The link between employee attitudes and job performance has been described as the “Holy Grail” of organizational psychology (Landy, 1989). On one hand, positive employee attitudes can lead to exceptio

Published online in Wiley Online Library (wileyonlinelibrary.com).

DOI:10.1002/hrm.21822 Correspondence to: Clare Kelliher, Cranfield School of Management, Cranfield University, Cranfield, Bedfordshire MK43 0AL, Tel: +44 (0)1234 751122, Fax: +44 (0)1234 751806, [email protected]. FLEXIBLE WORKING, INDIVIDUAL PERFORMANCE, AND EMPLOYEE ATTITUDES: COMPARING FORMAL AND INFORMAL ARRANGEMENTS LILIAN M. DE MENEZES AND CLARE KELLIHER In the context of a wider trend to individualize human resource management, this article examines the relationship between fl exible working arrangements and indi- vidual performance. Drawing on a range of theories, this article also examines potential indirect effects on employee performance via job satisfaction and organ- izational commitment and analyzes whether these relationships vary according to whether the arrangement was set up through a formal process or negotiated informally between the employee and his or her line manager. Extant research has tended to focus on formal arrangements, however, informal arrangements are widespread and may better accommodate work-life preferences, thereby poten- tially fostering more positive attitudes from employees. Survey data from 2,617 employees in four large organizations with well-established fl exible working poli- cies are analyzed. Results from structural equation models show average positive indirect effects from informal fl exible working, but also negative direct effects from formal fl exible working. When two forms of fl exible working amenable to being set up by both formal and informal means are examined separately, formal arrange- ments for fl exibility over working hours are found to be negatively associated with performance, but also a source of greater job satisfaction; informal remote work- ing arrangements have positive indirect effects via organizational commitment and job satisfaction on worker performance. © 2016 Wiley Periodicals, Inc.

Keywords: fl exible working arrangements, fl exibility over working hours, for- mal and informal arrangements, job satisfaction, organizational commitment, performance, remote working, structural equation models T his article examines the relationship between flexible working arrangements designed to accommodate employees’ needs (e.g., remote working, flexitime, compressed working) and individual performance. Drawing on a range of theories, it addresses potential indirect effects on employee performance via job satisfaction and organizational commitment. In addition, it analyzes whether the associations vary according to whether the arrangement was set up via a formal organiza- tional policy on flexible working or negotiated informally between the employee and his or her line manager. 1052 H UMAN RESOURCE M ANAGEMENT , N OVEMBER –D ECEMBER 2017 Human Resource Management DOI: 10.1002/hrm working policy, rather than actual flexible work- ing (e.g., Budd & Mumford, 2006; Wood, de Menezes, & Lasaosa, 2003). In addition, few stud- ies have considered actual individual performance.

This article, in contrast, analyzes the association between having a flexible working arrangement and actual employee performance, as measured by individual performance ratings. Indirect and direct associations with performance, via organizational commitment and job satisfaction, are hypoth- esized using competing theories, which are tested on data from a large sample of mostly professional workers in organizations with well-established flexible working policies. Importantly, this article also examines whether these associations differ according to how the flexible working arrange- ment was established: through the organization’s formal process or negotiated informally between the employee and his or her line manager.

Examining informal arrangements is impor- tant because evidence shows that formal policies are not the only means for employees to access flex- ible working arrangements and that, in practice, informally negotiated changes to working arrange- ments are widespread (Kelly & Kalev, 2006; Kossek, Lautsch, & Eaton 2005; Lambert & Waxman, 2005).

Indeed, it has been claimed that most flexible work- ing arrangements are informal (Gregory & Millner, 2009; Healy 2004; Troup & Rose, 2012). Yet, to date, knowledge about them remains very limited (Kossek et al., 2005; Rapoport, Bailyn, Fletcher, & Pruitt, 2002), because most studies have restricted their analysis to the outcomes of formal flexible working arrangements as observed by de Menezes and Kelliher (2011) in their systematic review of this literature. In some senses this is not surprising, because policy and legislative developments are largely directed toward formalized arrangements and processes via which employees can access flexible working. It is also likely that, by their very nature, formal arrangements are easier to identify and therefore study. The few studies that examine informal arrangements confirm their widespread nature and have identified a range of potential outcomes (see, for example, Anderson, Coffey, & Byerly, 2002; Richman, Civian, Shannon, Hill, & Brennan, 2008; Troup & Rose, 2012). As such, from the limited evidence available, no clear picture of how they may differ from formal arrangements emerges. Hence, this study examines flexible work- ing in organizations where formal flexible working policies were well established, but where informal arrangements were also made outside of formal processes.

The link between different forms of flexible working arrangement and employee performance and the role that employee attitudes may play is Flexible working has become increasingly common in many countries in recent years, with many employers offering some form of flexible working to their employees and significant num- bers of employees taking advantage of these oppor- tunities (see, for example, Chartered Institute of Personnel and Development [CIPD], 2012; Matos & Galinsky, 2012, 2014; Skinner, Hutchinson, & Pocock, 2012; Tipping, Chanfreau, Perry, & Tait, 2012). In some countries, greater availability of flexible working arrangements has been encour- aged by governments. The European Union policy on job quality advocates that employees should be able to exercise some control over their working arrangements (European Commission, 2012), and in the United Kingdom recent legislation has extended the legal “right to request” flexible working to all employees. Likewise, Australian legislation gives sev- eral employee groups the right to request flexible working.

Flexible working has attracted significant research attention. Studies have analyzed the associations with organizational and individual per- formance, employee attitudes, health, and well-being. While it has been observed that flexible working arrangements have been introduced in Europe due to managers’ con- cern with performance, implying a positive association with perfor- mance (Ortega, 2009), a systematic review of the literature (de Menezes & Kelliher, 2011) concluded that a “business case” for offering flexible working arrangements had not been demonstrated. Others have argued that, despite no evident direct link with performance, flexible working arrangements are inexpensive for employers and popular with employees, so may foster positive employee outcomes that could enhance performance (Bloom & Van Reenen, 2006). This view is consistent with findings from meta-analyses suggesting positive associations between certain types of flexible working arrange- ments and employee attitudes (e.g., Baltes, Briggs, Huff, Wright, & Neumann, 1999; Gajendran & Harrison, 2007) and between forms of work-family support and employee attitudes (Butts, Casper, & Yang, 2013).

Although several existing studies have exam- ined the associations between flexible working, employee attitudes, and performance, in many cases they are based on the existence of a flexible Importantly, this article also examines whether these associations differ according to how the flexible working arrangement was established: through the organization’s formal process or negotiated informally between the employee and his or her line manager. Human Resource Management DOI: 10.1002/hrm F LEXIBLE W ORKING , INDIVIDUAL PERFORMANCE , AND EMPLOYEE ATTITUDES 1053 informal arrangements as covert arrangements outside the scope of formal policies: “Supervisors can permit more flexibility than is formally allowed, encouraging employees to take time off unofficially, so that flexibility becomes invisible to higher-level managers” (Eaton, 2003, p. 147).

Here, a distinction is made between formal and informal flexible working in the following way. Formal arrangements are those that have been made via the organization’s flexible working policy, which normally involves a written request from the employee to be considered by his or her manager in conjunction with the human resources (HR) department. 1 Informal arrangements are those that have not gone through this process, but rather emerge from a discussion or negotiation between the employee and his or her line manager. These informal arrange- ments tend to relate to flexibility over working hours or remote work- ing, which do not require changes to the official contract of employment.

Both forms of arrangements may be seen as similar to i-deals (Rousseau, 2005), or more specifically flex- ibility i-deals (Hornung, Rousseau, & Glaser, 2008) and represent an individualization of the employ- ment relationship (Rousseau, 2005).

I-deals are defined as being individ- ually negotiated, heterogeneous, of benefit to both the employee and the employer, varying in scope, and initiated by employees (Rousseau, Ho, & Greenberg, 2006); however, the literature is less clear about who the arrangement is made with.

Some studies suggest that it is made with the organization or employer (Bal, de Jong, Jansen, & Bakker, 2012; Rousseau, 2005) suggesting a formal arrangement, whereas others suggest they are made between the supervisor and the employee (Rosen, Slater, Chang, & Johnson, 2013), which could be either formal or informal.

The focus of the present study is on arrangements initiated by the employee (although access is con- trolled by the employer) (Beck, 2013) that are spe- cific to the individual, but a distinction is made between those that are made via the formal pro- cess and those that are made informally. Flexible Working Arrangements and Performance Several extant studies have examined the link between giving employees some choice over their of interest to researchers, managers, and policy makers. This article builds on extant research in a number of ways. First, it explores the relation- ship between having a flexible working arrange- ment (specifically remote working and flexibility over working hours) and employee performance, using individual performance ratings. Second, the mediating roles of organizational commitment and job satisfaction in this relationship are exam- ined. Third, whether there are differences in these relationships according to whether the flexible working arrangement was established formally or informally is examined. Remote working and flex- ibility over working hours were chosen for further analysis because they are types of flexible work- ing that lend themselves to being established by either formal or informal means because they do not need to involve changes to the formal con- tract of employment. Furthermore, the focus here is on flexible working arrangements that are avail- able to all employees, not just those with parent- ing and caring responsibilities. This is in line with recent calls in the work-life literature for the adop- tion of a perspective on “life,” which goes beyond caring and domestic activities (de Janasz, Forret, Haack, & Jonsen, 2013; Ozbilgin, Beauregard, Tatli, & Bell, 2011). Background and Hypotheses Flexible Working Arrangements: Informal and Formal Flexible working arrangements in this article are taken to be arrangements that allow employ- ees to vary the amount, timing, or location of their work (de Menezes & Kelliher, 2011) and are designed to enable them to balance the demands of their work and nonwork lives more effectively.

These include, for example, remote working, flex- ibility over working hours, and reduced hours.

In a review of the literature, de Menezes and Kelliher (2011) observed that few studies distin- guished between formal and informal arrange- ments. The small number of studies that have specifically examined informal arrangements have not, however, defined these arrangements in a consistent way. For example, Richman et al.

(2008) identified informal arrangements simply as those that are occasionally used, while Troup and Rose (2012) described them as those that “were negotiated in a need-based way with supervisors or management” (p. 474). For Hall and Atkinson (2006), the immediacy and responsiveness to employee needs are key, because they defined informal arrangements as the ability to alter planned working time on an ad hoc basis at short notice. Eaton (2003) went further and referred to The focus here is on flexible working arrangements that are available to all employees, not just those with parenting and caring responsibilities.

