andrew ross on Sun, 9 May 1999 20:42:19 +0200 (CEST)

[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

<nettime> Sweated Labor in the Garment and High Tech Industries

       Applying the Anti-Sweatshop Model to High Tech Industries 	
                         by Andrew Ross
          (published in Spring issue of New Labor Forum)

	A few years ago, Suck, the irreverent webzine in San Francisco
( that offers daily comments on the latest media buzz,
cunningly revealed that the staff of Wired magazine occupied a floor in a
building full of garment sweatshops.  Wired had been the town crier of the
information revolution, and an obliging booster for the global chamber of
commerce that lives or dies by the Nasdaq index.  The sweatshops were a
chronic outbreak of an old disease that had once been contained, not
unlike tuberculosis itself, the historic scourge of the
turn-of-the-century garment industry.  Here, then, we had a picture of
geeky Gen X wordsmiths testdriving the latest software geewhizzery just
fifteen feet above the traditional labors of greenhorn immigrant
seamstresses.  Suddenly, the century-long gulf between the postindustrial
high tech world, for which Wired is the most glittering advertisement, and
the preindustrial no-tech worldfor which the sweatshop is the most sordid
advertisement--appeared to have dissolved.  In San Francisco's
"Multi-Media Gulch," where parcels of the new Internet industries are
located, this kind of juxtaposition is not uncommon.  But the scale of the
contrast between these two workplacesbetween the nineteenth and the
twenty-first centuryis much more pronounced in New York City, where the
ragged strip of Silicon Alley cuts through areas of old industrial loft
space that were once, and are again, home to the burgeoning sweatshop
sector of the garment industry.  Manhattan's downtown concentration of
Webshops--those much-romanticized laboratories of the brave new
technological futureshares much the same urban space as the new
sweatshops, where patterns of work for large portions of the immigrant
population increasingly resemble those in the very early years of the
century, before the first elements of industrial democracy were adopted
into law. 
 	As it happens, the comparison between these workplaces had its echoes in
the earlier period.  Then, the sweatshop's primitive mode of production
and the cutter's artisanal loft co-existed with semi-automated workplaces
that would very soon industrialize into economies of scale under the pull
of the Fordist factory.  Today, the sewing machine's foot pedal is still
very much in business, no longer competing with steampower, of course, but
with the CPU, which, at the higher end of garment production, is used to
govern Computer-Assisted Design, ensuring fast turnaround and just-in-time
supplies for the volatile seasonal trade in fashion lines.  In fact, the
sewing machinethe basic sweatshop technologyhas barely changed in almost a
hundred and fifty years, which affords it a rather unique place in
industrial history.  The main reason for its long survival has to do with
the physical limpness of fabric.  As a result, there is a portion of
garment production that cannot be fully automated, and that requires human
attention to sewing and stitching and assembly.  With a cheap labor supply
in abundance, developing countries often begin their industrialization
process in textiles and apparel, because of the low capital investment in
the labor-intensive end of production.
	In many of the free trade export zones of the developing world,
workplaces for electronic assembly keep company with garment assembly. And
in the industrial sectors of many other countries, workers in chip
manufacture, data processing, and digital programming experience many of
the same debilitating barriers to fair labor standards as do garment
workers.  Yet there has been much less attention to these workplaces in
the public mind and media.  The glamor of high technology carries a
powerful mystique, and its preeminence has eclipsed the capacity of moral
abhorrence to rectify exploitation.  Are there lessons from the recent
anti-sweatshop campaigns that can be applied to combat labor abuses in
this other industry?
	There are many reasons for the flourishing of garment sweatshops,
both in the poor countries of the world, and in the old metropolitan
cores: these include regional and global free trade agreements, the advent
of subcontracting as a universal principle, the shift of power towards
large retailers and away from manufacturers, the weakening of the labor
movement and labor legislation, and, last but not least, the transnational
reach of fashion itself, especially among youththe international mass
consumer wants the latest fashion post-haste, which necessitates
turnaround and flexibility at levels that disrupt all stable norms of
industrial competition. 
	This is not the place to describe any of these features in any
depth, and I will assume that most readers of this journal are familiar
with them.  Indeed it's fair to say that the larger public awareness of
the conditions of low wage garment labor is relatively far advanced.  The
public at least recognizes the situation even if it chooses, for the most
part, to ignore that much of the clothing we wear is made illegally and
under atrocious conditions.  The anti-sweatshop movement of recent years
has been immensely successful in penetrating the public domain of media
space.  Part of this success is due to the repugnance attached to the term
"sweatshop" which commands a moral power, second only to slavery itself,
to rouse public opinion.  But a more significant factor has been the
elastic nature of the movement's coalition-building.  The involvement of
interfaith and human rights groups, students, and NGOs alongside workers
has offered a stunningly effective model for transnational activism. 
These coalitions have demonstrated that organized labor cannot go it
alone, and that there is no alternative, in the age of global economies,
to this kind of activism.  Ultimately, of course, the endpoint must lie
with an appreciable alteration of consumption.  The challenge now lies in
making an impact at the point of sale, which is to say, reforming consumer
psychology to the level at which criteria of style, quality, and
affordability are all well served by appeals to the advantages of paying a
living wage.  Surprisingly, we are further forward than anyone could have
imagined just a few years ago.
	The same cannot be said of high technology.  The gulf between the
fashion catwalk and the garment sweatshop is nowhere near as great as the
gulf between the high-investment glitz and mega-hertz wizardry at the top
of the cyberspace chain, and the electronic sweatshops at the bottom. Why
is this the case? 

