Go back to John Quiggin's home page

Feb 1995


Employment and unemployment in Australia - A long-term perspective


A Submission to the House of Representatives standing committee on long-term strategies

John Quiggin


Senior Fellow

Centre for Economic Policy Research,

Australian National University


Visiting Associate Professor

Department of Economics

James Cook University






The basic causes of unemployment may be found in changing patterns of labour demand and in the failure of society to respond appropriately to those changes. In particular, the growth of the community services sector has been constrained by limits on public spending.A national economic policy program for full employment would consist of the following major elements:

* a macro-economic policy oriented towards growth in employment and output;

* substantial expansion of community and public services employment eventually involving the creation of up to 350000 new jobs;

* public investment projects funded by issuing government bonds;

* expansion of labour market programs to provide a job guarantee for all long-term unemployed workers;

* an incomes policy based on the principles of the Accord;

* progressive tax reforms including a jobs levy and elimination of regressive tax concessions; and

* replacement of payroll tax by alternative forms of tax.


Employment and unemployment in Australia - A long-term perspective

In considering the future of work in Australia, it is natural to focus on our biggest social and economic problem -- unemployment. Along with John Langmore MHR I have set out an analysis of unemployment and a policy program to restore full employment in a recently published book Work for All.

We argue that the basic causes of unemployment may be found in changing patterns of labour demand and in the failure of society to respond appropriately to those changes. In particular, the growth of the community services sector has been constrained by limits on public spending.

This submission presents some of the key features of our argument. Changes in employment and unemployment are described with particular emphasis on the past twenty years. A theoretical analysis of the problem of declining labour demand and the role of the public sector is followed by the outline of a program to restore full employment.

1. Employment

Although unemployment has risen sharply over the past twenty years, this has not been the result of an absolute loss of jobs. Total employment fell during the recessions of 1974-5, 1981-3 and 1989-92, and employment has declined fairly steadily in sectors such as agriculture and manufacturing. However, in most years, employment in the economy as a whole has risen. Total employment rose at an annual rate of 1.5 per cent between 1973 and 1992, a total increase of nearly two million jobs. Unfortunately, this has not been enough to provide new jobs for a growing labour force. The labour force grew at an annual rate of two per cent, from under six million to 8.5 million over the period 1973-92, a total increase of 2.8 million. Thus, the economy generated little more than two-thirds of the necessary jobs over this period.

Figure 1 shows how the distribution of the labour force between sectors of the economy has changed over the period 1966-92. The most notable features are the rise in unemployment and the decline in the traditional areas of primary and secondary employment. In 1966, agriculture, mining and manufacturing employed thirty-five per cent of the work-force, of which manufacturing alone contributed twenty-four per cent. This proportion has fallen steadily. In 1992, these industries employed only eighteen per cent of the work-force. Total employment in these industries has dropped while the labour force has doubled in size. There are encouraging signs, such as the rapid growth in manufactured exports, that the decline in employment in these industries may be slowing. However, with industries such as textiles, clothing and motor vehicles still contracting, it is unlikely that the decline in manufacturing employment as a proportion of the labour force will be reversed or even halted.

Figure 1 Employment and Unemployment by Sector 1966-92

The infrastructure sector (transport, communications, electricity, gas and water) provides large numbers of jobs for manual workers and skilled trades, similar in general character to those in agriculture, mining and manufacturing. This sector grew almost as rapidly as the labour force over the period 1966-82. Since then, however, employment has contracted. The most important long-term factor has been labor-shedding by government business enterprises such as Telecom, the railways and electricity authorities.

Employment in wholesale and retail trade has generally kept pace with the growth of the labour force. Job growth in this sector has been sufficiently rapid to avoid making any net contribution to unemployment, but not rapid enough to offset, even partially, the decline in employment in primary and secondary industries and in the infrastructure sector. It seems unlikely that there will be any dramatic change in this area. Developments such as extended trading hours, which might lead to more employment, are likely to be offset by increased use of labor-saving technology.

The only industries in which employment has grown more rapidly than the labour force as a whole are in the service sector--financial services, community and public services and recreation and personal services. After keeping pace with the growth in the work-force through the sixties and seventies employment in recreation and personal services grew strongly in the eighties, which was primarily a reflection of growth in the tourism industry. This growth is likely to continue. According to some estimates, tourism may generate as many as 200 000 new jobs by the end of the century. Welcome though this is, it is only a small part of the required growth of nearly two million new jobs if full employment is to be restored, and it must be set against job losses in other sectors.

The major growth in employment during the sixties and seventies was in the community services sector, consisting of health, education and other services mainly provided or funded by government. Over the period 1966-74, growth in this sector was rapid enough to offset most of the decline in agriculture, mining and manufacturing. The Whitlam government's expansion of the Commonwealth government's contribution to education, and the provision of increased access to health services through Medibank were the most visible causes of expansion. Even conservative governments of this period responded to the public demand for better education and health services. Expansion of the community services sector continued, though at a slower rate, through the late seventies and early eighties.

