Cause & Effect, Means & Ends, Seeds & Fruit: Reflections on role, history and future of labour law reform

Paper to the 21st Labour Law Conference,
Workplace Research Centre & Law School, Sydney University

The organisers have done the right thing by us with this topic, not least because it allows me, in front of such an informed audience, to eschew recitation of the ACTU view of the granular specifics of labour law as it stands.  I can do that in my sleep, and would in all probability put you to sleep doing it.

Instead I want to go through three things: the connection between productivity and industrial regulation; the purpose of industrial regulation (which, I submit, we have occasionally forgotten); and where this has all taken us.

What it does, what it’s for, and where we are.

Or roughly, and with apologies to Ralph Waldo Emerson, “cause and effect, means and ends, seeds and fruit.”

Cause & Effect
The first thing to do in considering Australia’s productivity performance is to separate myth from reality.  

Labour laws are not the cause of our productivity problems, and they’re not the solution. As we have comprehensively demonstrated , while Australia has issues with productivity it is for reasons almost entirely unconnected with labour law. This is a conclusion reached by the recent review of the Fair Work Act  and shared by almost everyone without an IR axe to grind. The short version is that WorkChoices didn’t fix the problem, and Fair Work hasn’t made it worse, and the construction phase of the mining boom sucking in capital and labour in advance of output is a key culprit.

We have had one strong cycle of labour productivity growth, that coincided with the Keating Government’s introduction of the enterprise bargaining. In general, it is difficult to see strong connections between labour law changes and productivity performance (in fact, Australia’s productivity performance under centralised wage fixing and behind tariffs walls was quite strong – not a number you hear from the “reform” crowd). In summary, the available evidence is that IR policy makes little difference to productivity.  

As a tool to improve national productivity performance, IR reform has been of limited utility – a result which would be repeated if we tried it again.

Those who posit that further IR change would move the national needle in respect of productivity performance find no support in the data.  In fact, the principle observable effect is a distributional one: who has captured the value of productivity improvements.

During the 1960s and early 1970s, hourly labour income grew at around the same pace as productivity. This means that the gains from productivity growth were shared evenly between labour and capital.

In the mid-1970s, labour income rose faster than productivity. At the time this was commonly referred to as the “real wage overhang”. An “overhang” of this sort is equivalent to a rise in labour’s share of national income.

By the late-1980s, the “overhang” had been eliminated, as real wage restraint was exercised under the Accord framework in return for gains in the social wage. Labour’s share of national income had returned to the levels of the 1960s and early ‘70s. In the 1990s, the gains were distributed fairly evenly between capital and labour. Total labour income in the economy grew at around the same pace as productivity.

In the 2000s, real wages growth has ‘decoupled’ from productivity, and has lagged behind. This is equivalent to saying that the labour share of national income has fallen, while the returns to capital as a share of income have risen.  Just as the relative growth rates of labour income and productivity were used to diagnose an “overhang” in the 70s, it now appears that Australia has a “real wage underhang” – wages haven’t kept pace with productivity growth.

This is an important story. In the 90s, labour and capital shared equally in the gains from productivity growth, and that growth was rapid. In the 2000s, the benefits went disproportionately to capital, and productivity growth slowed. This suggests that an environment in which workers feel like they stand to gain from productivity growth will be one in which growth is more rapid. That is, fair distribution helps to ensure good overall outcomes.

We’ve said it this way, productivity may indeed be almost everything, but distribution is not nothing.  Productivity growth matters – it’s the way to get sustainably rising  national income and living standards. At this point it’s customary to use a somewhat tortured metaphor  about slices of pies and tides lifting boats. For the record, the tide needs to lift all boats, not just the big ones. And it’s not a binary choice between growing and dividing the pie.

Means & Ends
So what is labour law for?  Perhaps a good deal of the conflict has been caused by inability to agree or clearly articulate on what the purpose of the industrial system is.  And if  we can’t agree on that, little wonder we can’t agree on what it should look like.

I’d like to posit two linked purposes: a rights purpose and a distributional purpose.

The rights purpose is clear enough. Without labour law disproportionate power lies with the employer. The historic heart of Australian industrial regulation is the recognition that the bargain between an employer and an employee is inherently unequal, absent the intervention of the state in the form of legislation and the existence of trade unions. This fact must be balanced against the appropriate limits of managerial prerogative.

Neo-classical economics might say employment is a free-exchange in a market, but this ignores the power relationship. Labour markets are not remotely well represented by some frictionless Walrasian process where workers receive their marginal product.  Witness the gender pay gap, and the lack of premium paid for insecure work, or the failure of the deregulated US labour market to clear.

