CHAPTER 5
The presence in the labour market of trade unions capable of exerting market power ensures that relative wages will deviate from their market rates and, therefore, that labour will be misallocated. The evidence shows that this is the case in Australia. (1) This conclusion will, of course, stand irrespective of the mechanism by which the influence of union market power is transmitted into the system for administering wages. Hence, just as collective bargaining between employers and trade unions will result in distortions of the structure of relative wages, so too will a system of compulsory arbitration which is responsive to union power.
It is a matter of casual observation that the Australian arbitration tribunals respond to the industrial power of both employers and unions. (2) However, it is often alleged that the arbitrators inject a double dose of distortion into the system by applying certain social criteria in wage fixing. The literature on Australian wage determination is peppered with words like "equity", "fairness" and "egalitarianism" when referring to the workings of the arbitral system. The arbitrators are widely perceived to be engaged in a levelling exercise -- ensuring that everyone in the same skill class gets the same pay and that the differences between the pay of those in different skill classes never become very large. If this is true, then the structure of relative wages is most unlikely to produce an efficient allocation of labour. The egalitarian structure of wages which it is alleged results from the arbitral processes will distort the allocation of labour by providing inadequate incentives for labour to find employment in its most productive uses.
If wage differentials between different skill classes are too narrow, in the sense that they offer too low a rate of return on the investment required to acquire the skill, then shortages of skilled labour will result. If both prosperous and impoverished firms and industries are constrained to pay similar wages for labour of the same skill then labour shortages will occur where employment should grow and labour surpluses will appear where employment should contract. If wages are uniform within occupations, the most productive workers cannot be rewarded. There is no incentive to seek out jobs where one's productivity is maximised.
If all that has been said about the tendency of the arbitral system to create an egalitarian wage structure is true, then it suggests that there will be major inefficiencies in the Australian labour market. But there are no countries in the industrialised world in which wages are generally determined by the free play of market forces, so the extent of the deviation of the Australian wage structure from the market solution cannot, in practice, be estimated. Moreover, the Australian wage structure can, in these circumstances, only be relatively more or less inefficient than that of other countries.
The comparison is usually made with respect to systems of decentralised collective bargaining. In particular, the US model of a collective bargaining system, where bargaining is conducted mainly at plant or company level, is viewed as one which is likely to be relatively responsive to market forces. This is so because the context in which bargaining occurs and the constraints on the exercise of union power will be determined by the local economic circumstances of each plant or company rather than by some average of the economy or industry. As a result, it would be expected that the bargains struck would be tailored to the economic conditions in the plant or company. This would tend to produce a structure of relative wages which reflected the differences between prosperous and less prosperous plants and companies and the conditions of supply and demand for different skills. There is no necessary presumption, however, that collective bargaining per se will tend to produce a greater dispersion of pay than an arbitral system within skill classes or occupations within the bargaining unit. But it would be expected to produce a greater dispersion of pay within skill classes or occupations across bargaining units, because different plants and companies will tend in general to pay relatively high or low wages.
