This Version Jan 96
The intensification of work and the polarisation of labor
Department of Economics
James Cook University
FAX + 61 77 814149
Phone + 61 77 81 4798
+ 61 77 251269
The intensification of work and the polarisation of the labor force
The dominant theme in microeconomic reform and competition policy, has been the need for people to work harder.
The interaction of these microeconomic reform policies imperatives with trends in technology and relative prices has produced an increasing polarisation of the labor market. At one pole, there is a group of full-time workers facing demands for steadily increasing hours of work and work intensity. At the other pole are those who are unemployed or in insecure often part-time employment.
This is a cross-section view of the problem. Because these trends have only been present for one or two decades, it is more difficult to get a lifetime view. One interpretation of the data is that the polarisation of the labor market present at any point in time reflects an underlying polarisation of the population as a whole. An alternative interpretation involves a life-cycle view, in which people move from the periphery to the core labor force and perhaps back again.
Both interpretations imply, prima facie, a serious misallocation of work and leisure. On the first view, some individuals or households have too little leisure throughout their working lives to enjoy the relatively large incomes they earn, while others have involuntary leisure and inadequate consumption. On the second view, the misallocation occurs within the lifecycle, with people being massively overworked during their childraising years, then facing decades of idleness.
Private sector bias
The main focus of this paper will be on the second theme. However, in order to understand the impact of microeconomic reform and work intensification, it is necessary to observe the way in which the economy has been distorted by a misplaced faith in private sector dynamism. The most common manifestation of this faith is the belief that we will reap rewards for opening up our economy in the form of a re-invigorated manufacturing sector, exporting large volumes of 'elaborately transformed manufactures'. Advocates of this idea sometimes industry policies of various kinds, but the dominant view is that the only necessary policy is the provision of a 'level playing field' undistorted by tariffs, subsidies and the like.
This view has been disputed at length. I want to make a more fundamental point. The idea of a level playing field is meaningless when the size and shape of large sectors of the economy is determined primarily by government policy. The main sectors where government policy is the predominant factor in determining the allocation of resources are the community services sector (including health and education), the infrastructure sector (including communications and transport) and the financial services sector. Together these sectors account for more than one third of total employment, output and investment. When combined with public administration and defence, and a range of smaller areas of activity where government policy is the primary determinant, it is likely that more than half of all economic activity is determined mainly by government policy rather than directly or indirectly by consumer demand. Many of the services in these sectors have been provided by private enterprises, and the reliance on private provision is increasing. But this does not alter the fact that the allocation of resources between these sectors, and between the policy-determined sector and the rest of the economy is determined primarily by government policy.
Yet policy discussion in Australia does not involve any serious attempt to consider the optimal allocation of resources. Instead, resource allocation is driven by dogmas and shibboleths, such as the need to reduce the share of public expenditures and revenues and to 'deregulate' capital markets. In Langmore and Quiggin (1994), it was argued that the result was a serious misallocation of resources including an overexpanded financial sector and inadequate provision of services such as health and education. The net result, it was argued, was a contraction of labor demand as well as a misallocation of resources. A program for full employment must involve a significant increase in the relative size of the community services sector. These arguments have been ignored by government, primarily because they would imply an increase in taxation.
The main implication of the belief in private sector dynamism has been the acceptance of high levels of unemployment.
Increases in work intensity
The characteristic feature of the labour market over the decade since 1985 has been increasing work intensity. This is in part due to developments in the labor market, but the trend has been amplified by the policy package of microeconomic reform.
Reform of government business enterprises
The area of microeconomic reform most obviously linked with increased work intensity has been the reform of government business enterprises. The special case of waterfront reform may be regarded as falling into this class, even though the stevedoring employers were nominally private. Prior to the period of microeconomic reform, the popular view of the public sector as a whole was one of underworked government employees. The clichés of the tea-sipping public servant and the idle wharfie and railway worker were part of common knowledge and had sufficient correspondence to reality that they could not be dismissed. Along with these clichés went the belief that these areas of inadequate work intensity represented a major economic problem. The view that the poor performance of the waterfront industries represented an obstacle to increased 'competitiveness' was and remains strong (see, for example, BIE 1995). Political programs, particularly those of the Liberal Party have also frequently assumed the existence of large potential savings from the elimination of fraud, duplication and waste in the public sector.