Labour markets are characterised by a fundamental power imbalance between employee and employer.  Employment relationships involve bargaining quasi-rents, where the parties to the negotiations typically and almost universally have vastly different bargaining power.  Monopsonistic employment relationships – where employers have some discretion over the wage they will pay – are widespread and endemic.  Labour markets are segmented.  Employment agreements are complex and highly incomplete, with an asymmetry of information between the parties and are typically entered into for an open-ended duration.  

Labour law is a deliberate and necessary intervention in a market because most workers will not get a fair outcome without it. It’s about worker voice, and redistribution of power.  It’s about fair treatment, and a giving people a measure of control, however modest, over their working lives.

Central to the efficacy of a regime of labour law is the role played by organised labour. Without free, independent, democratic trade unions, as actors in system and voices in civil society, the system will not work properly.  We cannot have a ceasefire in the IR wars while the legitimacy of unions is contested – openly or otherwise. The same goes for collective action by workers. Collective bargaining can’t work without an accessible right to strike. Moreover, that right (and other means of exerting pressure – endogenous to the firm or otherwise) must be exercised from time to time to keep the effect.

Second is a distributional purpose. Labour law is a way to influence the distribution of market incomes. The minimum wage, the Award safety net and collective bargaining are powerful tools to protect and increase the living standards of lower and middle income people . Without this intervention the share of national income flowing to labour will be low at a macro-level. Labour law is one way of dealing with inequality and its consequences, including social exclusion.  Without it, firms are likely to compete on wages, not innovation and quality. There is also a consequence if we don’t pull this policy lever. If we don’t seek to effect the distribution of market incomes, we double-down on post-income re-distribution (taxes and transfers) to deal with matters like poverty and inequality. The ability to address matters like rising earnings inequality  then becomes a function of having a tax system of sufficient heft to fund re-distribution and a political process that sustainably delivers it.  

These are big “ifs”, particular when the principle proposal of business in relation to the taxation system is a “tax-mix switch” from taxes on capital to taxes on consumption.  I note that such a shift would itself require a further redistribution to ensure that those in the bottom half of the income distribution, with their low marginal propensity to save, are not worse off.

You can legislate for effect fairness and distribution. And we should. But what you can’t do is legislate to require workplaces be productive and harmonious workplaces.

This view amounts to what I might suggest is a legalist conceit that statute can require people to get on with each other and act in a manner consistent with rational self -interest. Simply put, you can’t do anything of the kind. You can certainly, with efficacy and precision, legislate to restrain bad behaviour. You can certainly, and you should, set the basic rules of the game and a floor of rights. But you can’t make people love each other.  For my money, the suggestion that you can legislate to ensure productivity outcomes at an enterprise is as naïve a view as saying that all conflict can be avoided.

Seeds & Fruit
So, having said all that, where have we been and what’s next?

I am in my 19th year as a Trade Union official. I have now worked with four wholesale re-writes of Commonwealth industrial relations law. This doesn’t count other changes of a more minor nature.  

Swings of the wild and mild kind have occurred with an intensity that suggests a certain mania in the polity about IR – an abnormally agitated need to “do something”.    No other developed country has engaged in changes of such frequency and consequence. The presence of IR at the hyper-partisan edge of  our politics is probably the central factor.  Australian IR types who have explained this recent history to foreigners are met with considerable bewilderment.  

The organisers invited speakers to consider the question “what would a more stable regime of Labour Law look like?” At the headline level at least, my answer is that we have it now.

It will no doubt be said that is convenient for me call “time”  now, with the Fair Work Act in place, and that my real motivation is, if you like, a Thermidorian Reaction – a move to consolidate and secure the gains of the revolution.

My response is to that is simple.

First, for all the sound and fury, it remains demonstrably the case that the critical reform point was the introduction of the Industrial Relations Reform Act 1993 (Cth), which saw the transition from centralised wage-fixing to enterprise bargaining (underpinned by awards to act as a safety net).

Second, if you asses all the change, there is a compelling evolutionary logic to the structure of the Fair Work Act.  As we said to the Review Panel into the Fair Work Act ( as part of an analysis  of the five most recent iterations of Commonwealth IR law ) the Fair Work Act’s key elements are a product of logical evolution. There are six key elements to this:  a national system, a modernised safety net, collective bargaining with good faith obligations, rights to representation, protection in relation to termination of employment and an independent umpire.

In this analysis, WorkChoices was  an interregnum.  It’s the embarrassing uncle of Australian public policy – nobody admits to inviting him and most people are pleased if he’s not mentioned.  The expansion of management prerogative by the Howard Government in relation to termination of employment, individual contracts and restrictions on bargaining were the radical point of departure from the evolution of our modern industrial relations regime.

