“Unemployment is a lack of jobs”. This viewpoint in engrained in our thinking. But is it time to say, instead, that unemployment is too many people looking for work? How could we manage not labour demand, but labour supply? If we upend the view that economic growth must be the core goal of economic policy, other routes forward appear on the horizon.
Economic policy has long had growth as a preeminent goal. We see unemployment, accordingly, as not just a problem for the unemployed, but a waste for society, a lost opportunity. Maximising growth means maximising work. However, this thinking is now challenged from three fronts: the green movement, debates on automation, and “post-work” critiques of work-centered social norms.
At stake here is not whether we should tolerate involuntary unemployment, but whether high levels of employment can and should be sustained. It is whether it might be better to want less work, and what might be done to reconcile our willingness to work to a world in which there is less work to do.
Replacing jobs lost to automation – does it happen automatically?
Since the outbreak of industrialization, we have experienced a tremendous increase in labour productivity. Robots haven’t yet stolen our jobs, however. This is despite widespread fears that this might happen – fears which stretch back at least to the 1960s. Historically, the loss of employment in some sectors (such as agriculture) has been compensated for in others (manufacturing, and latterly services) as real incomes have risen and patterns of consumer demand have shifted. The dominant view of mainstream economists such as David Autor is that this process is likely to continue.
Nonetheless, it is hard to look back over the last decades without noticing that this process –that of maintaining aggregate demand in the face of growing productivity and prosperity – has not always appeared automatic, or straightforward. Indeed, some authors in the Keynesian and classical traditions, such as Magdoff and Foster, argue that success so far has been contingent on certain historically specific conditions and interventions. It is worth viewing the question of the future of work within this historical context.
What has kept us working? A short history of demand stimulus
The 1950s and 60s are remembered as a golden age of high growth and high employment. They followed the austerity and deprivation of the Great Depression and the war years, in which households built up extensive savings. They saw rising expectations of future prosperity and the development of the welfare state, making people feel progressively more secure. They also saw a wave of new life-changing consumption goods, such as cars and suburban homes, TVs and other household appliances. All of these conditions are likely to have provided a sustained stimulus to consumption.
But it couldn’t last. From the late 60s, the growth engine seemed to falter. Keynesian demand management no longer seemed so easy when the oil shocks hit, with unemployment rising. The post war rebound had mysteriously run out of bounce.
With the rise of neoliberalism from the 1970s onwards, politicians and central banks found other ways to keep demand rising. The suppression of organised labour, and a tolerance for higher unemployment, reduced inflationary pressures, allowing for an unprecedented progressive reduction in the rate of interest over four decades. The effect of these interest rate cuts has been amplified by new financial practices, financial deregulation and the emergence of a new “buy now, pay later” consumer culture. This allowed for an explosion in borrowing by households and companies, accompanied by the dramatic inflation of asset prices – shares and real estate in particular – and growing volatility.
As productivity has advanced, the maintenance of employment levels has required ever-lower interest rates with ever-higher private and public debts, ever more inflated asset-price bubbles, and ever-deepening recessions in between. The IT revolution has given a boost, and briefly fostered a belief in a gravity-defying “New economy” in the late 1990’s, but ultimately its effect has been limited.
In the recent Great Recession, when interest rates finally hit zero, central banks had to turn to large scale money creation (quantitative easing) to reflate asset prices and keep banking sectors solvent. Recovery from this recession has only been achieved by means of yet more private and public borrowing and asset-price reflation.
Can we keep on stimulating demand?
How far can this process go? Interest rates remain near zero in much of the industrialised world. Central bankers worry about whether they have enough “ammunition” left, should another recession strike. Just as the conditions that created the postwar boom eventually faded, so it seems that the conditions that kept things going in the neoliberal era may also be running out of steam. We are having to try ever harder to maintain demand.
