Job Destruction and Job Creation:
A Beginners Guide
by Wolfgang Kasper
Click here for PDF version

Contrary to much political rhetoric, it is private enterprise that creates jobs, not governments, whose interventions often kill jobs

A low rate of unemploymentÑmeasured as the percentage of those in the working-age population seeking work at the prevailing wageÑis widely considered an important goal of economic policy in Australia. Unemployment is thus an excess of the demand for jobs (or employment) over the supply of jobs by enterprises at a point in time.

A simple model of the labour market

Standard economic analysis can help to clarify the demand-and-supply situation in the jobs market.

First, we have to ask: What determines the supply of jobs? Time-tested theory relates employment to production, through what economists call an Ôaggregate production functionÕ. It seems plausible that expectations of additional units of output (for example, measured as billions of dollars of national product in real terms) lead to expectations of additional demand for labour input (refer to Graph 1 opposite, upper panel). The relationship between inputs of labour and outputs of national product tend to be such that the marginal output-input ratio or Ômarginal productivity of labourÕ (MPL, depicted by the gradient of the national production function) becomes gradually flatter as production increases. What this means in plain English is thatÑfor a given capital stock, given skills and technology, given natural resources, and given economic structuresÑadditional workers add less and less to overall output until the marginal product of additional workers (MPL) becomes zero (middle panel of Graph 1).

When businesses are exposed to a reasonable degree of competition, they will seek to hire workers as long as additional workers add to their profitability. If we consider production level [Y/P]* and assume that the average real wage level (wages adjusted for inflation) stands at [w/P]*, the business sector will over the medium term seek to offer jobs to L* number of workers, no more and no less. Were business to offer fewer jobs (say, L**), profitability could be increased by adding workers. The marginal workers would contribute more to output (have a higher productivity) than they cost (namely the prevailing wage [w/P]*, see central panel in Graph 1). If, on the other hand, more people were employed than L* (say L***), businesses would make a loss, as the marginal workers would cost more than they add to output ([w/P]* > MPL***). In other words, there is a tendency for job offers to adjust to the real wage level.

If the real wage level rises, then fewer jobs will be profitable and fewer jobs will be offered. The marginal productivity of labour (MPL) is thus equal to the demand for labour (or the supply of jobs). The amount of labour that people seek to supply also depends on the real wage. The higher it is, the more people will plan to work (lower panel in Graph 1), which reflects the labour market. This allows us to depict unemployment for alternative real-wage levels as the difference of supply of and demand for labour (expressed differently: the excess of jobs sought over jobs offered).

How to eliminate unemployment?

Graph 1 can help us to discuss various ways in which unemployment can be eliminated.

(a) Cut nominal wages: One obvious way is to reduce the real wage by cutting nominal wage rates [w]. The number of jobs offered will increase and the number of job seekers will fall until there is an ÔequilibriumÕ in the jobs market, the demand for and the supply of jobs being compatible with each other.

A high flexibility of nominal wages is how job markets used to function in 19th century Australia, how they work in many parts of the flexible US economy and how they still function in many third world countries. The normative argument against this state of affairs is that wage cuts impose hardship on workers. But what is harder to take for someone willing to work: to be out of work or to take home a smaller pay packet? When Ansett Airlines went broke in 2001-02, many pilots and cabin staff welcomed new jobs with Virgin Blue even if that meant a pay cut. For many people, the social life and the stimulation on the job are preferable to sitting idly at homeÑbut this is a subjective, normative issue. In addition, one can argue that it is often not desirable to shrink wage incomes, as this might reduce demand for product and hence the employment of others.

(b) Erode the real wage by inflation: Another way to eliminate un-employment is to encourage inflation (rising price level P) and keep nominal wages constant, so that the real wage [wnotional/P] is reduced. When trade unions and government regulations made nominal wages downwardly rigid in the late 19th and early 20th centuries, industrial economies began to experience high and durable unemployment levels. In the 1930s, these levels became socially intolerable.

British economist, John Maynard Keynes, suggested that unemployment could be lowered by the surreptitious means of keeping nominal wages low and encouraging ÔmildÕ inflation (printing money, public deficit spending). When this recipe was tried in a major way in Western countries in the 1970s, it failed conspicuously. Trade unions had the muscle to negotiate for inflation-adjusted (real) wages and there were many automatic inflation clauses in wage contracts. The ÔKeynesian trickÕ proved to be impossible. AustraliaÕs Labor governments of the 1980s and early 1990s nevertheless tried the trick again: They concluded a political deal with the unions to control wage levels (ÔThe AccordÕ) and went for cautious demand expansion. But rising public debt levels, the governmentÕs labour-market regulations, union influence and public resistance to inflation prevented the hoped-for drop in unemployment numbers.

