ARTÍCULO / ARTICLE

WANTING MORE OR WANTING LESS? MISMATCHES BETWEEN ACTUAL AND PREFERRED WORKING HOURS IN SPAIN, 2005-2014

¿QUEREMOS MÁS O MENOS? DESENCUENTROS ENTRE LAS HORAS DE TRABAJO QUE TENEMOS Y LAS QUE PREFERIMOS EN ESPAÑA, 2005-2014

Margarita Torre Fernández

Universidad Carlos III de Madrid

margarita.torre@uc3m.es

ORCID iD: http://orcid.org/0000-0002-1051-9717

CONTENTS

INTRODUCTION Top

Researchers and politicians long assumed that hour mismatches were fairly infrequent. It was only in the early 1990s, when Schor (1991Schor, J. B. 1991. The Overworked American: The Unexpected Decline of Leisure. New York: Basic Books.) published her seminal work The Overworked American, that researchers began to look at the mismatches between people’s actual and preferred working hours.

Hour mismatches are caused by social, political and economic forces (Reynolds 2004Reynolds, J. 2004. "When Too Much is Not enough: Actual and Preferred Work Hours in the United States and Abroad". Sociological Forum 19(1):89-120. http://www.jstor.org/stable/4148808.; Reynolds and Aletraris 2006Reynolds, J. and L. Aletraris. 2006. "Pursuing Preferences: The Creation and Resolution of Work Hour Mismatches". American Sociological Review 71(4):618-638. http://dx.doi.org/10.1177/000312240607100405.), and have important consequences. On the one hand, employees who feel overworked are more likely to feel angry and resentful at work and to experience health problems (Galinsky, Kim and Bond 2001Galinsky, E., S. S. Kim and J. T. Bond. 2001. Feeling Overworked: When Work Becomes Too Much. New York: Families and Work Institute.), as well as to face work-family conflicts (Jacobs and Gerson 2000Jacobs, J. A. and K. Gerson. 2000. "Do Americans Fell Overworked? Comparing Ideal and Actual Work Time". Pp. 71-95 in Work and Family: Research Informing Policy, edited by Toby L. Parcel and Daniel B. Cornfield. Sage Publications.; 2004Jacobs, J. A. and K. Gerson. 2004. The Time Divide: Work, Family, and Gender Inequality. Cambridge, MA: Harvard University Press.). On the other hand, those who work shorter hours than they might wish are more likely to feel depressed and suffer from low self-esteem (Dooley, Prause and Ham-Rowbottom 2000Dooley, D., J. Prause and K. A. Ham-Rowbottom. 2000. "Underemployment and Depression: Longitudinal Relationships". Journal of Health and Social Behavior 41:421-436. http://dx.doi.org/10.2307/2676295.). Hour mismatches are also linked to lower levels of organizational performance (Roehling, Roehling and Moen 2001Roehling, P. V., M. V. Roehling and P. Moen. 2001. "The Relationship between Work-Lif Politics and Practices and Employee Loyalty: A Life Course Perspective". Journal of Family and Economic Issues 22:141-170. http://dx.doi.org/10.1023/A:1016630229628.) and may lead to work behaviors that impair productivity, including absenteeism, lateness, use of sick-time and other forms of shirking (Clark 2001Clark, A. E. 2001. "What really Matter in a Job? Hedonic Measurement using Quit Data". Labor Economics 8:223-242. http://dx.doi.org/10.1016/S0927-5371(01)00031-8., Holtom, Tidd and Lee 2002Holtom, B., T. Lee and S. Tidd. 2002. "The Relationship between Work Status Congruence and Work-Related Attitudes and Behaviors". Journal of Applied Psychology 87(5):903-915. http://dx.doi.org/10.1037/0021-9010.87.5.903.). Understanding how mismatches occur and how they can be solved is, therefore, a crucial task for individual, family and organizational life.

The prior literature refers mostly to periods of economic growth, but there has been little research into the effects of recession on working time mismatches (Bell and Blanchflower 2011Bell, D. and D. Blanchflower. 2011. "Young People and the Great Recession". Oxford Review of Economic Policy 27(2):241-267. http://dx.doi.org/10.1093/oxrep/grr011.; Bluestone and Rose 1998Bluestone, B. and S. Rose. 1998. "The Macroeconomics of Work Time". Review of Social Economy 56(4):425-441.). This study contributes to existing research about the ways in which the quality of work has been affected by the economic crisis. The existing research has addressed changes in the intensity of work, job security, job control and skill development (Gallie 2013Gallie, D. (ed.). 2013. Economic Crisis, Quality of Work, and Social Integration: The European Experience. Oxford: Oxford University Press. http://dx.doi.org/10.1093/acprof:oso/9780199664719.001.0001.). Here, I examine hour mismatches in Spain between 2005 and 2014. I begin by asking how the level of hour mismatches has changed over time and how such mismatches are produced. I then test three possible competing predictions for trends in hour mismatches during the recessive economic cycle. The first, the Decreasing mismatches scenario, posits that the selective elimination of low-status jobs in periods of economic crisis will drive reductions in the level of hour mismatches. The second, the Increasing mismatches scenario, anticipates that workers will want more hours than they usually work during recessions to tide them over during the ups and downs in the labor market. Third, the Increasing divide scenario suggests that inequalities between different strata of the labor force will intensify during economic downturns.

