Jobs for young people: Is a wage subsidy a good idea?

Executive Summary

ALMOST three quarters (72 per cent) of South Africa's unemployed workers are younger than 34. The unemployment rate of people younger than 25 is almost twice the national average (49 per cent compared with 25 per cent). And the 50 per cent of South Africans aged 15 to 24 who want jobs and actually have them is significantly fewer than the 80 per cent which the OECD reports as the norm in other emerging market economies. We have a national crisis of youth unemployment.

The government has begun to acknowledge the seriousness of the situation, and has proposed a range of interventions to address it. Some of these are of doubtful merit or contradict other government policies or proposals. One initiative, proposed by the National Treasury, is to introduce a wage subsidy for young workers. The logic of this idea is that employers will be encouraged to employ more young and inexperienced workers if the costs of doing so are subsidised.

The Treasury has proposed that the wages of all workers between the ages of 18 and 29 earning less than R60 000 a year - the current tax threshold - should be subsidised for a period of two years. Registered employers would receive the subsidy via credits on their PAYE accounts.

In the first year, the subsidy would comprise 50 per cent of monthly wages up to R2 000 a month, reducing to nothing for wages of R5 000 a month or more. The value of the subsidy would decline by 50 per cent in the second year. Young people who are currently employed would be eligible only for one year and at the lower rate.

The Treasury estimates that the programme would subsidise 423 000 workers. Of these, 245 000 jobs would be created in any case, and the remaining 178 000 would be created in response to the subsidy.

The Treasury estimates that some 45 000 workers would drop out of the labour force after having benefited from the programme, so the net result would be 133 000 more people employed by 2015, when the programme would end. The programme would cost R5 billion over three years. Each job would have cost an average of R37 000, although the actual subsidies would be considerably less.

In November 2010, CDE convened a Round Table of senior economic policy-makers, economists, and leaders from business and civil society to examine the key issues surrounding this proposal. It was also attended by two international experts: Professor Paul Romer, a renowned growth economist from Stanford University, and Prof Hian Teck Hoon of Singapore Management University, a leading labour market economist and expert on wage subsidies. Questions addressed included: Would a wage subsidy raise employment levels, by how much, and at what cost? Is this an appropriate response to the unemployment crisis? And would it be sustainable?

While some issues of policy design were discussed, the main focus of the Round Table was the feasibility, desirability and possible impact of the proposed wage subsidy.

A job-centred approach to growth

Employment is vital for South Africa's social, economic and political development. It is the key mechanism for addressing mass poverty. In addition, many other benefits flow from higher levels of employment. Many of the skills needed to improve a worker's employability - punctuality, discipline, the ability to work with others, and so on - are most easily acquired on the job.

This is especially important in South Africa. Millions of people are the products of South Africa's dysfunctional education system and have had few opportunities to acquire skills. For many, the workplace is the institution in which they are most likely to be able to acquire skills. Jobs also generate a sense of accomplishment, dignity and participation.

While welfare grants can provide a modest income, they cannot offer any of these other benefits, making employment growth vital to creating a more inclusive society.

As the National Treasury has stated, the proportion of working age adults with jobs may be the best measure of social inclusion in a modern society - far better in many respects than the Gini coefficient.

On this measure, Brazil, where close to 70 per cent of adults have jobs, is a far more inclusive society than South Africa, where barely 40 per cent of adults have jobs even though levels of income inequality are similar.

If South Africa is to be more inclusive, far more people must find jobs. But what kind of jobs?

South Africa's labour laws mean that any job that is fully compliant with the law is also, from an employer's point of view, a reasonably expensive job. This is especially true when compared with the costs of employment in other developing countries. Minimum wages in South Africa's clothing industry are two or three times higher than those for similar jobs in Swaziland and Lesotho, let alone India, Vietnam, Bangladesh and Pakistan.

To justify this, levels of productivity have to be high, resulting in employers' offering fewer jobs but for more skilled workers.

In recent decades the South African economy has generated fewer and fewer new jobs for every unit of economic growth. While some of this reflects increased productivity, much of the decline in job creation can be attributed to rising wages and increased labour market regulations, which have made employers more reluctant to hire workers, particularly those who are young and unskilled.

Employers, who might be willing to incur the costs of employing highly productive workers with significant skills, are much more reluctant to do so for workers whose lack of skills and experience mean they will be less productive. This is a key reason why so many young, inexperienced workers are unable to find work.

While South Africa has sought to ensure that jobs are well paid and well protected, dynamic emerging economies in Asia have sought to expand the absolute number of jobs as rapidly as possible. Many of these jobs do not pay high wages or offer very good conditions of employment, but they do pay better than almost any alternative form of employment for unskilled people. Critically, they exist in very large numbers.