The biggest and longest test of the Keynesian recipe to Ôprint jobsÕ by inflating demand has been conducted by Japan since the late 1980s; one stimulus package after the other was launchedÑand the result has been rising unemployment. It is now widely accepted that the Keynesian idea of demand injection, accompanied by nominal wage control, may work once, but that repeated injections work like heroinÑwe become inflation addicts and get less and less of a ÔliftÕ from another demand injection.

(c) Reduce labour supply: Labour supply is reduced if people are paid for not working; that is, if unemployment is subsidised to help the jobless. Then, the labour supply curve shifts to the left. This of course makes the prevailing real wage more tenableÑno wonder organised labour unions favour taxpayer-financed subsidies to the unemployed. The dole removes some low-cost competition for workers.

In recent years, successive Australian governments have reduced the eligibility for the dole and have made it harder to remain unemployed. This has made a real contribution to improving the jobs situation.

Other ways to reduce labour supply is of course to cut the work week, to subsidise early retirement from the workforce or to keep young people longer at school or university. Both these types of labour market policy reduce the capacity to produce and hence lower living standards.

(d) Increase labour productivity and the demand for labour: A fourth way to eradicate unemployment is to shift the demand-for-labour schedule to the right (Graph 2 opposite). The assumption made so far was that there is a given production function, with given levels of skill, capital, resources, technology and enterprise. But this is a short-term and static view, and as such it is misleading. In reality, we live in a dynamic world of growth, where the production function is shifted upwards by investment in new tools, better human knowledge, skills and work practices, the tapping of new natural resources, technical innovation and structural changes. All this is driven by enterprise, as long as there is economic freedom and small government (low taxation). Of particular importance is the removal of counterproductive work practices (a deregulated labour market). In a deregulated, dynamic economy, the production function becomes an upwardly mobile feast. This means at the same time that the demand for labour schedule shifts to the right (Graph 2). At the given real wage [w/P]1, job offers increase to L2. Alternatively, higher real wages can now be earned if the workforce remains at L1.

The drawback for some is that everyone has to compete.1 Union powers and special privileges have to be controlled and governments have to remain small and non-interventionist (no hand-outs in exchange for favours to the party in power). But the advantages of this job creation strategy are massive: a higher real wage (higher living standard) is compatible with full employment.

This approach to job creation has been tried, within limits, by successive Australian governments since the 1980s: reduction of protection and subsidies, tax reform, less subsidisation to yesteryearÕs shrinking industries and regions, less favouritism and partial labour market reform. It has led to a moderate Ôproductivity explosionÕ and a new confidence and optimism.2 The 1990s have reconfirmed that rising productivity is good for employment and that measures to raise the production function (called Ôsupply-side policiesÕ) are the most promising escape from unemployment. Much more could be done, however.

Having reached this stage, the reader may ask: What about the frequently heard assertion by politicians that the government created so and so many jobs? The fact is that most jobs are created by private enterprises, that is people with capital and ideas who employ others. Governments only create jobs when they put more bureaucrats or workers on the government payrollÑand the taxes to pay their wages and salaries are a burden on private employers and hence on private job creation. As we saw, the attempt to inflate demand in order to Ôprint jobsÕ is largely discredited by experience and depends on workers tolerating a surreptitious erosion of real wages. When governments try to take credit for jobs growth (or are blamed for unemployment) that is based on tenuous and indirect connections, such as the removal of obstacles to employment by microeconomic reforms, as discussed in point (d) above. Beyond that, it is just ill-informed political propaganda.

By way of a summary: who kills jobs?

Like a map, the simple model underlying the two graphs abstracts from many time lags, frictions and complications to assist us with navigation through a complex problem. In reality, job creation lags behind the cyclical fluctuations in overall demand, so that watching unemployment figures as the key indicator of economic health is a bit like driving with your eyes fixed on the rear vision mirror. It is accident-prone. We must also be aware that behind such big aggregates as Ôjob demandÕ and Ôjob supplyÕ is a diverse variety of different jobs and skills.

In a living economy, unemployment is also the difference between the ÔbirthsÕ and ÔdeathsÕ of jobs. There is no deep mystery as to who causes still-born jobs that never eventuate and who kills existing jobs. From what was said above, we can identify the chief culprits:

(a) One major jobs killer is a wage explosion. When, in the wake of booming demand, unions were able to use their power to obtain massive nominal wage increases in the early 1970s and in 1981, many jobs disappeared and unemployment roseÑnot because of a cyclical downturn but because of wage rises. Similarly, the judges engaged in Australian industrial relations cases, who often display little understanding for economic side effects, have time and again ÔgrantedÕ wage rises far in excess of productivity growth and thus killed jobs.3

(b) Another big jobs killer is the social welfare state. By diminishing the incentive to seek work, especially at the lower end of the skills range, easy access to social welfare reduces the supply of labour and protects people from re-skilling or moving to areas of high labour demand.

