As the world recovers from the Great Recession, it has become increasingly difficult to discern the true trend of events.
On the one hand, we measure recovery by our success in regaining pre-recession levels of growth, output and employment. On the other, there is a disquieting sense that today's "new normal" may be slower growth and higher levels of unemployment.
So the challenge is to formulate policies to provide work for all who want it in economies that, as currently organised, may not be able to do so. This issue is much more acute in developed than in developing countries, though interdependence makes it, to some extent, a common problem.
The problem has two aspects. As countries become more prosperous, one would expect their growth rates to slow. In earlier times, growth was fuelled by capital scarcity: capital investment attracted a high rate of return, and this created a virtuous circle of saving and investment.
Today, capital in the developed world is abundant; the saving ratio declines as people consume more; and production shifts increasingly to services, where productivity gains are limited. So economic growth - the rise in real incomes - slows.
This was already happening before the Great Recession, so generating full-time jobs that pay decent wages was becoming ever more difficult. Hence the growth of casual, discontinuous, part-time jobs.
The other aspect of the problem is the long-term increase in technology-driven unemployment, largely owing to automation. In one way, this is a sign of economic progress: the output of each unit of labour is constantly rising. But it also means that fewer units of labour are needed to produce the same quantity of goods.
The market's solution is to re-deploy displaced labour to services. But many branches of the service sector are a sink of dead-end, no-hope jobs.
Immigration exacerbates both aspects of the problem. A large part of migration, especially within the EU, is casual - here today, gone tomorrow, with none of the costs associated with full-time hiring. This makes it attractive to employers, but it is low-productivity work, and it increases the difficulty of finding steady employment for many people.
So, are we doomed to a jobless recovery? Is the future one in which jobs are so scarce that many workers will have to accept a pittance to find any employment, and become increasingly dependent on social transfers as market-clearing wages fall below the subsistence level?
Or should western societies anticipate another round of technological wizardry, such as the internet revolution to produce a new wave of job creation?
It would be foolish to rule out the last possibility.
Capitalism has a genius for reinventing itself. It has seen off all of its challengers, and there are no new ones in sight.
But there is also a more troubling possibility: if, by proceeding on our current profligate path, we succeed in making natural resources scarce, we will require a new wave of technology, regardless of the cost, to rescue us from calamity.
But let's put these grim prospects aside, and ponder what a civilised solution to the problem of technology-driven unemployment would look like. The answer, surely, is work-sharing.
To the Anglo-American economist, any such proposal is anathema, because it smacks of the dreaded "lump of labour" fallacy - the idea, once popular in trade-union circles, that there exists only a certain amount of work, and it should be shared out fairly.
Of course, this is a fallacy when resources are scarce.
Some economists expected that, at some point in the future, mankind would attain a "stationary state" of zero growth. Then we would require only a certain amount of work - much less than we perform now - to satisfy all reasonable needs.
The choice would then be between limitless technology-driven unemployment and sharing out the work that needed to be done. Only a workaholic would prefer the first solution. Unfortunately, such people seem to be in charge of policy in the US and the UK.
Many other European countries are adopting the second solution. Work-sharing schemes, in many different forms, are becoming the norm in Holland and Denmark, and have made inroads in France and Germany.
The key element in any such approach is to separate work from income. A Danish law enacted in 1993 recognises a right to work discontinuously, while also recognising people's right to a continuous income. It allows employees to choose a "sabbatical" year, which could be divided into shorter periods, every four or seven years.
Unemployed people would take the place of those on leave, who, for their part, would receive 70 per cent of the unemployment benefit they would get if they lost their jobs (typically, 90 per cent of one's salary).
The idea of a universal basic income, paid to all citizens, independent of their position in the labour market, is a logical next step.
This will not be everyone's cup of tea. And, as I suggested above, all schemes aimed at easing the burden of work and increasing the amount of leisure risk falling victim to our genius for conjuring up new disasters.
After all, both capitalism and economics need scarcity to justify their existence, and will not give it up readily.
Robert Skidelsky, a member of Britain's House of Lords, is professor emeritus of political economy at Warwick University