Personnel Flexibility and Red Tape in Public and Nonprofit Organizations: Distinctions due to Institutional and Political Accountability

Academics and journalists have depicted government bureaucracies as particularly subject to administrative constraints, including the infamous red tape and personnel rules that sharply constrain pay, promotion and dismissal and weaken their relations to performance. Research on these topics has often focused on public organizations alone or on comparisons of public and private organizations. The analysis reported here extends this research to include nonprofit organizations. Certain theoretical perspectives would predict sharp differences between public and nonprofit organizations, while others would predict no differences. Using survey data from managerial-level respondents in state government and nonprofit organizations in Georgia and Illinois, this analysis compares perceptions of red tape and personnel rule constraints in public and nonprofit organizations. We investigate whether or not public and nonprofit respondents differ in their perceptions about levels of organizational red tape, and about whether formal rules enable or constrain managers in promoting and rewarding good employees and removing poor performers. The results indicate sharp public and nonprofit differences, with public managers reporting higher perceived organizational red tape and lower levels of personnel flexibility. In addition to public and nonprofit comparisons, the analysis takes into account other factors that might influence public and nonprofit managers’ perceptions of red tape and personnel flexibility in their organizations, including individual motivations to choose the job, the respondent’s state (Georgia or Illinois), and others.


Introduction
Historically, academics, journalists, and satirists have depicted government bureaucracies as particularly subject to administrative constraints, including "red tape" or burdensome rules and procedures that hinder performance. General red tape reduction initiatives in government have pursued this widely accepted view (e.g., Gore 1993). Major reforms have sought to eliminate red tape due to personnel rules because reformers contend that such rules constrain managers' ability to reward or discipline employees. Convinced that rigid pay systems fail to reward good performance, reformers in governments around the world have sought to implement performance-based pay systems. After decades of such efforts, however, experts still emphasize the need for reform (Ballenstedt 2007), suggesting that previous reforms have had little effect in loosening the constraints (Ingraham 1993;Kellough and Lu 1993). In addition, as described below, survey research across decades has indicated that government managers are much more likely than managers in business firms to say that personnel rules make it hard to reward a good performer with higher pay and fire a poor performer.
This apparent persistence of red tape and constraining personnel rules across time and in spite of reform initiatives raises questions about why the constraints endure. These questions have implications for theories of public and nonprofit organizations. In this analysis we pursue these implications by adding to the evidence not another comparison of government and business managers, but a comparison of public and nonprofit managers' perceptions of red tape and personnel constraints. One possible theoretical perspective would posit that government oversight imposes red tape and constraining personnel rules. Business firms have more independence from government oversight in the United States and many other nations, because firms are purportedly governed by market forces such as competition and reliance on market sales for resources. Most government agencies do not sell outputs in competitive economic markets, so governmental authorities such as legislative bodies impose controls upon them, often in the form of constraining rules. This perspective predicts that public and nonprofit managers will differ in response to questions about red tape and personnel rule constraints. Nonprofit managers will be more like business managers, in that they operate with more autonomy from government oversight.
An alternative perspective predicts no differences between public and nonprofit managers, since they both operate in not-for-profit, public-serving contexts. The absence of profit as an incentive and source of information reduces decision-makers' incentive to maximize the efficiency of organizational structures and procedures through such steps as eliminating red tape and personnel rules. 2 In addition, nonprofit status impedes the use of such motivating procedures as profit sharing. Both government and nonprofit organizations have value-laden missions that make performance goals harder to specify for use in such procedures as performance-based pay systems. Thus, most government and nonprofit organizations should be very similar because they usually share not-for-profit status.
