Demographic Assumptions by Alternative,
death rate 2/
At age 65|
The total fertility rate for any year is the average number of children who
would be born to a woman in her lifetime if she were to experience the birth
rates by age observed in, or assumed for, the selected year, and if she were to
survive the entire childbearing period. The ultimate total fertility rate is
assumed to be reached in 2021.
2 The age-sex-adjusted death rate is the crude rate that would occur in the enumerated total population as of April 1, 1990, if that population were to experience the death rates by age and sex observed in, or assumed for, the selected year.
3 The life expectancy for any year is the average number of years of life remaining for a person if that person were to experience the death rates by age observed in, or assumed for, the selected year.
The ultimate total fertility rates, as shown in table II.D2, are unchanged from last year's report. Due to additional data for the years 1994 and 1995, the total fertility rates during the first 25 years of the projection period are, generally, slightly lower than those in last year's report.
Mortality values shown in table II.D2 are different from values in last year's report. Additional data for the years 1992-95, resulted in overall lower mortality rates estimated for the starting year 1996 and slightly higher rates of mortality reduction during the first 25 years of the projection period. The ultimate rates of reduction in mortality are the same as those used in last year's report. In addition, the age-sex adjusted death rates are different because the population used for weighting the central death rates was changed from the enumerated total population as of April 1, 1980, to the enumerated total population as of April 1, 1990.
In addition to the assumptions discussed above, many other factors are necessary to prepare the estimates presented in this report. Section II.H includes a discussion of many of those factors.