The present recession in the US economy has been notably different from previous episodes of economic slump in atleast two ways. Firstly, the growth in employment which started falling since early 2001 continued to fall even during the first quarter of January, 2004. In the year 2001 alone, there were1.5 million net jobs lost according to official estimates on payroll employment. The following year added less than 0.4 million jobs. Employment growth during 2003 has been inadequate to make up for the massive job-losses of the previous two years. The situation in the opening months of 2004 has see-sawed so far (see Table 1). The slowest to respond has been the manufacturing sector, which actually triggered the recession when the boom in the IT sector burst at the end of the 1990s. In all the 21 industries that constitute manufacturing sector, employment declined, and 17 of the 21 saw losses exceeding 10%. Congressional budget office estimates the manufacturing sector net jobloss to be 3 million between July 2000 and January 2004 ^[1]. This has brought down the employment in manufacturing down to 14.3 million, lowest since July 1950. Unemployment rates that had dipped below 4% during the late 1990s shot up to above 6% (see Graph 1). Analysts say that even this high rate might be a conservative estimate since there are scores of people who faced with long periods of unemployment have stopped reporting themselves as part of the labour force ^[2]. Thus the employment situation is adverse not only because of the severe intensity of the jobloss but also due to the extended period during which the employment situation has worsened. The present cycle is one of the longest lasting declines in employment since 1933.

The second tendency, which makes the present US recession distinct from earlier periods of recession, is the enormous fiscal and monetary stimulus being provided by the Federal government. Direct demand injection in the form of substantial increases in defense and defense-related expenditure, orchestrated through the war against terrorism, has supplemented traditional supply side measures. The latter include substantial reduction in interest rates (see Graph 2) and major tax cuts. Probably for the first time ever, massive increases in war expenditure by the US government were accompanied by heavy tax relief ^[3].

The following paragraphs explore the unemployment issue in some detail. Why despite a turnaround in output growth evident since 2002/3, did the employment growth continue to drag behind? ^[4] Our method would be to analyse the patterns of expenditure in the two major sectors – government and private business and relate it to the puzzle of slow employment growth.

Government Spending
Military-fuelled growth, or military Keynesianism, was first theorised by Kalecki in 1943. Kalecki argued that capitalists and their political champions tended to bridle against classic Keynesianism; achieving full employment through public spending made them nervous because it risked over-empowering the working class and the unions. The military was a much more desirable investment from their point of view, although justifying such a diversion of public funds required a certain degree of political repression, best achieved through appeals to patriotism and fear-mongering about an enemy threat - and, inexorably, an actual war ^[5].

The neo-liberal right wing in the US has adopted the strategy of military Keynesianism with much enthusiasm. Between 2002 and 2004, total budgetary outlay grew at an annual rate of 7-8%, up from an average of 3.46% during 1991-9. National defense expenditure which had been declining in absolute terms every year since the end of Cold War, registered an annual growth of 12-17% between 2002 and 2004. Including the costs of two major military operations in Afghanistan and Iraq, the US defense budget increased from $300 billion in the year 2000 to $460 billion in 2004, i.e., by 52.9%.

Department of homeland security is another area where resources have flown generously. On March 1, 2003, approximately 1.8 lakh personnel from 22 different organizations around the government became part of the Department of Homeland Security, whose mission was to make America more secure. 17 of the 19 budget functions contain at least some funding for homeland security activities. The President’s budget for 2005, includes $47.4 billion as resources for homeland security activities, a 15% increase over 2004 level and a 130% increase over 2002. If we compare these figures with other budgetary allocations, say the estimated outlays on Education, Training, Employment, & Social Services (Function 500) or Community and regional development (Function 450), over 2004 and 2005, outlay on Function 500 is estimated to increase by $1.8 billion (2%) to reach a level of $89 billion, whereas outlay on Function 450 would actually fall in absolute terms by $1.7 billion to reach a level of $17 billion. Obviously, the disproportionately large spending on defense and related activities to combat the supposed threat of terrorism is being made up through cuts in spending on social sector, infrastructure, development and other heads. (see Graph 3)

What as been the underlying pattern of employment generation of this nature of public spending?
The rising level of sophistication of defense equipment and overall defense systems has meant rising capital intensity of production and consequently falling job-creation capacity of defense spending. A report in the Washington post in 1986 wrote that 1 out of every 20 jobs in the US were directly or indirectly related to military spending ^[6]. More recent estimates say that this percentage has dropped to 3.2% ^[7]. Military expenditure, which has always been extremely capital intensive has become even more so. Reports widely speak of absence of new recruits of military personnel even during the war on Afghanistan and Iraq.

In the absence of direct recruitment of military personnel by the Pentagon, the route by which defense expenditure can generate employment is through defense contractors. The heavy government outlays on security, military operations, and other types of defense expenditure have financed multi-billion dollar contracts to politically-connected giant corporations. The largest of these contracts would be more than $20 billion, and the average one would be atleast a few billions. In fact, each of these contracts has the capacity to regenerate the economy of an entire region. For instance, the $4 billion contract awarded to the California company Northrop Grumman to work on the Star Wars missile defense program, has a possibility to regenerate several of the computer firms, as much of the modern security paraphernalia depends on Silicon Valley computer technology.