The results reported below show strong support for the first perspective -public managers are much more likely to perceive higher levels of red tape, and lower levels of personnel flexibility (more difficulty rewarding a good performer with higher pay, more 2 Economic property rights theorists contend that the absence of transferable ownership rights creates a fundamental distinction between business firms and government organizations (e.g. Alchian and Demsetz 1972;De Alessi 1980;see Bozeman 1987, 52 for further discussion). These theorists contend that absence of ownership and property rights for government organizations dilutes incentives for efficient organizational decisions, such as decisions to reduce red tape and enhance pay and disciplinary procedures. One can extend these theoretical postulates to nonprofit organizations, since no individual has a legal claim to personal control of the proceeds or property of a nonprofit organization. Hence incentives related to red tape and personnel rules should not differ between public and nonprofit organizations. difficulty in firing a weak performer, and a weaker linkage between rewards and performance). This finding comes from results of the third National Administrative Studies Project (NASP) survey (see Appendix I) to investigate three primary research questions: (1) Do managers and others in managerial-level positions in the public and nonprofit sectors differ in their perceptions of the levels of red tape in their organizations? (2) How do perceptions of personnel flexibility (a measure of perceived constraints under personnel rules) vary between managers in the public and nonprofit sectors? (3) How convincing is this evidence? We take into account other responses on the survey that could influence such perceptions, such as individual motivations to choose a job, work history, demographic characteristics, organizational function/mission, and other variables in analyzing public and nonprofit managers' perceptions of red tape and personnel flexibility in their respective organizations.

Literature and Hypotheses
The question of whether public, nonprofit, and private organizations differ significantly in their characteristics has remained controversial and inconclusively resolved for many decades. The more specific question of whether the three types of organizations differ in levels of red tape and of personnel system constraints has followed a similar pattern, with practitioners and theorists adopting opposing stances.
Major developments in practice and public policy reflect the strong belief that government has too much red tape and too many personnel constraints, and more of these maladies than nongovernmental organizations. The Civil Service Reform Act of 1978, for example, sought to streamline procedures for dismissal. In various states in the U.S., reform legislation has loosened rules about firing, discipline, and pay and decentralized such decisions from central personnel agencies to the operating government agencies (Walters 2002). These reforms are driven in part by decades of sharp complaints from government managers and employees (Ballendstedt 2007;Macy 1971;Thompson and Rainey 2003). Often, the calls for reform include the assertion that government needs to adopt personnel rules and procedures similar to those in business. However, in striking contrast, government personnel administrators and experts have argued for decades that the claim that one cannot fire a bad government employee is a myth. They contend that an administrator can fire a bad government employee simply by following proper procedures (e.g., U.S. Office of Personnel Management 1999).
In the past, most disputes about whether government and business differ in levels of red tape and in personnel system constraints have often failed to consider the nonprofit sector. This omission requires attention, because government increasingly relies on the nonprofit sector administering government contracts and programs (Salamon 1998;Smith & Lipsky 1993), including consistent increases in the number of nonprofits providing government services such as programs for children, the elderly, and the homeless, and public health and mental health programs (Feiock and Jang 2007). As the nonprofit sector expands to administer government goods and services and draws upon a similar, public-service minded workforce, it is increasingly important to understand distinctions between the public sector and the nonprofit sector, including differences in red tape and personnel flexibilities.
As for theoretical implications, they, too, involve diametrically opposing views. As pointed out earlier, one can make a theoretical case that government organizations differ from nonprofits due to government controls. One can also theorize that governments and nonprofits show no differences because they share nonprofit status. A generic tradition among organization theorists would also contend, through a different theoretical rationale, that a public versus nonprofit distinction predicts no significant differences among organizations. This perspective condemns distinctions such as public versus private and profit versus nonprofit as crude stereotypes. Instead, the generic commonalities among organizations warrant more attention than such crude distinctions (e.g., Thompson 1962). Where there are differences among organizations, they are better explained by differences in such factors as technology, task, and size, rather than by public, nonprofit, or private status. One can cite no stronger expression of this perspective than an observation from Nobel Laureate Herbert Simon. Discussing economic rewards in organizations, Simon stated bluntly that, "…Everything said here about economic rewards applies equally to privately owned, nonprofit, and government-owned organizations. The opportunity for, and limits on, the use of rewards to motivate activities towards organizational goals are precisely the same in all three kinds of organizations…" (1995, 283, footnote 3).
Simon, the preeminent behavioral scientist ever to study organizations, thus forcefully asserts that reward systems (which include the personnel rules governing them) are "precisely the same" in public, nonprofit, and private organizations.