This is in line with recent calls in the work-life literature for the adoption of a perspective on “life,” which goes beyond caring and domestic activities. 1054 H UMAN RESOURCE M ANAGEMENT , N OVEMBER –D ECEMBER 2017 Human Resource Management DOI: 10.1002/hrm of arrangement but it was higher for those with a formal arrangement.

Affective organizational commitment has been described as a mediating path through which job design or redesign initiatives can be channeled to facilitate proactiveness and performance (Jafri, 2010; Thomas, Whitman, & Viswesvaran, 2010; Yee, Yeung, & Cheng, 2010). In the wider human resource management literature, the high involve- ment, high performance, and high commitment systems models predict that cultivating organi- zational commitment will translate into perfor- mance (Lawler, 1986; Osterman, 1995; Wood & de Menezes, 1998). In essence, the opportunity to customize working arrangements through both formal and informal processes may foster positive sentiments toward the organization, resulting in enhanced individual performance. We therefore hypothesize:

Hypothesis 1a: There is a positive association between informal fl exible working arrangements and individual performance, mediated by organizational commitment.

Hypothesis 1b: There is a positive association between formal fl exible working arrangements and individual performance, mediated by organizational commitment.

The Relationship with Job Satisfaction Hackman and Oldham’s (1975) job character- istics model implies that the basic characteristics of a job (e.g., skill variety, task identity, task signif- icance, autonomy) influence psychological well- being, which can affect individual performance.

Allowing employees some autonomy over their working arrangements may give them a sense of independence (Tietze, Musson, & Scurry, 2009) that could be linked to enhanced job satisfaction and in turn higher performance (Dodd & Ganster, 1996; Scandura & Lankau, 1997). Employees with informal flexible working arrangements have been found to place high value on their arrangements and also have a perception of control, which might suggest a positive influence on job satis- faction (Hall & Atkinson, 2006). Karasek’s (1979, 1989) model further proposes that higher job dis- cretion enables workers to cope better with higher job demands and may buffer associated adverse effects. Spector (1986), examining the relation- ship between perceived control and employee outcome variables and drawing on 101 studies, predicted that higher perceived control would be associated with greater motivation and job sat- isfaction. Furthermore, job control and enrich- ment have been found to be associated with job working arrangements and performance. Using social exchange theory (Blau, 1964), it has been argued that enhanced employee performance may be an act of reciprocation between the employee and the employer (Golden, 2001, 2009; Kelliher & Anderson, 2010), thus where an employee exer- cises a flexible working option, a feeling of obli- gation is generated toward the employer. Taking a different but related perspective, Konrad and Mangel (2000) used Akerlof’s (1982) gift exchange theory to propose a positive association between the provision of work-life programs and produc- tivity. The basis of gift exchange is that if the employer provides a “gift” to the employee by pay- ing wages or other benefits above what is required by the market, the employee will in turn respond with a “gift” of performance above the norm.

Thus, employers may offer a gift in the form of choice over working arrangements, expecting the employee to respond with enhanced effort or per- formance. In the case of informal flexible working arrangements, Atkinson and Hall (2009) observed that having an informal arrangement creates a sense of obligation on the part of the employee and a consequent need for reciprocation, likely to result in behaviors that are seen to be valued by the manager who has granted the arrangement and might involve additional effort. To date, however, there has been little examination of this relation- ship for any form of arrangement with reference to independent, comparable data on individual employee performance. The Relationship with Organizational Commitment By offering employees the opportunity to work flexibly, organizations may also foster a perception of organizational support in helping them manage the interface between their work and nonwork lives. In line with the norm of reciprocity (Gouldner, 1960), this may stimulate behaviors such as loyalty, attendance, and punc- tuality (Settoon, Bennett, & Liden, 1996) and may therefore imply an indirect link between flexible working arrangements and performance via orga- nizational commitment. A number of studies have found a relationship between flexible working and organizational commitment (see, for example, Chow & Keng-Howe, 2006; Gajendran & Harrison, 2007; Kelly & Moen, 2007). Notwithstanding the problems with definitions discussed above, a study of how formal and informal approaches to flexibility relate to organizational commitment showed little difference between the forms (Eaton, 2003). Looking at engagement and expectations to remain with the organization, Richman et al.

(2008) found positive associations for both forms Human Resource Management DOI: 10.1002/hrm F LEXIBLE W ORKING , INDIVIDUAL PERFORMANCE , AND EMPLOYEE ATTITUDES 1055 retain high-caliber staff, reduced sickness absence, and a reduced need for workspace where staff work remotely), employees may not feel under obligation to yield anything additional, such as increased effort, in return. Furthermore, if employ- ees believe that they would be able to access a similar arrangement with another employer, then this is less likely to influence their attitudes in a positive way, because no special favors are seen to have been done.

In the case of informal arrangements made between employees and their line managers, managers exercise discretion in allowing changes to be made to the working arrangement. While managers’ decision latitude may be constrained by the practicality of redesigning work to accom- modate employees’ needs (Furunes, Mykletun, & Solem, 2011) and by their competence to manage the work arrangement (Earl & Taylor, 2015), the employee may none- theless feel that the manager has personally made efforts to accom- modate his or her needs; hence, there may be a belief that something is owed in return (Gouldner, 1960).

The belief that through an informal arrangement employees’ specific needs have been accommodated may also foster feelings of organi- zational commitment, or at least commitment to the manager, but as Coyle-Shapiro and Shore (2007) observed, for many employees the line manager in effect represents the organization. With job satisfaction, informal flexible working arrange- ments may be seen to offer greater discretion than formal arrangements and to be aligned more closely with individual circumstances, unlike, for example, a formal flexitime scheme that may offer only predesignated windows of flexibility.

Hall and Atkinson (2006) have observed that informal arrangements may be characterized by ongoing negotiations between the employee and his or her manager to the benefit of both par- ties, but note that this is unlikely to be evident with formal arrangements. However, other studies while reporting that formal schemes are seen to have greater legitimacy, suggested that informal arrangements may question equity and mana- gerial control (Fogarty, Scott, & Williams, 2011) and could result in counterproductive workplace behaviors (Beauregard, 2014).

An important further difference between formal and informal approaches is the potential satisfaction (de Menezes, 2011; Hammer, Neal, Newson, Brockwood, & Colton, 2005; Kelliher & Anderson, 2008, 2010), which can in turn be linked to performance (Wood, van Veldhoven, Croon, & de Menezes 2012).

In the human resource management litera- ture, an organization’s ability to motivate and retain its human capital has often been asso- ciated with its adoption of high-performance work systems (Becker & Huselid, 1998; Boxall & Purcell, 2003; Capelli, 2000; Pfeffer, 1994; Wright, Gardner, & Moynihan, 2003), and job satisfaction has been identified as a potential mediator in the link between organizational policies and perfor- mance (Purcell & Kinnie, 2007). Thus, if employee choice over their working arrangements (through either a formal or informal process) is indicative of job autonomy, drawing on the job character- istic model and high-performance work systems theory, we hypothesise:

Hypothesis 2a: There is a positive association between informal fl exible working arrangements and individual performance, mediated by job satisfaction.

Hypothesis 2b: There is a positive association between formal fl exible working arrangements and individual performance, mediated by job satisfaction. Differences between Informal and Formal Flexible Working Arrangements The small number of studies that have attempted to examine differences in formal and informal flexible working arrangements have produced mixed results and failed to clarify dif- ferences in the way in which they may relate to performance. This may in part be related to the definitions used, as discussed earlier. Nevertheless, conceptually, it is reasonable to propose that there might be distinctions in outcomes between formal and informal arrangements. In the case of a formal arrangement, the employee has taken advantage of a benefit offered by his or her employer. Under these circumstances, it could be argued that there may be little perceived need to give back some- thing in return. In other words, formal arrange- ments accessed through an organizational policy are more likely to generate a sense of entitlement to change working arrangements (den Dulk, Peters, & Poutsma, 2012; Lewis & Smithson, 2001; Stavrou & Ierodiakonou, 2015).

Likewise, in the context of much discussion, both in the management press and inside orga- nizations, about the business benefits of flexible working (such as enhanced ability to recruit and The small number of studies that have attempted to examine differences in formal and informal flexible working arrangements have produced mixed results and failed to clarify differences in the way in which they may relate to performance. 1056 H UMAN RESOURCE M ANAGEMENT , N OVEMBER –D ECEMBER 2017 Human Resource Management DOI: 10.1002/hrm there are legal provisions in relation to the right to request flexible working), and there is no per- ceived need to reciprocate, outcomes from formal flexible working arrangements, such as organiza- tional commitment, could be weaker than with an informal arrangement.

In a similar vein, if flexible working arrange- ments become the norm, Akerlof’s Gift Exchange theory would imply that they will no longer be perceived as a gift or an incentive for enhanced effort. Consequently, any positive association between flexible working arrangements and per- formance would weaken over time. Along simi- lar lines, it has been observed that the positive association with employee outcomes is weaker in organizations where flexible working policies had been in place for longer than where they had been introduced more recently (Kelliher & Anderson, 2010). Given the lack of literature and empirical evidence on flexible working arrangements and the effect of time, it would be difficult to formal- ize a hypothesis; we therefore propose:

The associations between fl exible working arrange- ments and performance are moderated by the length of time the individual works fl exibly.

Nonetheless, if flexible working is essentially perceived as an element of job control, neither its length of use nor its prevalence in the organiza- tion should influence the association with job sat- isfaction and performance.

In this article, the outcomes of formal and informal flexible working arrangements will be considered. The analysis will first examine sev- eral types of flexible working that offer employ- ees some choice over their working arrangements together (including flexibility over working hours, staggered hours, remote working, and part-time working) and then examine remote working and flexibility over working hours separately, because these are types that are amenable to both formal and informal arrangements.

Method The article draws on survey data from 2,617 respondents in four organizations in the United Kingdom, all of which were large, multinational companies drawn from the pharmaceutical, utili- ties, banking, and consulting sectors. Each com- pany had offered a range of flexible working arrangements to employees for several years and had been identified as active promoters of flexible working by campaigning organizations and the media.

Data were collected via a questionnaire distributed by email as a hypertext link to all security of the arrangement. In the case of formal flexible working arrangements, once the arrange- ment has been made, it is likely to be harder, although not impossible, for the employer to withdraw from the arrangement, which itself is likely to involve a formal process. With informal arrangements, it is likely to be easier for a man- ager to change his or her mind and require the employee to return to a standard working arrange- ment. Faced with a lack of security of the arrange- ment, the employee may attempt to deliver a high level of performance to “protect” his or her arrangement (Hutchinson, 2012; Kelliher & Anderson, 2010). Informal arrangements may also be at risk if either the line man- ager or the employee moves to a new role. A new line manager may or may not be willing to honor an existing informal arrangement. As such, any positive influence on organizational commitment and job satisfaction may be mitigated by the potential insecurity of an informal arrangement.