On this basis we would expect the wage structure produced by a system of decentralised collective bargaining to be less egalitarian and less uniform than that produced by the arbitral system of wage-fixing employed in Australia. However, it should not be imagined that concepts such as "comparative wage justice" are unique to Australia. In almost all developed countries there is a counterpart of comparative wage justice. In the USA it arises in the form of "pattern bargaining" where the outcome of some particular bargain is taken as a pace-setter and attempts are made to imitate (flow-on) its terms in subsequent bargaining in other companies. Clearly, to the extent that pattern bargaining occurs, there will be a tendency to uniformity in negotiated rates of pay across companies in the USA. In the UK comparative wage justice goes under the name of "comparability", and in some European countries it is called "horizontal equity". There are other conventions which are common to both collective bargaining systems and the Australian arbitration system and this has led Dr Joe Isaac to comment that:
The criteria on which wages are fixed are substantially the same under both collective bargaining and compulsory arbitration. The interpretation of these criteria and the relative weights attached to them, however, vary from one situation to another under both systems. It is likely that in most cases the compromise reached under one system would not be very different from the compromise under the other system. (3)
We must be clear about the nature of the matter at issue. An economic evaluation of the wage structure which the arbitration system has created relative to the wage structure which the market would have produced in the absence of arbitration, trade unions, etc. is not feasible. One looks instead, therefore, to a comparison between the Australian wage structure and wage structures in other countries where collective bargaining is the predominant mode in wage-fixing. The results of such a comparison will simply tell us whether compulsory arbitration appears to produce a different wage structure from that which collective bargaining produces. There tends, however, to be a presumption that if such differences do in fact exist, they would indicate that arbitration yields a greater deviation from the market solution than collective bargaining does. A comparison between the wage structures in Australia and the USA is probably the most useful from this point of view. But, if we wish to know what the Australian wage structure might have been like if collective bargaining rather than arbitration had been chosen as the means of fixing wages, a comparison between the Australian and British wage structures is probably appropriate. This is because the structure, extent and philosophy of Australian trade unionism is more akin to that of Britain than it is to that of the USA or continental European countries. Indeed, one might go further and suggest that a comparison between the Australian and British wages structures will give an indication of what might result from shifting the emphasis away from arbitration in Australia towards a system of collective bargaining. This is, of course, the most practical aspect of such comparisons since it has a direct relevance to the current debate on the relative merits of arbitration and collective bargaining as options for the future of the Australian system of wage determination.
THE EMPIRICAL EVIDENCE
There is a variety of empirical evidence on the extent to which the Australian wage structure differs from that in other countries where collective bargaining is the main method of wage fixing. (4) Four main aspects of the wage structure have been examined in the various pieces of research:
- The inter-industry structure of wages [Hughes (1973), Norris (1980), Rowe (1982)];
- The inter-occupational structure of wages [Brown et al., (1978), Norris (1980), Norris (1985)];
- The intra-occupational dispersion of pay [Brown et al, (1978, 1980)];
- The inter-personal dispersion of earnings [Norris (1977, 1985)].
On the traditional view, that the effects of the arbitration system on relative wages are egalitarian, the prior expectation in each case would be that the Australian wage structure would be less dispersed than that in the UK or the USA.
THE INTER-INDUSTRY STRUCTURE OF WAGES
Since most awards effectively apply to industries, an obvious starting point in an analysis of the wage structure is that which exists between industries. Given the nature of the Australian wage fixing system, we might expect a fairly uniform rate of growth of award wages in different industries as the criterion of comparative wage justice and flow-ons of various kinds generalise the terms of some initial award. In contrast, one might expect the rate of growth of wages in different industries under collective bargaining to be disparate because of differences in the economic circumstances of each industry and the industrial power of its trade unions. This is particularly so when one takes into account the very great differences in the fortunes of declining industries, such as textiles, and expanding industries, such as electronics, in most Western economies over the last decade or so. The prediction would therefore be that the dispersion of average industry wage rates in Australia would be less than in the UK or USA.
Hughes (1973) was one of the first to address this question empirically. He selected a matched sample of over sixty industries in Australia, the UK and the USA and computed average earnings for each. Note that it was earnings that were computed, not award rates or collective agreement rates. Average earnings in each case will include over-time pay and, in Australia, over-award pay. Hughes used this data to compute a relative wage index in which the average earnings in each industry are expressed as a percentage of the median for all industries in each country. The data point was 1962-63 and scatter diagrams were used to present the evidence.
Hughes found that the inter-industry wage structure was similar in Australia and the UK. High wage industries in Australia were also high wage industries in the UK and similarly for low wage industries. Moreover, high and low wage Australian industries were generally similar amounts above and below the industry median wage as were their UK counterparts. Hence the dispersion in industry wages was generally comparable between the two countries.
