[Economic Report of the President (2004)]
[Administration of George W. Bush]
[Online through the Government Printing Office, www.gpo.gov]

 
Chapter 2


The Manufacturing Sector



The manufacturing sector was affected by the latest economic
slowdown earlier, longer, and harder than other sectors of the
economy and only recently have manufacturing employment losses
begun to abate. Over the past several decades, the manufacturing
sector has experienced substantial output growth, even while
manufacturing employment has declined as a share of total
employment. This chapter examines recent developments and long-term
trends in manufacturing and considers policy responses.

The key points in this chapter are:

  The severity of the recent slowdown in manufacturing
was largely due to prolonged weakness in business
investment and exports, both of which are heavily
tied to manufacturing.
  The manufacturing employment decline over the past
half-century primarily reflects striking gains in
productivity and increasing consumer demand for
services compared to manufactured goods.
International trade plays a relatively small role.

  Consumers and businesses generally benefit from the
lower prices made possible by increased manufacturing
productivity, and strong productivity growth has led
to real compensation growth for workers. The shift of
jobs from manufacturing to services has caused
dislocation but has not resulted, on balance, in a
shift from "good jobs" to "bad jobs."

  The best response to recent developments in
manufacturing is to focus on stimulating the overall
economy and easing restrictions that impede
manufacturing growth. This Administration has
actively pursued such measures.

Manufacturing and the Recent Business Cycle

This section looks at the characteristics and causes of the recent
economic downturn with particular focus on the manufacturing sector.
Output in manufacturing held up relatively well in the recent
recession, but employment declined sharply. Data released over the
past few months are encouraging regarding the prospects for recovery
in the manufacturing sector.

The Recent Downturn in Manufacturing Output

Manufacturing output dropped 6.8 percent from its peak in June 2000
to its trough in December 2001. This was a larger decline than that
for real GDP, which fell only 0.5 percent from its peak in the
fourth quarter of 2000 to its trough in the third quarter of 2001.
This gap is not out of line with historical experience:
manufacturing output has dropped much more than real GDP during
past business cycles (Chart 2-1). What is more unusual is that
the recovery in manufacturing output has been far weaker than the
recovery in real GDP.

As discussed in Chapter 1, Lessons from the Recent Business Cycle,
investment demand was especially weak during the recent recession.
A slowing of demand for equipment investment disproportionately
hurts the manufacturing sector because nearly all business equipment
involves manufactured products. The rest of final demand, in
contrast, involves a mix of manufactured goods, agricultural
products, services, and structures. The industries within
manufacturing contributing most to the downturn in manufacturing
output were those primarily associated with the production of
business equipment. In particular, slower growth in production of
computers and other electronics, machinery, and metals accounts
for nearly two-thirds of the swing in manufacturing output from
its rapid growth in the late 1990s (an annual rate of 6.9 percent)
to its decline in the 18 months after






mid-2000 (an annual rate of -4.6 percent). Some parts of
manufacturing saw especially difficult times. The metalworking
machinery industry, of which the hard-hit tool and die industry
makes up 40 percent of employment, has seen its payrolls decline
by almost 25 percent from mid-2000 to the end of 2003. Real
production in the metalworking machinery industry fell by more
than 35 percent over this period.

The timing of the manufacturing slowdown also strongly suggests a
link to the decline in business investment (Chart 2-2).
Manufacturing output declined substantially in the middle of 2000,
months before real GDP turned downward around the fourth quarter
of 2000. This pattern mirrors that of business investment in
equipment and software, which also peaked in mid-2000--well before
the overall economy. The prolonged period of weakness in
manufacturing output also bears a notable similarity to the
sluggish recovery in investment in equipment and software.

Lackluster demand for U.S. exports has been another source of
weakness in the manufacturing sector over the past three years.
Exports have been depressed, in part due to slow growth in other
major economies. Since the fourth quarter of 2000, the average
annual rates of real GDP growth in the euro area and Japan have
been less than half that of the United States. Industrial supplies
and capital goods make up the bulk of U.S. goods exports. Lower
exports of manufactured goods can account for all of the decline
in exports since 2000.