                                  Whizzkids or Cyberdrudges?

	Cyberspace, for want of a better term to describe the virtual
world of digital communication and e-commerce, is not simply a libertarian
medium for free expression and wealth accumulation.  It is a
labor-intensive workplace.  Masses of people work in cyberspace, or work
to make cyberspace possible, a fact that receives virtually no recognition
from the so-called digerati like John Perry Barlow or Kevin Kelly, who
edit, and write for Wired magazine, let alone the pundits and managers who
are employed to pump hot air into the great Internet Stock Bubble. Indeed,
it's fair to say that most people have little real sense of the material
labor that produces their computer technologies, nor are they very
attentive to the industrial uses to which these technologies are put in
the workplaces of the world.  Like all other sectors of the economy, the
high tech industries have been penetrated by the low-wage revolution--from
the janitors who service Silicon Valley in California to the part-time
programmers and designers who service Silicon Alley in New York.  Just as
Silicon Valley once provided a pioneering model for flexible
post-industrial employment, Silicon Alley may be poised to deliver an
	 New York's new media sector has seen the biggest job growth of any
urban industry in the metropolitan area for decades.  Web design,
programming, and marketing in this industry accounts for over 100, 000 new
jobs in the last three years.  But these are entirely non-unionized
workplaces, plagued by overwork, low wages, flexible contracts, and job
insecurity.  Over half of the jobs are filled by contract employees or
perma-temps, with no benefits, or paid vacations.  70 hour weeks without
overtime are a way of life for Webshop workers who are commonly signed up
on the promise that their "sweat equity" will lead to a pot of stock
proceeds at the end of the rainbow.  For the majority, the stock plan
which will deliver the big payoff never comes.  Startups have a high
failure rate, and those that merge with giants like AOL, Microsoft, and
Sun generate fortunes for their investors and founding entrepreneurs but
seldom any returns for employees.  Upgrades in software programs can put
skilled workers out of a job overnight.  The average salary for
full-timers on Silicon Alley is about $37k, which is hardly a starvation
wage, but still well below old media industries, like advertising, at
$71k, and television broadcasting, at $85K. 
	Silicon Alley is the face of the future, not because its whizzkids
are designing a brave new wired world, but because this new industry is a
prescient example of the kind of workplace that skilled technicians will
increasingly confront.  Its structures of capital financing, and the sheer
volatility of its labor landscape make the New Media workforces
notoriously difficult to organize.  While many traditional labor unions
have been slow to use Websites and Internet technologies in their own
organizing efforts, organizing is almost non-existent in the Internet
industries themselves.  In January 1999, over thirty unions sponsored a
Labor Online conference, organized by the Brooklyn College Graduate Center
for Worker Education, to chart the labor movement's use of high tech, and
to examine the future of organizing in high tech
(  Union reps showed up in force, but there
were very few employees from Silicon Alley. 

