[Marxism] Sam Gindin view of Delphi concession demands, second wave of attacks on US workers

Fred Feldman ffeldman at bellatlantic.net
Sat Nov 19 13:16:13 MST 2005


~~~~~~~~~~~~~(((( T h e  B u l l e t ))))~~~~~~~~~~~~~~~~~~~~~~~~~~ 

A Socialist Project e-bulletin ....  No. 5  .... November 14, 2005
________________________________________________________________________
____ 


GM, the Delphi Concessions and North American Workers: Round Two? 

--Sam Gindin-- 

It is important to recall that until the 1970s, collective bargaining in
the United States and Canada was largely about workers demanding
improvements from their employers. But a new era in collective
bargaining erupted at the end of the 1970s that was soon dubbed
'concessionary bargaining'. Corporations were now the ones making the
demands. Tensions had been building through the decade, with
corporations increasingly asserting that they could no longer both
maintain profit rates and meet workers' demands and expectations.
Governments also intervened to shift and enforce the balance of power in
society through 'neoliberalism' through the 1980s,transforming the
policy 'rules of the game'. 

We are about to see the second wave of this attack on the North American
working class. The first wave did not, of course, ever subside: the
story of the past quarter century is a litany of increased pressure on
the job, insecurity over keeping decent jobs, longer hours and increased
debt to hang on to consumption standards, concern over social services,
growing inequality, and the weakening of trade unions. This is all about
to get dramatically worse. 

The events signalling this second wave are unfolding in the American
auto industry. The United Auto Workers (UAW), under threat from General
Motors (GM), opened its collective agreement to save GM over $1 billion
in health care costs. UAW is currently in negotiations at Delphi (a
components division spun off from GM in 1999) where the corporation is
threatening to go bankrupt if workers don't cut their wages to $10/hr
from the current $26/hour as well as surrender health benefits and more
or less reduce the union to a dues collection agency that overseas a
non-union workplace. 

None of this is entirely new: American (and to a lesser extent Canadian)
workers in the airline and steel sectors are all too familiar with
concessions linked to corporate restructurings from bankruptcy. But
given the pattern-setting role that the auto sector has always played,
the impact of these latest developments should not be underestimated.
And given that the US-Canadian auto industry is the most integrated
cross-border industry anywhere in the world, workers on the Canadian
side of the border will not be immune from the concessions pressures. 

Several key questions emerge.  Will this be just an isolated bad-news
story -- a tsunami-warning -- that we can only hope misses us or at
least doesn't do any damage? Or will it be a wake-up call warning us
that if you're not fighting back, you're only waiting for things to get
worse? Is what's coming inevitable or can it be resisted? The first
concessions wave led the Canadian Auto Workers (CAW) in a direction that
challenged neoliberalism, free trade and ultimately their UAW parent
union.  What will the CAW response be this time? 

Any effective counter-response will demand some creativity on the part
of workers and their unions, and a crucial beginning is that workers in
the US and Canada start talking about why this is happening and what
might be done. The following points, we hope, might contribute to the
discussion of what an anti-concessions campaign must take up. 


1. You can not privatize the welfare state. 

The first generation of postwar autoworkers used the good times to
achieve a host of social benefits, health care and pensions being the
most important. But in bad times, and especially when the competition
didn't carry the same costs, those benefits came under attack. Defending
them in bargaining had its limits both because the companies were in
trouble and because winning benefits which other workers did not have
left autoworkers relatively isolated. 

The response to the recent GM attack on the health care benefits of
autoworkers should have been - as some American rank-and-file workers
insisted - to call for a national health care program that extended this
crucial benefit to everyone, not taking it away from those happened to
have some protection. The UAW did eventually make this point, but only
after they had made the concessions on cutting healthcare costs for the
company. Had they challenged GM and put the larger issue on the national
agenda, the union might have been a catalyst for a larger struggle (and
for taking a step towards reviving the potential social leadership role
of unions more generally). But declaring this only after the precedent
that UAW workers would bear the costs of cutbacks was set reduced the
UAW statement to empty rhetoric. 

For Canadian workers, this may seem beside the point, as Canada already
has a national health care plan. But isn't that health care system now
under attack? And if this leads to having to negotiate an increasing
share of health care privately with companies, what would happen to the
CAW's ability to negotiate wages and other benefits? More important,
however, is a larger lesson from the UAW concessions: workers and unions
who get too far ahead of other workers when the situation favours them,
will inevitably get in trouble when the winds change. Workers in leading
sectors will eventually be dragged down if other workers are being
pushed to the margins. Progress for workers has to be generalized or
those gains will be vulnerable to reversal. 


2. The problem isn't 'out there' from globalization, it is in North
America. 

