Gross state product by industry, 1977-96.

Beemiller, Richard M. ; Downey, George K.

In this article, the Bureau of Economic Analysis (BEA) presents new estimates of gross state product (GSP) for 1995 and 1996 and revised estimates for 1977-94.(1) The new and revised GSP estimates are consistent with the estimates of gross product by industry for the Nation that were published in the November 1997 Survey of Current Business.(2) The GSP estimates incorporate the results of the most recent annual revisions of State personal income and of the national income and product accounts.(3)

The following improvements have been incorporated into the GSP estimates as part of BEA's continuing effort to update and to better integrate these estimates with the national estimates of gross product originating (GPO) by industry and the national input-output accounts:(4)

* Data from BEA's 1992 benchmark input-output accounts were incorporated into the estimates of purchased services in manufacturing for 1988-96.(5)

* New State source data on natural gas output from the Department of Energy were incorporated into the estimates for transportation and public utilities for 1992-96.

* Newly available State source data on sales, on sales taxes, and on gross receipts taxes were used in the allocations of national commodity taxes by industry for 1977-96.

* New source data were incorporated into the State estimates for mining for 1993-94.

GSP for each State is derived as the sum of the gross state product originating in all industries in the State. In concept, an industry's GSP, or its value added, is equivalent to its gross output (sales or receipts and other operating income, commodity taxes, and inventory change) minus its intermediate inputs (consumption of goods and services purchased from other U.S. industries or imported). Thus, GSP is the State counterpart of the Nation's gross domestic product (GDP). In practice, GSP and GPO estimates are measured as the sum of the distributions by industry of the components of gross domestic income--that is, the sum of the costs incurred and incomes earned in the production of GDP.(6)

The GSP estimates are prepared for 63 industries (see appendix A). For each industry, GSP is presented in three components: Compensation of employees, indirect business tax and nontax liability, and "property-type income."(7) The relationship between these components and the components of GPO and GDP is shown in appendix B.

The estimates of GSP are prepared in current dollars (see table 6, which follows the text) and in chained (1992) dollars (see table 7). State estimates of GSP and its components for all industries are "controlled" to national totals of GPO and its components for all industries.(8) The estimates of real GSP are derived by applying national implicit price deflators to the current-dollar GSP estimates for the 63 detailed industries. Then, the same chain-type index formula used in the national accounts is used to calculate the estimates of total real GSP and real GSP at a more aggregate industry level.(9)

[TABULAR DATA 6 and 7 NOT REPRODUCIBLE IN ASCII]

Real GSP is an inflation-adjusted measure of each State's output that is based on national prices for the goods and services produced within that State. Real GSP may include a substantial volume of output that is sold to other States and countries. To the extent that a State's output is produced and sold in national markets at relatively uniform prices (or sold locally at national prices), GSP does a reasonable job of capturing the differences across States that reflect the relative differences in the mix of goods and services that the States produce. However, real GSP does not capture geographic differences in the prices of goods and services that are produced and sold locally.

The first part of this article discusses the relative performance of various States and regions in terms of growth rates, industry shares of State totals, shares of the Nation, and per capita GSP. The second part discusses the revisions to the GSP estimates and the major sources of the revisions.

Growth Rates, Shares, and Per Capita GSP

Comparisons of GSP growth rates and shares of GSP across industries or States and of per capita GSP across regions provide indications of the relative performance of industries, States, or regions. For example, comparing the growth rate of real GSP for an industry with the growth rate of total real GSP indicates whether that industry is raising or lowering the State's growth rate. Comparing the share of total GSP in current dollars that is accounted for by the GSP of an industry over time indicates whether that industry's claim on the State's resources is increasing or decreasing. Comparing per capita GSP for a region with per capita GSP for the Nation over time provides an assessment of the long-term trends in the relative economic performances of regions.

Real growth rates

The rate of growth in real GSP for the Nation in 1995-96 was 3.2 percent (table 1).(10) GSP increased in all States except Alaska and Hawaii.

