Bahnisch & Quiggin on IR changes (crossposted at LP)

Here’s a paper Mark Bahnisch and I have written on the government’s proposed IR changes, aimed at Online Opinion. We’ll try for a more complete analysis when we both get a bit more spare time. Comments appreciated. Crossposted here at LP

Justice Henry Higgins, the architect of the 1907 Harvester Decision of the Arbitration Court which enshrined the principle of a living wage, defined the scope of the Australian arbitration system as a “new province for law and order�. Almost a century on, John Howard claims that industrial relations reform is “one of the great pieces of unfinished business in the structural transformation of the Australian economy�. The great strikes of the 1890s demonstrated that a labour relations regime based on the law of master and servant was untenable as a matter of justice. Federation saw the entrenchment of Commonwealth powers to establish institutions to conciliate and arbitrate industrial disputes across state borders. The States followed suit with their own dispute settling and wage-fixation tribunals. Distinctive about the Australian arbitral model was its recognition of unions as bargaining parties with legal rights, and wage-fixation on national and industry levels through awards. Awards also comprehensively stipulated employment conditions. Over-award collective bargaining was commonplace particularly in industries with well-organised unions and tight labour markets. But overall, the system delivered high wages and employment security, and was supported by both the ALP and the Liberal Party and in general by employers.

Economic turbulence in the 1970s, and a shift in the labour share of GDP and increased bargaining power by unionised workers led to a backlash. For much of the 1980s, increases in wages arose primarily from annual national wage cases, in which the crucial determinant of the outcome was bargaining between the ACTU and the Hawke Labor government under successive versions of the Accord. As the Accord process came under strain in the late 1980s, individual unions sought to regain a more prominent role in wage bargaining, while policymakers sought to increase flexibility. The outcome was the system of enterprise bargaining, in which unions reached agreements with individual employers.

Reflecting the political compromise that led to the adoption of enterprise bargaining, unions focused on the bargaining aspect, while employers focused on flexibility and the opportunity to buy out restrictive award conditions. The election of the Howard government saw three key changes to industrial law. The vestiges of compulsory unionism were swept away, and unions kept on a very tight rein by the Office of the Employment Advocate, and an individual stream of bargaining – Australian Workplace Agreements – was introduced. Federal Awards were simplified to 20 allowable matters. One of the most significant changes was the removal of restrictions on casual employment, which led to an increasing casualisation of the service workforce. The actual take-up of AWAs by employers has not been impressive – except in sectors such as the Federal public service and communications where government pressure played a major role, and in mining where the introduction of AWAs has been the culmination of a decade long de-unionisation campaign by major firms. The AIRC proved more interventionist than anticipated, continuing to set new employment standards through test cases (for instance, redundancy payments in small businesses which the government is keen to wind back). Overall, real wage rates and employment have improved under the Howard government, but with very patchy gains in some sectors, and a continuing rise in inequality. The improved bargaining position of employees as labour market participation increases and skill shortages bite is no doubt a motivating factor in business’ desire for further IR reform, a desire that can now be fulfilled with the Government’s Senate majority.

Unlike the period leading up to the introduction of Enterprise Bargaining in 1991, there have been few sustained attempts to make the policy case for reform. The BCA embarked on a major research programme in the late 1980s with numerous studies arguing for deregulation. As David Peetz has argued, studies recently commissioned do not bear out the arguments for productivity which dominate the rhetoric of the Government and business. The consensus of the academic literature is that high performance work places are perfectly compatible with active unionism and sensible collective bargaining. Nevertheless, Howard’s longstanding desire to reform IR now has a chance of fulfilment, despite the fact that a political storm over the mooted changes is fast gathering.

The reform proposals have two main elements. The first is the amendment of unfair dismissal laws, so as not to apply to enterprises with less than 100 employees.

The second element, is a substantial extension of previous trends, aimed at reducing the role of unions, awards and arbitration, while increasing that of direct contracting between employers and individual employees. In particular:
1. A number of conditions are to be removed from the scope of awards;
2. Minimum conditions will be set by legislation rather than through the awards process;
3. Minimum wages will be set by the Fair Pay Commission, rather than the AIRC;
4. State tribunals will be abolished and replaced by a single national system
5. EBAs as well as AWAs will be assessed by the OEA and both will require substantially less scrutiny.

Most significantly, the reframing of the no disadvantage test will allow AWAs to undercut award minima, effectively making the award protections optional for employers. This key change has attracted little comment to date.

It is difficult to make definitive predictions about the results of particular changes in industrial relations systems. There are a number of reasons for this, including the inherent complexity of employment relationships, the fact that systems differ between jurisdictions and over time in ways that cannot easily be measured and the confounding effect of changes in the labour market, including those associated with macroeconomic cycles.

This point can be illustrated in relation to unfair dismissal laws. A priori arguments are inconclusive Supporters of such laws make the point that, other things equal, the easier it is to dismiss employees, the higher will be the rate of dismissal, and therefore the higher the level of unemployment. Opponents counter that employers will be unwilling to take on staff if they are unable to dismiss those who turn out to be unsatisfactory.

