Trade Deficits and Jobs

August 2008

Russell Roberts roberts@gm u.edu George M ason University

http://www.cafehayek.com http://www.econtalk.org http://www.m ercatus.org

www.mercatus.org A Persistent and Growing Merchandise Trade Deficit

U.S. Merchandise Trade Balance, 1960-2005 (billions of dollars) 1968 1970 1986 1972 1988 1974 1990 1960 1976 1992 1962 1978 1994 1964 1980 1996 1966 1982 1998 1984 2000 2002 2004 100

-100

-200

-300

-400

-500

-600

-700

-800

-900

Source: BEA, http://bea.gov/bea/di/table1.xls

• Merchandise trade balance—exports of goods minus imports of goods

• I have tried to make it as scary as possible:

• Excludes services (America exports more services than it imports since 1971) • Not deflated by either the price level or the size of the economy

• Until 1976, roughly zero—sometimes positive or sometimes negative but always somewhat small.

• Beginning in 1976, U.S. has run a trade deficit every single year

• Since 1976, U.S. has imported $6 TRILLION worth of goods more than it has exported.

2 Little or No Impact of Trade Deficit on Jobs

U.S. Total Employment, 1939-2005 (millions of jobs) 1949 1947 1961 1945 1975 1943 1959 1973 1957 1987 1941 1971 1985 2001 1939 1955 1969 1999 1953 1983 1967 1981 1997 1951 1965 1995 1979 1993 1963 1977 1991 1989 2005 2003 160

140

120

100

Source: BLS, Current Employment Statistics, Series Id: CEU0000000001

• No obvious impact of trade deficits on employment

• Recessions reduce number of jobs—otherwise, steady growth

• In 2005, over 40 million more jobs than in 1976, the beginning of persistent, growing trade deficits

• Trade affects the kind of jobs in the economy, not the numbers of jobs

• Economy produces jobs for people who want them. Population is higher but the proportion of the population working is also higher after 1976 than before.

3 What About Manufacturing Jobs?

Manufacturing Employment, 1939-2005 (millions of jobs)

0 1987 1989 1991 1977 1979 1981 1983 1969 1985 1971 1973 1959 1975 1961 1963 1965 1951 1967 1953 1955 1995 1941 1993 1957 1997 1943 1999 1945 2001 1947 2005 2003 1949 1939

Source: BLS, Current Employment Statistics Survey, Series ID: CEU3000000001

• No apparent difference between pre-1976 and post-1976

• Manufacturing employment surged during WWII, Korean War and Vietnam War

• Between 1965 and 2000, fairly stable fluctuating between 16 and 20 million jobs

4 But What if We Control for the Size of the Labor Force?

Manufacturing Jobs as a Proportion of Total Jobs, 1976-2005

0.25

0.2

0.15

0.1

0.05

0 1987 1983 1979 1996 1992 1988 1984 1980 1976 1997 1993 1989 1985 2000 1981 2003 1977 2001 2004 2002 2005 1998 1994 1990 1986 1982 1978 1999 1995 1991

Source: BLS, Table B-1, Employees on Non-Farm Payrolls by Industry Sector, from the Current Employment Statistics Survey, Series Id’s: CEU0000000001 and CEU3000000001

• Manufacturing jobs as a proportion of total employment HAVE fallen steadily since 1976

5 BUT THAT TREND BEGAN LONG BEFORE 1976

Manufacturing Jobs as a Proportion of Total Jobs, 1939-2005

0.4

0.35

0.3

0.25

0.2

0.15

0.1

0.05

0 1951 1949 1947 1969 1945 1967 1943 1965 1985 1941 1963 1983 1939 1961 1981 1959 1979 1957 1977 1955 1975 1997 1953 1973 1995 1999 2001 2003 2005 1993 1971 1991 1989 1987

Source: BLS, Table B-1, Employees on Non-Farm Payrolls by Industry Sector, from the Current Employment Statistics Survey, Series Id’s: CEU0000000001 and CEU3000000001

• Ratio peaked in 1944 and has been declining long before persistent running trade deficits and the growth in globalization.