Experience in Asia and elsewhere shows that once high levels of employment have been reached, productivity gains and progress up the industrial value chain lead to a rapid rise in workers' incomes and quality of life. It also leads to much higher rates of growth for the economy as a whole.

This is the process that has lifted hundreds of millions of people out of poverty over the past 40 years. And, apart from the discovery of previously untapped natural resources, it is the only process that has ever improved the quality of life of large numbers of people in poor countries both rapidly and sustainably.

The story of Singapore's rapid development is instructive. There, economic growth has generated a rapid change in its areas of comparative advantage. In the 1960s, shortly after independence, when unemployment was high, the country concentrated on attracting manufacturing firms, particularly in the garment and textile sectors. This was low-skilled work, but it lifted the incomes of the poor. In the 1970s, Singapore began to produce simple electronics, after which it began making hard-drives and semi-conductors. Since then, it has moved further up the value chain, with the country now having a comparative advantage in bio-medical research and development. Thus, over the decades, the country has steadily moved up the ladder of comparative advantage, with very positive implications for average wages as well as economic growth.

The Treasury's case for a wage subsidy

In response to the need to create jobs for young, unskilled workers, the Treasury has proposed the introduction of a wage subsidy for young workers at the bottom of the pay scale. Such a subsidy would narrow the gap between the costs employers incur when employing these workers and those workers' likely levels of productivity. The Treasury argues that this would induce firms to employ more young people.

The subsidy directly lowers the cost of employment to the employer. And, by getting inexperienced young people into jobs, it would enable them to acquire skills which will raise their productivity. This is also why the Treasury proposes that the subsidy would be offered for only the first two years of employment. This is on the basis that the beneficiary's rising productivity reduces the need to subsidise his or her employment.

The risks of implementing a youth employment subsidy

There is a strong case for instituting wage subsidies in South Africa. Promoting employment would generate far more inclusivity than increasing welfare payments. At an appropriate level, these subsidies could also stimulate significant employment creation. There is good evidence from countries as diverse as the United States, Belgium and Singapore that wage subsidies do encourage employment and help to reduce poverty.

Nonetheless, there were a number of concerns about the introduction of a wage subsidy.

These included:

Cost and sustainability: The number of jobs created by a wage subsidy depends on how many workers it induces firms to employ. This depends on the size of the subsidy, but larger subsidies raise issues of affordability and sustainability.

Waste: Wage subsidies can be more efficient than other forms of public policy, but they can also generate wasteful spending when jobs that would have existed in any event are also subsidised.

 Employer response: Many employers may feel that it would be too risky to take on potentially unsuitable (albeit subsidised) workers unless the process of dismissing them was made considerably less onerous.

Opportunity costs: The funds devoted to a wage subsidy might be better used for other government initiatives.

Sustainable jobs: The Treasury's proposal assumes that, after two years of subsidised employment, a young worker will be productive enough for employers to keep him without the subsidy. Whether this will happen is impossible to know.

Creating new distortions: A wage subsidy could result in the growth of businesses whose only rationale is to absorb public money through subsidies. Employers might also replace unsubsidized workers with subsidised ones.

Two of the most significant concerns were that a wage subsidy would do nothing to address the current regulatory regime, which plays a key role in raising the costs of employment, especially of unskilled and inexperienced workers; and that the Treasury's proposal was too small to impact significantly on South Africa's unemployment crisis or on the politics of reforming the labour market.

The wage subsidy and the costs of employment

In some respects, the Treasury's proposal represents a significant break from existing government policy. For the first time, an intervention is being proposed that takes seriously the negative impact of high employment costs on people's chances of finding work. It suggests that some policy-makers have recognised that high and rising employment costs are a key reason for South Africa's unemployment crisis. This injects a dose of realism into the debate about the choices facing the country. However, this does not necessarily mean that a wage subsidy is the best way to create jobs.

Is a wage subsidy desirable in principle?

Employment incentives are common across the world, with OECD countries spending an average of about 0,15 per cent of GDP on programmes of this sort. In South Africa, this would amount to R15 billion over three years, or three times more than the costs of the programme that has actually been proposed.

There are many reasons why governments subsidise wages, and the relevant literature suggests that these programmes are quite effective in relieving poverty or increasing employment levels. In South Africa, however, questions need to be asked about the desirability of introducing wage subsidies when it is existing labour market regulation that has ensured that labour costs have risen quickly over the past few decades. In these circumstances, a wage subsidy - which transfers some of the costs of employment from employers to taxpayers - may not be the most appropriate way to close the gap between employment costs and productivity. Instead, it may be more appropriate to address directly the reasons for high and rising labour costs.

