The figure above is a curve which was all the rage after it was published by two European economists in 1988 the Calmfors and Driffill curve. Calmfors and Driffill’s idea was that you could get caught between two stools.
The curve models unemployment against the kind collective bargaining in the labour market. On the one hand so the theory went, you have collective bargaining at the enterprise level as occurred (according to the theory) in Japan. Things work out OK in this scenario. That’s because, although workers have the ability to collectively bargain their collective interest is in not raising wages beyond a level at which the firm the workers work for will encounter strong disciplines on its market share from others in the same industry.
The middle part of the curve is with ‘industry wide’ collective bargaining such as occurred in Australia to some extent before the Accord and to a lesser extent now. We call it ‘pattern bargaining’. This creates a situation where the collective interest of those within the industry is to jack up their wages higher than they would in the first case because, so long as an important part of the competition is with other domestic firms, all their labour costs go up together thus minimising the downsides of higher wage demands to the industry collective that is bargaining.
Then there’s collective bargaining which is centralised nationally. Here all workers have a collective interest in avoiding a wage/price spiral and the result is similar (in macro-economic terms) to the result in the first case. Wage restraint is maintained in this case, not by competitive forces but by collective interest.
I like this approach because it says that competition, collectivism etc are not universally either good or bad, that context matters.
I think of this curve as I try to work out what I think about the labour market reforms.
In this discussion I’m going to be very general and assume labour market regulation is to generate more equal outcomes within the labour market. And I’ll associate high levels of labour market regulation with high levels of unionisation and/or union power in the market.
It seems clear that labour market regulation flattens labour market returns. Just as an example, compare the US where unions have limited power and labour market regulation is week the wage difference between the 10% lowest paid workers and the 10% highest paid workers is 440%. That figure was in a paper from 1999 and is very likely somewhat greater now. In Norway, a country generally considered to be subject to much greater unionisation and labour market regulation, the figure is (for me) a healthier 98%.
So long as I’m happy that this is not doing too much harm to efficiency or to other things I value like people’s basic rights (including their right to the opportunity to be rewarded for their labour), and avoiding the mediocrity which collective bargaining often involves I’m in favour of flattening income distribution compared with the a free market outcome.
If pushed I’d justify this argument in favour of more equal outcomes by saying that (up to some point) inequality of outcomes contributes to inequality of opportunity in a range of ways I won’t go into here. But it’s a subject I’ve tried to discuss a little on this blog and I hope I get time to develop it further.
Now if we assume that the labour market interventions I’m talking about promote greater equality of wage incomes, that is likely to be able to make an important contribution to income equality more broadly within the society. However it should be noted that it is not a ‘direct’ policy which would operate on all incomes and this involves certain costs.
Given this, as John Quiggin showed in a theoretical paper (and good on him for that but it is intuitively clear in any event), labour market regulation can efficiently promote greater equality of income along-side other policies. The reasoning is the same as the reasoning behind the slogan which equates efficiency with broad but low tax distortions the broader the base, the lower the rate.
As taxes distort behaviour roughly in proportion to the square of the tax so there’s efficiency to be had in broadening the base. Likewise there’s efficiency to be had in broadening the base equality promoting policies in our society.
Now assume that labour market regulation generally drives up labour costs. In the absence of other policies this will often create ‘outsiders’ by swelling the ranks of the unemployed. That’s true even at both ends of the Calmfors and Driffill curve, as at these points the collective that is bargaining is a collective of employed people who do not represent the unemployed.
As an aside one may note that while the Accord was in place or at least until towards the end of its life labour market regulation was actually a major vehicle for the gradual reduction of unit labour costs and this was a direct contribution to bringing outsiders inside the labour market. So that was an exceptional use of labour market regulation to which these observations obviously do not apply. (This is not an argument that we should (or should not) still be pursuing centralised collective bargaining, but it is an observation about the way labour market regulation and unions worked at that time).
More generally though the effect of labour market regulation at least outside the corporatist contexts of some (mainly) European countries has been to improve the conditions of labour market ‘insiders’ the employed with the unpleasant and unintended consequence of swelling the ranks of the outsiders.
‘Happiness’ studies not to mention our own intuition confirm that grave unhappiness is visited upon the unemployed. Thus, even assuming relatively inelastic responses to wage rises and other encumbrances on unit labour costs, I am generally much more concerned about one person becoming unemployed particularly for a long period than I am about several people taking a lower wage. (This is in addition to other arguments about the inefficiency of labour market regulation in effecting redistribution. For instance the fact that:
1. a lot of lower income workers are young and will enjoy higher incomes later in life when their productivity rises and
2. others are part of high income households (doctor’s wives is the colloquial expression) even if they will never themselves join the ranks of the highly paid.
What follows from this for me is that, in the absence of really serious supplementary policies to mop up the collateral damage it generates, labour market regulation on its own (at least as it is working in Australia at present) does a great deal of damage to the most vulnerable in our society alongside the good it does for those not made ‘outsiders’.
Now there are countries which use labour market regulation as part of a comprehensive and generous attempt to engineer egalitarian outcomes in the labour market particularly the Scandinavian countries. Here, in addition to a degree of centralised collective bargaining a l¡ Calmfors and Driffill if you’re at risk of becoming long term unemployed, few expenses are spared to help and hassle you into getting work. Money is spent on retraining, job creation and so on. This has both efficiency and equity benefits.
But, though we have sometimes used labour market institutions to internalise the outsiders, Australia is not and never has been a country with a particularly generous set of supplementary policies to ensuring that those who become outsiders are kept in touch with and assisted to get back on the inside.
The closest we got to my knowledge was the “Working Nation” package. One important idea was to bring in a Scandinavian style ‘job compact’. Trouble was it wasn’t funded properly so (from memory) one had to wait for eighteen months before one got onto the program. That’s pretty pathetic. I don’t think it is acceptable to simply ban people from selling their labour at a price they want to without making sure they have all the help (and if necessary hassle) they need to address the situation.
It makes for easy politics and easy thinking to assume that there are one set of policies that are ‘worker friendly’ and another set that are the opposite. I think that depends on context. As Andrew Leigh has pointed out, unfair dismissal restrictions are ‘worker friendly’ to some workers (those in jobs) and not for others (those without). If we’re going to keep them on the scrap-heap to defend the rights of the ‘insiders’ then I for one don’t think we’re doing the right thing.
Here’s a quadrant which summarises some of this and some suggestions as to where different countries fit. I’d welcome thoughts approving and otherwise on my thoughts and suggestions for further development.
Also in light of the passions that the subject (understandably and rightly) arouses, please don’t get into sloganeering.