new unionism>> State of the Unions (Part 2)

State of the Unions - Part Two
by Peter Hall-Jones for www.newunionism.net

The story so far
In the first part of this discussion (available here) we looked at a series of “raw” union membership numbers, from 1998 to 2003, and found that union numbers appeared to have held their ground. In fact membership had gone up in more countries than it had decreased. We also found reason to believe that a large amount of growth was going unrecorded.

We were as surprised by this result as most of our audience. Thanks to Eric Lee at Labourstart the story got great coverage, and the response level was astonishing. Among the many replies we received were no less than eighteen from within the ILO itself. While these are confidential, I think it would be safe to say that the overall message was: “thanks for doing this - we should be doing it ourselves - alas, there is just too much politics involved”. In the meantime two senior ILO figures confirmed this in detail. To mix a metaphor, so politically convoluted is this can of worms that it looks unlikely that the ILO will ever catch it by the tail.

Two external experts, both of whom do statistical work in this area, also contacted us to point out huge flaws in the data. As one of them put it:

In the meantime readers of the story have started to suggest we instigate a global petition demanding a proper study. While we are not necessarily opposed to this idea, New Unionism is a network, not an organisation. If anybody pursues this matter it should be the trade unions themselves, as elected representatives of working people. We would also suggest that the ILO per se is not the problem, so much as certain of its constituents who insist on playing games with figures.

All this aside, our primary concern is that the public is being left with a false picture. The last proper study is now ten years old. At the time it showed that the union movement was in quite serious numerical decline. While this was true back then, indications are that since then the picture has changed, and in fact the decline may have been arrested as far back as nine years ago. Rather incredibly, nobody knows for sure.

In the light of this we decided to write a second part to the article. In doing so we can only repeat, in our rather carping tone, what we said earlier. There is no solid evidence to back up the widespread belief that union numbers are declining internationally!

Moving right along
lies

In this second part of the story we want to look at another angle which the media pushes at every opportunity. "Unions are a thing of the past; this is proven by the decrease in their membership density". Membership density is the proportion of the workforce which is unionised. And yes, it would appear from the figures that in a majority of countries membership density is slowly decreasing. Is this decline in density proof that workers are turning their backs on trade unionism? Yes it is, if you ignore reality. And that is what the media has been doing, and training us to do.

Before going any further let’s look at how union density is measured. If we say that a given country has 20% union density, then what we (normally) mean is that one in five employed workers is a union member. ie

The media tends to swallow such figures whole. While there is nothing inherently wrong with union density as a measurement, the way that it is interpreted can lead to some very false conclusions. Let’s look more closely at each of the three components of the fraction: the numerator, the vinculum, and the denominator.

By the way, isn’t “vinculum” a cool word? I challenge you to find a way of slipping it into a sentence within the next 24 hours. You can use the comments field on the right to tell us how you did it.

The Numerator

This is what we concentrated on in the first part of the story, but let's repeat a small section of the argument here. In retrospect, it almost seems as if we have defined our numerator, in this case, to give the lowest possible result. For instance we don’t count China. While there are very good reasons for being cynical about trade unionism in China, globalisation is changing things. Recent measures to improve workers’ rights there are being pushed hard by the All China Federation of Trade Unions. The opposition is coming not from the Chinese government, but from multinationals who have relocated their manufacturing bases there. Can we really afford to count China’s union membership as zero? While we discount these 150,293,965 people (2005), we simultaneously add up all the unionised jobs which have disappeared into China and put them on the negative side of the ledger! There is no easy way to deal with this question, but in an age of globalisation we must clearly find a more rational accounting system.

Imagine that China suddenly became a fully-fledged capitalist power. Imagine that they then gave unions full autonomy, and employers full permission to give them a kicking. Would it be reasonable to then start counting every single “member” who left (or was discovered to be non-existent) on the negative side of the ledger? This is effectively what we did with Eastern Europe.