Manufacturing Employment in Recent Years


Manufacturing employment declined more than manufacturing output
during the recent downturn, just as overall employment declined
more than overall output. Manufacturing employment declined 16
percent from June 2000, the peak of manufacturing production, to
December 2003--a steeper decline than in recessions on average
(Chart 2-3). In fact, the recent drop in manufacturing employment
was the largest cyclical decline since 1960.

As with the overall economy, the weakness of manufacturing
employment relative to output during and after the recent recession
has been reflected in rapid productivity growth (Chart 2-4). From
the fourth quarter of 2000 through the third quarter of 2003,
productivity in the nonfarm business sector and in the manufacturing
sector rose more than 4 percent at an annual rate--appreciably faster
than in recessions on average since 1960. This rise has allowed
businesses to increase output without a corresponding increase in
labor input.




Signs of Recovery in the Manufacturing Sector

Data for the second half of 2003 suggest a noticeable firming in
the manufacturing sector. Orders and shipments of capital goods
began to increase around the middle of 2003. Industrial production
rose at an average annual rate of 5.9 percent during the second
half of the year, the largest six-month gain since the first half
of 2000. In addition, the new orders index from the Institute of
Supply Management's monthly survey of purchasing managers rose to
its highest level in two decades, indicating widespread optimism
that activity is picking up. Moreover, some of the factors that
have historically affected firms' production decisions support a
further strengthening--the cost of capital is low by the standards
of the last decade and manufacturers' profits are well above their
levels of two years ago.

Although manufacturing employment fell throughout 2003, recent
developments hint at improving employment conditions for the
sector as a whole. To be sure, some industries continue to
lag--for example, textiles, apparel, printing, and petroleum and
coal industries have seen employment fall substantially more than
overall manufacturing employment since mid-2003. More broadly,
however, the rate of decline in overall manufacturing employment
eased noticeably in the fourth quarter of 2003, with the smallest
quarterly loss in three years. In addition, the rise in
temporary-help services since the spring of 2003 is consistent with
a future rebound in permanent employment. The temporary-help sector
supplies a substantial share of its workers to the manufacturing
sector, and over the past decade has tended to lead movements in
the permanent payrolls of manufacturing firms (Chart 2-5).






Long-Term Trends

To place the recent experience of the American manufacturing sector
in perspective, this section examines the evolution of the
manufacturing sector as a whole over the 50 years from 1950 to 2000
along three key dimensions: output, productivity and demand, and
employment.

Manufacturing Output over the Long Term


Manufacturing output increased dramatically from 1950 to 2000,
with particularly strong growth in the 1990s (Chart 2-6).
Manufacturing industrial production, a measure of real
manufacturing output, increased more than sixfold from 1950 to 2000
before declining in the recent recession. Over the same period,
annual growth in manufacturing industrial production averaged 3.8
percent, faster than real GDP growth of 3.4 percent. From 1990 to
2000, manufacturing industrial production expanded at an annual
rate of 4.6 percent, outpacing real GDP growth by more than a
percentage point. Per capita consumption of manufactured goods has
also risen: consumption of goods excluding food and fuel more than
quadrupled in real 2000 dollar terms from $1,400 per person in 1950
to $6,000 per person in 2000.





In contrast to real manufacturing output, nominal manufacturing
output (the dollar value of manufacturing output) has grown more
slowly than nominal GDP (the dollar value of GDP). As a result,
the share of nominal GDP accounted for by manufacturing roughly
halved, from 29 percent in 1950 to 15 percent in 2000 (based on
GDP by industry data available when this Report went to press; that
is, prior to the 2003 benchmark revision of the National Income and
Product Accounts).


Manufacturing Productivity and Demand over the Long Term


Two factors are driving the declining share of manufacturing in
U.S. nominal output. First, and most significant, productivity
growth in manufacturing lowered the relative price of manufactured
goods, but demand did not respond proportionately. Second, imported
manufactured goods increased their market share.