The scenarios are tested using individual data from the Spanish Labor Force Survey (LFS) from 2005 to 2014. The survey includes specific questions about preferred, actual and contracted monthly working hours. Specifically, people were asked how many hours per week they were expected to work according to their contracts and how many hours they actually worked at their principal job. They were also asked if they would be willing to work more hours. In addition, the survey includes relevant information about labor conditions and individual attributes. The empirical part of this paper is organized in two main blocks. The first explores trends in the level of working hour mismatches, its origin and its profile. Multivariate analyses are then performed to test to what extent observed patterns are explained by a broad range of individual and labor market settings.

The findings reveal that mismatches increased significantly between 2005 and 2014, especially in recession years. This increase was mainly due to the increase in the number of people wanting to work more hours, while the number of people desiring fewer hours has fallen over time. Also, as predicted by the Increasing divide scenario, mismatches have grown particularly fast among fixed-term, low-status occupations, part-time workers and women. Consequently, inequality between high- and low-status workers had increased by the end of the period. The paper ends with a discussion of the implications of this study for scholarly research, and for organizational, individual and family life. 

HOUR MISMATCHES IN THE LITERATURE Top

Overwork occurs when workers’ desired hours of labor supply are exceeded by hours of labor demanded at their current pay rate. Conversely, underwork affects those who want to work more hours than they actually do. Both of them have different but important consequences for individuals (Reynolds and Aletraris 2006Reynolds, J. and L. Aletraris. 2006. "Pursuing Preferences: The Creation and Resolution of Work Hour Mismatches". American Sociological Review 71(4):618-638. http://dx.doi.org/10.1177/000312240607100405.). In the first place, undesired working hours have consequences for organizational performance. On the one hand, the literature highlights that people working fewer hours than they want to display a lack of affective commitment (Van Emmerik and Sanders 2005Van Emmerik, I.J.H. and K. Sanders. 2005. "Mismatch in Working Hours and Affective Commitment: Differential Relationship for Distinct Employees Groups". Journal of Managerial Psychology 20(8):712-726. http://dx.doi.org/10.1108/02683940510631462.; Roehling, Roehling and Moen 2001Roehling, P. V., M. V. Roehling and P. Moen. 2001. "The Relationship between Work-Lif Politics and Practices and Employee Loyalty: A Life Course Perspective". Journal of Family and Economic Issues 22:141-170. http://dx.doi.org/10.1023/A:1016630229628.) and satisfaction (Clark 2001Clark, A. E. 2001. "What really Matter in a Job? Hedonic Measurement using Quit Data". Labor Economics 8:223-242. http://dx.doi.org/10.1016/S0927-5371(01)00031-8.). In contrast, feeling overworked impairs workers’ performance, raises stress levels and increases the risk of injury (Dembe et al. 2005Dembe, A. E., B. Erickson, R. G. Delbos and S. M. Banks. 2005. "The Impact of Overtime and Long Work Hours on Occupational Injuries and Illness: New Evidence from the United States". Occupational and Environmental Medicine 62:588-597. http://dx.doi.org/10.1136/oem.2004.016667.). Second, time mismatches affect the work/family balance. Too much time at work can undermine personal and family welfare, whereas too little time can endanger economic security and depress living standards (Jacobs and Gerson 2004Jacobs, J. A. and K. Gerson. 2004. The Time Divide: Work, Family, and Gender Inequality. Cambridge, MA: Harvard University Press.). Meanwhile, underwork has been found to be associated with gender inequality. Gender differences in hours of paid work translate into gender differences in experience, earnings and promotion opportunities (Budig and England 2001Budig, M. J. and P. England. 2001. "The Wage Penalty for Motherhood". American Sociological Review 66:204-225. http://dx.doi.org/10.2307/2657415.). Finally, time mismatches have important consequences for personal well-being. In this regard, researchers have found that people working shorter hours than they would like are prone to lower self-esteem (Prause and Dooley 1997Prause, J. and D. Dooley. 1997. "Effect of Underemployment on School-Leavers’ Self-Esteem". Journal of Adolescence 20:243-260. http://dx.doi.org/10.1006/jado.1997.0083.), while those working longer hours than they wish tend to suffer from psychological and physical problems (Galingsky et al. 2001Galinsky, E., S. S. Kim and J. T. Bond. 2001. Feeling Overworked: When Work Becomes Too Much. New York: Families and Work Institute.).

