Almost nothing in economics is more important, or mysterious, than productivity. It means the amount of stuff - goods, services, economic value - produced for a given amount of input.
It is productivity that separates today's rich, modern consumer societies from subsistence farmers living on the edge of starvation.
The Industrial Revolution created technologies such as electricity, turbines and internal combustion engines that supercharged productivity, which is why our lives are incomparably better than those of our great-great-grandparents.
If productivity slows, we can expect our living standards to stagnate, no matter what other steps we take. The problem is, productivity growth is slowing. After decades of robust growth, it flatlined in the 1970s and 1980s, only to surge ahead again in the 1990s and early 2000s. But for the past decade or so, productivity has stumbled again, and the worry is that the information technology revolution was only a minor, short-lived reprieve from an inexorable stagnation.
This pessimistic case has been made by economists Robert Gordon and Tyler Cowen. As they see it, scientists and engineers have simply picked most of the low-hanging fruits of knowledge available in the universe. A new report by the Organisation for Economic Cooperation and Development (OECD), however, paints a more nuanced picture.
Much of economist Paul Romer's research is about "excludability", or the degree to which companies can stop other companies from learning their secrets. Excludability means new technologies don't necessarily flow from one company to another. Mr Romer has shown that excludability is, at least in theory, important to economic growth.
The organisation looked at productivity not at the global or national level, but at the corporate level. Different companies have different technologies, different management systems and different levels of talent.
What the OECD found was startling. At a small number of companies, productivity growth hasn't slowed at all. If you look at only these "global frontier" companies, there has been no productivity slowdown at all!
This is especially true in services industries. The top performers have blazed ahead, while other companies have stagnated or even become less productive.
Why is this happening? Here we should turn to the research of New York University economist Paul Romer, one of the most influential theorists of economic growth.
Much of Professor Romer's research is about "excludability", or the degree to which companies can stop other companies from learning their secrets. Excludability means that new technologies don't necessarily flow from one company to another. Prof Romer has shown that excludability is, at least in theory, important to economic growth.
If technology has become more excludable - if ideas and technologies are not spreading from company to company the way they used to - then we're in trouble. The OECD report suggests that this is happening, but it doesn't give us a clear answer as to why. In fact, no one really knows.
One possibility is that technology "spillovers" between companies - the term for when methods and practices jump from one company to another - are slowing as globalisation runs its course.
Companies that interact with each other via supply chains will naturally tend to exchange ideas, as each company in the chain sees what the others are doing. The burst of globalisation in the 1990s and early 2000s might have allowed a huge transfer of knowledge along these supply chains; that burst might be coming to an end as the integration of East Asia into the global economy is completed.
Another possibility - one not suggested by the OECD report - is that intellectual property law is making it harder for companies to use ideas developed at other companies. There has been an explosion in the number of patents granted in the United States since the early 1980s. In Japan, the increase has been even more dramatic. Some of the fastest growth has been in patents for business methods - exactly the kind of thing that ought to diffuse across companies and equalise productivity.
In earlier ages, businesses could freely copy each other's way of doing things; now, it is often illegal. Some level of patent protection, of course, is necessary to encourage innovation. But many economists believe that we now give out far too many patents, often for innovations of questionable originality. This is something we would expect to increase the gap between the most productive companies and the rest.
Whatever the reason for the divergence between companies, we need to find it and fix it if we can. The divergence could be affecting a lot more than productivity.
A torrent of research in the past decade suggests that much of the increase in wage inequality in developed countries is due to differences in wages between different companies - work for a good company and you get better pay, work for a bad one and you're out of luck. Fixing the productivity divergence might help us fight inequality as well.
So governments in the developed world should be thinking about how to break down the walls between companies. Our economic future could depend on it.
We have been experiencing some problems with subscriber log-ins and apologise for the inconvenience caused. Until we resolve the issues, subscribers need not log in to access ST Digital articles. But a log-in is still required for our PDFs.