The same techniques were applied to the comparison between Australian and US industries. Again there was a general similarity in the distribution of industries by average earnings, but it was not as close as in the comparison between Australia and the UK. After correcting for the proportions of females employed in each country, the high wage industries in the US are shown to be proportionately higher above the median than are Australian high wage industries, and US low wage industries lower below the median than their Australian counterparts. In rough quantitative terms, an industry which is "... twenty per cent above or below in Australia would be about thirty per cent above and below in the United States". (5)
Hughes concludes that the evidence suggests that, in 1962-63, "... the Australian Arbitration framework did not at that time exert much of an equalising effect on the industrial wage structure". (6) This conclusion reflects Hughes's view that only the comparison with the UK is wholly relevant, the comparison with the US being of "dubious relevance" due to institutional differences between the two countries and because "full employment" prevailed in Australia but not in the USA. This latter argument holds that full employment itself, almost independent of the wage fixing machinery, will tend to equalise wage increases across industries since "... economic pressures are likely to induce a substantial flow-on of wage gains to other sectors of the labour force". (7) Hence, since full employment had characterised both the Australian and UK labour markets since the 1950s, we might expect relatively compact wage structures whereas, because the USA has never really achieved full employment, we would expect a more dispersed wage structure there.
Rowe (1982) takes up the point that the dispersion of earnings may tend to be higher in times of high unemployment than in times when government pursues full employment policies. In periods of high unemployment, of course, any tendency of the arbitration system to equalise wage increases across industries will become more evident when compared to the experience of countries where collective bargaining is the predominant method of wage fixing. Rowe examines this issue by repeating Hughes's analysis for the years 1931-32, 1935-36, 1947-48 and 1955-56 and making comparisons between Australia and the UK but not the USA.
The period 1931-32 was the height of the Great Depression in both Australia and the UK. The evidence of the scatter diagrams does not, however, indicate that the inter-industry wage structure was more egalitarian in Australia than in the UK. Indeed, although one cannot draw firm conclusions, the evidence suggests that the reverse was, if anything, true. Similarly, the evidence for the other years, altogether spanning a wide rage of macroeconomic circumstances, shows no tendency for the Australian inter-industry wage structure to be more egalitarian than that of the UK. Norris (1980) adds supporting evidence for 1976, although the industries in his sample are so aggregated that one must query whether there is any meaning at all in his comparisons. Rowe computes coefficients of variation for the wage structure in each country which summarise the evidence.
The coefficient of variation is a measure of the relative dispersion of each wage distribution. The larger the estimated coefficient of variation, the more dispersed is the wage distribution.
Table 1 clearly shows that the dispersion of inter-industry wages in Australia has not consistently been less than that of the UK. Indeed, in four of the six periods considered, Australia displayed a greater degree of dispersion than the UK.
TABLE 1
Coefficient of variation of inter-industry wages (%)*
| Australia | UK | |
| 1931-32 | 16.28 | 14.19 |
| 1935-36 | 18.31 | 15.53 |
| 1947-58 | 11.15 | 8.80 |
| 1955-56 | 9.19 | 11.26 |
| 1962-63 | 9.39 | 11.95 (Hughes) ** |
| 1976 | 13.00 | 11.00 (Norris) ** |
* Taken from Rowe (1982), Table 1, p.260
** Figures for samples used by Hughes (1973) and Norris (1980) added (by Rowe).
There is no support to be found for the hypothesis that arbitration in Australia has been egalitarian in its effects on the inter-industry wage structure relative to the UK, either in times of high or low unemployment, in the Hughes (1973) or Rowe (1982) analyses. However, there are some reasons to be wary of an inter-industry analysis as a general test of the proposition.
Each industry is simply a mixture of different grades of labour, each paid a wage struck for that grade. Hence, average earnings for an industry are an average of the wage paid to each grade employed in it, weighted by the proportion which each grade comprises in the industry labour force. Now, since awards specify the wage rate of each (occupational) grade of labour, an egalitarian approach by arbitrators would presumably tend to compress wage differentials between grades of labour. If this had happened it would still be perfectly consistent with the findings of Hughes (1973) and Rowe (1982) although the egalitarian hypothesis would be true. This would be so because the wages of the unskilled would be high, and the wages of the skilled low, relative to the comparator country at the same time as industry average earnings were similar. The evidence on this issue is considered when we discuss the inter-occupational structure of wages.