Due to the significant impact that working hours have on individual, family and organizational life, it is crucial to understand how working hour mismatches originate and how they can be solved. From a micro-level perspective, research has revealed that the desire for fewer or more hours of work changes over a person’s lifetime. Studies show that middle-aged workers are the most likely to want shorter hours (Jacobs and Gerson 2000Jacobs, J. A. and K. Gerson. 2000. "Do Americans Fell Overworked? Comparing Ideal and Actual Work Time". Pp. 71-95 in Work and Family: Research Informing Policy, edited by Toby L. Parcel and Daniel B. Cornfield. Sage Publications.; Stewart and Swaffield 1997Stewart, M. and J. Swaffield. 1997. "Constraints on the Desired Hours of Work of British Men". Economic Journal 107(441):520-535. http://dx.doi.org/10.1111/j.0013-0133.1997.175.x.). On the one hand, this is because the middle-aged are those who spend the most time at work, even when all workers have similar preferences. Importantly, middle-aged workers are also more likely to experience conflict between work and family roles (Jacobs and Gerson 2001Jacobs, J. A. and K. Gerson. 2001. "Overworked Individuals Or Overworked Families? Explaining Trends in Work, Leisure and Family Time". Work and Occupations 28:40-63. http://dx.doi.org/10.1177/0730888401028001004.). Family responsibilities tend to reduce the desire and capacity to work more hours, making dual-earner couples and single parent families the groups most likely to feel the squeeze between the demands and rewards of family life (Jacobs and Gerson 2001Jacobs, J. A. and K. Gerson. 2001. "Overworked Individuals Or Overworked Families? Explaining Trends in Work, Leisure and Family Time". Work and Occupations 28:40-63. http://dx.doi.org/10.1177/0730888401028001004.; 2004Jacobs, J. A. and K. Gerson. 2004. The Time Divide: Work, Family, and Gender Inequality. Cambridge, MA: Harvard University Press.). This is particularly so among women, as they continue to do a larger share of household chores (Coltrane 2000Coltrane, S. 2000. "Research on Household Labor: Modeling and Measuring the Social Embeddedness of Routine Family Work". Journal of Marriage and the Family 62:1208-1233.).

From a different perspective, literature in the field of economics has attributed hour mismatches to market failures (Altman 2001Altman, M. 2001. "Preferences and Labor Supply: Casting some Light into the Black Box of Income-Leisure choice". Journal of Socio-Economics 33:199-219. http://dx.doi.org/10.1016/S1053-5357(01)00099-3.; Sousa-Poza and Henneberger 2002Sousa-Poza, A. and F. Henneberger. 2002. "An Empirical Analysis of Working Hours Constraints in Twenty-One Countries". Review of Social Economy 60:209-242. http://dx.doi.org/10.1080/00346760210146235"). The well-established model of labor supply assumes that individuals are able to choose any number of working hours. This does not mean that workers have a free choice of hours with their current employer, but rather that there is a feasible set of jobs for them that provide a free choice of hours (Steward and Swaffield 1997). However, prior research has found that these models are not applicable to all employees (Machin et al. 2003Machin, S., A. Manning and L. Rahman. 2003. "Where the minimum wage bites hard: the introduction of a national minimum wage to a low wage sector". Journal of the European Economic Association 1:154-180.), and has underscored the role of labor market constraints in creating mismatches between individual preferences and behaviors (Steward and Swaffield 1997Stewart, M. and J. Swaffield. 1997. "Constraints on the Desired Hours of Work of British Men". Economic Journal 107(441):520-535. http://dx.doi.org/10.1111/j.0013-0133.1997.175.x.; 2008Stewart, M. and J. Swaffield. 2008. "The other margin: do minimum wages cause working hours adjustments for low-wage workers?". Economica 75:148-167.).

HOUR MISMATCHES IN RECESSION PERIODS. HYPOTHESES Top

Hour mismatches may be related to the business cycle. Despite extensive research on hour mismatches in growth periods, however, the impact of recession on the phenomenon has been largely ignored. Recent research has explored changes in the skill structure of the workforce, the control that employees can exercise over their work, the intensity of work, job security and work-family conflict (Gallie 2013Gallie, D. (ed.). 2013. Economic Crisis, Quality of Work, and Social Integration: The European Experience. Oxford: Oxford University Press. http://dx.doi.org/10.1093/acprof:oso/9780199664719.001.0001.).[1] I build here on prior research and examine changes in the quality of work focusing on workers ability/inability to choose their preferred working hours and the consequences for personal and family well-being.

The existing literature makes different predictions about the way that the quality of work might be affected by economic crisis. The most optimistic voices contend that recession may have a positive impact on work quality. This claim draws on Schumpeter’s ideas (Schumpeter 1970Schumpeter, J. A. 1970. Capitalism, Socialism and Democracy. London: Routledge.), which posited that economic crisis contributes to the eradication of low-status jobs. There are two reasons for this assertion. On the one hand, employers have incentives to benefit highly skilled employees during economic downturns because they have invested more in their skill development and they are more difficult to replace. On the other hand, scholars examining past recessions have found that less skilled employees and those in technologically less sophisticated sectors have tended to be disproportionately vulnerable to unemployment (Gershuny and Marsh 1993Gershuny, J. and C. Marsh. 1993. "Unemployment in Work Histories". Pp. 66-114 in Social Change and the Experience of Unemployment, edited by D. Gallie, C. Marsh and C. Vogler. Oxford: Oxford University Press.; Gallie et al. 1998Gallie, D. et al. 1998. Restructuring the employment relationship. Oxford: Clarendon Press.). To the extent that workers in high-status jobs are less willing to experience hour mismatches (Clarkberg and Moen 2001Clarkberg, M. and P. Moen. 2001. "Understanding the Time -Squeeze: Married Couple’s Preferred and Actual Work-Hour Strategies". American Behavioral Scientist 44:1115-1136. http://dx.doi.org/10.1177/0002764201044007005.), this selective reduction of workers in low-status occupations will cause a progressive drop in the general level of hour mismatches. I refer to this as the Decreasing Mismatches Scenario (S1).

