Estimating disguised unemployment in major middle-income countries by means of non-linear input–output analysis, 2000–2014

Abstract

According to the disguised-unemployment hypothesis, significant wage differences between sectors in less-developed countries result from segmented labour markets and overcrowding of the flexible market segment. So stated, this hypothesis implies a way to measure non-open unemployment: by the amount of labour that must be withdrawn from the market for relative wages to change. Indeed, it is possible to undertake the exercise of comparing the actual employment of a country with a simulated ‘non-dualistic’ employment by means of a non-linear input-output model and taking the US wage structure as a benchmark. This simulation experiment was carried out for seven middle-income countries (Brazil, China, Indonesia, India, Russia, Mexico, and Turkey) using data from the 2016 Release of the World Input–Output Database. The results of the study are consistent with the disguised-unemployment hypothesis, as well as with related literature.

KEYWORDS:

• Disguised unemployment
• dualistic labour markets
• middle-income countries
• non-linear input–output analysis

Acknowledgments

The author thanks the three anonymous referees and the editor for their valuable comments, and also thanks the International Input-Output Association for acknowledging his work.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 It should be noted that the disguised-unemployment hypothesis is not the only hypothesis explaining the great differences observed in the marginal product of labour in less-developed countries. Actually, some scholars have proposed that a phenomenon like this may be due to unobserved differences either in the skills of workers or in the disutility of different types of work (Gindling, 1991; Magnac, 1991; Maloney, 1999, 2004; Pratap & Quintin, 2006; Rosenzweig, 1988). This alternative hypothesis is not considered in the present study.

2 Even today, there is frequent confusion in the literature between the general notion of disguised unemployment and the specific notion of surplus labour, to the point that almost all studies that refer to disguised unemployment actually refer to surplus labour.

3 It should be pointed out that the notions of disguised unemployment and informal unemployment do not necessarily imply each other. Informal employment is often characterized by its legal status, referring to all economic activities of workers that, in law or practice, are not covered or insufficiently covered by formal agreements. Conversely, a worker is said to be in a state of disguised unemployment if his/her marginal product is less than the marginal product of another worker with the same skills. In this way, a worker may be in a state of disguised unemployment even when his/her economic activities are covered by formal agreements (e.g. state-subsidized disguised unemployment). Furthermore, it is possible that the marginal product of a worker in the informal sector is similar to that of another worker with the same skills in the formal sector (e.g. working ‘under the table’ to avoid taxes).

4 Actually, J. Robinson (1936) pointed out that ‘the phenomenon of disguised unemployment may be regarded as a special case of a change in relative wages’. Morishima and Murata (1971) mathematically developed this idea and rigorously formulated a definition of disguised unemployment based on Robinson’s insight.

5 These are the seven largest middle-income countries in terms of GDP. However, this sample is more heterogeneous in terms of GDP per capita than an alternative sample composed of the seven largest high-income countries in terms of GDP (the so-called ‘G-7’: Canada, France, Germany, Italy, Japan, UK and the US). Specifically, during the period 2000–2014 the coefficient of variation of GDP per capita for the G-7 was less than 0.15, while for the sample considered in the present study the coefficient of variation of GDP per capita is around 0.50.

6 Non-competitive imports are defined as imports of those commodities which can be produced in the country, but only at a much higher price than the world market price after tariffs and transport costs.

7 Analysing the case of the input–output table for the Spanish economy, Fernández-Vázquez (2015) found no statistical evidence to reject the null hypothesis ${\beta }_{ij}=1$ for almost all sectors. Also, Prof. Fernández found ${\delta }_j$ to differ significantly from one, obtaining ${\delta }_j<1$ in all sectors considered. These findings can be considered as evidence in favour of the simplification made here.