Another issue arising from the inter-industry wage analysis concerns the use of average earnings rather than award rates as the measure of the industry wage. Arbitrators fix only the award rate but any over-award payments will be included in average earnings. If over-award payments are essentially a response to market pressures, the structure of average earnings will be partly determined by arbitration awards and partly by market forces. Hence, it may be that the Australian wage structure is not egalitarian as a result of the effect of market forces and despite the efforts of the arbitrators. If this is so, it calls into question the need for a comprehensive arbitration system.
THE INTER-OCCUPATIONAL WAGE STRUCTURE
The most widely accepted theory of occupational wage differentials is the theory of human capital. Differences in the wages of different occupations, after allowing for non-pecuniary benefits, are held to arise from differences in the cost of the education and training needed to acquire the skills necessary to engage in each occupation. Hence, the more "skilled" an occupation is, the more likely it is to be costly to acquire the necessary skills and the higher the income that will subsequently be earned. (8)
The return to investments in education and training which enhance an individual's earning capacity can be measured as a rate of return on the cost of the investment in much the same way as rates of return on physical capital or interest rates on deposits are measured. The hypothesis that arbitration tends to yield an egalitarian wage structure implies that wage differentials will be unduly compressed and this, in turn, implies relatively low rates of return on investments in human capital. A consequence of relatively low rates of return to human capital is likely to be shortages of labour in the occupations affected as people choose to make investments in higher yielding alternatives.
Comparisons between the Australian inter-occupational wage structure and those in other countries have concentrated on direct observation of earnings rather than rates of return to human capital. (However, some recent research makes the latter possible for some occupations.) (9) There are some considerable difficulties in making international comparisons of this kind, however. Probably the most important is the problem of identifying the same occupation, or even broad skill class, in the statistics of different countries. Moreover, there are many definitional problems in securing comparable wage data.
At a global level Phelps-Brown has produced an interesting set of international comparisons of the ratio of the wage rate of skilled manual workers to unskilled workers.
Most of the calculations relate to the building tradesman/labourers differential but for Australia they are computed from twenty skilled and twenty unskilled occupations in Melbourne. (10)
For reasons already mentioned these data permit only very tentative conclusions to be drawn. In general the Australian skill differential has been low by the standards of the other countries listed but by 1961 had risen above the differential in the UK. As in the other countries, the Australian differential declined over the period as a whole but rather less evenly than in the others. The influence of the Arbitration Commission can be detected in some periods. For example, the Commission's policy of adjusting the basic wage, but not margins, in line with prices in the late 1920s and early 1930s accounts in large part for the relatively low differential observed in 1928-30. However, from 1938 until 1953 the Australian differential was very close to that of the UK and actually exceeded it by 1961. Hence the evidence suggests that prior to 1938 the Australian skill differential was relatively low by comparison with the UK and by comparison with the USA and Canada it was relatively low throughout the period. On the basis of this evidence we cannot rule out the possibility that the arbitration system played an egalitarian role, at least during some periods, in the manual inter-occupational wage structure.
Norris (1980) considers the same Australian data but compares them with a different series for the UK from the one used by Phelps-Brown. The outcome is very similar and Norris concludes that in general the differences between Australia and the UK in respect of the skill differential are small, but that in the inter-war years there is evidence of a more significant compression in the Australian differential.
More recently Brown et al. (1978) have produced some further evidence on the manual skill differential in the course of a larger study of earnings structures in Adelaide and Coventry in the UK. The data relate to 1974.