The main problem that the GM and Delphi workers face isn't competition
from China or Mexico or even Japan but issues which   can be directly
addressed at home. As Steve Miller, the head of Delphi said in a recent
speech: "...in the auto industry, Toyota, Nissan, and Honda are
competing from assembly plants in our back yard...the old oligopoly has
crumbled, not so much from globalization, but from upstart domestic
competition" (October 28, 2005). In the parts sector, 80% of the US
industry is non-union and many of these plants pay less than half the
wages at Delphi (non-union parts also increase the incentive to
outsource even more from the assembly plants). 

The issue is not so different in Canada where the overall industry is in
fact doing well, but non-union auto majors are winning a larger share of
the market. Here too Toyota and Honda won't be organized through
business as usual and neither will the parts industry, where the level
of unionization was once close to 80% and is now approaching 40%. Unless
the CAW shows the same verve which unions showed in the 1930s when they
were able to organize workers in spite of times much tougher than today
(and in spite of dramatically fewer resources than today's unions),
breakthroughs in unionization simply won't happen. In the 1930s, for
example, mineworkers sent 100 organizers to organize steel workers so
miners would not be isolated. 

Although the auto companies are global, production is overwhelmingly
regional: cars sold in North America are largely assembled in North
America and made of parts produced here. This makes organizing all the
more possible, especially if it is seen in cross-border terms. Why
couldn't the CAW and UAW, for example, jointly declare: that the 10
major parts plants will be organized; that the longer it takes, the more
disruption the entire industry will face; and that there is no point
moving form the US to Canada or vice-versa because we will be there to
organize (and into Mexico as well)? And why would the UAW not put US$200
million of its ever increasing and unused $900 million strike fund to
such use, if only to defend its own members? 


3. Opposition to free trade is nevertheless necessary. 

Blaming globalization and free trade for everything can be a diversion
from more basic issues. Yet corporate mobility does remain a threat and
this will increase as we escalate our fights. If we see the issue not as
other workers taking our jobs, but as the freedom of corporations to do
what they want with production versus the ability of workers to
influence their lives and communities, then fighting free trade is a
matter of democracy (workers' freedom versus corporate freedoms), of
joining with other in the community to fight the unilateral power of
corporations, and of international solidarity to avoid the ratcheting
down all global working standards. 

To limit corporate threats of shutting down plants, it makes sense to
revive a variant of the former Canada-US auto pact and use the leverage
of the market to assert that investing in North America is a condition
for making profits here. Such a pact to constrain corporations and gain
some controls over investment flows would necessarily be extended to
include Mexico and Mexican workers. This couldn't be done alone: it
would mean a commitment on the part of unions far beyond anything to
date to join the global justice movement. In turn, such a campaign might
offer the wider movement the kind of concrete example it needs of
alternatives to free or simply fairer trade in favour of planned trade. 


4. There is a need to question what we produce. 

The big 'no-no' within auto unions in North America is questioning what
kind of products workers are making.  This was not always the case. In
the early 1950s, the UAW was a national leader in calling for small but
safe, fuel-efficient vehicles. Leaving this decision to the companies
has neither helped auto workers nor consumers. Time and again, the
companies gave up on this less-profitable part of the market to
concentrate on higher-profit big vehicles only to see its competitors
use this as a base for taking market share.  Now, an important part of
the problems at the Big Three of GM, Ford and Chrysler are not only cost
but the product. Where are autoworkers on this issue today? 

The issue has been avoided in part because of the belief that the
companies know best and in part because any criticism might hurt sales
and therefore the jobs people depend on. The problem is that whether or
not the companies know what their doing in terms of their own interest,
there is no reason to think it coincides with the collective interest of
auto workers or workers more generally.  And had we been pushing for
vehicles (and an entire transport policy) more sensitive to
environmental concerns - as we were warned to do by environmentalists
pointing to the trajectory of global warming and the inevitability of
rising gas prices - auto and transport sector jobs might actually be
more secure today. 

Consider one example. The Ford engine plant in Windsor makes large
engines. It has been clear for some time that this could not last. Why
is the union not out front mobilizing publicly for Ford to move to
develop new kinds of engines, to convert the Windsor facility to produce
them, and to make any monies given to Ford by the Canadian and
provincial government conditional on such changes? This may not offer
immediate answers to those laid-off, but it would position the union,
both in the community and nationally, as leading on a social issue and
this would be part of developing the capacity to perhaps influence the
direction of Ford and positively affect jobs down the road. 


5. There is space to negotiate decent contracts. 

It needs to be pointed out that the auto industry is not leaving North
America but competing to come in. Overcapacity is more of an issue than
plants leaving. 

