Technology and people: The great job-creating machine

In a world of driverless cars, self-service checkouts and online banking, will there be any jobs left for humans?

In a paper co-authored with my colleagues Debapratim De and Alex Cole, we mapped the historical relationship between jobs and technology by analysing census data from England and Wales since 1871. The data offers powerful reassurance that the ongoing wave of technological innovation will not render humans obsolete.

The paper was shortlisted for the Society of Business Economists' annual Rybczynski Prize and can be found here.

Last week, The Guardian also wrote an article on our paper which has generated a lot of interest and debate.

This week's Monday Briefing summaries our key findings.

On the tech and jobs debate we are unabashed optimists. The role played by technology in boosting employment often goes overlooked because of its more conspicuous destructive effects. This has created a skewed policy debate, one which underplays the complementary nature of the relationship between technology and labour.

We are all familiar with the destructive side of this story, one in which technology replaces people, raising productivity and lowering prices. Agriculture was the first major sector to feel this process. In 1871 it accounted for 6.6% of the workforce of England and Wales. Today it stands at 0.2%, a 95% decline. Since the mid twentieth century manufacturing has gone through the same process, with employment falling from 38% of the workforce to 8% today.

In dynamic, growing economies, this is inevitable. A recent Deloitte Report, Agiletown: the relentless march of technology and London's response, co-authored with Oxford academics Michael Osborne and Carl Frey, shows that in the UK this process is currently underway with 35% of jobs at high risk of being automated over the next two decades.

The flip side of this destruction is productivity growth which is the main driver of welfare and living standards. This process started with the Industrial Revolution and over time it has destroyed routine jobs, both cognitive and manual. In most cases, this destruction has meant that fewer humans are deployed as sources of muscle power and more are employed in jobs involving the nursing and care of others. Technology has been saving us from dull, repetitive and dangerous work.

Take washers and launderers, for example. In 1901, in a population in England and Wales of 32.5 million, 200,000 people were engaged in washing clothes. By 2011, with a population of 56.1 million just 35,000 people worked in the sector, most in launderettes or commercial laundries.

While this destructive side is clearly observed the way in which technology and innovation create jobs tends to command less attention. Our paper identifies the following three channels through which innovation has created jobs since 1871.

First, and most obviously, jobs grow in the burgeoning technology sectors. In the last 35 years, the number of information technology managers working in the UK has risen by a factor of 6.5, to over 327,000; and the number of programmers and software development professionals has risen by a factor of almost three, to 274,160.

This is a dynamic process and, in time, technologies also become obsolete and are supplanted. The census data show that the number of people employed as telephone and telegraph operators rose by a factor of forty in the 100 years to 1971. Since then employment has shrunk as automated switchboards, the internet and mobile telephony have taken off.

Technology and labour are often seen as being alternatives, and therefore in competition. But in many sectors they are highly complementary. Demand for specialist services such as medicine, business and professional services, marketing, design and education have increased as incomes have risen. The application of technology to these sectors has raised productivity and improved outcomes. The nature of work has become more professional, specialised and less routine; and employment has soared. This is the second channel through which technology creates work.

Medicine, a highly knowledge-intensive sector, illustrates this process. Until the reforms of Florence Nightingale and others in the second half of the nineteenth century nursing was widely seen as a low-status, low-skilled role, closer to domestic service than medicine. The 1871 census recorded only 28,000 "sick and invalid" nurses in England and Wales. By 2011 there were almost three-quarters of a million nurses in England and Wales. Today, nursing is a graduate-only profession.

In a more knowledge-intensive economy, the demand for services which help customers benefit from improvements in specialist knowledge seems to soar. Accountancy is one such service. In 1871, there were 9,832 accountants in England and Wales. Much of the work that accountants performed then such as record keeping and calculating is now performed by machines. Freed of routine tasks, accountants have taken on a wide range of more complex, advisory roles.

Far from destroying jobs, technology has helped transform the nature of accountancy and increased demand for its services. The 2011 Census records 215,678 accountants in England and Wales, a twenty-fold increase in the last 140 years.

The third channel by which technology creates work is the most indirect and subtle. Innovation makes goods and services cheaper and, in doing so, increases consumers' disposable incomes, raising demand for new products and services and creating new jobs.

Better productivity and growing international trade has shrunk the share of food in the retail price index basket from 34.8% in 1950 to 11.4% in 2014. Similar forces have reduced the share of clothing and footwear costs in the spending basket from 9.7% to 4.5% over the same time. The real price of cars has halved in the UK in the last 25 years, and its share in the RPI basket has almost halved.

The resultant rise in disposable incomes has enabled consumers to spend more on personal services, such as grooming, driving, for example, the employment of hairdressers. In 1871, there was one hairdresser/barber for every 1793 citizens of England and Wales; today there is one for every 287. Another profession where this effect is visible is bartending, where, despite the decline in the traditional pub, the number of people employed in bars has risen four fold in the 60 years to 2011.

The stock of work in the economy is not fixed. The last 144 years demonstrate that when a machine replaces a human the result, paradoxically, is faster growth and, in time, rising employment. Indeed, one only need look at the UK's employment figures, which have more than doubled since 1871, to see that technological change has coincided with the creation, not destruction, of work.

Machines will continue to reduce prices, democratising what was once the preserve of the affluent and furnishing the income for increased spending and jobs in new and existing areas.

We cannot predict how today's technology will shape tomorrow's society, or which goods and services its citizens will demand. Who, for instance, could have predicted 60 years ago the role coffee shops, gyms and mobile telephony would play in our lives in the early twenty-first century? Nonetheless, it is not hard to think of pressing, unmet need even in the rich world: the care of the elderly and the frail, lifetime education and retraining, health care, physical and mental well-being.

History points to two challenges in navigating this new world. If the pace of adoption of technology is accelerating, society will need to prepare for higher levels of technological unemployment. And the way in which change increasingly rewards high-level education and skills suggests that income inequality may yet widen. Education, training and the distribution of income are likely to be central to the political debate for many years to come.