Bureau of Labor Statistics projections of industrial and occupational employment are developed in a series of six interrelated steps, each of which is based on a different procedure or model and related assumptions: labor force, aggregate economy, final demand (GDP) by consuming sector and product, industry output, industry employment, and employment by occupation. The results produced by each step are key inputs to following steps, and the sequence may be repeated multiple times to allow feedback and to insure consistency.
Labor force projections are based on assumptions about the future size and composition of the current population, as well as on the trends in labor force participation rates of different population groups. Projections are made for more than 136 separate age-sex-race and ethnic groups.
The Census Bureau prepares the population projections. BLS develops participation rate projections using data from the Current Population Survey (CPS) conducted for BLS by the Bureau of the Census. The size and composition of the population are affected by the interaction of three variables: births, deaths, and net immigration. The Census Bureau makes different assumptions for each variable—preparing various combinations of these assumptions. More information about population projections is available from the Bureau of the Census web site.
For this latest round of projections, the interim population projections of the resident population of the U.S. provided to the BLS in 2003 by the Census Bureau was used as the base for the labor force and other projections. The BLS selected the middle population scenario as the base for the labor force and other projections. The size and composition of the population affect not only the labor force projections, but the projected composition of GDP and the levels of employment in various industries and occupations.
BLS currently disaggregates the various race and ethnicity categories into 5-year age groups by sex. Participation rates for these groups are smoothed, using a robust-resistant nonlinear filter and then transformed into logits. The logits of the participation rates are then extrapolated linearly by regressing against time and then extending the fitted series to or beyond the target year. When the series are transformed back into participation rates, the projected path is nonlinear.
After the labor force participation rates have been projected, they are reviewed from the perspectives of the time path, the cross section in the target year, and cohort patterns of participation. The labor force level resulting from the projection is also compared with the labor force derived from an econometric model that projects only the total civilian labor force.
The projected participation rate for each age-sex-race and ethnicity group are multiplied by the corresponding projection of the civilian noninstitutional population to obtain the labor force projection for that group. The groups are then summed to obtain the total civilian labor force.
The aggregate economic projections are developed using a commercially provided econometric model of the U.S. economy—the Macroeconomic Advisers, LLC WUMMSIM Model of the U.S. Economy (MA model). The MA model comprises 134 behavioral equations, 409 identities, and 201 exogenous for a total of 744 variables which describe all facets of aggregate economic performance. Estimates for exogenous variables are provided to the model and a solution of the behavioral and identity equations generated. Finally, the results are evaluated with regard to previously formulated targets for various key indicators of economic behavior.
The principal exogenous assumptions underlying the MA model fall into the categories of monetary policy, fiscal policy, government spending, energy prices and supply, and demographic assumptions. Primary targets, or variables used to assess the behavior of a given set of projections, include the rate of growth and demand composition of real GDP, the labor productivity growth rate, the inflation rate, the level of the unemployment rate, and the international trade related issues. Many solution rounds may be necessary to arrive at a balanced set of assumptions which yield a believable and defensible set of results.
Personal consumption expenditures
are projected in the MA model at an aggregate level. Consumption expenditures for 88 national income and product account categories are estimated for the 2006-2016 period by regressing each of the 88 categories against aggregate consumption and relative prices. These 88 category estimates are then aggregated to the level of total PCE from the macro model and adjusted as necessary to insure consistency between aggregate PCE and the detailed estimates. A bridge table is then used to distribute consumption spending for each of the 88 categories among the 201 producing industries for the 2006-2016 period.