• Something else is causing this long-term trend, something that is unrelated to globalization.

6 Productivity Gains in Manufacturing

Manufacturing Output, 1959-2005 (2002=100)

120.0

100.0

80.0

60.0

40.0

20.0

0.0 1987 1978 1988 2005 1972 1982 2004 1998 2003 2001 1997 2000 2002 1979 1994 1999 1996 1963 1973 1983 1993 1990 1995 1967 1992 1989 1991 1980 1964 1974 1984 1968 1981 1965 1975 1985 1959 1969 1966 1976 1986 1960 1970 1977 1961 1971 1962

Source: Table B-51, Economic Report of the President, from Board of Governors of the Federal Reserve

• America is not being “hollowed out.” The manufacturing sector is strong.

• Manufacturing output falls with recessions but otherwise rises steadily over time

• Since 1959, the economy is 4.5 times bigger as measured by real GDP. But manufacturing output is 4.7 times bigger over the same time period. America produces 4.7 times more stuff today than in 1959 with FEWER manufacturing workers

• Increased productivity (better educated workers working with more sophisticated machinery) is the dominant cause of the reduction in manufacturing employment

7 Do Trade Surpluses CREATE Jobs?

Agricultural Trade Balance, 1935-2005 (billions of dollars) 1999 1981 1987 1969 1997 1979 1985 1967 1995 1949 1977 2005 1983 1937 1965 1993 1947 1975 2003 1957 1935 1963 1991 1945 1973 2001 1955 1961 1989 1943 1971 1953 1959 1941 1951 1939 30

-5

Source: Economic Research Service, U.S. Department of Agriculture

• The U.S. has exported more dollars worth of food than it has imported, a trade surplus in food, every year since 1963.

• If you argue that deficits cause job loss, you have to argue that a surplus should create jobs. But there are fewer than half the number of workers in the agricultural sector than there were in 1963, despite increases in population and increases in labor force participation that have doubled the overall labor force.

• The decline in the importance of agriculture as a source of employment is caused by the same thing reducing manufacturing employment: productivity. We don’t need as many people to produce a particular quantity of food.

• Trade surpluses don’t create jobs. Deficits don’t destroy jobs.

8 Trade Surpluses Don’t Create Jobs

Agricultural Employment, 1948-2005 (millions of workers)

0 1948 1956 1962 1972 1950 1960 1970 1980 1954 1964 1974 1984 1952 1958 1968 1978 1988 1996 1994 1966 1976 1986 1998 2002 2000 2004 1982 1992 1990

Agricultural Employment as a Percentage of Total Employment, 1948-2005

0 1988 2002 1984 1998 1980 1994 1990 1968 1974 1964 1970 1976 1954 2004 1960 1966 1962 1972 1950 1956 1952 1958 1986 1948 2000 1982 1996 1978 1992

Source: Bureau of Labor Statistics, Current Population Survey, Tables A-1 and A5

9 Why Don’t Trade Deficits Destroy Jobs?

Why would a trade deficit destroy jobs? The argument is that imports destroy jobs and exports create jobs. So if imports exceed exports, there will be net job destruction. This mechanical approach to job creation ignores the dynamic nature of the job market.

Consider a world where every American wakes up to find a free car in the driveway, a gift from the Japanese auto industry. In the glove compartment is a note explaining that this gift will be repeated every year. In some way, this is the ultimate trade deficit—a set of imports with zero counterbalancing exports.

What will be the impact from this gift on the number of jobs in America and on America’s standard of living? It will devastate employment in the auto industry. But will total employment fall by the number of jobs lost there? A lot of industries are going to be expanding because people no longer have to pay $25,000 for a car. People will now be able to buy things they couldn’t afford to buy before the gift. So the decrease in the demand for labor is going to be offset by an increase in demand for labor in industries outside of the car market. The American standard of living will rise in exactly the same way it would if American carmakers figured out a cheaper way to make cars. Both changes— innovation or free cars from the Japanese—make Americans richer.

