Abstract
We propose an aggregate measure of employment deprivation among households that follows a methodological framework developed to measure wellbeing. This index verifies a set of reasonable axioms that other available measures do not: increases in three relevant employment deprivation elements-incidence, intensity and inequality. Incidence captures how many households in a population are touched by a lack of employment, while employment deprivation intensity reflects how far households are, on average, from being non-deprived of employment. Finally, employment deprivation inequality increases with the concentration of unemployment among few households. Based on this index, we analyze employment deprivation across the European Union using information from Labor Force Surveys during the current “Great Recession.” Our results provide evidence on the relevance of incorporating the household dimension to identify unemployment profiles, with a variety of implications, in terms of household wellbeing. Specifically, we show that countries with similar (intermediate) unemployment rates differ in their patterns of employment deprivation once the structure of employment across households is incorporated.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
The literature about the penalties of unemployment is profuse (see, for example, Sen 1997). The relation between workless household status and poverty, mental health in the family or children’s wellbeing is also discussed, for example, in Nickell (2004), Dew (1991), Pedersen et al. (2005) and Lindo (2011).
Any empirical analysis using our approach that is restricted to the standard definition of unemployment would be easily extended to a large number of countries in the world. A more comprehensive definition of employment deprivation including underemployment is more data demanding and makes cross-country comparisons more difficult to tackle in a wide range of countries with different Labor Force datasets. However, even in this case, if one is ready to make some further assumptions about the preference for more hours of work, our approach can be straightforwardly used.
Browning et al. (2014) provide an extensive review of the different contributions as a result of considering collective models.
In this same line, Shorrocks (2009b) enhanced the measurement of unemployment accounting for duration.
The normalization of individual employment gaps is not essential in our framework. Non-normalized employment gaps could also be used.
In the empirical exercise, this threshold will be determined directly using the information on desired hours of work reported by each individual. However, in the case one wants to incorporate any specific household labor supply model in which the household total number of working hours is jointly determined (in line with Ashenfelter and Heckman 1974 and subsequent literature), the individually reported number of hours could be corrected accordingly, in order to take into account the joint decision on hours and the degree of substitutability of individual labor supply between different household members.
Underemployment is understood here as including individuals who work less than the usual full-time hours but are willing to work more and are available to do so.
This threshold allows the researcher to set a minimum degree of employment deprivation that a household should have, in order to be considered deprived. In particular, if the value of the threshold is 0.5, a household will be counted as deprived if, for instance, half of all active household members are unemployed or if all active members are employed but work half the number of working hours they wish. This threshold allows the analysis to focus on severe deprivation more straightforwardly.
“First among them is the fact that employment imposes a time structure on the working day. Secondly, employment implies regularly shared experiences and contacts with people outside the family. Thirdly, employment links an individual to goals and purposes which transcend his own. Fourthly, employment defines aspects of status and identity. Finally, employment enforces activity. It is these objective consequences of work in complex industrialized societies which help us to understand the motivation to work beyond earning a living; to understand why work is psychologically supportive, even when conditions are bad, and, by the same token, to understand why unemployment is psychologically destructive” (Jahoda 1979: 423).
Note, however, that we are only measuring involuntary unemployment here; thus, when a person is out of the labor market as the consequence of any within-family agreement, as an inactive person, she will not be considered employment-deprived.
Note, here, that when the individual employment gap is a dichotomous variable, employment deprivation will always be equally distributed among deprived household members.
The index can be computed with the user-written Stata command unemp.
Note that if τ = 1, the index is invariant to different values of γ or α because in this case, by construction, employment deprivation is equally distributed among the deprived, both within and across households (\(g_{ij}^{\gamma } = 1, \, u_{i} = 1\)).
It is important to note that for any α > 1, provided τ < 1, households with a greater employment deprivation index will have a disproportionally larger impact on the aggregate index, as a consequence of the assumed social preference for equality. The higher α is, the larger this impact. The role of τ is to make the contribution of households with small deprivation zero, in order to concentrate exclusively on those who are severely deprived. Recall that the results do not vary with γ or α if τ = 1.
“Eligible households” means that households composed solely of students are excluded.
If we were interested in calculating the share of households in which none of the members are working, we could calculate our index using \(\omega_{i} = \frac{1}{N}\), with N being the number of eligible households. Using different values for τ allows us to change the minimum household employment deprivation level when selecting who is counted within the employment-deprived population.
If \(\omega_{i} = \frac{1}{N}\), H indicates the proportion of employment deprived households (\(H = {\raise0.7ex\hbox{$q$} \!\mathord{\left/ {\vphantom {q N}}\right.\kern-0pt} \!\lower0.7ex\hbox{$N$}}\)).