These data refer to average earnings from a sample of firms surveyed in each city at the same time, and the personal involvement of the authors in the survey (in Adelaide) offers some guarantee that problems of definition, etc. are likely to be small. The figures themselves leave no doubt that skilled manual workers in Coventry enjoy a considerably larger wage differential over semi-skilled and unskilled workers than do their counterparts in Adelaide. Semi-skilled workers, however, enjoy a broadly similar differential over unskilled workers in both cities.
These findings suggest either that the British skill differential rose sharply relative to that of Australia after the early 1960s or that the data in Table 2 and those in Table 3 are measuring different phenomena. The confusion is compounded when one observes in Norris (1980) that the differential of a fitter over a labourer in engineering in 1975 was "27% in Australia (NSW) and 25% in Britain; the corresponding figures in 1973 were 29% and 28%". Clearly the time, place and definition of the measured wage differential are vital in such comparisons, and until some reasonably comprehensive and consistent data are available we are unable to draw a conclusion.
TABLE 2
Ratio of wage rate of skilled manual workers to unskilled
| USA | Canada | Australia | UK | |
| 1912-14 | 198 | 201 | 136 | 150 |
| 1920-22 | 169 | 229 | 126 | 127 |
| 1928-30 | 178 | 235 | 124 | 134 |
| 1938-40 | 170 | - | 128 | 130 |
| 1945-47 | 148 | 207 | 126 | 126 |
| 1951-53 | 137 | 203 | 116 | 116 |
| 1959-61 | 127 | 161 | 125 | 113 |
Source: E.H. Phelps-Brown, The Inequality of Pay, Oxford, OUP, 1977, Table 3.2, p.73.
TABLE 3
Selected ratios of wage of skilled manual, semi-skilled, unskilled 1974
| Adelaide | Coventry | |
| Fitter/Forklift | 116.8 | 136.5 |
| Fitter/Storeman | 117.6 | 137.4 |
| Fitter/Cleaner | 128.0 | 183.2 |
| Electrician/Forklift | 115.8 | 128.5 |
| Electrician/Storeman | 117.0 | 127.9 |
| Electrician/Cleaner | 129.8 | 145.8 |
| Forklift/Storeman | 102.1 | 99.2 |
| Forklift/Cleaner | 113.8 | 114.4 |
| Storeman/Cleaner | 110.7 | 114.7 |
Source: Brown er al. (1978), Table 2, p.36.
In a more recent study Norris reverses his earlier conclusion that the wage structure in Australia was no more compressed than that of Britain [Norris (1985)]. Norris reviewed a number of existing studies as well as analysing new data on occupational wage differentials in Australia and Britain. His analysis of thirty-two occupations reveals that in almost every occupation in which average earnings are less than 83 per cent of the average for all occupations, the Australian wage is higher than the British wage as a percentage of the average. Taken together with the other evidence this leads him to conclude that, while the differences in wage structures are small '... virtually every difference we identified ... pointed in the same direction, that relativities are narrower in Australia.
Although the available evidence does not permit a firm conclusion to be drawn, there is a strong suspicion that the Australian skill differential may be lower than that in the UK, USA and Canada as a consequence of the activities of the Australian arbitrators. If this was so, it would imply that rates of return to investments in training were lower in Australia than in these other countries and that, all things being equal, shortages of skilled labour are likely to be more widespread in Australia than in the other countries in consequence. In fact, although shortages of skilled labour are common enough in Australia, they do not appear to be any worse than in the other countries. An explanation for this is that Australia's immigration programme, with its criterion of "skills in demand", may systematically neutralise shortages of skilled labour. So long as the absolute wage of a skilled worker in Australia is no lower than in the UK it will still be worthwhile for skilled immigrants to come to Australia since their investment in training will have been cheaper in the UK. This is so because the cost of training is largely made up of the income foregone while undertaking training and that is measured by the wages which can be earned in unskilled work. The result would be, however, that Australians are disproportionately employed in lower-skilled jobs because training for skilled occupation is not economically worthwhile.
