Unemployment: still hoping for a miracle?

John Quiggin

Australian Research Council Senior Fellow

Quiggin, J. (2000), Unemployment: Still hoping for a miracle?, International Journal of Manpower 21(5), 374-83.

 

 

 

 

 

Abstract

The Australian unemployment rate has fallen to its lowest level since 1989, and looks likely to fall further, perhaps even reaching the 5 per cent target set in the 1994 White Paper. However, this relatively favourable economy has been achieved only after a cyclical expansion so long and robust that it has been widely regarded as ‘miraculous’. It is important, therefore to consider whether recent reductions in unemployment will be maintained when the current expansion ends, and whether alternative policies could produce stronger and more sustainable growth in employment.

Keywords

Unemployment, multifactor productivity, human capital, public sector

JEL code

E2,J6

 

 

Unemployment: still hoping for a miracle?

The Australian unemployment rate has fallen to 7 per cent, its lowest level since 1989, and looks likely to fall further, perhaps even reaching the 5 per cent target set by the then Labor government in the White Paper, Working Nation (Commonwealth of Australia 1994). In itself, there is nothing remarkable in this achievement. Unemployment fell as low as 5.9 per cent in the expansion of the 1980s before jumping to 11 per cent in the subsequent recession. There is widespread optimism that the microeconomic reforms of the past fifteen years have laid the groundwork for sustainable growth, but similar optimism was also widespread in the late 1980s.

There are however, important differences between the current expansion and that of the 1980s. First, a number of countries have achieved unemployment rates below 5 per cent and there are some grounds for supposing that this achievement will be sustainable. Second, inflation rates have been very low throughout the decade, and inflation appears likely to remain quiescent. Finally, although the current account deficit is about as large, relative to GDP, as during the late 1980s, the advocates of the ‘new view’, that the use of restrictive monetary policy to control the current account deficit is inappropriate, have prevailed in the policy debate.

The main reason for pessimism is the fact that unemployment is at the bottom of the policy agenda, just as it has been for all but a few years in the past two decades. In the absence of a serious policy response, we are reduced to ‘hoping for a miracle’. Only if the Australian economy continues to grow for a period unparalleled even during the postwar long boom can unemployment be reduced to rates below 5 per cent and kept there.

The paper is organised as follows. An assessment of the outlook for employment and unemployment in Australia and the recent claim that Australia has produced a ‘miracle economy’ shows that the critical issues are the current account deficit and the implications of the return of price stability. A survey of international experience is used to assist understanding of these issues. The final section deals with a range of policy options to achieve and sustain a return to full employment.

The outlook for employment and unemployment in Australia

The dynamics of employment and unemployment

The dynamics of employment and unemployment can be described very well by some simple identities. The following formulation is taken from Quiggin (1993, 1994).

(1) Employment growth rate = Output growth rate — labour productivity growth rate

(2) Labour force growth rate = Working age population growth rate + Change in participation rate

(3) Change in unemployment rate = labour force growth rate — employment growth rate

With the exception of the economic growth rate, the exogenous variables in this system of equations are fairly predictable, at least in the absence of substantial changes in policy. The rate of growth of labour productivity has generally been around 1.5 per cnet per year. (The claim that productivity growth has accelerated recently will be discussed later.) The underlying rate of growth of the labour force is about 1.5 per cent per year, but during periods of expansion, this is augmented by growth in the participation rate. On average growth in participation absorbs about half the growth in employment. On the basis of these assumptions, it can be estimated that 4 per cent annual output growth will produce annual employment growth of 2.5 per cent, of which 1.5 percentage points will be absorbed by population growth and 0.5 percentage points by participation rate effects, yielding a reduction in unemployment of 0.5 percentage points per year.

Quiggin (1994) observed:

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 8 per cent.

