Manufacturing USA

Can there be a restoration and even an increase in the role and prominence of US manufacturing in the context of capitalist globalization today?
Short Answer: NO


The Decline in Manufacturing Employment

The manufacturing sector of the U.S. economy has experienced substantial job losses over the past several years. In January 2004, the number of such jobs stood at 14.3 million, down by 3.0 million jobs, or 17.5 percent, since July 2000 and about 5.2 million since the historical peak in 1979. Employment in manufacturing was its lowest since July 1950
Statistics provided in 1998-2005 charts below confirm this trend is not just a product of the 2000 recession.
 
Manufacturing Employment
 
(Millions of jobs graph only goes to 2004)


 
The drop in manufacturing employment since the beginning of the recession
The recent manufacturing decline relative to services largely reflects the weak demand for capital goods in the United States and for both capital and consumer goods among its major trading partners. In the United States, the demand for machinery and other capital equipment slumped after the investment surge of the late 1990s and was only beginning to recover in 2003. The resulting loss in production in industries producing capital goods severely reduced employment in the sector. Meanwhile, tepid growth overseas and a high U.S. real exchange rate meant weak demand for U.S. goods among the nation's major trading partners. Consequently, U.S. exports have been weaker during the 2001 recession and the recovery thus far than during and after most previous recessions, while imports have grown about as fast as they typically have after previous recessions. Clearly a strong dollar combined with cheaper real mfg labo costs in developing countries is contributing significantly to job loss in mfg.
 
U.S. Exports and Imports of Goods
 
(Percentage difference from peak value)




 
a)    Note. Average of the seven recoveries during the 1949-1990 period, excluding the recovery in 1980 from the recession that year because that recovery was so short-lived.

Long-Term Influences
Shift in Demand Away from Manufactured Goods
The share of consumer spending devoted to manufactured goods has declined over time both in the United States and in other industrialized nations. As consumers' income has risen, they have increased their purchases of goods but boosted their spending on services--including medical care, notably--even more. In 2000, 42 percent of U.S. consumer spending was devoted to goods, down from 53 percent in 1979 and 67 percent in 1950. Likely factors contributing to that shift are an increase in the value of time resulting from rising real (inflation-adjusted) wages and married women's increased participation in the labor force, which has led households to substitute some purchased services for tasks formerly performed in the home.
Manufacturing Productivity
Over recent decades, U.S. manufacturers have continually invested in more and better capital goods and manufacturing techniques in order to remain competitive in world markets. That investment has enabled them to raise their output and keep pace with overall economic growth without a corresponding increase in the number of workers that they employ. Since 1979, the productivity of manufacturing workers has grown at an average annual rate of 3.3 percent, significantly faster than the 2.0 percent growth of labor productivity in the nonfarm business sector overall.(2)
Improvements in productivity are economically beneficial, as they permit—but do not necessarily guarantee-- greater profits, higher real wages, and lower prices. But while the prices of manufactured goods have indeed fallen consistently relative to other prices, those lower prices have not led to increased sales: the share of gross domestic product (GDP) accounted for by manufacturing output has been roughly constant over the past half-century. Strong growth in productivity and a slower rate of growth in the demand for manufactured goods have necessarily entailed a decline in manufacturing's share of total employment.
 
Output and Employment in the Manufacturing Sector
 
(Log scale)


 
Sources: Congressional Budget Office; Department of Labor, Bureau of Labor Statistics; Department of Commerce, Bureau of Economic Analysis. Note: The vertical bars indicate periods of recession as defined by the National Bureau of Economic Research.
The gains in manufacturing productivity have continued recently, even through the downturn in 2001. Since the peak of the last business cycle in March 2001, labor productivity in manufacturing has risen at an average annual rate of 5.5 percent, faster than its average annual rate of growth during previous postwar recessions and the early part of the ensuing recoveries. Yet medium incomes are not rising at all, and, in fact, are falling or flat for all but the very highest paid occupations.
Competition from Foreign Producers
A portion of the long-term decline in employment in some manufacturing industries can be linked to the expansion of trade. The gains from trade arise as nations specialize in the goods and services that they can produce efficiently relative to other countries. Thus, the expansion of trade necessarily involves changes in the mix of products. The United States has specialized in products requiring a highly skilled labor force even as lesser jobs have shifted to countries where labor is less skilled. In the apparel sector, for example, the number of jobs in this country has declined from over 900,000 in 1990 to less than 300,000 today.
Some observers have specifically attributed recent job losses in manufacturing to a surge in the bilateral trade deficit with China. From 1992 to 2003, the trade deficit with China grew from $18.3 billion to $124.0 billion, which is larger than the deficit with any other country. However, much of the increase in imports from China reflects a shift away from imports from other Asian countries rather than an increase in total imports. In fact, while U.S. imports attributable to China increased from 5 percent in 1992 to 12 percent in 2003, the share of imports from other Pacific Rim countries declined from 34 percent to 21 percent
 