Norris (1980) also offers evidence on the inter-occupational wage structure for professional workers. The median earnings of ten professional occupations relative to the median earnings of all male workers in each of Australia and the UK are compared and it is found that in seven of the ten occupations relative earnings in the UK are higher, but the differences are mostly quite small. While the evidence suggests a slightly greater dispersion of inter-occupational earnings due to these groups in the UK, it is doubtful if the arbitration system can be greatly implicated in the matter because it has no jurisdiction over most of the professions in the comparison.
THE INTRA-OCCUPATIONAL WAGE STRUCTURE
The wages paid to the members of any given occupation may be expected to vary both within and across the establishment in which they are employed. There are a number of reasons for this. Employers will generally wish to reward high-quality and experienced labour; firm-specific skills may also attract a premium in order to keep them in the firm; employers may practise wage discrimination; workers may not have sufficient information about better-paying jobs in their occupation in other firms or may, due to inertia, not be prepared to search them out and move to them; and individual firms face different labour and product market circumstances which generate differences in occupational pay across firms. Observed intra-occupational pay dispersion has, in various studies, been in some measure associated with all of these factors.
Arbitrated wage awards, which set rates of pay for occupations, might be expected to impose a measure of uniformity on the intra-occupational pay structure. In contrast, collective bargaining in the UK and the USA is highly decentralised, usually based on the company or plant. It might therefore be expected that a greater dispersion of pay would be found within occupations under collective bargaining than under compulsory arbitration. The economic consequence of a uniform rate of pay within occupations would be that "... the more productive workers will have no incentive to move to better paid jobs (because there would be none) and effort and efficiency are likely to suffer". (11)
Brown et al. examine this issue by comparing intra-occupational wage structures in three cities -- Adelaide, Coventry, and Chicago. Their findings are published in two papers, Brown et al. (1978) and Brown et al. (1980). The methodology used is to estimate coefficients of variation of occupational earnings in each of the three cities. Table 4 sets out some of their findings.
TABLE 4
Dispersion of Individual Standard Earnings of Occupation
| Occupation | CV % |
| Adelaide Factory Cleaner (M) Driver (3-6 ton truck) (M) Forklift Truck Driver (M) Electrical Fitter (M) Ledger Mechanist (F) Typist (F) | 10.45 19.66 11.05 7.73 11.10 11.82 |
| Chicago Janitor (M) Truck Driver (M) Forklift Trucker (M) Maintenance Electrician (M) Keypunch Operator (F) Typist (F) | 14.06 5.70 15.10 12.19 12.80 14.44 |
| Coventry Non-Foundry Labourer (M) External Truck Driver (M) Internal Truck Driver (M) Electrician (M) Accounting m/c Operator (F) Copy Typist (F) | 13.54 13.78 12.57 10.53 12.41 11.27 |
Source: Brown et al., (1980) Table 2, p.225.
With the exception of truck drivers, the general picture is one of broad similarity between intra-occupational pay dispersions in all three cities. While the estimated coefficients of variation for Adelaide are generally slightly lower than those of Chicago and Coventry, they are hardly substantial. In the case of truck drivers there is some evidence to suppose that special factors differentiate the cities. In Adelaide implicit food and accommodation expenses may be included in the earnings figure while in Chicago multi-employer bargaining may have resulted in fairly uniform rates of pay.
After considering this and much more evidence Brown et al. come to the conclusion that the differences in intra-occupational pay dispersion between the three cities are not large enough significantly to influence labour market behaviour. In terms of the incentive for workers to search for better-paying jobs, the more dispersed pay structure offers potentially greater rewards. However, given the actual differences in pay dispersion set out in Table 4, the potential returns to search differ between cities by quite trivial amounts. For example: "The marginal increase in the expected maximum gain for the Chicago employee would be greater than that for the Adelaide employee as a result of their fourth search by an amount equivalent to about 7 hours pay on a year's salary." (12) Hence, in terms of the allocative effect of intra-occupational pay dispersion, the slightly less dispersed Adelaide wage structure does not appear to imply potential labour market misallocation relative to Chicago or Coventry.