In Canada, because of the $0.85 dollar (in terms of the US$) and health
care costs, workers in the Big Three continue to have space to negotiate
decent contracts. AS the CAW pointed out in a recent presentation,
Canadian Big three workers are $10 per hour cheaper (or $20 thousand per
worker yearly) than in the US. It is true that Canadian parts workers
must confront the falling level of unionization in both Canada and the
US. But much of the low-wage, low-capital section of the Canadian
industry departed in the 1980s, leaving an industry that is mid to high
tech in value-added and quality-based. This sector has the advantage
that it is not as easily moved, and that it must be located close to
just-in-time assembly plants.  Of all the vehicles assembled in North
American, 1 in 6 are assembled in southern Ontario, with a huge parts
industry therefore arrayed around Ontario as well. 

A crucial question, however, is what to focus on in bargaining. Working
time stands out for three powerful reasons. First, it is quite amazing
that while productivity has been growing (output per hour has doubled
since the first wave of concessions in the 1980s), workers are left with
less and less of their own time. Second, while higher wages in the Big
Three increase the gap with other workers, more time off is solidaristic
in terms of sharing existing jobs. Third, and this is especially
important in the US, the attempt to limit the impact of job loss through
income security and higher pensions has increased costs for the Big
Three in a way that disadvantages them relative to non-union assembly
plants. But paid time off is something the non-union plants tend to
follow to avoid unionization (or at least it can become a major issue in
organizing). So negotiating paid time off is actually a better response
even from the narrow perspective of 'competitiveness'. 

The time to negotiate paid time off may not be only when things are
going well. It may be that this is more likely to be achieved when there
are layoffs and the issue is solidarity to limit the layoffs (the
original UAW Ford contract in the 1940s provided for going to 32 hour
weeks before layoffs took place, a reflection of the solidaristic
culture then). Solidarity may also be invoked to limit overtime when
some are called back and many remain off work. In most cases, the
company will plan to reduce the workforce even when the upturn comes and
so limiting overtime might become a permanent union policy. 


6. Militancy is not enough. 

Worker militancy is fundamental to everything else. If there no struggle
over everyday issues and wages and benefits, there is unlikely to be
struggles over anything. Parts workers do have power - in some ways even
more power than workers in assembly - because they can shut down a
significant range of assembly plants. 

But militancy itself comes up against barriers that are real and not
just propaganda: non-union plants, products that are not selling,
corporations threatening to move abroad. Sometimes this demands new
strategies. For example, if one plant is constantly disrupting overall
production, the companies may move it. But if disruptions are
strategically spread across various plants at different times, no one
plant can be targeted by the companies. 

This strategy, too, will come up against limits. The key is that when
workers come up against a wall in fighting back, the question must not
be how to retreat but how to knock down that wall or how to scale it.
That is when we have to go beyond the everyday role of the union and
raise larger issues, deepen the involvement of the members, and build
broader class and social alliances. 


7. A rethinking of unions is needed. 

A common thread running through all of this statement is that: (a) in
fighting concessions and building unions, the need is to act now, and
not wait for further initiatives from the corporations; and (b) the
constant importance of building the capacity of workers to respond so we
can, in fact, have more meaningful options in the future. Over the past
almost thirty years of neoliberalism, corporations and the economy have
gone through remarkable transformations. Unions too have changed, but
not always in positive ways and not always in ways adequate to taking on
the new economic and political challenges. It is therefore central to
any successful working class response that workers think about their
unions and ask how they too might be transformed. This is the difficult
but increasingly unavoidable question that workers and unionists, and
socialists and the Left more  generally, cannot avoid. 


Sam Gindin teaches at York University and is retired from the CAW. 

********************************************************************** 

Other Resources on Delphi and the 'New Concessions' 

(1) The Delphi concessions can be found at
http://www.detnews.com/2005/autosinsider/0511/02/A01-369555.htm . 

(2) An important statement on the fightback to the Delphi concessions
can be found at the UAW New Directions Movement web-site at
http://www.uawndm.org/uawndm/retire.htm. 

(3)  A 'Hold Delphi Accountable' petition on the Delphi bankruptcy
proceedings can be signed at
http://www.thepetitionsite.com/takeaction/570770279?ltl=1131905012 or
off the Monthly Review website at http://mrzine.monthlyreview.org/. 

(4) For a discussion of an alternate North American auto industry,
addressed to American, Canadian, and Mexican auto workers, see the
Socialist Project pamphlet on the auto industry: 'Concretizing Working
Class Solidarity: International Solidarity beyond Slogans' at
http://www.socialistproject.ca. 

(5)  For a list of resources and labour reform groups in the US go to:
http://www.uniondemocracy.com/AUDLinks/RNFLinks.htm.

This email was cleaned by emailStripper, available for free from
http://www.papercut.biz/emailStripper.htm





More information about the Marxism mailing list