With the adoption by the Commonwealth government of severely tightened fiscal policy in 1984, the financial basis for further growth was removed. The sector was particularly affected by the sharp cut in grants to the states and local government. Since state and local governments account for over three quarters of total public sector employment, the effects of this funding cut have been severe. The proportion of the work-force employed in the community services sector has remained almost unchanged since 1984. Even this result has only been achieved through a combination of increased state borrowings, reduced capital expenditure and heavier reliance on regressive and inefficient state taxes such as payroll and gambling taxes. Together, these expedients have permitted the states to hold their current expenditure (excluding transfer payments) roughly constant as a share of GDP.

The financial sector grew slightly over the sixties and seventies. But with the Fraser government's first moves towards financial deregulation in the late seventies a period of rapid growth began. Between 1978 and the peak of the asset market boom in 1989, the number of people employed in the finance sector doubled. Since the recession, job growth in the financial sector has stopped. There is little likelihood of strong renewed job growth in this sector in the future. Provided asset markets are organised in such a way that profits are available to well-informed or well-financed players, asset trade can continue whether or not it yields any social benefit. This helps to explain the fact that rapid growth in the financial sector has been associated with very weak productivity performance in the private sector as a whole.

The expansion of the financial system was assisted by the boom mentality of the eighties. Asset market operators such as Bond, Elliott, Skase and Holmes-a-Court were hailed as 'entrepreneurs'. Economic fundamentalists joined the chorus of praise with arguments showing that the 'market for takeovers' would improve the efficiency of corporate management. As the wealth of the entrepreneurs grew, at least on paper, it appeared that the financial sector did indeed have the key to a new prosperity.

In fact, the entrepreneurs had merely rediscovered the oldest fallacy of speculation. In a rising market, leverage pays. The quickest way to get rich is to borrow money to buy assets, use those assets to borrow more, and so on. Increasing demand drives prices up and creates more wealth. The fallacy lies in supposing that this growth can go on indefinitely. In every speculative boom, this fallacy is reinvented in a new form. Sometimes, as in the mining booms of the sixties and seventies, it is based on the notion that certain products are so special that their wealth-generating potential is unlimited. In the boom of the eighties the fallacy was based on the idea that financial deregulation would generate unlimited opportunities for increased efficiency.

It is now clear that the growth of the financial sector was unsustainable. Technological change has greatly increased productivity as measured by transactions per worker. Hence, financial sector employment can continue to grow only if the number of financial transactions rises rapidly relative to real product. This is unlikely at the household level. Indeed, the steady increase in transaction fees gives households an incentive to economise on transactions and this has been reflected in declining employment in the retail end of the banking industry. Transactions by business can only be cost-justified if they increase the efficiency with which final output is produced. There is no evidence that this has been the case. Indeed, an attempt to estimate the contribution of financial sector capital and labour to private business productivity suggests that employees in the financial sector, at best, make no contribution to final output and, at worst, have a negative impact. The prospects are therefore for future contraction in financial sector employment.

In summary, if the continuing contraction in employment in mining, manufacturing and agriculture is to be offset, the potential growth areas of community, recreation and personal services must grow more rapidly than they have over the past twenty years. Since many of these services are funded wholly or in large part by government, it is necessary to face the problem of funding the expansion of services.

2. Unemployment

The rise in unemployment has been the most dramatic economic phenomenon of the past twenty years, yet its causes have received little serious attention. It is first necessary to observe that the truly remarkable event is not the current period of high unemployment but the long period of full employment that preceded it. From the time of European settlement in Australia to the beginning of World War II, unemployment was an ever-present threat for working men and women. Although reliable measures of unemployment are hard to obtain for this period, rates of five to ten per cent were common. In the Depressions of the 1890s and 1930s unemployment was far worse. At the depth of the Great Depression of 1929-33, one in three trade union members was unemployed. Although there was some recovery after 1933, unemployment remained high throughout the thirties.

Unemployment fell rapidly as Australia mobilised for World War II. This in itself was nothing new. A boom in World War I had been followed by a severe recession as returned soldiers flooded the labour market. Unlike World War I, however, World War II was followed by a period of unparalleled prosperity and growth which lasted nearly thirty years. Australia entered the post-war world with an explicit commitment to the maintenance of full employment, embodied in the 1945 White Paper entitled Full Employment in Australia. The White Paper was the defining document of economic policy for the thirty years between 1945 and 1975. For the first time, the Australian government accepted an obligation to guarantee full employment and to intervene as necessary to implement that guarantee. During this period, unemployment virtually disappeared. Rates of unemployment below one per cent were common. The unemployment rate rose above two per cent only in isolated periods of recession.

Unemployment began to rise in the early seventies (see figure 3.2). The first big shock came with the credit squeeze of 1974 and the subsequent recession. Unemployment approached five per cent in 1975 and stayed around five to six per cent for the remainder of the decade. Another severe recession began in late 1981 and continued until 1983. When recovery commenced in 1983 the unemployment rate was over ten per cent.

