How will a job-search subsidy create jobs?

Reads 7,973
Neil Rankin, 5 February 2013
A job-search subsidy has been proposed as a measure to help people find employment. At least three criteria need to be met to create new jobs for those who receive the subsidy. First, it needs to be used only to search for jobs or to remove the financial constraints that prevent people from searching for jobs; second, firms need to recruit through the channels which subsidy holders actually use to seek employment; and third, the relative cost of labour needs to fall.


It is well known that the price of transport is high in South Africa. These high costs are a barrier to individuals searching for work and applying for jobs when the applicant has to present himself or herself for selection or an interview. Thus, the idea of a job-search subsidy, mooted by the ANC, seems sensible. The question is whether this is so. And, if the subsidy is going to create jobs, what mechanisms are necessary for it to be successful?

Three criteria for success

For a job-search subsidy to create jobs for those who receive it, at least three criteria need to be met. The first is that job searchers should spend the subsidy money received on truly searching for jobs. There is evidence that searching for jobs is constrained by a lack of funds – and that an inflow of income to households from, for example, government social grants (such as old-age pensions) could encourage and enable the members of a household to search for jobs or move to a new area in order to find work (see, for example, Hosegood et al. 2009; Posel, et al. 2006). This being so, a job-search subsidy may indeed encourage people to seek jobs. In addition, the effectiveness of a job-search grant could be enhanced by tying it to a monitored search programme. Monitoring programmes in other countries require job seekers to prove that they have searched for employment by, for example, supplying copies of letters of application, registering with temporary placement agencies, and giving proof of participation in selection procedures.

The second criterion that needs to be met is that firms actually have to recruit through the channels that the subsidised job seekers will be using. Yet it seems that companies, especially those with high visibility, are approached directly by many young people seeking jobs – including people simply arriving at their doors. Surveys by the African Microeconomic Research Umbrella (AMERU) at the University of the Witwatersrand suggest that the average firm has more than one person approaching them for work each day. As a result, firms have easy access to a pool of potential workers – especially unskilled ones. No special effort is required.

These surveys also show that the most common way in which firms disseminate information about available jobs is via their own workforce, rather than by advertising or through other channels. A job-search subsidy may help work seekers who hear about jobs through such word-of-mouth networks but who cannot afford to travel to apply in person. However, those excluded from networks will not benefit much.

Then again, if the search subsidy increases the number of job seekers at the gate, it may  perversely lead to more recruitment through networks. Sorting applicants is a major cost of the recruitment process for firms. One way to reduce sorting costs is to recruit via the firm’s own workforce – it limits the number of people who hear about jobs and shifts some of the responsibility of sorting onto its own workforce. (It seems likely that existing workers will not refer unsuitable candidates since this may reflect badly on them.)

Another way is to limit the pool by asking for certain qualifications (such as matric) which have little to do with the ability to do the job – it reduces the work involved in screening and eliminating candidates. This is particularly the case for low-skilled jobs where applicants have fairly similar skills.

In addition, since employers do not know the productivity level of unsolicited applicants, it is costly for the firm to sort through large numbers of applications and gain (potentially) only small improvements in the match between worker and firm. Again, the cheaper and more certain channel of recruitment would be through their own workforce – thus limiting the impact of a job search subsidy.

Lastly, for firms to create more jobs (in circumstances where there is no increase in the demand for their products) requires that the relative cost of labour should fall, compared to the cost of capital. The cost of labour refers not only to the employee’s wage but could also include the uncertainty associated with unknown productivity levels. A larger number of potential candidates may lead to better matches between job vacancies and employees, and thus higher productivity (and lower labour costs). On the other hand, an increase in the supply of labour, with no increase in demand, would place downward pressure on wages. Thus, in the absence of growing demand, the mechanism through which a search subsidy actually creates jobs could be via lower wages. It is unlikely that the ANC had this route in mind when they suggested the policy.

Comparing a search subsidy and a wage subsidy

For a search subsidy to stimulate the creation of large numbers of jobs, firms’ cost of employment would have to be reduced. This means that individuals that are searching for jobs would have to accept lower wages than those prevailing in the market. It is not clear whether this would be possible with minimum wages, bargaining council extensions of minimum wages and other regulations which prevail in the labour market.

These implications of the job-search subsidy contrast with those of the proposed wage subsidy – which allows firms to claim tax breaks on employment, thus reducing the cost of employment directly and increasing firms’ demand for labour. Since the wage subsidy is given directly to firms, there is not as much pressure for them to reduce wages; wages should not drop much (wages could even rise if the subsidised worker is able to negotiate with the firm for a share of the subsidy). For a job-search subsidy to create as many jobs as a wage subsidy, wages will have to fall by more than any wage reduction occurring in the case of a wage subsidy.

Fairness in access to jobs

A job-search subsidy could also have redistributive effects. There are many reasons why a policy which does not necessarily create jobs but which redistributes employment opportunities could be supported. A major argument is that it would make access to jobs fairer. Currently, job-information flows favour those with connections to people with jobs (through family, friends or other networks). Searching for a job is also easier if someone in the family is employed and can help pay for transport. For example, young people in poor households with no one employed lose on both fronts – they hear about fewer jobs and cannot afford to travel to search or apply for jobs. A job search subsidy could redistribute job opportunities to the poor and marginalised, provided that firms actually hire via direct applications from job searchers. If this is the aim of the policy, it needs to be made clear.


Hosegood, V., Case, A. & Ardington, C., 2009. Labor Supply Responses to Large Social Transfers: Longitudinal Evidence from South Africa. American Economic Journal: Applied Economics, 1(1), pp. 22-48.

Posel, D., Fairburn, J. A. & Lund, F., 2006. Labour migration and households: A reconsideration of the effects of the social pension on labour supply in South Africa. Economic Modelling, 23(5), pp. 836-853.

Download article

PDF icon Rankin Feb 2013 Job search subsidy FINAL.pdf
Post a commentary

This comment facility is intended for considered commentaries to stimulate substantive debate. Comments may be screened by an editor before they appear online. To comment one must be registered and logged in.

This comment facility is intended for considered commentaries to stimulate substantive debate. Comments may be screened by an editor before they appear online. Please view "Submitting a commentary" for more information.