It seems opportune to revisit my 2006 paper, ‘Equal Opportunity in Australia: myths and reality’1. One’s life chances depend in good part on one’s innate qualities and character, but are dependent at least as much on the circumstances of one’s birth. It is common knowledge that some countries enjoy more income and occupational mobility than others, and that experience is highly sensitive to the scale and mix of policy instruments used by governments. Australia has seen its policy environment change over the last forty years in very fundamental ways, some of them quite positive to the environment and others perhaps less so. It is time to look back at why some countries do well on equality of opportunity and some do not.
What does equality of opportunity mean?
When politicians pontificate about equality of opportunity and ‘a fair go for everyone’, they often mean little more than equality under the law. This can be called formal equality of opportunity (FEOP). Many Australians, however, have something much more ambitious in mind than this classical liberal vision. Their norm is substantive equality of opportunity (SEOP). It refers to a situation in which everyone is able to develop their full potential, irrespective of the original circumstances of their birth and childhood. SEOP differs fundamentally from FEOP in two ways.
Firstly, it takes a longer term perspective. FEOP is about ensuring the best person wins at any point in time: individuals competing in schools or the labour market at any one time are judged only on the attributes they bring that are relevant to the duties of the position, without any bias on grounds of race, ethnicity, gender, caste hierarchy or nepotism. SEOP extends this focus, being concerned with risk factors and handicaps in early childhood and teenage years.
Secondly, on the narrow FEOP perspective, only minimal intervention is expected by governments in order to deter overt or covert discrimination. By contrast, under SEOP, governments would be expected to actively intervene to ensure that:
- as children, citizens are not unduly impeded by lack of parental wealth, status and power in achieving their full education potential; and
- as adults, citizens are not impeded by location, inadequate access to training or skill-enhancement, and poor availability of health, housing, welfare or networking from achieving their full employment potential.
Social mobility refers to the ease and frequency by which people move up the social hierarchy during their lifetime and between generations, irrespective of their different backgrounds and starting opportunities. Low inter-generational elasticity – say 0.15 – translates into a high rate of social mobility.
Intra- and inter-generational mobility – comparisons of different countries
Much information on vertical mobility is short range – intra-generational – such as from the lowest deciles or occupational grade to the one just above it. Short range income mobility is still important for equality of opportunity. It shows, for example, that much observed ‘poverty’ in any given year is transitional rather than sustained; that is, it reflects passing events such as joblessness and studying. In many of the countries studied, ‘permanent’ inequality represents about two thirds of cross-sectional annual inequality. In other words, approximately two thirds of the inequality observed in a single year tends to be persistent. However the trend in long-run inequality is not significantly different from the trend in annual inequality and the ranking of countries in regard to inequality is much the same whether cross-sectional or longitudinal analysis is used.
The literature on inter-generational mobility tells a broader story. While still somewhat inconclusive and not always consistent, studies tend to suggest that in the US, the association of fathers’ and sons’ incomes is appreciably stronger, and a child’s economic background a better predictor of later earnings or school performance, than in the Scandinavian countries.
What of the trend in mobility? Whether measured by the father/son association or by the persistence of relative poverty, mobility has been generally stable, with a slight tendency to for it to increase in many European and Scandinavian countries.
What are the reasons for differences in mobility in different countries?
It has surprised many people that social mobility in the US is considerably lower than in social democratic countries like Sweden, Norway and the Netherlands. The US has been a world leader in economic liberalisation and structural change; although Europe has in more recent years started to embrace liberal market policies, the gap with the US remains appreciable. In theory, market liberalisation should make the employment market more merit-based and blinder to socio-economic status, gender, ethnicity and other such factors. Further, market liberalization ought theoretically to widen the employment choices for low-paid workers and the opportunities for upward wage mobility. In addition, the US has a cultural advantage: it is less influenced by historical class distinctions and/or hierarchies than ‘old Europe’, allowing more upward mobility for persons of less privileged origins.
So why has all this failed to give the US a sustained edge on equality of opportunity? The missing link is political ideology on redistribution. While structural change, freer markets and economic growth create ‘more room at the top’, the role of the State may be decisive in determining who fills these places: the existing poor or people from a relatively high social class to begin with.
A 2008 OECD report2 highlights the fact that ‘social mobility is lower in countries with high inequality, such as the Italy, the United Kingdom and the United States, and higher in the Nordic countries where income is distributed more evenly’. It adds that ‘greater inequality stifles upward mobility between generations, making it harder for talented and hard working people to get the rewards they deserve. Ignoring inequality is not an option’.