After the election of the Labor Government in 1983, the combination of wage restraint and moderately growth-oriented economic policy, with favourable circumstances including the end of the drought and a recovery overseas, delivered rapid growth in employment. Nearly one and a half million jobs were created between 1983 and 1989. Despite this, the unemployment rate only fell to 5.7 per cent in 1989, so that, at the end of the long upswing of the eighties there were still nearly half a million people out of work--more than there had been when Australia entered the 1981-3 recession.

With the onset of recession in late 1989, unemployment rose rapidly. The unemployment rate peaked at 11.3 per cent (seasonally adjusted) in December 1992. Despite six quarters of recovery (defined as positive growth in national income) unemployment stayed high, exceeding one million in February 1993 and remaining at around eleven per cent throughout 1993. Strong economic recovery in 1994 saw a decline in unemployment, but the rate of recovery in the labour market has already slowed in response to recent increases in interest rates.

Figure 2 Unemployment in Australia 1966-94

The general pattern of rising unemployment is common to most countries in the OECD, and it would be a mistake to focus specifically on Australian causes of rising unemployment. The underlying problem has been a fall in the demand for labour throughout the Western world. This in turn may be traced to the failure to make an appropriate social adaptation to change. Because of technological and structural change, fewer workers are needed in the traditional employment areas of primary and secondary industry, and in the routine clerical areas of the service sector. The reduced needs for workers in traditional employment areas should be an opportunity for more people to be employed in meeting social needs for improved health, education, welfare, research, the arts and environmental services. However, because these services have been mainly provided, or at least financed, by the public sector, the necessary transfer can only be achieved through higher levels of taxation and public spending. Such measures have been vigorously resisted. The result is a recurrence of the paradox, observed by the 1945 White Paper, of mass unemployment side by side with unsatisfied social needs on a large scale.

A Hysteresis

The rate of unemployment has risen greatly over the past twenty years, however, the process has not been a steady one. Rather, each recession has seen a large increase in unemployment, but each subsequent recovery has yielded only a small reduction. As a result, the rate of unemployment has ratchetted upwards. The most popular ideas used to explain this process are associated with the term 'hysteresis'. Hysteresis originally referred to the lagged effects of exposure to magnetic fields, by which a metal that has previously been exposed to a field is more susceptible to subsequent magnetisation. The implicit assumption associated with the term 'hysteresis' is that episodes of high unemployment engender a greater likelihood of unemployment in the future. Once the level of unemployment has risen, the base level of unemployment appears to rise permanently.

As well as losing work skills, people who are unemployed tend to lose contact with the informal networks through which many, perhaps most, jobs are filled. People who are in work, or who have been unemployed for only a short time, often have friends, business contacts and former colleagues who know of job vacancies and are able to make recommendations on their behalf. Employers have a natural preference for hiring through networks of this kind (assuming their recommendations have worked out well in the past) rather than taking the risk of employing someone about whom little is known. The longer a person is unemployed, the more they lose contact with these networks, and the less willing employers are to take the risk of hiring them.

There are various explanations of hysteresis. The most plausible is the skill atrophy model, which stresses the loss of skills and demoralisation associated with unemployment. Therefore, the longer people are unemployed the less likely they are to regain employment at any given wage. The probability of finding a job is lower for the long-term unemployed than for those who have been unemployed for only a short time. Any policy that generates unemployment permanently reduces both the welfare of those who lose their jobs and the productive capacity of the economy as a whole. The skill atrophy model has direct implications for long-term unemployment. It implies that a rising trend in unemployment will be associated with increasingly severe long-term unemployment, as a growing proportion of unemployed workers are marginalised.

The tendency for the unemployment rate to ratchet up after each recession makes it imperative to avoid repeating the policy mistakes of the late eighties. Getting macro-economic policy wrong, by tightening fiscal and monetary policy too far, will make it impossible to achieve a return to full employment. But getting macro-economic policy right will not be enough. Even if steady growth in total output is maintained for the rest of this century, there is little hope of a return to full employment without a basic change in economic policy.

B Prospects for the nineties

What will happen to unemployment for the remainder of this century? No one knows for sure. However, if current policies are continued it is unlikely that the rate of unemployment will fall much below eight per cent.

One way of looking at the future is to make projections of the main variables that affect unemployment--population growth, economic growth, labour productivity and labour force participation rates, and derive their implications for rates of employment and unemployment. These variables are related by some very simple equations

% Employment growth rate = % Economic growth rate - % labour productivity growth rate

% Labour force growth rate = % Population growth rate + % Change in participation rate

% Change in unemployment = % labour force growth rate - % Employment growth rate

From these equations, it is possible to show that the rate of unemployment is determined by economic growth, population growth, productivity growth, and the participation rate.

