The cost of innovation has risen, and productivity has suffered13 October, 2017 / Articles
Were there far fewer undiscovered ideas out there than in our more primitive past, how would people know? This is not an idle question; decoding the mysteries of nature, from atmospheric pressure to electricity to DNA, allowed people to bend the natural world to their will, and to grow richer in the process. A dwindling stock of discoverable insights would mean correspondingly less scope for progress in the future—a dismal prospect. And some signs suggest that the well of our imagination has run dry. Though ever more researchers are digging for insights, according to new research, the flow of new ideas is flagging. But that uncovering new ideas is a struggle does not mean that humanity is near the limits of its understanding.
The development of new ideas—meaning scientific truths or clever inventions—allows economies to grow richer year after year. Adding more workers or machinery to an economy boosts GDP, but only for a while. Applying ever more men with hoes to the cultivation of a field will generate diminishing returns in terms of crop yields, for instance; wringing more from the soil eventually requires the use of better seed-stock or fertiliser. Unless humanity finds new ways to do more with the same amount of labour and capital, growth in incomes peters out to nothing.
Dwindling growth in incomes is not a bad description of what has happened in much of the industrialised world in recent decades. Meagre rises, in turn, lead some to conclude that there are simply not many breakthroughs left to be uncovered, of the sort that lifted living standards during the Industrial Revolution. That, for instance, is the view of Robert Gordon, an economist at Northwestern University, whose bleak book, “The Rise and Fall of American Growth”, reckons that the era of economic revolution is behind us.
Is it? A recent paper by Nicholas Bloom, Charles Jones and Michael Webb of Stanford University, and John Van Reenen of the Massachusetts Institute of Technology, provides relevant evidence. Though striking an agnostic position as to whether humanity has used up all its eureka moments, they nonetheless conclude that new ideas are getting more expensive to find. The authors consider four different case studies, within which they compare research “inputs” (such as the money spent on researchers and lab equipment) and outputs. The number of transistors that can be squeezed onto a microchip has doubled with reassuring regularity for half a century, every two years or so—a phenomenon known as Moore’s Law (after Gordon Moore, a founder of Intel). Yet this success has been achieved by pouring more and more resources into the effort over time. The research productivity of each scientist participating in the battle to cram in transistors has correspondingly tumbled.
Much the same is true in other fields of inquiry, such as efforts to raise crop yields and extend life. As the authors acknowledge, squeezing oranges dry is not a problem if new oranges keep arriving: ie, if new lines of research appear even as others are exhausted. Yet they reckon that, across the economy as a whole, the notion that the cost of ideas is rising holds true. Since the 1930s, the effective number of researchers at work has increased by a factor of 23. But annual growth in productivity has declined (see chart).
Not the only fruit
Despair is premature, however. The effort to find new, growth-boosting ideas is not necessarily hopeless, just complicated. Whether herding more researchers into the laboratory raises growth might depend on how intensively the resulting brainstorms are used, for example. Across the global economy, many countries have yet fully to exploit ideas already in use by firms at the frontier of scientific knowledge. The problem, in other words, is not that oranges are in short supply or are already squeezed dry, but rather that of the ten workers at the juice bar, only one has learned to do the squeezing. Investments in education and training, to expand the share of workers that can use new ideas, or in the quality of management, to improve how effectively ideas are applied within firms, would do wonders for growth, even if the world’s scientists are idly scratching their heads.
Analysing the supply side of the innovation equation in isolation can also be misleading. The demand for new ideas, and, correspondingly, the incentive to tackle difficult questions, also matters. In his analysis of the Industrial Revolution, Robert Allen, an economic historian then at Oxford, sought to explain why it started in Britain rather than anywhere else. Supply-side factors, such as improved literacy and stronger property rights, certainly played a part. But it was the demand for labour-saving innovation, prompted by Britain’s relatively high wages at the time, which gave tinkerers a strong incentive to develop and hone the steam engine and its applications.
Put differently, researchers are often like the drunken man searching for his keys under the streetlight, because that is where the light is. Until some pressure is applied to encourage him to look elsewhere, the search will often prove fruitless. It is easy to see why firms might take a lackadaisical approach to some research questions. Disappointing wage growth across advanced economies is a deterrent to the invention and use of labour-saving innovations. Persistently high rates of profit give big firms plenty of money to plough into fancy research labs, but also suggest that the competitive pressures which might prompt them to exploit the resulting discoveries are weak.
The accumulation of knowledge is in some ways a burden. The more is known, the more researchers must absorb before they can add to the stock of human knowledge—or the more they must collaborate with other researchers to combine their areas of expertise. But the incomplete exploitation of currently available knowledge is in some way reassuring. It suggests that people are underperforming relative to their potential: both in how they use available ideas and in how they uncover new ones.