Simon's observation is obviously quite at odds with the views of reformers and experts described earlier, and its theoretical implications are enormous. Should a theory of public and nonprofit organizations posit that civil service systems and "system-wide rules" (Warwick 1975, 162) produce no difference in personnel rules and red tape among government, nonprofit, and business organizations? For many years administrative theorists in political science and economics have contended that government organizations tend toward high levels of rules, red tape and personnel constraints than business firms (e.g., Dahl and Lindblom 1953, 458;Downs 1967, 59;Warwick 1975, 162;Wilson 1989, 120-121). Theoretical explanations for the higher incidence of red tape often assert that political authorities exercise much more oversight over government agencies than over nongovernmental organizations. In the absence of market-provided information such as sales and profits government authorities emphasize compliance with rules and procedures to assure accountability and performance. Others emphasize accountability pressures in government that reduce incentives and opportunities to streamline administrative procedures and to reduce red tape. This theoretical orientation depicts government auspices as a distinctive context for organizations.
These ongoing disagreements show the need for more evidence. Boyne (2002), in reviewing studies comparing survey responses of public and private respondents, points out that many of the published studies analyze results of the same surveys. There are even fewer direct comparisons of public and nonprofit organizations. The situation points to the need for more studies with different samples, including more large sample empirical research comparing "sectors" on personnel dimensions and expanding to include nonprofit organizations. More comparisons between the public and nonprofit sectors can add evidence about institutional distinctions such as the presence or absence of profit indicators, motivations, and incentives. If we find differences between government and private nonprofit organizations that resemble the differences researchers have found between government and private for-profit organizations, this provides evidence that the differences may not derive from the presence or absence of a profit indicator and incentive. Rather, this would provide evidence that public versus nonprofit differences result from other institutional distinctions, such as the presence or absence of governmental ownership, oversight and personnel structures.

Research on Red Tape and Personnel Rule Constraints
In recent years, serious analysis of red tape (Bozeman 2000) has led to empirical examination and improved conceptualization. For example, Pandey and Scott (2002) have shown evidence that managers' responses about red tape differ from their responses on formalization measures, which organizational researchers use to measure the general intensity of formal rules and procedures, as opposed to the unduly constraining effects of red tape. Researchers have shown evidence that measures of general organizational red tape, including the one used in the present study, relate to such factors as alienation (Dehart-Davis and Pandey 2005), public service motivation (Scott and Pandey 2005), and, in complex ways, to performance (Brewer and Walker 2005), as well as to other variables.
In comparing public and private managers, researchers have found large differences in the way the two groups of managers responded to questions about red tape, and whether their organization's personnel rules constrain pay and dismissal. Other studies have found that public managers perceive higher levels of red tape in personnel rules (Baldwin 1990;Bozeman and Bretschneider 1994;DeHart-Davis and Pandey 2005), information red tape (Baldwin 1990), and red tape as a delay in task completion (Bretschneider 1990). These results are very consistent with studies of public and private managers' perceptions about the constraining effects of personnel rules on dismissal of poor performers and on rewarding good performers with higher pay. The studies individually have limited samples, but together show striking consistency in their support for differences between public and private managers. Rainey (1983), for example, reported that a sample of federal and state government middle managers in Ohio differed sharply from middle managers in Ohio business firms, on questions about personnel flexibility (i.e. absence of constraints on tying pay and dismissal to performance). Later, in the early 1990s a survey in the Syracuse, New York area found similar differences between large and diverse samples of government and business managers, on the same questions (Rainey, Pandey, and Bozeman 1995). Table 1 shows the mean scores for two of the questions from these surveys: (1) Even if a manager is a poor performer, formal rules make it hard to remove him or her from the organization and (2) The formal pay structures and rules make it hard to reward a good manager with higher pay here. Also indicated on table 1 are additional surveys administered in Florida and Denver (Moon and Bretschneider 2002;.