To what extent the contracts translate into real production and employment boosters would of course depend on the extent of excess capacity and inventory holdings of these firms. Most of the orders for supplies of tanks, ordnance, missiles, shipbuilding, aircraft, engines, computer technology would be enforced over a longish period of time and therefore involve considerable production lags. Research on defense systems, including homeland security, a major contributor to the increments in defense budget similarly has a long gestation period. In the short-run, there is less chance of many of these spending heads stimulating production and employment significantly.

War profiteering has raised the profit margins of defense contractors enormously ^[8]. Yet the transmission from very high profits to higher employment is not at all obvious. A case in point is General Dynamics, one of the top 5 military contractors. Between 1991 and 1993, stock price of the company rose 553%, but General Dynamics downsized its workforce in the early 1990s by 80% from 1,02,000 to 21,000 ^[9].

Overseas defense spending on war and occupation of Afghanistan automatically represents leakage from US public spending and would therefore lower the multiplier effect of public spending. For instance, the present state of affairs in Iraq has forced the giant corporations engaged in reconstruction work to employ security persons to the tune of 200,000 for its staff, which would mean an additional cost of $1 billion. However, there are two factors that have prevented the overseas factor from becoming too strong a force in pulling down domestic employment. One, the actual war or the reconstruction budget was only a relatively small part of the overall rise in defense spending. Second, barring physical constraints that would encourage the US firms in Iraq and Afghanistan to spend locally, US contractors have been found to place a lot of the reconstruction orders with their offices in the US. Bechtel, the engineering giant was awarded a $680 million contract to evaluate and repair Iraq's power, water and sewage systems. The work has mainly been assigned to its offices in the US.

Finally, in comparison to military expenditure the job creation capacity of various other sectors that are part of federal government’s budgetary functions is much higher (see Graph 4). Thus as far as jobs are concerned, military spending is a much worse investment than other federally funded programme. The new orientation of the US budget with cuts for non-homeland non-defense functions, would have caused some erosion of employment growth.

Private Business Spending
The private business sector is in the peculiar position of being responsible for the massive job-loss, and yet structural constraints have prevented generation of any fresh stimulus from within the industrial sector, especially manufacturing. Table 2 compares the contribution to real GDP growth by various national income categories for the three years 2001 to 2003. It can be clearly observed that for two consecutive years, 2001 and 2002, contribution to real GDP growth of gross non-residential private domestic investment was negative. But for the steady expansion of personal consumption spending and public spending, the repercussions for real output growth would have been much more severe during the recession years.

Private business spending has been slow to respond to monetary stimulus by the Fed. Since May 2000, the Federal funds rate was eased 13 times and by some 550 basis points (refer to Graph 2). The reason why the comfortable liquidity position failed to revive investment in capacity has to do with the level of capacity utilization of US industries. Average capacity utilization rate during 1993-2000 was 82.4% for all industries, which fell steadily since May/June 2000 to reach a trough of 74.4% by Dec.2001 and has remained between 74-76 % eversince. Even after the demand for industrial goods started picking up, firm managers have preferred to run-down inventory stocks rather than increasing capacity utilization. This is reflected in the declining inventory/sales ratio which has continued to fall in the first quarter of 2004 to reach the lowest level in the past 8 years (see Graph 5). In such a scenario, employment growth in industry would automatically be sluggish.

To avoid hiring workers, US firm managers have also encouraged existing workers to work extra hours. Just like the slack in capacity, there appears to be a slack in labour use. The average working hours of workers in private industry that had dropped during the course of the recession is now beginning to climb up (see the slightly upward trend in Graph 6 in the recent months). Adjustment by employers of laying-off the less productive workers during the downturn, has supposedly, resulted in increased productivity such that the same output now requires less labour hours ^[10].

During the 1990s the development of the manufacturing sector in the United States was led particularly by the growth in computers and electronic products. In 1990, the computer and electronic products industry accounted for 5 percent of real total manufacturing output. By 2001, its real share had grown to 28 percent – a more than fivefold increase. To a great extent, the tremendous growth in the computer and electronic products industry through 2001 masked declines in other manufacturing industries. With growth, however, came greater vulnerability to changes in the fortunes of this particular industry. When the bubble burst in the demand for semiconductors, computers and telecommunications equipment in late 2000 and early 2001, workers throughout the United States were laid off, generating ripple effects that extended throughout manufacturing and the broader economy. Approximately 19 percent of the 2.4 million jobs lost in manufacturing sector since March 2001 were shed in the computer and electronic products industry ^[11]. Another 11 percent of these 2.4 million jobs were lost in the machinery industry – a sector which includes semiconductor processing equipment. Hence, while manufacturing employment was buoyed by increased demand for high-technology products in the nineties, it now suffers the flipside of this relationship— the downturn in the global demand for computers and electronic products has contributed in large part to the overall slow pace of recovery.