Is government doing enough to reduce employment costs?

In general terms, the outlook for labour market reform that would reduce labour costs is poor.

As recently as December 2010, the government proposed amendments to the Basic Conditions of Employment, Labour Relations and Employment Equity Acts, and proposed a new law - the Employment Services Bill - aimed at governing labour brokers. There is widespread agreement that these proposals would have increased the cost of employment and resulted in more unemployment.

As a result, the ‘social partners' represented in NEDLAC - national government, organized business and organised labour - agreed in April 2011 that the four bills should be re-drafted.

A process of consultation is currently under way regarding these bills. It is not yet clear what will emerge from this. The fact that these bills were drafted at all, however, suggests that key government institutions want to increase labour market regulation rather than decrease it. It is especially telling that government proposed these measures even after its own regulatory impact assessment concluded that the changes would reduce employment. It seems unlikely, therefore, that reforms will emerge from this process that will significantly reduce the direct and indirect costs of hiring young, unskilled work-seekers.

There is no indication that the proposed wage subsidy is part of an incremental strategy to reform the labour market. If the wage subsidy is an isolated initiative, is it worth implementing anyway?

Is the National Treasury's proposal worth implementing?

If the Treasury's estimates are correct, and the subsidy would increase employment levels by 133 000 at an overall cost of R37 000 per job, it would use public funds more efficiently than some other interventions. In fact, to the extent that these jobs would generate taxable profits for employers, they would actually cost government less than this estimate.

By contrast, the Expanded Public Works Programme creates short-term, low-wage jobs at an annual cost to the taxpayer of R100 000 for the equivalent of a full-time job. Similarly, the Industrial Development Corporation's newly announced fund, which will provide concessionary finance to job-intensive investments, will target projects that create jobs at an average cost of between R250 000 and R500 000 a job.

From the point of view of employment creation, the proposal also appears to be far more efficient than other subsidies, such as the estimated R18 billion that public support for the automotive industry cost the economy in 2009/10. While there are other benefits associated with having this industry, and while getting accurate figures is not easy, one estimate suggests that this support costs the economy between R200 000 and R600 000 per job, depending on whether employment in upstream industries is included. Including down-stream workers as well lowers the annual cost per job to R60 000. But, unlike the estimated once-off R37 000 cost per job of the wage subsidy, these costs are incurred by the economy every year.

An important advantage of the Treasury's proposal is that if new jobs are not created, the money would not be spent. Furthermore, because the jobs created would be in the private sector, they would probably be in areas of the country (and sectors of the economy) in which sustainable employment is a more plausible outcome than is the case with other proposals.

All this - and the fact that the programme is expected to cost less than 0,2 per cent of anticipated government spending over the next three years - suggests that, on its own terms, the wage subsidy is a sound use of public resources. There are, however, other issues to consider.

The proposal to subsidise the employment of young workers whose salaries fall below the income tax threshold is intended to increase the chances that people in this demographic will find work. Subsidising their employers is one option for achieving this. An alternative would be to engage directly in the process of reforming the labour market in order to reduce the costs of employment - in general, or for this group of workers.

The proposed subsidy does not do this even though there are ways in which the two kinds of intervention could have been aligned. For example, when a proposal to institute a wage subsidy was endorsed by the team of Harvard-based economists that advised the National Treasury on South Africa's economic strategy in 2006/7, it was deemed ‘essential' to link the subsidy to the creation of a probationary period during which subsidised workers could be dismissed on a ‘no questions asked' basis.

No one has tried to calculate how many more jobs might be created if the wage subsidy were linked to a probationary period of this kind. This could be considerably more than the 133 000 that the Treasury projects for a subsidy on its own. By allowing employers to dismiss unsuitable workers relatively quickly, a probationary period could increase the number of jobs created, help target subsidies to the most productive and capable workers, and increase the likelihood that those employees who were retained by their employers would be kept on when their subsidies expired.

Such a proposal might provoke political resistance. However, if the government is serious about dealing with mass unemployment, it will have to confront these sorts of issues. As it stands, the Treasury's proposals - assuming its projections are accurate - would have only a modest impact on the crisis of unemployment.

Could this proposal be scaled up?

In principle, a subsidy would have a greater impact if it were larger. This could be accomplished either by making each subsidy larger or by extending the length of time for which beneficiaries are eligible. Doing this would increase the costs per job and the overall costs of the programme.

While this raises questions of affordability and sustainability, it may be that a programme costing R15 billion over three years but which also created significantly more jobs would be worth having.

South Africa would then be spending as much as OECD countries spend on wage subsidies, but it would still be less than 1 per cent of the national budget. If 350 000 new jobs were created, the subsidy programme would more than double the chances of a young person finding work in the next three years.