Despite variations in the approaches and definitions adopted in the literature, it would seem rea- sonable to argue that there are likely to be differences between formal and informal arrangements, based on variations in perceived legitimacy and security of the arrangement and a sense of entitlement or obligation.

Therefore, we hypothesize:

Hypothesis 3: Formal and informal fl exible working arrangements differ in their associations with performance.

Hypothesis 4a: The association with organizational commitment differs depending on whether the fl exible work- ing arrangement was established via formal or infor- mal means.

Hypothesis 4b: The association with job satisfaction differs depending on whether the fl exible working arrangement was established via formal or informal means.

In other words, indirect and total effects on performance may vary with the form of arrange- ment. For example, if formalized arrangements lead to a sense of entitlement (as might be the case in countries like the United Kingdom, where Despite variations in the approaches and definitions adopted in the literature, it would seem reasonable to argue that there are likely to be differences between formal and informal arrangements, based on variations in perceived legitimacy and security of the arrangement and a sense of entitlement or obligation. Human Resource Management DOI: 10.1002/hrm F LEXIBLE W ORKING , INDIVIDUAL PERFORMANCE , AND EMPLOYEE ATTITUDES 1057 indicate nonresponse biases; the distributions of job satisfaction and commitment, though based on different scales, are consistent with those obtained from a larger survey of employees in the United Kingdom (van Wanrooy et al., 2013). After removing inconsistent cases, the analysis was based on 2,617 individuals.

Respondents were asked if they worked flex- ibly, with three possible answers: no; yes, infor- mally; yes, formally. If yes, they were then asked to indicate the type of flexible working, which included among several options: “Is this remote working?” and “Is this flexitime?” By using these three questions, respondents who had any flex- ible arrangement or a combination of arrange- ments were included in the category flexible working. Formality or informality was assessed by the answers to the second part of the first ques- tion. Out of flexible workers, remote working and flexibility over work- ing hours were inferred by a positive response to the question concern- ing the specific arrangement. As shown in Table  I, 59.8 percent of respondents (1,565) had a flexible working arrangement, and most arrangements were informal. As might be expected, there is signifi- cant overlap between remote work- ing and flexibility over working hours, because most remote workers (68.36 percent) also reported having flexibility over working hours.

The sample is nearly evenly split by gender (49.2 percent female), and 97 percent of the respondents are approximately equally distributed across three companies, with the remaining 3 per- cent working in the bank. The gender distribution of flexible workers is shown in Table II. Thirty-five percent of formal flexible workers were male, and out of those who described themselves as informal flexible workers, 58 percent were male.

As shown in Table III, most flexible workers were aged between 30 and 49. Overall, chi-square tests (5 percent significance level) indicate that the use flexible working varies with age and gen- der, except in the case of remote working. employees in the divisions being researched in each company, who were mainly professional workers. By focusing on professionals, the element of choice over working arrangements can be bet- ter captured, because professionals are more likely to be able to exercise discretion over their work- ing hours, location, and effort (Felstead, Jewson, Phizacklea, & Walters, 2002; Golden, 2012; Ibarra, 1999). In comparison to previous studies, the data set offers several advantages. First, it includes responses from both those with and those without a flexible working arrangement. It further distin- guishes between those with formal and informal arrangements. This allows for direct comparisons to be made among three separate groups: no flex- ible working arrangement, formal flexible work- ing arrangement, and informal flexible working arrangement. Second, like a small number of other studies (e.g., Allen, 2001; Hammer et al., 2005; Lapierre & Allen, 2006), it includes responses from actual flexible workers, not just those who have access to, or perceived access to, flexible working arrangements (e.g., CIPD, 2012; Matos & Galinsky, 2012; van Wanrooy et al., 2013). Third, the proportion of male respondents in the sample is large, both with formal and informal flexible working arrangements, and may therefore allow for more general observations to be made, than with analyses largely based on female respondents (Casper, Eby, Bordeaux, & Lockwood, 2007; Eby, Casper, Lockwood, Bordeaux, & Brinley, 2005). It is important to note that this study includes all employees with flexible working arrangements and is not limited to parents and carers.

Overall, a response rate of 24 percent was achieved, which is significantly higher than both the estimated participation of adults in US surveys (18.1 percent by Bickart and Schmittlein, 1999) and that reported by Wright and Schwager (2008) in their study of the effectiveness of online sur- veys. Considering the distributions of response rates for data collected from individuals reported by Baruch and Holtom (2008), a response of 24 percent is not statistically significantly below the average (5 percent significance level, in one-tailed test based on the means and standard deviations of response rates in their study). Response was encouraged by including a cover letter from a senior manager in each organization and ensuring the language used was meaningful for the orga- nization. One possible reason for nonresponse is that the employees targeted mainly worked in high-pressure environments and therefore may not have taken the time to respond, which is a common problem in data collection. As described below, preliminary analysis of the data does not This allows for direct comparisons to be made among three separate groups:

no flexible working arrangement, formal flexible working arrangement, and informal flexible working arrangement. TABLE I Distribution of Flexible Workers   Formal Informal Total Flexible Working 490 1,075 1,565 Flexibility over Working Hours134 902 1,036 Remote Working 239 779 1,018 1058 H UMAN RESOURCE M ANAGEMENT , N OVEMBER –D ECEMBER 2017 Human Resource Management DOI: 10.1002/hrm but nonresponse was found to be random. The average (median) respondent had significantly met or exceeded expectations, which is to be expected with the assessment of performance of employees in post.

Job satisfaction was measured by a 10-point additive scale based on two items (Schneider, Hanges, Brent Smith, & Salvaggio, 2003): Consi- dering everything, how satisfied are you with your job? Considering everything, how would you rate your overall satisfaction with your company at the present time? The correlation between the items is equal to 0.8, and the measure is reliable (Spearman Brown coefficient = 0.89).

Organizational commitment was measured using Cook and Wall’s (1980) British Organi- zational Commitment Scale, which has nine items and uses a 7-point Likert scale. A one-fac- tor model explains 52 percent of the variance in the data, and loadings vary between 0.5 and 0.8.

Although there was some residual correlation between items, the additive scale was judged to be reliable (Cronbach’s alpha = 0.81), and therefore items were not deleted to improve the fit of the factor model.

The distributions of the outcome variables show few outliers, and no significant gender dif- ferences were found. The average individual in the sample is satisfied with his or her job (mean = 7.21, standard deviation = 1.86) and committed to the organization (mean = 27.82, standard deviation = 6.84). Because flexible working arrangements may vary between organizations, and as Tables I through III suggest associations with individual characteristics, control variables were indicator dummy variables for the employees’ age (refer- ence category: less than 30) and gender (reference category: male), as well as organization.

Table IV summarizes the two-way associations in the data. Employees have either an informal or formal working arrangement, hence the negative correlations between forms. Rows 7 and 8 illus- trate positive associations between flexible work- ing arrangements and employee outcomes, which vary in significance. Row 11, however, indicates no association between gender and performance ratings. Taken together, rows 7, 8, and 11 suggest differences in outcomes from different flexible working arrangements, which are examined below.

Missing values were found to be random and treated as such. The distribution of performance ratings is skewed, as previously indicated. Analysis Procedure The mediation model, where an independent variable is directly associated with an employee attitude that is linked to his or her performance The length of flexible working was established by the question: “How long have you been work- ing flexibly?” Answers were coded as less than one year, between one and two years, between two and four years, and greater than four years, and the distribution is nearly evenly split among the four groups. The median respondent had been work- ing flexibly between one and two years, and 26.2 percent of respondents had been working flexibly for over four years.

Respondents were asked to report their last performance rating, which is taken as a proxy measure of performance. There were some minor variations made to the wording of the questions to fit with the terms used in each organization.

While this is a self-reported measure of perfor- mance, it does not rely on the respondents’ own judgment about their performance. Responses were then coded into four general categories:

1 = improvement required or expectations not met (5.8 percent); 2 = good or expectations met (26.3 percent); 3 = high, significantly meeting or exceeding expectations (48.6 percent); 4 = excel- lent or outstanding (18.9 percent). Of the 2,617 respondents, 195 did not answer this question, TABLE III Age and Flexible Working   Up to 29 30–39 40–49 ≥ 50 Total General (Any) Flexible Working Informal 237 459 284 91 1071 Formal 65 259 126 39 489 Flexible Working Hours Informal 155 414 258 72 899 Formal 33 61 24 16 134 Remote Working Informal 137 360 215 65 777 Formal 29 106 77 25 237 TABLE II Gender and Flexible Working   Male Female Total Flexible Working Informal 619 454 1,073 Formal 172 318 490 Flexibility over Working Hours Informal 495 406 901 Formal 57 77 134 Remote Working Informal 456 322 778 Formal 126 113 239 Human Resource Management DOI: 10.1002/hrm F LEXIBLE W ORKING , INDIVIDUAL PERFORMANCE , AND EMPLOYEE ATTITUDES 1059 rating, is the basis of the analysis. Because different flexible working arrangements are to be compared, the independent variable in the model is a binary indicator of group membership that is equal to zero for those in the benchmark group (i.e., not a flexible worker, while testing Hypotheses 1 and 2) and equal to one for those in the other group (e.g., flexible worker with a formal arrangement). The analysis of different subsamples enables an assess- ment of the sensitivity of the results obtained from the larger samples where outcomes from any flexible working arrangement (flexible working) are examined. Using MPlus (Muthén & Muthén, 1998–2010), standardized coefficients and esti- mates of direct and indirect effects on employee performance are obtained.

Job satisfaction and organizational commit- ment are often correlated, for example, as shown in meta-analyses (e.g., Kaifeng, Lepack, Jia, & Baer, 2012; Thomas et al., 2010). This is also observed here, because the correlation between the employee attitudes is high (r = 0.72) and could have allowed for the esti- mation of a common attitudinal factor. However, the hypotheses in this study are based on differ- ent theories of associations with specific dimensions of employee attitude, and may unveil distinct effects of forms of flexible working on each dimension of employee attitude. Consequently, the paths to employee attitudes are tested separately, by estimating models where the mediator is either orga- nizational commitment (Hypothesis  1) or job satisfaction (Hypothesis 2).