[TABULAR DATA 1 NOT REPRODUCIBLE IN ASCII]

By State, the growth rates ranged from an increase of 8.3 percent in Utah to a decline of 5.6 percent in Alaska. The five States with the fastest rates of growth in red GSP were Utah (8.3 percent), Nevada (7.8 percent), Delaware (6.3 percent), Oregon (5.9 percent), and New Hampshire (5.9 percent) (chart 1).

[CHART 1 ILLUSTRATION OMITTED]

In Utah, the major contributors to the growth in real GSP were finance, insurance, and real estate, mainly depository institutions; services, mainly business services; and manufacturing, mainly industrial machinery and equipment (table 2).(11) In Nevada, the major contributors were construction, retail trade, and mining, mainly metal mining. In Delaware, the major contributors were finance, insurance, and real estate, mainly depository institutions, real estate, and insurance carriers; and services, mainly health, business, and "other" services. In Oregon and New Hampshire, the major contributors were manufacturing, mainly electronic and other electric equipment; and services, mainly business, health, and "other" services.

[TABULAR DATA 2 NOT REPRODUCIBLE IN ASCII]

The five States with the slowest rates of growth in real GSP were Alaska (-5.6 percent), Hawaii (-0.1 percent), Rhode Island (0.2 percent), Wyoming (0.7 percent), and Idaho (1.6 percent). In Alaska, the major contributors to the decline were mining, mainly oil and gas extraction; transportation and public utilities, mainly pipelines, except natural gas; and government, mainly State and local government. In Hawaii, the major contributors to the decline were construction and government, mainly State and local government. In Rhode Island, the major contributor to the slow growth was finance, insurance, and real estate, mainly depository institutions. In Wyoming, the major contributors to the slow growth were mining, mainly oil and gas extraction; and agriculture, forestry, and fishing, mainly farms. In Idaho, the major contributor to the slow growth was agriculture, forestry, and fishing, mainly farms.

Shares of current-dollar GSP

Industry shares.--In 1995-96, the share of U.S. current-dollar GSP accounted for by private services-producing industries increased 0.2 percentage point, from 62.9 percent to 63.1 percent (table 3).(12) The share accounted for by private goods-producing industries was unchanged, at 24.7 percent.(13) The share accounted for by government declined 0.2 percentage point, from 12.4 percent to 12.2 percent.(14)

[TABULAR DATA 3 NOT REPRODUCIBLE IN ASCII]

By State, the change in the share of the private services-producing industries ranged from increases of more than 1.0 percentage point in Delaware, Utah, and Idaho to a decline of nearly 2.0 percentage points in North Dakota. In Delaware, the largest increase in share was in services, mainly business and health services; in Utah, the largest increase was in finance, insurance, and real estate, mainly depository institutions; and in Idaho, the largest increase was in wholesale trade. In North Dakota, the largest declines were in transportation and public utilities, mainly electric, gas, and sanitary services, and in finance, insurance, and real estate, mainly depository institutions.

The changes in the share of the private goods-producing industries ranged from an increase of 3.0 percentage points in North Dakota to declines of more than 1.0 percentage point in Idaho and Delaware. In North Dakota, the largest increase was in agriculture, forestry, and fishing, mainly farms. In Idaho and Delaware, the largest declines were in durable goods manufacturing; in Idaho, the decline was mainly in electronic and other electric equipment, and in Delaware, the decline was mainly in motor vehicles and equipment.

For government, the change in the share ranged from an increase of 0.2 percentage point in West Virginia to declines of more than 1.0 percentage point in North Dakota and South Dakota. The increase in share in West Virginia was mainly in Federal civilian and State and local government. The declines in North Dakota and South Dakota were mainly in Federal civilian and State and local government.

State shares.--Chart 2 shows the relative size of the various State economies in terms of each State's share of current-dollar GSP and of personal income for the Nation. The 14 States that have the largest GSP together accounted for nearly two-thirds of the U.S. total; the five largest States are California (12.6 percent), New York (8.0 percent), Texas (7.2 percent), Illinois (4.9 percent), and Florida (4-7 percent). The 20 States that have the smallest GSP--mostly States in the West and in New England--together accounted for less than 10 percent of the U.S. total. This pattern of the relative size of the State economies that is based on GSP is paralleled by the pattern of the relative size of the State economies that is based on personal income.