It may be noted in Table 4 that the only skilled manual occupation in the sample was "electrician", and that the dispersion of electricians' pay in Adelaide was significantly lower than in Chicago or Coventry. Similar findings apply to almost all skilled manual workers in the full study. Brown et al. are unable to account for this phenomenon and "... speculate that it is connected with the fact that [an Australian skilled worker] is also anomalous in the degree of compression of his pay differential over the less skilled manual workers". (13)
THE INTER-PERSONAL DISPERSION OF EARNINGS
Finally, Norris has examined the distribution of individual earnings across the labour force in Australia and the UK [Norris (1985)]. If the Australian arbitration system tends to promote egalitarianism in specific aspects of the wage structure, we might expect that it will be manifest in a relatively narrow dispersion of individual earnings. Of course, many influences on the distribution of individual earnings, other than the system of wage fixing, may influence relative dispersion, but as a piece of additional evidence it is worth considering [see Phelps-Brown (1977), Ch. 9)].
Norris compares dispersions of individual earnings on the basis of an Australian survey taken in 1981 and the British New Earnings Survey 1981. Although the two samples are not entirely comparable Norris argues that reasonably reliable conclusions may be drawn from them.
The measure of dispersion used is the ratio of various percentiles to the median of each distribution. Norris's findings are summarised in Table 5.
TABLE 4
The Dispersion of Weekly Earnings of Adult Workers in Britain and Australia, 1981
| Percentile Points as a Percentage of Medium | ||||
| Males | Females | |||
| Australia | Britain | Australia | Britain | |
| Lowest decile | 71.9 | 65.6 | 77.6 | 68.0 |
| Lowest quartile | 82.7 | 79.8 | 87.8 | 80.6 |
| Upper quartile | 125.4 | 129.5 | 119.6 | 129.8 |
| Highest decile | 154.4 | 167.7 | 144.1 | 172.5 |
| Mean | 109.6 | 111.1 | 105.6 | 111.2 |
Notes:
(1) Australian source data exclude government managerial employees.
(2) British source data confined to those whose pay was not affected by absence.
(3) All differences save at mean for males significant at 5% level.
Source: Norris (1985)
From the table it may be observed that Australian male workers in the lowest 10 per cent of the earnings distribution earn 71.9 per cent of median earnings while their British counterparts earn only 65.6 per cent of the median of their distribution. Similarly, the best paid 10 per cent of male workers in Australia earn 154.4 per cent of the median while their British counterparts earn 167.7 per cent of the median. At all points throughout the distributions a similar pattern may be observed and shows that the dispersion of individual earnings in Australia is significantly lower than in the UK.
While it is impossible to be sure that the differences in individual earnings dispersions are associated with the arbitration system, it is at least consistent with the hypothesis that the arbitration system has been egalitarian in its effects on the wage structure.
CONCLUSIONS
A study of the empirical evidence on the extent of differences between the Australian wage structure and the wage structure of countries where collective bargaining is the predominant mode of wage determination does not settle any of the arguments. Instead, it points to the need for further research of a comprehensive and detailed kind. Moreover, the inherent weakness of international comparisons in an analysis of this kind are highlighted by the very inconclusive nature of the findings reported.
The only evidence that tends to support the hypothesis that the influence of arbitration on the Australian wage structure is relatively egalitarian, is Hughes's (1973) finding that the US inter-industry wage structure is apparently rather more dispersed than that of Australia, the finding that, at least in some periods, the wage differential of skilled manual labour over semi-skilled and unskilled labour is relatively small, and Norris's conclusion that the dispersion of individual earnings is significantly lower in Australia than in the UK. For the rest, there is apparently a fairly high degree of similarity between the Australian and overseas wage structures.
In part at least, the similarities in wage structures here and abroad are accounted for by the fact that collective bargaining in the UK and the USA, as well as in other countries, trades in the same sort of moral currency as the arbitration system in Australia. "Fairness" and "equity" are concepts that pervade trade union attitudes to wage determination in the collective bargaining process and they give rise to criteria that correspond closely to those which Australian arbitrators employ.