Gross private domestic investment
is initially projected by the MA model for private investment in equipment and software, nonresidential and residential structures, and business inventories. The PIES categories are estimated in greater detail using a system of regression equations that sets GDP, capital stock, and the cost of capital as explanatory variables. In all, projections are made for 13 categories of private investment in equipment and software. The estimates are then aggregated to the level of the macro model control and adjusted as necessary to insure consistency between the macro model aggregate and the detailed estimates. Business inventories, on a commodity basis, are extrapolated based on lagged values of commodity output. These are also aggregated and adjusted to conform to the macro model aggregate of the change in inventories. The controls for nonresidential and residential structures are taken directly from the macro model. All the category controls, with the exception of business inventories, are then distributed to producing sectors using projected bridge tables.
is initially projected by the MA macro model for export goods and services, and import goods and services. Distributional models are used to allocate the forecasted macro model data to a commodity basis. For both exports and imports, these commodity estimates are controlled back to the MA macro model aggregates and are adjusted as necessary to ensure consistency between the detailed estimates and the macro model. Other factors are also considered, including energy forecasts, existing and expected shares of the domestic market, expected world economic conditions, and known trade agreements.
is projected by the MA model for three major government categories: Federal defense, Federal nondefense, and State and local government. Projections for each major category include estimates for four categories: compensation, consumption of general government fixed capital, gross investment, and all other consumption expenditures. These are further disaggregated based upon past trends and expected government political and policy changes. For State and local government, expenditures are no longer subdivided between education and noneducation functions. Finally, each of the twelve expenditure categories is allocated to the appropriate industry sector or sectors, such as electric utilities or hospitals.
Projected industry output is derived using a set of projected input-output tables. One of these tables, projected final demand, results from the preceding step. In addition, the projected market share and direct requirements tables must be created. The projected market share table is initially based on the last historical table. The projected direct requirements is extrapolated based on historical trends. These two tables are used to create direct requirements tables. These requirements tables yield the projected levels of industry and commodity output required to satisfy projected final demand.
The next step is to project the industry employment necessary to produce the projected output. To do so, projected output is used in regression analysis to estimate hours worked by industry. From these hours data, projected wage and salary employment by industry is derived. For each industry, the share of self-employed and unpaid family workers is extrapolated using historical data, and the total employment by industry is derived. Finally, the implied output per hour is calculated for each industry and used to evaluate the projected output and employment.
Employment by occupation
An industry-occupation matrix is used to project employment for wage and salary workers. The matrix shows occupational staffing patterns—each occupation as a percent of the work force in every industry. It includes more than 300 detailed industries and more than 750 detailed occupations. Data for staffing patterns in the base-year matrix come primarily from the BLS Occupational Employment Statistics surveys, which collect data from employers on a 3-year cycle.
Projected occupational staffing patterns for each industry are based on anticipated changes in the way goods and services are produced, and then are applied to projected industry employment. The resulting employment is summed across industries to get total wage and salary employment by occupation. For occupations that are projected not to change their share of industry employment, employment will grow in line with the growth of the industries in which they are concentrated. For example, health care support occupations are expected to grow rapidly, mainly because the health care industry will grow quickly.
Employment in an occupation also may grow or decline as a result of many factors that affect its share of industry employment. For example, rapid growth is expected among social and human service assistants as employers increasingly rely on these workers to undertake greater responsibility for delivering services to clients. Rapid growth also is expected among computer systems analysts as technology advances, and as organizations place more emphasis on network applications and on maximizing the efficiency of their computer systems. Conversely, productivity improvements, such as automation and developments in computer software, will result in slower than average growth among office and administrative support workers, machine operators, and assemblers—thus lowering their proportion of the workforce. The projected-year matrix incorporates these expected changes.
Data on self-employed workers, unpaid family workers, and workers who have a primary or secondary wage and salary job in agricultural production, forestry, fishing, or private households in each occupation come from the Current Population Survey. Workers in these groups for each occupation are projected separately for the economy as a whole rather than by industry, and are added to the projections of wage and salary workers to obtain total projected employment for each occupation.
Last Modified Date: December 4, 2007
Please help us improve our website by participating in the survey that begins below!
Our Pledge of Privacy
The answers you provide on this survey will be used by the Bureau of Labor Statistics (BLS) to improve the website. Please note:
Click on the yes button below to take the survey after you have finished visiting our website. The survey will be available in the taskbar at the bottom of your screen when you are ready to take it.