The Role of Financial Burden in Nonprofit Sector Commitment

Evidence about millennial work motivations and the increasing importance of compensation questions the durability of the donative labor hypothesis in explaining nonprofit sector commitment. Nonprofit graduate education offers an employment pipeline into the sector, but what if the importance of compensation is partly driven by the financial burden accrued from education? Could it be that financial burden contributes to choices about work and commitment to the nonprofit sector? Using longitudinal data of nonprofit education alumni, we inquire about their sector commitment in light of the financial burden from their degree. Findings of this exploratory study offer a starting point for future research into how nonprofit education alumni view career opportunities in the nonprofit sector.

The primary assumption of the labor donation hypothesis is that nonprofit employees are willing to accept lower wages than their for-profit counterparts and, as such, donate a portion of their labor to the nonprofit organizations for which they work (Hansmann, 1980). This assumption may no longer hold true. Recent studies have found that the nonprofit sector attracts mission-driven employees while easily losing them because of the inability to pay competitively (Kim & Charbonneu, 2020;Ng & Johnson, 2019). More specifically among the millennial generation-the main generation now entering and increasingly dominating the workforce-findings show that compensation influences younger workers' career decisions (Abouassi et al., 2019;Johnson & Ng, 2016;Walk et al., 2019).
Whereas the evidence supporting the relationship between compensation and commitment to the nonprofit sector is mounting, not much is known about other factors-aside from membership in the millennial generation-that could explain why compensation appears to be an emerging consideration. One potential explanation is the role of graduate education and the accompanying financial burdens in light of high education costs. The growth of nonprofit graduate education degree offerings and enrollments signal that more students seem to opt into the sector by means of their education choice (Kuenzi et al., 2020;Mirabella et al., 2019). Yet, these diplomas come with an associated expense, and nonprofit professionals with graduate degrees accrue on average of $50,000 in debt (Berkshire, 2012). When nonprofit alumni leave with their diploma in hand, they may be heavy laden with debt, and gainful employment, regardless of sector, may be the most pressing priority. Thus, given these potentially heavy debt loads, we are interested in further understanding the impact financial burden has on an individual's commitment to the nonprofit sector.
This descriptive study examines tangible financial burden in the form of student debt, but also considers the potential impact of perceived financial burden and the worthiness of the monetary investment the degree necessitated, on nonprofit sector commitment, controlling for education, a factor associated with sector commitment (Johnson & Ng, 2016). We use longitudinal survey data from 71 nonprofit education alumni of three nonprofit graduate degree programs (e.g., nonprofit management and philanthropic studies) in the Southeastern and Midwestern United States. Our findings and suggestions for future research further the understanding of career commitment in the professionalizing nonprofit sector that values specialized, albeit costly, graduate training. These findings also have important implications for nonprofit organizations as they consider how to attract and retain top talent.

Nonprofit Sector Commitment
A sizeable number of studies explore individuals' motivations for choosing work in one sector over another; however, these studies often capture motivation at one point in time with less focus on the variables that influence an individual's commitment to that sector long-term. However, sector commitment, the dedication of an individual in continuing a career within a specific sector of employment, has received increased attention among public and nonprofit researchers as a means of understanding and engendering retention (see Walk et al. 2019 for a summary of this research). Sector commitment is an extension of Weisbrod's (1988) theory of managerial sorting, which posits individuals will opt for opportunities within the sector whose incentives match their preferences, and conceives that individuals will stay in a sector that aligns with these preferences. Research on sector commitment has examined the role of intrinsic and extrinsic rewards, and findings imply that rewards and commitment have a complex relationship with recruitment and retention of employees (e.g., Chetkovich, 2003;Johnson & Ng, 2016;Lee & Wilkins, 2011;LeRoux & Feeney, 2013;Ng & Johnson, 2019).
Prior research describes nonprofit sector commitment as driven by a complex set of motivations, including a combination of both intrinsic and extrinsic rewards (LeRoux & Feeney, 2013;Ng & Johnson, 2019;Walk et al., 2019). Evidence suggests that the nonprofit sector offers more to individuals who are motivated by intrinsic factors such as a public service orientation, and altruism (LeRoux & Feeney, 2013;Mirvis, 1992). These motivations may help explain why nonprofit employees seek opportunities and remain in a sector that typically offers lower wages than in government or for-profit organizations (Cohen, 2010). Yet research increasingly documents extrinsic factors as drivers of sector commitment, pointing out how wages might outweigh the motivational impacts of altruism and public service orientation. Particularly, Johnson and Ng (2016) show that millennials (referring to people born between 1981-1996) who have a college education and are in managerial positions are especially likely to report low or no nonprofit sector commitment. Further, while some past studies assume sector commitment is stable over time, there is evidence that it diminishes with time spent in the sector (Berkshire, 2012;Walk et al., 2013). This is especially true for younger employees when considering starting a family in light of low compensation (Walk et al., 2013). Therefore, while mission and intrinsic rewards may influence an individual to work in the sector in the first place, other factors such as extrinsic rewards may mitigate their commitment to remaining.