Within a collective bargaining system it might be expected that these concepts would be modified by market forces to a greater extent than in Australia. It may be that the differences which do appear to exist in the international comparisons are reflecting this tendency. But it seems as though the factors which are common to both systems, through the very different institutional arrangements for wage determination, dominate the final outcomes.
While recourse to moral arguments in collective bargaining mirrors many of the criteria of compulsory arbitration, the presence of over-award pay in Australia permits a measure of flexibility in the wage structure that seems likely to accommodate some of the same forces which determine bargaining outcomes. In particular, it appears that over-award pay reflects a number of firm-specific factors, including product and labour markets. (14) Hence, both compulsory arbitration and collective bargaining may be viewed as systems that respond to similar sets of forces. (15) But the relative importance of different aspects of these forces is almost surely different between the two systems. The empirical evidence reviewed in this paper suggests, however, that the differences are greatly outweighed by the similarities. But this is not conclusive and further research is called for.
ENDNOTES
1. C. Mulvey, "Wage levels: Do unions make a difference?", in J. Niland (ed.), Australian wage fixing: processes and outcomes, Sydney, Allen and Unwin, forthcoming, 1985.
2. K. Hancock, "The arbitration tribunals and the labour market", Growth, 33, September 1983.
3. Cited in B. Hughes, "The wages of the strong and the weak", Journal of Industrial Relations, March 1973.
4. The works considered in this paper are:
- B. Hughes, "The wages of the strong and the weak", Journal of Industrial Relations, March 1973;
- W. Brown et al, "Occupational pay structures under different wage fixing arrangements: a comparison of intra-occupational pay dispersion in Australia, Great Britain and the United States", British Journal of Industrial Relations, July 1980;
- K. Norris, "Compulsory arbitration and the wage structure in Australia", Journal of Industrial Relations, September 1980;
- W. Brown et al, "How far does arbitration constrain Australia's labour market?", Australian Bulletin of Labour, September 1978;
- L.G. Rowe, "Reason, force and compromise: egalitarian wage structures under bargaining and arbitration", Journal of Industrial Relations, June 1982;
- K. Norris, "The wage structure; does arbitration make any difference?", in J. Niland (ed.), Australian Wage Fixing: Processes and Outcomes, Allen and Unwin, 1985;
- K. Norris, "The dispersion of earnings in Australia", Economic Record, December 1977.
5. Hughes (1973) p 17.
6. Ibid, p 20.
7. Ibid, p 4.
8. It is of interest that the Arbitration Commission explicitly recognised this point as early as 1913. In the Builders' Labourers' Case in that year Higgins remarked: "I have to keep steadily in view the recognized practice of treating men of special training or gifts as entitled to higher wages than other workers, and I must do nothing to encourage lads in the idea that they will be as well off in life if they do not apply themselves to the attainment of special skills in industrial work, as if they do so." (Builders' Labourers' Case 1913, 7 CAR 210. Cited in D. Plowman, S. Deery and C. Fisher, Australian Industrial Relations, McGraw-Hill, 1980.
9. P.W. Miller, "Education and the distribution of earned income", paper presented at NILS-CEPR-BLMR joint seminar on Understanding Labour Markets in Australia, held at Flinders University, May 1982.
10. Figures compiled from D. Oxnam, "The relation of unskilled to skilled wage rates in Australia", Economic Record, June 1950, and K. Hancock, "The wages of the workers", Journal of Industrial Relations, March 1969.
11. Brown et al. (1978) p 32.
12. Ibid.
13. Brown et al. (1980) note 24, p 230.
14. Brown et al. "Supply-side and demand-side pressures in wage determination: Some Australian evidence", mimeo, May 1982.
15. This general argument has been consistently advanced by Isaac. See for example J.E. Isaac, "Economics and industrial relations", Journal of Industrial Relations, December 1982.
No comments:
Post a Comment