Microeconomic reform has been highly effective in eliminating areas of inadequate work intensity. Employment in the railway and waterfront industries has been cut in half. In the public service, work intensity increased in two main ways. First, the steady application of staff ceilings and 'productivity dividends' has maintained continuous pressure to reduce staff numbers. Second, large reductions in the numbers of routine clerical positions have been combined with a relative (and in many cases, absolute) expansion in employment at senior levels, where most employees worked long hours even before reform. Large increases in labor productivity have also been achieved in government business enterprises such as Telstra and the electricity authorities where, at least in the popular wisdom, there was not a great deal of slack to begin with.
Yet the results in terms of the overall performance of the economy have been disappointing, to put it mildly. Per capita output growth in the microeconomic reform has been slower than in the 1950s and 1960s and no faster than in the turbulent decade from 1973 to 1983. In large measure, this may be explained by the observation that policy has been directed at the wrong problems. In the case of waterfront reform, for example, the central problem was the existence of around 3000 surplus employees. The loss to the economy can be best measured by the amount those employees could be expected to produce in alternative employment. In most cases, the answer is probably zero -- the likelihood of a 50 year old retrenched wharf laborer with a large redundancy package seeking and finding work is not very high. But even if we supposed that every one of these workers could be re-employed at average weekly earnings, the resulting social gain would be only around $100 million a year, or about 0.02 per cent of GDP. Larger estimates that have been put around are based in part on a confusion between efficiency gains and transfers, and in part on a confused belief that the waterfront is, in some sense, strategic. Waterfront employers, and the import and export industries they serve have gained condiderably more than $100 million a year from reform, because waterfront workers were being paid more than they could earn in the market. The effect of waterfront reform was to redistribute much of this rent away from employees; moreover, the government picked up much of the tab in the form of lump-sum payments. The idea that the waterfront is strategic is evident in the discussion of BIE (1995). It may be first observed that Australia's poorest performance is in the handling of containers, which mostly contain inbound consumer goods -- not most people's idea of a strategic activity. But if the waterfront is truly strategic, in the sense that the benefits of rapid low-cost handling of goods are greater than would be estimated from the total expenditure on waterfront services, the most appropriate response is a subsidy. The fact that no-one has suggested this indicates the confused thinking behind the idea that waterfront reform would generate large economic benefits.
The same point may be made in relation to other areas of work intensification in government business enterprises. In a situation of chronically high unemployment, the net social payoff from identifying and eliminating areas of labor slack will in general, be considerably less than the direct saving to the employer, particularly if the workers involved are unskilled or are old enough to make attractive the option of permanent withdrawal from the labor force. In the case of government business enterprise reform, even the direct savings to employers have not been all that great. Most government business enterprises are fairly capital intensive, so that even large proportional reductions in labour inputs yield only small savings in total costs.
The Hilmer reforms and competition policy
The linkage between competition policy and the intensification of work is less obvious than in the case of government business enterprises. However, the primary implications of the competition policy reforms implemented under the COAG agreements of 1995 involve a presumption in favor of competitive tendering and contracting of a wide range of government services, and of increased reliance on competition for government business enterprises more generally. By contrast the implications for the private sector, including an examination of arrangements covering the sugar industry and suburban newsagencies, are relatively trivial.
There seems to be little doubt that competitive tendering and contracting frequently leads to budgetary cost savings, though the magnitude of these savings has been disputed. In some cases, particularly those where public employees are replaced by contract workers, the cost savings may be traced fairly directly to pay reductions. However, in many cases the main saving arises from a smaller number of workers doing the same job. Prima facie, it seems reasonable to infer that cost savings arise primarily from increased effort.