So much is clear from the fact that, with the exception of the issue of statutory individual contracts (a feature basically without parallel in overseas jurisdictions), the essential components of the current architecture (for all the “calls for reform”) are seriously contested only at the margins.

But there’s a catch.

I’ve previously complained that “calls for reform” are so misused in the IR space that my instinct is to reach for my revolver.   Sadly, the weapon leaves its holster often. Shrill calls for “reform” remain common but, a few outright cranks notwithstanding, the basic framework is conceded and complaint confined to matters which are down in the long grass.

I’ve already expressed the view that change will not move the needle on productivity or macro-economic performance but they matter very much to individuals.  The problem is that the claims made in respect of the effect of the offending matters of detail are outlandish.

First, they tend to ignore the obvious counterfactuals. For example, given the claims made about unfair dismissal laws regularly since their introduction, it’s a wonder, 20 odd years later, that anyone has a job.  About 800 000 Australians leave a job at involuntarily at the initiative of their employer in a year . In 2011-12, about 1.5% of them filed an unfair dismissal claim, and 85% of those settled for somewhere between zero and four weeks’ pay – figures that include unpaid statutory entitlements. In 2011-12, 74 proceedings resulted in re-instatement (including by conciliation) or about 0.001% of people sacked by their boss that year.  Based on the statistical incidence, business has more to fear from insolvency, fire, fair trading proceedings and any other number of catastrophic hazards than it does from unfair dismissals.  The reality is we have very high levels of mobility between jobs – gross flows  – un-retarded by unfair dismissal laws.

Second, when put in a position where something more than “feelpinions” are required the complaints collapse. Penalty rates do not mean you can’t get a long-black on a Sunday. They are not destroying the retail industry. In fact the retail “industry remains broadly robust and generally profitable”  despite other factors like online competition. Put over the jumps in process where real evidence of probative value is required, these claims crumble to dust.

In succession since the start of Fair Work we’ve seen  breathless predictions of  an inflation driving “wages breakout” and of surging waves of “militancy” about the cripple the economy. All these predictions have come to nought – we have low inflation, wage growth at or around long-term trends and industrial disputes (based on even recent comparisons) at trace levels. It seems that “high labour costs” is the next push – an attack on the absolute current level of wages now that predictions of escalation have proved to be nonsense.  But what’s the alternative for a developed, rich country? In releasing the White Paper on Australia in the Asian Century, former Prime Minister Gillard said our future was as a high-wage economy. She was right – as a statement of  basic  fact, and of natural ambition. We don’t have a choice, and who in their right mind would want otherwise?

Of course as Gough said “only the impotent are pure”. Trade unions are not beyond making claims that the sky is falling, or inconsistent claims about the system. In one prominent union (which I shan’t name) affection for arbitration has largely replaced a long-held prejudice against “the boss’s court”.   But the interests of working people and their unions (and the national interest) are served by a fact-based debate.   The ACTU prides itself on engaging with the facts in the big national debates.  We will continue to do so.

And the facts are that we have no significant economic problem to which IR “reform” is the answer.  And so my principle point is that no real good can come of this. The motive purpose behind calls for industrial relations regulation to be subject to further partisan comprehensive review (as a precursor to “reform”) is a dishonest one.  It’s one part attack on the legitimacy of organised labour and one part attempt to reduce workers’ rights and entitlements while admitting neither motivation.  After all, a reduction in collective and individual rights is the only plausible outcome of such a process. A dispassionate evaluation of the evidence does not provide any support for the contention that the removal of workers’ rights is the most pressing priority for governments seeking to improve the wellbeing of the Australian people or our economic performance.

You know the story about our economic performance of course. Twenty-one years of continuous economic growth, low public sector debt, a tax-GDP ratio well below the OECD average, and a highly target-efficient welfare system .  We have low inflation, interest rates and unemployment and modest wage growth. There are significant issues being thrown up by the ongoing structural changes in our economy. These are being driven by the complex interaction of commodity prices, patterns of investment, the currency and technological change.

But not by the choices we have made on labour law. It’s not the source of these issues and  not a “solution” – cause and effect, means and ends, seeds and fruit.  In truth, we have a largely deregulated and flexible labour market. No jobs for life here.  The balance sheets of Australian corporates don’t groan under the weight of pensions and health-care costs – workers bear the investment risk for the former and we socially insure for the latter.

So let’s move on from IR wars.  The organisers ask us to get “beyond groundhog day”. I promise you this:  if in a few years ground-hog day is still with us, I’ll get out my revolver and shoot Punxsutawney Phil myself.