Pessimistic voices, on the contrary, contend that economic crises tend to undermine work quality. In times of recession and rising unemployment, both the fear of redundancy and high levels of unemployment undermine workers’ ability to resist management initiatives. As a consequence, employers are enabled to offer working hours that do not match employees’ preferences. Indeed, in some economic contexts cutting the number of working hours has traditionally been the first response of business to recession, before they move to temporary or permanent lay-offs (Rones 1981Rones, P. L. 1981. "Response to Recession: Reduce Hours Or Jobs?". Monthly Labor Review October.).[2]-[3] Furthermore, average overtime has historically fallen in recessions (Hetrick 2000Hetrick, R. L. 2000. "Analyzing the Recent Upward Surge in Overtime Hours". Monthly Labor Review February 2000.). Altogether, cuts in working hours, lay-offs and increasing vulnerability in economic downturns may result in workers wanting to work more hours than usual so that they can ride out market fluctuations (Bell and Blanchflower 2011Bell, D. and D. Blanchflower. 2011. "Young People and the Great Recession". Oxford Review of Economic Policy 27(2):241-267. http://dx.doi.org/10.1093/oxrep/grr011.). I define this trend as the Increasing mismatches scenario (S2).

Recent research, which has its beginnings in the early 1990s, builds on the last two approaches to argue that economic crisis may accelerate the process of labor market polarization (Kalleberg 2011). Earlier theories of labor market segmentation highlighted the gap existing between employees in a primary labor market who benefited from multiple advantages in terms of employment conditions and those in a secondary labor market who suffered from cumulative disadvantages (Doeringer and Piore 1971Doeringer, P. B. and M. Piore. 1971. Internal labor markets and manpower analysis. Lexington, Mass: Heath Lexington Books.; Kalleberg et al. 1981Kalleberg, A. L., M. Wallace and R. P. Althauser. 1981. "Economic segmentation, worker power and income inequality". The American Journal of Sociology 87(3):651-683. http://dx.doi.org/10.1086/227499.). Later, in the early 1990s, researchers noticed a continued and increasing differentiation in the quality of work between high-status and low-status occupations (Kalleberg 2011; Kalleberg, Reskin and Hudson 2000Kalleberg, A. L., B. Reskin and K. Hudson. 2000. "Bad Jobs in America: Standard and Nonstandard Employment Relations and Job Quality in the United States". American Sociological Review 65:256-278. http://dx.doi.org/10.2307/2657440.). This gap is presumably wider in economic recessions because workers in low-status occupations are a first-choice option for dismissals, making them particularly vulnerable. Consequently, low-status workers will be more likely to want to work longer hours than their relatively more secure higher-status peers in an attempt to both earn more money and to hold onto their jobs (Bluestone and Rose 1997Bluestone, B. and S. Rose. 1997. "Overworked and Underemployed: Unraveling and Economic Enigma". The American Prospect 31:58-69.; 1998Bluestone, B. and S. Rose. 1998. "The Macroeconomics of Work Time". Review of Social Economy 56(4):425-441.). In other words, inequality between high- and low-status workers will increase in recession periods, resulting in what I call the Increasing divide scenario (S3).

I test these alternative scenarios through the study of hour mismatches in Spain from 2004 to 2015. Building on previous literature, I analyze the determinants of hour mismatches over time and how they vary according to workers occupational groups, type of contract, hours worked and gender. These dimensions are designed to capture the main differences between high- and low-status workers. On the one hand, managerial and professional workers enjoy higher levels of stability and job security than workers in low-status occupations (Shaumann 2006Shauman, K. A. 2006. "Occupational sex segregation and the earnings of occupations: What causes the link among college-educated workers?". Social Science Research 35:577-610. http://dx.doi.org/10.1016/j.ssresearch.2004.12.001.; Filer 1990Filer, R. K. 1990. "Compensating Differentials and the Male-Female Wage Gap: A Comment". Social Forces 69:469-474. http://dx.doi.org/10.1093/sf/69.2.469.). Moreover, high-status occupations pay higher salaries, which help to combine work and the burden of family responsibilities (Pettit and Hook 2009Pettit, B. and J. L. Hook. 2009. Gendered Tradeoffs: Family, Social Policy, and Economic Inequality in Twenty-One Countries. New York: Russel Sage.). In contrast, workers in low-status occupations are more likely to have lower pay, less job security, part-time schedules and temporary employment, which make them particularly vulnerable to market fluctuations (Kalleberg, Reskin and Hudson 2000Kalleberg, A. L., B. Reskin and K. Hudson. 2000. "Bad Jobs in America: Standard and Nonstandard Employment Relations and Job Quality in the United States". American Sociological Review 65:256-278. http://dx.doi.org/10.2307/2657440.). Finally, it seems reasonable to expect different supply-side responses to recession between the sexes due to gender differences in the distribution of household chores and family responsibilities.

DATA Top

I analyze data from the Spanish Labor Force Survey (LFS) from 2005 to 2014. The Spanish Labor Force Survey (LFS) is a rolling quarterly survey of families carried out by the Spanish National Institute of Statistics (INE in the Spanish acronym). Its main objective is to obtain data on the population in relation to the labor market, working people, the unemployed and economically inactive people. It also includes detailed information about the ideal number of working hours, making it suitable as a data source for the purposes of this paper.

The dependent variable. Measuring time

The LFS includes three different questions regarding working hours, which have been formulated in exactly the same terms throughout the period from 2005 to 2014, thus enabling comparison. The answers to these questions provide exact information about contractual hours, actual working hours and preferred hours of work:[4]

1. Contracted working hours: What are/were your total agreed or contracted hours each week? Please, do not take lunch time into consideration.

