Since the global financial crisis of 2008, aggregate productivity growth has been sluggish in many developed and developing countries; however, there are stark differences across economies. Understanding the reasons behind this trend is essential for policy-makers.
In collaboration with the National Institute of Economic and Social Research, this new report from Cambridge Industrial Innovation Policy contributes to the understanding of how sectors account for aggregate productivity gains and losses and how this differs across economies.
The report looks at sector-level data in a sample of eight economies, at different stages of development, which account for over half of the world’s economic output: China, France, Germany, Korea, Taiwan, Singapore, the United Kingdom and the United States.
Six takeaway messages emerge from this report:
- While some variations across economies exist, sectors such as finance, mining, information and communication, and manufacturing tend to have above-average levels of productivity and to experience faster productivity growth. With the exception of mining, we find that these sectors also tend to contribute the most to aggregate productivity growth.
The magnitude of their contributions, however, depends on their relative size across economies. This basic, often overlooked, point has important implications for the way in which we analyse productivity growth. The rate at which national productivity grows is determined by the combined performance of individual sectors of the economy. An economy’s aggregate productivity grows not only when its sectors become more productive but also when the participation of sectors with above-average levels of productivity increases.
- In the last two decades aggregate labour productivity growth has been largely explained by the labour productivity growth that takes place within sectors. However, we also find considerable effects on aggregate productivity growth as a result of ongoing structural change. When looking at the market economy only – that is, excluding sectors such as real estate, education and public administration – the growth of the participation of sectors with below-average productivity levels, at the expense of above-average productivity sectors, has more than halved the UK’s overall productivity growth.
- Manufacturing is one of the sectors with the fastest labour productivity growth among the economies analysed during the 1998–2017 period. Although deindustrialisation was a dominant trend in this period, the manufacturing sector contributed significantly to productivity growth in economies where it accounts for over 20% of GDP: almost half of the aggregate labour productivity growth in Taiwan; roughly a third in Korea and China; and a quarter in Germany.
In contrast, manufacturing has made a negative contribution to productivity growth in the UK and US, where the participation of manufacturing in the economy has been the lowest among the economies analysed (10% and 11% of total output in 2017, respectively). In particular, the loss of manufacturing has imposed a penalty on UK productivity growth of three-quarters of a percentage point, on average, each year for the last two decades.
- The shrinking of manufacturing has gone hand in hand with the expansion of service activities. Service sectors whose contribution to aggregate productivity growth has increased over the last two decades in the economies analysed include both activities with productivity levels that are above average, such as financial and insurance activities and professional, scientific and technical activities, and more labour-intensive activities with below-average productivity levels, such as wholesale and retail trade, human health and social work activities, and administrative and support services.
- Productivity is widely considered to be one of the major determinants of prosperity and is high on the agenda of many countries. Restricting policy analyses to aggregate productivity growth across the whole economy risks overlooking the variation between sectors within national economies and limits the evidence available to policy-makers. Designing effective policy interventions requires detailed information of sector-specific characteristics, including: factors constraining productivity growth, high-growth sub-sectors, the potential for adopting productivity-enhancing technologies, and interdependencies across sectors.
- Because productivity measures are based on value added measures, they suffer from similar limitations to those found in the measurement of GDP, from how value is mainly determined by market transactions, and the related undervaluing of non-market activities, to not accounting for the environmental and social costs of production. Although different methods have been developed to adjust the productivity measures of non-market services to quality, these are far from perfect and tend to underestimate the non-market value of the contribution of these sectors to society.
Policy-makers should exercise caution when using productivity measures to monitor the performance of predominantly non-market sectors such as public administration and healthcare. Confusing productivity measures with efficiency metrics could lead to drawing the wrong conclusions about, for example, the optimal size of a sector or the adequate wage levels.
In addition to the summary and full reports, eight economy-specific studies have been produced for