Abstract
The value that employees attach to the intrinsic aspects of work is important for whether or not job quality issues should have a central place on the social agenda. This article examines whether the importance that British employees attach to intrinsic job quality changed between 1992 and 2006. It uses two nationally representative surveys of employees. It finds no evidence to support the view that there has been a shift towards instrumental job preferences. On the contrary, it shows that intrinsic job preferences rose over the period. The growth in importance of intrinsic orientations is associated with rising levels of education and parental encouragement in education, the improvement of people’s jobs with respect to skill, learning opportunities and employee involvement and higher incomes and security.
Introduction
The importance employees attach to the intrinsic aspects of work is of central importance to debates about the quality of work. Since the Lisbon summit of 2000, the European Union has placed the achievement of better jobs as one of the key objectives of the Union alongside that of higher competitiveness (European Commission, 2008). But how important is the intrinsic quality of work tasks for employees and is its importance increasing or diminishing? The literature provides quite diverse speculative scenarios. Some have argued that employees are becoming increasingly instrumental, focusing primarily upon the economic rewards of work; others have suggested that needs for self-realization in work will become more salient. If it is the case that employees have been becoming less attached to the intrinsic aspects of work, then policies of job enrichment would seem of relatively low importance. If, on the other hand, expectations have been rising, there should be an increasing urgency to find ways of improving the intrinsic quality of jobs in the interests both of employee motivation and of employee well-being. However, there is currently little empirical evidence about the way in which aspirations with respect to intrinsic job quality have been changing. This article seeks to redress this lack by comparing the importance that British employees attached to job quality (that is, the quality of their work tasks) over the decade and a half between 1992 and 2006 and by examining the major factors that affect the salience of intrinsic job preferences.
Growing instrumentalism or rising needs for self-realization in work?
An influential early contribution to debates about job preferences in British sociology (Goldthorpe et al., 1968) advanced the view that employees were coming to attach less importance to the intrinsic nature of the work and were increasingly selecting jobs primarily on the basis of the extrinsic rewards they offered, particularly in terms of income. As workers became more affluent and traditional patterns of community life were replaced by the growing importance of the nuclear family, family and leisure life would come to have an increasingly dominant place in people’s overall values. A concern that jobs should provide opportunities to use initiative and skills would be replaced by an emphasis on the income that the job generated as a way of meeting non-work aspirations.
However, in more recent decades, there have been important aspects of social change that raise doubts about the expectation of increased instrumentalism. First, there has been a sharp rise in educational levels. This may have had an influence directly through longer experience of study or indirectly through the way it affects parental upbringing. Cross-sectional studies indicate that those with more advanced education attach a greater importance to intrinsic rewards from work (Gallie et al., 1998; Rose, 2005; Russell, 1998) and some longitudinal evidence confirms a causal effect (Lindsay and Knox, 1984). The experience of education may instill an expectation of progressive self- development, leading to a desire for work that provides opportunities for continued learning and skill development. The effects of this could be amplified through inter-generational transmission. There is an association between parental education and intrinsic job values (Johnson et al., 2007). By extension, rising levels of parental education across time may lead to children placing a greater value on personal development and self-realization.
There was also little in the original argument that anticipated the major change in the gender composition of the workforce. The early contributions to the debate about job preferences were based upon research into the attitudes of male workers. But the following decades witnessed a rapid expansion in women’s labour market participation. At the same time there was a marked rise in women’s educational levels and a decline in the time that mothers took away from employment on childbirth. The increased integration of women into the mainstream workforce raises the possibility that a decline in traditional role identifications may have encouraged a shift away from instrumentalism to a greater emphasis on the quality of experience in the job.
A major critique of the ‘instrumentalism’ thesis was that job preferences may be affected as much by the work context itself as by developments in the family and community (Beynon and Blackburn, 1972; Brown, 1973; Gallie, 2007). Longitudinal studies subsequently confirmed that the quality of people’s jobs has an effect on the importance they attach to intrinsic work values (Mortimer and Lorence, 1979; Mortimer et al., 1996). In the 1960s, when the ‘instrumentalism’ thesis was being developed, the increased use of large-scale mechanized technologies was providing declining opportunities for self-fulfilment through work as a result of the deskilling of manual work. It was understandable that workers in this context adaptively redefined their life priorities in order to preserve a meaningful sense of identity.
