Scientific Data Documentation
Population Projections Of The US, 1992-2050
*SEE CENSUS PROJECTIONS 1992-2050 DATASET NAMES FOR DSN.
GENERAL INFORMATION
Population Projections of the United States, by Age, Sex. Race, and Hispanic
Origin.: 1992 to 2050 {machine readable data File} / prepared by the Bureau
of the Census, Population Division, 1992.
Users who purchase these diskettes should also purchase the publication,
"Population Projections of the United States, by Age, Sex, Race, and
Hispanic Origin: 1992 to 2050." (Current Population Reports P251092) This
report includes an analysis of the results of these projections as well as
detailed information on the assumptions and methodology used in generating
the series. To order this publication, contact the Statistical Information
Staff at 301/763-5002. Questions concerning the methodology, analysis of the
data, or file layout should be addressed to Jennifer Day, Population
Projections Branch at 301/763-1902.
SUBJECT-MATTER DESCRIPTION
Middle series projections of the resident population of the United States by
age, sex, race, and Hispanic origin, from 1992 to 2050.
The data are consist of eight files: six include data on the projected
population, one file shows the projected components of change, and one file
consists of the Armed Forces overseas population.
For July 1 of each year, 1992 to 2050, the population file presents the
middle series of projections classified by age, sex, race, and Hispanic
origin. These projections are based on July 1, 1991 estimates, consistent
with the 1990 census, as enumerated and projected forward using the
inflation-deflation variant of the cohort-component method. Projections are
provided for 102 age categories (yearly total, individual years 0 to 99, 100
and over), four race categories White; Black; American Indian, Eskimo, and
Aleut; and Asian and Pacific Islander), Hispanic origin, and the four races
without the Hispanic origin population (non-Hispanic race groups).
The component file presents data 1992 through 2050. This file includes the
annual July 1 and January 1 populations in addition to the annual number and
rates of birth. death, net immigration, natural increase, and net change.
UNIVERSE DESCRIPTION
Resident Population of the United States
TECHNICAL DESCRIPTION
File Type
ASCII
File Size
File name Number of bytes Records Columns
---- ---- ------ -- ----- ------- -------
P1992_00 203,796 916 220
P2001_10 226,440 1020 220
P2011_20 226,440 1020 220
P2021_30 226.440 1020 220
P2031_40 226,440 1020 220
12041_50 226,440 1020 220
COMPONENT 61,362 590 91
AFO 10536 49 213
Sort Sequence
Series Number by Year by Age.
EXPLANATION OF DATA FIELDS
Series: All data on the diskettes are based on the middle series.
The letter "A" represents the middle series data.
Year: Annually from 1992 to 2050. Population data are as of July 1.
Birth, death, net immigration, natural increase, and net
change are for the calendar year.
Age group: The total population is code 999 which is the first record for
each year. This is followed by 0 to 99 which are individual
ages and 100 which is the centenarian population.
Race and
Hispanic
origin: The data are shown for the Total; White; Black; American Indian,
Eskimo, and Aleut; and Asian and Pacific Islander populations.
Data are also shown for the Hispanic origin population, and the
non-Hispanic White; non-Hispanic Black; non-Hispanic American
Indian, Eskimo, and Aleut; and non-Hispanic Asian and Pacific
Islander populations.
Persons of Hispanic origin may be of any race.
In the components of change File, each group is represented by a letter:
A = Total population
B = White population
C = Black population
D = American Indian, Eskimo, and Aleut population
E = Asian and Pacific Islander population
F = Hispanic population
G = non-Hispanic white
H = non-Hispanic Black
I = non-Hispanic American Indian, Eskimo, and Aleut population
J = non-Hispanic Asian and Pacific Islander population
RECORD LAYOUT
Components of Change file: COMPONEN
Location Length Type Data
-------- ------ ---- ----
2 1 Series letter
3 1 Race and Hispanic origin
4-7 4 Numeric Year
8-17 10 Numeric July 1 population
18-22 5 Numeric Rate of Net Change per 1,000 mid-year pop.
23-27 5 Numeric Rate of National Increase per 1,000
28-32 5 Numeric Rate of Birth per 1,000 mid-year pop.