                                Going Down the Chain           

	Many of the readers of this article probably want their computers
to go faster, and yet most people who work with computers, or according to
schedules set by computers, already want them to go slower.  For those who
view the persona l computer as an artisanal tool of comparative advantage
with which to compete in the field of skills, resources, and rewards, it
makes sense to respond to the heady promise of velocification in all of
its forms: the relentless boosting of chip clock speed, of magnification
of storage density, of faster traffic on Internet backbones, of higher
baud rate modems, of hyper-efficient data-base searches, and rapid
data-transfer techniques. 
	For workers who are not masters of their own work environment, the
speed controls of technology are routinely used to regulate their labor.
These forms of regulation are well documented: widespread workplace
monitoring and electronic surveillance, where keyboard quotas and other
automated measures are geared to time every operation, from the length of
bathroom visits to the wasted productivity claimed by personal e-mail.
Software is programmed to control and monitor task performance.
Occupationally, this world stretches from the high-turnover
burger-flippers in MacDonalds to the offshore data entry pools in Asia and
the Caribbean (and, arguably, stretches all the way to include
front-office managers, who complain about their accountability to
inflexible productivity schedules).  Low-skilled information processing,
in particular, is characterized by musculo-skeletal, and psychological
disorders, chronic stress and fatigue, and reproductive problems. Women
occupy a huge majority of low-skilled offshore data entry jobs, while men
dominate the professional end.  In general, advanced automation has
enabled the global outsourcing of low-wage labor, and the wholesale
replacement of decision-making by expert systems and smart tools.  Our
processed world thrives on undereducation, undermotivation, and
underpayment; and it appears to be primarily aimed at the control of
workers, rather than at tapping their potential for efficiency, let alone
their native ingenuity. 
	For most low-wage employees who work with computers, there is
simply nothing to be gained from going faster; it is not in their
interests to do so, and so their ingenuity on the job is devoted to ways
of slowing down the work regime, beating the system, and sabotaging its
automated schedules.  The cumulative loss of productivity from computer
downtime caused by worker sabotage on the job is one of the biggest of all
corporate secrets.  This alone would help explain the "productivity
paradox"; it has yet to be empirically proven that the introduction of
information technology into workplaces boosts productivity.
	If we go further down the chain of high tech production, we find
ourselves in the semiconductor workplaces, where the operating machinery
of computers is manufactured in the least unionized of all goods-producing
industries.  While 56.2 percent of steel workers, 54.6 percent of
automobile workers, 43.8 percent of telecommunications workers and 23.7
percent of workers in durable goods manufacturing overall are unionized,
only 2.7 percent of workers in electronics and computer equipment belong
to unions.  One of the world's fastest growing industrial sectors,
semiconductor factories (or "fabs") have mostly been concentrated in the
U.S. West Europe and Japan, but the passage of NAFTA and GATT has extended
the mobility of the industry.  Lower wages and weak environmental
standards in Southeast Asia, Central America and the Caribbean have drawn
off much of the new investment.  Even within the U.S., it is the
Southwest, with its sparse union activity and softer environmental and
safety regulations that has attracted many of the new fabs.  The new