U.S. Imports from China and from Other Pacific Rim Countries
 
(Percentage of total imports)




 
Sources: Congressional Budget Office; Department of Commerce, Bureau of the Census.
a. Australia, Brunei, Hong Kong, Indonesia, Japan, Korea, Macao, Malaysia, New Zealand, Papua New Guinea, Philippines, Singapore, and Taiwan
 
Productivity and Income.

While productivity has increased steadily since 1975, incomes measured in either hourly wages, or total compensation (including benefits, income from capital—pensions, savings, etc) have not kept pace, except for professions and very highly skilled workers at the very top. It should be noted a broad range of very detailed studies of historically wages and worker incomes relative to productivity in all advanced capitalist countries demonstrate that worker incomes DO track productivity over the long run, although it is arguable that social upheavals from labor struggle – THE CLASS STRUGGLE -- play an essential role in regulating and enforcing this trend. If this is true then we are due for a major social upheaval, since a divergence in productivity and income, beginning at the bottom of the scale but gradually working its way upwards, has been underway since approximately 1975. This correlates broadly with both the decline in the US mfg sector, the decline in the strength of industrial unions, and first surge in “runaway shops” of the late 60’s and early 70’s. The fact the median household income did not reflect an absolute decline until the 2000 recession and afterwards only underscores how long this trend has been underway and how, despite big gains in income due to the explosion of  high-tech related occupations in the 90’s, it has now become the dominant trend. Even the high-tech occupational incomes have been flat compared with their own productivity growth.

All that being said, it should be noted: in truth, productivity is not easily measured, especially when a significant number and proportion of outputs are service oriented, informal, or are intangible. Manufacturing productivity IS easily measured. However many estimates of productivity in services or intangibles rely on some questionable assumptions where real data is highly imperfect, or on very weak linkages and correlations with manufacturing indexes. Example:  computer cpu increases in processing power have grown at a polynomial rate since the 70’s (at a power of between 2 and 3) per year. However the software a worker uses to compose a document has grown in complexity (use of processing power) at a nearly comparable rate. How do you measure the productivity of a worker’s ability to insert a huge and expanding array of formatting features into his/her document? It is very difficult to find accurate quantitative measures. Another example of bad correlations: In NAFTA negotiations, Mexican and U.S. economists had no reliable data on Mexican agricultural labor productivity. Much work in the agricultural sector was in fact informal, compensated in kind or barter, and unrecorded in any form. NAFTA negotiators decided to use the ratio of Mexican to US manufacturing productivity as a substitute. The ratio was in error by a factor of nearly 70! That is, Mexican agricultural productivity was nearly 100 times less productive than US agricultural productivity (whereas Mexican manufacturing productivity for comparable industries was only 30% less than that in the US). Consequence: terms of trade negotiated in NAFTA were disastrous for Mexican agriculture. Millions left farms in 94-96, found no work in the quickly saturated manufacturing sector, and immigrated to the US by any means they could.

Conclusions: Our focus should be on raising income, not manufacturing, per se:

1.    While there is some debate on how much weight each of various causes of this (income lagging productivity) phenomenon should be given, there is little debate about the list of causes:
a.    Weak bargaining power of workers to recapture gains from productivity.
b.    Absence of nationalized health care.
c.    Immigration, law of supply and demand in the labor markets, especially when 6-10 million workers have no legal protection.
d.    Competition with cheap, often forced or slave labor: Transnational corps have become truly transnational; supply chain technology and infrastructure guarantee their independence from ANY one national government.
e.    Relative decline in investments in human capital in the US – esp. education, retraining, etc.
2.    Manufacturing decline is a fundamental trend of the scientific-technological revolution’s continuing re-divisions of labor, reflecting increased productivity. While this trend can be modified or smoothed or alleviated by reforms in trade, immigration, health care and labor-law policy, its fundamental course is an objective feature of economic development, and cannot be altered.
3.    Globalization is here to stay. The negative impact of global competition on US incomes can only fundamentally be countered by:
a.    Advances in the class struggle in developing (primarily manufacturing) nations to improve world incomes;
b.    More favorable and progressive trade agreements for developing nations, allowing them sufficient protection, while promoting freer, completely reciprocal trade zones between advanced capitalist trading partners. Together, these reforms outlined by Stiglitz in his program of 2007 will promoting more equitable global development, and growth for all.
c.    Advances in the US and advanced capitalist partner workforces to new and broad-based high-tech – high-skill levels alongside an enhanced ability to bargain for the productivity gains inherent in that transformation.
4.    Service occupations and incomes are in fact the key battle ground in raising working class incomes, as they constitute the fastest growing occupations,  invading significant parts of manufacturing labor markets (the bane of organizing attempts by the UAW at Toyota, in fact). They are also the source of both the highest and lowest income categories in the workforce. To the extent science and technology are increasingly able to automate the repetitive or algorithmic parts of these occupations: the occupations will become proportionally unsuitable as sources of surplus value. In other words they become less and less “proletarian”, in the Marxist sense, under the impact of technology. Raising their incomes thus becomes theoretically and practically problematic from the standpoint of the traditional collective bargaining and workplace-based “class struggle” forms of organization, requiring some adjustments.
a.    Consider the labor contracting business in the US, a rapidly growing sector of the workforce in almost every industry and service, as a special case, a kind of substitute for service labor in general. In many ways (though not all), the union hiring hall of yore could deliver these services no less efficiently, and at greater profit to the workers, than the existing forms in which services are delivered. Other than the legal, tax and liability preferences afforded the “corporate” form of economic organization there is little economic justification for a sharp division between “owners” and “workers” in these firms. Contract labor firms require virtually no capital other than human capital. Beyond reasonable management fees and incentives, there is really no cause for the labor being traded as “labor-power” in the Marxist sense. And, in fact, it is not in many instances. What is really being traded is the market value of the actual “service” or labor product, of which the contract worker is being paid a proportional part.
b.    Thus, an important means, perhaps the only means, of raising many service incomes is to demand true proportional returns on the value of the traded labor product; i.e. a division of the capital earnings, or profits, of the contracting firm. (Of course, an obstacle here is risk – who assumes it? Everyone wants a “just” share of the profits, but what about losses?).
c.    Raising incomes has to become a political question where, in effect, getting a raise is directly on the ballot. Contract services as well as many other kinds of service work occur in environments that make traditional union recognition and collective bargaining virtually impossible. The reforms of the Employee Free Choice Act, while absolutely necessary, are in my opinion not sufficient to overcome these obstacles, even if they were to actually pass Congress. And the indirect character of the means by which the  benefits would accrue to all from its passage is going to make a very difficult mass fight given the much degraded numeric strength of the labor movement. Further, the verbal assurances of support from many members of Congress will have to be subjected to Reagan’s “trust but verify” test once the pressure from employers is brought to bear.

d.    In many areas of service work, chiefly education and health and other public services, worker organizations must move beyond another boundary of traditional union organization, despite the many roadblocks put in their path:  they must assume responsibility not only for their wages, benefits and working conditions, but also welcome – and demand -- public responsibility and ownership for the services they deliver. There is a crisis in many areas of service that goes to the heart of the investments the United States is making in its people. The working class has the capacity, the skills, and the know-how to get it done. And done right.




Additional charts of interest:

Exports, Imports and Net Exports share of GDP


 

Total Manufacturing Value added compared to total Manufacturing compensation 87-97 (in millions $$):





Total mfg VA compared to Total Comp 98-05 (millions $$)


 

Mfg Percent of GDP 1947 – 1986



 
Mfg share of GDP relative to selected services 47-86



 Mfg share of GDP relative to financial services




Average Real Hourly wages



 


Compensation portion of value added 1987-2005



More charts when I get to Chicago!!


 
Source: BLS

Productivity and Income.

While productivity has increased steadily since 1975, incomes measured in either hourly wages, or total compensation (including benefits, income from capital—pensions, savings, etc) have not kept pace, except for professions and very highly skilled workers at the very top. It should be noted a broad range of very detailed studies of historically wages and worker incomes relative to productivity in all advanced capitalist countries demonstrate that worker incomes DO track productivity over the long run, although it is arguable that social upheavals from labor struggle – THE CLASS STRUGGLE -- play an essential role in regulating and enforcing this trend. If this is true then we are due for a major social upheaval, since a divergence in productivity and income, beginning at the bottom of the scale but gradually working its way upwards, has been underway since approximately 1975. This correlates broadly with both the decline in the US mfg sector, the decline in the strength of industrial unions, and first surge in “runaway shops” of the late 60’s and early 70’s. The fact the median household income did not reflect an absolute decline until the 2000 recession and afterwards only underscores how long this trend has been underway and how, despite big gains in income due to the explosion of  high-tech related occupations in the 90’s, it has now become the dominant trend. Even the high-tech occupational incomes have been flat compared with their own productivity growth.