[ Table 1 about here] Table 1 indicates that public sector respondents compared to private sector respondents consistently expressed more agreement with the two statements. Table 2 provides a more straightforward look at the sharp differences. In the Ohio survey, over eighty percent of the public managers felt that the statements about constraints on higher pay and about the difficulty in firing were true, while much smaller percentages of private business managers said the statement was true. In the New York survey, about 15 years later, nearly ninety percent of the public managers agreed that it is hard to remove a poor manager and reward a good manager with higher pay, compared to very small percentages of the business managers who agreed.
These differences remained large even when taking into account numerous control variables.
[ Table 2 about here] These strong findings come from samples representing different levels of government, government agencies and business firms, geographic regions, and time periods. Quite significantly, they contrast sharply with Nobel Laureate Herbert Simon's aggressive claims about reward conditions being "precisely the same" in government, business, and nonprofit organizations. Still, given the vastness and complexity of the three "sectors," these surveys offer limited evidence. Further studies can provide evidence about the generalizability and durability of such findings. The comparison of a public and a nonprofit sample in the current study provides evidence about whether the differences will be similarly strong.

Nonprofit Organizations: Relevant Research
While there are very few direct comparisons of public and nonprofit organizations on red tape and flexibility of personnel procedures (Lan and Rainey 1992), there is research on nonprofit organizations relevant to this topic. As do public organizations, private nonprofit organizations pursue public missions without providing financial gains to stockholders or individual owners.
Nonprofit organizations balance public and private interests, often engage in activities similar to those of government organizations (e.g. health care, education, social services, and disaster relief), but operate apart from the control of the market and the state (Anheier and Seibel 1990) and outside of the constraints of government personnel systems.
Increasing interest in research comparing government and private nonprofit work environments has led to diverse findings that indicate similarities, but also important differences in attitudes, motives, and perceptions in the two workforces. For example, research indicates that nonprofit workers report higher job commitment compared to public sector employees (Borzaga and Tortia 2006;Light, 2002;Mirvis and Hackett 1983). On the other hand, studies have found that that both public and nonprofit workers are public service-minded (Rotolo and Wilson 2006) and place greater importance on personal relations than workers in the private for-profit sector (Rawls et al. 1975).
Although research indicates that nonprofits share some similarities with public organizations, such as dependence on external sponsors (Corder 2001) and level of managerial professionalism (Berman and West 1998), nonprofit organizations do not operate within civil service systems. Such distinctions lead to claims that nonprofits are more flexible and responsive than many government institutions (e.g., Hall 1992). Nonprofit organizations manage a complex labor force composed of paid and volunteer staff (Brudney 1998a(Brudney , 1998bWeisbord 1997) and must negotiate their personnel policies with boards of directors and funders (Corder 2001;Hall 1992;Weisbrod 1988). Observers nevertheless, contend that nonprofit organizations generally operate under more flexible rules and fewer constraints than government organizations (Eggleston and Zeckhauser 2002;Hall 1992), but are these claims true? Nonprofit organizations do face challenges and constraints due to external controls and multiple stakeholders including boards, trustees, service recipients, volunteer and paid labor, and government and corporate investors (Salamon 1998). Some studies have found that personnel rules in nonprofit organizations are not as highly formalized as in government organizations (Kalleberg et al. 2006), but formalization is distinct from red tape and constraining personnel rules that we analyze in the current study (Pandey and Scott 2002;Welch and Pandey 2007). In general, red tape researchers note a distinction between formalization and red tape -formalization is neutral, whereas red tape is negative (Bozeman 2000;Bozeman & Kingsley 1998;Bozeman & Scott 1996;Pandey & Scott 2002). 3 Some researchers argue that sector differences are narrowing as nonprofit and public organizations become more commercial and businesslike (Kettl 1993;Weisbrod 1988). This makes it all the more important to analyze whether this narrowing might be taking place, and what differences it may be making. For example, if this narrowing is having substantial effects we should find that managers in state government and nonprofit organizations do not differ in their perceptions of red tape and hiring, firing, and promotion systems, taking into account the 3 Formalization is defined as "the extent to which rules, procedures, instructions and communications are written" (Pugh, et al., 1969: 75). Red tape, in contrast, is defined as rules, regulations, and procedures that have a compliance burden but do not achieve the functional objective of the rule (Bozeman 2000). Red tape is inefficient and has negative outcomes -there is no such thing as good red tape. It is possible that highly formalized organizations have low levels of red tape and organizations with weak internal controls can have high levels of red tape (Bozeman and Loveless 1987). Welch and Pandey (2007) offer a methodological test that red tape and formalization are different concepts, and conclude that although formalization is positively associated with general organizational red tape it is not a significant determinant of procurement red tape. Thus, while formalization and red tape are both concerned with rules, regulations and procedures, they remain distinct concepts. possibility that public and nonprofit organizations attract, select, and retain different types of people (Harpaz and Snir 2003;Schneider 1987;Schneider et al. 1995)? On the other hand, research indicates that most of the factors that increase red tape in the public sector are external sources such as govenment civil service and purchasing systems that should be independent of the alleged narrowing described above (Bozeman and Bretschneider 1994;Brewer and Walker 2006). We hypothesize that nonprofit managers, who are not subject to civil service protections and restraints, will report more personnel flexibility in their organizations, compared to public managers.