For each type of flexible working arrangement and employee attitude (organizational commit- ment, job satisfaction), the mediation model with performance as the final outcome is estimated three times, by selecting a subsample of the data that corresponds to those to be compared, while testing Hypotheses 1 and 2: not working flexibly and working flexibly with an informal arrange- ment; not working flexibly and working flexibly with a formal arrangement; and to test Hypothesis 3, working flexibly with an informal arrangement and working flexibly with a formal arrangement.

Goodness of fit of each model is judged by sev- eral criteria (chi-square tests, information criteria, and root mean square error measures). The signifi- cance of standardized estimated indirect (via job satisfaction or organizational commitment) and estimated direct effects on performance is then assessed, by comparing the respective probability values (p-values) with the threshold. If significant TABLE IV Correlations   1 2 3 4 567891011 1. Informal fl exible working 2. Formal fl exible working –0.40** 3. Informal fl exibility over working hours 0.76** –0.30** 4. Formal fl exibility over working hours –0.30** 0.72** –0.22** 5. Informal remote working 0.77** –0.31** 0.57** –0.22** 6. Formal remote working –0.27** 0.67** –0.20** 0.58** –0.21** 7. Organizational commitment 0.05* 0.06** 0.04 0.04 0.09** 0.06** 8. Job satisfaction 0.06** 0.06** 0.03 0.05** 0.08** 0.05** 0.72** 9. Gender –0.12** 0.15** –0.06* 0.09** –0.10** –0.01 –0.02 0.04* 10. Age 0.06** 0.07** 0.10** 0.04* 0.08** 0.09** –0.05** 0.06** –0.09** 11. Performance rating 0.07** –0.08** –0.04 –0.07** 0.12** –0.02 0.18** 0.23** 0.02 –0.16** 12. Length time 0.00 0.00 0.09** 0.03 –0.05 –0.04 0.07* –0.10* –0.05* 0.25** –0.05 Note: Forms of fl exible working arrangements above are binary variables that are equal to 1 if used and 0 otherwise (N = 2, 617).

Phi coeffi cients are reported for binary variables; otherwise Spearman correlation coeffi cients are reported. *: 5% signifi cance level, **: 1% signifi cance level. The hypotheses in this study are based on different theories of associations with specific dimensions of employee attitude. 1060 H UMAN RESOURCE M ANAGEMENT , N OVEMBER –D ECEMBER 2017 Human Resource Management DOI: 10.1002/hrm Results Flexible Versus Nonfl exible Workers:

The Added Value of Flexible Working Arrangements Table V summarizes the results from testing Hypotheses 1 and 2 using mediation models where flexible workers are benchmarked by nonflexible workers. It shows standardized estimates of direct and indirect effects of informal and formal flexible at the 5 percent level, these estimates indicate differences relative to the benchmark working arrangement.

Having identified significant links, the poten- tial moderation of the length of time working flexibly is then considered. This variable and its interactions with the indicator variable are then included in the specific paths of the models that were found to be significant. The new models and estimates are then assessed using the same criteria. TABLE V Assessing the Effect of Flexible Working—Standardized Estimates    Estimate SEp-Value n R 2 Informal Flexible Working Via Organizational Commitment 0.020** 0.003 .001 Total (direct + via OC) 0.036 0.019 .064 2,109 0.31 Via Job Satisfaction 0.016** 0.004 .000 Total (direct + via JS) 0.034 0.019 .078 2,109 0.31 Informal Flexibility over Working Hours Via Organizational Commitment 0.005 0.003 .054 Total (direct + via OC) 0.006 0.018 .721 2,403 0.31 Via Job Satisfaction 0.005 0.003 .096 Total (direct + via JS) 0.004 0.018 .827 2,403 0.31 Informal Remote Working Via Organizational Commitment 0.011** 0.003 .000 Total (direct + via OC) 0.045* 0.019 .018 2,312 0.31 Via Job Satisfaction 0.014** 0.004 .000 Total (direct + via JS) 0.044* 0.019 .020 2,312 0.32 Formal Flexible Working Via Organizational Commitment 0.013** 0.004 .001 Total (direct + via OC) –0.036 0.023 .121 0.31 Via Job Satisfaction 0.016** 0.004 .000 Total (direct + via JS) –0.039 0.023 .100 1,527 0.31 Formal Flexibility over Working Hours Via Organizational Commitment 0.007* 0.003 .044 Total (direct + via OC) –0.040 0.022 .066 1,638 0.32 Via Job Satisfaction 0.010** 0.004 .010 Total (direct + via JS) –0.04 0.022 .066 1,638 0.32 Formal Remote Working Via Organizational Commitment 0.009** 0.003 .003 Total (direct + via OC) –0.012 0.022 .557 1,772 0.3 Via Job Satisfaction 0.010** 0.004 .005 Total (direct + via JS) –0.014 0.021 .514 1,772 0.3 Note: Independent variable is a binary indicator equal to 1 when the employee has the arrangement and 0 when the employee does not work fl exibly.

Control variables: gender (ref. category = male), age (ref. category = less than 29), company.

*: 5% signifi cance level, **: 1% signifi cance level. Human Resource Management DOI: 10.1002/hrm F LEXIBLE W ORKING , INDIVIDUAL PERFORMANCE , AND EMPLOYEE ATTITUDES 1061 commitment is positively associated with perfor- mance (standardized coefficient = 0.12, p-value = .00). As highlighted in Table V, the indirect effects on performance via organizational commitment (estimate = 0.007) and the total effect (estimate = 0.045) are positive. Consequently, the subsample that works remotely supports Hypothesis 1a.

Hypothesis 1b: Formal Flexible Working, Organizational Commitment, and Performance Considering formal flexible working, the model fit is satisfactory (SRMSE = 0.024). There is a positive association between formal flexible work- ing and organizational commitment (standard- ized coefficient = 0.11, p-value = .00) and between organizational commitment and performance (standardized coefficient = 0.12, p-value = .00).

Yet, the direct association with performance is negative (standardized coefficient = –0.05, p-value = 0.04). Consequently, positive indirect effects via organizational commitment (estimate = 0.016, p-value = .00) and negative direct effects (estimate = –0.04, p-value = .04) are found, with the total effect that is shown in Table V being insignificant (p-value = .121).

Benchmarking formal flexibility over work- ing hours by not working flexibly, the model fits the sample (SMRMSE = 0.02) and shows a posi- tive association with organizational commitment (standardized coefficient = 0.05, p-value = .03), which is positively associated with performance (standardized coefficient = 0.13, p-value = .00).

However, the direct association between flexibil- ity over working hours and performance is also negative (standardized coefficient= –0.05, p-value = .03). Estimated effects are positive indirect (esti- mate = 0.01, p-value = .04) and negative direct (estimate = –0.07, p-value = .03), resulting in the total effect that is shown in Table V (estimate = –0.04, p-value = .066), which is insignificant at the 5 percent level. As above, the positive indirect effects of formal flexibility over working hours via organizational commitment are diluted.

Similarly, when formal remote working is examined, the model (SRMSE = 0.024) shows posi- tive association with organizational commitment (standardized coefficient = 0.08, p-value = .00), but no direct association with performance (stan- dardized coefficient = –0.02, p-value = .3). Indirect positive effects via organizational commitment are positive (estimate = 0.01, p-value = .00), but the estimated total effect is insignificant (p-value = .56).

In summary, positive indirect effects, as high- lighted in Table V, are observed thus supporting Hypothesis 1b. Formal flexible working arrange- ments can lead to greater organizational commit- ment, but do not result in higher performance. working arrangements, their standard errors and significance (p-values), the subsample sizes (n) and the variance in performance that is explained by each model (R 2). Generally, about 30 percent of the variance in performance ratings is explained; indi- rect effects of flexible working arrangements via organizational commitment and job satisfaction are significant, except when focusing on infor- mal flexibility over working hours. As previously observed, gender is independent of performance; however, respondents aged between 30 and 49 were more likely to have reported higher perfor- mance when compared to the reference category (less than 29) and this was common to all models (5 percent significance level). Below, the associa- tions with each outcome are examined in detail.

Hypothesis 1a: Informal Flexible Working, Organizational Commitment, and Performance Considering flexible working in general, the model fits the sample (standardized root mean square error [SRMSE] = 0.023, which is less than the threshold of 0.05). Positive direct associations between informal flexible working and organiza- tional commitment (standardized coefficient = 0.08, p-value = .00) and between organizational commitment and performance (standardized coefficient = 0.013, p-value = .00) are supported.

However, there is no direct association between informal flexible working and performance (stan- dardized coefficient = 0.03, p-value = .18). As shown in Table V, the estimated effects from informal flex- ible working are indirect via organizational com- mitment (estimate = 0.02, p-value = .00) and total effects are not significant at the 5 percent level; hence, Hypothesis 1a is only partially supported.

Similarly, when benchmarking informal flex- ibility over working hours against not working flexibly, the model fits the sample (SRMSE = 0.026) and confirms a positive association with organi- zational commitment (standardized coefficient = 0.04, p-value = .04), which is linked to performance (standardized coefficient = 0.12, p-value = .00). As above, there is no direct association between infor- mal flexibility over working hours and performance (standardized coefficient = 0.002, p-value = .93). As shown in Table V, p-values of estimated effects on performance are greater than .05, so that indirect and direct effects are insignificant. Consequently, this subsample rejects Hypothesis 1a.

Examining informal remote working, the model fit is satisfactory (SRMSE = 0.023). A positive association with organizational commitment is observed (standardized coefficient = 0.094, p-value = .00), the direct association with performance is insignificant at the 5 percent level (standardized coefficient = 0.03, p-value = .07). Organizational 1062 H UMAN RESOURCE M ANAGEMENT , N OVEMBER –D ECEMBER 2017 Human Resource Management DOI: 10.1002/hrm (standardized coefficient = –0.05, p-value = .02).

As highlighted in Table V, indirect effects of for- mal flexible working through job satisfaction are positive (estimate = 0.02, p-value = .00), thus sup- porting Hypothesis 2b. Nonetheless, there is nega- tive direct association with performance (estimate = –0.05, p-value = .02) and the total effect is insig- nificant (p-value = .10).

The model of having formal flexibility over working hours versus no flexible working fits the subsample (SRMSE = 0.026, n = 1638) and shows positive association with job satisfaction (stan- dardized coefficient = 0.07, p-value = .01), which is linked to higher performance (standardized coefficient = 0.15, p-value = .00). However, having formal flexibility over working hours is negatively associated with performance (standardized coef- ficient = –0.05, p-value = .02). There are positive indirect effects (estimate = 0.01, p-value = .01), which support Hypothesis 2b, but negative direct effects (estimate = –0.05, p-value = .02) lead to a total effect that is insignificant (p-value = .066).