destinations for "toxic flight" are Albuquerque, Phoenix, and Austin. 
	The hazards to workers and to the environment in the sterile,
dustless "clean rooms" of these fabs (designed to protect silicon wafers,
and not workers) are already excessive, and likely to multiply with each
new generation of components.  Semiconductor manufacturing (which produces
over 220 billion chips a year) uses more highly toxic gases (including
lethal ones like arsine and phosphine) and chemicals than any other
industry, its plants discharge tons of toxic pollutants into the air, and
use millions of gallons of water each day; there are more ground water
contamination and EPA Superfund sites in Silicon Valley than anywhere else
in the US.  The ecological footprint of a single silicon chip is massive. 
Despite the public perception that these are light manufacturing
workplaces, microchip workers suffer industrial illnesses at 3 times the
average for other manufacturing jobs, and studies routinely find
significantly increased miscarriage rates and birth defect rates among
women working in chemical handling jobs.  The more common and
well-documented illnesses include breast, uterine and stomach cancer,
leukemia, asthma , vision impairment, and carpal tunnel syndrome . In many
of these jobs, workers are exposed to hundreds of different chemicals and
over 700 compounds that can go into the production of a single work
station, destined for technological obsolescence in a couple of years12
million computers are disposed of annually, which amounts to 300, 000 tons
of electronic trash that are difficult to recycle.  Very little
occupational health research exists that analyses the impact on the human
body of combining several of these compounds, and research on reproductive
hazards, in particular, has been seen as a women's issue and is therefore
underfunded and underreported. 
	Increasingly, the "dirtier" processes of high-tech production are
located in lower income communities with large immigrant populations in
the U.S. or are being dispersed throughout the developing countries,
augmenting existing patterns of environmental and economic injustice. With
them will go modular industries like the printed circuit board sector, and
other electronic assembly operations where immigrant workers, employed at
rock bottom wages with even less benefits than in the clean rooms, use
solders and solvents that are almost as toxic as those handled by chip
workers.  Large U.S. and Japanese companies are steadily relocating their
fabrication lines to Thailand, Vietnam, Malaysia, Indonesia, Mexico,
Panama, Costa Rica, and Brazil.  In ten to fifteen years time, the
geography of high tech global production may increasingly resemble that of
the garment industry.
	Fabrication flight is accelerated at the least sign of an
organizing drive with teeth.  Nonetheless, groups like the Silicon Valley
Toxics Coalition (see have played a leading role in
coordinating semiconductor activism, formulating the Silicon Principles,
and petitioning companies like Intel, GTE and Motorola to establish a Code
of Conduct to protect the health, safety and human rights of workers in
its factories, and of communities where the factories are located. Through
the Campaign for Responsible Technology, an international network is now
being formed to make links with local labor, environmental, and human
rights groups around the world.  Much of the groundwork for this was laid
at a European Work Hazards convention in Holland in March 1998, which
brought together activists with the common goal of holding companies to
Codes of Conduct through the acceptance of independent workplace
monitoring.  Clearly some kind of local accounting is needed since
transnational companies tend to export hazards to countries where labor is
least organized and where media and government whistleblowing is least

