Tech-ing on economic theory
David Smith assesses Bill Gates’ views on whether standard economic theories can be applied to technology.
Normal economic rules, Gates pointed out, rely on an upward-sloping supply curve and a downward-sloping demand curve. If that sounds complicated, it is not. The supply curve slopes up because the higher the price, the more that businesses will be prepared to produce. The demand curve slopes down because the lower the price, the more that people will be prepared to buy the product.
This in turn depends on another economic rule, which is that companies will carry on producing until the marginal cost of each extra unit is equal to the marginal revenue – the amount that the company brings in from selling that unit. In other words, there is no point in carrying on producing beyond the point at which the extra revenue to be gained from sales is less than the additional cost of producing.
So far, so straightforward, and many will remember this from economics lessons. Gates, however, said that software, and other aspects of our technological age, are different. Software has a very high set-up cost – the cost of initially developing it – but a very low marginal cost. Once it has been developed, the cost of producing more units of it can both be extremely low, in some cases non-existent, and also show no tendency to rise over time. There is therefore no natural limit to the amount that can be produced profitably, because even very low marginal revenues will exceed zero or near-zero marginal cost.
It is not just software. Most of us remember a progression in buying music – from relatively expensive-to-produce long-playing records, to cassettes and CDs. Now, as HMV discovered to its cost last Christmas, the most common way of buying music is via downloads, which have very low marginal costs. The same applies to DVDs versus streaming movies. Gates’s software no longer comes in a box and a set of CDs but in the form of a download and an access code. Microsoft Office is typically bought on an annual subscription.
The laws of economics
So has Gates discovered something that changes the laws of economics? No. Economics has always embraced the idea of products with high fixed costs but very low marginal costs. Economies of scale rest on the idea, indeed the practical observation, that the more that is produced, the lower the average cost – and in many cases the lower the marginal cost. For traditional products, this reduction in costs came in the form of automation and repetitive tasks. The belief that specialisation in a particular task lowers costs goes all the way back to 18th century economist Adam Smith. It is also the basis of mass manufacture. It means that the cost of the 1,000th or 10,000th unit will be less than that of the first.
Economists have also identified diseconomies of scale – when production gets to a certain level it may be necessary to invest in a new factory, or pay expensive overtime – but these are less important than economiesof scale. On analysing the Gates argument, it is also worth noting that, while distribution is cheap, costs do not stop. Software developers are constantly working on their products and sending out fixes and updates.
The bigger point is that economists are onto this. In their book Capitalism Without Capital: The Rise of the Intangible Economy, Jonathan Haskel and Stian Westlake describe this very phenomenon. Haskel was appointed last year to the Bank of England’s Monetary Policy Committee.
Intangible investment, they point out, now exceeds tangible investment in plant machinery, buildings and vehicles. They give the example of Bodypump, an exercise programme developed by Les Mills, which is widely used in gyms around the world. Mills, a former New Zealand Olympic track and field athlete, not only sells the programme – the marginal cost of which is low – but makes money out of providing training to instructors.
For Haskel and Westlake, intangible investment should satisfy what they call the four ‘S’s. It should be ‘scalable’, so that once it has been developed it can be easily and cheaply replicated. The second S is ‘sunk’; once an intangible investment has been made, unlike physical assets such as vehicles and buildings, it might be hard to sell on.
The third S is ‘spillovers’. Intangible investment does not just benefit the firm undertaking it. When Apple launched the iPhone, it established the smartphone category, with others following suit. The final S is ‘synergies’, because, as they write, “ideas and other ideas go well together”.
The point is that economics has adapted to the rise of technology and intangibles like software. Nothing said by Gates would have been seen as a serious challenge by most economists.
The tech revolution
And this is not the only area where the tech billionaires have been interacting with economists. The billionaires, including Tesla’s Elon Musk and Facebook’s Mark Zuckerberg, think that the coming tech revolution will have such a profound impact on jobs that it will be necessary to introduce a universal basic income (UBI).
On the face of it they have a point. Credible estimates suggest that up to 35% of current jobs in Britain could be replaced by technology over the next 20 years. That ostensibly suggests the mother and father of an unemployment problem.
Economists have in the past warned about the impact of technology on employment. In his 1930 book Economic Possibilities for Our Grandchildren, John Maynard Keynes wrote that their challenge would be filling their leisure time, as technology and rising productivity was set to reduce the average working week to no more than 15 hours.
Keynes was wrong, and he was wrong for a simple reason. While technology does indeed replace many jobs, it does not reduce the overall demand for labour. The prosperity created by new technology – and in normal circumstances higher productivity – creates new wants and whole sets of new jobs.
So, when we say that many of the jobs that exist now did not exist a quarter of a century ago, when we were in the foothills of the rise of the internet, that has to be right. Many of the jobs that will exist in 25 years from now will include many that we are currently unaware of.
Similarly, most economists regard UBI as a flawed policy. Welfare systems are complicated because people’s needs are. Giving everybody a UBI would be unfair to those with these additional needs, while providing a guaranteed income to those who until now have been perfectly able to earn one. If the UBI were to be anything other than a token amount, it would require very high tax rates to fund it.
These things are never set in stone. It is possible that at some stage technology will evolve so much that there is genuinely no need to employ very many people. That, however, is a very long way off. Economics, meanwhile, would adapt to the change.
About the author
David Smith, economics editor, The Sunday Times.