When formal remote working is benchmarked against no flexible working arrangements (SRMSE = 0.02), formal remote working is positively linked to job satisfaction (standardized coefficient = 0.08, p-value = .00), which is positively correlated with performance (standardized coefficient = 0.14, p-value = .00), but there is no direct association between formal remote working and performance (standardized coefficient = –0.02, p-value = .26). As above, the positive effect of formal remote work- ing is indirect via job satisfaction (estimate = 0.01, p-value = .01) in line with Hypothesis 2b, but the total effect is insignificant (p-value = .51).

Thus, as observed with organizational commit- ment, there are positive indirect effects of formal flexible working via job satisfaction, but the overall association between formal flexible working and performance is not significant. Hence, the impact of formal flexible working on performance is via the employee attitude. As a whole, results from for- mal arrangements do not vary significantly with the type of flexible working arrangement.

Comparing Informal and Formal Flexible Working Arrangements The results thus far suggest that there are dif- ferences in outcomes from formal and informal arrangements. Formal flexible working arrange- ments have been found to be negatively associated with employee performance. Formal flexibility over working hours may lead to positive employee attitudes, while there is no evidence in regards to informal flexibility over working hours. Hence, the 12 models that were summarized above gener- ally support Hypothesis 3. Hypothesis 2a: Informal Flexible Working, Job Satisfaction, and Performance For job satisfaction as the mediator, when infor- mal flexible working is benchmarked by no flex- ible working the model fit is satisfactory (SRMSE = 0.027). There is a positive association with job satisfaction (standardized coefficient = 0.1, p-value = .00), but no direct association with performance (standardized coefficient = 0.02, p-value = .33).

Job satisfaction, however, is positively associated with performance (standardized coefficient = 0.16, p-value = .00). Direct effects of informal flexible working on performance are insignificant (esti- mate = 0.03, p-value = .078). As shown in Table V, indirect effects via job satisfaction are positive (0.016, p-value = .00) and support Hypothesis 2a, but the total effect on performance is insignificant.

Comparing informal flexibility over working hours with no flexible working leads to a model that fits the data (SRMSE = 0.027). Job satisfaction (stan- dardized coefficient = 0.04, p-value = .09) and performance (standardized coefficient = –0.001, p-value = .94) are independent of having an informal arrangement over working hours. Job satisfaction is positively associated with performance (standardized coef- ficient = .13, p-value = .00). Neither direct nor indirect effects are signifi- cant (p-values > .05), and thus this subsample rejects Hypothesis 2a.

The model that compares infor- mal remote working against no flex- ible working arrangements (SRMSE = 0.025) supports a positive associa- tion between remote working and job satisfaction (standardized coefficient = 0.09, p-value = .00), which is positively associated with performance (standardized coefficient = 0.15, p-value = .00); yet, it shows no significant direct association between informal remote working and performance (standardized coefficient = 0.03, p-value = .10). As highlighted in Table V, the effects of informal remote working on performance are positive. In contrast to the above, the data on informal remote workers support Hypothesis 2a.

Hypothesis 2b: Formal Flexible Working, Job Satisfaction, and Performance Comparing formal flexible working to no flex- ible working, the model fit is satisfactory (SRMSE = 0.03). The associations flexible working with job satisfaction (standardized coefficient = 0.11, p-value = .00) and job satisfaction with perfor- mance (standardized coefficient = 0.14, p-value = .00) are positive. However, formal flexible work- ing is negatively associated with performance Formal flexible working arrangements can lead to greater organizational commitment, but do not result in higher performance. Human Resource Management DOI: 10.1002/hrm F LEXIBLE W ORKING , INDIVIDUAL PERFORMANCE , AND EMPLOYEE ATTITUDES 1063 (SRMSE = 0.024) and shows no increase in orga- nizational commitment from a formal work arrangement (standardized coefficient = 0.04, p-value = .14). Organizational commitment is positively associated with performance (stan- dardized coefficient = 0.13, p-value = .00) but the impact of formal flexible working on per- formance is lower than where the arrangement is informal (standardized coefficient = –0.08, p-value = .00). As summarized in Table VI, there is no significant difference in indirect effects via organizational commitment (estimate = 0.003, p-value = .156). Estimates of both direct and total effects are negative, meaning that direct and indirect effects via organizational commitment from formal flexible working arrangements are lower than those that could be obtained from informal flexible working arrangement.

When flexibility over working hours is exam- ined, the model fits the sample (SRMSE = 0.026).

There is no difference in the association with organizational commitment (standardized coef- ficient = 0.04, p-value = .20), which is positively Table VI focuses on the mediation models of for- mal flexible working arrangements benchmarked by informal arrangements, thus giving insights into the added value of having formal arrangements as well as the differences between pairs of arrange- ments. It shows the standardized estimates of the effects (direct, indirect via an employee attitude, and total), their standard errors (SEs) and probabil- ity values (p-value) based on which significance is assessed, subsample size (n) and estimated R-square (R 2). Significant estimates are highlighted and sup- port Hypothesis 3 that there are differences in out- come between formal and informal arrangements.

However, the evidence is insignificant with regards to indirect effects via organizational commitment, and mostly the formalization of the arrangement is linked to a decrease in reported employee perfor- mance (negative direct effect).

Hypothesis 4a: The Associations with Organizational Commitment and Performance Comparing formal and informal work- ing arrangements, the model fits the data well TABLE VI Added Value of Formal in Relation to Informal Arrangements   Estimate SEp-Value n R 2 Flexible Working Direct –0.083** 0.023 .000 Via Organizational Commitment 0.005 0.003 .156 Total –0.078** 0.023 .001 0.30 Direct –0.083** 0.023 .000 Via Job Satisfaction 0.004 0.005 .403 Total –0.079** 0.023 .001 1,560 0.30 Flexibility over Working Hours Direct –0.058* 0.028 .034 Via Organizational Commitment 0.005 0.004 .220 Total –0.053 0.028 .057 1,033 0.29 Direct –0.065* 0.027 .019 Via Job Satisfaction 0.013* 0.006 .046 Total –0.052 0.028 .065 1,033 0.30 Remote Working Direct –0.054 0.029 .064 Via Organizational Commitment 0.003 0.005 .512 Total –0.051 0.030 .085 1,014 0.28 Direct –0.057* 0.029 .049 Via Job Satisfaction 0.003 0.006 .640 Total –0.054 0.030 .068 1,014 0.29 Note: Independent variable = 1 if formal arrangement and 0 if informal arrangement.

Control variables: gender (ref. category = male), age (ref. category = less than 29), company.

*: 5% signifi cance level **: 1% signifi cance level.

Gender coeffi cients were not signifi cant in all models. Age (30–39, 40–50) and company coeffi cients signifi cant. 1064 H UMAN RESOURCE M ANAGEMENT , N OVEMBER –D ECEMBER 2017 Human Resource Management DOI: 10.1002/hrm moderate the association with employee atti- tudes was examined. Table IV showed associations between the length of time and organizational commitment (positive) and job satisfaction (nega- tive). Nonetheless, the duration of the flexible working arrangement was unrelated to both job satisfaction and organizational commitment (p-values > .05) in all models. Furthermore, the interactions with different types of flexible work- ing arrangements were also insignificant. Discussion and Conclusion In this article, the relationships between having a flexible working arrangement (specifi- cally remote working and flexibility over work- ing hours) and employee performance have been examined. The mediating roles of organizational commitment and job satisfaction in this relation- ship, largely neglected in extant studies, have also been assessed. In addition, whether there are dif- ferences in these associations depending on how the working arrangement was established, either formally or informally, was examined for flexible working arrangements combined and specifically for remote working and flexibility over work- ing hours. These two types of arrangement were investigated separately because they lend them- selves to being established formally or informally, without involving changes to the formal contract of employment.

Positive association between having a flex- ible working arrangement and organizational commitment and job satisfaction are consistent in the findings and in line with the findings of several previous studies (e.g., Almer & Kaplan, 2002; Cotti, Haley, & Miller, 2014; Harris & Foster, 2005; Hooker, Neathey, Casebourne, & Munro, 2007; Maxwell, Rankine, Bell, & MacVicar, 2007).

Beyond this corroboration, this study has fur- thered understanding of the relationship between flexible working and performance. It has shown, irrespective of the mediator, or the form of the arrangement (formal or informal), that flex- ible working arrangements can lead to positive employee attitudes, which may contribute indi- rectly to employee performance.

As hypothesized, the mediating role of organi- zational commitment may be explained by refer- ence to social and gift exchange theories (Akerlof, 1982; Blau, 1964) as discussed earlier. This is based on the notion that employees reciprocate support in managing the interface between their work and nonwork lives with increased commitment to the organization. The mediating role of job satisfac- tion is explained by reference to the autonomy open to employees over their working arrange- ments (Hackman & Oldman, 1975; Karasek, 1979, related to performance (standardized coefficient = 0.13, p-value = .00). As highlighted in Table VI, those with formal flexibility over working hours reported lower performance than those with an informal arrangement for working hours, because the direct effect is negative.

By contrast, the model that evaluates the impact of formal remote working (SRMSE = 0.02) shows no difference in the associations via orga- nizational commitment. Overall, Hypothesis 4a is rejected.

Hypothesis 4b: The Associations with Job Satisfaction and Performance When formal is compared to informal flex- ible working, the model that accounts for indi- rect effects via job satisfaction fits well (SRMSE = 0.02). The link between the indicator variable and job satisfaction is insignificant (standardized coef- ficient = 0.02, p-value = .40), thus the association with job satisfaction does not vary according to whether the arrangement is formal or informal.

Consequently, effects via job satisfaction are the same, thus rejecting Hypothesis 4b. Nonetheless, the association with performance differs: direct and total effects of formal flexible working are negative (Table VI). When the two forms are compared, for- mal arrangements have a potentially lower impact on performance than informal flexible working.

By contrast, benchmarking formal flexibility over working hours with informal flexibility over work- ing hours (SRMSE = 0.028) shows a positive associa- tion with job satisfaction (standardized coefficient = 0.17, p-value = .04) that counterbalances a nega- tive association with performance (standardized coefficient = –0.15, p-value = .02). The overall total effect, as shown in Table VI, is not signifi- cant (p-value = .065), and there is no difference in the association with performance. Nevertheless, having formal flexibility over working hours is linked to greater effects via job satisfaction (esti- mate = 0.013), thus supporting Hypothesis 4b.

Considering, remote working, the findings confirm what was observed with regard to orga- nizational commitment. The model also fit the data well (SRMSE = 0.02) and showed lower per- formance than those with a formal remote work- ing arrangement, but no differences in the direct or the indirect association via job satisfaction.

