Why Don't We Have a Job Guarantee?

Basic income is often derided as impractical despite the evidence of success in past (albeit limited) trials and the willingness of governments to conduct further experiments. In contrast, the job guarantee is advocated as being more realistic (because closer to our belief in the value of work), and tends to enjoy more favourable coverage in the media, yet it has never been successfully trialled let alone implemented. Why is this? One explanation is that basic income experiments are often limited in scope to the unemployed, reflecting an ideological assumption about the desirability of integrating them into the labour market or bypassing the perceived disincentives of traditional welfare. In other words, schemes sold as basic income trials are often no more than attempts to reform existing unemployment benefits in a "market-friendly" way. A universal basic income (UBI) trial would need to address a full cross-section of society (the majority would be employed) and would necessitate a parallel tax and benefits system. In order to gauge pro-social benefits and the wider effects on the labour market, such a trial would need to run for longer than a couple of years.

The problem then is one of scale and the same challenge has historically limited the opportunity for a job guarantee (JG) system to be trialled. It's worth emphasising at this point that temporary job creation schemes, such as India's rural employment guarantee, or those aimed at the young or long-term unemployed in developed nations, are not job guarantees in the sense that the term is used by JG advocates. The chief feature of the JG is that it provides a permanent option that any adult can choose rather than being a temporary, counter-cyclical programme or a targeted relief. This make it particularly suitable for unemployment black-spots in an otherwise buoyant economy - for example, where a large plant closes or a major employer goes bust in a small town. When a worker is laid off by the private sector, she can immediately get a job provided by the public sector (this may actually still be a job operationally in the private sector, in that the provision of work projects may be outsourced, but her wage is paid by the state). Once the private sector starts hiring again, the worker would be incentivised to move back by the prospect of better pay and a better career fit.

The traditional arguments against the JG are essentially critiques of government: the enormous cost and potential for waste, the state's managerial incompetence and the crowding out of the more dynamic private sector. The solution to government is usually "the market", but this suffers in the case of a JG because the actual jobs market has already passed judgement in the form of unemployment, which leads some on the right to insist that the only solution to worklessness is to abolish minimum wage levels and severely reduce benefits to price labour back into the market. This has proven socially disastrous when tried. While the anti-government argument is plausible its material significance is rarely assessed. There is plenty of waste in the private sector and the social benefits of maintaining local communities may be considerable, so all of these imputed downsides might be tolerable given the upside benefits of employment and demand stimulus. A more honest criticism would be that capitalism still requires a "reserve army" of the unemployed, whether to maintain its political power (per Michael Kalecki) or to guard against inflation in the interests of rentiers (the NAIRU concept).

A key macroeconomic argument in favour of a JG is that it is a form of automatic stabiliser, maintaining aggregate demand in a recession and unwinding naturally through the labour market during a recovery. In counterpoint to this, JG sceptics fear that the system could become too attractive, thereby making labour "sticky" - i.e. reluctant to move into the private sector when vacancies arise, which would drive up the cost of labour for the private sector and so impede any recovery. The automatic stabiliser then becomes a fetter. While JG schemes often propose payment of the minimum wage, this could still make them more attractive than many private sector jobs because of a less stressful working environment or because of better fringe benefits. For example, schemes proposed in the US would provide access to healthcare, which is significantly better than what most minimum wage jobs in the private sector provide, though some advocates admit that this is intended to force the private sector to offer comparable benefits. In the UK, the NHS means that employment benefits would not be as great a differentiator (though they might be in the case of childcare). Even so, there would still be the possibility that some workers would prefer tidying-up public gardens to delivering pizzas.

The implied dynamic is that workers smoothly move between regular employment and the JG of their own volition, but for this to happen they must generally consider a JG job a second-best alternative to regular employment but must also be prepared to upgrade from the JG as soon as regular vacancies become available. The wage paid for a JG job must be sufficiently attractive to prevent recourse to residual benefits (assuming these are still available for hardcore refuseniks), while also being sufficiently unattractive to ensure almost no one would stick around once there were employment opportunities in the private sector. This implies both a significant differential between unemployment benefit and the JG, and between the JG wage and the minimum wage available in the private sector. The former is easily achieved (there would be widespread support to be parsimonious with those who refused to work) but the latter less so. In an era when low-wage private sector employment is precarious and fringe benefits have largely disappeared, the JG wage would have to be significantly lower to prompt workers to leave the security of a JG job.