More detailed studies support this conclusion. The main source of efficiency gains explicitly noted in studies of garbage collection by Domberger et al is the replacement of fixed 'task and finish' payments (sometimes for tasks which have become less onerous since the time rates were set) with piecework rates. Productivity gains from such changes in payment schedules will arise primarily from increased effort. Furthermore, the observation that tasks have become less onerous since rates were initially set implies that the new piecework rates will embody an effective reduction in wages in the absence of increased effort. Ganley and Grahl (1988) cite a number of cases of increases in working hours or reductions in working conditions associated with contracting out of garbage collection.
There is a tendency, criticised in Quiggin (1992), to regard output gains arising from increased effort as a free good. Such a view has no basis in neoclassical economic theory, although it may be supported by X-efficiency arguments (Liebenstein 1966). Such arguments require the existence of unexploited gains from trade, arising from distortions in the bargaining process (in addition to any element of union monopoly power). The implied claim is that the wage bargain embodies excess on-the-job leisure, in the sense that both workers and employers would prefer a bargain with higher wages and higher work intensity. There is no clear reason why bargains should be systematically biased in this way.
The argument that apparent increases in technical efficiency are more likely to reflect increases in work intensity gains strength from consideration of the tasks contracted out. Cleaning is a job with very limited capacity for either technical or organisational innovation. Cleaners typically work alone or in pairs, using equipment which has not changed substantially for decades. Furthermore, cleaners are normally paid for completion of a specified task, with no direct monitoring of time spent on the task, so that there is little scope for unions to impose restrictive work practices. The claim that there exist unexploited changes in cleaning methods permitting a twenty per cent increase in output with no increase in effort or reduction in service quality seems inherently implausible.
In the case of garbage collection, there is more scope for technical innovation (for example through the use of large 'wheelie' bins). However, such changes would probably have been captured explicitly in the detailed data available to Domberger et al. The simplest source of increased labour productivity in tasks of this kind is speed-up -- in this case, requiring collectors to run faster. Up to a certain point, speed-up yields efficiency gains as well as transfers, since the truck and driver are used more efficiently. However, speed-up can easily be pushed to the point where the risk of acute injury or repetitive strain injury increases at a rate that outweighs any benefits of more effective capital utilisation. Contracting out may well create incentives for employers to disregard these costs, either because of moral hazard associated with workers compensation insurance or because of the availability of workers who are unaware of the risks of injury.
Taken together, government business enterprise reform and competitive tendering and contracting account for the bulk of the GDP gains from Hilmer and related reforms estimated at 5.7 per cent by the IC (1995). A large component of the remaining estimated gains arise from highly speculative suggestions about the benefits of relaxing building codes and moving from direct regulation to self-regulation. Thus, although the re-analysis by Quiggin (1995) suggests that the IC's estimates of benefits from government business enterprise reform and competitive tendering and contracting are greatly overstated, they remain the principal source of the smaller estimated social gain of 0.5 per cent of GDP derived in that study.
Labor market reform
The tendency towards increasing working hours became apparent earlier in the United States. In Australia, however, significant reductions in work intensity continued at least until the 1970s. Important aspects of this trend included the general shift to four weeks annual leave, extension of long service leave, widespread adoption of maternity leave and a reduction of standard working hours from 40 to 38 hours (implemented as a monthly rostered day off) in many industries. The introduction in the Australian Public Service of the flexitime system, under which working hours could be varied at the employees' discretion, was another important step. Although the flexitime system was not formally adopted outside the public sector, impressionistic evidence suggests a reduction in the degree to which working hours were tightly controlled by employers and particularly in the prevalence of devices such as timeclocks, over this period.