On top of this we don’t count most other countries at all, because they do not have the infrastructure needed to maintain solid statistics. These, of course, tend to be the low-wage countries –the logical targets of employers seeking to relocate production. How many of these newly-industrialised workers become union members? Don’t know; don’t count. The fact that our numerator is looking more or less static should be seen as a real triumph for workers and their unions, not a cause for woe!

The Denominator
The trouble with using the denominator as part of the mechanism for evaluating unionism is that it is in the control of employers (including governments, as employers of the public service). If an employer takes on more workers, then union density automatically drops.

Between 1970 and 2000 the size of the world’s labour force more than doubled(1). Over the same period the number of women in the labour force rose by 7% (to 40%)(2). Casual, informal and temporary employment have all become commonplace, and workers have become far more mobile between jobs. Privatisation and outsourcing have also pushed previously accessible workers off the union's immediate radar. While all this has been going on huge sections of production, and many parts of the service sector, have been transferred to the developing world. What this all adds up to is a generation of unprecedented expansion and churn.

The media points to this period as the era of union decline. Technically they are right. But density can only stay level if the union recruits and re-recruits at a level equal to the churn. There was no invisible hand increasing their resource base to allow for this, and so what happened over the last generation was that unions had to find ways to absorb the full impact of this churn. They had to take it on the chin and just do their best to keep up. In some countries they have had to recruit and re-recruit up to 20% of the workforce each year just for their total membership to stand still!

Again, if the ILO and EIRO figures are anything to go by, they have more or less succeeded in doing this.

Visser's figures(3), which are probably the most accurate, report on density over 30 countries. He finds a drop of less than 2%. Given the increase in the labour market over the same period, and the parallel decline in the unionised manufacturing base, this indicates that unions have been doing a lot better than just holding their ground.

...and the Vinculum
As we saw above, that little dividing line between the numerator and the denominator (as in ⅓) is called a vinculum. Let’s imagine a hypothetical workplace with three workers, one of whom (ie ⅓) joins a union. If he/she then recruits a colleague, then the two of them appear as figures in both the numerator and the denominator. In effect, the second member has made a conscious decision to cross the vinculum.

No doubt this all sounds rather silly and self-evident, but when we refer to the real world we soon see that it is one of the key issues when it comes to the misinterpretation of density figures.

Somewhere around the beginning of the 1970s the developed nations reached the point where mass markets were “saturated”. There were enough cars. There were enough washing machines. There were enough toasters. Producers had to stop thinking about how to make production faster and more efficient, and start thinking about how to compete on quality or price. We need funkier cars. We need more buttons on our washing machines. We need cheaper toasters. For that matter, we need 12" platform boots. New strategies started to develop such as “niche marketing” and “product differentiation”. From now on the cake was going to stay more or less the same size, give or take the odd burst of market expansion (remember Nixon’s famous swig of Coca Cola with Mao Tse Tung?). The point of the game became to sneak a bigger slice.

This led employers to start demanding more elbow room. They resented government regulation - it interfered with their ability to shift resources around rapidly, and to implement socially-dubious solutions. They also wanted more flexibility from their workforce. And as for unionisation, well, it was like trying to drive a car with the handbrake on.

Thatcher and Reagan were the darlings of this era. They slashed government controls and tore into trade unions with a vengeance. Did a worker really want to join his/her union, at a time like this? Employers were making it quite clear that unionism had gone the way of the dinosaurs. In many countries that vinculum was starting to look rather like a hurdle. Perhaps one might even say obstacle.

Before this, and it might seem almost impossible to believe this nowadays, governments were generally supportive of trade unions. Here is a collection of quotes from US Presidents:

It was the end of the Cold War which sealed the change in climate. Now, in most countries, governments could fairly openly take the side of employers. The way was open for a boom in neo-liberalism. There was no need to fear the spread of revolutionary sentiments anymore. And besides, if you helped employers enough, wealth would just "trickle down" naturally. And what was good for investors was good for the economy, which in turn was good for the whole nation. There was no need for grand compromises anymore - unless perhaps it was election year. And to the extent that they could sway public opinion against unions, governments could then undermine unions still further. Thus, the last ten years have been a time of general collaboration between governments and employers to weaken the position of trade unions. As one small example, but an interesting one, the very first act of the incoming Bush administration was to cancel the Executive Order mentioned above.

