How do economists measure productivity

How do we even measure our productivity gains?

The current weak productivity gains are surprising given the large-scale technological changes that are expected in the context of the digital revolution. Of course, new technologies first need a certain amount of time before they are fully productive. The companies initially incur set-up costs, learning effects and jobs which, from a business point of view, do not yet lead to any turnover. In the initial phases of a new technological age, weak productivity gains are initially to be accepted. There is a lack of complementary factors (such as organizational capital) and possible production externalities such as an increase in competence or knowledge in other companies. This is also shown by historical examples such as electrification. On the contrary, it is argued that the weaker productivity growth reflects the absence of significant technological innovations. In order to achieve technological progress nowadays, one must invest in research and development to a much greater extent than in the past. Accordingly, the overall economic weakness in productivity reflects declining productivity gains in research.

The search for the appropriate yardstick

Could measurement problems also be a reason for the statistically proven slowdown in productivity? This thesis is not new. Rather, there is the general statistical problem of depicting structural change and the associated changes on the goods and factor level adequately and promptly. This measurement problem may have been exacerbated with digitalization, which is increasingly permeating economic life. This would then not only have corresponding effects on the level measurement of overall economic output, but also on its dynamics.

Where can potential measurement problems arise?

To investigate this, the definition of productivity can be helpful as a starting point. Productivity is generally a performance indicator in which the result of economic activities (output) is related to the underlying production factors (inputs). A corresponding productivity can be shown for each selected input. This applies in the business context as well as in the macroeconomic context. This empirically raises the question of whether output and input are defined and measured appropriately and in a timely manner.

In macroeconomic growth models, macroeconomic economic output is usually created using the three production factors of labor, capital and technical knowledge. If work is understood more as a physical contribution, then its qualitative dimension, i.e. the so-called human capital, is interpreted as its own determinant or, in a simplistic way, as technical progress. The same applies to the environment and natural raw materials if these are not explicitly booked under the capital factor. An improvement in the institutional framework or an intensification of the international division of labor over trade and capital are also to be interpreted as technical progress. The technical knowledge, which is often broadly delimited in the growth calculation, thus has the character of a residual. It includes all increases in production and productivity that do not result from changes in the explicitly defined production factors such as labor and capital.

A major advantage of this measurement approach is that the data on labor and capital input is based on internationally agreed classifications and measurement methods. This is what enables comparisons between countries. The information on macroeconomic economic output and the factors labor and capital as well as their weighting factors (usually the income shares) are mostly based on data from the national accounts and the definitions on which this set of calculations is based.

The right amount is crucial

Ultimately, the growth empiricism shows that the actual macroeconomic growth in added value is not only determined by the use of labor and capital. The difference between actual economic growth and the growth contributions of the explicit factors labor and capital is referred to as the growth in total factor productivity (TFP), tracing back to Robert Solow. This residual thus includes all increases in production or productivity that do not result from changes in the production factors labor and capital, but from all other changes in economic life. But all errors in the measurement of the two explicit production factors labor and capital are ultimately included in the residual and in TFP growth.

How we perceive our economic life and what economic policy conclusions result from it depends on the underlying statistical methods and classifications. This applies to all productivity indicators. In addition to the peculiarities of the definition, this also reflects the statistical limitations, for example resulting from model estimates.

To the guest post on inclusive-productivity.de