My previous article on Why Blame Neo-Liberal Economics, which argued that neo-liberal economics was not a main cause of increasing inequality, drew an unusually large and mostly critical response. While it is not feasible to respond to all the detailed points that my many critics have raised, in this response I propose to focus on two big issues: (i) what is neo-liberal economics and how does it influence policy outcomes, and (ii) why has inequality increased since the 1980s. I will also briefly discuss the policy implications that flow from my analysis.
In addition, a large part of the motivation for my original article was the extent to which those alleging a strong link between neo-liberalism and inequality rely on conjecture and even ideology, rather than carefully examining the evidence. For that reason, my response will rely heavily on what the numerous studies of the evidence have concluded.
What is neo-liberal economics
I should first begin with an apology. Clearly I made a mistake in not defining neo-liberal economics. I offer the not very good excuse, that I have been in good company – most of the critics of neo-liberal economics do not define it either.
I took my definition of neo-liberalism from what I understood the critics of the micro-economic reforms of the 1980s and 1990s to be complaining about. These reforms involved:
- floating the Australian dollar,
- financial deregulation and (later) changing the type of regulation,
- largely eliminating protection, and a shift in industry assistance in favour of more generic assistance and less specific industry assistance,
- tax system reform,
- decentralising wage determination in favour of more enterprise bargaining,
- some measure of privatisation accompanied by the introduction of competition policy.
I think these reforms can be broadly characterised as making greater use of markets through a degree of market liberalisation, and it is the impact of these reforms, frequently characterised as representing ‘neo-liberalism’, that I am discussing. Furthermore, as someone who was heavily engaged in the development of many of these reforms, I don’t think they were inspired by economic rationalist views that markets are always right. Neither did these so-called neo-liberal micro-economic reforms represent an attempt to introduce ‘smaller government’. Instead, on any objective measure the government is significantly larger today than it was at the time of the Whitlam Government – a time which I suspect so many of my critics now look back on nostalgically. Today Australian Government Budget outlays represent 25.1 per cent of GDP, quite a lot more than the 21.7 per cent share in 1974-75 (the last full year of the Whitlam Government). Similarly, the number of pages of government regulation has increased over the same time period, although the nature of the economic regulation has changed to improve the working of markets by improved prudential regulation of financial institutions to increase investor protection and to improve consumer protection through labelling and health and safety regulation, and ensuring competition in the markets for goods and services. Industrial awards continue to set minimum standards for pay and conditions in the labour market, and environmental protections have also increased.
The fundamental reason for many of these reforms was because the previous regulations were just not working. For example, the previous attempts to ‘manage’ the Australian currency were proving counter-productive, signalling the likely next movement to the speculators, and giving them a one-way bet. Similarly, protectionism wasn’t working. These sorts of regulatory changes did not, however, involve any attempt to reduce government responsibilities, as implied by the critics of neo-liberalism. Rather they represented an attempt to improve the effectiveness of government intervention as governments found that often they would be more effective if they focussed more on managing markets and creating the right incentives and disincentives, and relied less on administrative controls that were increasingly being evaded or proving unworkable (Keating, 2004). Today the equivalent approach would be to price carbon as the best way to reduce carbon emissions, rather than relying on direct action policies, and it is curious that this reliance on pricing carbon is now supported by many of the critics of neo-liberalism.
My sense is that it may be privatisation that most attracts the animus of the critics of the neo-liberal economic reforms. There are, however, a couple of points to be made. First, the majority of economists long ago concluded that competition was a much more important determinant of corporate performance than ownership. Furthermore, the government ownership of commercial entities such as airlines and banks did not seem to be achieving any social purpose, with the government owned airlines and banks behaving and having to behave in much the same way as their competitors to survive. While, the funds from the sales of these government businesses helped fund other more important government activities, and I have no sense of any public pressure to use scarce government funds to re-purchase these companies.
Recently there have been many allegations that the privatisation of public utilities has led to higher prices. However, the evidence shows that prices initially fell following many privatisations and the introduction of competition policy at the same time. While clearly further large increases in the prices of electricity in particular have been foreshadowed, there are a number of reasons which have nothing to do with privatisation; such as the recent increases in gas prices, and the impact of policy uncertainty on investment.
Nicholas Gruen has argued that public utilities can be managed with reasonable efficiency, citing the case of Australia Post. I agree that Australia Post is a success, but It is important to note that Australia Post is now subject to considerable competition; evidence of my point that competition is more important than ownership. Furthermore, as a former Board member, I know that the Board was very conscious of how that success was only made possible by the corporatisation of Australia Post, which freed up Post from excessive Ministerial direction and allowed it to compete.