But before we go any further, let’s take a reality check. Sure, we might be able to list a series of barriers, country by country, which have been erected to dissuade or outright prevent workers from unionising. This is familiar ground. But it wouldn't necessarily prove that workers want to join a unions, would it? Perhaps they are just staying away out of choice; because they have simply decided that unions are a spent force?

The Representation Gap
We are indebted here to Harvard’s Richard B. Freeman, among others, for his studies on the “the representation gap”. This is the measure of unsatisfied demand for union membership. It can be defined in various ways, each producing rather different results, but whichever way one looks at it, there seem to be significantly more workers who support unions, and want to join one, than there are actual members.

Let’s look at the United States, for example. According to Freeman’s latest study, based on 2005 data(5), more Americans want to join unions today than at any other time in the last generation. 53% of non-union workers want a union, while little more than a third say they would vote against one.

"If workers were provided the union representation they desired in 2005, then the unionization rate would be about 58% — almost eight times higher than the actual rate of 7.4%.”(6)

As an interesting aside, studies also suggest that the representation gap is greater among young workers (7). This challenges the conventional notion of young people are turning their back on trade unions. A study in Canada, Australia and New Zealand confirmed the finding (8).

But why would so many workers in America, the land of the free, feel they were unable to join a union? Well, for a start, a full 25% of the workforce is now prevented from doing so by law! This figure just kept on growing under the Bush administration. The "Kentucky River" ruling denied the right to unionize professional workers. The "Lechmere" ruling denied non-union members the right to meet with unionized members on company property. But the law is not the only instrument which has been used to shut unions out. In fact much of what has been done is not just in breach of ILO standards, it is in breach of US law. The AFL-CIO has assembled a set of statistics which paint the picture vividly.

As Tom Kochan, co-director of the Institute for Work and Employment Research at MIT said: "It’s a risk, rather than a right, to join a labor union."

At the time of writing the Employee Free Choice Act has passed through the House of Representatives. If passed into law, it would make it harder for employers to intimidate workers who want to join a union. But President Bush has already announced that he intends to veto the Act. As we can see in the case of the USA, the mighty vinculum can be a very, very powerful tool when it comes to keeping union density low.

What about other countries? Is this representation gap just a US aberration? Freeman says: “The gap between what workers want and obtain in representation is greater in the United States than in any other advanced english-speaking country”. Nevertheless, the gap is still very large elsewhere. In looking at five other countries he found(10):

In Canada:
25% of non-unionised workers would, all things considered, prefer to belong to a union. This would almost double the size of the movement (from 29% in 2004)

In the United Kingdom:
10% of non-unionised workers would be “very likely” and 26% “quite likely” to join a union. It is hard to say how “soft” this latter figure is, but we could probably predict that another 20% of the workforce would join. Again, this more than doubles the size of the movement (from 29% in 2004)

In Ireland:
A staggering 64% would be likely to join if management supported unions. This is reduced to 28% if management opposes. If we apply only the latter figure, this would increase the movement’s size by about 75% (from 35% in 2004)

In Australia:
17% would be very likely to join and 22% fairly likely to join. Shall we say that this would amount to a 27% increase? If so, the movement more than doubles again (from 22% in 2004). Since then new legislation has placed heavy restrictions on trade union activities and made it easy for employers to push workers onto individual contracts.

In New Zealand:
11% would be “very likely” to join and 21% fairly likely. Again, this would probably double the size of the movement, though perhaps a little less.

Okay, but that’s just five wealthy and developed nations. Is there a representation gap in other countries? Well, it would appear that employers and governments can work together pretty brutally when it comes to dissuading unionists elsewhere. Here are some examples from 2005(11):

In Bangladesh:
Hundreds of striking workers were injured, some seriously, in three separate attacks by police. Three trade unionists were killed following police intervention in one demonstration.

In Brazil:
Eight people defending the rights of rural workers were murdered. Another employee died of strangulation when a demonstration was broken up by the police.