The same thing has happened over the last century in agriculture. As farmers have become more innovative, we get more food at lower prices using fewer workers. That creates wealth, not poverty. In 1900, agriculture employed 40% of the American work force. Today, that number is under 2%. New jobs have come along to replace the lost farming jobs. And the new jobs pay well because we don’t have to pay as much as we once did for food. It has been gloriously good for America that we don’t need as many people farming as we once did.

Would it make any difference if that decrease in farm employment had come from foreigners willing to sell us food cheaply or technological change that made agriculture more efficient? Both lead to cheaper food and fewer workers necessary to grow food in the United States. Both increase the standard of living of the average American.

Is this dynamic view of the job market accurate? Look at the data. Imports have surged over the last 50 years. The trade deficit has ballooned over the last 30 years. Yet employment has grown steadily. Banning imports would eliminate the trade deficit. But the number of jobs in America wouldn’t change—we’d just find ourselves trying to make all the cars and all the steel and all the watches that we used to import. Those industries would grow. Others would shrink because there wouldn’t be enough workers to go around and our demand for many goods would fall as cars and steel and watches became more expensive leaving less money for other things. America would be starkly poorer.

Self-sufficiency is the road to poverty. Trade lets us cooperate and allows others to make things for us that we could only make for ourselves at greater expense.

10 Is Anything Related to the Trade Deficit?

Capital Account Balance, 1960-2005 (billions of dollars) 1979 1993 1971 1985 2002 1977 1994 1969 1961 2000 1992 1967 1984 1976 1990 1968 1982 1960 1999 1974 1991 1966 2005 1983 1997 1975 1989 1981 1998 1973 1965 2004 1996 1988 1963 1980 1972 1986 1964 2003 1978 1995 1970 1987 1962 2001 900

800

700

600

500

400

300

200

100

-100

Source: BEA, Table A-1, U.S. International Transactions Accounts Data

• The capital account balance is close to a mirror image of the pattern of the merchandise trade balance

• Zero or close to zero for a long time, then persistent and growing surpluses

• A sign of the attractiveness of dollar-denominated assets as a store of value. Includes government treasuries but also corporate equity and bonds. The trade deficit is not a measure of debt or of how much America owes the rest of the world.

• A capital account surplus allows a nation to consume more than it produces—a trade deficit

• The trade deficit and the capital account surplus are determined simultaneously by a wide array of factors. Neither is the cause of the other.

• A trade surplus and a trade deficit are sustainable as long as the U.S. remains an attractive place to invest relative to the rest of the world.

11 An Economics Rorschach Test

Capital Account Balance, 1960-2005 (billions of dollars) 1963 2002 1998 1994 2003 1999 1995 1991 1987 2004 1983 2000 1979 1996 1975 1992 1971 1988 2005 1984 2001 1980 1997 1976 1993 1972 1989 1968 1985 1964 1981 1960 1977 1973 1990 1969 1986 1965 1982 1961 1978 1974 1970 1966 1962 1967 900

800

700

600

500

400

300

200

100

-100

Merchandise and Services Trade Balance, 1960-2005 (billions of dollars)

100

-100

-200

-300

-400

-500

-600

-700

-800 2005 2004 2003 2002 2001 2000 1999 1998 1997 1996 1972 1995 1994 1993 1992 1991 1973 1990 1989 1988 1987 1986 1963 1974 1985 1984 1983 1982 1964 1981 1980 1979 1975 1965 1978 1976 1966 1977 1967 1968 1969 1970 1960 1971 1961 1962

In the bottom picture, I’ve added the balance of trade in services to the earlier picture of the merchandise trade balance to illustrate how closely the capital account surplus and the trade balance in goods and services mirror each other. The attractiveness of America as a place to invest says the same thing as Americans buy more from foreigners than foreigners buy from the United States. Another way to say it is that our imports of capital and goods and services are roughly equal to our exports of capital and goods and services. The only discrepancy from this equality is currency flows.