Money isn’t everything: job satisfaction, nonmonetary job rewards, and sub-baccalaureate credentials

Limitations of Earnings

Research often looks at earnings outcomes, and policymakers now propose rating colleges on earnings outcomes.¹ The rationale behind this proposed policy is that students often take out student loans to attend college without knowledge about whether their post-college jobs will enable them to repay their loans. If graduates’ earnings are not better than the earnings of high school graduates, students should be warned, and colleges should be sanctioned. Earnings are important, especially with the importance of student loans in college finance, but an earnings-centered perspective can be criticized on several grounds.

First, economic theory suggests that some individuals will sacrifice pay to get better training, especially in early jobs (Doeringer and Piore, 1971; Piore, 2002; Holzer et al. 2011; Stern 1978.1982). Some qualitative research indicates that young employees are aware of this trade-off (cite withheld). Young adults’ low-paid jobs may offer better training and long-term career outcomes than higher-paid jobs.

Second, some colleges help graduates obtain career-relevant jobs, and even discourage highly-paid jobs (Rosenbaum, et al., 2006). College occupational faculty report that they advise graduates to look for career-relevant jobs that use their skills and provide job advancement. They also advise graduates to seek autonomy and responsibility that will develop their skills. Some occupational faculty discourage some highly-paid jobs, such as dangerous security jobs, strenuous warehouse jobs, and dead-end jobs. Graduates who take such jobs may see early earnings advantages disappear 10–20 years later, particularly if they are injured or if they are less willing to tolerate those job conditions as they get older.

Earnings may not correspond to whether or not a job is “good” or satisfying to the individual who holds it. This paper extends prior research to examine whether young adults’ ratings of various non-monetary job rewards are more related to job satisfaction than earnings and identify which credentials lead to better nonmonetary rewards.

Job Satisfaction and Nonmonetary Job Rewards

Job satisfaction is the usual indicator of job quality, and research has found a wide array of nonmonetary job rewards related to job satisfaction. In the earliest research, organizational psychology considers job satisfaction the usual indicator of job quality. It is related to five core job characteristics: skill variety, task identity, task significance, autonomy, and task feedback (Hackman and Oldham, 1975). Surveys of national samples indicate that job satisfaction is strongly associated with similar job conditions—autonomy and occupational self-direction, job pressures, organizational structure and position, job uncertainties, and rewards and protections (Clark, 2005; Davis et al. 2009; Miller, 1980). A recent review concludes that a good job provides relatively high earnings, potential for earnings and career growth, adequate fringe benefits, autonomy (self-direction over work), flexibility to address non-work needs, and relatively secure jobs (Kalleberg, 2011, pp. 7–9).

Using national survey data, Jencks, Perman, and Rainwater (1988) have found that adults’ job satisfaction is less related to earnings than to non-monetary job rewards, such as autonomy and career-relevance. They found that the effect of these non-monetary job conditions on job desirability is twice that of earnings, in a study of adults of all ages (Ibid.). More recent research has found comparable findings (Oreopoulos and Salvanes 2011). However, diverse-aged samples like these include individuals who started their careers in different labor market conditions that no longer exist, so they may not be informative to current students.

The literature on job quality has placed considerable confidence on subjective ratings. Despite researchers’ concerns about respondents’ subjective judgments of career relevance, objective indicators require researchers to make coding judgments that raise other doubts (Davis, Smith, and Marsden, 2009). Many job attributes that individuals value are hard to measure objectively, and individuals’ subjective ratings (including job satisfaction) provide indicators that are “closely related to other, more complicated objective measures” (Clark et al. 2008). We concur with researchers who have concluded that subjective ratings are valuable indicators about aspects of jobs that are otherwise difficult to measure, and subjective ratings are the best indication of individuals’ own reactions to their jobs.

Measures

Add Health asked participants about properties of their current job in 2008 that we have coded as 17 variables in four categories: career relevance, benefits, working conditions, and job quality. Career-relevance was coded as three binary variables: whether the job is related to a career, prepares for a career, or is part of a career. Benefits includes whether the job offers health benefits, retirement benefits, and vacation benefits. Working conditions includes whether jobs were day shifts, irregular hours, physically strenuous, desk jobs, repetitive, unstable (number of times fired), supervisory, or supervisory of managers. Job quality includes status, autonomy, variety, and job satisfaction.

Add Health asked about three fringe benefits: health, vacation, and retirement. These variables are strongly correlated, but other job dimensions are measured by a single item and aren’t highly correlated. The creation of factors for fringe benefits might make benefits better measured but would reduce comparability with the other dimensions. The three items for career relevance are strongly correlated because they came from a single item with four non-ordered categorical responses. Job rewards are all measured by individual items.

Associations between Job Satisfaction and Job Rewards

The usual analyses implicitly assume that earnings is either the only or the most important dimension of good jobs. We find that other dimensions are more or equally as important. 6 out of the 17 variables we studied were correlated with job satisfaction (r=0.10 or higher). Those 6 variables were earnings, perceived SES, autonomy, repetitiveness, job related to career goal, and job is part of a career (“career-relevant,” Table 1a). In addition, these correlations do not change very much within different levels of education credentials (Table 1a). These findings indicate that a “good job” is multidimensional and relies on several factors, aligning with previous literature.

Consistent with prior literature, we find that career-relevance and autonomy are consistently most highly correlated with job satisfaction. These correlations persist at comparable levels within education categories. The relationships between job rewards and satisfaction are not mediated by education, even though education is associated with many job rewards, as we show below.

