Go to Brad DeLong's Home Page
MEMORANDUM FOR DEPUTY ASSISTANT TO THE PRESIDENT BOWMAN CUTTER
Through: Alicia Munnell, Asst. Secretary of the Treasury for Economic Policy
From: Brad DeLong, Dep. Asst. Secretary of the Treasury for Financial Analysis
Subject: ASSESSING GLOBALIZATION AS A CAUSE OF BLUE-COLLAR WAGE DECLINES: CON
"Globalization" is shorthand for the telecommunications, trade, and transport revolutions that allow enterprises to efficiently organize value chains that span continents. Henry Ford managed his automobile assembly line from Detroit, and so had to employ blue-collar workers in Detroit: his early twentieth-century span of control extended no further-and blue-collar workers in Detroit earned high wages because Ford could not play them off against potential workers in Florida, let alone workers in Thailand.
Today firms are increasingly able to spin their webs of enterprise across continents, and to move almost any component of the value chain that is labor-intensive to places where labor is cheap without losing efficiency. M.I.T. Press typesets its books in Hong Kong. Apple produces computers in Singapore (and Ireland). Swissair has announced a move of its accounting department to Bombay. But conventional wisdom is that the largest shift in relative demand for workers is a shift away from demand for high-priced U.S. blue-collar workers-that the collapse in relative blue-collar wages seen in the 1980s is the result of "globalization," and that the collapse is likely to continue in the 1990s as globalization accelerates.
Successful management of the ongoing process of "globalization" is essential if America is to fulfill its economic potential. But there is a tendency to overstate the impact that globalization has had to date-especially a tendency to incorrectly attribute a large share of the fall in the relative wages and relative numbers of blue-collar workers.
I believe that to date the decline in blue-collar wages and numbers has other causes-technology and demand. It is not due to trade or globalization, or at least a smaller component of the past decline is due to trade and globalization than the conventional wisdom thinks.
Note that I do not want to make any claims about how important "globalization" will be in the future. I am not yet confident enough that I have the pulse on long-run trends to want to commit to a position on future trends. This memo merely makes an argument that globalization--which is likely to be more important in the future--has not been very important in the past as a source of significant downward pressure on the wages and numbers of America's blue-collar workers.
Here are seven reasons why I believe that globalization has been overrated:
The U.S. is not facing increasing competition from low-wage countries. Existing U.S. trading partners have become richer-and seen their manufacturing wages increase-faster than newly-industrializing low-wage economies have been able to begin exporting to the United States. The wage level of the average import into the United States has grown from 59 percent of the U.S. level in 1975 to 73 percent of the U.S. level today.
Imports are not the principal cause of falling blue-collar wages, or falling demand for blue-collar workers. The most rapid growth in the relative share of imports occurs in the 1970s, during which blue-collar wages keep pace with productivity and with the wages of white-collar workers.
Why hasn't trade been a factor placing significant downward pressure on U.S. blue-collar wages and jobs? Because U.S. international transactions are of three types, only one of which has the potential to put significant downward pressure on the wages of blue-collar workers.
- Trades of U.S. goods for foreign goods: the decrease in blue-collar job demand because we do not produce what we import is roughly offset by the increase in blue-collar job demand by export-producing firms.
- The trade deficit component--trades in which the U.S. imports foreign goods and foreigners take the dollars from their sales and invest them in America. The decrease in blue-collar demand in industries that produce the goods we import is roughly offset. Blue-collar job demand in construction, machine tools, and other investment-goods industries rises because foreign capital finances more investment in America than would take place otherwise.
- Trades in which we import goods and export services: this is the only component that could put significant downward pressure on American blue-collar wages and jobs. Instead of managers, professionals, and technicians in Chicago employing blue-collar workers in Gary, managers, professionals, and technicians in Chicago employ blue-collar workers in Thailand. We trade the services of what Labor Secretary Reich calls "symbolic analysts" for goods produced by their blue-collar workers. But to date this component of trade is relatively small.
While in the future U.S. demand for blue-collar workers may be depressed if trade of our service exports for their manufactured products continues to grow rapidly, to date net service exports are still only a small share of national product. If we shifted the workers and resources that produce all service exports into producing goods, we would see an increase in blue-collar jobs (and a decrease in service sector jobs) of perhaps 600,000.
In 1962 36% of the workforce were blue-collar "production workers"; in 1992 26% of the labor force falls into the blue-collar category. If today we had the same occupational distribution of employment that we had in 1962, there would be 12,000,000 more blue-collar jobs. Thus "Globalization"--trading our services for their goods--is responsible for only five percent of the relative decline in the share of blue-collar jobs in our economy over the past three decades.