This analysis was based on the assessment that the average rate of economic growth over the 1990s was unlikely to exceed 4 per cent, and would probably be reduced even further by a slowdown or recession at some point in the decade. In fact, even though the economy experienced one of the longest expansions on record, the prediction was quite close to the mark, with unemployment remaining above 8 per cent until late 1998. The estimated rate for September 1999 was 7.2 per cent.

Australia as a 'miracle economy'

The analysis of the previous section is based on the assumption that the behavior of the economy is unlikely to change fundamentally. An alternative view is that fundamental change has already taken place, to the extent that Australia is now a ‘miracle economy’. This view is put forward by Parham (1999), using estimates of partial and multi-factor productivity generated by the Australian Bureau of Statistics (ABS).

The key results of the analysis are presented in Table 1

Period

MFP

Output

Inputs

Capital-Labour

1964-65 to 1968-69

1.3

5.1

3.8

4.5

1968-69 to 1973-74

1.6

4.5

2.9

4.5

1973-74 to 1981-82

1.3

2.0

0.7

4.3

1981-82 to 1984-85

1.2

1.8

0.6

4.2

1984-85 to 1988-89

0.8

4.4

3.6

0.7

1988-89 to 1993-94

1.1

1.7

0.6

2.8

1993-94 to 1997-98

2.4

4.6

2.2

2.4

Thus, the rate of multifactor productivity growth in the market sector of the economy for the period since 1993-94 is estimated at 2.4 per cent, compared to 1.3 per cent for the period from 1965-65 to 1968-69, at the peak of the long boom.

There are, however, some important difficulties with this analysis. The most important relates to the dating of ‘productivity cycles’. The ABS analysis, based on the time-series properties of the productivity estimates themselves, identifies seven such cycles in the period since 1964-65, whereas a conventional business cycle analysis would identify only four. In particular, the business cycles of the 1980s and 1990s are divided into two separate productivity cycles ( There was a brief slowdown around 1986 which might possibly justify a breakup of the 1980s into two business cycles, but there are no grounds for supposing that the Australian business cycle reached a cyclical peak in 1993-94. )

Table 2 shows the effect of a cyclical breakdown based on business cycles rather than peaks and troughs in productivity estimates.

Period

MFP

Output

Inputs

Capital-Labour

1964-65 to 1973-74

1.5

4.8

3.3

4.5

1973-74 to 1981-82

1.3

2.0

0.7

4.3

1981-82 to 1988-89

1.0

3.2

2.2

2.2

1988-89 to 1997-98

1.7

3.0

1.3

2.6

 

An analysis based on two cycles over the 1980s and 1990s is more consistent with economic theory. Standard production theory implies that an increase in multifactor productivity would lead to an increase in input use since the rate of return to investment would increase as would the marginal product of labour. With the cyclical breakup in Table 1, there appears to be a negative rather than a positive correlation between productivity and input demand. In particular, input demand is shown to be at its highest point since the long boom for the cycle from 1984-85 to 1988-89, when productivity growth is at its weakest. More generally, with the cyclical breakup in Table 1 input demand fluctuates sharply over the four cycles since 1981-82. The obvious explanation is that the cyclical breakup is incorrect.

Another disturbing feature of the analysis is the estimate that the labour input for the market sector has grown by only 1.4 per cent, whereas employment for the economy as a whole has risen by more than 2 per cent per year over the same period. The discrepancy arises primarily because of rapid growth in hours worked in the property and business services industry, which is not classified as part of the market sector (Hours worked in this industry grew by nearly 40 per cent in only four years.). The growth in the property and business services industry appears to arise primarily from contracting out of activities formerly undertaken within the market sector. Hence, a more appropriate analysis of productivity growth would treat this industry as part of the market sector. The effect would be to eliminate most of the apparent improvement in the rate of multifactor productivity growth.

Finally, the ABS data which is based on the Labour Force Survey, does not take any account of the increase in work intensity that has taken place over the last decade. Work intensity is difficult to measure, but it can be defined fairly simply in economic terms. If workers would be indifferent between working 40 hours per week under one set of conditions and working 44 hours per week under a second set of conditions, then the intensity of work is 10 per cent greater in the first case.