[CHART 2 ILLUSTRATION OMITTED]

Per capita GSP, 1979-96

From 1989 to 1996, per capita GSP as a percentage of the U.S. average increased from go percent to 93 percent in the low-income regions (the Southeast, Southwest, Rocky Mountain, and Plains regions) and declined from 108 percent to 106 percent in the high-income regions (the New England, Mideast, Far West, and Great Lakes regions) (chart 3). In contrast, from 1979 to 1989, per capita GSP as a percentage of the U.S. average declined in the low-income regions and increased in the high-income regions. This pattern of regional convergence in the 1990's after divergence in the 1980's is also evident in the estimates of per capita income.(15)

[CHART 3 ILLUSTRATION OMITTED]

Revisions to the Estimates

In general, the revisions to GSP as a percentage of the previously published estimates for 1977-94 are small. However, the revisions for 1992-94 are larger than those for 1977-91 because of larger revisions to the more recent source data.

Impact of the revisions

Current-dollar estimates.--For 1994, the five States with the largest upward percentage revisions were New Mexico, Louisiana, Indiana, Vermont, and Connecticut (table 4). The five States with the largest downward percentage revisions were Delaware, Wyoming, Hawaii, Alaska, and New York. For all 10 States, the revisions mainly reflect the statistical changes incorporated into the current-dollar estimates of GSP for these industries: Mining in Alaska, Louisiana, and Wyoming; transportation and public utilities in Wyoming and Alaska; manufacturing in New Mexico, Indiana, Vermont, and Connecticut; finance, insurance, and real estate in Delaware, Hawaii, New York, and Connecticut; and services in Connecticut and New York.

[TABULAR DATA 4 NOT REPRODUCIBLE IN ASCII]

Real growth rates.--For 1993-94, the States with the largest upward revisions in the growth rates of real GSP were Alaska, New Mexico, Louisiana, and Illinois (table 5). The four States with downward revisions were Delaware, Wyoming, Hawaii, and Mississippi. The revisions for all these States mainly reflect the incorporation of statistical changes into the current-dollar estimates.

[TABULAR DATA 5 NOT REPRODUCIBLE IN ASCII]

Major sources of the revisions

For the industries that had a major impact on the States with large revisions to current-dollar GSP, the sources of the revisions were either revisions to the national estimates Of GPO by industry or revisions to the State source data.

For finance, insurance, and real estate and for the transportation portion of transportation and public utilities, the revisions mainly reflect the incorporation of the revised estimates of national GPO for these industries.

For the gas utilities portion of transportation and public utilities, State estimates for 1992-96 of property-type income are now based on data for the volume of interstate natural gas movements and on data for deliveries of natural gas to final consumers from the Department of Energy (DOE). Previously, the estimates for 1992-94 were based on State data for wage and salary disbursements.

For services, the revisions mainly reflect the incorporation of the revised estimates of proprietors' income by State.(16)

For manufacturing, estimates of the cost of purchased services for 1992 are now based on the 1992 benchmark input-output accounts (on a Standard Industrial Classification basis) and on wage and salary disbursements by State.(17) Bemuse of a lack of source data for 1988-91, the estimates of the cost of purchased services for these years are derived from "straight-line" interpolations between the 1987 and the 1992 estimates of the ratio of purchased services to value-added-in-production. The estimates for 1993-96 are extrapolations that used new data on value-added-in-production by State from the Census Bureau.(18)

For the coal mining and the oil and gas extraction portions of mining, GSP estimates for 1993-96 continue to be extrapolated from the 1992 benchmark year estimates, using data on value of production that is calculated from DOE source data by multiplying the quantity produced by the average price. When DOE suppresses the price data in order to protect the confidentiality of the data, BEA now uses GSP prices to develop substitute estimates for the suppressed data; previously, DOE price data for adjacent States were used.

Tables 5-7 and appendix A and B follow.

Acknowledgments

The estimates of gross state product (GSP) were prepared by staff in the Regional Economic Analysis Division under the direction of John R. Kort, Chief, and George K. Downey, Chief of the Gross State Product by Industry Branch. Hugh W. Knox, Associate Director for Regional Economics, provided general guidance.