(c) Over the longer run, the greatest jobs killer is the regulatory system that inhibits entrepreneurs from shifting the production function upwards, creating well-paying jobs: the maintenance of obstacles to foreign trade and investment by the Federal government, the many regulations of production and trade, the support by elected parliaments of favours to special interest groups, subsidies to uncompetitive industries and regions,4 and regulations of work and production which hinder the flexible adjustment and the skill acquisition needed to cope with changing circumstances. If some State governments inhibit the tapping of natural resources, for example by unilaterally anti-cipating Kyoto-style prohibitions on burning our copious black coal, one of AustraliaÕs greatest natural assets, then this hampers the upward movement of the production function and destroys jobs.

There may be good reasons why certain jobs will be killed, for example regulations to ensure public health and safety, social compassion for poor people, political opportunism to favour particular firms and regions, or intentions to protect the environment. Open political debate should in these cases explicitly weigh up the employment consequences with the other policy objectives. But it is wrong to deny, as is often the case, that the welfare state, Greenhouse policies or the industrial relations system create a dysfunctional labour market and cost jobs.

The track record and the future
There has been some unemployment throughout the 20th century, whereas there had been consistent labour shortages in the 19th century, partly mitigated by the attraction of immigrants.5 The 1950s and 1960s were a period of low unemployment, but the 1970s to 1990s, after the two above-mentioned wage explosions and Federal inflation policies, saw a Ôratchet effectÕ: every recession brought a rapid rise in unemployment, followed by only a partial reduction in the following upswing. The gradual economic deregulation during the last two decades of the 20th century, and the reduction of subsidies for being unemployed, have in recent times helped to ease the unemployment situation somewhat. Obviously, much remains to be done.

It is desirable to do more to raise labour demand, because unemployment is not only a disheartening experience, but also unjust: a lack of a job is the single most powerful source of poverty. Unemployment inflicts burdens on many of those young people who have poor job prospects and may indeed not get the chance to get their foot on the Ôjobs-and-skill-learning ladderÕ. Arguably, the most important aspect of high employment is that it empowers ordinary people. When labour and skills are short, willing workers are appreciated, and the ÔbossesÕ compete amongst each other to attract them. The long history of deficient employment opportunities and wage fixing in Australia led to persistent unemployment. Many observers conclude that this is the result of class warfare. The reality is that it is an unintended consequence of price fixing, which has generated an entire Ôdisadvantage industryÕ. If we had one generation of high employment and free labour markets, the apparent disadvantage would vanish.

There are, therefore, powerful grounds for arguing for full labour market deregulationÑabolishing AustraliaÕs odd and outdated industrial relations system, where quasi-judicial officials set wage levels centrally, and withdrawing union privileges to form cartels and impose counter-productive work practices. Once workers and employers are able to negotiate freely for work conditions throughout the economy (also with the help of experienced agents), the production function will shift upwards sufficiently to give everyone a job opportunity and rising real incomes.

The most important step in this direction is that people learn to cut through interest group propaganda and think clearly about who the killers and the creators of jobs areÑand who deserves political priority: special, organised interest groups or the population at large, including the young.

Endnotes
1 W. Kasper, Property Rights and Competition: An Essay on the Constitution of Capitalism, CIS Policy Monograph 41 (Sydney: The Centre for Independent Studies, 1998).
2 Productivity Commission, Microeconomic Reform and Australian Productivity: Exploring the Links, vol. 1 (Canberra: AusInfo, 1999), 23-33.
3 P. McGuinness, The Case Against the Arbitration Commission, CIS Occasional Paper 11 (Sydney: The Centre for Independent Studies, 1985).
4 In politics and the media, there is a frequent confusion between the preservation of certain employment structures (for example, by subsidy to a loss-making firm) and the fostering of high employment levels. The former often undermines the latter aim, because resources are artificially channelled into declining activities, thus hindering the conditions for general jobs growth.
5 The evidence about whether immigration adds to unemployment or leads to net job creation seems clear to me: on balance, new immigrants add to aggregate demand and create more jobs, as long as they are skilled and willing to work. Immigrants also help with new ideas and make the production apparatus more flexible, raising the production function. But this is a matter of migrant selection: the case is less clear when the share of welfare-dependent refugees and family reunion migrants goes up and skilled and business migration goes down.

Policy is the quarterly review of The Centre for Independent Studies. For more information on subscribing to Policy, click HERE

If you are interested in the Centre's activities and publications, why not subscribe to e-PreCIS, our regular email update on the latest news and events.

(e-PreCIS requires html capable email facilities, such as Microsoft Outlook Express or Netscape Messenger)