Moreover, most of our service-sector exports are not the products of high-paid "symbolic analysts." The largest service-export categories are travel and transportation: someone making the bed of a busload of Japanese tourists on their way to Yellowstone. Travel and transportation provide a strong demand for low-wage workers.
The shift to a post-industrial economy--the decline in blue-collar workers as a share of the non-agricultural labor force-has been underway for most of this century. It is not a new trend that has emerged in the last ten, or even the last twenty years. It is hard to blame a phenomenon that is seen only in the past decade or two-falling blue-collar wages-on trends that have been underway for the better part of a century.
The manufacturing wage in the country providing the average import to the United States has not fallen but risen over time.
In 1975 the average import to the United States came from a country where manufacturing compensation (at then-current exchange rates) was 59.6% of the United States. In 1991 the average import came from a country where manufacturing compensation was 73.9% of the United States level.
Existing U.S. trading partners have become richer--and seen their manufacturing wages increase--faster than newly-industrializing low-wage economies have been able to begin exporting to the United States.
Non-oil imports from low-wage countries are lower relative to U.S. national product than they were fifteen years ago.
Japan was our second-largest trading partner in 1975, and is our second-largest trading partner today. In 1975 Japan was a low-wage country--with manufacturing compensation 47 percent of the U.S.--and so imports from Japan did put downward pressure on U.S. wages in import-competing industries. But in 1991 Japanese manufacturing compensation was 93 percent of the U.S., and today it is higher than U.S. levels: imports from Japan no longer count as imports from places where labor is cheap.
Imports from low-wage countries as a share of GDP have shrunk because those low-wage countries today that have become export platforms to the U.S. are less important trading partners than those countries that were low-wage producers in 1975 but are high-wage producers today.
Non-oil imports from very low-wage countries are lower relative to U.S. national product than they were fifteen years ago.
Taiwan, Singapore, and Korea were very low-wage economies in 1975, with wage levels one-twentieth that of the United States. Today their manufacturing wage levels are about a third of American levels.
Thus a good deal of what would have been downward pressure on American wages had real incomes in Korea, Singapore, and Taiwan remains constant has disappeared: they are now middle-income economies, not poor economies. So U.S. firms actually face less relative competition from workers in very poor economies than they did two decades ago.
Merchandise imports as a share of total goods output rose from six percent in 1960 to around fifteen percent in 1975-a period during which blue-collar wages and average productivity grew relatively rapidly.
Since 1975 merchandise imports have grown no more rapidly than in the previous period-from fifteen to twenty-five percent of total goods output. Yet blue-collar wages, which rose rapidly before 1975, have stagnated since.
The biggest leap in the share of motor vehicles that are imported came from 1960 to 1970--from 2 to 16 percent--a decade that saw blue-collar worker wages rise rapidly.
The period since 1980, that has seen the most rapid relative decline in blue-collar wages, has seen next to no change in the import share of motor vehicle sales.
The U.S. trade surplus in services in 1991 was only $60 billion--one percent of GDP.
If we had shifted the workers and resources that produce these service exports into producing goods, we would have seen an increase in blue-collar jobs (and a decrease in service sector jobs) of perhaps 600,000.
In 1962 36% of the workforce were blue-collar "production workers"; in 1992 26% of the labor force falls into the blue-collar category. If today we had the same occupational distribution of employment that we had in 1962, there would be 12,000,000 more blue-collar jobs.
"Globalization"-trading our services for their goods-is responsible for only five percent of the relative decline in the share of blue-collar jobs in our economy over the past three decades.
Moreover, most of our service-sector exports are not the products of high-paid "symbolic analysts." The largest service-export category ($53 billion) in the table above is travel: someone making the bed of a busload of Japanese tourists on their way to Yellowstone.
Travel and transportation make up 56% of service exports, and travel and transportation have a strong demand for low-wage workers. Even expanded service exports place only limited downward pressure on employer demand for less-skilled workers.
The shift to a post-industrial economy-the decline in blue-collar workers as a share of the non-agricultural labor force-has been underway for most of this century. It is not a new trend that has emerged in the last ten, or even the last twenty years.
Fully three-fifths of non-agricultural workers were blue-collar workers in 1900; only a quarter are in this category today. But the decline in the blue-collar job share has been gradual and steady: its pace has not significantly accelerated since the 1920s.
The popular image of the non-agricultural workforce-of assembly-line workers, white-collar managers, and bosses, of large mass-production firms as the representative places in which Americans work-was never completely true. Even at the beginning of the 1930s blue-collar industrial workers made up only half of America's non-agricultural workforce, and blue-collar industrial workers in manufacturing and construction made up only 28 percent.
Go to Brad DeLong's Home Page
Professor of Economics J. Bradford
DeLong, 601 Evans
University of California at Berkeley; Berkeley, CA 94720-3880
(510) 643-4027; phone (510) 642-6615 fax