Productivity, as measured by output per hour worked, has grown more
rapidly in manufacturing than in the overall nonfarm business sector
over the last three decades. From 1950 to 1973, manufacturing
productivity grew at about the same pace as productivity overall.
Over the period from 1973 to 1995, manufacturing productivity growth
exceeded productivity growth overall by about 1 percentage point per
year. The disparity is even wider over the period from 1995 to 2000,
when manufacturing productivity grew at an annual rate nearly 2
percentage points higher than nonfarm business productivity (Chart
2-7). An hour of work in manufacturing produced about four times as
much in 2000 as it did in 1950, whereas an hour of work in the
nonfarm business sector produced less than three times as much in
2000 as it did in 1950.

This dramatic productivity differential has contributed to a decline
in the price of manufactured goods relative to services, which in
turn helps to explain the difference between the behavior of nominal
and real manufacturing output. Increased labor productivity in a
sector means that fewer hours are required to make a given amount of
output. This reduces the cost of production and, typically, the
relative price of that output. In the same way, relative prices tend
to increase in sectors that have experienced less productivity
growth, such as services. For example, the falling prices of
computers and other electronics have contrasted sharply with the
rising costs of services. This example is confirmed by the aggregate
data: the average price of consumption goods relative to services
fell more than 50 percent between 1950 and 2000. In contrast to the
nearly ninefold increase in the prices for






services, prices for durable goods (goods such as cars and
refrigerators that are expected to last, on average, three years or
more) rose by a factor of only 212 and prices for nondurable goods
rose by a factor of about 5 from 1950 to 2000 (Chart 2-8). Expressed
another way, to equal the buying power of $100 worth of durable goods
in 1950, a consumer would have spent $250 in 2000, while for $100
worth of services in 1950, a consumer would have spent $890 in
2000.

The slower growth of manufactured goods prices has increased the
purchasing power of incomes relative to what it otherwise would have
been, but the portion of this increase that Americans have allocated
to manufactured goods has not been large enough to maintain
manufacturing's share of nominal output. The boost to real income
from the relative price decline of manufactured goods has supported
demand not only for these goods but also for services such as health
care and financial advice. That is, Americans have used the resources
made available from the relatively slow growth in manufacturing
prices to buy many things, not just manufactured goods. Increased
demand for services, combined with rising relative prices for
services, is reflected in the fact that health services and
business services each




have increased their share of total nominal output about 4
percentage points since 1950. The finance, insurance, and real
estate industry has increased its share a dramatic 912 percentage
points. The opposite trend has held for manufacturing, in which
relative price declines have not been fully offset by increases
in demand. This explains why the share of manufacturing in total
nominal output has roughly halved since 1950. (All calculations of
industry share of nominal GDP are based on the pre-benchmark data
available when this Report went to press.)

In other words, U.S. demand for manufacturing products has been
relatively price inelastic. That is, demand has not been very
responsive to price declines. For example, a family that purchased
a car may have reacted to lower relative car prices (and the
increased real income they create) by paying for college or hiring
a home health care aide, rather than by putting those gains toward
the purchase of another car. As a numerical example of inelastic
demand, suppose that people buy 10 compact discs at $20 each (for
a total expenditure of $200). Now suppose the price falls from $20
to $10.  If people buy twice as many compact discs at $10, the value
of overall sales will still be $200 (20 compact discs at $10 each).
But if people increase their purchases to 15 compact discs, the
value of overall sales will be only $150, a decline of 25 percent.
This is similar to what has happened in manufacturing. Productivity
gains have tempered price increases, and demand has not responded
strongly enough to keep nominal revenues constant as a share of
nominal GDP.

A second factor that has led to a decline in manufacturing's share
of GDP is that Americans are purchasing more goods from abroad.
Goods purchases as a share of total domestic purchases have been
declining for about 30 years. The share of domestically produced
goods has fallen somewhat faster, particularly in the 1970s and
1990s. Domestically produced goods were 91 percent of overall
domestic goods purchases in 1970; by 2000, they had fallen to 68
percent. In other words, imports have made up an increased share
of goods bought in the United States (Chart 2-9).

Growth in exports of manufactured goods from the United States over
the past several decades has offset only some of the growth in
imports (Chart 2-10). As a result, net imports of nonagricultural
goods (imports minus exports) have risen materially, reaching about
30 percent of manufacturing production in 2000 (based on the
pre-benchmark data available when this Report went to press)
(Chart 2-11). In relation to the overall economy, net
nonagricultural goods imports have also risen, but remained below
5 percent of GDP in 2000. China has been a growing source of
manufacturing imports, although this growth has not been a major
factor in the increase of the U.S. trade deficit (Box 2-1).