The magnitude of the task confronting us is set out in a recent analysis of this kind undertaken for the Committee for the Economic Development of Australia (CEDA) by Fred Argy, former director of the Economic Planning Advisory Council (EPAC). Argy shows that if the trends of the last two decades were to continue, the labour force would grow at 2.1 per cent a year, while employment would grow at 1.8 per cent a year. This would imply that the unemployment rate would rise by about 0.3 percentage points each year, leading to a rate of nearly thirteen per cent by the turn of the century.

This is a fairly pessimistic scenario, since it does not allow for any cyclical recovery from the present recession. However, it should not be discounted. The adoption of a labour-shedding orientation by large private and public employers and the availability of labour-saving technological innovations not exploited in the eighties have allowed a recovery from the recession, as measured by GDP growth, without any significant recovery in employment and with rising levels of unemployment.

Under current policies it is unlikely that an average GDP growth rate above 4.5 per cent will be achieved over the next seven years. The generally restrictionist stance of macro-economic policy, and the inability of the current set of policy instruments to deal with balance of payments problems mean that any higher growth rate would be regarded as unsustainable. A rate of 4.5 per cent would still be significantly better than that achieved over the past twenty years.

Because of the general trend towards labour shedding and the likely contraction of the low productivity financial sector, it is likely that labour productivity will rise fairly rapidly, probably at around 2.5 per cent a year, as compared to 1.6 per cent over the last twenty years and around three per cent in the sixties. However, because of the increased substitution of capital for labour, total factor productivity will grow more slowly.

This implies a employment growth rate of two per cent a year, or more than one million new jobs over the next seven years. Although this sounds impressive, and would be hard to achieve under current policies, it would make very little impact on current levels of unemployment. It is likely that because of natural increase and immigration, the potential labour force will grow at a rate of about 1.2 per cent a year.

In addition, the participation rate is rising gradually over the long term, although it has declined in the current recession. The increased labour force participation of adult women more than offsets longer periods of schooling and earlier retirement. The increasing participation rate contributes long-term growth in the labour force of about 0.3 per cent a year. Hence, the average rate of growth in the labour force is likely to be around 1.5 per cent a year, leaving employment growth around 0.5 per cent a year, or a total of around 3.5 per cent over the next seven years.

At least a third of this growth is likely to be absorbed by discouraged workers and other hidden unemployed returning to the labour force after the present recession (this implies a cyclical increase in participation rates of about 0.7 per cent). Hence the net reduction in unemployment is likely to be around 2 to 2.5 percentage points. In the absence of a policy change, the rate of unemployment is likely to be above eight per cent for the rest of the century, even assuming very favourable macro-economic outcomes.

The implications for long-term unemployment are even grimmer. Increases in long-term unemployment tend to trail increases in unemployment. People who lose their jobs do not become long-term unemployed (by the current official definition) until a year later. Work done by Chapman, Junankar and Kapuscinski provides a basis for modelling the relationship between the general unemployment rate and the rate of long-term unemployment. On the basis of this work, the unemployment patterns projected above would imply that around fifty per cent of the unemployed, that is 300000 to 400000 will be long-term unemployed and that this pattern will persist for most of the remainder of this decade.

Of course, higher economic growth rates would imply more employment growth. Argy envisages a GDP growth rate averaging five per cent for the next seven years, and four per cent thereafter. He presents a number of policy changes aimed at making such growth rates achievable and sustainable. If achieved, this growth rate would imply a stable rate of unemployment around five per cent.

There is much to commend in the proposals made by Argy. However, a full employment strategy based solely on attempts to boost the rate of economic growth seems unrealistic. Even with macro-economic management greatly superior to that of the past few years, it will be impossible to avoid at least some years of slow growth, say, two years of two per cent growth out of the next seven years. To achieve an average economic growth rate of five per cent, it would be necessary to maintain growth rates over six per cent in normal years. In the absence of fundamental changes in policy, this would almost certainly imply a balance of payments crisis and a rapid reversion to restrictionist policies. Even the 4.5 per cent average growth scenario requires that growth in most years should be over five per cent.

3. Labour demand

Much evidence now suggests that the main underlying cause of unemployment is a slowdown in the demand for labour. Labour supply growth arising from rising population and a youthful demographic structure has been fairly steady over the post-war period. Increasing participation among adult women has been partly offset by declining participation of youth and men aged between fifty-five and sixty-five. The problem is that the demand for labour, after growing steadily in the fifties and sixties, has not risen sufficiently to absorb growth in the labour force since about 1973. The result is a rising unemployment trend.

Attention has naturally been focused on those areas where jobs are disappearing, and most notably on the manufacturing industry. Analyses of job losses in manufacturing have centred either on technological change or on competition from imports, particularly from the 'Newly Industrialising Countries' (NICs) of Asia. Seen in this light, declining labour demand is the product of inexorable external forces. Little more can be done than to ease the pain of adjustment.

But this defeatist analysis takes no account of the opportunities offered to society by technological change and international trade. Declining relative employment in manufacturing is the common experience of most mature industrialised economies. Manufacturing employment declined as a proportion of the workforce for most of the post-war period. Furthermore, technological change, at least as measured by total factor productivity growth, was faster in the fifties and sixties than at present. There is no reason to believe that technological change leads inevitably to unemployment.