Work intensity may arise because of a faster pace of work (as with the Taylorism of the early 20th century), because of the elimination of formal or informal breaks (which would usually be counted as work time in the Labour Force Survey) or because of increases in stress. If the increase in work intensity over the past decade is taken to be the equivalent of a 10 per cent increase in working hours, the apparent improvement in productivity growth disappears and is replaced by a further deterioration.

Comparison of the last two business cycles

The rhetoric of the miracle economy suggests that the present business cycle is radically different from the last one. As has been argued above, the real differences are considerably more modest. The rate of output growth over the (incomplete) business cycle of the 1990s (starting at the cyclical peak of 1988-89) is quite similar to that of the 1980sbut the rate of employment growth has been slower in the 1990s. Hence, after eight years of expansion, the unemployment rate has yet to fall below 7 per cent, compared with the rate of 5.9 per cent achieved at the peak of the seven-year expansion in the 1980s.

The relatively weak employment growth of the present cycle has two main causes. First, the increase in working hours and work intensity for both full-time and part-time workers has reduced the employment growth associated with any given increase in output in all industries. evidence

Second, economic policy has been oriented towards cutting back the relatively employment-intensive health, education and community services sectors. Thus, changes in the industry mix have been less favourable to employment than in previous cycles.

If the present economic expansion is more promising that of the 1980s, it is because it seems likely to last longer. The view that the current expansion is more robust than that of the 1980s rests in part on the fact that the economy was unaffected by the Asian economic crisis. However, the economy also appeared to shrug off the stock-market crash of 1987, only to enter a severe recession two years later.

A second argument is that microeconomic reform has rendered the economy more flexible. Again, this view was widely held in the late 1980s. As then Treasury Secretary, Chris Higgins, observed in 1989,

We have had a decade of remarkable and fundamental economic and social policy reform; reform which in all its major contours and, arguably, in 99 per cent of its detail, is efficiency-enhancing.

The success of the economy in weathering the ‘banana republic’ crisis of 1986 was widely seen as heralding the emergence of a new more flexible economy. In fact, as the recession of 1989-91 showed, the economy was as vulnerable as ever to misjudgements in fiscal and monetary policy. The fact that Australia was unaffected by the Asian crisis while New Zealand entered yet another recession illustrates the same point. Australia's superior performance was the result of more skilful (or perhaps more fortunate) management of monetary policy and the exchange rate, not of better implementation of microeconomic reform.

There are, however, three important differences between the current cycle and that of the 1980s. First, inflation is a less serious problem, both in reality and in the minds of policymakers. Second, although in objective terms the current account is at much the same level as in the 1980s, it is less important to policymakers and therefore less likely to induce the adoption of restrictive policies. Finally, there are some signs of an international trend towards lower rates of unemployment, in contrast to the sharp rise in the 1970s, and the gradual upwards drift in the 1980s.

The return of price stability

To determine the implications of low, and apparently stable, inflation rates for policies aimed at the restoration of full employment, it is necessary to examine the relationship between inflation and unemployment. Analysis of this relationship has been based on two competing ideas - the Phillips curve and the ‘natural rate of unemployment’ .

The macroeconomic policy framework of the 1950s and 1960s was based on the stated objective of stabilising the economy at its full employment level, which was assumed to be consistent with price stability. In practice, full employment in Australia was interpreted to mean an unemployment rate between 1 and 2 per cent and price stability was interpreted to mean mild price inflation, at annual rates of around 2 per cent. Unemployment rates above 2 per cent were interpreted as evidence of inadequate demand, requiring the adoption of stimulatory fiscal and monetary policy, while inflation rates above 2 per cent were interpreted as evidence of excess demand requiring the adoption of contractionary policies. This approach was represented in terms of a control-theoretic framework in which policy instruments were used to minimise deviations from the full-employment target.