So sooner or later, advancing automation may translate into falling demand for labour, instead of rising living standards. There might certainly be ways to postpone this: a new arms race; income redistribution to put money in the pockets of people more likely to spend it; or a Federal Job Guarantee. But it is one thing to use a measure like a job guarantee scheme as a countercyclical tool, quite another to try to fight a loosing battle against a secular trend.
This may be a good moment to ask whether it makes sense to keep working so much. Should we insist on trying to maintaining full employment at a high level? Or should we shift to maintaining full employment at a progressively lower level?
Might less work be a good thing?
Like never before, people are now discussing whether we might be working more than is good for us. Without getting deep into those arguments, it is maybe worth rehearsing a few:
- Consumption brings diminishing returns. With limited time on our hands, it is actually rational to work less as income increases.
- Consumption is increasingly sustained by zero-sum games. Competitive, status-driven consumption makes no net addition to well-being.
- Unnecessary work is increasingly created by zero-sum games within production. Many marketing industry functions are notably given as examples of such putative “bullshit jobs”.
- Much work is done better outside the market. The commodification of some forms of reproductive labour, such as care work, expands opportunities for capital, “creates jobs”, and adds to GDP. Yet we often find more value in care work that is personal, not commodified, and on self-directed, non-alienating activity..
- Our ecosystem places safe limits on production. It is very difficult to reconcile the stimulation of demand, to maintain high employment in the face of advancing automation, with the need to limit the further disruption of our global ecosystem.
This said, there is also significant socially useful work to be done that is not yet being done. One clear case is the need for an emergency response to climate change. A Green New Deal would be a one-off attempt to readapt our productive systems to ecological constraints. It would justify a boost to employment, but not the maintenance of wasteful employment elsewhere.
Similarly, in the global north there is a risk of forgetting the scale of the income gap with the south (and the fact that some of our apparent productivity stems from the cheapness of labour there). A serious response to this from the North would also justify a surge of activity.
How can we have full employment, but at a lower level?
If we want to switch away from the maximisation of growth and jobs, we need to abandon the reflex of seeing unemployment as always a lack of jobs. We need to start seeing it as people wanting to work too much. We need to see labour supply not as a given, but as a proper object of policy intervention.
There are at least three ways to influence labour supply, in order to adjust it to demand:
- Working time limits, such as a four-day week, function to ration out available employment more fairly, whilst limiting total labour supply. Their downside is that they block some people from working as much as they would like – it’s one-size-fits-all.
- A universal basic income functions to reduce incentives to seek work, by disconnecting income from the supply of labour. However, a UBI is also a one-size-fits-all policy, rationing out income without paying attention to differential income needs and earning potentials.
- Needs-driven redistribution, such as non-means-tested welfare spending, also functions to reduce incentives to seek work by disconnecting income (in this case public health and education services, pensions, childcare allowances, etc.) from the supply of labour. Such redistribution takes approximate need and potential to work into account, shifting the burden of work onto those better able to perform it.
In some ways, these tools are complementary. Working time limits help to protect workers from abusive employers and have a very direct effect. Needs-driven redistribution compensates for the one-size-fits-all nature of working time limits or a UBI. A UBI appears most relevant at high levels of automation when it might become needed to complement needs-based provision, to further disconnect income from work and further reduce the level of employment we seek. It should be added that these are not measures to use counter-cyclically (for which demand management is still needed), but acyclically, to adjust how much employment counts as “full”.
As it gets harder to sustain growth, and as the desirability of growth is brought into question, we must confront the problem of how to manage a shrinking pool of paid work. We have to give up the employment maximisation reflex. But at the same time, we cannot accept mass, involuntary unemployment as just inevitable and somehow to be coped with. We have to think about what we can do to ensure that we collectively want to work just as much as we collectively can.
John Marlow is an economist working in the financial sector. He tweets @JohnTMarlow.
To support Renewal’s work encouraging debate and exchange across the British left, and receive the whole journal every quarter, please subscribe.