The claim that competition from the NICs can explain today's unemployment is similarly unconvincing. In the fifties and sixties, Japan, with a population as large as that of the 'Four Tigers', (Hong Kong, Singapore, South Korea and Taiwan) combined, achieved growth rates as high as those of the NICs in the seventies and eighties. This did not prevent Australia and other developed countries from achieving strong economic growth and full employment.

Although there are reasons for supposing that the impact of technological change and global competition on wages and employment is likely to be more severe in the next decade than it has been in the past, it is a mistake to treat the declining demand for labour as the result of external or technological shocks. A central problem is, moreover, that whereas in the post-war period, the benefits of technological improvement were channelled into the expansion of much needed services, those benefits are now being squandered in the form of mass unemployment. This necessary reallocation of resources was achieved in the sixties when productivity growth was more rapid than at present. It is reasonable to propose a similar reallocation of resources in the nineties.

A Technological change and labour demand

The most striking technological development of the last fifteen years has been the rise of the personal computer. Desktop machines costing a few thousand dollars are now more powerful than the mainframe computers of the sixties and seventies, which cost millions of dollars and occupied entire rooms.

The initial introduction of personal computers, in the form of word processors, was accompanied by predictions of widespread job losses. It was suggested that the availability of easy editing of documents, in place of re-typing, would allow a single word processor operator to replace four or five typists. More generally, it was predicted that just as automation had reduced employment in manufacturing, the new computer technologies would reduce employment in clerical work and middle management.

These predictions were not fulfilled in the eighties. Studies conducted during the eighties failed to find any correlation between the adoption of personal computer technology and improvements in productivity. The lack of rapid productivity gains reflected attempts to introduce the new technology with no change in existing structures. Initial attempts by management to exploit the new technology focused on the ability to count keystrokes. Demands for higher keying rates led to an outbreak of repetitive strain injuries, and ultimately to a redesign of keyboard tasks that probably reduced throughput below the level associated with manual typewriters.

Word processors did not lead to one typist doing the work of four. But as they became easier to use, they made it possible, and appealing, for other workers to do their own typing. Resulting increases in potential productivity have taken some time to filter through the system. Productivity gains were held back in many organisations by the reluctance of managers to give up the executive status symbol of a personal secretary, or to be associated with the 'menial' activity of typing.

The later development of desktop publishing and database systems also failed to produce easily measurable productivity gains. The new systems were used primarily to perform new administrative and management tasks and to increase the quality of document presentation rather than to perform existing tasks with fewer workers. The proliferation of internal newsletters and annual reports, and their increasingly professional appearance, is a notable example.

The pattern of technological change leading to increased administrative activity, rather than higher productivity, was particularly evident in the financial services sector. Technological changes of the late seventies and eighties, such as the introduction of automatic teller machines and electronic trading of shares and currency greatly reduced the cost, and particularly the labour cost, of financial transactions.

However, this did not result in declining employment. On the contrary, during the eighties, the financial services sector took over from community services as the main growth area in employment. The combination of financial deregulation and the lower costs of financial transactions meant that it was highly profitable to discover and exploit minor divergences in the prices of different assets, and new trading opportunities were created by the relaxation of controls on takeovers.

The choice of increased administrative output, rather than reduced employment, reflected the introduction of the new technology into unchanged structures. Thurow (1993) describes how this process worked in relation to accounting:


The problem is seen graphically in accounting. During the period when accounting was being computerised, from 1978 to 1985, the number of accountants on American payrolls rose 30 per cent, from 1 million to 1.3 million, while output was rising only 16 per cent. Accounting productivity fell, despite the computerisation of accounting.

  Computers made accounting faster, but that speed was used, not to reduce the employment of accountants but to increase the frequency and types of accounting. Old accounts that in the past had been calculated every three months were now ordered up every day. Whole new systems of accounts (management information systems, cost accounting, inventory control, financial accounting, etc.) that had previously been impossible to calculate were put online. Yet there was no evidence that all of these new accounts improved decision making enough to justify their cost. In fact, given the huge increases in white-collar employment required to generate all of this new information, there was evidence to the contrary. But power and style called for ordering up all of those new accounts, and so it was done.

  To the boss, more information seems like a free good. He orders it from subordinates and the cost of acquiring it appears on the budgets of his subordinates. Subordinates can neither refuse to provide the requested information nor know if the information is valuable enough to justify the costs of its acquisition. To the subordinate costs are irrelevant. They are not even calculated. One does what one's boss orders.

Thurow's analysis is penetrating, but it does not tell the whole story. Another factor leading to the massive increase in accounting was the impact of financial deregulation and particularly the take-over boom. Before deregulation, accounting data was required primarily as an input to production and marketing and also to convey information to shareholders. In the eighties, it increasingly became a strategic tool as firms sought to identify take-over targets and protect themselves from takeover. Company activities were driven more and more by asset market considerations, and less and less by considerations of long-term productive efficiency.