The discovery of the Phillips curve (Phillips 1958) was initially seen as providing an empirical verification of the theory underlying this approach to demand management. However, the natural tendency of economists to apply their microeconomic training led to a subtly different (and ultimately disastrous) interpretation, sometimes referred to as the ‘policy menu’. The idea of the policy menu was that governments could pick any point on the Phillips curve, thereby trading off inflation against unemployment. With the choice framed in this way, it was inevitable that the selected trade-off would involve tolerating higher inflation in the hope of permanently reducing unemployment.

This idea was criticised by Friedman (1968) who argued that this approach would work only until inflationary expectations adjusted to the new higher rate, at which time unemployment would return to the ‘natural rate’ consistent with equilibrium in the labour market. The ‘stagflation’ of the 1970s, in which high rates of unemployment were combined with accelerating inflation confirmed Friedman's prediction and were widely interpreted to mean that, in the long run, there is no trade-off between unemployment and inflation, so that the long-run Phillips curve is a vertical line at the ‘natural rate’.

Friedman's argument was never fully accepted. In particular, estimates of the natural rate showed a disturbing tendency to track movements in the actual unemployment rate. The dependence of current unemployment on past rates of unemployment, referred to as hysteresis (Blanchard and Summers 1987), casts doubt on the usefulness of the concept of a natural rate of unemployment. Nonetheless, during the 1970s and 1980s, when most developed countries experienced inflation rates consistently above 5 per cent, there was little evidence of a stable trade-off between unemployment and inflation.

With the return of the low inflation rates that gave rise to the original Phillips curve, the trade-off between the inflation and unemployment objectives is evident once again. There is a clear association between deflation and depression. If the empirical evidence on this point is accepted, the idea of a completely vertical Phillips curve must be discarded. If the long-run Phillips curve is negatively sloped for negative rates of inflation, it is not reasonable to suppose that it becomes vertical as soon as the inflation rate becomes positive.

Even in the absence of any long-term reduction in unemployment arising from higher inflation, there may be a trade-off between stability of real output and stability of the inflation rate. This idea is supported by the much greater output variability experienced in New Zealand, which has a very tightly specified inflation target, compared to Australia, which tolerates larger variations in the inflation rate (Bean 1999). It seems likely that higher variability of output will correspond both to more variable unemployment and to a higher average unemployment rate.

The current account and the sustainability of growth

The long economic expansion in Australia since 1992 has been accompanied by growth in the current account deficit, expressed as a proportion of GDP. A similar pattern has emerged in the United States, while New Zealand has experienced even larger current account deficits, but without any sustained growth.

Before the move to floating currencies, large current account deficits were generally thought to be unsustainable and therefore to require a policy response, even in the absence of any other macroeconomic imbalance. This ‘old view’ that current account deficits may be so large as to be unsustainable is still widely held, and dominated the policy debate in the late 1980s, the last time at which such large deficits were experienced.

The alternative ‘new view’ is that, in a deregulated environment, movements on the capital account represent voluntary transactions between foreign lenders or investors and Australian borrowers. Since the current account deficit is merely the obverse of the capital account surplus, determined by the total volume of these voluntary transactions, it is argued that a large current account deficit is not a policy problem and that the deficit should not be a target of policy (Pitchford 1992).

Before considering the issues, it is worthwhile to examine what it means to say that the current account deficit is, or is not, sustainable. One way of assessing the sustainability of a given level of the current account is to examine the effect of the deficit on foreign debt and equity holdings. At present, the net foreign debt is equal to about 40 per cent of GDP, a level which has remained broadly stable over the past decade, after rising rapidly during the 1980s. Estimates of the value of foreign holdings of equity are harder to make, but it seems likely that this value is also close to 40 per cent of GDP, so that the value of total obligations is around 80 per cent of GDP.