One reason for framing the issue in these terms is the sense that 133 000 new jobs are simply too few to make a meaningful difference to the employment prospects of the jobless. Equally important, this number may also be insufficient to shift the debate about labour market policy.

In practice, a much larger programme could help change the terms of the debate and start convincing those who would like to increase regulation of the labour market that creating more jobs actually requires lowering the costs of employment.

Concluding remarks

The government has announced or introduced a large number of proposals, policies and programmes with the stated aim of creating more jobs. Some of these would create a small number of relatively high-skilled and capital-intensive jobs for which the majority of the unemployed are not unsuitable. Others have created a larger number of temporary ‘make-work' jobs that offer transitory and modest relief from poverty with little prospect of a sustained improvement in the beneficiaries' quality of life. Both approaches are expensive, and neither offers a realistic longterm solution to South Africa's crisis of unemployment.

The Treasury's wage subsidy proposal is also not a comprehensive response to the crisis of unemployment. It is, nonetheless, a step in the right direction and should be introduced. It would be preferable, however, if it formed part of a series of initiatives.

One option would be to introduce a national wage subsidy and to encourage and support a variety of policy experiments at the provincial, municipal, or even sub-municipal levels. Another would be to subsidise wages in some sectors or areas and introduce other initiatives in others, so that the impact of different approaches could be assessed. Such policy experiments could include:

• Allowing employers to offer young people jobs at wages lower than the established minimum wages;
•  Allowing young people to opt out of employment regulations altogether as a part of their employment contract; and
• Establishing special economic zones with different rules and regulations aimed at promoting low-cost, export-oriented manufacturing firms.

These are all potentially useful initiatives. Perhaps the most straightforward and appropriate policy change to link with the institution of a wage subsidy is the one originally proposed by the Harvard-based team advising the Treasury: any new hire who benefits from a wage subsidy should be subject to a ten-week probationary period during which employers can dismiss them on a ‘no questions asked' basis. If this were implemented, employers would be less concerned about being stuck with unsuitable employees, while new hires would have every incentive to show that they are suitable. The results would be that only those people who are most likely to be suitable would be subsidised, dramatically improving the efficiency and effectiveness of the wage subsidy itself.

Locating wage subsidies in a wider programme of labour market reform and experimentation could extract far greater value out of government spending. The key is to try to leverage as many jobs as possible from policy initiatives of this kind, and to shift the policy debate to the fundamental reasons for South Africa's high and rising labour costs.

There is merit in the Treasury's proposal, but South Africa could do better. First prize would be to address directly the high costs of employment, especially of young, unskilled and inexperienced workers. While this would be politically difficult, the benefits could be significant.

Even if political circumstances effectively rule out more fundamental labour market reform, serious questions remain about the current Treasury proposal. Chief among these is why it is so modest. If the Treasury is confident about its projections, why it is not pushing for a more ambitious programme? Is it concerned that it would cost too much? Is it seeking to avoid a more serious political battle?

That said, there is also an argument that the modesty of the proposal is a virtue. The success and affordability of a wage subsidy is not guaranteed. If there is some chance of failure, the relative modesty of the proposal insulates it from incurring significant costs.

Whether or not the proposed wage subsidy succeeds, questions need to be asked about what government will do next. If the programme does succeed, will more resources be devoted to subsidising employment, or will a new debate about addressing the costs of employment be instigated? And if the proposal fails to increase employment, what lessons will government (and others) draw from this experience? Will efforts to tackle the costs of employment be expanded and enhanced, or will there be a retreat from any reform at all?

These are vital questions because, even if it achieves its stated goals, the Treasury's proposal will make no more than a modest dent in South Africa's massive challenge of unemployment.

Addressing unemployment is South Africa's most pressing national priority. As the Treasury itself recognises, a relatively small wage subsidy programme would not have a very significant impact. Nonetheless, this proposal - if seen as a learning experiment with vital policy implications - does seem a sound use of public funds. It certainly holds more promise than many other government initiatives costing considerably more. It would be a tragedy if this modest experiment was defeated or abandoned.

The Treasury's proposal is premised - rightly - on the recognition that for South Africa to become a more inclusive society, much more needs to be done to help unskilled and inexperienced young people to get jobs. This is an important first step.

The Round Table and this publication were funded by Business Leadership South Africa. The funders do not necessarily agree with the views expressed in this report.

This is an executive summary of the Centre for Development and Enterprise report, Jobs for young people: Is a wage subsidy a good idea? CDE Round Table no 17 (August 2011). The full-length publication is available here - PDF.

Click here to sign up to receive our free daily headline email newsletter