33-37 5 Numeric Rate of Deaths per 1,000 mid-year pop.
38-42 5 Numeric Rate of Net Immigration per 1,000 mid-yr. pop.
43-52 10 Numeric January 1 population
S~ 10 Numeric Net Change
63-72 10 Numeric Natural Increase
73-82 10 Numeric Births
83-92 10 Numeric Deaths
93-102 10 Numeric Net Immigration
Population Files
File Name File Description
---- ---- ---- -----------
P1992_00 US resident population projections 1992 to 2000 (middle series)
P2001_10 US resident population projections 2001 to 2010 (middle series)
P2011_20 US resident population projections 2011 to 2020 (middle series)
P2021_30 US resident population projections 2021 to 2030 (middle series)
P2031_40 US resident population projections 2031 to 2040 (middle series)
P2041_50 US resident population projections 2011 to 2050 (middle series)
Location Length Type Data
-------- ------ ---- ----
2 1 Character Series
3-6 4 Numeric Year
7-10 3 Numeric Age
11-20 10 Numeric Total population
21-30 10 Numeric Total male population
31~0 10 Numeric Total female population
41-50 10 Numeric White male population
51-60 10 Numeric White female population
61-70 10 Numeric Black male population
71-80 10 Numeric Black female population
81-90 10 Numeric American Indian, Eskimo, and Aleut male
91-100 10 Numeric American Indian, Eskimo. and Aleut female
101-110 10 Numeric Asian and Pacific Islander male population
111-120 10 Numeric Asian and Pacific Islander female population
121-130 10 Numeric Hispanic male population
131-140 10 Numeric Hispanic female population
141-150 10 Numeric White non-Hispanic male population
151-160 10 Numeric White non-Hispanic female population
161-170 10 Numeric Black non-Hispanic male population
171-180 10 Numeric Black non-Hispanic female population
181-190 10 Numeric Amer Indian,Eskimo,& Aleut non-Hisp male pop.
191-200 10 Numeric Amer Indian,Eskimo,& Aleut non-Hisp female pop.
201-210 10 Numeric Asian & Pac Islander non-Hisp male population
211-220 10 Numeric Asian & Pac Islander non-Hisp female population
Armed Forces Overseas file: AFO
The Armed Forces overseas file contains the unrounded population estimates of
Armed Forces overseas as of July 1,1991. Data are shown for total population
and ages 17 through 64. This population is assumed to remain constant through
out the entire projection period (1992-2050). The projected population
figures for the United States are for resident population only. To calculate
the Total Population including Armed Forces overseas, the population in this
file needs to be added to each projection year. Past population projections
completed by the Bureau of the Census presented the Total Population
including Armed Forces overseas.
Location Length Type Data
-------- ------ ---- ----
1-3 3 Numeric Age
4-10 7 Numeric Total population
11-17 7 Numeric Total male population
18-24 7 Numeric Total female population
25-31 7 Numeric White total population
32-38 7 Numeric White male population
39-45 7 Numeric White female population
46-52 7 Numeric Black total population
53-59 7 Numeric Black male population
60-66 7 Numeric Black female population
67-73 7 Numeric American Indian, Eskimo, and Aleut total pop.
74-80 7 Numeric American Indian, Eskimo. and Aleut male pop.
81-87 7 Numeric American Indian, Eskimo, and Aleut female pop.
88-94 7 Numeric Asian and Pacific Islander total population
95-101 7 Numeric Asian and Pacific Islander male population
102-108 7 Numeric Asian and Pacific Islander female population
109-115 7 Numeric Hispanic total population
116-122 7 Numeric Hispanic male population
123-129 7 Numeric Hispanic female population
130-136 7 Numeric White non-Hispanic total population
137-143 7 Numeric White non-Hispanic male population
144-150 7 Numeric White non-Hispanic female population
151-157 7 Numeric Black non-Hispanic total population
158-164 7 Numeric Black non-Hispanic male population
165-171 7 Numeric Black non-Hispanic female population
172-178 7 Numeric Amer Indian,Eskimo,& Aleut non-Hisp total pop.
179-185 7 Numeric Amer Indian,Eskimo,& Aleut non-Hisp male pop.