The failure in the late 1970s of union campaigns for a 35 hour week signalled an end to this trend. From the mid 1980s onwards, the trend was reversed. Enterprise agreements and individual contracts rarely involved reductions in total hours and frequently involved increases. Moreover, where flexibility had previously implied greater discretion on the part of the employee it now implied that working hours would be varied according to the requirements of employers. A trend encapsulating both these tendencies was the shift from employment on the basis of a fixed working week plus overtime payments at time and a half or double time rates for any extra work with an annual salary based on a commitment to work extra hours as required without additional payment.
A number of factors contributed to the increase in work intensity. First, there was the general decline in the bargaining position of labour, reflected in the declining real wages of unskilled and semiskilled workers in the United States and in the rise in unemployment elsewhere in the OECD. The decline in the bargaining position has been reflected in declining union power and in the increasing tendency for industrial disputes to be initiated by employer demands for changes in working conditions, usually in the direction of increased work intensity.
For a variety of reasons, both economic and cultural, employers tend to demand long working hours. The simplest economic factor arises in the short term, when wages are sticky. Every increase in work intensity or unpaid working hours represent a pure gain for employers. Longer term gains arise from reductions in the fixed costs of taking on additional staff, including costs of initial assessment and subsequent monitoring. There are also reduced costs of capital broadening, the provision of worker-specific capital such as desks and computers that is used more intensively by a smaller number of workers working overtime than by a larger number working standard hours. At an industry level, employers as a group have an interest in demanding longer hours since this implies an effective increase in the supply of labour and therefore a reduction in the equilibrium price.
Cultural factors are also important, to a degree that economists are frequently reluctant to admit. The 'tournament' character of managerial hierarchies and the need for ostentatious displays of commitment means that managerial staff tend to work very long hours, longer than those that would yield the optimal ratio of output to effort. Anecdotal evidence suggests moves towards that soem firms are moving towards an unofficial version of the rule, overtly imposed in Japan, that 'no-one leaves before the boss'. As the power of management increases, and contract employment is extended to larger numbers of workers, there is a tendency for the cultural norms of management to become the benchmark against which working conditions are assessed.
Labour market policy has reinforced this tendency. Guidelines on enterprise bargaining have stressed the need for all wage increases to be justified by productivity gains. This idea makes sense at the aggregate level, since, assuming the wage share is held constant, the rate of increase in real wages must be equal to the rate of productivity growth. At an industry or enterprise level, however, the idea is nonsensical. The underlying rate of productivity growth delivered by technological change differs across industries. However, it is obviously not sustainable for nominal wages for workers engaged in similar tasks to grow steadily in those industries with rapid technological progress while remaining static in industries with a mature technology. In either a centralised or a fully decentralised system, this will not occur. The principles of comparative wage justice and of market equilibrium are at one in insisting that identical workers must receive identical wages, regardless of where they are employed. Assuming that wages grow in line with aggregate productivity, this means that wages will grow more slowly than physical productivity in some industries and more rapidly than productivity in others. Since the return to capital must also be equalised, prices for the output of mature industries must rise relative to the prices of technologically progressive industries. This change in prices will offset differences in physical productivity so that, in value terms, the rate of productivity growth will be much the same in all industries.
The enterprise bargaining system has been characterised by persistent denial of this logic. Strenuous attempts have been made to hold to the principle that each individual wage increase must be justified by an increase in physical productivity. In industries with little capacity for improvements in technical productivity, this demand can be met only by increases in work intensity. In principle, this effect might be offset by reductions in work intensity in high productivity industry. But any such reductions in intensity leave employees in those industries vulnerable to competitive pressure, enhanced by the widespread adoption of contracting out in both private and public sectors. The result is that the enterprise bargaining process has been associated with a combination of static real wages for given occupations combined with a general increase in work intensity. To the extent that real wages have risen over time, this has been primarily the result of changes in the composition of employment rather than of occupation-specific increases in real wages.