H1a: Perceptions of organizational red tape vary between employees in the public and nonprofit sectors.
H1b: Public managers compared to nonprofit managers will report higher organizational red tape.

H2a: Perceptions of personnel flexibility vary between employees in the public and nonprofit sectors.
H2b: Nonprofit managers compared to public managers will report higher personnel flexibility.
Since this is the first study comparing nonprofit and state government managers' perceptions of red tape and personnel flexibility we need to eliminate as many alternative interpretations of the findings as possible. Perceptions of red tape and personnel rules can relate to other factors besides the public versus nonprofit distinction. Hence we take into account factors such as the respondents' motivations to choose their current jobs, because selection into a job and sector may relate to respondent perceptions of red tape and personnel flexibility.
In addition, previous research indicates that a person's desires for extrinsic rewards (e.g. increased salary and benefits) and intrinsic rewards (e.g. feelings of accomplishment and a desire to serve the public) are related to the person's perceptions and attitudes about the organization in which she or he works. Researchers have investigated the motivations and values related to selecting work in a particular sector (Jurkiewickz et al. 1998;Lyons et al. 2006;Mirvis and Hackett 1983), including public service motivation (e.g., Hondeghem 2008a, 2008b), a desire for job security, the opportunity for equal opportunity hiring, and the need to control the amount of time spent at work (Izraeli 1990). Both nonprofit and public sector employees report higher intrinsic rewards than those in for-profit organizations and a lower likelihood of receiving extrinsic rewards such as increased pay or promotion (Mirvis and Hackett 1983;Wittmer and Coursey 1996). Workers in the nonprofit sector, compared to the for profit sector, are more people oriented (Rawls et al. 1975) and care more about serving the public interest (Wittmer and Coursey 1996) and advancing the organization's public service mission (Light 2002). In comparison, some research indicates that public sector workers value opportunities for advancement and intellectually stimulating and challenging work more than nonprofit workers (Crewson 1995, 94;Lyons et al. 2006). These diverse findings suggest that public managers and employees might differ from nonprofit comparison groups on attitudes and perceptions that might influence perceptions about red tape and personnel rules, and hence need to be taken into account in our analysis.
There is a great deal of research investigating work life and motivation among public, nonprofit, and for-profit private sector workers (e.g., Wechsler 1990, 1991;Light 2002;Steinhaus and Perry 1996). People who are interested in job security may have a higher tolerance for red tape and personnel rules and restrictions. Those driven by a desire for career advancement may have strong negative perceptions about personnel inflexibilities especially when those personnel restrictions impede promotion and advancement. Individuals driven by a desire to serve the public interest may tolerate higher levels of personnel restrictions (Scott and Pandey 2005) or they may resent them as an impediment to the organization's mission (Moynihan and Pandey 2007). Given the variations in research on work life by sector, we predict that motivations for selecting a job in the public or nonprofit sector (public service motivation and a desire for job security, career advancement, and financial benefits) will be related to perceptions of red tape and personnel flexibility. Public and nonprofit organizations can also differ in the external institutional context in which they operate, and we represent such variations by comparing respondents from Illinois to respondents from Georgia. We expect to find variation across states in perceptions of red tape and personnel flexibility for both nonprofit and state managers. First we expect that perceptions of rule flexibility will differ between managers in Georgia and Illinois due to state government human resource reforms. Georgia underwent reforms in the late 1990s that have resulted in about 72% of Georgia state government employees being at-will-employees. The Georgia reforms also included a pay-for-performance initiative aimed at tightening the linkage between pay raises and performance evaluations. In addition, Georgia restricts the number of issues open to grievances more than Illinois (Hays and Sowa 2006). Illinois' public sector has not expanded its at-willemployment beyond its standard 20% of the workforce and continues to offer a wide range of issues open to grievances. We expect that the expansion of at-will hiring in Georgia will drive lower perceptions of red tape and more positive views of personnel flexibilities compared to perceptions among Illinois state managers.
