ACTU Research Officer Tim Harcourt comments on 'The Economic Effects of Labour Market Deregulation' by Peter Dawkins.
It is a great honour for me to represent the Australian trade union movement in this timely tribute to Emeritus Professor Joe Isaac.
I had the privilege of working with Joe - we overlapped by one week! I started work at the (then) Australian Conciliation and Arbitration Commission (ACAC) in February 1987 - the week of Joe's retirement. I had just started my first full-time job as associate to the newly appointed Deputy President, Professor Keith Hancock. Keith, who has himself retired this year, took up Joe's mantle as the resident economics expert in the ACAC and later the Australian Industrial Relations Commission (AIRC).
Without getting too much into the debate about retirement ages I always thought that the industrial relations community missed out by having Joe retired from the Commission in 1987. I thought we should have pass a "Nelson Mandela amendment", especially for Joe Isaac so we could continue to benefit from his widespread experience and wisdom in the Commission.
But, fortunately for all of us Joe's previous life as an academic was resumed after his 'retirement' from the Commission. In fact he has since been prolific as a researcher and has provided wonderful advice as a teacher. Two generations of Harcourts have benefited from Joe's wisdom and many careers in both the academic world and the practical world of industrial relations have sprung from Joe's inspirational teaching and research.
So, at the outset, on a personal note, and on behalf of the ACTU I would like to thank Joe Isaac for his wonderful contribution to the welfare of the Australian workers, in particular, and to the industrial relations community in general.
2. Introductory Comments on the Peter Dawkins Paper
Like a contemporary enterprise agreement, I have received Peter Dawkins's paper in instalments. However, I understand Peter is a busy man, and he has done his best to get most of it to me in time. I will comment on what I have received and try to treat the general directions of Peter's arguments fairly.
At the outset, Peter has written on a difficult topic. This is not just due to the breadth of the topic "labour market deregulation", and the theoretical difficulties involved, but also because it is a very politically sensitive topic. It thus needs to be dealt with some caution. However, can I say up front, that whilst I do not share Peter's views in economics and politics, I do think he has attempted to deal with this difficult question in a professional fashion. [ I should also declare that I am a former student of Peter's when he was in South Australia. I was fortunate enough to be taught Labour Economics by economists of diverse views from Barry Hughes, Owen Covick, Tom Sheridan and Bill Mitchell on the one hand, to Peter Dawkins and Dick Blandy on the other.]
I have arranged my comments on the paper under four headings.
First, I wish to discuss some historical questions.
Second, I wish to comment on the discussion of deregulation and how the labour market is supposed to function.
Third, I wish to comment on the issue of globalisation and labour markets.
Fourth, I will comment on the use of international evidence.
3. Historical Perspective
Firstly, I want to comment on Peter's view that deregulation in Australian Industrial Relations is some 'new phenomenon'. He comments:
"Australia's system of Commonwealth and State tribunals and awards and associated principle of compulsory arbitration has set the tone of Australia's labour market regulation for most of the twentieth century. Since the late 1980s, however, the degree of centralisation has been gradually relaxed and the ability of employing organisations to have greater influence over their own wage and employment conditions has increased. This process could be described as a partial deregulation of the labour market......."
[Dawkins (1997), p.3]
I do not agree with this simple view that Australia has always based one kind of wages/ IR system and that somehow this has led the Australian labour market to perform differently than labour markets in other industrialised countries.
The Australian wages/IR system has gone through a number of different historical developments, both centralised and decentralised - over the century with the basic wage, the basic wage plus margins, the total wage, indexations, partial and indexation, two tiers, supplementary payments, safety nets, "the Living Wage", overaward bargaining, industry bargaining, enterprise bargaining etc. etc.
I think Dawkins' analysis is limited and a historical. I would argue that history shows that our wages system has been:
- adaptable and flexible in the face of macro-economic changes (eg. external 'shocks' such as the Terms of Trade crisis in 1985-86);
- adaptable and flexible in terms of award and workplace reform;
- efficient in responding to industrial disputes (saving Australia from the lengthy disputes experienced in other industrial countries);
- socially progressive in terms of equal pay for women (Australia has a more favourable female-male earnings ratio than most other countries); family leave; equal pay for Aboriginal and Torres Strait Islanders; equitable wage outcomes for workers whose first language is not English (relative to other industrialised countries);
- adaptable in balancing economic, social and industrial objects.
In fact, to understand the history of the wages system, you need go no further than the works of Joe Isaac himself. In 1993, Isaac notes:
"Industrial relations reform is not a new phenomenon in Australia. It has been going on through most of our history in order to deal more effectively with industrial disputes and economic requirements.