2. Actual working hours: How many hours do you normally work each week? Please, do not take lunch time into consideration.

3. Preferred working hours:

• 3a. Would you be willing to work more hours than you currently do?

• 3b. Would you be willing to work fewer hours than you currently do, bearing in mind that your earnings would be lower depending on how many hours you work?

These questions are a useful measure of mismatches because they allow respondents to evaluate their present situation and report their actual preferences rather than their likely preferences in a hypothetical situation. Based on these questions, I define three dependent variables. First, the variable hour mismatches scores 1 when respondents want to work a different amount of hours and 0 otherwise. I then split the sample to distinguish people who feel underworked from those who feel overworked. Specifically, I define the variable More Hours, which scores 1 for those who want to spend more time in paid work and 0 otherwise, and the variable Fewer Hours, which scores 1 for those who want to spend less time in paid work and 0 otherwise. Following prior research, the study focuses only on employees. Self-employed respondents are excluded because, by definition, they have greater freedom to determine their labor supply. Also, I exclude personnel from the armed forces and those employed in agriculture, as they do not face the same constraints as other employees (Böheim and Taylor 2004Böheim, R. and M. P. Taylor. 2004. "Actual and Preferred Working Hours". British Journal of Industrial Relations 42(1):149-166. http://dx.doi.org/10.1111/j.1467-8543.2004.00308.x.).

Independent variables

The variable Fixed-term contract distinguishes between those employed under a permanent contract and those who are not. The distribution of temporary and permanent jobs is not the same across industries. For example, temporary employment is particularly high in the construction industry in Spain (about 60% of temporary employment, plus a significant presence of self-employed workers). This amount, however, is much lower in, for example, the manufacturing industry (around 20%) or financial services. In order to capture employees’ insecurity in different industries I control for the sector of activity.

Occupations are grouped into five main categories, namely managers and professionals; technicians; non-manual skilled workers; manual skilled workers and unskilled workers. This makes it possible to control for differences between occupations, such as security levels, salaries, work and family facilities, and so on.

The variable Number of working hours specified in contract is designed to control for variations in the continuum of working hours. I also use this variable to create a measure for full-time workers (those working more than 35 hours per week) and part-time workers (those working less than 35 hours per week). In addition to contracted hours, I include two variables to capture the effect of both paid and unpaid Overtime. This differentiation is crucial, as the number of unpaid overtime hours has soared in recent years in Spain. According to LFS data, about 52% of overtime hours were unpaid in 2010. This data is 14 percentage points higher than in 2008, immediately after the crisis started. The analysis also includes measures for having a Second Job and working for the Public Sector. Finally, year dummies are included to control for variations across years.

Finally, drawing on previous literature, the analyses include controls for variations in working hour mismatches due to sex (1 if male, 0 if female), age, age squared and civil status, in line with previous research. Also, I include a measure for immigrant status, as most of the immigrants in Spain are economic migrants and constitute a vulnerable collective of workers (Castro and Rosero-Bixby 2011Castro Martín, T. and L. Rosero-Bixby. 2011. "Motherhood and Transnational Borders. Immigrant’s Women Fertility in Spain". Revista Internacional de Sociología Monográfico nº1. La Migración en España: Perspectivas Innovadoras, 105-137. ISSN:0034-9712.).

Table 1. Means and standard deviations of variables for full sample. Average 2004-2015 [Descargar tamaño completo]

METHODS Top

The primary goal of this article is to explore how the probability of hour mismatches at work varies over time according to the type of worker. Logistic regression is normally used to model dichotomous outcome variables. However, coefficients in binary regression models are often confounded with residual variation (unobserved heterogeneity) (Allison 1999Allison, P. D. 1999. "Comparing Logit and Probit Coefficients across Groups". Sociological Methods and Research 28(2):186-208. http://dx.doi.org/10.1177/0049124199028002003.; Williams 2009Williams, R. 2009. "Using heterogeneous choice models to compare logit and probit coefficients across groups". Sociological Methods and Research 37(4):531-559. http://dx.doi.org/10.1177/0049124109335735.; Mood 2009Mood, C. 2009. "Logistic Regression: why we cannot do what we think we can do, and what we can do about it". European Sociological Review 26(1):67-82. http://dx.doi.org/10.1093/esr/jcp006.). Unobserved heterogeneity refers to the variation in the dependent variable that is caused by unobserved variables. Serious biases could occur due, for example, to changes in the composition of the labor market in the recession years. To address this critical issue, I used heterogeneous choice models that allowed me to control for this potential source of heterogeneity[5] (Williams 2009Williams, R. 2009. "Using heterogeneous choice models to compare logit and probit coefficients across groups". Sociological Methods and Research 37(4):531-559. http://dx.doi.org/10.1177/0049124109335735.; 2010Williams, R. A. 2010. "Fitting heterogenous choice models with Oglm". The Stata Journal 10(4):540-567.). In a first step, I regress the dependent variables including the set of variables specified in the previous sections. In a second step, I split the sample and fit models separately by the most relevant dimensions. Concretely, I run separate regressions for people with fixed-term contract, permanent contract, full-time employment, part-time employment, managers and professionals, technicians, non-manual skilled workers, manual skilled workers, unskilled workers, women and men. This allows assessing to what extent different groups of workers behave differently during the recession years. Additionally, as suggested by Mood (2009Mood, C. 2009. "Logistic Regression: why we cannot do what we think we can do, and what we can do about it". European Sociological Review 26(1):67-82. http://dx.doi.org/10.1093/esr/jcp006.),[6] I calculated marginal effects to interpret the substantive effects of coefficients. All regressions are estimated using the following specification:

where the numerator is the choice equation that models the effect of a series of variables on the outcome and the denominator is the variance equation that models the effect of a series of variables on the variance in outcomes. To be more specific, the x values are the explanatory variables, including the individual and occupation-related measures defined above. The z values define groups with different error variances in the underlying latent variable, which are occupational category and year in this study. β and γ are vectors of coefficients and σ is the variance.