In economic terms, the problem starts with the fact that an employment contract has a lot of implicit terms. In the management literature, it’s often referred to as a psychological contract. Once both parties have committed to the relationship, each has the opportunity to cheat on these commitments. How this works out depends on institutional rules, the state of the labour market and so on. Whatever happens there are going to plenty of people who perceive the outcome as unfair, and plenty of cases where this perception is accurate.

What about the empirical evidence? As often happens, the literature on unfair dismissals starts out with a big publication finding clear-cut results, only to descend into a morass of contradictory findings. A paper by Lazear in 1990 found strong negative correlations between the strength of employment protection laws, proxied by severance pay, and desirable labour market outcomes such as employment and participation rates, hours worked and so on.. But Lazear’s results have not stood the test of time. More recent research suggests that employment protection laws lower the variance of employment and unemployment but have no clear effect on average levels.

In a comparison between neoliberal labour market institutions and alternatives involving either collective bargaining or centralised wage-fixation, one feature is clearly evident. Neoliberal institutions produce substantially more unequal outcomes. This is evident both from comparisons over time and from comparisons between countries. The US, where the labour market has always had most of the main neoliberal characteristics, displays easily the highest inequality. The reforms undertaken in New Zealand and the UK show up clearly in rising levels of inequality, overtaking European countries that were initially less egalitarian. In the US, where declining rates of unionisation and an even more extreme form of neoliberalism have produced a dramatic shift in the distribution of income. Low-income families have experienced almost no income growth since 1970. Wages for workers with high-school education or less have actually fallen, but this has been offset by longer hours of work and increased female participation.

Defenders of neoliberal institutions argue that growth in inequality has been offset by stronger employment growth resulting from more flexible labour markets. The evidence is decidedly mixed. Until the present business cycle, from early 1990s, there was little evidence to support it. For most of the past decade, the English-speaking economies have outperformed those of the EU and Japan. It is too early to judge whether this is merely the outcome of cyclical timing, or whether there is a sustainable gain in employment. Our view, based on the huge current account deficits being run by all the English-speaking countries, is that a severe cyclical correction lies ahead. Only when the macroeconomic imbalances have been resolved will it be possible to make a clear judgement.

Employment relationships are complex, but the outcome of bargaining depends on two factors. The first is the state of the labour market. The second is the balance of bargaining power. Usually, the state of the labour market is more important, but it’s largely determined by exogenous macroeconomic shocks originating not in the labour market but in the financial sector or the world economy. The reforms proposed by the Howard government will tilt the balance strongly in favour of employers. The likely outcome is a substantial increase in inequality of incomes, and in day-to-day relationships within the workplace.

65 thoughts on “Bahnisch & Quiggin on IR changes (crossposted at LP)

  1. And all the while the neglected, underpaid labourers are continuously becomming increasingly poorer and unhappier. When will our government listen to the voices of our people and equalise our rights? Probably not until they step down from the parliamentary red carpet and into the labour force for a taste of their own medicine.

  2. To Mark Culley – brilliant. I really like your point about the “indirect discrimination” – very good argument.

    I just wanted to add that employees should not have to bear the brunt of ignorant employers. It is actually not that difficult to dismiss a poorly performing employee.
    1. Make sure they’re aware that they’re not performing properly and tell them why – no, it doesn’t have to be in writing.
    2. Make sure you given them a chance to improve.
    3. Make sure you tell them that their job’s on the line if they don’t improve.
    4. After they’ve had a reasonable time to demonstrate improvement and if they haven’t improved – terminate their employment and give them appropriate notice.

    Simple really.

    How come small business can manage to get their tax arrangements right but they don’t seem to understand simple employment law.

    Also, the economic arguements are irrelevant. Ask yourself – do you really want to live in a country that endorses unfairness – to use John Howard’s words – it’s unAustralian.

  3. “More recent research suggests that employment protection laws lower the variance of employment and unemployment but have no clear effect on average levels.”

    If we care about the unemployed more than the employed, and about the longterm unemployment more than the shorterm unemployed, then is a good thing, right?

  4. James Farrell, I don’t follow the “therefore”. Privatisation of gains could describe any profit share/wages share regime for dividing up economic growth or national income so long as tax rates are not too high. High wages are still private income. So how does privatised gain link to an equality/efficiency trade-off? Or does “equality/efficiency trade-off” just mean a consumption/investment trade-off?

  5. Andrew, I’d prefer to promote churn through wage subsidies than through facilitating dismissals, even if the equilibrium effect was the same.

    But on reflection, I think this result must refer to restrictions on mass layoffs rather than individual dismissals – my estimates above suggest that the magnitude of effect for any restriction on individual dismissal would be too small to detect over background noise.