Contributing staff members were Richard M. Beemiller, Gerard P. Aman, Michael T. Wells, Clifford H. Woodruff III, and Tasie Anton. Michael T. Wells prepared the box on, and the calculations of, the industry contributions to changes in real GSP.

(1.) For the previously published estimates of GSP, see Howard L. Friedenberg and Richard M. Beemiller, "Comprehensive Revision of Gross State Product by Industry, 1977-94," Survey of Current Business 77 (June 1997): 15-41.

(2.) See Sherlene K.S. Lum and Robert E. Yuskavage "Gross Product by Industry, 1947-96," Survey 77 (November 1997): 20-34.

(3.) See Wallace K. Bailey, "State Personal Income. Revised Estimates for 1958-96," Survey 77 (October 1997): 24-43; and Robert P. Parker and Eugene P. Seskin, "Annual Revision of the National Income and Product Accounts: Survey 77 (August 1997): 6-35.

(4.) For a discussion of the relationship between the estimates of GPO by industry and the estimates of value added by industry from the 1992 benchmark input-output accounts, see "Note on Alternative Measures of Gross Product by Industry." Survey 77 (November 1997): 84-45.

(5.) Purchased services are subtracted from the estimates of value-added-in-production by State from the Bureau of the Census. BEA's definition of value added differs from that of the Census Bureau; BEA's definition excludes the cost of purchased services, includes sales, excise, and other indirect business taxes, and reflects inventory change valued at replacement cost. Manufacturing is the only industry for which information from the 1992 benchmark input-output accounts is used to estimate GSP.

(6) The difference between private GDP and private gross domestic Income is the statistical discrepancy. In the GSP estimates, insufficient information is available for allocating the statistical discrepancy to States. In the national estimates of GPO by Industry, the statistical discrepancy Is not allocated by industry. For more information on the statistical discrepancy, see Parker and Seskin, 19.

(7.) Property-type income is the sum of corporate profits, proprietors' income, rental income of persons, net interest, capital consumption allowances, business transfer payments, and the current surplus of government enterprises less subsidies Property-type income at both the national and State levels Includes proprietors' income as a capital share of production; however, an unknown portion of proprietors' income represents the labor share of production.

(8.) If the initial sum of the State estimates differs from the national total for an industry, the difference between the national total and the sum-of-State total is allocated to the States.

(9.) For additional information, see J. Steven Landefeld and Robert P. Parker, "BEA's Chain Indexes, Time Series, and Measures of Long-Term Economic Growth," Survey 77 (May 1997): 58-68 and Friedenberg and Beemiller, 28-29.

(10.) The rate of growth in real GDP--BEA'S featured measure of U.S. output--was 2.8 percent in 1995-96. GSP for the Nation differs from GDP because GSP, like GPO, is derived from gross domestic income (GDI), which differs from GDP by the statistical discrepancy. In addition, GSP excludes, and GDP and GPO include, the compensation of Federal civilian and military personnel stationed abroad and government consumption of fixed capital for military structures located abroad and for military equipment, except office equipment. Finally, GSP and GDP may differ because of differences in revision schedules. For an accounting of the differences (in current dollars) in 1996 between GSP for the Nation and GPO, GDP, and GDI, see appendix B. For a discussion of the relationship between GPO and GDP see Lum and Yuskavage, 20.

(11.) The GSP estimates in chained (1992) dollars are usually not additive for periods other than the base year for example, we the residual "not allocated by industry" in table 7, which is calculated as the difference between the sum of the industry detail of real GSP and total real Gap. The value of "not allocated by industry" reflects the nonadditivity of detailed real GSP estimates that results from the formula used to calculate real GSP; it also reflects the nonadditivity of detailed GPO estimates that results from differences in the national source data (in both current dollars and prices) that are used to estimate GPO by industry and the expenditures measure of real GDP.

As one moves further from the base year, the residual tends to become larger, and using chained-dollar estimates to calculate component contributions to red growth may be misleading. In table 2, an exact formula for attributing GSP growth to the industries is used, so these estimates provide accurate measures of the contributions of the industries to the percentage change in real GSP for 1995-96. See the box "Calculation of Industry Contributions to changes in Real GSP."