Organisational structures and the deleterious effects of financial deregulation delayed the impact of new technology during the eighties, but it cannot be deferred indefinitely. The delayed impact of technological change is now being reflected in a surge of labour shedding, and rapid increases in labour productivity.

B Organisational structure

Until recently white-collar workers were largely insulated from the impact of technological and structural change. Although few Australian firms have an official commitment to notions of lifetime employment, and most have been willing to dismiss blue-collar workers in response to changing market conditions, lifetime employment has been an effective reality for many white-collar employees. Employers generally regarded white-collar employees as having a long-term stake in their companies. In addition, the total white-collar employment of most companies was growing over time. For these reasons, mixed perhaps with a little old-fashioned class bias, employers were reluctant to lay off white-collar staff in response to temporary economic difficulties. The same has been true until recently in the United States. In the 1981-2 recession, ninety per cent of the firms that laid off blue-collar workers did not lay off a single white-collar worker (Thurow, 1993).

This has all changed in the nineties and is partly the result of the gradual impact of technological change on organisational structures. More and more functions previously performed by central units, such as typing pools and data processing, have been taken over by individual sections. This in turn has reduced the value of a multi-layered organisational hierarchy. Small businesses have grown at the expense of large firms, and, among large firms, decentralised structures have been more successful than the traditional unitary hierarchical structure. A fundamental change in organisational structure is under way.

This process intensified after the onset of recession in 1989. Employers, forced to cut costs, realised that many clerical and management employees could be dispensed with permanently. Many companies were forced to cut back their administrative staff during the recession, and found that many of the tasks that had been added during the eighties were not really necessary. Meanwhile, the number of staff required to perform basic tasks such as payrolls and the management of employee records had declined greatly.

Although this process was initially driven by the need to cut costs, it was rapidly perceived as an opportunity for profit and promotion. In the period of growing white-collar employment, managers achieved success and prestige by increasing the number of workers under their control. 'Empire building' was the order of the day. The reverse is now the case. Managers are competing to identify areas in which employment can be reduced. Indeed, as Paul Keating (1993) has pointed out, labour shedding has become an end in itself, even when it does not serve the long-term interests of the organisation, let alone those of society as a whole. This change in corporate culture and organisational structure means that as economic activity picks up, employment in manufacturing, wholesaling and retailing will not grow nearly as fast because of the permanent changes which have occurred.

Organisational change has also intensified the rate of blue-collar job loss. A major element of micro-economic reform has been the breaking of union resistance to compulsory redundancies, even in highly profitable firms. At the same time, management innovations, such lean production methods learnt from Japan, 'just-in-time' inventory control and quality circles have accentuated the trend towards smaller organisational work-forces and less labour-intensive methods of production. Organisational style areas using these innovations has altered permanently, so that at best only modest employment growth can be expected in these areas as normal economic activity is restored.

C International factors

During a period in which Australian economic policy has been dominated by fundamentalism, economic performance has been notably poor. Australian economic fundamentalists have often used movements in the world economy as an alibi, arguing that their policies would have worked much better if it were not for shocks such as the international recession. This alibi ignores the fact that the recession itself has been caused in large measure by the fact that other countries have pursued the same policies as Australia--financial deregulation, macro-economic contraction (most notably in Germany and the United Kingdom) and public sector cutbacks. More fundamentally, changes including international financial deregulation have produced an international economic environment biased in favour of contractionary, high unemployment policies.

Countries in East and South-East Asia have enjoyed rapid economic growth and low levels of unemployment. The most dramatic performances have been those of the Four Tigers. These countries have enjoyed rates of growth of per capita GDP around ten per cent, with Singapore and Hong Kong now approaching the income levels of the OECD countries. Less dramatic, but still substantial growth rates have been achieved by Thailand, Malaysia and Indonesia.

It is natural to ask whether the good performance of these countries is linked to the economic problems of the Western countries. In a simplistic form, this question is posed in terms of industries 'moving off shore' to take advantage of cheap labour in Asia. A more sophisticated version of the same argument is developed in orthodox trade theory as the 'factor price equalisation theorem'. This result states that under free trade, workers with the same skill levels will earn the same wages wherever in the world they are located. The most obvious implication is that the wages of unskilled labour in Australia must ultimately be driven down to the levels prevailing in Asia. Skilled labour can continue to earn a premium but only as long as the required skills cannot be hired more cheaply elsewhere in the world.

An obvious difficulty with this argument is the question of why the factor price equalisation theorem should be relevant now, when large disparities between more and less developed countries have been maintained for the last hundred years. The rise of the Four Tigers alone is not a sufficient explanation. Their combined population is only about seventy million, and their combined GDP not much greater than that of Australia. Such a small addition to the labour force of the developed world (considerably less than that arising from natural increase and migration over the past ten years) could scarcely have much impact on wages. Even adding the remaining NICs into the picture does not change this argument. Although their combined population is nearly 250 million, employment is still predominantly agricultural, and their total manufacturing output is smaller than that of Australia.