186-192 7 Numeric Amer Indian,Eskimo,& Aleut non-Hisp female pop.
193-199 7 Numeric Asian and Pacific non-Hisp total population
200-206 7 Numeric Asian and Pacific non-Hisp male population
207-213 7 Numeric Asian and Pacific non-Hisp female population
DETAILED METHODOLOGY
General Information
NOTE: These projections are consistent with the 1990 census as enumerated.
NOTE: This is the first projections report which presents data for four race
groups (White; Black; American Indian, Eskimo, and Aleut; and Asian and
Pacific Islander) and the Hispanic origin population. The race groups
also are separated into their Hispanic and Non-Hispanic components.
NOTE: Many of the trends described here are substantially different from
those shown in the previous projections (Current Population Reports,
Nos. 995 and 1018). These differences are primarily due to the
changeover to the 1990 census base and three significant changes in our
assumptions: 1) Future fertility is assumed to remain near current
levels; 2) Future immigration is assumed to remain near current levels;
and 3) Fertility and mortality differentials by race/ethnic groups are
assumed to continue their current trends. The future course of
population change could be substantially different -- depending on the
actual trends in births, deaths, and net migration.
The projections shown here supersede the information contained in
Current Population Reports, Series P-25, No. 1018 and Series P-25,
No 995. The methodology used to generate the projections in this
report is similar to that used for the earlier reports. However, the
base population data and several assumptions have changed.
The Cohort-Component Framework
Six sets of data are required to generate these population projections
using the cohort-component model. These are a base-year population, projected
survival rates, future net immigration statistics, 1990 inflation/deflation
rates, and an estimated Armed Forces overseas population. The most difficult
aspect of producing these population projections involves deriving the base-
year starting points and rates. Each data set is organized into 16 different
race/ethnic/sex matrices with a cell for each year of age 0 to 100 and over.
The sixteen matrices are White; Black; American Indian, Eskimo, and Aleut;
Asian and Pacific Islander - by Hispanic origin (Hispanic and Not Hispanic)
and by sex. The sum of all the cells in all 16 matrices equals the total
population.
Starting with a July 1, 1991 modified population estimate based on the
1990 census, each cell is inflated by Demographic Analysis to correct for
persons not included in the population count in 1990. Then each age/race/
ethnic/sex cell is survived forward to July 1, 1992 by applying the
appropriate survival rate.
The population under 1 is created by first calculating the population
of women exposed to the risk of child bearing. Generally this involves
averaging the July 1,1991 and July 1,1992 inflated female population, of
each race/ethnic group by single years of age between the ages 14 through 49.
Then, the corresponding age/race/ethnic specific fertility rate is applied
to this averaged female population to produce, after aggregation, the total
number of births by race/ethnicity for that fiscal year. The assumed sex
ratio for each group is used to divide the births into males and females.
Finally, the number of births by sex and race are survived forward to July 1,
1992.
After the births are calculated, net immigration by age/sex/race/
ethnicity are added. Then the movement of the population of Armed Forces
overseas are applied to the population by detailed group. Next, the
population is deflated to be consistent with the 1990 census count. A small
pro rata adjustment is made to the deflated age estimates in each sex/race/
ethnic group to bring them in exact agreement with an independent estimate
of the total population of each group.
Finally, the 16 groups are summed, creating the groups most frequently
requested, and are displayed in this report. This includes adding
the Hispanic and Not Hispanic groups for each race to make a total
population by age/sex for each race, and adding the 8 Hispanic origin
matrices to get the total age/sex Hispanic origin population. As these groups
are added, new total rates are derived for these new groups. The same set of
procedures when applied to the July 1,1992 population would generate the July
1, 1993 population. This process is continued until 2050.
The Base Population
The beginning population of these projections is the July 1, 1991
estimate. Because this estimate reflects births by race of child, a slight
re-organization of the zero and one-year-old populations by race was made to
account for the new mother rule of race designation of births used in these
projections. These estimates are consistent with the 1990 census count, but
cannot be directly compared to the published results by age and race because
modifications were made to the data to correctly place each in an appropriate
age and race category. This was done to adjust for age misreporting and the
reporting of an unspecified race in the 1990 census.