More generally, whether the government should be directly responsible for the provision of a service or contract it out, seems to me to be a matter of judging horses for courses. For example, a long time ago – well before the advent of neo-liberalism – governments started contracting out construction activity, such as road building – and this seems to have received community-wide acceptance. Similarly, the community has long supported the system whereby medical services are mostly provided by private doctors who are then reimbursed totally or partially by the state. What may be more controversial is the contracting out of some decisions, which in the end affect the nature of government programs. For example, the Keating Government in its One Nation package of assistance to long-term unemployed people decided to contract out the provision of employment and training advice and assistance. The reason was that although the former Commonwealth Employment Service had a reasonable record of assisting the typical job seeker, its culture was built around the provision of the same uniform service to all its clients, whereas long-term unemployed people required more individual assistance which non-government providers were better at delivering. On the other hand, I personally question the increasing contracting out of the analytical and advisory functions of the public service. Too often the use of a commercial consultancy firm results in the government getting the sort of analysis and advice that it wants to receive, rather than frank and fearless advice. This is probably because of the pressure on the consultancy firm “to please” when it wants to gain another future contract.
Finally, I would like to comment briefly on what my definition of neo-liberal economics doesn’t cover. I suspect that my critics would argue that neo-liberalism is not just about improving and making more use of managed markets to achieve long-held government objectives, but that neo-liberalism has had an impact on the nature of our society and relations between the state and communities (for example, see the comment by Eva Cox). I acknowledge that this might be a possibility, but it is still a conjecture and unproven. The reality is that there are numerous reasons why our society and culture may have become more individualistic, and I don’t see a more individualistic society as necessarily being a consequence of measures to improve the functioning of markets. Indeed, it seems equally likely that some of these measures were more in response to a more individualistic society and culture, rather than its cause (for example, the contracting out of personalised assistance to the most disadvantaged unemployed persons). In this regard, it might be noted that individualism has always been central to the Liberal Party’s values since its foundation in the early 1940s – well before the advent of modern neo-liberalism. In addition, individualism has been assisted by changes in the way we live: marriage break-up is much more common, most of us no longer travel together on public transport, and our entertainment is more likely to be spent watching TV or a video than engaging in community activities or even in group activities in the home, such as singing around the piano. These changes have all encouraged the development of a more individualistic society and culture, without any assistance from neo-liberal thought. In short, I am inclined to think that neo-liberalism is providing a scape-goat for a number of changes in our society and culture that some people do not approve of, but which have other more complex causes than neo-liberalism.
Causes of the rise in inequality
As I said in my previous article, ‘one thing that we are all agreed upon is that the evidence published by the OECD (and others) clearly shows that inequality has risen in most advanced countries since the early 1980s’. While no-one has disputed the finding that according to the OECD data, the increase in Australian inequality has been less than in other countries, although I should have added that the distribution of household disposable income in Australia was relatively unequal to start with back in 1980. In addition, I further noted that ‘as both the IMF and OECD have been saying for some time, inequality is bad for economic growth and governments can make a difference’. But if governments are to make an effective difference to the inequality of incomes we need to understand better the causes of the increase in that inequality.
My contention has been that technological change was the most significant cause of the increase in inequality across the OECD group of member countries, but this has been heavily disputed in many of the responses. What disappoints me is that so many (although not all) of those who disagree, have failed to provide statistical evidence and analysis of cause and effect to support what is too often conjecture. So in what follows I will spell out at greater length my reasons for continuing to think that technology is the main cause of the increase in income inequality since the early 1980s.
First, it is useful to list the principal possible causes of changes in the distribution of earnings, noting that earnings account for about three quarters of total household incomes among the working age populations. In sum these possible causes of increasing income inequality are:
- the nature of technological change
- regulatory changes
- financialisation which describes the expanding scale and scope of financial markets and activity and is often closely associated with rent seeking behaviour
- changes in the tax/transfer systems
Technology, and particularly information and communication technology has had its greatest impact on relatively routine tasks, reducing the share of middle-level jobs such as clerical occupations and the operation of basic machinery (Blinder, 2oo6; Borland & Coelli, 2015; Frey & Osborne, 2013; IMF, 2017). This hollowing out of the number of middle-level jobs has most affected manufacturing, which has always been at the forefront of introducing new technology. As a result manufacturing output has continued to increase in most OECD economies, while total manufacturing employment has declined. Across all industries, the increased share of jobs at the top and bottom of the job distribution has increased the dispersion of jobs, which in turn has resulted in an apparent increase in inequality as measured. In addition, technological change has often been skill-biased, resulting in an increase in relative wage rates for highly paid skilled professionals. However, one and possibly the main reason why inequality did not rise as much in Australia as in the United States is that the number of people with skilled qualifications expanded rapidly in Australia, while it stagnated in the United States, largely because of the very high up-front costs of getting a degree in the United States. As a result, relative wage rates have not changed much in Australia, while the wage premium for skills has increased dramatically in the United States.