Student Debt and Financial Burden
A professional's "personal financial considerations" may weigh heavily as they seek to maintain their overall financial health and meet financial obligations as they pursue professional opportunities (Ng & Johnson, 2019, p. 2). These financial considerations include such factors as their job's compensation as well as personal obligations, such as home mortgages or rent, child-rearing expenses, healthcare expenses, and education-related debt. Financial industry research, however, has documented that student debt weighs heaviest in terms of personal repayment obligations (Lewin, 2011). Further, Lange and Byrd (1998) documented the psychological burden student debt holds for graduates.
This burden may be manifested in the debt-holders' employment decisions. Accruing evidence suggests the significant role education debt has as a driver of job choice and even sector choice. Hausdorf's (2007) research among university graduates reveals financial need rather than career interest and educational investment as the driver of employment decisions. Similarly, studies found that individuals with higher debt burden took positions with higher paying base salaries post-graduation (Minicozzi, 2005;Phillips et al., 2014;Rothstein & Rouse, 2011;Velez et al., 2019). Higher levels of student debt increase the likelihood of individuals working outside their field (Huelsman, 2015;Minicozzi, 2005) and is also related to career choice: Graduates with debt are less likely to take public interest positions or public service positions (Field, 2009;Minicozzi, 2005;Rothstein & Rouse, 2011).
Most of the evidence on debt and its impact on career choices and sector commitment is found outside the nonprofit sector. For instance, lawyers with higher levels of debt were more likely to work in private firms (Sieg & Wang, 2017;Wright & Christensen, 2010). In a study of social work students (at both the baccalaureate and master's level), researchers indicated that the majority of students (76%) had accumulated student debt, and that debt levels among this population were higher than when measured a decade earlier (Unrau et al., 2020). Further, the economic hardships faced by these students had impacts in their ability to stay in school and eventually engage in social work careers, potentially compromising career stability in the social work field. Chetkovich (2003) remarked on the relationship between debt and sector choice, "Among these policy students, those who enter the private sector hope to gain skills, credibility, and experience; to make enough money to pay off debts and live comfortably; and to enjoy a resource-rich and fast-paced environment" (p. 670). Related research on undergraduate students found, educational debt has a marginal impact on initial job choice and that rising educational debt may discourage students from choosing public and nonprofit sector jobs, despite high levels of public service motivation (Ng & Johnson, 2019). These findings indicate that students' financial needs outweigh more personal and intrinsic motivations such as desire to serve the public, work in a cause career, or do work associated with their chosen degree.
The nonprofit practitioner audience, particularly Young Nonprofit Professionals Network (YNPN), along with The Chronicle of Philanthropy and the Nonprofit Leadership Alliance, have also studied the role of compensation among the nonprofit sector workforce (Berkshire, 2012;Rendon, 2019;Solomon & Sandahl, 2007). Despite limited generalizability since the samples were derived from YNPN membership, findings show that among nonprofit professionals who had been in the sector five years or less, 65% reported college loan debt, and 30% responded that their debt burden was $50,000 or higher (Berkshire, 2012). Of those with debt, the majority (80%) noted that their debt burden was an influential factor in shaping their career decisions, noting a trade-off between work that is meaningful and earning enough to pay off student loan debt. Nonprofit professionals were more likely to hold a graduate degree and highly likely to report "they had to stretch to make ends meet" (Rendon, 2019, p. 4). As Paul Schmitz, founder of Public Allies, explains: "The combination of low pay and student debt could cause many young people to flee nonprofits" (quoted in Berkshire, 2012, para. 29).