However, the evolution of work in more recent decades may have helped to reverse earlier tendencies towards increasing instrumentalism. A significant body of research suggests that the nature of technical change, in particular with increased automation, has the potential to improve the opportunities for skill use (Applebaum, 2002; Badham, 2005; Chase and Karwowski, 2003). Moreover, some prominent management theories have come to emphasize the importance of skill enrichment and greater employee involvement in meeting rising demands for product quality in an increasingly competitive international environment (Walton, 1985), although it can be questioned how far this has informed actual management policy (Legge, 2001). Certainly, there has been a marked overall tendency for jobs to become more skilled over time, although there is also evidence of polarization (Autor et al., 2003). Moreover, the period has seen an attempt by employers to build up identification with work and the company by introducing new forms of direct consultation with employees (Willman et al., 2009). A number of studies have found that such practices increase employees’ intrinsic rewards from work and hence their organizational commitment (Applebaum et al., 2000; Wood and De Menezes, 2007; although for a more sceptical view, see Godard, 2001; Ramsay, 1977).
Finally, it is notable that the underlying assumptions about the motivational effects of increased prosperity were in sharp contrast to those offered by some influential psychological theories of the time. For instance, Maslowian and neo-Maslowian theories of psychological needs (Alderfer, 1969; Maslow, 1954) postulated that higher incomes, far from undermining the importance that people attached to the nature of their jobs, would lead to a higher priority for self-realization in work. Individuals, it was argued, have a hierarchy of needs in which they initially give priority to material and security needs, but attach a greater importance to higher order needs, in particular a concern with personal self-realization, once their basic needs have been met. This implies that higher incomes will be accompanied by an increasing desire by employees to make use of, and develop, their skills in the sphere of work. There has certainly been a growth in real incomes over time. There has been much debate about underlying changes in job security. While there is no consistent evidence that jobs have become less stable as a result of new employer ‘flexibility policies’, job security is subject to a strong cyclical pattern, reflecting the business cycle (McGovern et al., 2007).
These different arguments lead then to a number of contrasting hypotheses about the way intrinsic job preferences have been changing:
The importance of intrinsic job preferences may have declined due to a shift in life priorities made possible by rising incomes.
Increased levels of education may have strengthened intrinsic preferences either through their direct impact or their indirect effects on parental socialization.
Higher levels of integration into the labour market may have led to normative shifts that encouraged a greater concern of female workers with the intrinsic aspects of work.
Improved job quality may have increased the salience of intrinsic preferences by increasing opportunities for self-realization and a more positive self-identity.
The importance of intrinsic preferences may have increased as basic economic and security needs became less salient due to the protection offered by rising incomes.
The following sections of the article first discuss the data that are drawn upon and the key measures of job preferences. Second, an attempt is made to present a clearer picture of the changes in intrinsic job preferences between 1992 and 2006. The third section examines whether the socialization, family, job quality and employment factors thought to be influential in affecting intrinsic job preferences are related to intrinsic preferences in a way that is consistent with the theoretical arguments. Finally, the article examines how far change in such factors can account statistically for change over time in the importance attached to the intrinsic quality of work.
Data and measures
Despite the diverse scenarios about the way in which job preferences may be changing in British society, firm evidence about change over time has been missing. To address this gap, the article draws on the rich source of data on work experience provided by the British Skills Surveys, 1 a series of cross-sectional national surveys, providing a representative picture of the British workforce (for full details see Felstead et al., 2007; Gallie et al., 1998). The analyses are based on employees aged 20 to 60 in two of these surveys – carried out in 1992 and 2006 – which alone provide detailed information on job preferences. They have sample numbers of 3461 and 4049 employees respectively, 2 with overall response rates of 72 per cent in 1992 and 62 per cent in 2006. The characteristics of the samples on the key variables used in this study can be found in Table 5.
In both years, job preferences were assessed by presenting people with a list of different things people may ‘look for in a job’ and then asking them to rate the importance they themselves attached to each. As pilot work in 1992 showed that responses tended to be positively skewed, the response scale was designed to create a finer distinction among those who thought specific factors were important, giving the options of essential, very important, fairly important and not very important. The 15 items in the list (which were rotated in the interviews to avoid order effects) were: friendly people to work with; good promotion prospects; good pay; good relations with your supervisor or manager; a secure job; a job where you can use you own initiative; work you like doing; convenient hours of work; choice in your hours of work; the opportunity to use your abilities; good fringe benefits; an easy work load; good training provision; good physical working conditions; a lot of variety in the type of work.