(12.) Private services-producing industries are defined to consist of transportation and Public utilities; wholesale trade; retail trade, finance, insurance, and real estate; and "services."

(13.) Private goods-producing industries are defined to consist of agriculture, forestry, and fishing; mining; construction; and manufacturing.

(14.) A decline in share does not necessarily indicate a decline in the level of GSP. For example, the share of government declined, but GSP for government increased $31.8 billion (see table 3).

(15.) See Duke Tran, "Personal Income and Per Capita Personal Income by State and Region, 1997," SURVEY 78 (May 1998): 11.

(16.) See Bailey, 25-26. The incorporation of the revised estimates led to revisions in proprietors' income for 1994 that ranged from 10.4 percent in Connecticut to -9.0 percent in Alaska.

(17.) See footnote 5 and see Ann M. Lawson, "Benchmark Input-Output Accounts for the U.S. Economy, 1992," Survey 77 (November 1997): 36-85. The incorporation of the new estimates of the cost of purchased services resulted in revisions to manufacturing GSP for 1992 of less than 3.0 Percent in most States.

(18.) The new data are from the annual survey of manufactures that was released in 1997.

RELATED ARTICLE: Calculation of Industry Contributions to Changes in Real GSP

For current-dollar GSP, the sum of the component industries is equal to total GSP so the changes in total GSP can be expressed as the sum of the changes in the component industries. This relationship can also be expressed in terms of percent changes by dividing by total GSP for year t - 1:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

where [GSP.sub.Total,t] is the value for total GSP in year t, and [C.sub.i,t] is the value of GSP for component industry i in year t.

However, for real GSP, a modified formula is used because chained-dollar measures of GSP by industry are not necessarily additive to total GSP Specifically, an industry's contribution to the total percent change in real GSP for years following the base year is calculated as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

where [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] is the implicit price deflator for total GSP in year t,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] is the chained-dollar value for total GSP in year t - 1,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] is the Paasche price index for total GSP in year t,

[p.sub.i,t] is the price index for industry i in year t, and

[q.sub.i,t] is the quantity produced by industry i in year t.

This formula can be expanded to yield

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

Because the State variables that represent the composites of prices in 1 year and quantities in another (such as [p.sub.i,t-1][q.sub.i,t]) are not directly observable, the contributions to growth are actually calculated using the following algebraically equivalent formula, which consists of combinations of prices and quantities for the same year and indexes of relative prices for years t and t - 1:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

This formula changes somewhat when the contributions to growth for years preceding the base year are calculated.

For more information, see "A Guide to the NIPA's," Survey of Current Business 78 (March 1998): 38-39.

RELATED ARTICLE: Data Availability

This article presents summary estimates of gross state product (GSP) by major industry group. The following GSP estimates for 63 industries for 1977-96 for States, BEA regions, and the United States are available from BEA on diskette: Current-dollar estimates of GSP and its three components--compensation of employees, indirect business tax and nontax liability, and property-type income--and real GSP estimates in chain-type quantity indexes, in chained (1992) dollars, and in fixed-weighted (1992) dollars (two diskettes)--product number RDN-0197, price $40.00.(1) The diskettes include a Windows program so that selected records from the data files can imported into computer spreadsheets.

To order using Visa or MasterCard, call the BEA Order Desk at 1-800-704-0415 (outside the United States, call 202-606-9666). To order by mail, send a check made payable to "Bureau of Economic Analysis, BE-53" to BEA Order Desk, BE-53, Bureau of Economic Analysis, U.S. Department of Commerce, Washington, DC 20230.

The GSP estimates and other regional economic information are available on BEA's Web site at <http://www.bea.doc.gov>. They are also available by subscription from the Commerce Department's STAT-USA on the Economic Bulletin Board and the Internet; to subscribe, call 202-482-1986, or go to <http://www.sta-usa.gov>.

For further information, E-mail <gspread@ bea.doc.gov>, or call 202-606-5340.

(1.) The GSP estimates will also be available on the State Personal Income CD-ROM that is scheduled to be released in the fall of 1998.