Study Context and Methods
This descriptive study explores the role of financial burden from nonprofit graduate degrees and nonprofit sector commitment. The sample was compiled in two phases. In phase 1 (summer 2017), we recruited recent alumni (past 5 years, n=700) of three nonprofit graduate degree programs to participate in an online survey. The programs are within public universities in the southeastern and midwestern United States with enrollment ranges less than 25 students, 50-75 students, and 75-100 students. One program is accredited by the Network of Schools of Public Policy, Affairs, and Administration (NASPAA), another is accredited by NASPAA and is a member of the Nonprofit Academics Center Council (NACC), and the third is a member of NACC only. The first phase secured 184 responses (26.3% response rate), of which 153 were usable given missing data (>30%). In Phase 2 (summer 2019), a survey was distributed to those who indicated their willingness to receive follow up surveys (n=155). Of those, 75 responded for a 48.4% response rate. Four respondents were dropped due to missing data (>75%) resulting in a final matched sample of 71. Both surveys were delivered using an initial invitation and two or three reminder emails. A raffle for gift cards was offered during both phases to increase the response rates.
A longitudinal design allowed us to evaluate changes in sector commitment over time, as well as to identify changes in individual circumstances such as age, marital status, and employment. In sum, the surveys captured aspects related to nonprofit sector commitment, financial burden of the graduate degrees, alongside demographic information. How each concept was operationalized is described below.
Nonprofit sector commitment was assessed in wave 1 and 2 using a question developed by Johnson and Ng (2016). Participants were asked, "How do you currently feel about building a career in the nonprofit sector?" and were provided with four options: (1) "I am 100% committed to building a career in the nonprofit sector", (2) "I will only leave the nonprofit sector for the right professional opportunity", (3) "I will be looking for the best job regardless of sector", and (4) "I do not plan to build a career in the sector". Given that there were limited numbers of responses to categories 3 and 4, we collapsed these into one to capture those who were agnostic towards the sector. Category 1 captures those fully committed to a career in the nonprofit sector, and category 2 captures those conditionally committed to a career in the nonprofit sector.
Graduate education can entail financial burden that goes beyond student debt, especially for those who did not rely on loans but rather on work to fund their graduate education. Further, financial burden can be assessed in more objective ways such as loan amounts as well as more subjective ways with regards to alumni perceptions. We therefore devised measures capturing financial burden in multiple ways. First, we assessed perceived financial burden by asking: "What was the financial burden you experienced from earning your graduate degree from the [University Name]?" Respondents were prompted to indicate their responses using a slider with a scale of 0 (no financial burden) to 100 (significant financial burden). Second, we further inquired about the kinds of financial assistance that alumni sought out and/or received to fund school-related expenses. Respondents were presented with 5 options (financial assistance from their university, financial assistance from other source, student loans, work study, financial support from family) with response categories of "1=yes, I sought or applied for this type of assistance, 2=yes, I received this type of assistance, 3=no, I did not seek or apply for this type of assistance. We combined categories 1 and 2 to reflect an overall need for financial assistance regardless of actual receipt of said assistance. Further, those who received student loans were also asked about the amount of student loans for the graduate degree. We recoded the initial 9 options into: 1=loans<$19,999; 2=$20,000-$49,999; and 3=loans $50,000 and greater. Finally, we inquired about the perceived worthiness of the degree with regards to the costs to capture the respondents' evaluation of their monetary investment into their education. On a 5-point scale ranging from 1=strongly disagree to 5=strongly agree, respondents were asked to respond to the question: "This degree was worth the money I invested in it." As few respondents indicated disagreement/neutrality with this statement we collapsed categories 1-3 into one for analysis.
Employment-related information was captured by three questions. First, we asked respondents about their current employment status. Response categories were: 1=full-time (35 hours or more per week); 2=part-time (less than 35 hours/week); 3=unemployed or looking for work; 4=in school full-time; 5=caring for family full-time; 6=retired; 7=other (please describe). After recoding the few responses in the 'other' category (e.g., "self-employed, full-time" was recoded into "full-time employment"), we recoded all responses into: 1=fulltime employment; 2=part-time employment; and 3=not employed. Second, we inquired about the respondents' work sector to reflect sector choice (the initial 3 options were recoded into 1=employment in the nonprofit sector, and 0=employment in all other sectors), and current salary (10 initial options recoded into: 1=<$49,999; 2=$50,000-$79,999; and 3=$80,000 or greater).
Demographic information collected at time 1 was age (in years), race (recoded initial 7 options into: 1=white, and 0=other), gender (recoded three options into: 1=female, and 0=male). The time 2 survey asked about marital status (recoded initial 5 options into: 1=single, and 0=other) as well as children in the household (total number recoded to: 1=children, and 0=no children). Note: Rounded to two decimals. Significant differences in italics. Table 1 presents summary statistics of variables that were collected at time 2 only or that were collected at time 1 and remained stable over time. The sample is predominantly female and white with an average age of 41 years. Of those who were employed full-or part-time (92.7%, see Table 2), 66.7% worked in the nonprofit sector, whereas 33.3% worked in other sectors. A sizeable number of nonprofit alumni earned more than $80,000 (37.5%). Perceived financial burden ranged from 0 to 100 with a mean of 49, a median of 51, and a mode of 80 (8 respondents). Alumni sought out and received financial assistance from various sources; coming from the university (63%), student loans (53%), or other financial sources (36%) were among the most often mentioned. Overall, nonprofit alumni perceived their degree to be worth the money (2.22). Table 2 presents differences between variables that were collected over time. Since time 1 data collection (2 years prior), fewer nonprofit alumni are single and, on average, more nonprofit alumni report to have children. Most notably, though, is the change in self-reported sector career commitment: Fewer nonprofit alumni report to be sector agnostic (a decrease from 16% to 6%; t(67)=2.77, p=0.007), while more nonprofit alumni indicate being fully committed to the nonprofit sector (an increase from 46% to 65%; t(67)=-2.85, p=0.006).