All that being said, it should be noted: in truth, productivity is not easily measured, especially when a significant number and proportion of outputs are service oriented, informal, or are intangible. Manufacturing productivity IS easily measured. However many estimates of productivity in services or intangibles rely on some questionable assumptions where real data is highly imperfect, or on very weak linkages and correlations with manufacturing indexes. Example:  computer cpu increases in processing power have grown at a polynomial rate since the 70’s (at a power of between 2 and 3) per year. However the software a worker uses to compose a document has grown in complexity (use of processing power) at a nearly comparable rate. How do you measure the productivity of a worker’s ability to insert a huge and expanding array of formatting features into his/her document? It is very difficult to find accurate quantitative measures. Another example of bad correlations: In NAFTA negotiations, Mexican and U.S. economists had no reliable data on Mexican agricultural labor productivity. Much work in the agricultural sector was in fact informal, compensated in kind or barter, and unrecorded in any form. NAFTA negotiators decided to use the ratio of Mexican to US manufacturing productivity as a substitute. The ratio was in error by a factor of nearly 70! That is, Mexican agricultural productivity was nearly 100 times less productive than US agricultural productivity (whereas Mexican manufacturing productivity for comparable industries was only 30% less than that in the US). Consequence: terms of trade negotiated in NAFTA were disastrous for Mexican agriculture. Millions left farms in 94-96, found no work in the quickly saturated manufacturing sector, and immigrated to the US by any means they could.

Conclusions: Our focus should be on raising income, not manufacturing, per se:
5.    While there is some debate on how much weight each of various causes of this (income lagging productivity) phenomenon should be given, there is little debate about the list of causes:
a.    Weak bargaining power of workers to recapture gains from productivity.
b.    Absence of nationalized health care.
c.    Immigration, law of supply and demand in the labor markets, especially when 6-10 million workers have no legal protection.
d.    Competition with cheap, often forced or slave labor: Transnational corps have become truly transnational; supply chain technology and infrastructure guarantee their independence from ANY one national government.
e.    Relative decline in investments in human capital in the US – esp. education, retraining, etc.
6.    Manufacturing decline is a fundamental trend of the scientific-technological revolutions continuing re-divisions of labor, reflecting increased productivity. While this trend can be modified or smoothed or alleviated by reforms in trade, immigration, health care and labor-law policy, its fundamental course is an objective feature of economic development, and cannot be altered.
7.    Globalization is here to stay. The negative impact of global competition on US incomes can only fundamentally be countered by:
a.    Advances in the class struggle in developing (manufacturing) nations to improve world incomes;
b.    More favorable and progressive trade agreements for developing nations, promoting more equitable global development.
c.    Advances in the US and advanced capitalist partner workforces to new and broad-based high-tech – high-skill levels alongside an enhanced ability to bargain for the productivity gains inherent in that transformation.
8.    Service occupations and incomes are growing in their own right, as well as invading significant parts of manufacturing labor markets (the bane of organizing attempts by the UAW at Toyota, in fact). To the extent science and technology are increasingly able to automate the repetitive or algorithmic parts of these occupations: the occupations will become proportionally unsuitable as sources of surplus value. In other words they become less and less “proletarian”, in the Marxist sense, under the impact of technology. Raising their incomes thus becomes theoretically and practically problematic from the standpoint of the traditional collective bargaining and workplace-based “class struggle” forms of organization, requiring some adjustments.
a.    Consider the labor contracting business in the US, a rapidly growing sector of the workforce in almost every industry and service, as a special case, a kind of substitute for service labor in general. Other than the legal, tax and liability preferences afforded the “corporate” form of economic organization, there is little economic justification for a sharp division between “owners” and “workers” in these firms. Contract labor firms require virtually no capital other than human capital. Beyond reasonable management fees and incentives, there is really no cause for the labor being traded as “labor-power” in the Marxist sense. And, in fact, it is not in many instances. What is really being traded is the market value of the actual “service” or labor product, of which the contract worker is being paid a proportional part.
b.    Thus, an important means, perhaps the only means, of raising many service incomes is to demand true proportional returns on the value of the traded labor product; i.e. a division of the capital earnings, or profits, of the contracting firm. (Of course, an obstacle here is risk – who assumes it? Everyone wants a “just” share of the profits, but what about losses?)