"You can see the computer age everywhere but in the productivity statistics" Thus quipped the Nobel Prize winning economist Robert Solow in the late eighties – 1987 for the sake of accuracy – while talking about what was later to become known as the Solow paradox. Between the seventies and eighties, computers had advanced enormously, their computing power having increased by hundreds of times in the space of ten years. Why, then, had productivity increased by just two or three percentage points?
While economists debated the matter, time passed, and in the nineties productivity made up for lost time, growing considerably over the following decade. Whether that was down to computers or perhaps globalisation is still open for discussion. What is certain is that this wave of strong productivity growth brought with it another, perhaps even more compelling paradox, known as big decoupling. Essentially it is this: why, especially in the United States, was increased productivity coupled with a fall in wages and employment?
One answer was recently put forward by Martin Ford in his essay “A Jobless Future”, published recently in Italy, in which Ford states his concerns about “technology displacing humans. We are only now waking up to this fact because it is becoming huge and progressing beyond our control". In his opinion, this trend will simply keep deteriorating: "Machines are beginning to think. Yes, they can handle everything; but it is in their head that something big is happening".
If it were just a matter of machines replacing people, that would be simple. For example, the knowledge economy has marginal costs which tend to zero. In a perfect competitive market – which is an abstract model, but an important benchmark all the same – this phenomenon would bring the price to zero; a dominant trend, if we think about Google and its mail hosting and translation services, for example. Added to this is the fact that globalisation, with its low cost labour markets, has further exacerbated this trend: what machines don't do is done by people who cost much less than their Western counterparts. Not to mention market concentration at a global level or problems related to intellectual property. The latter, from file sharing to free software, has had a deflationary impact, on the one hand increasing the diffusion of knowledge and on the other contributing to eroding the bargaining power of some workers and the purchasing power of their salaries. Some workers, that is: because the second perverse effect of this second paradox is the polarisation of the labour market, which tends to reward employees with greater skills and penalise those with fewer ones. In other words, the flat wage curve conceals a gigantic polarisation. High wages have become very high. Low wages have remained low. And average wages have also become low.
Is it possible to ensure that the increased computing power of hardware and software is reflected in an increase in productivity?
The great question of the new digital economy depends on the solution of these two paradoxes. Is it possible to ensure that the increased computing power of hardware and software is reflected in an increase in productivity – contrary to what still happens in Italy, where productivity has remained stagnant for at least twenty years? And is it possible to ensure that an increase in productivity does not result in reduced wages and increased inequalities?
The solutions are on the agenda. The first is called minimum income, or negative income tax. The second is called public (or private) wage insurance – actually put forward by Solow – which some think could be financed by a small levy on every purchase. The third is called continuous training: as the world moves fast, workers must adapt quickly, continuing to learn new things throughout their whole career, perhaps through courses organised by the very company they work for. The debate is open. But undoubtedly, that's (almost) all there is to it.