----------------------------------------------------------------------

Box 2-1: China and the U.S. Manufacturing Sector

The recent decline in employment in U.S. manufacturing has
coincided with a sizable increase in the overall U.S. trade
deficit and a sharp increase in the U.S. bilateral trade deficit
with China. In part because of the high visibility of Chinese
imports, which are primarily everyday consumer goods, these
events have raised concerns that imports of Chinese goods come at
the expense of American manufacturing workers.

China's Trade with the World

While China's exports and imports grew quickly starting in the
early 1990s, China's trade with the rest of the world has been
modest until very recently (Chart 2-12). The growth in China's
trade has been well balanced in that increased exports to the
world have been matched by rising imports from the world.
According to data from China's official statistical agency, China
has had a trade deficit with the world excluding the United States
for several years. China recently ran trade deficits with a number
of other countries, including industrial countries such as Germany
and Japan.

China's Trade with the United States

China has a significant trade surplus with the United States, its
most important export market and the destination of one-quarter of
all Chinese goods exports. The U.S. trade deficit with China--about
$124 billion through November 2003 at an annual rate--is the single
largest bilateral goods and services trade deficit for the United
States. The next-largest bilateral deficit is with Japan, at $66
billion through November 2003 at an annual rate.

The U.S. trade deficit excluding China has also risen dramatically
since the mid-1990s and is about 312 times larger than the bilateral
deficit with China (Chart 2-13). China's share of the overall U.S.
trade deficit in goods has actually fallen since 1997--exactly the
period over which trade with China grew rapidly.

Greater trade with China does not appear to have contributed to an
increased overall U.S. trade imbalance, as the higher share of U.S.
imports from China has been more than offset by a declining share of
imports from other Asian countries. The share of U.S. imports from
the Pacific Rim as a whole has fallen since the mid-1990s (Chart
2-14). Restrictions on imports from China would be expected to
increase imports from other low-cost foreign producers, rather than
to increase production and employment for American manufacturers.
That is, any job gains from reduced Chinese imports are more likely
to occur in other developing countries rather than the United States.

U.S. exports to China have grown strongly in the last several years,
with exports to China up more than 60 percent since 2000. As of the
third quarter of 2003, China was the sixth-largest U.S. export
market. Exports to China have grown even while exports to the rest
of the world have stagnated (Chart 2-15).

The Impact of Trade with China on U.S. Manufacturing Employment

Imports from China affect the prospects for domestic firms with which
they compete, and this impact often extends to workers and
communities associated with these firms. This is especially the
case for firms that make items that are relatively intensive in the
use of less-skilled labor, as these are goods in which China has a
comparative advantage in production. This may raise the question of
whether imports from China are a primary factor in the displacement
of American manufacturing workers.

A closer look at the data indicates this is not the case. The low
level of U.S. imports from China before the mid-1990s suggests that
declines in employment prior to that period were not due to U.S.
trade with China. The data on more-recent job losses in manufacturing
indicate that China is not a primary factor in these declines, either.
With the exception of apparel, the largest job losses have occurred
in export-intensive industries for the United States, and job losses
in U.S. manufacturing have been mainly in industries in which imports
from China are small. For example, the computer and electronic
equipment industry accounts for 15 percent of all manufacturing job
losses since January 2000, but imports from China were only 8 percent
of U.S. output in 2002. Other export-intensive industries that have
suffered large job losses include fabricated metal products
(9 percent of manufacturing job losses and 2 percent of U.S.
output), machinery (10 percent and 2 percent), and transportation
equipment (12 percent and 0.4 percent).

----------------------------------------------------------------------






Manufacturing Employment over the Long Term


Employment in manufacturing as a share of total employment peaked in
the early 1940s at about one-third of all farm and nonfarm workers.
By 2000, it had declined to just below 13 percent (17 million out of
135 million employees). Employment in service-providing sectors
(including transportation, wholesale and retail trade, finance,
insurance, and real estate, and services) increased from 35 percent
of payroll employment in the early 1940s to 65 percent (86 million
workers) of all employees in 2000 (Chart 2-16).