Suppose that the trend rate of growth of nominal GDP is 5 per cent (3.5 per cent real growth and 1.5 per cent inflation). Then nominal foreign indebtedness must grow at the same rate if the ratio of foreign debt to GDP is to remain stable. If unrealised capital gains on foreign equity are disregarded, the current account deficit is equal to the growth in nominal foreign indebtedness. Hence, stability will be maintained if the current account deficit is equal to 5 per cent of foreign indebtedness or 4 per cent of GDP. Conversely, if the current account deficit remains at 6 per cent of GDP, and nominal growth at 5 per cent, foreign indebtedness will be stable at 120 per cent of GDP.

Recent international experience suggests that the model implicit in the new view, in which foreign exchange balances are nothing more than the aggregate of individually rational transactions, is flawed. International capital markets are dominated by sentiment about particular national economies, and countries with high levels of debt are particularly vulnerable to reversals of sentiment.

This does not necessarily imply that it is appropriate to react to a current account deficit by tightening monetary policy, as was done in the late 1980s with disastrous effects. It does, however, imply that large current account deficits should be a signal for a tightening of prudential regulation of the financial system, to reduce vulnerability to sudden changes in market sentiment.

Even if the current account deficit is not regarded as a target for policy, it is useful as an indicator that output is above the level that is sustainable in the medium term, and hence as a leading indicator of a cyclical downturn. In this context, it is worrying that the current account deficit has already reached levels suggesting a likely future downturn in economic activity at a time when the unemployment rate is still well above the low point of 5.9 per cent reached in the 1980-89 cycle. Although the present cycle may show some improvement, the level of the current account deficit suggests that it will be difficult to achieve and sustain an unemployment rate below 5 per cent.

International experience

Unemployment rates in developed countries rose rapidly during the 1970s. During the 1980s, a pattern of hysteresis emerged with unemployment rising rapidly in periods of economic crisis, then declining gradually when economic growth resumed, but remaining at a higher level than during the previous cycle. The upward ratcheting of unemployment took place against a background of gradually declining inflation over the 1980s. For most of the 1990s, there was no clear trend in unemployment, with rates rising in some countries and falling in others.

Assessment of underlying trends in the 1990s was complicated by the shock to the German economy arising from the absorption of the former East Germany into the Federal Republic. Because of Germany's central role in the European Monetary System, this shock had a deflationary effect on all countries. This deflationary effect was reinforced, for much of the decade by the requirement to satisfy the Maastricht conditions for participation in the European currency union which came into effect in 1999.

Over the last few years, however, a general downward trend in unemployment has become apparent. The declining trend first became evident in the English-speaking countries and a number of smaller European countries, notably Austria, Denmark, Norway and the Netherlands. Austria and Norway managed to maintain near-full employment throughout the long slowdown, while the Netherlands and Denmark have succeeded in reducing unemployment rates to less than 5 per cent on an OECD standardised basis. More recently, unemployment has declined in the majority of European countries, including France and Germany. In addition, Ireland's remarkable growth performance has finally been reflected in lower unemployment. The only important exceptions to this pattern have been Japan, where the economic stagnation of the 1990s has produced large scale unemployment for the first since World War II and New Zealand, where the strong employment growth of the early 1990s has not been sustained.

In general, the reduction in measured unemployment overstates the improvement in labour market conditions. A variety of forms of hidden unemployment have proved less tractable than overt unemployment. The forms of hidden unemployment vary greatly with institutional arrangements, but they include premature retirement (particularly important in Austria), reliance on sickness benefits (the Netherlands) and incarceration (the United States).

International comparisons of unemployment rates, particularly English-speaking countries have tended to focus on a comparison between the strong performance of the United States, on the one hand, and the poor performance of France and Germany, on the other. The other English-speaking countries are treated as intermediate cases, in the process of moving from European-style sclerosis to American-style free markets.