                                   Sullying the Name

	The model for such a campaign already exists in the anti-sweatshop
movement.  Struggles for a workable code of conduct that respects worker
safety, union rights, and a living wage already have a proven record of
successes and failures.  One of the more successful pressure points has
been to tarnish the company name with high-profile media exposes.  The
integrity of a company's brand name is all-important and its vulnerability
is amply reflected in management's skittishness about sour publicity.
Companies must keep their brand names clean, because it is often the only
thing that distinguishes their product from that of their market
competitors.  If that name is sullied it does not matter whether they have
access to the very cheapest labor pool in the worldall is lost. 
	The most recent round of anti-sweatshop campaigning has involved
the integrity of some of the more prominent U.S. varsity names in a
$2.5billion sector of the garment industry.  In the winter of 1998-99,
college presidents were asked to review and sign a code of conduct
governing the labor conditions under which licensed articles, bearing the
college name, are manufactured.  This code of conduct had been prepared by
the Collegiate Licensing Company (CLC), and was loosely based on the set
of regulatory provisions drawn up by the Apparel Industry Partnership
(AIP), a task force of garment companies, organized labor, and human
rights and religious groups convened by the Clinton administration in
1996.  At this point, virtually all of the labor, religious and human
rights groups had withdrawn from the AIP, in the understanding that its
provisions are unenforceable. It is now widely accepted that the AIP
agreement is virtually consistent with current industry policies, and will
not have any appreciable impact on the exploitative practices associated
with those policies.  The CLC Code ran into student opposition for much
the same reasons, relating to the same sticking points: a) Absence of a
"disclosure" provision, whereby garment companies would be required to
disclose the locations of factories where their products are manufactured
or assembled. Without disclosure, it is impossible to monitor the working
conditions at these factories. 
 b) Absence of a provision guaranteeing a living wage, as opposed to a
minimum wage.  In most countries, including the U.S., the minimum wage
does not ensure subsistence levels of living for working families, and is
therefore a subpoverty wage.
	The national mobilization of students, through United Students
Against Sweatshops, resulted in major concessions on both of these points. 
At many campuses, students secured agreements similar to an initial
settlement reached with the Duke University administration. Colleges would
pull out of the CLC if full disclosure of factories has not occurred
within twelve months.  At the same time, they have committed to funding
interim studies on the criteria for a living wage in countries around the
world.  Plainly attentive to the need to keep their names sweat-free,
college presidents were pulled into the bargaining pit, while the media
coverage of student sit-ins and mass rallies expanded public consciousness
of the issues.  The campaign reaffirmed what was rapidly becoming common
sense--no one wants the name of their company or institution mentioned in
the same paragraph as that of Nike.
	Will the same strategies work in high-tech industries? There is no
reason why the brand names of AT&T, Phillips, Intel, IBM, Hewlett Packard,
Toshiba, Samsung, and Fujitsu cannot be publicly shamed in the same way as
Nike, The Gap, Guess, Phillips Van Heusen, and Disney.  Their name
recognition has long been a fixture of the mass market, and is
increasingly a mark of distinction on the advertising landscape.  Above
all, it is important not to underestimate public outrage.  Far from
apathetic, public concern has been inflamed by revelations about labor
abuses in the industrialized and non-industrialized world, where workers
are physically, sexually and economically abused to save 10c on the cost
of a pricey item of clothing.  Unlike clothing, consumption of high tech
goods is not yet a daily necessity, but the rate of market penetration in
the last twenty years has been phenomenal.  It cannot be too long before
high tech household items are as disposal, and subject to the same
volatile seasonal turnover, as fashion goods.  At that point, the high
tech market will be fully within the orbit of consumer politics on the
scale of boycott threats.
	In concluding, perhaps it is worth reviewing why so little
attention is paid to high tech labor issues in the flood of commentary
directed at cyberspace.  One reason certainly has to do with the lack of
any tradition of organized labor in these industries.  The fight against
the garment sweatshop was a historic milestone in trade union history, and
gave rise to the first accords on industrial democracy.  So, too, the
recent campaigns have been on the leading edge of the resurgent labor
movement, at least in the US.  Nothing comparable exists in the high tech
workplaces of the new information order..  They have emerged in a climate
intrinsically hostile to the principles of trade unionism.  Indeed,
high-tech industry lobbyists have been zealous leaders in efforts to
undermine the existing protections of labor legislation.  A second reason
has to do with ideology of the clean machine.  In the public mind, the
computer is still viewed as the product of magic, and not industry. It is
as if computers fall from the skies, and they work in ways that are
entirely beyond our understanding.  The fact that we can repair our car
but not our computer does not help.  As a result, the process of
manufacturing is obscured and mystified.  A third reason probably has to
do with the special treatment afforded to microelectronics by state
managers, aware of the industry's strategic importance to the national
economy.  Currently, nations assess their competitive standing in world
trade by the growth of their advanced technology sectors, and so
government's regulatory eye often looks the other way when abuses and
hazards proliferate.  Lastly, there is the utopian rhetoric employed by
the organic intellectuals and pundits of cyberspace.  Choose any one of
the bestselling books that extol the praises of the new frontier of
virtual life.  Chances are there will be no mention of the crippling
workplace injuries sustained in manufacturing of the new clean machines,
nor any recognition of the cruel outsourcing economies that export
low-wage labor to piecework contractors, whether local or far-flung. There
is a complete and utter disconnect between the public discourse of the New
Media whizzkids, intellectuals, and entrepreneurs, and any awareness of
high-tech workplace hazards.  As long as we separate the world of ideas
and high-tech buzz from the testimony and experience of the workplace,
people simply will not make connections between the two. 
	The successes of anti-sweatshop garment organizing have come as a
surprise to many seasoned activists, long accustomed to being shut out of
the media, to dealing with the often stony indifference of the public, and
to watching the cruel march of corporate armies across the killing fields
of labor.  In the case of information technology, the time is ripe for
capitalizing on the climate for such successes.  Perhaps we can exercise a
little foresight, and anticipate the public appetite for responding to
such abuses.  The history of the Internet should remind us that nothing is
impossible, and what was unimaginable just a few years ago is a fact of
life today. 