Taking together the different types of arrange- ments, Hypothesis 4b is rejected by flexible and remote working, but is supported by flexibility over working hours. Time as a Moderator The proposition that the length of time the individual had been working flexibly would Human Resource Management DOI: 10.1002/hrm F LEXIBLE W ORKING , INDIVIDUAL PERFORMANCE , AND EMPLOYEE ATTITUDES 1065 job satisfaction, a sense of independence enabled by having flexibility (Tietze et al., 2009) may be enhanced by exercising some choice over work- ing arrangements. However, this is not the case with having flexibility over working hours on an informal basis, because the results indicate that this type of arrangement may lead to greater job satisfaction when formalized. Formal flexible working arrangements were found to be negatively associated with perfor- mance. It is possible that those with a formal flex- ible working arrangement simply perform poorly compared to other groups. Reduced face time with their managers and coworkers may have a detri- mental effect on performance. Less close super- vision, or fewer opportunities for training for example may contribute to lower performance.

Equally, where employees’ performance is depen- dent on coordination with coworkers, this may be hampered by some forms of flexible working.

However, the process itself may also have a role to play. When making a formal request to change working arrangements, employees may be more likely to see themselves as tak- ing advantage of a benefit that they are entitled to and consequently see no need to reciprocate in relation to performance (Lewis & Smithson, 2001). In addition, the extent to which an employee believes that the formal process operates fairly could be important here.

Beauregard (2014) argues that per- ceived unfairness in the allocation of work-life initiatives can lead to counterproductive work behaviors, which could then result in lower performance rat- ings. A formal process may be more susceptible to perceptions of unfairness, partially because it is likely to be more readily observable and may involve a more bureaucratic and lengthy process.

However, other studies have suggested the reverse, that the existence of informal arrangements may be seen to be less equitable (Fogarty et al., 2011; Golden, 2009).

Several studies have highlighted the impor- tance of line manager support for flexible work- ing (see for example Bagger & Li, 2014; Ryan & Kossek, 2008). With a formal arrangement, the extent to which the line manager is supportive of the arrangement is less clear. Unlike an infor- mal arrangement, which is only likely to be established if the line manager is supportive, the decision to grant a formal arrangement is unlikely to be made by the line manager alone and there- fore the existence of such an arrangement is less of an indicator of his or her support. Line managers 1989). Essentially, the discretion given to employ- ees over when and where they work is likely to foster enhanced job satisfaction.

While investigating differences in performance ratings between forms of arrangement, in the first instance, mean and variance of performance scores were examined. The variances in the distribution of performance ratings were the same, irrespective of the arrangement form (nonflexible, formal flex- ible, informal flexible). However, there were dif- ferences in means (p-value = 0) between flexible workers with an informal arrangement (mean = 2.87), flexible workers with a formal arrangement (mean = 2.67), and those who did not work flex- ibly (mean = 2.79). Thus, having a flexible working arrangement established through an informal pro- cess appears to enhance performance. By contrast, the formal process may have a negative effect on performance, as indicated by the negative direct effects of formal arrangements on performance that were reported above. There are several pos- sible explanations for these findings.

Those with informal arrangements may have received higher performance scores due to increased effort exercised in return for the flexibil- ity extended to them, in line with social or gift exchange theories, discussed earlier. It could be that an informal arrangement is more likely to engen- der a social exchange relationship because it will have resulted from direct negotiation between the employee and his or her line manager (Atkinson & Hall, 2009). As such, it may be possible to accom- modate personal circumstances to a greater degree than where the arrangement is set up through a formal mechanism (Eaton, 2003; Hall & Atkinson, 2006). Employees may feel that it is their line manager who has offered them the benefit and therefore reciprocate with exercising greater effort. Specifically, in line with gift exchange, line managers may offer a greater degree of custom- ization in the anticipation that this will enhance performance. In addition, with an informal flex- ible working arrangement the employee may take greater steps to protect the security of this arrange- ment (Kelliher & Anderson, 2010), such as increas- ing effort. However, no direct positive association between informal flexible working and perfor- mance rating was established, but rather indirect links via employee attitudes. Previous studies have suggested that flexible working can enhance effort (Golden, 2001, 2009; Kelliher & Anderson, 2010), but it may be that effort, or employees’ perception of enhanced effort, does not necessarily translate into higher performance ratings.

Because the relationships between having informal and formal flexible working arrange- ments and performance were also mediated by Having a flexible working arrangement established through an informal process appears to enhance performance. 1066 H UMAN RESOURCE M ANAGEMENT , N OVEMBER –D ECEMBER 2017 Human Resource Management DOI: 10.1002/hrm 2012), can increase job satisfaction. Furthermore, where employees are required to work additional time, due to business demands, they are also likely to be able to recoup this later at a time of their choosing. These differences underscore the impor- tance of examining different types of flexibility separately (Allen et al., 2013; Butts et al., 2013).

Because the positive outcomes from having a flexible working arrangement were not subject to the duration of the flexible working arrange- ment, it appears that autonomy is important in understanding the link with employee attitudes and performance is less likely to dilute over time.

This finding adds further support to the job char- acteristics model (Hackman & Oldham, 1975) as a means to explain the outcomes of flexible work- ing arrangements.

Given that the data included a higher propor- tion of males than most extant studies, additional analyses investigated whether the findings could be moderated by gender but found no evidence.

Nonetheless, future studies should examine how individual characteristics may affect the associa- tions between forms of flexible working arrange- ments and employee outcomes, including, for example, health and well-being.

Limitations This is a cross-sectional study, with assessments of causality relying on the theories underlying the models tested; however, reverse direction of cau- sality cannot be ruled out. A longitudinal study of flexible workers with more general measures of performance from different sources would enable a clearer assessment of potential outcomes from different arrangements to be made. In this study, performance has been measured by performance ratings reported by the respondent, which are sub- ject to accurate recall and reporting. Furthermore, factors such as work climate, or the fit between working patterns and business activity, which may influence individual performance have not been addressed. Moreover, a limited number of flexible working types were considered and some were subject to overlap. Butts et al. (2013) suggest that in relation to work-family support the posi- tive effect on attitudes increases with the number of policies. This study was based on professional employees in organizations where flexible work- ing policies were well established. It is possible that by examining other types of employees or in organizations where flexible working arrange- ments were less well established, different results might have emerged. Although no effect for the length of time the arrangement had been in place was observed, it is noteworthy that most arrange- ments had been in place for less than four years. who are not completely supportive of a flexible working arrangement may find their assessment of the flexible worker’s performance, consciously or otherwise, influenced by their beliefs about the arrangement. In addition, managers, who have not received adequate training in manag- ing flexible workers, may find it more difficult to manage and assess the performance of employees that are mostly away from the workplace or pres- ent at different times, thus resulting in lower rat- ings. Likewise, several studies have suggested that employees who request flexible working arrange- ments are perceived as being less serious about their careers by managers (Leslie, Manchester, Park, & Mehng, 2012), which may affect the way in which their performance is rated. However, analysis of the variance in performance ratings in this sample did not show that their distribution varied between informal, formal, and nonflexible workers; therefore, it cannot be concluded that line managers’ ratings of perfor- mance were biased. In line with the call by Allen, Johnson, Kiburz, and Shockley (2013), these findings sug- gest a need to understand the role of informal workplace and supervi- sor support more fully.

When the two types of flexible working were examined separately, there was no difference in the media- tion via organizational commitment.

Yet having a formal arrangement for flexibility over working hours may lead to greater job satisfaction than will having an informal arrange- ment. Formal mechanisms for flex- ibility over working hours tend to be based around total hours worked, which may be counter to the nature of managerial and profes- sional work (Kalleberg & Epstein, 2001; Perlow & Porter, 2009), where employees are often expected to put in the hours required to get the job done.

Formal flexitime schemes often have a predefined window of choice over working time, together with prescribed “core time” and an agreement on the total number of hours to be worked. In prac- tice, this may mean that employees actually work less under such a scheme, which in turn influences performance ratings negatively and therefore the positive association with organizational commit- ment does not translate into performance. Formal flexibility over working hours supports mediation via job satisfaction, consistent with the job char- acteristics model. The ability to adjust working hours according to the employees’ preferences and importantly to feel that the arrangement is secure, protected by the formal system (Hutchinson, Having a formal arrangement for flexibility over working hours may lead to greater job satisfaction than will having an informal arrangement. Human Resource Management DOI: 10.1002/hrm F LEXIBLE W ORKING , INDIVIDUAL PERFORMANCE , AND EMPLOYEE ATTITUDES 1067 More generally, these findings raise questions about the formalization of human resource man- agement (HRM) processes and practices in organi- zations. Research by Storey, Saridakis, Sen-Gupta, Edwards, and Blackburn (2010) found that formal- ization of HRM influences employee perceptions of job quality in a negative way and argue that it is because they impinge on autonomy and discre- tion. For large organizations offering flexible work- ing arrangements such as those in this study, it is important to examine how line managers can be afforded greater autonomy in the implementation of HR practices to yield the potential benefits from informal and more personalized arrangements.

This study adds to knowledge and under- standing about the consequences of individualiz- ing the employment relationship. In the context of flexible working arrangements customized to meet the needs of employees, the positive associ- ation with employee attitudes was confirmed. In addition, performance outcomes were found to vary according to the way in which the arrange- ment was established, adding to knowledge about the outcomes of informalization of the employment relationship. This finding is impor- tant in a context, where informal flexible work- ing arrangements are prevalent, even where a formal right to request flexible working arrange- ments exists. Note 1. In countries where the right to request is enshrined in law, there is normally an obligation on the employer to respond in a particular way and in a designated time frame. Management practices may take several years to affect performance (Powell, 1995) and therefore studies examining flexible working arrangements over a longer period of time may reveal more insight into the effect of time.

Implications An important message for managers is the role flexible working arrangements may have in gener- ating positive organizational outcomes. Allowing employees choice over their working arrange- ments may have a marked influence on how they view their employment relationship, which may translate into greater organizational commitment and job satisfaction, which are then associated with performance gains.

The differences observed in formal and infor- mal arrangements are also important for manag- ers and policymakers. These findings highlight a tension between promoting fairness and equity through formalizing processes and an apparent detrimental impact on performance. However, as discussed above, the processes may also influence results. In order not to disadvantage those with formal arrangements, it is important to examine the implementation of flexible working and to provide support for managers who manage flex- ible workers and for flexible workers themselves.