The two main reasons for this shift from the manufacturing sector
to service-providing sectors in the labor market are related to
the explanations for the declining nominal share of manufacturing
output. First, increased demand for services and relatively slow
productivity growth in service-providing sectors have led to rising
demand for workers in these sectors. In manufacturing, inelastic
demand for manufactured goods and faster productivity growth have
lowered the relative demand for manufacturing workers.

Second, manufacturing employment likely has fallen in response to
the transfer of manufacturing jobs abroad. The jobs affected have
generally been those involved in the production of goods requiring
relatively low skills.






Indeed, this is part of the explanation for the rapid growth in
manufacturing productivity over the last 50 years (Chart 2-16). The
relatively highly-skilled American manufacturing workforce has been
increasingly focused on higher-productivity activities. This shift
can be seen by looking at compensation for the industries in which
employment decreased or increased the most from 1950 to 2000 (Table
2-1). With a few exceptions, employment fell dramatically in
industries with relatively low-skilled jobs and rose dramatically
in industries with relatively highly-skilled jobs.

This specialization is a natural outcome of the opening of economies
all over the world to trade. As a result of such specialization,
world efficiency increases and world output goes up as countries
focus on the activities in which they are relatively more
productive. All countries that participate in trade benefit from
this increased output.

The effect of long-term productivity improvements on the shift to
service-providing jobs is far more important than increased
manufacturing imports. Two simple hypothetical exercises can help
to illustrate this. In the first exercise, imagine that
manufacturing productivity was fixed at its value in 1970. To match
the actual amount of manufacturing output in 2000, one-third of
total U.S. nonfarm employment would have been required by
manufacturing, compared with the 13 percent required at 2000
productivity levels. That is, without the increase in
manufacturing productivity, manufacturing's share of nonfarm
employment would have increased 8 percentage






points rather than decreased 12 percentage points from 1970 to 2000.
As a second exercise, imagine that trade in manufactured goods was
balanced in 2000, so that net exports were zero, but assume that
the share of manufacturing employment in 1970 and productivity
growth from 1970 to 2000 were their actual values. This would
raise the amount of manufactured goods produced in the United
States. Manufacturing employment as a share of total nonfarm
employment, however, would have been only 1 percentage point
higher--14 percent, compared with the actual figure of 13
percent--if there had been balanced trade in manufactured goods
in 2000.

The Effects of Domestic Outsourcing and Temporary
Workers on Measurement of Manufacturing
Employment

The decline in manufacturing employment in the official statistics
may somewhat overstate the number of actual manufacturing production
jobs that have been lost.  Changing business practices in the
manufacturing sector have led to both the outsourcing of
nonproduction work that used to be done "in house" and the
increased use of temporary workers. Manufacturing firms that once
employed lawyers or accountants in their legal or finance
departments might now hire outside consultants to perform these
services. Counting this outsourcing as a decline in manufacturing
jobs is somewhat misleading, because these workers provide
services whether they are working for a manufacturing firm or an
outside firm.

Similarly, manufacturing firms are increasingly using temporary
workers, especially during periods of uncertain demand. Such
workers, previously counted as manufacturing employees, are now
counted as service-sector employees in the payroll employment data,
although many of them still produce manufactured goods. The way in
which employment statistics capture the increased use of
outsourcing and temporary workers thus overstates the shift from
manufacturing to service-providing jobs.

Much of the outsourced work is taken on by industries that make up
the employment category "Professional and Business Services," which
includes the temporary-help services industry. The professional and
business services category covers a rapidly growing sector of the
labor market, so it is likely that the understatement of
manufacturing employment has increased over time. Professional and
business services grew from just under 3 million employees in 1950
to over 16 million employees in 2000 (Chart 2-17). Employment in
subgroups of this category increased substantially in the 1990s
(Chart 2-18).







Results from academic studies can be used to estimate the
understatement of employment in the manufacturing sector, bearing
in mind that outsourced jobs are not necessarily comparable to
permanent ones (for example, a temporary worker may receive fewer
benefits than a permanent employee). One widely-cited study
estimates that about one-third of all temporary-help services
employees work in the manufacturing sector. If the official
manufacturing employment statistics are adjusted by this amount,
the decline in the level of manufacturing employment in the 1990s
is eliminated.