A factor analysis, carried out on the pooled data for the two years using a Varimax rotation, revealed four factors. The first, where the highest loading items were pay, promotion, fringe benefits and security can be interpreted as an ‘extrinsic’ dimension. The second, where the main items were the use of initiative, work you like doing, the use of abilities and the variety of the work, was clearly an ‘intrinsic’ dimension. The third brought together ‘convenient hours of work’ and ‘choice in hours of work’, followed at some distance by ‘an easy work load’. The interpretation of this dimension is not unambiguous but it may reflect a preference for a job permitting work-life balance. The final factor centered on good relations with one’s supervisor and friendly people to work for, indicating an emphasis on the social climate of the job. As separate analyses for each year showed that the same items loaded most highly on each factor for 1992 and 2006, the factor scores derived from the pooled data were used. 3
The pattern of change – the decline of intrinsic job preferences?
Did the importance of intrinsic job preferences decline or increase between 1992 and 2006? Table 1 presents the separate items grouped under the dimensions to which they related most strongly. It gives two summary measures for each year: the proportion of employees who regarded this aspect of a job as ‘essential’ or ‘very important’; and a scale that captures the overall pattern by scoring the response categories from 4 for ‘essential’ to 1 for ‘not very important’.
Job preferences 1992 and 2006
Note: E+VI=Essential+Very important
There was no evidence that the importance of intrinsic preferences declined over the period. Instead the importance attached to all of the intrinsic items was higher in 2006 than in 1992. In contrast, the picture was more varied with respect to the main extrinsic items. There was an increased importance attached to pay, promotion prospects and good fringe benefits (although the last of these remained a rather low priority). However, there was little change with respect to job security and physical working conditions and there was some decline in the importance attached to training opportunities.
Employees also attached more importance in 2006 than in 1992 to having friendly colleagues and (albeit to a lesser extent) to good relations with the supervisor. The most marked changes of all were in the importance attached to choice and convenience in hours of work, although it should be noted that even in 2006 these remained relatively low in the list of priorities. 4
Taking the factor scores for different job preference dimensions, it can be seen in Table 2 that the importance people attached to the intrinsic aspects of work, to the social climate in the workplace and to the opportunities for work-family balance was greater in 2006 than in 1992. However, there was no increase in the importance they attached to the extrinsic features of work.
Job preference dimensions 1992 and 2006
Note: Sig ***=p<0.001 n.s.=not significant.
In short, it would seem that the prediction of growing instrumentalism is not supported. This had argued that the importance of extrinsic preferences would grow at the expense of concern with the intrinsic quality of jobs. It was correct that the importance attached to pay had increased (although this was not the case for the wider spectrum of extrinsic job preferences). But this did not imply that employees became less concerned about the intrinsic quality of their jobs. To the contrary, the importance of preferences relating to the intrinsic features of work was higher in 2006 than in 1992, as was concern about the quality of social relations at work and the availability of working hours that make it easier to combine work with non-work life.
Determinants of intrinsic job preferences
Given the central concern with the implications of job preferences for the quality of work agenda, the analysis is focused on the determinants of the strength of intrinsic work preferences. 5 The earlier theoretical discussion pointed to a number of factors that could be important in determining the strength of intrinsic job preferences. Broadly these can be grouped into four types of factor: early socialization in terms of education; the nature of people’s family life; the skill level and quality of their jobs; and the extent to which income and security from work provide for more basic needs. This section considers whether or not these characteristics are empirically associated with the strength of intrinsic work orientations. 6
Socialization, gender and the family
The argument from early socialization suggests that people’s basic preferences are established when they are relatively young and then come to affect their choice of jobs. Central factors for the salience to people of intrinsic preferences are likely to be the importance that parents attach to self-development and exposure over time to formal education. It is also possible that there have been important gender-specific changes in socialization, with women increasingly growing up in a cultural climate in which employment careers were regarded as the norm.
In the absence of longitudinal data, attempts to capture early circumstances depend on retrospective self-report and are necessarily rather approximate. The assumption is that parental encouragement about education provides a key tracer for early effects, since it is clearly directed to and values personal skill development. It is also likely to have been a memorable experience, increasing the reliability of recall. Both the 1992 and 2006 surveys contained a question that was designed to tap this. It asked: ‘When you were at school, how much interest would you say your parents took in how you were getting on there?’
Education in itself has been shown repeatedly to be strongly related to intrinsic job preferences and education levels have been rising sharply over the period. Education is measured by the highest qualification level that people obtained, distinguishing five levels: no qualifications, poor lower secondary, lower secondary, upper secondary and tertiary education. A summary measure can be constructed by scoring the different levels from zero (no qualifications) to 4 (tertiary education).
