William Jefferson Clinton is the first Democratic president of the postwar era to have presided over a decline in manufacturing jobs in the American Southeast.  According to the U.S. Bureau of Labor Statistics, monthly manufacturing payroll employment increased ten percent in Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana, Mississippi, North and South Carolina, Tennessee, Virginia, and West Virginia under Harry S. Truman, 14 percent under John F. Kennedy, 27 percent under Lyndon B. Johnson, and six percent under James E. Carter but declined four percent under President Clinton.

How severe is this manufacturing contraction?  Economists disagree about the factors that caused the Great Depression, but the event taught them a great deal about how to measure contraction.  Gottfried Haberler, in his 1937 book, Prosperity and Depression  (Harvard University Press), identifies three times series—employment, production, and income—that can be analyzed to determine whether an economy or a specific private-industry sector is expanding or contracting.  Today, economists use three measures of these time series: duration, depth, and diffusion.  The period of contraction must be sufficiently long (duration), feature a substantial decline (depth), and affect a sufficient number of industries (diffusion).  Duration is measured by calculating the difference between the cyclical peak and trough; depth, by the degree of decline; and diffusion, through the use of indices that compare the percentage of sectors increasing during designated intervals with those in decline.

The duration of the ongoing contraction in Southeast manufacturing employment is the longest since the monthly data series started in 1939—in other words, since the Great Depression.  Southeast manufacturing employment peaked on Mr. Clinton’s watch in January 1995 at 4,785,100 workers.  U.S. manufacturing employment peaked more than three years later, in April 1998.  There were 4,096,900 employed in Southeast manufacturing in November 2002, which means a net loss of 688,200 manufacturing jobs in the region.  Measured from its January 1995 peak, the contraction lasted 95 months.  Before the current recession, which started in March 2001, the average duration of declines in Southeast manufacturing employment was only 14.2 months.  (These declines occurred from August 1948 to July 1949; September 1953 to July 1954; November 1956 to May 1958; September 1960 to February 1961; August 1969 to March 1971; December 1973 to March 1975; October 1979 to July 1980; September 1981 to January 1983; and June 1989 to April 1991).

After World War II, Southeast manufacturing employment grew during all ten U.S. expansions and declined in the recessions that followed.  From April 1991 to January 1995, Southeast manufacturing jobs increased 5.5 percent, the second-smallest gain of the postwar era.  By comparison, Southeast manufacturing employment expanded 48 percent under Kennedy and Johnson; 26 percent under Truman and Eisenhower; 15 percent under Nixon; 22 percent under Ford and Carter; and 15 percent under Reagan and Bush.

The depth of Clinton’s contraction is the greatest of the postwar era.  Through December 2002, Southeast manufacturing employment declined 14.7 percent from its peak.  The old record (14.4 percent) was set between December 1973 and March 1975.  Manufacturing employment fell 17 percent in the World War II demobilization (November 1943 to February 1946).

Those who downplay the significance of these declines argue that they are limited to the Southeast’s textiles sector.  Employment in textiles-producing nondurable goods—“Textile Mill Products” and “Apparel and Other Textile Products”—has fallen by more than 220,000 since 1995.  The greatest declines in employment under President Clinton occurred in North Carolina (70,900); South Carolina (43,000); Tennessee (42,500); Alabama (35,400); and Mississippi (21,800).  Yet the contraction has been more dispersed.  Some components actually started to contract in 1994.  Southeast manufacturing employment suffered seven lean years during this portion (1994-2001) of Mr. Clinton’s tenure.

The contraction’s diffusion is evident in Bill Clinton’s home state of Arkansas, where employment in 80 percent of manufacturing components declined between the state’s June 1995 peak and January 2001, when Clinton left office.  These included durable goods such as lumber and wood, furniture and fixtures, primary metals, industrial machinery, electronic equipment, and instruments and miscellaneous manufacturing; and nondurable goods such as food and kindred products, textile mill products, apparel and other textile products, paper and allied products, printing and publishing, chemicals and allied products, petroleum and coal products, and leather and leather products.  Employment grew in only three manufacturing components (fabricated metals, transportation equipment, and rubber and miscellaneous plastics) and was neutral in stone, clay, and glass.

A depression is a recession that is major in both scale and duration.  Using this definition, which is widely accepted among economists, Southeast manufacturing employment is depressed.  But what about production and income?  Some economists argue that advances in technology mean fewer manufacturing jobs are required for growth in production.  This
argument has been a staple of Federal Reserve policy statements since the mid-1990’s.  Has this occurred in the Southeast?  Only in the durable-goods manufacturing sector, where real Gross State Product (GSP) increased, but at a rate less than the U.S. average.  (GSP is the state counterpart of U.S. Gross Domestic Product.  Real data is adjusted for inflation.)  Real nondurable-goods manufacturing GSP declined in 11 of 12 states (Georgia was the exception) and fell 15 percent for the region (1995-2000).  However, Total Real Manufacturing GSP (durable and nondurable) actually declined in four states (Louisiana, Mississippi, Virginia, and West Virginia) in the period.

These long-term declines in employment and nondurable production are synonymous with depression.  Southeast manufacturing personal income (1995-2000) grew at a lower average annual rate than at any time since the 1930’s but Mississippi is the only state in which it declined.  (Nondurable income in the Carolinas also fell.)  The 2002 income data, which is scheduled to be released later in 2003, could provide more evidence.  Manufacturing income fell in 10 of 12 states in 2001.  The last time it fell in that many Southeast states was 1949.  The last time manufacturing income fell for two consecutive years was during the Great Depression.