One way round this is to make the JG wage variable, turning it up to the minimum wage level during periods of high unemployment and turning it down during periods of private sector jobs growth. But this assumes that labour is largely fungible and ignores the possibility of differential employment growth across sectors and geographies. If construction is expanding, should we reduce the pay of retail workers currently on a JG scheme? If London and the South East is booming, should we reduce JG wage rates in the North to encourage mobility? An answer might be to make rates variable by sector and location, but that is potentially destabilising for workers (it undermines the concept of a "guarantee") and may even end up producing sub-optimal behaviour, such as encouraging people to move to high unemployment areas to guarantee a liveable JG wage. The suspicion is that a JG can only work if it is both parsimonious (to encourage movement back to the private sector) and coercive (to prevent mass non-cooperation in the face of low pay). This might be one reason why it has never been fully trialled.

There has recently been a tendency to consider a basic income and a job guarantee as complementary rather than as alternatives: "the only cost-effective policy for comprehensive welfare is a combination of a modest basic income with job offer by local authorities below the minimum wage." In effect, the basic income would be parsimonious - enough to live on but not enough to flourish - making the JG wage sufficiently attractive to maintain employment, while the differential between the latter and a minimum wage job would encourage a speedy return to the private sector. This appears to combine the best of both worlds: security from destitution, meaningful employment and incentives to re-enter the labour market. A more progressive interpretation of this idea is that "Pairing a job guarantee with a UBI would mitigate the risk that the 'guarantee' would transmogrify under political pressure into a punitive workfare program. Pairing a UBI with a job guarantee would mitigate the risk that we neglect the broader project of integrating one another into a vibrant society, that we let a check in the mail substitute for human engagement".

Leaving aside the assumption that a UBI is anti-social (most basic income advocates emphasise its pro-social benefits), I'm not convinced that a basic income safety net would make the JG less coercive, even if it is no longer technically workfare as the entire income is not dependent on cooperation. The problem is that most of the unemployed would be obliged to accept JG jobs because the basic income would be insufficient, but those jobs are unlikely to be of high calibre because they would be paid at a much lower level that the worst jobs in the private sector. They won't be intrinsically rewarding or entail training in marketable skills. This is by design - the aim is to keep any viable minimum wage jobs in the private sector - but the result is more likely to be demoralising make-work jobs with little social benefit. There is also the challenge of preventing JG jobs being covertly used to service the private sector at a cost below the minimum wage. Given the public sector's reliance on outsourcing, drawing a clear line between public and private jobs is hard at the best of times. If the JG system was highly devolved, as most schemes propose, there would be obvious scope for corruption and abuse.

While basic income has remained conceptually consistent over the years, but has been given salience by the growth in precarious work and fears of technological unemployment, the job guarantee has had to reinvent itself to keep up with contemporary concerns. One significant change in JG advocacy is the move away from the idea that it will discipline labour (by institutionalising the reserve army of the unemployed) towards the twin ideas that it will embolden private sector workers in an era of weak unions (because they have an alternative) and provide upward pressure on private sector wages and employment benefits (to encourage mobility). This reflects both the growth in wage inequality and (more locally) the prominence of healthcare as an issue in the US, though there are clearly easier ways to address both problems than using the JG as a Trojan Horse (e.g. reversing anti-union laws and implementing universal healthcare). This latest iteration assumes that JG jobs would be relatively attractive, but the whole premise of the JG as an automatic stabiliser is that they would be sufficiently repellent to avoid sticky labour. It's not obvious that this fine balance can be struck, given the difference across sectors and geographies, while the notion of JG pay rates being selectively varied to address differences and encourage movement into the private sector undermines the idea of an automatic mechanism.

The job guarantee remains a superficially attractive idea that plays to our beliefs that work is intrinsically beneficial, that welfare should entail some quid pro quo, and that maintaining the calibre of labour for the benefit of the private sector is the responsibility of the public sector. All are dubious beliefs, in my view, but they are clearly commonly-held. Despite this broad acceptance, proposals for a comprehensive job guarantee system have invariably dwindled into marginal or targeted schemes that look remarkably like workfare or make-work. This probably reflects the dominance of the anti-state argument: a fear that a comprehensive job guarantee system would disrupt the "natural" labour market and lead to the harmful expansion of the public sector. Schemes that attempt to mitigate this fear through public sector jobs that would be non-competitive with the private sector ignore that the boundary between the two is not permanently fixed. Capital will not support a JG system that is in any way attractive to workers, or that restricts its ability to extract rents from the public sector, even if it helps support aggregate demand. As Michael Kalecki noted, it's about power.