THE LEVEL AND ORIGIN OF WORKING HOUR MISMATCHES. DESCRIPTIVE OVERVIEW Top

The first step in the analyses is to explore how the level of mismatches developed from 2005 to 2014. The solid line in Figure 1 below shows the percentage of workers who would like to change their working hours. Although figure was 18% in 2005 but had dropped to 15% in 2008, the percentage of dissatisfied people rose sharply right after the economic crisis broke, climbing to a maximum of 21% in 2013.

The long-dashed and short-dashed lines correspond to people feeling overworked and underworked, respectively. In line with prior research (Bluestone and Rose 1997Bluestone, B. and S. Rose. 1997. "Overworked and Underemployed: Unraveling and Economic Enigma". The American Prospect 31:58-69.; 1998Bluestone, B. and S. Rose. 1998. "The Macroeconomics of Work Time". Review of Social Economy 56(4):425-441.), there are more people who would prefer to work more hours after 2007. At the same time, the number of people wanting fewer hours decreases gradually until reaching 4% or so by the end of the observation period. To sum up, the breakdown of hour mismatches shows an increase in workers demanding longer working hours coinciding with the start of economic recession.

Figure 1. Working hour mismatches from 2005 to 2014 [Descargar tamaño completo]

In a second step, I explore how mismatches are originated. Table 2 shows the average yearly level of preferred working hours, contracted hours, total hours of work and both paid and unpaid overtime. The last column shows the total change between 2014 and 2005. As can be observed, increases in the level of mismatches coincide with decreases in the total number of contract hours (-1.6) and total hours of work (-1.8). Interestingly, the reduction in paid overtime coincides with the increase in unpaid overtime (-1.2 and 1.2, respectively). At the same time, there is a substantial increase in the number of desired hours of work. On average, workers in 2014 wanted to work 2.2 more hours than in 2005. These data suggest that employees wished to work more hours than they did before the crisis and not simply to return to their previous level of working hours. Thus, the data are in line with research claiming that workers seek to accumulate hours of works in recession periods, probably in an attempt to both earn more money and to hold onto their jobs (Bluestone and Rose 1997Bluestone, B. and S. Rose. 1997. "Overworked and Underemployed: Unraveling and Economic Enigma". The American Prospect 31:58-69.; 1998Bluestone, B. and S. Rose. 1998. "The Macroeconomics of Work Time". Review of Social Economy 56(4):425-441.),

Table 2. Preferred and actual hours of work. 2005-2014 [Descargar tamaño completo]

Finally, I examined matches and mismatches by contract type, full-time/part-time work, occupational group and gender. Table 3 focuses on 4 points in time, namely 2005, 2008, 2011 and 2014. The last column in the table shows the net difference between the last and the first year of the observation window.

First, the amount of people wanting the same hours is fairly stable among the most privileged workers. For workers employed under permanent contracts, the difference between 2005 and 2014 (last column) is virtually nil. The variation is also very low for managers and professionals, technicians and non-manual skilled workers, ranging from 1 to 3 percent points. However, the percentage points of workers wanting a different number of hours increases by 11 among temporary workers employed under fixed-term contracts and 7.6 points among unskilled workers. More specifically, about a third of temporary workers said they would like to work more hours than they did in 2014, almost 15 percent points more than in 2005.

While no variations were found in the amount of people wanting more hours between 2004 and 2015, a dramatic increase (almost 25 percentage points) of workers desiring more hours among those employed under part-time contracts can be observed. Similarly, the figure is eleven points higher among unskilled workers than in 2005, followed by skilled manual workers (9.1 points). Finally, the data show that women are more likely to seek more hours of work than men, and this difference was higher in 2014 than it was in 2005. All in all, the observed trends are clearly in line with the Increasing divide scenario. The multivariate analysis described in the next section delves deeper into this question.

Table 3. Matches and mismatches in working hours by type of contract and occupation [Descargar tamaño completo]

THE DETERMINANTS OF TIME MISMATCHES AT WORK. MULTIVARIATE ANALYSIS Top

The primary goal of this article is to examine the determinants of working hour mismatches in the context of economic recession. To examine this question, I regress time mismatches on the contract types under which respondents are employed, occupational group and a set of individual characteristics and work-related features. Table 4 shows the coefficients obtained for the heterogeneous choice model for Mismatch, wanting More hours and wanting Fewer hours, respectively.[7] Once the sample is broken down, the results clearly indicate that the two types of mismatches are intrinsically different, as the predictors frequently have opposite effects.