More specifically, for any \(\alpha > 1\), the Generalized Entropy index is: \(GE_{u}^{\alpha } = {{E_{u}^{\alpha } } \mathord{\left/ {\vphantom {{E_{u}^{\alpha } } {\alpha (\alpha - 1)}}} \right. \kern-0pt} {\alpha (\alpha - 1)}}\).
The Nordic countries (Denmark, Sweden and Finland) lack complete information on households, while Malta’s microdata are currently missing from the European Labor Force Survey data provided by EUROSTAT. Note also that we drop any individuals in the sample who live in institutional households and are not linked to a private household for the countries in which they are sampled (e.g., Germany).
For the case in which only members who are currently within the labor force are considered in the weights, see Gradín et al. (2012b).
These samples contain information on why an individual is in part-time work, in order to classify involuntary part-time workers as deprived. Note that this implies using nationally representative sub-samples for Spain, France, the Netherlands and Germany.
This number of hours is capped at the country’s full-timers’ usual working hours’ mode. This prevents our data from including unreasonably large numbers coming from either outliers or measurement error. In the case of part-timers, the gap is only positive if they wish to work more hours but cannot find a full-time job.
For cases in which the individual reported missing his or her usual hours of work, this value was estimated using the number of hours actually worked during the reference week in the main job.
An individual is considered to be unemployed if even if actively seeking employment, as he or she did not do any work for pay during the reference week and did not have a job or business from which she was absent for some reason during that week. If the individual has found employment to begin in three months but is available for work in the following two weeks she is also classified as unemployed.
Indeed, in Spain, there has been a disproportionally large increase in the proportion of involuntary part-timers (from about one third of the part-time employed right before the crisis to about half in 2010).
Note that the reason for this differs in these two countries. In the Netherlands, less than 5 % of part-timers wish to work more hours, and part-timers account for almost half the employed population. In Spain, on the contrary, part-timers are a smaller proportion of the employed population, 13 %, but nearly half of them want to work more hours.
The results differ, however, with \(\gamma = 0\) because, in this case, the index is sensitive to the proportion of deprived adults in the household, but not to the actual value of their unmet desired hours of work (thus, in this case, the index counts those fully and partially unemployed equally). In particular, in comparison to when \(\gamma = 2\), deprivation increases somewhat in some countries with large underemployment (e.g., Cyprus, Germany and Romania).
Inequality in these countries is even larger in relative terms, if it is measured only among the deprived using the coefficient of variation (more than 40 % above the average).
References
Apps, P. F., & Rees, R. (1996). Labor supply, household production and intra-family welfare distribution. Journal of Public Economics, 60, 199–219.
Apps, P. F., & Rees, R. (1997). Collective labor supply and household production. Journal of Political Economy, 105, 178–190.
Ashenfelter, O., & Heckman, J. (1974). The estimation of income and substitution effects in a model of family labor supply. Econometrica, 42(1), 73–85.
Bargain, O. (2008). Normative evaluation of tax policies: From households to individuals. Journal of Population Economics, 21, 331–371.
Becker, G. S. (1965). A theory of the allocation of time. Economic Journal, 75, 493–517.
Becker, G. S. (1973). A theory of marriage: Part 1. Journal of Political Economy, 81, 813–846.
Becker, G. S. (1974). A theory of marriage: Part 2. Journal of Political Economy, 82, s11–s26.
Blank, R. (1993). Why were poverty rates so high in the 1980s? In D. B. Papadimitriou & E. N. Wolff (Eds.), Poverty and prosperity in the late twentieth century (pp. 21–55). London: Macmillan Press.
Blank, R. (2000). Fighting poverty: Lessons from recent US history. Journal of Economic Perspectives, 14(2), 3–19.
Bonke, J. (2013). Pooling of income and sharing of consumption within households. Online-first in Review of Economics of the Household, doi:10.1007/s11150-013-9184-y.
Borooah, V. K. (2002). A duration-sensitive measure of the unemployment rate: Theory and application. Labour, 16(3), 453–468.
Brenke, K., Rinne, U., & Zimmermann, K. (2011). Short-time work: The German answer to the great recession. IZA Discussion Paper, 5780.
Browning, M., Chiappori, P.-A., & Weiss, Y. (2014). Economics of the family. Forthcoming in Cambridge University Press.
Chakravarty, S. R. (1983). A new index of poverty. Mathematical Social Sciences, 6, 307–313.
Chiappori, P. A. (1988). Rational household labor supply. Econometrica, 56, 63–89.
Chiappori, P. A. (1992). Collective labor supply and welfare. Journal of Political Economy, 100, 437–467.
Chiappori, P. A. (1997). Introducing household production in collective models of labor supply. Journal of Political Economy, 105, 191–209.