An historical view of the Australian IR system shows it to be a flexible institution in which changes in policies, principles and procedures - sometimes generated by legislation - have taken place. These changes reflect attempts to adapt the operation of the system to the changing social, economic and political environment".
[Isaac (1993), pp.10-11, cited in Harcourt (1997)]
Furthermore, I do not agree with the assertion put in the paper about the award system putting the labour market "in a strait jacket" [Dawkins (1997) p.4].
In fact, I would agree with Joe Isaac's assessment that the Australian system has not only accommodated but has encouraged workplace reform to generate the necessary efficiencies in the labour market.
The IR system in Australia has also been criticised on the grounds of its alleged inflexibility, rigidity and inefficiency compared to other countries. However research using international comparisons in fact show the Australian systems' flexibility and efficiency relative to IR systems in comparative economies.
Examples of the research include studies by:
- Coelli, Fahrer and Lindsay (1994) which examines Australian wage setting institutions in a study of 14 OECD countries.
- Withers, Pitman and Whittingham (1986) - comparing US, UK, Sweden and Australia.
- Brown, Hayles, Hughes and Rowe (1978), which show that arbitration did not constrain the Australian labour market when comparing local labour markets in Adelaide, Coventry and Chicago.
- OECD (1986) - the Dahrendorf Report - on relative wage flexibility.
- NLCC (1987) - which found that the Australian system was not inflexible by international standards;
- Committee of Review (1985) into Australian Industrial Relations Law of Systems 'The Hancock Committee' - which found that relative wage flexibility was not a key element to improving the allocation of labour.
[cited in Harcourt (1997)]
The studies showed that Australia does not have the rigid, inefficient, inflexible wages system that the critics claim. Both historical and cross-country studies show that the Australian IR systems have at the least the same (if not more) flexibility and efficiency than corresponding wages/IR systems of similar economies.
4. How Does the Labour Market Work?
I disagree with Dawkins on the grounds of theory and practice. First, I wish to deal with theory.
The studies cited above, suggest that the answers we get in applied labour economics research do not always seem to match the questions we ask according to neo-classical economic theory.
In fact, the reason I became interested in labour economics was because it did seem to be a unique branch of economics.
That is why I do not agree with the premise that you start with demand and supply curves in the labour market (like any other market) and then you make assumptions about what will happen with "interventions" (like trade unions and industrial tribunals).
In fact, I have always enjoyed Robert Solow's book, 'The Labour Market is a Social Institution'. His book can be summed up with the following quote:
".......the labour market is a social as much as an economic institution - and that the interaction between human beings cannot be interpreted in the same way as the supply and demand for dead fish."
This too has been noted by Joe Isaac:
"A distinguished British labour economist has said 'The management of fairness of pay is of great importance in the achievement of satisfactory labour productivity'. (Brown 1989:254) Violation of accepted notions of fairness will have adverse effect on cooperation and can lead to industrial action, covert and overt, and drain productivity. This applies as much to professional groups like the medical specialists in Victorian public hospitals as to the workers at Weipa."
[Isaac (1993) pp. 18, cited in Harcourt (1997)]
The market for labour is a market for a very different commodity - where social notions of fairness and equity come into play. Any move to 'deregulate' will not alter the fact that there is a human side to the labour market. People, unlike commodities, have concepts of fairness, and this will impact on their productivity and ultimately to the performance of their employees.
This is why there has been such a rich variety of literature on the economic of labour markets. I refer to:
- Quantity Constrained Models
- Human Capital Theory
- Search Theory
- Bargaining Theory
- Insider/Outsider Models
- Segmented labour markets
- Dual labour markets
- Monopsony/Monopoly Models
- Internal labour markets
- Efficiency wage models
- UV 'Beveridge' curves etc.
This literature has also benefited from by the contributions of industrial relations scholars such as Isaac, Willy Brown, Keith Hancock and Richard Freeman (who straddle both worlds of labour economics and industrial relations).
That is what makes labour economics so interesting. Any analysis that just uses demand and supply curves (and assumes they only behave in certain ways) be limited and not convincing once you look at the empirical literature.
Secondly I wish to deal with practical issues.
The first problem with the Dawkins' analysis is if you deregulate (by removing awards) surely you must have some other employment instrument in their place - namely an individual contract. Employers who may have been able to simply follow the award would now have to go to the trouble of making individual contracts with every members of their workforce.
The slow uptake of Australian Workplace Agreements (AWAs) under the Workplace RelationsAct shows that "re-regulation" is perhaps not always an attractive option to employers. In fact the debate should not be "deregulation" versus "regulation" but perhaps "individual regulation" versus "collective regulation".
