Why Developing Asia Should Not Fear Technological Progress
Even if someday robots are able to do most jobs, the decision to deploy machines will still rest with humans.
It is widely acknowledged that new technologies play a critical role in making economies more productive and creating better paying jobs. Indeed, throughout developing Asia, high-yielding crop varieties in agriculture, modern machine tools in manufacturing, and new information and communications technologies in services have done much to support food security, dramatically curb poverty, and improve labor market conditions.
Yet, there is growing concern about the potential downsides of new technology. In particular, the growing sophistication of robotics and artificial intelligence raises the possibility that an unprecedented degree of automation may render many jobs obsolete.
In textile and footwear manufacturing, for example, companies are testing “worker-less” factories, using completely automated production. In services, it is becoming technically feasible to automate even complex tasks such as customer support, book-keeping, and financial analysis.
How will these new technologies impact developing Asia’s ability to generate more and better jobs?
In contrast to some studies suggesting that automation could result in up to 50% unemployment, our new report How Technology Impacts Jobs argues that just as today’s advanced economies were able to negotiate the job displacement that accompanied the introduction of new technologies in certain sectors and occupations, so too will the economies of developing Asia.
We offer several reasons for this. A key one is that the productivity gains that new technologies generate often trigger a chain reaction that creates more jobs than are lost.
Consider automation resulting from new technology. While it reduces the number of workers required to produce a given level of output, the story does not end here. Productivity gains from automation typically reduce the costs of production; under fairly general conditions prices fall, spurring an increase in demand. The latter may be strong enough to even expand the number of jobs in factories that automate part of their production process.
More generally, productivity gains from new technology in one industry lower production costs in downstream industries through input–output channels, contributing to increased demand and employment across industries. Higher demand and more production in one industry raises demand for other industries as well.
Historically, this partly explains why automation in the production of textiles in the US was accompanied by a rise in employment, along with new jobs in upstream and downstream industries.
As the work of American economist James Bessen shows, US textile productivity soared with the introduction in 1814 of power looms. Rather than cause a reduction in employment, the price of textiles and textile products declined, raising demand so much that employment increased. It was only after many decades, once demand for textile products started to get saturated, that employment began to decline.
Is a similar process underway in developing Asia? Since long time-series on employment by sector in the region are scarce, our report uses the Asian Development Bank’s multi-regional input-output tables and employment data from 2005-2015 to show that in 12 developing Asian economies (accounting for 90% of the region’s total employment), rising demand more than compensates for jobs displacement on account of rising productivity.
The structural decomposition analysis of employment changes shows higher productivity is associated with a notional 66% decrease in employment, equal to 101 million jobs annually, for a fixed level of output.
However, concurrently higher demand for goods and services more than offsets this, with a notional 88% increase in employment, equal to 134 million jobs annually associated with rising incomes.
Of course, this is not to deny the possibility that some time in the future technology may become so advanced that most jobs could be carried out more effectively and at lower cost by machines than by humans. But, even if that day arrives sooner than anticipated, the decision to deploy machines will still rest with humans.
If new occupations in which humans have the comparative advantage do not arise, and this lead to mass unemployment, society always has the familiar tools of taxes and subsidies to alter the calculus of employing machines instead of humans.
Whichever way one looks at it, the future is ours to control.