While maintaining a fair and equitable process, there may be scope to encourage greater dialogue between the employee and his or her line manager within a formal process and thereby customize the arrangement to a greater degree, which may in turn facilitate reciprocal behaviors of benefit to the organization. LILIAN M. DE MENEZES is professor of Decision Sciences, Cass Business School, City, Univer- sity of London, which she joined in 2002. Her research contributions range from studies of the changing world of work to the development of forecasting and statistical methodology. Her publications include articles in European Journal of Operational Research, Human Relations, Industrial Relations, International Journal of Human Resource Management, International Journal of Production and Operations Management, Journal of Operations Management, and Journal of the Royal Statistical Society. She has accumulated industrial experience from work- ing in a range of projects in operations for consulting and fi nancial organizations.

CLARE KELLIHER, BSc, MA, PhD, is professor of Work and Organisation at Cranfi eld School of Management, Cranfi eld University. Her research interests center on the organization of work and the management of the employment relationship. She has a long-standing interest in fl exible working and has directed a major project concerned with examining the impact of fl exible working on performance, sponsored by seven companies. She is the author of many published papers and book chapters and regularly speaks at national and international con- ferences. Her recent book, New Ways of Organising Work: Developments, Perspectives and Experiences, coedited with Julia Richardson, was published by Routledge. 1068 H UMAN RESOURCE M ANAGEMENT , N OVEMBER –D ECEMBER 2017 Human Resource Management DOI: 10.1002/hrm Butts, M. M., Casper, W. J., & Yang, T. S. (2013). How impor- tant are work-family support policies? A meta-analytic investigation of their effects on employee outcomes.

Journal of Applied Psychology, 98(1), 1–25.

Capelli, P. (2000, January–February). A market driven approach to retaining talent. Harvard Business Review, 2000, 103–111.

Casper, W. J., Eby, L. T., Bordeaux, C., & Lockwood, A. (2007).

A review of methods in IO/OB work-family research.

Journal of Applied Psychology, 92(1), 28–41.

Chartered Institute of Personnel and Development (CIPD).

(2012). Flexible working provision and uptake. London, England: Author.

Chow, I. H., & Keng-Howe, I. C. (2006). The effect of alterna- tive work schedules on employee performance. Interna- tional Journal of Employment Studies, 14(1), 105–131.

Cook, J., & Wall, T. (1980). New work attitude measures of trust, organizational commitment and personal need non-fulfi lment. Journal of Occupational Psychology, 53(1), 39–52.

Cotti, C. D., Haley, M. R., & Miller, L. A. (2014). Workplace fl exibilities, job satisfaction and union membership in the US workforce. British Journal of Industrial Relations, 52(3), 403–425.

Coyle-Shapiro, J. A.-M., & Shore, L. M. (2007). The employee- organization relationship. Where do we go from here?

Human Resource Management Review, 17(2), 166–179.

de Janasz, S., Forret, M., Haack, D., & Jonsen, K. (2013).

Family status and work attitudes: An investigation in a professional services fi rm. British Journal of Manage- ment, 24(2), 191–210.

de Menezes, L., & Kelliher, C. (2011). Flexible working and performance: A systematic review of the evidence for a business case. International Journal of Management Reviews, 13(4), 452–474.

de Menezes, L. M. (2011). Job satisfaction and quality man- agement: An empirical analysis. International Journal of Operations and Production Management, 32(3), 308–328.

den Dulk, L., Peters, P., & Poutsma, E. (2012). Variations in adoption of workplace work-family arrangements in Europe: The infl uence of welfare-state regime and organi- zational characteristics. International Journal of Human Resource Management, 23(13), 2785–2808.

Dodd, N. G., & Ganster, D. C. (1996). The interactive effects of variety, autonomy, and feedback on attitude and per- formance. Journal of Organizational Behavior, 17(4), 329–347.

Earl, C., & Taylor, P. (2015). Is workplace fl exibility good policy? Evaluating the effi cacy of age management strat- egies for older women workers. Work, Aging and Retire- ment, 1(2), 214–226.

Eaton, S. (2003). If you can use them: Flexibility policies, organizational commitment and perceived performance.

Industrial Relations, 42(2), 145–167.

Eby, L. T., Casper, W. J., Lockwood, A., Bordeaux, C., & Brin- ley, A. (2005). Work and family research in IO/OB: Content analysis and review of the literature (1989–2002). Journal of Vocational Behavior, 66(1), 124–197.

European Commission (2012). EUR 25270—new skills and jobs in Europe: Pathways towards full employment. Lux- embourg: Publications Offi ce of the European Union. References Akerlof, G. A. (1982). Labor contracts as partial gift exchange.

Quarterly Journal of Economics, 97(4), 543–569.

Allen, T. D. (2001). Family-supportive work environments: The role of organizational perceptions. Journal of Vocational Behavior, 58(3), 414–435.

Allen, T. D., Johnson, R. C., Kiburz, K. M., & Shockley, K. M. (2013).

Work-family confl ict and fl exible work arrangements: Decon- structing fl exibility. Personnel Psychology, 66(2), 345–376.

Almer, E. D., & Kaplan, S. E. (2002). The effects of fl ex- ible working arrangements on stressors, burnout, and behavior job outcomes in public accounting. Behavioural Research in Accounting, 14(2), 1–34.

Anderson, S., Coffey, B., & Byerly, R. (2002). Formal organi- zational initiatives and informal workplace practices:

Links to work-family confl ict and job-related outcomes.

Journal of Management, 28(6), 787–810.

Atkinson, C., & Hall, L. (2009). The role of gender in varying forms of fl exible working. Gender, Work and Organisa- tion, 16(6), 650–666.

Bagger, J., & Li, A. (2014). How does supervisory family sup- port infl uence employees’ attitudes and behaviours? A social exchange perspective. Journal of Management, 40(4), 1123–1150.

Bal, P. M, de Jong, S. B., Jansen, P. G. W., & Bakker, A. B.

(2012). Motivating employees to work beyond retirement:

A multi-level study of the role of i-deals and unit climate.

Journal of Management Studies, 49(2), 306–331.

Baltes, B., Briggs, T., Huff, J., Wright, J., & Neuman, G. (1999).

Flexible and compressed workweek schedules: A meta- analysis of their effects on work-related criteria. Journal of Applied Psychology, 84(4), 496–513.

Baruch, Y., & Holtom, B. C. (2008). Survey response rate levels and trends in organizational research. Human Rela- tions, 61(8), 1139–1160.

Beauregard, T. A. (2014). Fairness perceptions of work- life balance initiatives: Effects on counterproductive work behaviour. British Journal of Management.

doi:10.1111/1467-8551.12052 Beck, V. (2013). Employers’ use of older workers in the reces- sion. Employee Relations, 35(3), 257–271.

Becker, B. E., & Huselid, M. (1998). High performance work systems and fi rm performance: A synthesis of research and managerial implications. In G. R. Ferris (Ed.), Research in personnel and human resources (Vol. 16, pp.

53–101). Stanford, CT: JAI Press.

Bickart, B., & Schmittlein, D. (1999). The distribution of sur- vey contact and participating in the United States: Con- structing a survey-based estimate. Journal of Marketing Research, 36(2), 286–294.

Blau, P. M. (1964). Exchange and power in social life. New York, NY: Wiley.

Bloom, N., & Van Reenen, J. (2006). Management practices, work-life balance and productivity: A review of some recent evidence. Oxford Review of Economic Policy, 22(4), 457–482.

Boxall, P., & Purcell, J. (2003). Strategy and human resource management. Basingstoke, England: Palgrave Macmillan.

Budd, J. W., & Mumford, K. A. (2006). Family-friendly prac- tices in Britain: Availability and perceived accessibility.

Human Resource Management, 45(1), 23–42. Human Resource Management DOI: 10.1002/hrm F LEXIBLE W ORKING , INDIVIDUAL PERFORMANCE , AND EMPLOYEE ATTITUDES 1069 tional outcomes? A meta-analytic investigation of medi- ating mechanisms. Academy of Management Journal, 55(6), 1264–1294.

Kalleberg, A. L., & Epstein, C. F. (2001). Introduction: Tem- poral dimensions of employment relations. American Behavioral Scientist. Special Issue: Time and the Employ- ment Relationship, 44(7), 1064–1075.

Karasek, R. (1989). Control in the workplace and its health- related aspects. In S. L. Sauter, J. J. J. Hurrell, & C. L.

Cooper (Eds.), Job control and worker health (pp. 129– 159). London, England: Wiley.

Karasek, R. A. J. (1979). Job demands, job decision latitude and mental strain: Implications for job redesign. Adminis- trative Science Quarterly, 24, 285–308.

Kelliher, C., & Anderson, D. (2008). For better or for worse?

An analysis of how fl exible working practices infl uence employees’ perceptions of job quality. International Jour- nal of Human Resource Management, 19(3), 421–433.

Kelliher, C., & Anderson, D. (2010). Doing more with less?

Flexible working practices and the intensifi cation of work.

Human Relations, 63(1), 83–106.

Kelly, E. L., & Kalev, A. (2006). Managing fl exible work arrangements in US organizations: Formalized discretion or “a right to ask.” Socio-Economic Review, 4(3), 379–416.

Kelly, E. L., & Moen, P. (2007). Rethinking the clock cork of work: Why schedule control may pay off at work and at home. Advances in Developing Human Resources, 9(4), 487–506.

Konrad, A., & Mangel, R. (2000). The impact of work-life programs on fi rm productivity. Strategic Management Journal, 21(12), 1225–1237.

Kossek, E. E., Lautsch, B. A., & Eaton, S. C. (2005). Flexibility enactment theory: Implications of fl exibility type, control and boundary management for work-family effective- ness. In E. E. Kossek & S. J. Lambert (Eds.), Work and life integration: Organisational, cultural and individual per- spectives (pp. 243–261). Mahwah, NJ: Erlbaum.

Lambert, S. J., & Waxman, E. (2005). Organization strati- fi cation: Distributing opportunities for balancing work and personal life. In E. E. Kossek & S. J. Lambert (Eds.), Work and life integration: Organizational, cultural, and individual perspectives (pp. 103–126). Mahwah, NJ: Erl- baum.

Lapierre, L. M., & Allen, T. D. (2006). Work-supportive fam- ily, family-supportive supervision, use of organizational benefi ts, and problem-focused coping: Implications for work-family confl ict and employee well-being. Journal of Occupational Health Psychology, 11(2), 169–181.

Lawler, E. E. (1986). High involvement management: Par- ticipative strategies for improving organizational perfor- mance. San Francisco, CA: Jossey-Bass.

Leslie, L., Manchester, C., Park, T., & Mehng, S. A. (2012).

Flexible work practices: A source of career premiums or penalties? Academy of Management Journals, 55(6), 1407–1428.