Findings
Before exploring the relationship of financial burden indicators (perceived financial burden, financial assistance, worthiness of degree) and nonprofit sector commitment, we investigated the relationship between sector commitment and sector choice. As indicated earlier, not all respondents work in the nonprofit sector. Indeed, drawing on time 2 data, those being fully committed to a career in the nonprofit sector reported to work in the nonprofit sector to a larger degree ( In the next set of analyses, we explored the relationship among financial burden indicators with results found in Figure 1. Perceived financial burden is negatively related to perceived worthiness of the degree with regards to costs. Particularly, those who did not perceive the Evidence of the frequency of each type of financial assistance is found in Figure 2. We further explored if specific types of financial assistance are related to perceived financial burden. Indeed, two of the five types of financial assistance are significantly related to perceived financial burden (see Table 3). First, we found when nonprofit alumni relied on financial support from their family they displayed lower levels of financial burden (M=34.45) as compared to those who did not (M=53.31; F(59)=2.40, p=0.0095). Second, alumni who sought out/received student loans (M=69) showed statistically significantly higher levels of perceived financial burden compared to those not having sought out student loans (M=30; F(60)=2.98, p=0.0018).
Among those who received student loans (N=21), we find that those who received more student loans, on average, reported higher financial burden. Particularly, those reporting loan amounts below $20,000 reported lower scores of financial burden (M=56.43, SD=19.73); those reporting up to $50,000 reported higher scores (M= 66.89, SD=12.32); and those having received amounts $50,000 and greater reported the highest levels of perceived burden (M= 81.80, SD=11.78; F(20)=4.12, p=0.03).
In the final set of analyses, we explored the relationship between financial burden indicators on nonprofit sector commitment (comparing those fully committed to those conditionally committed/agnostic) and sector choice. .31) and receipt of student loans (χ 2 =0.29, p=0.59) were unrelated to nonprofit sector commitment. However, those who perceived their degree worthy of the monetary investment were more likely to be fully committed to the nonprofit sector (χ 2 =6.85, p=0.03).