The coefficients in Table 3 show that a larger number of contracted hours and overtime is associated with a lower probability of mismatches, and in particular of feeling underworked. Conversely, it increases the likelihood of wanting fewer hours. In line with the findings in the previous section, the results confirm that temporary employees have a significantly higher probability of experiencing a Mismatch in comparison to people with a permanent contract. More specifically, fixed-term contracts increase the probability of wanting More Hours. For those desiring fewer hours of work, however, the coefficient is negative and not statistically significant. Moreover, the results confirm significant differences among workers in different types of occupations. As anticipated, manual skilled and unskilled workers are significantly more likely to experience a time mismatch. Specifically, they are significantly more likely than managers and professional workers to feel underworked and less likely to feel overworked after controlling for relevant intervenient variables. The dummies for years are statistically significant from 2009 onwards, when the probability of wanting more hours becomes significantly higher than in 2005. Exactly the opposite holds for the probability of wanting fewer hours. Finally, the variance equation in the lower half of Table 3 shows the coefficients for the two potential sources of heterogeneity, namely year and occupational group. The positive coefficient for non-manual skilled, manual skilled and unskilled workers in the variance equation tell us that there is more residual variability among low-status workers than there is for high-status workers after controlling for individual and occupation-related attributes. Interestingly, no variability is observed across years.

Table 4. OGLM regression of working hour mismatches, wanting more hours and wanting fewer hours. 2005-2014 [Descargar tamaño completo]

Up to this point, the analyses confirmed significant differences among groups of workers, but did not allow an assessment of whether these differences were greater in economic recessions than in times of economic prosperity. For this purpose, I split the sample by type of contract, hours worked, occupational group and gender, and ran separate regressions. Table 5 displays the predicted probabilities of wanting More hours for the various groups of workers and Figure 2 offers a visual comparison.

Figure 2 clearly shows that inequality between groups of workers has significantly increased during the recession, in line with the Increasing divide scenario. The left upper plot shows the probability of wanting more hours by type of contract from 2005 to 2014. The solid line represents temporary employees, while the dashed line refers to workers on permanent contracts. The likelihood of wanting more hours is higher for temporary workers over the whole period. The gap, however, widens significantly after the onset of the crisis in 2007. The probability of feeling underworked is 0.06 for workers employed under permanent contracts and 0.16 for temporary employees. In 2013, the gap widens to its maximum value. In this year, the probability is 0.09 and 0.27, respectively. That is, the risk of temporary employees’ feeling underworked in 2013 is three times higher than for permanent workers.

Full-time and part-time workers are represented in the upper right-hand corner of the figure. As for the type of contract, there is a substantial gap between the two groups of workers. Two differences emerge here. First, the probability of hour mismatches is concentrated in the group of part-time workers only, remaining low and stable among full-time workers. Second, the gap between full- and part-time workers increases very slightly until 2010, but sharply thereafter.

Table 5. Net effect of the probability of wanting more hours[8] [Descargar tamaño completo]

Figure 2. Predicted probabilities. Wanting more hours by type of contract, contracted hours, occupational group and gender [Descargar tamaño completo]

The probability trends for occupational groups are shown in the bottom left-hand corner of the figure. In general, the propensity to want more hours of work is always higher for unskilled and manual workers. However, the rapid increase in the probability is striking in both groups, compared to the more moderate rise in other occupational categories, and especially among managers, professionals and technicians. Finally, the plot appearing in the bottom right-hand corner reflects a convergence in the probability of wanting to work more hours among men and women until 2008. Thereafter, however, there is a substantial increase in mismatches for all workers, though this is particularly noticeable among women. This is true after controlling for other relevant factors such as temporary and part-time work, confirming that gender differences are not merely the result of women working in a secondary market.

DISCUSSION Top

Hour mismatches are not without consequences for individuals and family life, resulting in poor organizational performance and causing lower productivity and greater absenteeism. Prior theoretical research and empirical evidence suggest that mismatches in working hours reflect both workers’ own labor supply decisions (Altman 2001Altman, M. 2001. "Preferences and Labor Supply: Casting some Light into the Black Box of Income-Leisure choice". Journal of Socio-Economics 33:199-219. http://dx.doi.org/10.1016/S1053-5357(01)00099-3.; Sousa-Poza and Henneberger 2002Sousa-Poza, A. and F. Henneberger. 2002. "An Empirical Analysis of Working Hours Constraints in Twenty-One Countries". Review of Social Economy 60:209-242. http://dx.doi.org/10.1080/00346760210146235.) and employer preferences (Reynolds and Aletraris 2006Reynolds, J. and L. Aletraris. 2006. "Pursuing Preferences: The Creation and Resolution of Work Hour Mismatches". American Sociological Review 71(4):618-638. http://dx.doi.org/10.1177/000312240607100405.; Jacobs and Gerson 2000Jacobs, J. A. and K. Gerson. 2000. "Do Americans Fell Overworked? Comparing Ideal and Actual Work Time". Pp. 71-95 in Work and Family: Research Informing Policy, edited by Toby L. Parcel and Daniel B. Cornfield. Sage Publications.; 2004Jacobs, J. A. and K. Gerson. 2004. The Time Divide: Work, Family, and Gender Inequality. Cambridge, MA: Harvard University Press.; Böheim and Taylor 2004Böheim, R. and M. P. Taylor. 2004. "Actual and Preferred Working Hours". British Journal of Industrial Relations 42(1):149-166. http://dx.doi.org/10.1111/j.1467-8543.2004.00308.x.). However, much of this research was done in periods of economic growth and little is known about how hour mismatches vary in the downward part of the business cycle. The study tests alternative scenarios on the role of working conditions and the economic context. More specifically, I use data from the Spanish Labor Force Survey to examine the origin, evolution and determinants of working hour mismatches in Spain from 2005 to 2014.