A move towards individual contracts will probably eliminate this bias. However, this offect will be offset by the increase in employer bargaining power associated with individual contracts. The requirement for no net disadvantage will tend to encourage work intensification. Proof of net disadvantage will be relatively difficult in cases where a wage increase is traded for increased work intensity. Employers may well calculate that wage increases can easily be clawed back in the long run through inflation.
The statistical evidence
The statistical evidence on work intensity does not yield any easy answers. The basic data on hours worked per fulltime employee indicates a general tendency to increasing hours. However this data does not discriminate between standard hours worked on the main, overtime and hours worked in second jobs.
Overtime data is also complicated. The official data shows a slight increase in average hours of overtime per employee. Since overtime hours are worked almost exclusively by fulltime employees and the proportion of fulltime employees has declined substantially, average hours of overtime worked by fulltime employees have increased. However the official data is compromised by the fact that most overtime is unpaid and unreported.
Composition effects also complicate the picture. In general employment in sectors with high levels of paid overtime has contracted. That is, in the absence of composition effects the average number of overtime hours per fulltime employee would have increased more rapidly than actually observed. However, it may be that the composition effect reflects a shift from sectors where overtime is mostly paid to sectors where it is mostly unpaid.
At first blush, the data on multiple jobholding appears to support the supply side hypothesis. The decision to take on more than one job is presumably one which is made freely by the workers concerned. Indeed, multiple jobholding is often colloquially referred to as moonlighting. The associated image is one of a full-time employee taking on a second job, presumably without the knowledge or consent of the principal employer.
Detailed examination of the data reveals a different picture. In particular, the moonlighting image is inaccurate. The increase in multiple jobholding has taken place primarily among females. Although the time-series data on the characteristics of multiple jobholders is unsatisfactory, cross-section data suggests that most females with multiple jobs held two or more part-time jobs. That is, rather than being a moonlighter, the typical multiple jobholder looks more like a marginalised member of the labour force, cobbling together part-time jobs in order to earn something less than a fulltime wage.
To summarise all of this, the time series data generally supports the view that the increase in working hours for fulltime employees observed over the decade has taken place at the initiative of employers rather than employees. It would be useful to obtain information based on direct questioning about this. The ABS occasionally asks employees whether they would like to work more, less or the same hours. The last survey of this kind for fulltime workers was conducted in 1985 just as the current surge of work intensification was beginning. It showed that the majority of workers were satisfied with their current hours. Of those who were not however, rather more would have preferred longer hours (and presumably more pay) than would have preferred a reduction in hours. The analysis presented above suggests that this question would have been answered differently in 1995, but this remains to be seen.
Similar issues arise with respect to part-time work. ABS surveys have shown that most women with part-time jobs are satisfied with their existing hours, but most men with part-time jobs would prefer to work longer hours. Members of this latter group might reasonably be defined as 'under-employed'. On the basis of the trends outlined above, we might expect an increase in part-time employment being sought by men and women unwilling to accept the increasingly intense demands of full-time jobs. To the extent that this shift is driven by work intensification it must be regarded as a second-best response to inefficient labor market organisation rather than as evidence of increasingly flexible patterns of employment.
The distribution of work and work intensity
It is not immediately clear how to interpret the increasing polarisation observed in cross-section studies of the labor force. One interpretation suggests an underlying polarisation of the population into a group of secure, fully employed people with access to superannuation, and another group moving between part-time employment, insecure and temporary full-time jobs and hidden or overt unemployment. An alternative interpretation involves a life-cycle view, in which many, perhaps most people, start out in the peripheral labour force, but ultimately enter the core labor market as prime-age workers. After a couple of decades of intensive full-time work, people aged over 50 become increasingly vulnerable to displacement from the core labor force.
The same kind of issue has arisen in connection with the debate about 'ghettos'; that is the increasing tendency for people of similar incomes to cluster together in residential areas. The census data that shows this tendency cannot tell us whether the same people remain concentrated in low-income areas, or whether people are simply moving more rapidly to accommodation consistent with their income. However, to the extent that studies of such distinct phenomena as residential location and working hours reveal increasing polarisation of low and high income groups, it seems more reasonable to suppose that we are observing a polarisation of the population rather than a life-cycle phenomenon.