Globalisation can affect employment and the distribution of earnings in a number of ways, including through trade integration, financial integration, and technology transfers. In principle, trade integration can affect the structure of industry, but the reduction of trade barriers and the increase in trade can benefit both sides, as each country concentrates on what it does best (its comparative advantage), and exports and jobs increase in these sectors. In addition, the increase in incomes associated with the lower costs of imported products enables demand to increase, thus further offsetting the loss of jobs in sectors where a country is less competitive. Indeed, for a long time, the output of goods in the trading sector in most of the advanced OECD economies has continued to increase even while employment in this sector was falling, suggesting that increasing productivity through technological changes was the principal influence on earnings and their distribution during that time. Nevertheless, there are examples where job losses from increasing trade were significant; such as footwear, clothing and textiles, and more recently motor cars, although in the latter case it may have more to do with the failure of the Australian car industry to adapt to changing demands.
Product market regulation, labour market regulation and union density can all impact on employment and the distribution of earnings; indeed that may be their intended purpose. An OECD study (2011) found that between 1980 and 2008 most countries relaxed regulation of product and labour markets, and these changes increased total employment, while also increasing wage disparities. with the two effects tending to cancel each other out. The net effect of these regulatory changes ‘on trends in “overall earnings inequality” remains indeterminate in most cases’ (OECD 2011: 31).
The enormous increase in the size, scope and influence of financial markets has been most helpful in explaining the increased share of the top one percent of the income distribution, with executive remuneration being increasingly linked to share price and stock options. In effect, there is a de-facto alliance of executives with financiers. In addition, Piketty argues convincingly that executive pay is heavily influenced by ‘pay norms’ and that these have shifted under the influence of financialisation. Indeed, the capacity of top managers to develop and appropriate the rents that Paul Frijters points to in his response has probably been aided and abetted by this change in social norms. But as I am sure that Paul knows, the opportunities to appropriate rents typically flows from a monopoly situation, and often such situations are created by regulations. Accordingly, economists have favoured regulatory reform to improve competition where possible, or otherwise regulate to eliminate the monopoly rents as the best policy option. While most importantly, in most countries the increase in the share of the top one per cent does not explain most of the increase in inequality, although the United States has been an exception (Piketty, 2014: 296). Even in the United States, however, it has been the increase in earnings, and not income from capital, that accounts for most of the increase in the share of the top one per cent.
Although non-earnings income (capital and self-employed income) is less equally distributed than earnings, these other two sources of income are relatively too small to materially alter the conclusions reached about the causes of the increase in earnings inequality. For example, the OECD (2011: 34) found that ‘Even though its [capital income] share increased in most countries, it remained at the moderate average level of around 7% of total income’. While ‘the effect of self-employment on overall inequality remained modest … because self-employment income fell in most countries and accounted for only a relatively small share of gross labour income – between 3% and 13%, depending on the country’ (OECD, 2011: 36).
Finally, the international evidence is that the tax-transfer system was on average able to offset about half of the of the increase in pre-tax market incomes in OECD countries, and more than half in Australia, Canada, Finland and Sweden (OECD, 2011). And because they are more tightly targeted, transfers have a much greater impact on redistribution than taxes, including income taxes. However, most of that reduction in market income inequality occurred between the mid-1980s and the mid-1990s, and since then the adoption of ‘austerity policies’ (mostly introduced for reasons unrelated to neo-liberalism) have made the tax-transfer system less redistributive in many countries.
Australia too has followed this fiscal pattern. Between 1981-82 and 1996-97 the amount of redistribution achieved solely through the tax-transfer system stayed constant, reducing inequality by 34 per cent at the beginning of the period and by just over 35 per cent at the end (Johnson & Wilkins: 2006). In the 2000s, however, the amount of redistribution achieved through government intervention in Australia declined, although two thirds of this decline was accounted for by fewer people on benefits as unemployment fell, while the progressivity of benefits hardly changed over the period from 1999 to 20009 (Herault & Azipitarte, 2015). Thus most of the decline in government assistance did not flow from government policies. I also think it is drawing a long bow to assert that the fiscal tightening and tax cuts in the 2000s flows directly from the adoption of neo-liberal policies. Instead, I would contend that a policy approach that favours greater use of market instruments and incentives was not necessarily inspired by a neo-liberal attack on government redistribution. A more likely explanation in my view is that Australia elected the conservative Howard Government, and like previous conservative governments, the Howard Government was committed to lower taxes for the rich as part of its traditional conservative policies. As Paul Fritjers says in his comment, politics has everything to do with shifts in the income distribution caused by changes in the tax-transfer system.