We also analyze organizations by state. State-level regulation of nonprofit organizations varies across states in reporting requirements, state laws, tax codes, and tort law (Irvin 2005).
State regulation of nonprofit organizations in Georgia and Illinois creates conditions that should influence employees' perceptions. Differences in state regulations over nonprofits and in state government human resource reforms lead us to expect that perceptions of red tape and personnel flexibilities in nonprofit and public organizations will vary by state.
H4a: Perceptions of red tape will vary by state.
H4b: Perceptions of personnel flexibility will vary by state.

Data and Methods
This study analyzes data from the NASP-III questionnaire, which was administered to a random from the nonprofit sector. The overall response rate is 39% (43% for the public sector and 33% for the nonprofit sector). Fifty-five percent of the public sector respondents and one quarter of the respondents from the nonprofit sector work in Georgia. Although this response rate is lower than previous NASP studies, it is not surprising inasmuch as public sector respondents regularly respond at higher rates than those in private and nonprofit organizations. Unlike other public sector surveys, NASP III was administered without the assistance of state government and 4 NASP III included respondents from 102 public agencies and 306 nonprofit organizations. The organization names are too numerous to enumerate here, but include organizations in areas of public safety, natural resources, social services, education, housing, animal protection, environment, agriculture, telecommunications, human resources, law, finance, taxation, industry, science, research, health and medicine, commerce, utilities, architecture, building and real estate, engineering, tourism, and the arts. For a full list of the organization functions, please see the NASP III codebook available at http://www.uga.edu/padp/nasp.html. agency heads or managers. Previous studies have found that high-ranking individuals in nonprofit organizations are less likely to respond to surveys. The response rate remains higher than average response rates in general surveys of organizational employees (Kanuk and Berenson 1975). See Appendix I for details about the study approach and Appendix II for descriptive statistics.   Scott and Pandey 2005), and in more recent research (Boyne, Brewer, and Walker 2006;Walker 2005, 2006). Brewer and Walker (2005) describe the first two personnel flexibility items as measures of personnel red tape and use these items as both separate independent variables (2005) and as a summed index dependent variable (2006).

DeHart-Davis and Pandey (2005) and Pandey and Welch (2005) combine Performance-based
Removal and Performance-based Rewards plus two additional items to create a variable that they call human resources red tape. Our third item, Performance-based Rewards, is distinct from previous uses, in that it asks about the ability to reward a good employee, not just a manager.
We ran a principal components factor analysis on the three personnel flexibility items (see table 3). The three items load on a single factor, with an initial eigenvalue of 1.795. The principal components analysis suggests that the three items are strongly correlated and measuring a shared underlying concept, thus we created an index by summing the three items.

Independent Variables.
We use a dummy variable, Nonprofit, to test for variation in perceptions of personnel rules by sector. Nonprofit is coded one if the respondent works in the nonprofit sector (0=respondent works in the public sector).