Fertility
Assumptions
As in the past, three different future fertility levels are used. These are
derived from analysis of natality statistics for five groups of women by race
group and origin. Assuming that the pattern of fertility would be the same
for all Hispanic women, across all races, we derived levels for Hispanic-
origin women and the remaining four non-Hispanic race groups: non-Hispanic
White; non-Hispanic Black; non-Hispanic American Indian; and non-Hispanic
Asian. The levels for each total race group would then be the combination of
the Hispanic and non-Hispanic proportions of that race group. As these
proportions change, so would the combined fertility levels. In 1989, NCHS
changed their method of reporting race of births to reflect the race of the
mother. The projected fertility levels and rates in this report use this new
rule.
Several fertility assumptions in this report are founded on past
trends. First, this projection will not assume race/ethnic fertility
convergence. Historical data shows that the major differences between White
and Black fertility is timing, that is, Blacks tend to have their children
at earlier ages than Whites. After age 25, however, White and Black fertility
has been about the same. Yet, there is no compelling evidence of overall
convergence of Black-White or any race-ethnic fertility. Second, in the last
decade, many women delayed the start of childbearing until their late 20s or
30s. This recent shift to a new age pattern of childbearing is assumed to
remain constant. Third, since the end of the Baby Boom, completed cohort
fertility has remained about the same. Therefore, there appears to be no
reason to assume a change from the fertility levels for Whites, Blacks, and
American Indians.
The fertility levels in the middle series are based on the premise
that foreign-born women may have higher fertility than native-born women.
Thus, Hispanic and Asian fertility are assumed to decrease after 2000 as the
share of their fertility contributed by the foreign born will decrease. The
non-Hispanic White, non-Hispanic Black, and non-Hispanic American Indian
fertility would remain constant throughout the projected period.
In the low series, the fertility rates is assumed to fall for all
races and Hispanic origin, decreasing 10 percent for non-Hispanic White and
20 percent for the other four groups by the year 2010. The reverse is assumed
for the high series. The high series would have an increase in rates, up 20
percent for non-Hispanic Whites, and 10 percent for the other four groups by
2010.
Creation of 1990 Central Birth Rate
The birth rates for these projections are based on 1990 fertility rates. The
beginning rates were created using NCHS natality data and 1990 census
population figures. Because only provisional data were available for 1990,
1989 final detailed data were controlled by the provisional 1990 number of
births (corrected for under registration). Births for each race group --White,
Black, Asian, and American Indian -- were divided by a 1989 (1990 census
based) population. Because the base population for these projections was a
1990 census base, the base rates needed to be consistent with this new
population count and distribution. Using a time series model, the White 1990
birth rates were forecast, adjusted to the provisional 1990 birth totals.
Using a ratio to the White 1989 rate, the birth rates were derived for the
other races - Black, American Indian, and Asian.
Birth rates for the Hispanic origin population are based on NCHS
birth data. In 1989, these data included 47 states and the District of
Columbia, covering about 99.2 percent of the total US. Hispanic population
(based on 1990 census figures). The Hispanic rates were held constant at 1989
levels. Non-Hispanic rates were created by extracting the Hispanic data out
of each race for both population and births. Finally, each group's total
fertility rate was rounded down slightly.
Life Expectancy
Assumptions
As in the last Census projections, three basic city assumptions are used.
The middle life expectancy assumptions reflects a slow improvement in life
expectancy. The last 10-year trend of mortality improvement, from 1980 to
1990, is replicated, and some additional impact of AIDS is included .The
incidence of AIDS is projected to increase linearly until the turn of the
century. After 2000, mortality from AIDS will slowly decrease, returning to
1990 level of AIDS morality by 2050. The low life-expectancy series assumes
that current mortality rates will continue, with an increase over the next
15 years in deaths due to AIDS. This uses a 1990 base lifetable with AIDS
projected to increase linearly up to the year 2005, then remain constant. The
high life-expectancy assumption, or rapid improvement series, replicates the
pattern of mortality, between 1970 to 1980, thus ignoring the impact of AIDS.