1. "Global Labor Standards and the Apparel Industry: Can We Regulate
Production?  Harvard Trade Union Program, Harvard University (October
1998); Andrew Ross ed., No Sweat: Fashion, Free Trade and the Rights of
Garment Workers, (New York: Verso, 1997). 

2. Estimates based on two annual reports on new media employment, Coopers
and Lybrand. New York New Media Industry Survey: Opportunities and
Challenges of New York's Emerging Cyber-Industry (New York: New Media
Association, 1996 and 1997). The most significant figures in the report
show that the number of fulltime jobs increased by 28% from year-end 1995
to 1997, compared to a 44% increase for freelance jobs and 162% for

3. See Andrew Ross, "Jobs in Cyberspace," in Real Love: In Pursuit of
Cultural Justice (New York: NYU Press, 1998), pp. 7-34. 

4.  See Austin Bunn, "No-Collar Workers: Is There Room for Unions in the
New Media World?" Village Voice (January 12, 1999).  Two active high tech
organizations include FaceIntel, or Former and Current Employees of Intel,
begiun by ex-engineer Ken Hamidi, who operates a web site
( that documents complaints against the chip
manufacturer; and Mike Blain's WashTech union for Microsoft's 6,000
perma-temps ( . 

5. Ruth Pearson and Swasti Mitter, "Employment and Working Conditions of
Low-Skilled Information-Processing Workers in Less Developed Countries,"
International Labour Review, 132, 1 (1993), pp. 49-58

6. Barbara Garson, The Electronic Sweatshop (New York: Simon and Schuster,

7.  Chris Carlsson and Mark Leger eds., Bad Attitude: The Processed World
Anthology (London: Verso, 1990). 

8. David Bacon, "Silicon Valley Sweatshops:  High-Tech's Dirty Little
Secret," The Nation ,256, 15 (April 19, 1993), p. 517. 

9.  Lenny Siegel, The High Cost of High Tech (1985): Dennis Hayes, Behind
the Silicon Curtain: The Seductions of Work in a Lonely Era (Boston: South
End Press, 1989). 

10.  Leslie Byster, "The Toxic Chip," Environmental Action, 27, 3 (Fall
1995), pp. 19-23. 

11. Kevin Kelly, one of the founders of Wired, and the former publisher
and editor of Whole Earth Review, is a good example.  See his book Out of
Control: The Biology of Machines (London: Fourth Estate, 1994)

Andrew Ross 
Director, American Studies Program
New York University 285
Mercer St. 8th Flr. 
New York, NY 10003

Tel: 212-998-8538
Fax: 212-995-4371

#  distributed via nettime-l : no commercial use without permission
#  <nettime> is a closed moderated mailinglist for net criticism,
#  collaborative text filtering and cultural politics of the nets
#  more info: and "info nettime-l" in the msg body
#  URL:  contact: