Jobless Growth

Formation of the concept

Jobless growth names the phenomenon of a macroeconomy growing while employment holds steady or falls. The expression was coined by the American economist Nick Perna in the early 1990s. After the recession of 1990 and 1991 output recovered while the recovery in employment lagged, and the term was intended to explain that gap.

Three causes are offered. First, gains in labour productivity through automation and investment in information technology, so that less labour is needed for the same output. Second, change in industrial structure, as the centre of growth moves from sectors with high employment absorption to sectors with low absorption. Third, change in labour market structure, as firms meet a recovery with non-regular employment and outsourcing rather than regular hiring.

Use in South Korea

In South Korea the concept has served as a critical analytical tool applied to the export-led growth structure. Where capital-intensive and technology-intensive sectors such as semiconductors lead growth, the connection between the growth rate and employment growth weakens. A long-run decline in manufacturing employment has been observed together with movement into low-wage service jobs.

The diagnosis functions as a rejoinder to growth discourse. If it does not hold that a recovery in the growth rate leads to improvement in employment, it follows that policy aimed at aggregate growth alone cannot address the employment problem. The argument has been used as grounds for alternatives such as reducing working time, expanding social service sectors, and expanding public employment.

Debate

There are objections. One concerns measurement: the quality and the hours of employment must be considered alongside its quantity, and reading a reduction in working hours as jobless growth is an error. Another concerns the direction of causation: weak employment may originate in deficient aggregate demand or in labour market institutions rather than in technology.

The argument has revived as the adoption of artificial intelligence spreads. The point at issue is the relation between the speed at which automation displaces particular tasks and the speed at which new tasks are created, and judgements differ on whether technological change reduces aggregate employment or alters distribution and the composition of work. The question is not separable from the distributional problem of who receives the gains from rising productivity.