Further evidence against the life-cycle view is given by the rapid increase in unemployment rates for workers in the prime age groups 25-54 over a period when unemployment rates for the workforce as a whole have remained largely unchanged. In some ways, a life cycle view seemed more appropriate in explaining the unemployment experience of the 1970s when a third of all unemployed workers were aged between 15 and 19,a nd presumably had never held a permanent fulltime. Unemployment could then be interpreted in terms of a problem of 'school to work transition', an increasingly difficult rite of passage to the world of permanent full-time employment. In the economy of the 1990s, permanency is an obsolete category, and holding on to a job once employed is almost as much a problem as getting one in the first place.
A detailed welfare-theoretic analysis of the labor market developments outlined above is beyond the scope of this paper. The policy proposals that follow are based on the view that the polarisation of the workforce is highly undesirable, and should be avoided, even if some aspects of the necessary response cannot be shown to meet standard efficiency criteria.
The first part of any response to the current situation must be serious attempt to achieve substantial and sustained reductions in unemployment. As argued in Langmore and Quiggin (1994), the core of such a program must be an expansion of employment in the community services sector. The argument that such an expansion is economically justified is presented in detail in Langmore and Quiggin (1994) and Quiggin (1995b). The fact that it is popularly demanded is regularly shown by public opinion polls.
This core program should be complemented by an active labor market along the general lines of Working Nation, but without the artificial constraints imposed by the attempt to implement such a program in a context of overall public sector contraction. In particular, the idea that program places should be 'extra work options' rather than 'real jobs' that would otherwise be funded by local governments and voluntary organisations should be abandoned. In the context of expanded provision of community services, the more 'real jobs' that are funded by employment programs the better. In addition to these core proposals, a number of employment-friendly measures for the private sector and the corporatised government business enterprises sector should be implemented, including the abolition of payroll taxes, careful scrutiny of other employment-based charges such as the Superannuation Guarantee Levy and the imposition of employment community service obligations on government business enterprises.
The second main part of a labour market policy program should be an attempt to maximise the extent to which working conditions meet the needs and desires of workers rather than the cultural norms of employers. Given equal bargaining power, such an outcome might be achieved by unconstrained individual contracting between employers and workers. Since equal bargaining power is no more than a fantasy, the problem is more difficult. Legislation guaranteeing rights to maternity leave and specific protection for minimum annual leave entitlements would be a step in the right direction. More generally, there is a strong case for turning back from enterprise bargaining towards a system in which awards play a central role, while admitting some scope for local variations to meet the needs of particular enterprises.
After a long period during which society became steadily more equal, indications of increasing inequality are becoming apparent. Whether there has been an actual increase in inequality or more a halt to the equalising tendency of the past is not yet clear. But the danger of social polarisation is a real one. This is nowhere more apparent than in the polarisation of working hours and work intensity. A renewed commitment to full employment and to the rights of workers to control their own lives should be the basis of a policy response to this danger.
Domberger, S., Meadowcroft, S. and Thompson, D. (1986), 'Competitive tendering and efficiency: The case of refuse collection', Fiscal Studies 7(4), 69-87.
Industry Commission, (1995), The growth and revenue implications of Hilmer and related reforms, AGPS, Canberra.
Langmore, J. and Quiggin, J. (1994), Work for All: Full Employment in the Nineties, Melbourne University Press, Carlton, Victoria.
Liebenstein, H. (1966), 'Allocative efficiency vs X-efficiency', American Economic Review 56, 392-45.
Quiggin, J., (1995), The growth consequences of Hilmer and related reforms, , Community and Public Sector Union.
Go back to John Quiggin's home page
Go back to John Quiggin's Conference papers