Given the experience of the fifties and sixties, there would be no reason to suppose that the rise of the NICs should depress wages in the more developed countries. During that period wages in the rich countries rose rapidly, even though countries such as Japan (and to a lesser extent the southern European countries) were greatly increasing their output of manufactures and other traded goods.

If factor price equalisation is taking place now, it is the result of the policy changes of the seventies and eighties, most notably the deregulation of international capital flows. This has made it more difficult for any one country to maintain high levels of real wages for given skills. Any policies that are seen as redistributing income from capital to labour are met with capital flight and currency depreciation. In addition, multinational firms can play potential host countries off against each other to obtain the best possible deal in terms of taxation treatment and wages. The regular (though apparently inaccurate) claim that large amounts of capital are leaving Australia for New Zealand in response to the New Zealand labour market reforms are an illustration of this process.

Factor price equalisation might really begin to push wages down if China and India were added to the equation. Both of these countries were formerly isolated from the world economy. China under Mao was almost totally cut off from trade. India pursued an import replacement strategy which greatly restricted trade. Both countries, but particularly China, have undergone a re-orientation towards trade and exports. Their combined population of around two billion is twice that of the OECD. However, their impact on world markets so far has been small.

While competition from Asia is most significant from an Australian perspective, and probably for the world as a whole, local developments will affect other Western countries. Labour in the United States will be subject to competition from Latin America. The most direct manifestation of this is the North American Free Trade Agreement (NAFTA) which creates a free trade zone embracing Mexico, Canada and the United States. In addition, direct, mostly illegal, immigration has sustained a low wage sector within the US economy.

In Europe, the most direct competition will come from the formerly Communist countries of Eastern Europe. These countries have large numbers of skilled workers and economic policies aimed at the achievement of international competitiveness. In addition to the effects of import competition, many workers are eager to migrate to Western Europe. Despite increasing barriers to migration, it is likely that the flow of workers from East to West will be substantial.

Factor price equalisation cannot be more than a partial explanation of the growth in OECD unemployment and the general stagnation of wages since 1973. Nevertheless, the globalisation of the world economy and the entry of China and India into the world market is likely to put further downward pressure on wages.

Like technological change, the international economy provides both threats and opportunities. The contraction of traditional manufacturing industries will result in the loss of jobs, concentrated in manufacturing regions of the southeastern states. The general tendency towards higher levels of unemployment and depressed economic activity in the OECD countries creates great difficulties for any economic strategy based on exports.

On the other hand there are opportunities. Many of the jobs that have disappeared from the manufacturing sector were dirty, unskilled and dangerous. With a commitment to full employment these jobs could be replaced with work that is both more attractive and more socially useful. With appropriate industry policies, a smaller, more skill-oriented manufacturing sector could continue to play a significant role in the economy. As well, the lower cost of manufactured goods could release resources for expansion of the services sector. There are at least some rays of hope here. In a number of areas, industry policy has been employed successfully. Notable examples include the steel and motor vehicle industries. Exports of manufactures have increased, particularly for those exports characterised by high added value.

4. Fiscal crisis and the community services sector

The experience of the sixties and early seventies would have suggested that the main area of employment growth in the last part of the century should have been that of community services. The demand for services such as education and health, police and urban amenities, environmental protection and the arts tends to rise with income, and to increase as a proportion of total demand as income rises. Furthermore these services are highly labour-intensive and have been relatively little affected by technological change. For instance, despite massive changes in the content of education, and in educational philosophy, the average classroom is little different now from that of fifty or a hundred years ago. Efforts to improve the quality of education have generally focused on increasing the ratio of staff to students. Technological developments are unlikely to change this in the near future.3

In many respects this is also true of health care. Although there have been changes in equipment, these have generally not been labour-saving in nature. Instead the introduction of more sophisticated equipment and new medical procedures has generally required an increase in the employment of skilled workers.

Labour-saving technological change has been similarly limited in other parts of the community services sector, such as police services, environmental protection and the arts. This phenomenon was analysed in the sixties by the leading American economist, William Baumol. He argued that if productivity grew more slowly in the services sector than in other sectors such as manufacturing, and it was desired to maintain output in the services sector at least as a constant proportion of total output, it was necessary that resources should be progressively transferred towards the services sector.

Until the late seventies, Baumol's analysis was borne out in Australia. During the period 1966-80, employment in mining, agriculture and manufacturing fell from thirty-five per cent of the work-force to twenty-six per cent. Although this decline was associated with an increase in the rate of unemployment from rates of one or two per cent in the sixties to around five per cent in the late seventies, it was largely offset by an increase in the proportion of the workforce employed in community and public services from thirteen to nineteen per cent of the work-force. Other sectors of the economy remained broadly stable.