In terms of shares of overall nonfarm employment, adjusted
manufacturing shows a decline of 2.8 percentage points over the
1990s, compared with a drop of 3.1 percentage points in the reported
data. If outsourcing were also included, the decline in the actual
share of employment in the manufacturing sector would probably be
even smaller. In other words, at least one-tenth (and perhaps as
much as one-fourth) of the decline in manufacturing's share of
employment over the 1990s does not reflect a loss of manufactured
goods-producing jobs. Rather, it reflects how measurement
conventions used to calculate employment statistics account for
manufacturers' increased use of outsourced workers for tasks
previously performed internally. Another example of how
measurement conventions can affect, and confuse, the evaluation of
the manufacturing sector is in the definition of manufacturing
(Box 2-2).

---------------------------------------------------------------------
Box 2-2: What Is Manufacturing?

The value of the output of the U.S. manufacturing sector as
defined in official U.S. statistics is larger than the economies of
all but a handful of other countries. The definition of a
manufactured product, however, is not straightforward. When a
fast-food restaurant sells a hamburger, for example, is it providing
a "service" or is it combining inputs to "manufacture" a product?

The official definition of manufacturing comes from the Census
Bureau's North American Industry Classification System, or NAICS.
NAICS classifies all business establishments in the United States
into categories based on how their output is produced.  One such
category is "manufacturing." NAICS classifies an establishment as
in the manufacturing sector if it is "engaged in the mechanical,
physical, or chemical transformation of materials, substances, or
components into new products."

This definition is somewhat unspecific, as the Census Bureau has
recognized: "The boundaries of manufacturing and other sectorsï¿½ can
be somewhat blurry." Some (perhaps surprising) examples of
manufacturers listed by the Bureau of Labor Statistics are:
bakeries, candy stores, custom tailors, milk bottling and
pasteurizing, fresh fish packaging (oyster shucking, fish
filleting), and tire retreading. Sometimes, seemingly subtle
differences can determine whether an industry is classified as
manufacturing. For example, mixing water and concentrate to produce
soft drinks is classified as manufacturing. However, if that
activity is performed at a snack bar, it is considered a service.

The distinction between non-manufacturing and manufacturing
industries may seem somewhat arbitrary but it can play an important
role in developing policy and assessing its effects. Suppose it
was decided to offer tax relief to manufacturing firms. Because
the manufacturing category is not well defined, firms would have
an incentive to characterize themselves as in manufacturing.
Administering the tax relief could be difficult, and the tax relief
may not extend to the firms for which it was enacted.

For policy makers, the blurriness of the definition of
manufacturing means that policy aimed at manufacturing may
inadvertently distort production and have unintended and harmful
results. Whenever possible, policy making should not be based upon
this type of arbitrary statistical delineation.

---------------------------------------------------------------------


Effects of the Shift to Services on Workers'
Compensation




Many workers affected by the structural developments in
manufacturing have experienced difficult transitions. Studies
indicate that displaced workers have a significant chance of being
unemployed or employed in a part-time job for some time following
their job loss. Many of those who are able to find new jobs suffer
earnings declines compared to previous earnings. Furthermore,
workers also experience losses in earnings growth relative to what
they would have had if they had remained continuously employed.
Because of these effects, an often-voiced concern is that the shift
toward employment in services has meant that more Americans are
working in low-paying jobs.