The main findings from the study can be summarized in two points. First, the economic recession has increased mismatches, particularly because the number of people who want to work more hours than they actually do has risen inexorably in recent years. Second, this increase is especially noticeable among less privileged workers; specifically, temporary employees, part-time employees, people in low-status occupations and women. This is the net effect after controlling for individual attributes, work-related characteristics, and potential heterogeneity caused by year and changes in the occupational composition of the labor market. These findings are consistent with an Increasing divide scenario, which posits that not all workers face the same difficulties and problems. Thus, low-status workers are particularly vulnerable to economic fluctuations and more likely to want to work longer hours than their better-placed peers (Bluestone and Rose 1997Bluestone, B. and S. Rose. 1997. "Overworked and Underemployed: Unraveling and Economic Enigma". The American Prospect 31:58-69.; 1998Bluestone, B. and S. Rose. 1998. "The Macroeconomics of Work Time". Review of Social Economy 56(4):425-441.). As a result, polarization between high- and low-status workers will increase more quickly in recession periods than in periods of economic prosperity (Kalleberg 2011, Kalleberg, Reskin and Hudson 2000Kalleberg, A. L., B. Reskin and K. Hudson. 2000. "Bad Jobs in America: Standard and Nonstandard Employment Relations and Job Quality in the United States". American Sociological Review 65:256-278. http://dx.doi.org/10.2307/2657440.).

This study has both theoretical and empirical implications for future research. Conventional research has focused broadly on the study of wage gaps, occupational mobility and the risk of unemployment. More recently, research has examined different dimensions of work quality, such as changes in the control that employees can exercise over their work, job security and intensity of work (Gallie and Zhou 2013Gallie, D. and Z. Ying 2013. "Job Control, Work Intensity, and Work Stress". Pp. 115-141 in Economic Crisis, Quality of Work, and Social Integration: The European Experience, edited by Duncan Gallie. Oxford: Oxford University Press. http://dx.doi.org/10.1093/acprof:oso/9780199664719.001.0001.). The present study highlights the relevance of considering working hour mismatches in relation to inequalities in the labor market, as they shape another important source of differentiation among workers. Consequently, mismatches in working hours should arguably be added to the classic indicators of labor market inequality.

The results may also be of interest to policymakers concerned with guaranteeing the well-being of individuals, as well as good organizational performance. On the one hand, the analyses show that social structures often prevent people from getting what they might prefer. Thus, the results urge the need for organizations to allow their employees more influence in determining their own working hours to ensure affective commitment and satisfaction at work. Increasing commitment is desirable as it is assumed to decrease turnover or churn rates (Mowday, Porter and Steers 1982Mowday, R. T., L. W. Porter and R. M. Steers. 1982. Employee-organization linkages: The psychology of commitment absenteeism and turnover. New York: Academic Press.; Marsden, Kalleberg and Cook 1993Marsden, P. V., A. L. Kalleberg and C. R. Cook. 1993. "Gender Differences in Organizational Commitment Influences of Work Positions and Family Roles". Work and Occupations 20:368-390. http://dx.doi.org/10.1177/0730888493020003005.). On the other hand, support is required for people who feel underworked in order to improve economic security among low-status workers. The transformation of work in recent decades has considerably worsened the quality and pay of jobs in the low-status sector of the economy (Kalleberg 2012). Relative to high-status jobs, the insecurity and precarious nature of low-status occupations have risen to the point where some authors claim that class inequality has mushroomed (Cobble 2007Cobble, D. S. 2007. The Sex of Class: Women Transforming American Labor. Ithaca, NY: Cornell University Press.; McCall 2005McCall, L. 2005. "The Complexity of Intersectionality". Signs 30:1771-1800. http://dx.doi.org/10.1086/426800.). Analyses show that recent labor reforms (2010, 2011 and 2012) have failed to support work quality; an issue which policymakers should address.

Likewise, labor law reforms should be designed to mitigate the increasing differentiation between men and women. Findings confirm that economic crises aggravate the process of gender inequality. While both men and women are susceptible to feeling underworked during economic downturns, the increase in hour mismatches has been significantly higher among female workers. Gender differences in hours of paid work have important consequences for gender equality, as they translate into gender differences in experience, earnings and promotion opportunities (Budig and England 2001Budig, M. J. and P. England. 2001. "The Wage Penalty for Motherhood". American Sociological Review 66:204-225. http://dx.doi.org/10.2307/2657415.).

The empirical findings obtained from this study raise new questions that will require further research. Particularly, the use of panel data would improve our understanding of the emergence and resolution of mismatches in working hours. Longitudinal research would reveal how far these processes are driven by changes in actual or preferred work hours. Cross-national research would also be needed to assess the extent to which analyses yield similar results in different economic contexts. Finally, more investigation is required to account for the increasing gender divide. In this sense, further examination of how spousal characteristics can affect men’s and women’s supply side would seem particularly pertinent.

NOTES Top

REFERENCESTop

Sobre los autoresTop