In a regression of job rewards and background variables on job satisfaction (Table 1b), we find that, with the exception of earnings, five of the six job rewards that correlate with job satisfaction also have significant coefficients, after extensive controls.

Nonmonetary Job Reward Payoffs for each Education Credential

We now analyze whether different levels of education lead to different non-monetary rewards. We examine 5 education groups: non-graduates (students who attended college without earning a credential), certificates, associate’s degrees, bachelor’s degrees, and master’s degrees or higher. We find that all credentials lead to many non-monetary rewards, compared to high school graduates, but non-graduates have only a few nonmonetary rewards and at modest levels.

Table 2 shows how various college credentials are related to each job reward, after many controls (listed in footnote to table). For the binary (first 13) job rewards, we used regression and we report IRR in rows for each education credential. The final 4 used linear regression and report regression coefficients.

Non-graduates Payoffs: Some College but No Credential

Compared to high school graduates, non-graduates report a limited number of rewards. In terms of career-related variables, they are slightly more likely to report that their job offers career preparation, but say their job is less related to their career. Both relationships are weak and confidence intervals of coefficients mostly don’t overlap (preparation slightly overlaps) between non-graduates and certificates. Non-graduates were more likely than high school graduates to report receiving health, retirement, and vacation benefits and some better working conditions (desk jobs, perceived status, supervising others, less likely to work hard physically). Non-graduates did not differ in autonomy or repetitive work from high school graduates.

Non-graduates had some nonmonetary job rewards, but they had fewer rewards and generally at more modest levels than individuals with any college degree. Moreover, non-graduates were less likely to report having career-related jobs, and did not differ in autonomy and career relevance, which are the items most highly correlated with job satisfaction. Attending college without earning a credential yields few benefits.

Certificates Payoffs

Compared to high school graduates, individuals whose highest credential is a certificate have significantly higher levels of career-related variables, autonomy, and benefits, but do not differ from high school graduates for some working conditions.

Certificate holders are more likely to have jobs that prepare them for a career, career-relevant jobs, and jobs related to careers. Certificates also predict greater job autonomy, variety, and job status.

Certificate holders are more likely to be offered benefits than high school graduates. Certificate holders are also more likely than high school graduates to have desk jobs, less physical labor, and less job instability (number times fired), but their jobs are no better on day shift and irregular hours.

Overall, certificate holders are more likely to have some job rewards than high school graduates – particularly in autonomy, variety and career-related variables – which are job rewards that non-graduates do not get. However, certificate holders lack the across-the-board reward increases that come with associate’s, bachelor’s, and graduate degrees, as shown below.

AA Payoffs

AA graduates report more of almost all job rewards than high school graduates, with increases in career-related, benefits, working conditions, and quality. Compared with high school graduates, AA graduates are more likely to have jobs offering career preparation and much more likely to have career-relevant jobs, but are not less likely to work in jobs unrelated to their career. They are also more likely to receive vacation, health, and retirement benefits. In addition, they show increases in working conditions relative to high school graduates (more likely to have day shifts and desk jobs and less likely to have strenuous work). They also get more job autonomy, variety (less repetitive), job satisfaction, and job status.

In sum, AA graduates have significantly higher payoffs than high school graduates on most job rewards. They also enjoy a wider variety of job rewards than certificate holders, and are generally on par with bachelor’s degree holders, as described below.

BA Payoffs

Like associate’s degrees, bachelor’s degree graduates report more of almost all non-monetary job rewards than high school graduates, across all categories of relation to career, benefits, working conditions, and quality. Compared with high school graduates, BA graduates are more likely to report that their job is preparation for a career, that their job is part of a career, and that their job is related to their career. They are also more likely to be offered each of the benefits. In addition, they have increased likelihood of nearly all desirable working conditions (but not supervising others or supervising managers, which no credential predicts). Finally, they also report higher job quality. Bachelor’s degree holders report more job autonomy, variety, job satisfaction, and status.

Overall, bachelor’s degree graduates report increased likelihood of almost all rewards compared to high school graduates. The rewards they get are more wide-ranging than “some college” and certificates. However, associate’s degree graduates have nearly all the same job rewards as BAs, sometimes at about the same magnitude.

MA and Above Payoffs

Like associate’s and bachelor’s degree holders, graduate degree holders are better off in almost all respects than high school graduates. In addition, most coefficients are as large or larger than those for BA’s. In terms of relation to career, graduate degree holders are more likely to report that their job is part of their career (compared to high school graduates) and that their job is related to their career. Curiously, however, they do not differ from high school graduates in whether their job prepares them for their career (perhaps because their job is already part of their career).

Graduate degree holders also get all the working conditions payoffs that bachelor’s degrees get. Compared with high school graduates, people with graduate degrees are more likely to be currently employed, day shifts, and desk jobs; they are less likely to have strenuous work. The one exception: graduate degrees don’t get significantly lower irregular hours than high school graduates (perhaps because of health field demands). Graduate degree holders also report more job autonomy, variety, job satisfaction, and status.

Overall, graduate degree holders mostly have the same rewards that associate’s and bachelor’s degree graduates get. However, they often have larger coefficients, indicating higher levels or more widely received rewards.

This research uses data from Add Health, a program project directed by Kathleen Mullan Harris and designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris at the University of North Carolina at Chapel Hill, and funded by grant P01-HD31921 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, with cooperative funding from 23 other federal agencies and foundations. Special acknowledgment is due Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design. Information on how to obtain the Add Health data files is available on the Add Health website (http://www.cpc.unc.edu/addhealth). No direct support was received from grant P01-HD31921 for this analysis.