Discussion and Directions for Future Research
This descriptive study explored nonprofit sector commitment of nonprofit education alumni in light of their personal financial considerations. In what follows we discuss our findings and develop recommendation for future research.
We note that a majority of nonprofit alumni are currently working in the nonprofit sector. Since a nonprofit graduate degree is positioned as professional training for prospective nonprofit professionals, this is both confirming and encouraging. However, about one-third of alumni ended up working in other sectors post-degree. This indicates that sector commitment and sector choice are related, but not the same. Previous research mostly focused on sector choice over sector commitment (see, Johnson & Ng, 2016, for an exception). We recommend to study those alongside each other aiming to further disentangle the causal relationships between them.
The responses between the first and second wave of the survey reveal that nonprofit sector commitment may not be stable over time. Prior research has inquired about initial career choice (Ng & Johnson, 2019) or sector commitment at only a single point-in-time (e.g., Johnson & Ng, 2016;Tschirhart et al., 2008;Walk et al., 2013  to nonprofit sector commitment, while worthiness of the degree was. Whereas research from within and outside the nonprofit sector has depicted a direct linkage between education debt and sector commitment (e.g., Berkshire, 2012;Huelsman, 2015;Write & Christensen, 2010), we find a more nuanced picture in the case of nonprofit education alumni. One potential explanation and avenue for future research could be to test worthiness of the degree as a mediator of the financial burden-nonprofit sector commitment relationship.
We also saw that perceived financial burden can be significant when graduating with a nonprofit diploma, as respondents signaled a mid-range financial burden, on average, and a mode of 80. Student loans were the most influential factor related to perceived financial burden. Future research, however, should further investigate the role of other forms of financial support or assistance students can seek to alleviate financial burden such as part-or even full-time employment while pursuing the degree. In light of the finding that family support reduces financial burdens, future research should also elucidate the role of partner or family support in seeking a nonprofit graduate degree and commitment to the nonprofit sector.
Taken together our findings show that student loans seem to drive perceived financial burden, that financial burden is related to worthiness of the monetary investment, that worthiness is related to nonprofit sector commitment and, finally, that nonprofit sector commitment is related to sector choice. Especially in light of the increasing number of graduates with nonprofit education degrees (Mirabella et al., 2019), further longitudinal analysis deploying a larger sample size is needed to test this proposed relationship. We also recommend qualitative research to better understand the mechanism between sector commitment and sector choice in light of degree-related financial considerations.
Despite evidence from research on other graduate program alumni (Chetkovich, 2003;Wright & Christensen, 2010), we were unable to find a direct relationship between compensation and financial burden. Since nonprofit employees differ in significant ways from employees in other sectors (Lee & Wilkins, 2011;Mirvis, 1992), we may not find the same or highly similar relationships among nonprofit education alumni. One explanation could be that compensation levels are more salient for those with managerial roles, especially when compensation is not commensurate with work responsibilities (Johnson & Ng, 2016). As such, employment level is an important factor to consider in future studies. It also could be that compensation levels may indirectly rather than directly impact the financial burden-sector commitment relationship. We pointedly call for sector commitment research that considers employment and compensation levels to add nuance to our understanding of millennial work motivations in light of prior research that questioned the durability of the donative labor hypothesis and since millennials now constitute the majority of the workforce.
As an exploratory study, this research is not without limitations. First, the sample size limits the analysis and its inference. We recognize that between its small size and because it is a convenience sample, sampling bias may be present. Therefore, we caution against generalizing, although note that our sample did not vary significantly across programs. Next, our survey included a direct question about the amount of debt alumni held related to their degree. We further recognize this measure was problematic due to missing data and because most respondents chose to select a debt range rather than provide a precise amount. Therefore, future research could seek to employ a different measure as well as inquire to more recent alumni who may be able to recall actual amounts instead of ranges or perception measures. Related, salary does not capture other possible motivators such as fringe benefits or flexible schedules, so future studies should aim to include these in order to further isolate the impact of financial burden. Yet, given the limited evidence of previous studies that financial burden diminishes sector commitment, our study has merit in further disentangling the factors that influence commitment, identifying if and how sector commitment changes over time, and understanding the variables that impact the perceived value of a nonprofit graduate education.
While recognizing these limitations, the findings prompt practical considerations for nonprofit management graduate programs, nonprofits themselves, and for individuals that are considering a nonprofit management graduate degree. For example, given the relationship between financial burden and perceived worthiness of a student's degree found here, individuals may want to weigh their commitment to the sector against the comparative cost of different degree programs. Programs with significant nonprofit placement post-graduation and low costs could also utilize this information in marketing to students. Finally, enhancing financial assistance or reducing programmatic costs to offset burden may be necessary in order to continue to maintain nonprofit graduate education as a viable, worthwhile avenue into the nonprofit sector for all, regardless of ability to pay. This is especially important given equity and inclusivity concerns if these programs are an effective means of providing needed skills and competencies to individuals who aim to work in the sector.

Conclusions
Prior research has depicted education-related debt as having a bearing on employment decisions, but the findings of this exploratory study imply a more nuanced relationship for alumni of nonprofit-related graduate programs. In a sector known for its poor compensation prospects, and in a society increasingly holding educational expectations of paid professionals, how we understand the sector commitment of nonprofit professionals in light of their financial considerations is important. With the rise of a millennial workforce, it is a timely endeavor to understand if conventions, such as the donative labor hypothesis, adhered to by prior generations still hold under this generation. Nonprofit alumni who have recently invested in their graduate education are an appropriate means of inquiry and offer some initial evidence that the risk calculation of a nonprofit-related degree pays off in the form of professional opportunities that endear sector commitment. Future research should build upon these preliminary findings to update how we think of sector commitment, compensation, and educational outcomes in the nonprofit sector.

Notes
1. Given the low sample size, we collapsed those conditionally committed and agnostic into one group for analysis. Analysis on the three groups using Fisher's exact test yielded similar results.