Economic statistics are often revised to reflect the availability of new data, to incorporate new estimation techniques, to reflect a change in definitions, or to make different measures more compatible. In December 1991, the Bureau of Economic Analysis (BEA), an agency within the Department of Commerce, began to emphasize gross domestic product (GDP) over gross national product (GNP) as the most comprehensive measure of production in the U.S.

  The difference between GNP and GDP lies in the treatment of income from foreign sources. GNP measures the value of goods and services produced by U.S. nationals, while GDP measures the value of goods and services produced within the boundaries of the U.S. For example, all the income received by a U.S. proprietor owning factories in the U.S. and Mexico (ie. Ford Motor Co.) would be counted in GNP. GDP would include income from the U.S. factory, but exclude the income received from the factory in Mexico. At the same time, the profits earned by a Japanese-owned corporation in the U.S. (ie. Nissan) would be included in GDP but excluded from GNP. Over the last 3 years, GNP has been larger than GDP since U.S. nationals received more income from abroad than foreign nationals received from the U.S. You have never heard this on the news have you? The difference has been small however, amounting to about 0.2 percent of GDP in 1991. For many countries the difference is much larger. In Germany, for example, GNP is about 8.6 percent larger than GDP. What does this tell you? Does it say that Germany has exported more jobs from her boundaries than the U.S.? Does it say that the U.S. has exported about as many jobs to other countries as has been imported to the U.S.? A little different picture than the ones painted by the Textile Manufacturers Association and U.S. labor unions, don't you think.

  There are two major reasons for the switch from GNP to GDP. First, because GDP measures economic activity in the U.S., it more closely parallels other measures such as employment or industrial production, which do not distinguish among the nationalities of the employer or producer. Second, GDP is emphasized by the United Nations' System of National Accounts, which many other countries use as their framework for reporting and gathering statistics. Comparing U.S. performance with other countries' performance is easier when the standards are the same.

  Production, sales, income, and employment statistics tell analysts how well the economy is performing. But for advance reports on how the economy is likely to perform, analysts turn to a set of statistics called leading indicators.

  One example of a leading indicator is building permits, which are required before construction begins on a house or apartment building. Other statistics also serve as leading indicators. Orders for expensive consumer goods (durable good orders) or orders for new business equipment usually precede production and purchases of those goods. Employers tend to increase current employees' work hours before hiring new workers.

  Several leading indicators are often combined to form a leading index: a single composite statistical series that can give advance warning about business cycle movements. Combining several indicators helps keep temporary developments in one part of the economy from giving a false impression about where the overall economy is headed.

  Correctly interpreting movements in a leading index is difficult. Variation in lead times is a major problem, both for a single indicator and for a composite index. For example, building permits do not always lead housing starts by the same number of months. The more the lead time varies, the more difficult prediction becomes. When several series are combined, the problem is compounded because each series has its own varying lead time.

  Analysts may use average lead-time relationships, but to make a useful prediction they must then also incorporate factors specific to the given situation that may differ from the average. For example, the lead time might shorten if business invested in plant and equipment more rapidly than usual in order to take advantage of a temporary tax cut.

  The best known leading index is constructed by the Bureau of Economic Analysis in the Department of Commerce, and is released in the first few days of each month. The index is composed of 11 separate leading indicators, including a stock price average, a measure of raw materials prices, building permits, consumer expectations, new claims for unemployment insurance, the average work week in manufacturing, the inflation-adjusted money supply, and four measures of new orders and orders backlog.

  Averaged over all post World War II recessions (except the 1990-91 recession), the Commerce leading index in general showed significant decline more than 9 months before the overall economy entered a recession. However, the lead time varied from a 2 month warning to a 20 month warning. Moreover, there were at least three times when the leading index declined substantially but no recession followed.

  The index has given less advance warning of business cycle recoveries than downturns. On average, the index has turned up between 4 and 5 months before the overall economy began to recover.

 The leading index did not begin to decline substantially until July 1990, giving no advance signal of the 1990-91 recession. However, the index was virtually unchanged from the middle of 1989 through the middle of 1990, which many analysts interpreted as a sign of an upcoming slow-growth period.

  In February 1991, the index began to rise consistently, prompting predictions and presidential declarations of recovery in the second half of the year. The index slid in the last 4 months of that year, however, at the same time the recovery faltered.