In Burma:
Ten organisers from the independent national federation FTUB received jail sentences of up to 25 years. One of the organisers died in prison five months later.

In Cambodia:
Police used batons, rifle butts and tear gas against protesters, and union president Chea Mony was forced into exile, fearful for his life. His brother had been murdered the year before.

In Colombia:
Still the most dangerous country in the world to be a trade unionist, in 2005 more than 70 unionists were assassinated, and a further 260 received death threats.

In Ethiopia:
Public sector employees were deprived of the right to form a union, and members of the Ethiopian Association of Teachers were arrested, imprisoned and accused of high treason.

In Ghana:
Three miners were shot when police opened fire on peaceful protestors.

In Honduras:
The regional coordinator of the CGT union centre was murdered.

In India:
Over 300 protesting trade unionists were seriously injured by police.

In Iraq:
At least 13 trade unionists were murdered in an ongoing series of killings.

In the Philippines:
At least four trade unionists were murdered as a direct consequence of their trade union activities.

In South Korea:
The government refused to recognise the newly created migrant workers’ union, the president of which was arrested and savagely beaten by police.

In Turkey:
Police injured seventeen workers when they attacked a peaceful demonstration. In another incident twenty-three trade unionists were beaten and 543 dismissed.

In the United Arab Emirates:
About 130 construction workers, striking over the non-payment of wages, were brutally assaulted by thugs hired by their employer. They were then arrested and prosecuted.

In Zimbabwe:
Zimbabwe Confederation of Trade Unions leaders received death threats, suffered physical assaults, and many were arrested. A teachers’ union leader was tortured and telecommunication workers beaten for protesting against unfair dismissal.

It is impossible to say exactly how trade unionism is affected by atrocities and injustices such as those above, but there are certainly some things we should NOT be saying. For instance when we note that trade union density in Korea fell from 12% in 1997 to 11% in 2003, we should NOT be saying that workers in Korea seem to have made a conscious decision, on the balance of evidence, all things considered, that trade unionism is probably not the thing for them.

There is another point that we need to make about the common (mis)interpretation of union density figures. As Professor John Kelly from the London School of Economics has shown (among others), union density and union influence are not the same thing. France is a clear proof of this... it is not union members who make French strikes so large and successful, it is union supporters. Despite falling density numbers, last year the UK, USA, Australia and France all saw the largest union rallies they had seen in a decade. And when Spanish trade unions defeated the Aznar government’s unemployment reforms by means of a general strike in 2002, they did so with a membership level below 20%.(12)

In fact it is quite conceivable for unions to wield more power as density falls, especially where the fall is generated by state and/or employer interference which forces numbers down but builds increasing sympathy for the goals of the movement. This may well be exactly what we are seeing in the USA. You should think about this very carefully, President Bush, before you wield your veto against the Employee Free Choice Act.

In conclusion
Figures are odd things at the best of times, and especially so when they have politicised vinculi. What we might appear to be saying above is that union density of 20% in 2007 is different to union density of 20% in 1977. This would be absurd. No, what we are saying is that the correct interpretation of the two figures calls for reference to very different contexts; to ignore this means you will misread the results.

Joining a union in 1977 (in most countries) was dead easy - in fact it was often harder not to. They had political and legislative support, and most employers saw them as a socially necessary force. Those days are gone. The representation gap shows us that many, many people want to join unions but can't, or feel they can't. To rather oversimplify the case - the numbers do not prove that trade unionism is in decline, rather than active interference and denial of rights is on the rise.

If current union density figures were used to gauge the undermining of labour, rather than the strength of unions, the reportage around the latest set of statistics which emerged from the USA might have run like this:

The international union movement has come through a whole generation of doom and gloom. It has prevented an endless string of dire prophecies from becoming self-fulfilling. It has emerged into the 21st century with a series of practical initiatives for going global (see box right), and fresh and realistic plans to build worker solidarity across borders. The union movement is alive and well, and is living somewhere near you. Tear down the vinculum - give them a call! Or click here

And next time someone tells you that the trade union movement is a thing of the past, remind them of the story of Goldilocks and the Three Bears. It's a story they will have heard many, many times. The chances are they've also repeated it to others. But did that make them think, even for a moment, that it was true?