Lewis, S., & Smithson, J. (2001). Sense of entitlement to support for the reconciliation of employment and family life. Human Relations, 54(11), 1455–1481.

Matos, K., & Galinsky, E. (2012). 2012 National study of employers. New York, NY: Families and Work Institute.

Matos, K., & Galinsky, E. (2014). 2014 National study of employers. New York, NY: Families and Work Institute. Felstead, A., Jewson, N., Phizacklea, A., & Walters, S. (2002).

Opportunities to work at home in the context of work-life bal- ance. Human Resource Management Journal, 12(1), 54–76.

Fogarty, H., Scott, P., & Williams, S. (2011). The half-empty offi ce: Dilemmas in managing locational fl exibility. New Technology, Work and Employment, 26(3), 183–195.

Furunes, T., Mykletun, R. J., & Solem, P. E. (2011). Age man- agement in the public sector in Norway: Exploring man- agers’ decision latitude. International Journal of Human Resource Management, 22(6), 1232–1247.

Gajendran, R. S., & Harrison, D. A. (2007). The good, the bad, and the unknown about telecommuting: Meta-analysis of psychological mediators and individual consequences.

Journal of Applied Psychology, 92(6), 1524–1541.

Golden, L. (2001). Flexible work schedules: Which workers get them? American Behavioural Scientist, 44(7), 1157–1178.

Golden, L. (2009). Flexible daily work schedules in US jobs:

Formal introductions needed? Industrial Relations, 48(1), 27–54.

Golden, L. (2012). The effects of working time on productiv- ity and fi rm performance: A research synthesis paper.

Geneva, Switzerland: International Labour Offi ce.

Gouldner, A. W. (1960). The norm of reciprocity: A preliminary statement. American Sociological Review, 25(2), 161–178.

Gregory, A., & Milner, S. (2009). Work-life balance a matter of choice? Gender, Work and Organization, 16(1), 1–13.

Hackman, J. R., & Oldham, G. (1975). Development of the job diagnostic survey. Journal of Applied Psychology, 60(2), 159–170.

Hall, L., & Atkinson, C. (2006). Improving working lives: Flex- ible working and the role of employee control. Employee Relations, 28(4), 374–386.

Hammer, L. B., Neal, M. B., Newson, J. T., Brockwood, K. J., & Colton, C. L. (2005). A longitudinal study of the effects of dual-earner couples’ utilization of family-friendly work- place supports on work and family outcomes. Journal of Applied Psychology, 90(4), 799–810.

Harris, L., & Foster, C. (2005). Small, fl exible and family friendly: Work practices in service sector businesses. Lon- don, England: Department of Trade and Industry.

Healy, G. (2004). Work-life balance and family friendly poli- cies—in whose interest? Work, Employment and Society, 18(1), 219–223.

Hooker, H., Neathey, F., Casebourne, J., & Munro, M. (2007).

The third work-life balance employee survey: Main fi nd- ings. Brighton, England: Institute for Employment Studies.

Hornung, S., Rousseau, D. M., & Glaser, J. (2008). Creating fl exible work arrangements through idiosyncratic deals.

Journal of Applied Psychology, 93(3), 655–664.

Hutchinson, C. (2012, February 8–10). The impact of fl exible work arrangements on workers work-life outcomes: A study of public sector workers in South Australia. The Australian and New Zealand Industrial Relations Confer- ence, Surfers Paradise, Queensland.

Ibarra, H. (1999). Provisional sales: Experimenting with image and identity in professional adaptation. Adminis- trative Science Quarterly, 44, 764–791.

Jafri, M. (2010). Organizational commitment and employee’s innovation behaviour: A study in retail sector. Journal of Management Research, 10, 62–68.

Kaifeng, J., Lepack, D. J., Jia, J., & Baer, J. C. (2012). How does human resource management infl uence organiza- 1070 H UMAN RESOURCE M ANAGEMENT , N OVEMBER –D ECEMBER 2017 Human Resource Management DOI: 10.1002/hrm Maxwell, G., Rankine, L., Bell, S., & MacVicar, A. (2007). The incidence and impact of fl exible working arrangements in smaller businesses. Employee Relations, 29(2), 138–152.

Muthén, L. K., & Muthén, B. O. (1998–2010). Mplus user’s guide (6th ed.). Los Angeles, CA: Muthén & Muthén.

Ortega, J. (2009). Why do employers give discretion? Family versus performance concerns. Industrial Relations, 48(1), 1–24.

Osterman, P. (1995, December). Work/family programs and the employment relationship. Administrative Science Quarterly, 1995, 681–701.

Ozbilgin, M. F., Beauregard, T. A., Tatli, A., & Bell, M. P. (2011).

Work-life, diversity and intersectionality: A critical review and research agenda. International Journal of Manage- ment Review, 13, 177–198.

Perlow, L. A., & Porter, J. L. (2009). Making time off predict- able—and required. Harvard Business Review, 87(10), 102–109.

Pfeffer, J. (1994). Competitive advantage through people:

Unleashing the power of the workforce. Harvard, MA:

Harvard Business School Press.

Powell, T. C. (1995). Total quality management as competitive advantage: A review and empirical study. Strategic Man- agement Journal, 16(1), 15–37.

Purcell, J., & Kinnie, N. (2007). HRM and business perfor- mance. In P. Boxall, J. Purcell, & R. Wright (Eds.), Oxford handbook of human resource management (pp. 533– 552). Oxford, England: Oxford University Press.

Rapoport, R., Bailyn, L., Fletcher, J. K., & Pruitt, B. H. (2002).

Beyond work-family balance: Advancing gender equity and workplace performance. San Francisco, CA: Jossey-Bass.

Richman, A. L., Civian, J. T., Shannon, L. L., Hill, J., & Bren- nan, R. T. (2008). The relationship of perceived fl exibility, supportive work-life policies and use of formal fl exible arrangements and occasional fl exibility to employee engagement and expected retention. Community, Work and Family, 11(2), 183–197.

Rosen, C. C., Slater, D. J., Chang, C. H., & Johnson, R. E.

(2013). Let’ s make a deal: Development and validation of the ex post i-deals scale. Journal of Management, 39, 709–742.

Rousseau, D. M. (2005). I-deals: Idiosyncratic deals employ- ees bargain for themselves. New York, NY: M. E. Sharpe.

Rousseau, D. M., Ho, V. T., & Greenberg, J. (2006). I-deals: Idi- osyncratic terms in employment relationships. Academy of Management Review, 31(4), 977–994.

Ryan, A. M., & Kossek, E. E. (2008). Work-life policy imple- mentation: Breaking down or creating barriers to inclu- siveness? Human Resource Management, 47(2), 295–310.

Scandura, T., & Lankau, M. (1997). Relationships of gender, family responsibility and fl exible work hours to organi- sational commitment and job satisfaction. Journal of Organizational Behavior, 18(4), 377–391.

Schneider, B., Hanges, P. J., Brent Smith, D., & Salvaggio, A. N. (2003). Which comes fi rst: Employee attitudes or organizational fi nancial market performance. Journal of Applied Psychology, 88(5), 836–851.

Settoon, R. P., Bennett, N., & Liden, R. C. (1996). Social exchange in organizations: Perceived organizational sup- port, leader-member exchange. Journal of Applied Psy- chology, 81(3), 219–227. Skinner, N., Hutchinson, C., & Pocock, B. (2012). The big squeeze: Work, home and care in 2012. Adelaide: Centre for Work and Life, University of South Australia.

Spector, P. E. (1986). Perceived control by employees: A meta-analysis of studies concerning autonomy and participation at work. Human Relations, 39(11), 1005– 1016 .

Stavrou, E., & Ierodiakonou, C. (2015). Entitlement to work- life balance support: Employee/manager perceptual discrepancies and their effect on outcomes. Human Resource Management, doi:10.1002/hrn.21745 Storey, D. J., Saridakis, G., Sen-Gupta, S., Edwards, P. K., & Blackburn, R. A. (2010). Linking HR formality with employee job quality: The role of fi rm and workplace size.

Human Resource Management, 49(2), 305–329.

Thomas, J. P., Whitman, D. S., & Viswesvaran, C. (2010).

Employee proactivity in organizations: A comparative meta-analysis of emergent proactive constructs. Journal of Occupational and Organizational Psychology, 83(Part 2), 275–300.

Tietze, S., Musson, G., & Scurry, T. (2009). Homebased work:

A review of research into themes, directions and implica- tions. Personnel Review, 38(6), 585–604.

Tipping, S., Chanfreau, J., Perry, J., & Tait, C. (2012). The fourth work-life balance employee survey. London, England: Department for Business Innovation and Skills.

Troup, C., & Rose, J. (2012). Telecommuting: Does work- ing from home have positive or negative outcomes for all employees? Community, Work and Family, 15(4), 472–486.

van Wanrooy, B., Bewley, H., Forth, J., Freeth, S., Stokes, L., & Wood, S. (2013). The 2011 Workplace Employment Relations Study: First fi ndings. London, England: Depart- ment of Business, Innovation and Skills. Retrieved from https://www.gov.uk/government/uploads/system/uploads/ attachment_data/fi le/336651/bis-14-1008-WERS-fi rst - fi ndings-report-fourth-edition-july-2014.pdf Wood, S., & de Menezes, L. (1998). High commitment management in the UK: Evidence from the workplace industrial relations survey and the employer’s man- power skills practices survey. Human Relations, 51(4), 485–515.

Wood, S., de Menezes, L. M., & Lasaosa, A. (2003). Family- friendly management in Great Britain: Testing various perspectives. Industrial Relations, 42, 221–250.

Wood, S., van Veldhoven, M., Croon, M., & de Menezes, L.

M. (2012). Enriched job design, high involvement man- agement and organizational performance: The mediating roles of job satisfaction and well-being. Human Relations, 65(4), 419–446.

Wright, B., & Schwager, P. (2008). Online survey research:

Can response factors be improved? Journal of Internet Commerce, 7(2), 253–269.

Wright, P. M., Gardner, T. M., & Moynihan, L. M. (2003).

The impact of HR practices on the performance of busi- ness units. Human Resource Management Journal, 13, 21–36.

Yee, R. W. Y., Yeung, A. C. L., & Cheng, T. C. E. (2010). An empirical study of employee loyalty, service quality and fi rm performance in the service industry. International Journal of Production Economics, 124(1), 109–120. Copyright ofHuman Resource Management isthe property ofJohn Wiley &Sons, Inc.and its content maynotbecopied oremailed tomultiple sitesorposted toalistserv without the copyright holder'sexpresswrittenpermission. However,usersmayprint, download, oremail articles forindividual use.