Lifetables
The 1990 lifetable was based on NCHS death data (final 1989 and sample 1990)
for age, sex, race, and Hispanic origin. Race detail provided deaths only for
the White, Black, and "Other" populations. To obtain detailed deaths for the
American Indian and the Asian populations, the "Other" group's deaths were
split by subtracting death data provided by the Indian Health Service. The
remaining deaths were considered to be Asian. Although this technique
provided relatively adequate estimates of deaths for the Asian and American
Indian races, the quality of mortality estimates for these two groups still
remains questionable. The deaths for each race then were divided by their
appropriate population estimate for July 1,1990, creating death rates by
five-year age groups, sex, race, and Hispanic origin. For the oldest age
groups, where frequently age specific death data is often misreported, the
rates were corrected using data from the Social Security Administration. The
total death rates were then adjusted to agree with the estimated 2,162,000
deaths in 1990.
Non-Hispanic death rates by race were computed by extracting the
Hispanic deaths from the total deaths for each race. By applying the
proportional race distribution of the Hispanic origin population to the total
number of Hispanic deaths, the Hispanic deaths by race were derived. Thus,
the remainder of deaths in each race were non-Hispanic.
The death rates used for every projected lifetable was based on the
assumed change from the death rates used in the 1990 lifetable. In each
assumption, sex and race differentials were not assumed to converge, but were
determined by the rate of change applied to each individual group. The rates
of change were computed based on adjusted level death rates for 1970, 1980,
and 1990. In the middle series, lifetables were computed for 2000 -- a turning
point of the series, and 2050 -- the end point. For the alternative series, a
2005 lifetable was created for the low life-expectancy assumption, holding
the rest of the projected period constant at 2005 levels (i.e., AIDS gets no
worse or better), and for the high life expectancy series a 2050 lifetable
was made. Survival rates were extrapolated for each year between these
points.
Net Immigration
The net immigration component used in these projections is composed of
five types of migration, four which increase the population (immigration)
and one that decreased the population (emigration). In the low, middle, and
high net immigration, the same age/sex/race/ethnic ratio is used throughout,
based on a July 1, 1991 estimate of each migration type, but raked to the
alternative levels for each type.
The levels of legal and refugee immigration are heavily influenced by
Federal legislation and the political environment. Therefore, the middle
assumption for this type of immigration is based on current levels and
interpretation of current laws. The low alternative represents the legal and
refugee immigration experience during the 1980s, where as the high assumption
reflects the possibility of piercing the legal cap on this type of
immigration through modifications to the current law. Undocumented
immigration is difficult to measure. Currently, the Census Bureau's best
estimate adds about 200,000 net undocumented immigration the US. each year.
Yet, the wide range between the low and high alternatives reflects some of
the uncertainty of the actual number. Puerto Rican and civilian citizen
immigration levels exhibit the 1991 estimate of these components in the
middle series, equally bounded by the low and high alternatives. Emigration,
similar to the undocumented migration, contains some unknown qualities.
Similar to previous Census Bureau projections, emigration is considered a
constant, instead of a proportion of the total of all in-migration. However,
unlike other projections, the low series does not show a lower number for
emigration; in fact, it reflects a greater population exiting. Logically, if
conditions exist for low in-migration (for example, an economic downturn),
the same conditions or reasons would also drive more people out of the
country. The reverse is assumed for the high series.
RACE AND ETHNIC DEFINITIONS AND CONCEPTS
General Information
The race classification used by the Census Bureau generally adheres
to the guidelines the Federal Statistical Directive No. 15, issued by the
Office of Management and Budget, which provides standards on race and
Hispanic origin categories for statistical reporting to be used by all
Federal agencies. The race and Hispanic origin categories are defined as the
following:
American Indian, Eskimo, and Aleut
A person having origins in any of the original peoples of North America, who
maintains cultural identification through tribal affiliation or community
recognition.
Asian and Pacific Islander
A person having origins in any of the original peoples in the Far East,
Southeast Asian, the Indian subcontinent, or the Pacific Islands. This area
includes, for example, China, India, Japan, Korea, the Philippine Islands,
and Samoa.
Black
A person having origins in any of the black racial groups of Africa.
Hispanic
A person of Mexican, Puerto Rican, Cuban, Central or South American or other
Spanish culture or origin, regardless of race.
White
A person having origins in any of the original peoples of Europe, North
Africa, or the Middle East.