The second problem is the argument mentioned by Dawkins about arbitration and the existence of tribunals encouraging conflict and adversarial industrial relations. I do not agree with this hypothesis. I think it is a bit like blaming the existence of the family court for marriage breakdowns.
Further, the argument does not consider that Australian industrial disputes have typically been of shorter duration than other countries and we generally keep more accurate data than other countries. We also count a dispute more readily than other countries. For example, in Australia, a dispute is included in the official statistics if it causes the loss of ten or more working days. A dispute is only counted in the US if it involves at least 1,000 employees; in Japan if it is 'official', in Denmark if it leads to the loss of 100 working days and affects firms belonging to an employer's federation (see ACTU (1994)).
[Of course, authoritarian countries don't collect the disputes data at all.]
The third problem is the view that centralisation and regulation go hand in hand (as do decentralisation and deregulation). I do not think it is as simple as that. For example, the ACTU has over the years supported both centralised and decentralised wages systems. It has never had a view on what is a perfect wages system. However it is clear that the ACTU has supported collective regulation by awards and agreements over deregulation/ individual regulation. The current Government, however seems to want "deregulation" of wages and conditions but "regulation" of trade unions and trade union activity. The Reserve Bank, too, seems to want "regulation" of macroeconomic wage outcomes but "deregulation" of the wages system. There seem to be many ideas around about what we mean by "deregulation" and therefore what we want.
5. Globalisation and Labour Markets
The Dawkins paper discusses the "exogenous" vs "endogenous" question with respect to globalisation. Following this discussion, there is an account of the historical link between tariff protection and arbitration (eg Plowman's discussion of the Harvester Case and the Brigden Committee).
It is often argued that because of the historical occurrence of tariff protection (and 'the white Australia policy') being established at the same time as arbitration that the removal of tariff protection should automatically lead to the removal of arbitration. This thesis has been developed by journalist Paul Kelly (see Kelly (1992:'The End of Certainty') and by Dr. John Hewson. For example, Hewson said in his Alfred Deakin Lecture in 1992:
"It is the historic destiny of the Liberal Party to send industrial arbitration and protective tariffs to join white Australia in history's dustbin".
However, I am not convinced that there is a policy link between the two that says "both started at this point in history, therefore both should go". As women got the vote at approximately the same time period as tariff protection and arbitration does that mean that when we remove tariff protection we should also take the vote off women?
[It should also be noted that wage boards and industrial tribunals(and votes for women) in the colonies pre-dated federation and tariff protection].
Furthermore, I do not agree that trade liberalisation is the same policy response as labour market 'deregulation'.
No matter what its trade policy stance, a nation must develop social institutions (including labour market institutions) to ensure fairness and social stability. Whether an economy is 'closed' or 'open' it must deal with questions of income distributions, social welfare and labour market adjustment policy. In fact an influential book by a neo-classical economist, Dani Rodrik (see Rodrik (1997):'Has Globalisation Gone Too Far?') argues that the more open the economy is, the more need there is for social institutions to assist those affected by globalisation. In fact Rodrik has noted that the open economies (ie the Scandinavian economies, the Netherlands etc.) have tended to spend more on social welfare spending. Similarly, trade unions, labour market institutions, and 'solidaristic' type wage setting has occurred in 'open' economies with low or 'negligible' tariffs.
I would argue that trade unions, arbitration safety nets etc..... are necessary whatever the trade and investment regime of the nation. Furthermore, a fair and efficient set of labour market institutions is necessary for a small open economy to flourish with economic growth and social stability. Trade unions and labour market institutions have been shown to be a positive factor in improving an economy's trade performance (see Mishel and Voos (1992)).
Further, there are examples of unions in Australia assisting trade and competitiveness (eg the Accord Mark II negotiation in response to the terms of trade decline). I do not accept the proposition that trade unions and labour market institutions are inconsistent with the successful functioning of a small open economy.
6. Use of International Evidence
From the paper I received, there was some discussion for international evidence. The aim of this section of the paper was to compare a 'partially regulated' Australian labour market with a 'radically deregulated' New Zealand one, and (I assume) the deregulated USA and UK labour markets (the paper does not say whether the USA or UK have been 'partially' or 'radically' deregulated).
As with all international comparisons there is always difficulty with time periods chosen, which economies to compare, what variables to choose etc.
For instance, we may take issue with an Australia/New Zealand comparison because:
- the profound differences between Australia and New Zealand arbitration before reform;
- the fact that Australian economic performance over the past twelve years shows a higher cumulative employment and economic growth increase (see ACOSS/ACTU (1995));
- the fact that the performance of the Australian economy and labour market in turn, affects New Zealand;
- the difference in tax/social security structures.