The Scandinavians and many of the smaller European governments have generally been more successful in reducing education inequality. These policies have helped to reduce differences related to family background and wealth, generating benefits for income mobility which seem to have outweighed the disadvantages from having less flexible labour market and higher taxes.
Similar arguments apply to health and other public benefits. While populations become healthier as the economy expands, relative health outcomes do not necessarily narrow. In Nordic and some European countries, greater government involvement in reducing health inequalities has given poor families better access to medical care; in the US, low-income children tend to suffer from more health problems both in childhood and as adults. In many countries (notably the Nordic countries), greater investment in wider employment programs for the disadvantaged and more active redistribution have helped to more than nullify the liberal market advantages of the US and the UK.
What of Australia’s experience?
The most interesting question, from our viewpoint, is Australia’s performance. In my 2006 paper referred to above, I argued that, although Australia has a more unequal society than in Nordic and many European economies, it is among the most inter-generationally mobile. The paper relied rather heavily on Andrew Leigh’s early draft, as his piece was not completed by early 2006.
Although Leigh sent his revised final draft in 20073, it was overlooked by the OECD in its late 2008 paper. That is why the OECD-recorded inter-generational elasticity for Australia (of 0.15) is so surprisingly low. In his revised paper, Leigh now finds Australia’s inter-generational experience – viewed through four surveys conducted in 1965, 1973, 1987 and 2001-04 in which the relationship between the earnings of sons and fathers is analysed – takes Australia to a level of 0.2 to 0.3 – say 0.25. This puts Australia below the US experience and above the Nordic countries (although in the same ballpark as the UK).
For purposes of this discussion, I will refer to social mobility instead of inter-generational elasticity (high mobility translates into low inter-generational elasticity). The evidence suggests that inter-generational mobility tends to remain fairly stable over a forty year period. We need to focus on enduring characteristics of Australia that continue to make it more mobile.
In some areas, Australia made considerable progress on social mobility but so did the US. These included the boost to immigration, banning of racial discrimination, increases in university participation, improvement in gender inequalities and, for a time, a pull-back in unemployment. In other respects, Australia may have gone backwards, resulting in decreased inter-generational mobility, such as the increasing concentration of joblessness (distance between income quintiles was larger in the early 2000s), the abolition of federal inheritance taxes, further reductions in capital gains taxes, high income super concessions, tough new IR policies, a falling back in some low income groups without families and neglect of investment in public goods at a time when it would have been largely affordable.
On the other hand, there are some distinguishing features of Australia’s recent performance which could explain our improved inter-generational elasticity. Firstly, Australia has managed more sensitively the distribution of incomes, both under Hawke and later under Howard, in the form of special concessions for families. Indeed, the new comparative study lines up even better with OECD data on income distribution than before. Our welfare system is well targeted and means-tested.
Secondly, Australia’s national health coverage is far better than that in the US, (although some of this advantage may be being lost in the areas of dental treatment and elective surgery).
Perhaps Australia gained a little from its efforts to increase the compulsion on employees to go to work (increasing the relationship between net inflows and unemployment) and its promotion of free, competitive markets. Australia may have made important social progress in reducing disparities in education inequalities, such as the increased assistance given to disadvantaged students and the abolition of up-front university and tuition fees.
Looking further ahead, the Rudd Government has now taken steps to decentralise wage bargaining and protect the more disadvantaged workers, stopping the decline in level of trade unionisation (albeit with some unemployment risk). Rudd has also sought to reduce the digital divide, raised hopes for early development childhood education, improved assistance to disadvantaged children (although not undoing aid to public schools), lifted our game on health and taken a long string of measures to build up our investment in infrastructure, training and housing. Unemployment is likely to increase, but this will happen across the board. Unlike inequality, very few of these factors will impact on inter-generational mobility.
Countries enjoying relatively high levels of mobility tend to have governments which encourage free, competitive markets, regulate labour markets relatively lightly and engage in active social redistribution. The first two alone will not deliver optimal equality of opportunity. Active social intervention by governments is also required.
Fred Argy has advised governments from Menzies to Keating and has received an OBE and AM for services to economic planning. He has a long list of publications to his name.
*The author is indebted to Andrew Leigh for comments received on an earlier draft.
1. Australia Institute discussion paper no.85, April 2006. The original paper (of which these extracts comprise about 1/10th, has a full set of bibliographies and some of the selections have also been brought up to date. The document can be found at https://www.tai.org.au/file.php?file=DP85.pdf )
2. ‘Income inequality and Poverty rising in most OECD countries, 21 October 2008’, accessed at http://www.oecd.org/document/25/0,3343,en_2649_201185_41530009_1_1_1_1,00.html .
3. See B.E. Journal of Economic Analysis and Policy (2007) 7(2).