Another structural shift in US manufacturing industries has been the rising competition from cheap imports especially from China. Imports as a share of total demand for manufactured goods in the U.S. has increased steadily over the past few decades, and this trend continues. Between 1997 and 2003 this share increased from 17 percent to 26 percent ^[12]. China has been competing in industries such as textiles, stuffed toys, metal-furnitures etc, industries that are labour-intensive. Though analysts assure that U.S. manufactures are concentrated in capital goods industry, and trade-related lay-offs have been negligible during this recession ^[13], the strong Chinese presence in the labour-intensive sectors does add to the downslide of manufacturing employment.

Manufacturing sector jobs in USA account for only 12% of employed workforce. An overwhelming proportion of the jobs are now in the service sector: service jobs account for more than 80% of the total employment and 78% of GDP. The services witnessed low but non-negative growth in real GDP during the first two years of recession.(see Graph 7) Net jobs lost in the private service sector between 2001 and 2003 was 0.3 million (based on BLS data), which is 1/8th of the job-loss in the manufacturing sector over a similar period. Recent reports speak of a continuing streak of robust growth of the service sector over the past 13 month-period^[14].

Just as manufacturing jobs have been threatened by import competition from China, an imminent threat to service-sector jobs in the US arises from business process outsourcing to countries such as India. The present recession has been an opportune moment for many large service industries to move their operations offshore in search of cheap labour resources. In a static sense this has implied fewer jobs for US citizens. But the corporates assure that the long-run effects of lower labour costs and therefore higher profit margin would show up in higher profitability and lower average prices for the Americans. Forrester Research, a consultancy, guesses that 3.3m American service-industry jobs will have gone overseas by 2015.

Obviously there is clash of interest here: the US multinational lobby is pushing aggressively for more overseas-based operations so that labour costs are minimized, while the domestic worker unions are fighting to retain these jobs within USA in an already job-scarce economy. The developing country governments like India look upon the outsourced jobs as reciprocal gains for opening their economies to both free trade and investment flows from developed countries. The recent legislations against outsourcing imposed by several state governments in the US have invited sharp criticism. Raising protectionist barriers would reduce the employment opportunities for workers in the developing countries.

It must be mentioned here that social safety net, an essential buffer against economic cycles has been steadily withering away in the US. This has considerably added to the woes of the workers and their families. Unemployment insurance which is the only state programme worth its name has benefited less than 40% of the jobless workers during 2000/1 ^[15]. Many states have raised the minimum qualification for unemployment insurance to ridiculously high levels. The root of the problem, the state governments complain, lies in the inadequate financial resources that have kept the present unemployment benefits way below the 1991 level, not to speak of the levels during the recessions of mid 1970s or the early 1980s.

US government largesse on military is not only devastating poor developing nations, but also destroying the lives of ordinary Americans.
 
May 27, 2004.
 

[1] See What Accounts for the Decline in Manufacturing Employment? http://www.cbo.gov/showdoc.cfm?index=&sequence=0
[2] See Dean Baker, Employment Flat in February, Jobs Byte, March 5, 2004.
http://www.cepr.net/Bytes/Jobs_Byte_Mar_5_04.htm
[3] See Robert Pollin(2003) Contours of Descent, Published by Verso for a detailed analysis of the recent tax cuts and its distributional aspects.
[4] NBER dates the present recession to late 2000 when it began; it was over by the end of the year 2001.
[5] Quoted in Gumbel, Andrew (2004) 'How The War Machine Is Driving The US Economy Military Keynsianism Might Get Bush Re-Elected, But It Is Starting To Worry Economists’ in The Independent – UK, 1-9-4 http://www.rense.com/general47/usecon.htm
[6] See The Washington Post, 17 January 1986, cited in http://www.unu.edu/unupress/unupbooks/uu38ne/uu38ne07.htm
[7] See http://www.wand.org/getfacts/index/PentSp.pdf
[8] See Pratap Chatterjee and Herbert Docena (Winter 2003/2004) Occupation, Inc. http://www.southernstudies.org/reports/OccupationInc.htm
[9] Source: same as fn 7.
[10] Over the period 2001 to 2003 manufacturing productivity rose at the rate of 5% (calculations based on value-added data from Federal Reserve Board and hours data from Bureau of Labour Statistics). While the rate of productivity growth has slowed compared to the astounding rate of 6.3 percent experienced between 1996 and 2000, each year fewer and fewer workers are needed in manufacturing to produce the same level of output in value added terms.
[11] Calculations based on data from Bureau of Labour Statistics, and covers the period March 2001 to September 2003.
[12] The worsening trade deficit is evident from Table 2.
[13] The New Protectionism: The great hollowing-out myth, Economist, Feb 19th 2004
http://www.economist.com/printedition/displaystory.cfm?story_id=2446951
[14] http://money.cnn.com/2004/05/05/news/economy/ism_services/
[15] http://www.wsws.org/articles/2001/nov2001/unem-n07.shtml