With respect to differential gender socialization, the expectation would be that the importance of intrinsic preferences for women would increase the later the birth cohort in which they were brought up. Women born in earlier decades would have been raised in a culture which defined their roles primarily in terms of homemaking. In contrast, women born in more recent decades would have developed their values at a time in which employment was regarded as a legitimate and desirable life objective for women. The expectation for men would be rather different. If the instrumentalism argument were correct, it would be expected that older employees, brought up in a traditional work culture, would attach more importance to the intrinsic characteristics of jobs than younger employees who grew up in an environment in which leisure and home life had become more important to male identity. To try to capture such early normative factors, respondents have been grouped from both surveys into five birth cohorts: those born respectively from 1932 to 1945, 1946 to 1960, 1961 to 1972, 1973 to 1981 and 1982 to 1986.
How did such factors relate empirically to the salience of intrinsic job preferences? Table 3 presents the results of OLS regressions first for all employees and then for male and female employees separately. A first point to note is that the evidence does support the view that early educational socialization is associated with current job preferences. Those who had parents who were interested in their progress at school were notably more likely to emphasize intrinsic job preferences even when their own educational achievement was taken into account. But separate analyses for men and women show that the effects of early parental interest in educational progress are only significant for women. In contrast, personal educational experience, as reflected in the highest qualification obtained, independently relates strongly to the importance of intrinsic preferences and the effect is significant for both sexes. Additional interaction tests (not shown) indicate that the strength of the effect of education was very similar across the different birth cohorts with one exception – it was significantly stronger (for both men and women) for the cohort born between 1961 and 1972, perhaps reflecting the distinctiveness of the educational ethos in the aftermath of the 1960s.
Early socialization and family effects on employees’ intrinsic job preferences
Note: OLS Regression. Pooled data. Sig: ***=p<0.001; **=p<0.01; *=p<0.05 n.s.=not significant.
There is less support for the other socialization hypotheses. There is no difference between men and women in intrinsic orientation, contrary to the hypothesis of sex-specific role socialization. Further there is no evidence from the different birth cohorts of particularly marked changes in patterns of early socialization for either sex, once education is taken into account.
The argument that people’s job preferences will be affected by their involvement in the nuclear family receives some limited support. But the separate analyses for men and women indicate the effect of living with a partner is sex-specific. Having a partner is associated with a greater intrinsic orientation for men (contrary to the instrumentalism hypothesis), but makes no difference for women (contrary to what might be expected from theories of gender specific socialization). In contrast, having dependent children under the age of 5 in the household is associated with the lower salience of intrinsic preferences for both men and women. This is consistent with the view that family life may constitute an alternative source of life interest, although it may also reflect the financial burden of bringing up children.
Job quality - skills levels and organizational involvement
Where work tasks are low skilled, the expectation is that people will retreat into an ‘alienated’ instrumentalism in which they lose their aspiration for types of work that foster self-development. In contrast, higher skilled jobs are likely to increase personal identification with work and increase people’s desire to make use of their initiative and skills. Similarly the extent to which jobs provide learning opportunities may affect the importance attached to self-realization. Finally, writers on high performance management systems have emphasized the value of involvement through communication and consultation for the importance people attach to their work.
A commonly used measure of the general skill level of jobs is occupational level. The first digit classes of the Standard Occupational Classification (SOC), which is explicitly intended to reflect skill, are taken. The surveys also included three more direct indicators of skill requirements. The first is the qualification level required for the job. People were asked: ‘If they were applying today, what qualifications, if any, would someone need to get the type of job you have now?’ The highest qualification cited was allocated to one of five broad job qualification categories. 7 The second measure is concerned with the length of training time the person had received, asking people ‘Since completing full-time education, have you ever had or are you currently undertaking, training for the type of work you currently do?’ Those who had received training were then asked ‘How long in total did (or will) that training last?’ The third measure addresses the on-the-job learning required when entering the job, asking ‘How long did it take for you after you first started doing this type of job to learn to do it well?’ A summary index has been constructed for each measure by scoring the responses for the levels of required qualifications and for the length of training and on-the-job learning times. 8
There is a single item measure of the learning opportunities provided by the job. People were asked how strongly they agreed or disagreed that: ‘My job requires that I keep learning new things’. With respect to involvement in decision-making, people were asked whether they made suggestions about improving the efficiency of work, whether management held meetings in which they could express their views about what happened in the organization and whether they were involved in quality circles.
