This technical note provides background information about the source data and
estimating methods used to produce the estimates presented in the GDP news release.
The complete set of estimates for the fourth quarter is available on BEA's Web site at
www.bea.gov; a brief summary of "highlights" is also posted on the Web site. In a few
weeks, the estimates will be published in BEA's monthly journal, the Survey of Current
Business, along with a more detailed analysis of the estimates ("GDP and the
Economy").

Real GDP

Real GDP decreased 0.1 percent (annual rate) in the fourth quarter of 2012, following
an increase of 3.1 percent in the third quarter. The downturn in real GDP in the fourth
quarter reflected downturns in inventory investment, in federal government spending,
in exports, and in state and local government spending that were partly offset by an
upturn in nonresidential fixed investment, a larger decrease in imports, and an
acceleration in consumer spending.

Source Data for the Advance Estimate

The advance GDP estimate for the fourth quarter of 2012 is based on source data that
are incomplete and subject to revision. Three months of source data were available for
consumer spending on goods; shipments of capital equipment; motor vehicle sales and
inventories; durable manufacturing inventories; federal government outlays; and
consumer, producer, and international prices. Only two months of data were available for
most other key data sources; BEA’s assumptions for the third month are shown in table
A. Among those assumptions are the following:

*	an increase in nondurable manufacturing inventories,
*	an increase in non-motor-vehicle merchant wholesale and retail inventories,
*	a decrease in exports of goods, excluding gold, and
*	a decrease in imports of goods, excluding gold.

Federal government spending

Fourth-quarter federal government spending decreased at a 15.0 percent annual rate,
reflecting a large decrease in national defense spending. The decrease in national
defense spending is based on the Monthly Treasury Statement (MTS) for October,
November, and December from the Department of the Treasury, which shows a large
decrease in fourth-quarter outlays for Department of Defense-military programs other
than for military personnel. (The MTS shows a fourth-quarter increase in outlays for
military personnel, but that increase reflects special factors such as once-a-year lump
sum payments that BEA distributes across the quarters of the year, and an extra pay
day that BEA adjusts for in preparing accrual-based estimates.)

Effects of the Midwest Drought

BEA’s GDP estimates reflect the continuing effects of this summer’s extreme hot
weather and drought in the Midwest on farm production. For the most part, the effects
are embedded in the regular source data that are used by BEA and cannot be
separately identified in components such as personal consumption expenditures or
exports of goods.

The farm inventory investment estimates reflect the effects of the drought on farm
production, particularly for corn and soybeans. The estimates for the fourth quarter of
2012 are based on farm income statistics from the U.S. Department of Agriculture
released in November and on crop production reports for corn and soybeans released
through January. In current dollars, the estimates indicate that the drought reduced farm
inventory investment by about $24 billion in the fourth quarter, after reducing farm
inventory investment by about $28 billion in the third quarter and $12 billion in the
second quarter. Adjusting for inflation, the change in farm inventories added 0.11
percentage point to the fourth-quarter change in real GDP after subtracting 0.38
percentage point from the third-quarter change and 0.17 percentage point from the
second-quarter change.

For additional information about the methodology used to estimate the impacts of the
drought, see the box “Effects of the 2012 Midwest Drought on the NIPA Estimates” in
the October 2012 Survey of Current Business.

Prices

The price index for gross domestic purchases—the prices paid by U.S. residents for
goods and services wherever produced—increased 1.3 percent in the fourth quarter
after increasing 1.4 percent in the third. Excluding food and energy prices, the price
index for gross domestic purchases increased 1.1 percent after increasing 1.2 percent.

Disposable Personal Income, Special Dividends, and Accelerated Compensation

Real disposable personal income (DPI) increased 6.8 percent in the fourth quarter,
following an increase of 0.5 percent in the third. The acceleration in real DPI reflected
accelerated and special dividends that were paid by many companies late in the quarter
in anticipation of changes in individual income tax rates, as well as some acceleration in
wages and salaries.

Based on reports from more than 2,000 companies from Compustat, BEA estimates that
companies paid special or accelerated dividends of $39.5 billion in the fourth quarter, of
which $26.4 billion ($105.6 billion at an annual rate) was paid to persons and is included
in personal income.

In addition to the special dividends, there is evidence that some companies accelerated
their payment of bonuses or other types of irregular pay. In the absence of source data
that cover these types of payments, BEA made a judgmental adjustment to its estimates
of fourth-quarter wages and salaries, adding an additional $15 billion (annual rate) to the
regular estimate of wages and salaries. In May, BEA will incorporate fourth-quarter
source data from the BLS quarterly census of employment and wages, which covers
irregular pay.

The personal saving rate was 4.7 percent in the fourth quarter, compared with 3.6
percent in the third.

Hurricane Sandy

On October 29, Hurricane Sandy made landfall on the New Jersey coast and caused
major damage and disruption throughout the Northeast region. Like other disasters,
Hurricane Sandy disrupted many types of production through closures of factories,
offices, and transportation facilities, while causing certain types of production, such as
emergency services and rebuilding activities, to increase. These effects on production
are included, but not separately identified, in the source data that BEA uses to prepare
the estimates of GDP; consequently, BEA is not able to provide an estimate of the
overall impact of Sandy on fourth-quarter GDP.

While the destruction of fixed assets, such as residential or nonresidential structures,
does not directly affect GDP, national income, or personal income, BEA does track
disaster losses as part of its fixed asset accounts. BEA’s preliminary estimate is that
$35.8 billion in private fixed assets and $8.6 billion in government fixed assets,
measured at current cost, were lost due to Hurricane Sandy.

BEA also prepares estimates of the insurance benefits paid or received due to major
disasters. These benefits are recorded on an accrual basis in the quarter in which the
disaster occurs and are classified as capital transfers, and thus do not affect the
measures of GDP, personal income, or saving. BEA’s preliminary estimate is that
domestic and foreign insurance companies expect to pay benefits for losses related to
Sandy of $20.6 billion ($82.5 billion at an annual rate), and that the federal government’s
National Flood Insurance Program is expected to pay an additional $7.5 billion ($30.0
billion at an annual rate) in Sandy-related benefits.

For additional information, see “How are the measures of production and income in the
national accounts affected by a natural or man-made disaster?”

Brent R. Moulton
Associate Director for National Economic Accounts
Bureau of Economic Analysis
(202) 606-9606

• Entire Note and Table (PDF)
• Entire Note and Table (Text)
• Table Only (XLS)

• Interactive Tables