Clark, S., Hemming, R., & Ulph, D. (1981). On indices for the measurement of poverty. Economic Journal, 91(362), 526–551.
Clark, K. B., & Summers, L. H. (1979). Labor market dynamics and unemployment: A reconsideration. Brooking Papers on Economic Activity, 1, 13–60.
Deutsch, J., Flükiger, I., & Silber, J. (2008). On various ways of measuring unemployment, with applications to Switzerland. Research on Economic Inequality, 16, 259–284.
Dew, M. A. (1991). Effects of unemployment on mental health in the contemporary family. Behavior Modification, 15(4), 501–544.
European Commission (2010). Employment in Europe. Employment, Social Affairs and Inclusion, Report: http://ec.europa.eu/employment_social/eie/index_en.html.
Foster, J. E., Greer, J., & Thorbecke, E. (1984). A class of decomposability poverty indices. Econometrica, 52, 761–766.
Gradín, C., Cantó, O. & del Río, C. (2012b). Measuring employment deprivation among households. ECINEQ WP2012-247, February.
Gradín, C., del Río, C., & Cantó, O. (2012a). Measuring poverty accounting for time. Review of Income and Wealth, 58(2), 330–354.
Haveman, R., & Schwabish, J. (2000). Has macroeconomic performance regained its antipoverty bite? Contemporary Economic Policy, 18(4), 415–427.
Jahoda, M. (1979). The psychological meanings of unemployment. New Society, 49(883), 492–495.
Jenkins, S. P., & Lambert, P. J. (1997). Three ‘I’s of poverty curves, with an analysis of UK poverty trends. Oxford Economic Papers, 49(3), 317–327.
Lindo, J. M. (2011). Parental job loss and infant health. Journal of Health Economics, 30, 869–879.
Meyer, B. D., & Sullivan, J. X. (2011). Consumption and income poverty over the business cycle. NBER Working Series, 16751.
Nickell, S. (2004). Poverty and worklessness in Britain. The Economic Journal, 114(494), 1–25.
OECD. (2001). Employment outlook 2001. Paris: OECD.
OECD. (2009). Employment outlook 2009: Tackling the jobs crisis. Paris: OECD.
OECD. (2010). Employment outlook 2010: Moving beyond the jobs crisis. Paris: OECD.
Paul, S. (1991). On the measurement of unemployment. Journal of Development Studies, 36, 395–404.
Paul, S. (1992). An illfare approach to the measurement of unemployment. Applied Economics, 24(7), 739–743.
Paul, S. (2001). A welfare loss measure of unemployment with an empirical illustration. The Manchester School, 69(2), 1463–6786.
Pedersen, C. R., Holstein, B. E., & Köhler, L. (2005). Parents’ labour market participation as predictor of children’s well-being: changes from 1984 to 1996 in the Nordic countries. European Journal of Public Health, 15(4), 431–436.
Riese, M., & Brunner, J. K. (1998). Measuring the severity of unemployment. Journal of Economics, 67(2), 167–180.
Sen, A. K. (1976). Poverty: An ordinal approach to measurement. Econometrica, 44, 219–231.
Sen, A. K. (1997). Inequality, unemployment and contemporary Europe. International Labour Review, 136(2), 155–171.
Sengupta, M. (2009). Unemployment duration and the measurement of unemployment. Journal of Economic Inequality, 7(3), 273–294.
Shorrocks, A. (1995). Revisiting the sen poverty index. Econometrica, 63(5), 1225–1230.
Shorrocks, A. (2009a). Spell incidence, spell duration and the measurement of unemployment. Journal of Economic Inequality, 7(3), 295–310.
Shorrocks, A. (2009b). On the measurement of unemployment. Journal of Economic Inequality, 7(3), 311–327.
Thon, D. (1979). On measuring poverty. Review of Income and Wealth, 25(4), 429–439.
Watts, H. W. (1968). An economic definition of poverty. In D. P. Moynihan (Ed.), On understanding poverty. New York: Basic Books.
Weiss, Y. (1997). The formation and dissolution of families: Why marry? Who marries whom? And what happens upon divorce”. In M. R. Rosenzweig, O. Stark (Eds.). Handbook of population and family economics (pp. 81–123). Elsevier Science.
Acknowledgments
We acknowledge financial support from the Spanish Ministerio de Ciencia e Innovación (Grant ECO2010-21668-C03-03/ECON) and Xunta de Galicia (CN2012/178, and Grant 10SEC300023PR).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Gradín, C., Cantó, O. & del Río, C. Measuring employment deprivation in the EU using a household-level index. Rev Econ Household 15, 639–667 (2017). https://doi.org/10.1007/s11150-014-9248-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11150-014-9248-7