Johannesburg — IT's a most peculiar example of government spin. First you promise 500000 jobs by the end of the year. Then, a while later, you concede that maybe the target won't quite be met. But what you neglect to mention is that the jobs you promised were never real jobs in the first place: they were work "opportunities" in a public works programme.

That means they can involve anything from a few days' work clearing alien vegetation or planting trees to a few months of work renovating a school. They might even be slightly longer-term opportunities to earn a bit of money by helping at a creche or caring for the sick. What they are not is full-time, permanent, sustainable jobs.

Not that President Jacob Zuma said they were: when he made the promise in his June state of the nation address, he did talk of 500000 "job opportunities". He talked of them in the context of the second phase of the expanded public works programme, which aims to create nearly 4-million opportunities over five years. But when commentators understood this to mean proper jobs, the government and ANC spin doctors didn't correct them. Which is how they now find themselves having to dash the expectations they themselves created. It's also, perhaps, why Zuma reportedly told the ANC's national executive committee that even 100000 job opportunities could be considered a success, given the manner in which the economy was shedding jobs -- but also, according to City Press, suggested that the ANC needed a communications plan regarding the 500000.

As it happens, the target is not that wildly ambitious, as long as no one is under too many illusions about the duration of these jobs. In the year to March, the public works programme created 570000 opportunities, though these were equivalent to only about 180000 "person years of work", so they were clearly pretty short term on average, with daily pay ranging from about R40 to R80.

Since April, Public Works Minister Geoff Doidge reported last week, the programme created more than 220000 opportunities, so the government may just be on track to deliver on the 500000 by the end of the fiscal year, if not by the end of this calendar year.

How many days of work these offer the unemployed is not clear. But as a palliative for unemployment, especially when the economy has shed nearly 1-million jobs, the public works programme is important and its achievements are not insubstantial. It does provide some work and some income for poor households and may even provide useful training. It helps to deliver services to communities, some of which the government probably couldn't afford to provide if it had to rely on the relatively highly paid employees on the permanent payroll. And the government has over the past year introduced some innovations that should enhance the effect of the public works programme and get the public sector to absorb more labour -- paying incentives to provincial and local government to make infrastructure projects more labour intensive, for example.

It all helps. But it is not part of a job creation strategy in any true sense. Indeed, the trouble with the Zuma government is that for all its talk about "decent work" it doesn't actually have a job-creation strategy. Nor has it promised one any time soon.

Saving jobs is not the same thing. Zuma's state of the nation address put lots of emphasis on the "recession-rescue" framework agreement signed this year by business labour and the government; the ANC's alliance summit reaffirmed its commitment to it again at the weekend. But that agreement is mainly about trying to find ways to stave off job losses. It is not about ways to create new jobs. Not that preventing job losses is unimportant, and particularly so in an economy that, as it is, provides work for only a minority of people of working age -- only four out of every 10 adult South Africans have jobs, and in the case of young people fewer than half have ever been employed.

That's precisely the problem a strategy of job creation must tackle. So far, although there have been mentions from various ministers of the need for a more labour-absorbing growth path, there haven't been too many concrete proposals. One exception is the medium-term budget policy statement, which lists policy options available to the government, from expanding labour- intensive services to using the tax system to provide employment incentives to providing industrial incentives for labour-intensive sectors to issuing matriculants job vouchers.

Many of these options are controversial, inside and outside the government. But they need to be detailed and the policy trade-offs explored so that the government can start making some decisions about how it's going to deliver on what is, after all, one of its five priority areas. And if it wants to deliver on jobs, the government will ultimately also have to confront questions in political no-go areas, such as whether SA's system of labour market regulation is conducive to job creation (and, of course, whether destroying the labour broking industry will also destroy jobs).

But who will do the creative thinking required to ask the questions and devise a strategy, the hard work of investigating and costing options? Who better than that minister in search of a job description, Ebrahim Patel? He has at times couched his portfolio, economic development, in job-creation terms. But that's tended to get lost in the clamour for him to be in charge of "economic policy", whatever that might be. Is it just fiscal and monetary policy? Does it encompass trade and industrial policy? Does it stretch to the macroeconomics of energy policy and what we ought to do about Eskom?

So nebulous a notion of his portfolio risks defining him out of policy making altogether. And his efforts to build a large departmental bureaucracy could weigh him down, further diluting any effect he might have.

Far better surely to appoint himself "minister of jobs" and gather a small team of clever people, who could do some serious thinking about jobs and come up with some real solutions. Then perhaps the government might be able to spin public works into more permanent work.

Joffe is senior associate editor.