1 Harrod, J. and O'Brien, R., "Organized Labor and the Global Political Economy" in "Global Unions: Theory and Strategy of Organized Labor in the Global Politcal Economy", London: Routledge, 2002

2 ibid

3 Visser, Jelle, "A Century of Unions – trade union membership, employment, unemployment and density in thirty countries, 1900-2000", AIAS Working Paper, 2004

4 A copy of this interesting document can be found in New Unionism's Online Library here

5 Freeman, Richard B., " Do Workers Still Want Unions? More then Ever!", see http://www.sharedprosperity.org/bp182.html

6 ibid

7 "Using Freeman and Rogers (1999) definition, we estimate the ‘representation gap", the extent of unsatisfied demand for union membership, at 17.8 percent of the labour force. The gap is greater among younger and lower paid workers in smaller organizations in private sector service industries."
Haynes, P, Boxall, P and Macky, K, "Union Reach, the ‘Representation Gap’ and the Prospects for Unionism in New Zealand", from The Journal of Industrial Relations, Vol. 48, No. 2, 193-216 (2006)

8 “The Australian, Canadian, and New Zealand surveys indicate that young workers, those with low income, and workers in small workplaces are much more likely to have unsatisfied union demand (Boxall, Hayes, and Macky 2005; Campolieti, Gomez, and Gunderson 2005; Teicher et al. 2005). In other words, the more vulnerable groups of workers are less well covered by unions. As Campolieti, Gomez, and Gunderson put it, "Frustrated desire for unions is highest among the youngest and most economically disadvantaged.”
See http://www.press.uillinois.edu/journals/irra/proceedings2005/boxall.html

9 This is from an AFL-CIO leaflet available from our On-Line Library here Sources quoted in the document are:
1 and 3-8: Kate Bronfenbrenner, “Uneasy Terrain: The Impact of Capital Mobility on Workers, Wages and Union organizing,” September 6, 2000. A study of Chicago-area NLRB representation elections by University of Illinois-Chicago professors Chirag Mehta and Nik Theodore reported similar findings. Mehta and Theodore found that workers were fired illegally during 30 percent of organizing campaigns, employers force workers to attend one-on-one, anti-union meetings with supervisors during 91 percent of NLRB representation election campaigns, and employers hire consultants or union-busters to help them fight 82 percent of union organizing drives. See Mehta and Theodore, “Undermining the Right to organize: Employer Behavior During Union Representation Campaigns,” report for American Rights at Work, December 2005.
2. John Schmitt and Ben Zipperer, “Dropping the Ax: Illegal Firings During Union Election Campaigns,” Center for Economic and Policy Research, January 2007, http://www.cepr.net/ index.php?option=com_content&task=view&id=775&Itemid=8
9. National Labor Relations Board annual report, fiscal year 2005, Table 4.
10. Kate Bronfenbrenner, “Uneasy Terrain: The Impact of Capital Mobility on Workers, Wages and Union organizing, Part II: First Contract Supplement,” 2001. According to more recent data reported by the Federal Mediation and Conciliation Service, an even higher proportion of unions newly certified pursuant to the NLRB representation process are denied first contracts by employers: 45 percent. Federal Mediation and Conciliation Service annual report, 2004.
11-12: Peter D. Hart Research Associates, survey for the AFL-CIO, December 2006.
13. AFL-CIo calculation based on Peter D. Hart Research Associates survey, December 2006.

10 Freeman, Richard B., "Do Workers Still Want Unions? More then Ever!", see http://www.sharedprosperity.org/bp182.html

11 This data is all taken from the ICFTU's "Annual Survey of Violations of Trade Union Rights", 2006.
See http://www.icftu.org/survey2006.asp?language=EN

12 Kelly, Professor John, London School of Economics, "Labour Movement Revitalization? A Comparative Perspective", Countess Markievicz Memorial Lecture, Dublin, 7 April 2003