The decline in employment in primary and secondary industry continued steadily during the eighties, from twenty-six per cent to the current level of eighteen per cent. But employment in community and public services grew very slowly, rising only from nineteen per cent to twenty-one per cent. Continuation of the previous trend would have implied that about twenty-five per cent of the labour force should be employed in community and public services today and perhaps thirty per cent by the year 2000. If such a trend had continued, the outlook for unemployment would be considerably less gloomy.

A similar picture is obtained from data on expenditures. Australian general government outlays (Commonwealth, State and local) rose from thirty-three per cent of GDP in the early seventies to 42.5 per cent in 1985-6 before falling back to 39.6 per cent in 1990-91. However, outlays in the main areas of public service goods provision (defence, education, health, housing, energy supply, transport and communications) actually fell from the seventies onwards, declining from 18.6 per cent of GDP in the early seventies to 18.3 per cent in 1990-91. The rise in outlays was concentrated in increased transfer payments (largely due to the rise in unemployment) and increased public debt interest repayments.

The slowdown in public sector employment during the eighties reflected the 'crisis of the state' which afflicted many Western countries in the middle and late seventies. This involved an inability to maintain the upward trend in revenue as a proportion of GDP. Since GDP growth was also slowing, the rate of growth of revenue declined. Initially, this led to a period of large budget deficits. However, capital markets were unwilling to absorb these deficits. There was also a perception of a rising burden of debt, which was inaccurate in Australia, at least as far as public debt is concerned. These factors led to a decline in expenditure growth. Although transfer payments had constituted the largest area of increased expenditure, they proved the most difficult to constrain. Decisions taken during the post-war boom implied an expansion of 'entitlements', particularly for the elderly. Rising unemployment and, in many countries, rising rates of family breakdown implied an increased dependency ratio. Yet Australian governments have taken more stringent measures than those in most other OECD countries to wind back the growth of transfers.

5. Responses

The task of restoring full employment in Australia by the end of the century is a daunting one. Under current policies, it is likely that nearly a million people would still be unemployed in the year 2000. Of these, as many as 400 000 could be long-term unemployed, under the official criterion of at least twelve months of continuous overt unemployment. If full employment is defined as a three per cent rate of overt unemployment, these projections suggest a need to create at least 700 000 extra new jobs (over and above the 1.2 million jobs estimated to be created under existing policies). If allowance is made for increased labour force participation, even more rapid employment growth would be necessary.

The aim of this and later chapters is to present a set of policies aimed at maximising growth in employment. Although the future cannot be forecast with precision, these policies could, under favourable conditions, lead to a restoration of something approaching full employment. In view of the natural tendency to dismiss the idea of full employment as Utopian, it is important to put forward specific proposals to achieve this goal.

A national economic policy program for full employment would consist of the following major elements:

* a macro-economic policy oriented towards growth in employment and output;

* substantial expansion of community and public services employment eventually involving the creation of up to 350000 new jobs;

* public investment projects funded by issuing government bonds;

* expansion of labour market programs to provide a job guarantee for all long-term unemployed workers;

* an incomes policy based on the principles of the Accord;

* progressive tax reforms including a jobs levy and elimination of regressive tax concessions; and

* replacement of payroll tax by alternative forms of tax.

The scale of this program would be large indeed. The total gross expenditure involved would eventually be up to $15 billion a year, although a significant part of this spending would be recouped through increased revenue resulting from growth in employment and output and reduced outlays on job search allowance and related benefits. However, it should be remembered that the budgetary cost of unemployment is more than $20 billion a year, and that the full economic cost is much larger still.

Further, the programs proposed would yield great and enduring social and economic benefits--easier access to higher quality education and health care, safer and cleaner cities and more accessible and efficient transport and communications systems--in addition to the reductions in unemployment they would generate. The proposals would be financed through a combination of new and more progressive taxes, borrowing and reorientation of existing public spending.

These proposals are not put forward in a dogmatic spirit. Some elements of the program suggested here may prove impractical and there may be superior options which have not been considered in this book. However, this strategy gives some idea of the scale of the necessary response.


Unemployment is the greatest problem facing Australian society. The present recession is the product of contractionary macro-economic policies and the collapse of the asset price boom generated by financial deregulation. However, the underlying causes of rising unemployment may be traced to a declining demand for labour. Technological, organisational and structural changes have reduced employment in agriculture, mining and manufacturing and, more recently, in routine clerical and administrative areas of work.

Growth alone will not be enough to eliminate unemployment. It is necessary to reorient the economy so that resources are allocated to those areas with a strong, and potentially growing, demand for labour. This means renewed growth in the provision of community services in place of the current policy of contraction.

Society still has great unfulfilled needs for productive labour in areas such as education, and research, health and welfare, the arts and protection of the environment. The problem is that, because these sectors are largely publicly financed, existing economic structures have failed to allocate resources where they are needed. Instead, community services have been cut back in response to imagined fiscal problems and the priority given to cutting taxes. This has contributed to further unemployment, creating a fiscal problem and weakening demand for private sector goods and services.


Go back to John Quiggin's home page