There are a number of difficulties with this account. First, the neat correlation between the degree of market flexibility and the strength of economic performance breaks down when the smaller European countries are taken into account.

Second, the pitfalls of making comparisons between countries at different stages of the economic cycle are particularly acute in the present cases. Most of the English-speaking countries experienced recession around 1990 and a recovery which has been maintained (with a few interruptions) for most of the 1990s. In particular, the United States is experiencing what is, on all past experience, an unsustainable boom. By contrast, German political union and the convergence measures required for European currency union led to an unusually sustained slump during the mid-1990s, from which the main European economies are only now emerging. A valid comparison of the European and US models can only be made at the end of the current cycle.

Finally, there is the problem of current account deficits and sustainability. Most of the English-speaking countries are running large current account deficits, and most have a history of ‘stop—go’ economic growth, with periods of strong growth being halted by foreign exchange crises. In part, the large current account deficits of the English-speaking countries reflect . However, there also appears to be an increase in the long-run average level of the deficit, at least for the United States, New Zealand and Australia. In the latter two cases, the emergence of large structural deficits was clearly associated with financial deregulation.

Policy options

Over the last three years the favourable macroeconomic outcomes arising from sensible use of monetary policy have enabled Australia to reduce unemployment by about one percentage point, from just over 8 per cent to just over 7 per cent, despite the absence of any serious attempt by the government to address the problem. With luck, a further reduction may be achieved over the next year. But, even in a more favourable environment, it seems unlikely that full employment will be restored without effort.

Public expenditure

The relatively weak growth in employment over the current cycle has already been noted. Langmore and Quiggin (1994) argued that expansion of employment in the human services sector (health, education and community services) would increase the level of employment consistent with any given level of economic activity and, in particular, with any given level of the current account deficit.

Quiggin (1999) uses ABS input-output tables to estimate the demand-shifting effect of an increase in expenditure on human services. Each $1 billion of expenditure in the human services sector, generates about 25 000 additional jobs. By contrast, each $1 billion of private final consumption expenditure is associated with about 13 000 jobs. Hence, an increase of $1 billion in expenditure on human services, and a corresponding reduction in general consumption expenditure, would be associated with a net gain of 12 000 jobs. Moreover, the proportion of full-time jobs would increase.

Because of the central role of education in building human capital, increased employment in education is particularly desirable. In fact, however, education spending has been tightly constrained. Cuts in Commonwealth grants to the states under the Hawke and Keating Labor governments led to cuts in spending on primary and secondary education. The (conservative coalition) Howard government’s 1996 Budget extended the cuts to tertiary education.

The unexpected resilience of the Australian economy has offered a further opportunity to reverse cuts in education spending, while maintaining a budget surplus consistent with stable levels of debt and net worth for the public sector over the course of the macroeconomic cycle. This would create jobs in the short term and, more importantly, increase long-term investment in human capital. But the government has already indicated that both human capital investment and fiscal consolidation will be sacrificed to the imperative of cutting income tax rates. This is the same strategy adopted, with dismal results, by New Zealand in the late 1980s.

Working hours

It appears obvious that, if everyone worked fewer hours, there would be more jobs to go around. This has led to numerous suggestions for work-sharing and even as in France, to legislation limiting hours of work. In reality matters are not so simple. Some jobs cannot easily be shared, and some workers prefer longer hours and higher wages. Nevertheless, there is strong evidence to suggest that the recent increase in average hours of work for full-time workers is an undesirable outcome of the increased bargaining power of employers.

If it is accepted that a reduction in hours of full-time work is desirable, it is not immediately apparent what can be done. The attack on the award system commenced by the Hawke—Keating Labor government and carried on by the Howard Liberal—National coalition government has reduced the capacity of governments to affect general working conditions.