While the shift from the manufacturing sector to service-providing
sectors has been painful for many displaced from the manufacturing
sector, the average effect on compensation--and in particular on
new entrants into the labor force who have chosen to work in
services rather than manufacturing--has been less worrisome. Some
service-providing industries pay less than some manufacturing
industries, but much of the employment growth in service-providing
sectors has occurred in industries with higher than average
compensation. The third column of Panel A in Table 2-2 shows the
total compensation per full-time equivalent employee in five
service-providing industries relative to the average across
industries: for example, compensation in wholesale trade in 2000
was 27 percent higher than the average (which equals 100 percent).
The second column gives the change in employment from 1950 to 2000
for each industry: wholesale trade employment increased more than
4 million over this period. As Panel A reveals, four of the five
service-providing industries with the largest employment increases
paid compensation roughly at or above the average. Together, these
five service-providing industries can explain nearly two-thirds of
overall private employment growth from 1950 to 2000. Panel B of
Table 2-2 shows that three of the five manufacturing industries
with the highest job-loss rates paid less than the average
private-sector job in 2000. For example, apparel employment fell
nearly 600,000 from 1950 to 2000, and compensation of workers in
the apparel industry in 2000 was only 67 percent of the average.
As a result of the large increases in employment in some of these
high-paying service-providing industries, the gap between
compensation in service-providing sectors and manufacturing has
been closing over the last couple of decades.






The Transition in Context


Individuals and communities tied to declining industries experience
dislocation and distress. While many workers have made the
transition from manufacturing to the service sector, the transition
can be difficult. To ease it, the President has supported policies
for worker retraining accounts and has extended unemployment
insurance benefits when needed. The appropriate policy responses
to this transition will be discussed in more detail later in
this chapter. Before that, however, it is useful to place the
evolution of the U.S. manufacturing sector in a broader context.

First, the shift to a relatively more service-oriented economy has
involved substantial benefits for American consumers and producers.
Real incomes have risen, allowing consumers to purchase more goods
and services such as food, health care, transportation, and
education, while measures of the quality of life and life expectancy
have also increased. In addition, the growth of the
service-providing sector has generated new opportunities for
employment in industries such as information technology services,
financial services, and entertainment.

Second, the shift of employment away from lower-productivity
manufacturing toward higher-productivity manufacturing and
service-providing sectors reflects economic growth and development,
just as the shift away from agriculture toward manufacturing did in
the last century (Box 2-3). The relative shift from manufacturing
toward service-providing sectors has been shared by other advanced
economies over the last few decades (Chart 2-19). Manufacturing
employment declined from the mid-1990s to 2002 in a number of
countries whose economies are rapidly developing, including China,
Brazil, and South Korea. In fact, China, Brazil, South Korea, and
Japan had steeper percentage declines in manufacturing employment
over that period than the United States.






---------------------------------------------------------------------

Box 2-3: The Evolution of the U.S. Agricultural Sector


The evolution of U.S. manufacturing from 1970 to 2000 mirrors, in
important respects, that of U.S. agriculture from 1940 to 1970.
Total real farm output increased more than 60 percent from 1940 to
1970. Over the same period, employment in farming declined nearly
6 million, or almost two-thirds of the level in 1940 (Chart 2-20).
This translated into a decline in agriculture's share of total
employment of 15 percentage points, from 19.4 percent in 1940 to
4.4 percent in 1970.

While the histories of agriculture and manufacturing in the United
States differ in some ways, such as the prominent role of subsidies
in the agricultural sector, their similarities help put the
long-term story of the manufacturing sector in context.

In both sectors, a 30-year period of rapid productivity growth
substantially reduced the share of the American workforce needed to
meet demand for food and manufactured goods. Labor productivity in
agriculture nearly quadrupled from 1940 to 1970 (Chart 2-21), a
period that has been called the "second American agricultural
revolution." This productivity boom has been attributed to the
invention of new technologies, such as hybrid crop varieties, as
well as the widespread application of existing technologies, such
as machinery and conservation practices.

Agricultural productivity growth led to low growth in the price of
food, bringing substantial benefits to American consumers and the
U.S. economy as a whole and significantly improving U.S.
competitiveness in world markets. Despite the mid-century expansion
in the demand for agriculture's output, prices remained essentially
flat. After the run-up in demand and prices during World War II and
its immediate aftermath, agricultural prices increased only
4 percent from 1950 to 1970. The average price of all commodities,
in comparison, increased 35 percent from 1950 to 1970 (Chart 2-22).
The lack of food price inflation is mimicked by the low inflation
in manufacturing in the last few decades, with a sizable benefit
for American consumers in both cases.

