Many commentators aren't just pessimistic abour our current growth prospects. It is becoming increasingly popular to suggest that lower growth more generally might be the new normal, because productivity growth – driven by innovation – is slowing down. An article by Martin Wolf in Wednesday’s FT seemed sympathetic to the recent thesis of Professor Robert Gordon. Whilst Gordon’s theory isn’t as extreme as that advocated by green ideologues (i.e. that growth is unsustainable, the end of it should be welcomed and we should redistribute current wealth), it suggests that we should expect much slower growth going forward because innovation growth today, i.e. the growth of significant general purpose productivity-improving new ideas, is slowing down. Gordon suggests that the ‘third industrial revolution’, i.e. the communication and computer age, has produced much smaller improvements in productivity than those seen in the ‘second industrial revolution’, when huge advances in transport and energy technologies, but perhaps more importantly other innovations like domestic running water, transformed life in the West.
Gordon posits a thought experiment to emphasise his point. If we were offered the choice of having all the technologies up to 2002, or all of the technologies pre- and post-2002 except for running domestic water, then which would we choose? Most people would say the former with little thought – the ability to use running water had a much bigger effect on our lives. This is therefore taken as evidence that the importance of new innovations is shrinking, and as such, the extent of influence of new innovations on productivity is slowing.
Gordon concludes that slowing innovation growth coupled with significant economic headwinds, like ageing populations, poor educational attainment, environmental regulation and taxes, and debt overhangs will inevitably lead to lower growth for the US. Given that many of these same challenges face the UK, we should all be worried. Is this slow growth the new normal?
It’s certainly clear that innovations, that is new ideas, drive improvements in productivity – the long-run determinant of growth of GDP per capita. And Gordon is certainly right that growth shouldn’t be taken for granted – which is why the attitude of our political class is so complacent.
But overall it’s difficult to find his pessimistic conclusions persuasive, for several reasons.
First, as Gordon himself acknowledges, it is impossible to predict the future. By definition, we simply do not know what the future innovations will be, and history tells us it would be wrong to presume we are anywhere near the outer frontier of possibility today. In several of the examples in his paper he seems to neglect the potential future impact of current technologies. For example, Gordon suggests that travel speed has plateaued because planes now travel at the same speed as they did thirty years ago. But in terms of working, people travel for a reason. With video conferencing, non-essential meeting travel times have been reduced to zero. It’s simply wrong to suggest that advances since the jet plane have not been made, and with the potential to harness huge network effects set to be brought about via communication technologies, it does seem overly pessimistic that long-run growth should stagnate so significantly.
But given we can’t predict the future either, we of course don’t know that Gordon’s wrong.
Gordon’s main example, however, seems misleading. Of course we wouldn’t choose to swap running water for iPads, because running water was an innovation which helped meet basic human need. Most innovation, after all, entails standing on the shoulders of previous advances. But if given the choice of having to lose one of the internet (a later innovation) or jet planes, I think the choice would be much tougher than Gordon’s own thought experiment. In other words, it’s not at all clear that the passage of history shows future innovations are likely to have a diminishing impact on productivity.
As Professor John Cochrane of the University of Chicago notes, the other things that Gordon lists as headwinds to future growth seems to confuse ‘level’ effects with ‘growth’ effects – that is, things that make us less well-off compared to others with how fast new ideas more generally are implemented. What’s more, he fails to realise that some of those level effects are self-inflicted wounds: illiberal immigration policies, not following economically sound energy policies and running poor quality education systems.
Towards the end of his piece, in a paragraph about what Gordon’s analysis tells us, Martin Wolf claims ‘growth is not just a product of incentives’ and that lower tax rates would not influence the growth rate of new ideas. The link provided suggests the clear target here is the free-market influenced polices of the Romney-Ryan ticket in the US.
But is it really impossible that high-tax, high-regulated, big government economics might not just affect levels, but actually does affect the growth of new ideas?
Think of it this way. The traditional mechanism through which capitalism has generated sustained innovation has been through the magic ingredient of competition or contestability. In order to maintain profitability, firms have to continually develop new products, discover niches, or produce additional features. This is particularly true in a globalised world. Being able to make a new innovation which completely solves a human need or want therefore has the potential to generate a huge payoff for the individual or firm involved.
It comes as no surprise, therefore, that the areas which Wolf cites as being devoid of innovation in recent years seem to be industries which have been the most heavily regulated and devoid of market pressures, with high levels of government spending or subsidy and often crony-capitalist relationships between government and firms. What if the slowdown in innovation for big general purpose technologies which solve human need are precisely because governments have become more involved in attempting to solve the problems?
Indeed, the elephant in the room in the whole of Gordon’s paper is the effect the huge growth of government is likely to have on both the incentive and likelihood at innovation. Government simply cannot identify what the new innovations that will drive prosperity are going to be – but it can certainly harm the incentive to find them, divert resources into traditional technologies which limits the search for new ones, reduce the rewards for finding them through punitive taxes on success, or reduce the speed of diffusion of new ideas through misguided protectionist policies for trade and education. General efforts to liberalise the economies, fiscally and the supply-side, are therefore more likely to be conducive to conditions for high levels of innovation.
Consider UK government policy. Whilst we can’t talk about future unknown ideas, we can talk about things we do know. The UK government, at a time where the Chinese middle-class is growing is both restricting the freedom of the higher education sector, which has the potential to be a huge export market, and is calling for an overt ‘industrial strategy’ to improve the manufacturing sector, even though all economic evidence suggests demand for services increases at a much faster pace than manufactured goods as income increases. The second is likely to have a significant level effect for the UK – as Chinese incomes rise, our decision to divert more resources into manufactured goods is likely to be misguided.
But the university policy is also likely to slow the growth of new ideas. Universities are often the hubs of innovation, and we would imagine the chance of innovation is highest where the brightest students, whoever they may be, attend the best universities. Government policies across the Western world which prevents this matching can therefore have an effect on the growth of innovation at the frontier.
Energy is also vital. But while the US has taken a lead on new methods of extraction through hydraulic fracking for shale gas, for example, the UK government and others are making energy more expensive and seem hesitant about adopting new technologies.
Unfortunately, of course, it’s difficult to provide direct evidence of any damaging effects of big government on innovation precisely because any foregone innovation is unseen. But an overly pessimistic outlook because of a current slowdown of productivity-enhancing technologies provides a twin danger: that we both ignore the potential that this is the case, and furthermore use a leap of faith that growth will be permanently slower to introduce further policies which damage the mechanisms by which capitalism has generated an unmatched explosion of wealth and prosperity.