The economic and social context of these comparisons needs to be set.
There is also a problem of what countries you choose and how you interpret possible relationships. This was shown in the Cunningham lecture by Keith Hancock (see Hancock (1997) p.17-18] where he showed cross-country data on wage dispersion and employment/population ratios and thought there to be no particular relationship between the two despite the assertions of Access Economics that economies with high wage dispersion also experienced high rates of employment.
For example, if you were to look at employment to population ratios across countries it would not be clear whether a centralised or decentralised industrial relations system produces the higher (or lower) ratio.
This is illustrated in the following tables (see attachment). Table 1 shows employment to population ratios for several OECD economies that have been classified according to the Calmfors/Driffill analysis used by Dawkins (1997).The 'Prime age' persons category (that is 25-54 years ) is used for 1996 to abstract from the effects of diverse education, training, welfare and retirement systems on the top and bottom end of the age distribution in each country. From the data produced in table 1 it is not clear that moving to a more decentralised IR system would necessarily increase the employment to population ratio. In fact, according to table 1, the 'centralised' systems produced higher ratios on average than the 'decentralised' or 'intermediate' systems.
Table 2 shows 1996 employment to population ratios for countries classified by the ILO as being dominated by 'national/sectoral' or 'company level' bargaining. Again, the countries dominated by 'national/sectoral' (or centralised) bargaining have higher employment to population ratios than countries dominated by 'company level' (or decentralised) bargaining (see ILO (1997)).
This exercise shows that caution should be exercised when making international comparisons of data. It is not clear, for instance, from these observations that decentralising the IR system would guarantee an improvement in the national employment to population ratio.
7. Closing Remarks
In conclusion I have tried to make some comments on Peter's paper in the limited time available. My overall view is that is lacks historical and institutional context and is narrow in its theoretical focus. However, it is just that sort of institutional depth that Joe Isaac adds to rigorous economic analysis that makes his contribution so important for labour economists and industrial relations scholars so important. The Isaac-type approach I think would improve the paper. However, I do appreciate the integrity with which Peter Dawkins has approached the topic and regard his professionalism highly even if we do not agree on many of the questions involved in the labour market deregulation/re-regulation debate.
Thank you for giving me the honour to thank Joe Isaac on behalf of the ACTU and to add my personal congratulations and best wishes to Joe and Golda.
ACOSS/ACTU (1995) "Report of the Study Program on Structural Adjustment and Social Change: New Zealand. ACTU Doc No. 134/1995, ACTU, Melbourne.
ACTU (1994) "Trade Union and Industrial Relations for the Asia Pacific Observer" ACTU Doc No.109/1994, ACTU, Melbourne.
ACTU (1997) Living Wage Case - Submission and Exhibits.
Dawkins P (1997) "The Economic Effects of Labour Market Deregulation". Economics and Industrial Relations: Reappraising the Relationship. Conference in honour of Professor Emeritus, Joe Isaac, ANU, Canberra.
Hancock K J (1997) "The Needs of the Low Paid". The Cunningham Lecture for the Academy of the Social Sciences in Australia, Canberra, 10 November, 1997.
Harcourt T (1997) "The Economics of the Living Wage" Australian Economic Review, Vol. 30, No. 2, pp.194-203.
Hewson J.R (1992) "Liberal Values and Aspirations of Australia in the 1990s" Alfred Deakin Lecture, Melbourne University, Melbourne.
ILO (1997) World Labour Report 1997-98, International Labour Office, Geneva.
Isaac J E (1993) "How Important is Industrial Relations Reform to Economic Performance", paper presented to ACIRRT Conference, Sydney.
Mishel and Voos (1992) "Unions and Economic Competitiveness", M E Sharpe Inc. New York.
OECD (1997) Employment Outlook, July 1997, OECD, Paris.
Kelly P (1992) "The End of Certainty". Allen and Unwin. Sydney.
Rodrick D (1997) Has Globalisation Gone Too Far? Institute for International Economics, Washington D.C.
Solow R (1990) The Labour Market as a Social Institution Basil Blackwell Oxford.
Tim Harcourt, Research Officer, Australian Council of Trade Unions (ACTU), Melbourne.
Presented at the Economics and Industrial Relations: Reappraising the Relationship Conference in Honour of Professor Emeritus Joe Isaac, 4-5 December, 1997. Australian National University, Canberra.
These comments do not represent the views of the ACTU. I would like to thank Russell Lansbury for his comments and my ACTU colleagues Grant Belchamber and Jane Howard for their assistance.