The evolution of the agricultural sector has been good for the
economy on the whole, but it meant dislocation for millions of
agricultural workers--a process that continues today. Displaced
farm workers faced uncertainty regarding their next job and the
applicability of their skills in different sectors, just as
manufacturing workers do today. The 1940s and 1950s saw the rapid
growth of new industries that hired workers no longer needed on
farms. Manufacturing itself likely absorbed a substantial
percentage of former agricultural workers: nearly 8 million new
manufacturing jobs were created between 1940 and 1970, 2 million
more than the total decline in agricultural employment.

In the 1970s and 1980s, service-providing sectors likely absorbed
workers not needed in manufacturing. This continued in the 1990s,
as high-tech and financial services accounted for new employment
growth. Looking forward, it is difficult to predict which
industries will grow and require more workers. The past experience
of the adjustment in agriculture suggests that market forces will
continue to reshape the American workforce.

---------------------------------------------------------------------






The Role of Policy



Markets operating free from government intervention will, in most
cases, best allocate the Nation's resources across sectors. It is
generally a mistake to target government assistance to a particular
sector at the expense of other sectors, and manufacturing is no
exception. That said, government policy can play a positive role.
Policies targeted toward general education and training, such as
the President's landmark education reforms and proposed funding to
help displaced workers train for new opportunities, will help people
adapt to ongoing structural changes. The President's Jobs for the
21st Century plan will support students and workers by improving
high school education and strengthening post-secondary education
and job training.

The short-run performance of the manufacturing sector is closely
tied to fluctuations in overall economic activity. Policies that
increase aggregate output and economic growth will help to improve
the near-term outlook for the manufacturing sector. This
Administration put forward a six-point plan for the U.S. economy in
September 2003. The plan would help the manufacturing sector along
with the overall economy, and it includes the following components:


Making Tax Relief Permanent


The Administration has undertaken several important fiscal measures
to strengthen growth, including the 2001 tax relief program, the
March 2002 stimulus package, and the May 2003 Jobs and Growth Act.
These policies have already contributed to the current recovery in
manufacturing. The President has proposed making provisions of the
2001 and 2003 tax cuts permanent. These include measures that lower
the cost of capital and thereby encourage business investment.
Capital investment makes up a relatively large share of
manufacturers' costs, so a lower cost of capital provides a
particularly important benefit to manufacturers. Moreover,
manufacturers produce capital goods, so increased investment demand
particularly benefits manufacturing firms.


Making Health Care Costs More Affordable and Predictable


The President's proposals aim to reduce frivolous litigation, help
individuals save for future health expenses, and allow small
businesses to pool together to purchase health coverage. Health care
costs as a share of total compensation are one-third higher in
manufacturing than in service-providing industries. The President's
proposals will help manufacturers reduce the burden of increasing
health care costs.


Reducing the Burden of Lawsuits on the Economy


The President seeks to address the burden that lawsuits impose on
American businesses. For example, estimates suggest that roughly 60
companies entangled in asbestos litigation have gone bankrupt
primarily because of asbestos liabilities, displacing between
52,000 and 60,000 workers.


Ensuring an Affordable, Reliable Energy Supply


Initiatives include modernizing the electricity grid and
streamlining the process of acquiring permits for natural gas
exploration. This is vital for manufacturing, which makes up about
15 percent of nominal GDP but accounts for around one-quarter of
energy use in the United States.


Streamlining Regulations to Ensure that they are Reasonable and
Affordable


Research has shown that manufacturing bore about 30 percent of the
costs of regulation in the United States in 2000--nearly double its
share of nominal output.



Opening International Markets to American Goods and Services


This has become particularly important for the manufacturing sector.
While exports accounted for about one-sixth of American
manufacturing production in 1970, they made up nearly half by 2002.


Conclusion


The manufacturing sector in the United States has undergone
significant change in the last half-century. Productivity and real
output in manufacturing have risen dramatically, and faster than in
the economy as a whole. Productivity improvements have boosted real
income in the United States. However, because Americans have spent
much of their real income gains on services rather than
manufactured goods, manufacturing's share of employment has
declined. In the recent recession, manufacturing output and
employment were hit particularly hard. The President's policies,
aimed at stimulating the overall economy, easing restrictions that
impede manufacturing growth, and ensuring that workers have the
skills they need to be competitive, address the short-term
difficulties of the sector and ensure its long-term health.