To sum up, the actual examination of the evidence strongly supports the findings by the key international organisations – the OECD and the IMF – about the causes of increasing inequality, which were:
- ‘neither rising trade integration nor financial openness had a significant impact on either wage inequality or employment trends within OECD countries’ (OECD, 2011: 29). However, the OECD study did find that increased financial flows and technological change did have an impact on inequality, with technological change having the more comprehensive effect.
- ‘Technological progress, reflected in the steep decline in the relative price of investment goods, has been the key driver of the decline in labour’s share of production income in advanced economies, along with a high exposure to routine occupations that could be automated, with global integration playing a smaller role’ (IMF, 2017: 40). The IMF further concludes that ‘The evidence suggests that the impact of technological advancement and participation in the global value chains on the aggregate labour share in advanced economies comes through a reduced share for middle-skilled labour’ (IMF, 2017:40).
In short, the serious examinations of the evidence conclude that technological change has been the main driver of increased inequality over the last three or more decades since the early 1980s. Furthermore, it is difficult to imagine any connection between the nature of this technological change and neo-liberalism. Indeed, if further proof were needed, no-one in commenting on my original article sought to rebut my point that there is no obvious correlation between neo-liberalism and the types of government in Finland, Germany, and Luxembourg, which are three of the five countries where inequality has most increased, whereas Australia is a country where inequality has least changed.
Where do we go from here?
Before concluding I would like to add a few words on what does this discussion of the causes of increased inequality imply for future policy responses.
First, we cannot nor should we try to stop the structural changes in our economy generated by technology, or even by globalisation. Nor would rolling back market liberalisation achieve much in the way of improved equality. As some of my critics have acknowledged, these reforms have helped underpin the strong performance of the Australian economy over the last quarter century. However, I agree with those who are sceptical that economic growth on its own will ensure an egalitarian society as I don’t expect the fruits from that growth will trickle down equally to all member of our society.
Instead, there are a few policy responses that stand out and which should be pursued. First, Australian micro-economic reform has generally recognised that there can be some losers, and adjustment assistance has typically been provided. For example, when the Howard Government introduced the GST the compensation package actually cost more than was raised by the new tax. Failure to provide such assistance risks losing the consensus which successful reforms typically require. Second, increasing skills to assist in the adoption and adaptation to new technology is absolutely essential. This will improve the sharing of the gains from that technology as well as improving its rate of take-up and therefore future economic growth. Indeed, Piketty – contrary to the view of some of my critics who accused me of not reading Piketty – did conclude that:
‘the poor catch up with the rich to the extent that they achieve the same level of technological know-how, skill and education, not by becoming the property of the wealthy. The diffusion of knowledge is not like manna from heaven: it is often hastened by international openness and trade.’ (2014: 71).
‘To sum up: the best way to increase wages and reduce inequalities in the long run is to invest in education and skills’ (2014: 313).
I rest my case.
Michael Keating, AC was the Secretary of the Departments of Employment and Industrial Relations, Finance, and Prime Minister and Cabinet from 1983 to 1996. The ideas and evidence discussed above are drawn from a forthcoming book, co-authored with Professor Stephen Bell, Fair Share: Competing Claims and Australia’s Economic Future.
Blinder, A (2006). ‘Offshoring: The Next Industrial Revolution’, Foreign Affairs, vol. 85, pp. 113–28.
Borland, J & Coelli, M (2015). ‘Information Technology and the Australian Labour Market’, in CEDA (Committee for Economic Development of Australia), Australia’s Future Workforce? June 2015, CEDA, Melbourne, pp. 131–41.
Frey, C & Osborne, M (2013). The Future of Employment: How susceptible are jobs to computerisation, Oxford Martin School, University of Oxford.
Herault, N., & Aziptarte, F., 2015, ‘Recent Trends in Income Redistribution in Australia: Can Changes in the Tax-Benefit System Account for the Decline in Redistribution?’, Economic Record, 91, pp. 38-53.
IMF, (2017). World Economic Outlook, April 2017; Chapter 3: Understanding the Downward Trend in Labor Income Shares, pp.121-141.
Johnson, D, & Wilkins, R (2006). ‘The Causes of Changes in the Redistribution of Family Income in Australia, 1982 to 1997–98’, Social Policy Research Paper, no. 27, Department of Families, Housing, Community Services and Indigenous Affairs, Commonwealth of Australia, Canberra.
Keating, M (2004). Who Rules? How government retains control of a privatised economy, Federation Press, Sydney.
OECD, (2011). Divided We Stand: Why Inequality Keeps Rising, OECD Publishing, Paris.
Piketty, T (2014). Capital in the Twenty-First Century, Harvard University Press, Cambridge, MA.