The intrinsic quality of jobs may exercise an independent influence on job preferences, but it may also reflect self-selection into particular types of jobs. Those with stronger intrinsic preferences may choose jobs that offer better opportunities for skill use and self-development. In estimating the influence of work factors, the early socialization variables discussed in the previous section are then controlled for on the assumption that these will take account of at least some of the self-selection effect. Industry controls were also introduced but coefficients are not reported as they proved to be of little substantive interest.
Even when early socialization factors have been controlled for, the skill level and developmental opportunities of the job still stand out as having a strong association with intrinsic preferences (Table 4). This is true for both men and women. There is a strong gradient by occupational level: managers, professionals and associate professionals have a stronger intrinsic orientation. It is notable that even when broad skill position is taken into account through occupational level, one of the more detailed job skill indicators – the qualifications required for the job – is still highly significant. Opportunities for skill development are also associated with stronger intrinsic job preferences, as is shown by the significant coefficient for jobs requiring continuous learning.
Job quality and employment conditions effects on intrinsic job preferences
Note: OLS regressions. Pooled data. Controls include constant, variables in Table 5 and industry. Sig: ***=p<0.001; **=p<0.01; *=p<0.05 n.s.=not significant.
Organizational involvement procedures are less consistently related to a stronger intrinsic orientation. Those in jobs where there are opportunities to make suggestions about work and to participate in quality circles do show a higher intrinsic orientation. However, ‘consultative participation’, via meetings in which employees can express their views about wider developments in the organization, was not significant.
Overall job quality has an important effect over and above early socialization effects. Whereas the adjusted R2 for the socialization model was 0.12, it rose to 0.22 when job quality variables were added. A comparison of the models for the different sexes shows that the overall effect of job quality factors was strong for both men and women.
Employment conditions - pay and security
The final set of factors concerns the extent to which the conditions of employment allow employees to meet basic financial and security needs. The focus is on real income, taking log income adjusted between the years for inflation. For security, a measure is taken of how satisfied or dissatisfied people are with their job security, with a seven-point scale from completely satisfied to completely dissatisfied. The variables used to proxy early socialization characteristics are again included as a control for self-selection.
As can be seen in Table 4, the prediction that higher income and higher job security would be associated with a greater emphasis on intrinsic preferences is strongly supported by the results. In both cases the effects are highly significant for both men and women. The extent to which employment protects basic material needs does then seem to be an important factor in accounting for intrinsic job preferences, although the overall explanatory power of the ‘needs hierarchy’ model is less than that of the previous ‘job quality’ model (an adjusted R2 of .15 compared with .22).
Accounting for the increased importance of intrinsic job preferences
As can be seen in Table 5, many of the factors that were shown in the previous section to be associated with the strength of intrinsic job preferences became more prevalent over the period between 1992 and 2006. There was an increase in parental support for schooling: whereas 40 per cent of respondents in 1992 reported that their parents had taken a lot of interest in their schooling, the proportion had risen to 47 per cent in 2006. Similarly the data documents the familiar picture of a marked rise in educational levels. With respect to job quality, there was an overall increase in job skill levels. There was growth both in managerial and associate professional positions and a decline in the least skilled elementary occupations. Moreover, all of the direct skill measures indicate a rise in the skill levels of jobs from the early 1990s to the middle of the current decade. There was a concomitant increase in the extent to which jobs were conducive to self-development: the proportion agreeing strongly that it was the case that their job requires them to keep learning new things rose from 26 per cent in 1992 to 34 per cent in 2006. Employee involvement mechanisms became more prevalent over the period – particularly with respect to quality circles and the opportunities to suggest organizational improvements. Finally, in the period covered by the surveys, incomes rose and there was a marked increase in employees’ satisfaction with their job security.
Individual and work characteristics 1992 and 2006
The article now turns to examine how far these changing patterns account for the increased importance of intrinsic job preferences between 1992 and 2006. The first step consists of estimating the gross change in attachment to intrinsic preferences between the two years, by including a variable for year 2006 without any control variables. This coefficient is shown in Model 1 of Table 6. Then in Models 2 to 4 different sets of variables, representing the main theoretical arguments, are added into the regression as controls. The extent to which they reduce the initial year coefficient is taken as their effect in accounting for change over the period.
Changes in the year coefficient with different sets of controls
Note: Sample is held constant to full model (N=5729). Model 2 includes education, parental interest, sex, dependent children and cohort; Model 3=Model 2 + occupation, industry, skills and organizational involvement; Model 4=Model 2 + log pay and job security. Sig: ***=p<0.001; **=p<0.01; *=p<0.05.