One possible legislative response would be an general increase in annual leave from the present four weeks to, say, six weeks per year. This would be the equivalent of a 4 per cent increase in wages, or a reduction in standard working hours from 38 to 36.5 hours per week. Whereas a reduction in standard working hours could be negated by an increase in overtime work (paid or, more likely, unpaid), it would be more difficult to negate an increase in annual leave entitlements. Although some employees fail to use all their annual leave entitlements, they remain a small minority.

An increase in annual leave would be a particularly family friendly way of reducing working hours. For households with two working parents it would permit at least one parent to be home during school holidays, while still allowing for a family holiday.

Active labour market programs

The effectiveness of active labour market programs remains controversial. In Australia at least, the main difficulty is that few such programs have been maintained long enough to permit a serious evaluation, and most official evaluations have been highly politicised. The Working Nation program, which was introduced by the Labor government in 1994, cut back in 1995, and slashed by the incoming Liberal—National coalition government in 1996, is typical.

Official evaluations of theWorking Nation program have varied from glowing (Crean 1995) to circumspect (Department of Employment, Education, Training and Youth Affairs 1996) to damning (unpublished research cited by the Liberal—National government) according to the political demands of the occasion. Having initially condemned active labour market programs, the Liberal—National government introduced the ‘work for the dole’ program. Apart from the poll-driven political wrapping, this was simply a cut-down version of the job creation programs of Working Nation.

Non-official evaluations of Working Nation have generally been favourable, though this may reflect the fact that most such evaluations have been undertaken by supporters of active labour market policies. Piggott and Chapman (1995) presented an ex ante evaluation suggesting that the net budgetary cost per job created would be low, a view broadly supported by the ex post studies of Junankar and Kapuscinski (1997) and Stromback, Dockery and Ying (1997).

Tax policy

The low priority accorded to employment in Australia is nowhere more evident than in the case of tax policy. Although tax reform has been debated vigorously for two decades, the employment consequences of tax policy have barely been considered. Apart from some opportunistic point-scoring based primarily on the projections of computable general equilibrium models, the recent debate over tax policy conformed to this pattern.

Tax reform offered opportunities for a more employment-friendly policy. Some limited progress was made in integrating and simplifying the tax and welfare systems, but the Howard government’s primary concern was to improve the relative position of single-earner families. The need to eliminate ‘poverty traps’ associated with the high effective marginal tax rates generated by the interaction of the tax and social welfare systems received relatively little attention. In particular, proposals for some form of earned income tax credit, along the lines of that adopted in the United States, were disregarded.

The Ralph Review of business taxes (reference ….) could have addressed the employment effects of payroll tax, but any action on payroll tax was ruled out in advance under the terms of reference of the review. Payroll taxes, in the first instance, are taxes on jobs. In the long run, however, it can be shown that the effects of a payroll tax are similar to those of a GST. A GST is levied only on value added, and not on the value of purchased inputs to production, and therefore falls ultimately on labour and land (including mineral resources). Since the share of value-added attributable to land is small in most developed economies, it is approximately correct to say that, in the long run, a GST is equivalent to a payroll tax. However, long-run arguments of this kind rest on the presumption of full employment. Hamilton, Hundloe and Quiggin (1997) argue that a switch from payroll tax to a tax on emissions of carbon dioxide would yield a significant increase in employment.

Fiscal and monetary policy

During the long boom, it was generally assumed that the critical policy decisions affecting employment were macroeconomic policy decisions. Although Keynesians and monetarists disagreed on the relative merits of active and rule-based policies and of fiscal and monetary policy instruments, they generally agreed on this point.

Experience during the long slowdown since the early 1970s has shown that even with favourable fiscal and monetary policies it is difficult to achieve and maintain full employment. However, there has also been ample evidence that inappropriate fiscal and monetary policies can easily generate high unemployment. This is most evident in New Zealand, where after a decade of stagnation in the 1980s, the expansion of the 1990s was derailed by excessively tight monetary policy.