  6. gordon,
    In this instance ‘privatisation’ refers not to government / private, but to gains that are private to the individual. So, if I work harder a privatised gain would mean that I get the full benefit of the extra work as opposed to that increase being shared amongst all the firm.

    Using James’ terminology, my point was that, in a system of perfect equality there would be no gain that is privatised – i.e. my increased work / capital / thought / improved process would not benefit me any more than it would benefit the person down the road that was not even aware that I did it. Under this system the incentive to work harder, or at all, is greatly diminished – why work when any effort I make I do not see the benefit of?

    OTOH, if all gains are full privatised then we are in a complete dog-eat-dog situation. It would also be impossible, as all actions have some externalities – whether positive or negative, so both ends are impossible extremes.

    My contention is that there is some optimal level of inequality, where the incentive to work both harder and better is maximised while preserving the ability of society to function generally and not leaving too many people behind as they are unable to keep up.

    The article, to me at least, seems to be implying that a decrease in inequality is a desirable outcome. I was just making the point (in what is now a long winded way) that this has not been proven, but is implicit in the article.

  7. Andrew Reynolds, thanks. I was interested in your remark about an optimal level of inequality. I think the belief in such an optimal level is at least as widespread as the belief that reducing inequality is always and in all circumstances good. It is good to get both implicit beliefs out in the open. In present circumstances, I do think that inequality reduction is the more pressing issue. Maybe that makes me a social democrat.

    Some writers on economic growth suggest that inequality is a brake on growth, but examples I’ve seen are drawn from early-stage growth, and I’m not aware of estimates of “optimal” or “maximum tolerable” inequality even in this literature – which may reflect only my incomplete aquaintance with it. Recent increases in inequality in developed countries (as illustrated by Prof. Quiggin’s graphs in his 1 June 2005 post on IR and inequality) are of course on everybody’s mind, but again I am not aware of any estimates of optimal inequality in developed countries.

  8. I think a study along those lines would be an interesting exercise – but, unfortunately, not one I would have time for. Maybe PrQ could put one of the (many) minions other commentors seem to think he has onto the question.

    The difficulty, IMHO, would be in formulating the question – how would optimal be defined? Economic output (perhaps GDP, adjusted for environmental degradation or improvement, depletion or increase in human capital etc. etc.).

  9. It’s not entirely clear what Andrew means by dog-eat-dog, but I guess it covers several good justifications for compulsory redistribution:

    1. Luck as well as effort affects outcomes. Even those who make a big effort can lose their job or come unstuck in a variety of other ways. Of course people can insure voluntarily against income loss, but this is a market that will never work for everybody, due to information asymmetries. So a universal scheme (i.e. the wefare state) is more efficient.

    2. Children in poverty, through no fault of their own, are unlikely to realise their potential. They can’t borrow against future output, so it makes sense for each generation to subsidise the children of their poor peers.

    3. Absolute poverty breeds crime. Crime, as well a range of other ilegal but wasteful rent-seeking activities, is a rational response for people poor labour market prospects and no other means of support.

    4. Marked relative poverty inspires irrational anti-social behaviour, which also has external costs. People who feel abandoned by society will not develop a normal sense of social obligation. They will show less courtesy, and may neglect to flush toilets.

    These are all justifications from the point of view of rational individual self-interest. That is, they’re not drawn from any ethical philosophy that promotes altruism as inherently good. But they may be consistent with such philosophies. They are awkward for libertarians of the Nozick brand.

  10. James,
    I would agree with all of your points bar point 1. Luck does affect outcomes, but the information assymetries can be dealt with – this is a possible use for the unions. Creating a huge bureaucracy to deal with information assymetries is probably not the way to go.
    Point 2 I agree with wholeheartedly, but I am not sure an Education Department is the best way to deal with this.
    Points 3 and 4 I also agree with – but both of these flow from your point 1. Deal with the luck and information problems and 3 and 4 become much less of a problem.

  11. James Farrell, another consequence of dog-eat-dog is likely to be further inequality, as families accumulate wealth across generations. This is likely also to reduce social mobility, as bridges to advancement like education come to be seen as privileges rather than rights. I think we are already seeing a reduction in mobility in parallel with increases in inequality, at least in UK and USA (The Independent reported a LSE study by Blanden, Gregg and Machin on this in April).

  12. Gordon,

    I would agree about the reduvction in quality in the educational systems – governments tend to have a much shorter time horizon than parents.

  13. Bahnisch and Quiggin Live

    John Quiggin and I will both be speaking on Thursday at a forum organised by the UQ Union on the IR changes and their implications. If you’re in the vicinity, come and have a listen. Say hello afterwards too if you like – I might grab a spot of …

  14. […] The United States of America, as a result of it’s deregulated labour market, has some of the most striking inequality anywhere in the world. Do we really need more low-paying, casual jobs? CATHERINE MCGRATH: On the issue of regulated markets, Sweden, Norway, Iceland, Denmark, have highly regulated labour markets, but have higher productivity and higher wealth per capita than Australia. […]

Comments are closed.