The effect of the early socialization and family factors is strong. The initial coefficient is reduced by 25 per cent. The most important factors contributing to this were parents’ encouragement of education and personal educational experience. As shown in the 4th column (Model 3), the work context factors (including industrial sector) add another substantial reduction of the year coefficient and the model now accounts for 58 per cent of the initial year coefficient. It is the skill and learning qualities of the job that make the most substantial difference, on their own reducing the coefficient to 0.12. The 5th column (Model 4) shows that the employment conditions model, which adds job security and pay to the individual and family characteristics of Model 2, reduces the year coefficient even further. Looking at the separate effect of the two types of employment conditions, it is clear that it is real pay that has the strongest effect. With just pay included, the year coefficient is reduced to 0.11, while security on its own reduces it only to 0.19.
Finally, taking the full model, which brings together early socialization and family factors, the quality of the job and its employment conditions, nearly three-quarters of the initial year effect is accounted for – with the year coefficient reduced to 0.07.
Conclusions
The starting point was the importance of intrinsic job preferences for policies to improve job quality. If employees become more demanding about the intrinsic quality of job tasks, then the urgency of taking work reform seriously is stronger than if the importance attached to intrinsic factors is declining. The pattern of change in job preferences between 1992 and 2006 shows that, although the importance of pay had risen, this did not imply a decline in the importance attached to issues of intrinsic job quality. Rather, on all of the indicators, concern for the intrinsic quality of jobs had increased. This was evident even after controls had been introduced for changes in workforce structure and economic context.
A number of factors thought to affect intrinsic job preferences were examined, relating to early socialization, to the quality of the jobs and to the extent to which employment conditions protected people from economic deprivation and insecurity.
The evidence supported the claims that early socialization factors were associated with the salience of intrinsic job preferences. In particular, those with higher levels of education were markedly more intrinsically orientated than those with lesser education. For women, the importance that parents had attached to schooling was also important. In contrast, no significant effects were found for either sex or birth cohort – suggesting that, once educational expectations and experience were taken into account, there was no additional effect of changes in gender-specific socialization.
The data were also consistent with the view that the intrinsic quality of the jobs in which people work affects job preferences. Both skill content and learning opportunities were positively associated with intrinsic preferences. The evidence for employee involvement was less consistent. Consultative committees providing opportunities to discuss broader organizational issues made little difference. But more local involvement through quality circles and suggestion schemes was associated with stronger intrinsic orientations.
The evidence was consistent with the argument that the protection of more basic needs of economic sustenance and security increases concern for issues of self- realization in work. Those who felt more secure and who had higher rates of pay attached more importance to the intrinsic features of work. This undermined one of the key assumptions of the instrumentalism perspective, namely that higher pay would reduce people’s concern for the intrinsic quality of their jobs.
Finally, the analysis examined how far these different factors helped to account for the increase in the importance of intrinsic job preferences over time. Education certainly had a strong effect, but the impact of people’s work and employment conditions was even greater. In particular, the skill level of jobs and greater material well-being through higher incomes sharply diminished the initial year change effect, even after early socialization factors had been taken into account.
The comparison of job preferences between 1992 and 2006 points clearly to a rise in the importance to employees of the intrinsic quality of jobs – driven primarily by higher educational levels, higher skilled jobs and greater security in terms of pay and employment. But this occurred in a period of strong economic growth and rising prosperity. It remains to be seen how the value people attach to the quality of their work has been and will be affected by the economic crisis since 2008. The period covered by the data does not allow us to distinguish cyclical from longer term structural tendencies. Rising education levels may maintain an upward pressure on demand for good quality jobs. But other factors that have been shown to be significant for the importance attached to intrinsic preferences – such as income and job security – will be very sensitive to the economic cycle. Rising incomes and greater job security helped account for the rise in intrinsic preferences in a period of economic growth, but declining real incomes and greater job insecurity could be expected to undermine such preferences during and in the aftermath of economic crisis. The effect of the economic crisis on employers’ choices about skill development and the organization of work is difficult to predict. Only subsequent research can establish whether the employer workforce strategies that emerge from the current recession are such that they undermine or reinforce the growth of aspirations for self-realization and self-development at work.
Footnotes
Funding
We are grateful to the funders of the 1992 and 2006 surveys. The 1992 survey was funded by a consortium of employers, government departments and the Leverhulme Trust, the 2006 survey by a consortium of the ESRC and government departments. For full details of funders, see Gallie et al. (1998), 
.