Recent Australian fiscal and monetary policy has been characterised by a mixture of good luck and good management. The decision of the Reserve Bank not to raise interest rates in response to the depreciation of the dollar in the wake of the Asian economic crisis (a depreciation that was exacerbated by an organised speculative assault) displayed good management. The combination of low interest rates and a modest depreciation in the trade-weighted exchange rate ensured that strong domestic demand and diversion of exports to US and European markets offset the drop in demand from Asia. The government’s fiscal policy, in which deep cuts in 1996 were offset by a modest relaxation in 1997 and 1998 was fortunate in its timing, since the Asian crisis began just as the domestic economy was recovering.

The main weakness in current macroeconomic policy is the absence of an active role for fiscal policy. Except in the later stages of the 1989—92 recession, decisions on fiscal policy have been driven more by the politics of taxation than by an attempt to stabilise aggregate demand. The result has been an excessive reliance on monetary policy. Monetary policy has worked well over the most recent cycle, but it failed disastrously in the 1980s.

Concluding comments

With a favourable macroeconomic environment, and a general downward trend in unemployment around the world, the unemployment rate could be reduced to 5 per cent or less through concerted policy action. Unfortunately, there is little sign of such concerted action. The present government has the best opportunity in decades for restoring full employment. History will judge it harshly if it fails.

References

Bean, C. (1999), ‘Australasian monetary policy: a comparative perspective’, The Australian Economic Review 32(1), 64—73.

Blanchard, O. J. and Summers, L. (1987), ‘Hysteresis in unemployment’, European Economic Review 31, 288—295.

Crean, S. (1995), Working Nation: The First Year 1994-95, Department of Employment, Education and Training, Canberra.

Department of Employment Education Training and Youth Affairs (1996), Working Nation: Evaluation of the employment and education and training elements, Evaluation and Monitoring Branch, Economic and Policy Analysis Division, DEETYA, Canberra.

Friedman, M. (1968), ‘The role of monetary policy’, American Economic Review 68(1), 1—17.

Hamilton, C.,Hundloe, T., and Quiggin, J. (1997), Ecological tax reform in Australia: using taxes, charges and public spending to protect the environment without hurting the economy, Discussion Paper No. No. 10, Australia Institute, Canberra.

Higgins, C. (1991), ‘Opening address to the Australian Economic Policy Conference’, Centre for Economic Policy Research, Australian National University, Canberra.

Junankar, P. N. and Kapuscinski, C. (1997), Was Working Nation working? A failed experiment or a premature abortion?, Public Policy Program, ANU.

Langmore, John and Quiggin, John (1994), Work for All: Full Employment in the Nineties, Melbourne University Press, Carlton, Victoria.

Parham, D. (1999), The New Economy? A New Look at Australia's Productivity Performance, Productivity Staff Research Paper No. May, AusInfo, Canberra.

Phillips, A. W. (1958), ‘The relationship between unemployment and the rate of change of money wages in the United Kingdom, 1861-1957’, Economica 25(4), 283—99.

Piggott, J. and Chapman, B. (1995), ‘Costing the Job Compact’, Economic Record 71(215), 313-28.

Pitchford, J. (1992), Current account deficits, external liabilities and economic policy, Discussion Paper No. No. 262, Centre for Economic Policy Research, Australian National University.

Quiggin, J. (1993), ‘A policy program for full employment’, Australian Economic Review (2), 41—47.

Quiggin, J. (1994), ‘The White Paper and after: policies for full employment’, Just Policy 1(1), 27—38.

Quiggin, J. (1999), ‘The public sector as a job engine’, forthcoming in Bell, S. (ed.), Unemployment Crisis: Which Way Out?, Cambridge University Press, Cambridge.

Stromback, T.,Dockery, M., and Ying, W. (1997), Labour market programs and labour market outcomes, Curtin University and Centre for Labour Market Research.

Toffler, A. and Toffler, H. (1994), Creating a New Civilisation: The Politics of the Third Wave, Turner Publishing, Atlanta.