We include four variables measuring the reasons a respondent took the current position: respectively. Security Motivation and Career Advancement Motivation are saved factor scores from a factor analysis using an orthogonal solution and Varimax rotation of the following items: • Opportunity for advancement within the organization's hierarchy • The organization's pension or retirement plan • Desire for increased responsibility • Benefits (medical, insurance) • Few, if any, alternative job offers The factor analysis resulted in two dimensions: security and advancement, which represent 63% of the common variance in the initial correlation matrix. A reviewer of this article raised the interesting and important question of whether the "publicness" of a nonprofit organization (Bozeman 1987), and in particular the "financial publicness" (the level of funding from government) could explain similarities between nonprofit and public organizations. As nonprofits receive more government funding through contracts and grants, will their personnel rules and procedures become more like those of government organizations? The reviewer suggested that this possibility makes it important for us to control for financial publicness. Our data do not allow us to control for this variable, but the present analysis remains valuable without this control. The available evidence suggests that financial publicness should not have strong effects on the personnel flexibility variables in our study. Bozeman and Bretschneider (1994) compared financial publicness and "core public versus private" (a public versus private distinction) as predictors of personnel constraints and other dependent variables in R&D labs. They found that financial publicness related significantly to such variables as the types of products a research and development organization produced, but it showed no significant relationship to personnel constraints. Core publicness, on the other hand, showed highly significant relations to personnel system constraints. This indicates that presence or absence of a civil service system, which usually comes with "public" status, is more influential on personnel system characteristics than financial publicness. This important question  (Bridges and Villamez 1994). We also include an interaction variable for manager and tenure at the current job, Manager*Tenure. The variable, Promotion, is a dummy variable coded one if the respondent's current job was a promotion. 8 Research indicates the need to take into account previous work experiences because such experiences can shape perceptions of a current work environment (e.g., Dodd-McCue and Wright 1996). Experiences in one sector affect work behavior and perceptions when an individual switches sectors (Feeney 2008;de Graaf and Van der Wal 2008). Feeney (2008) finds that previous private sector work experience shapes public managers' perceptions of the public and nonprofit sectors and Boardman and colleagues (in press) find that public managers with private sector work experience have significantly different work attitudes compared to those who have spent their entire careers in the public sector. We use a dummy variable, Private, to control merits further attention in future research. In addition, if financial publicness is making nonprofit organizations more like public organizations, this makes our public versus nonprofit comparison more stringent. We should be less likely to find the difference that we report in this study. 8 It is possible that in some cases the current job was a demotion. Because demotions are uncommon in the public sector, we did not ask for this information. Furthermore, we felt any small amount of additional information gained would be at too great a price in terms of diminishing rapport with the respondent.
for previous private sector work experience. Private is coded one if the respondent's previous job was in the private sector. 9 We also control for the respondent's activities outside of the workplace. Total Civic Activities is an additive index of responses to a series of dummy variables listing organizations or groups to which the respondent might belong. Total Civic Activities, ranging from zero to eight, is an indication of the respondent's involvement in non-work organizations.
Previous research  finds that organizational size is significantly related to perceptions of red tape among public and private managers. Org. Size is a continuous variable indicating the number of full time employees in each respondent's organization. We also conducted analyses of the relationships between organizational function and the dependent variables and found that organizational function did not significantly affect the model or results (See Appendix III). 10 Finally we control for demographic characteristics, which may shape public and nonprofit managers' perceptions of red tape and personnel flexibility. Since many hiring preference programs have been designed with women and minorities in mind, female and nonwhite respondents may be more accepting of personnel restrictions, which they may see as protecting their interests and jobs. Female is coded one if the respondent is female. Age is a continuous variable. Nonwhite is coded one if the respondent self-reported any race other than white/Caucasian. 9 We also tested variables that indicate if the respondent's previous job was in private, nonprofit, and public sectors, tenure in the private sector, nonprofit, and public sectors, and if the current job was a sector switch. 10 We considered the possibility that organizational function is related to perceived red tape and personnel flexibility. The sample was drawn from about 306 nonprofit organizations and 103 state agencies. These organizations vary in type, mission, and size. We coded the organization data into 15 functional types of organizations. We ran the models first with a single categorical variable for organizational type and then with a set of 15 dummy variables. The organizational function variables were not significant in any of the models and did not significantly affect the relationships between the other variables or the model specificity. We include a table in Appendix III, which shows the mean responses for the rules items, by organizational function.
We use an OLS regression model to predict the red tape scale and the personnel flexibility index. In addition to the models predicting the dependent variables with sector as an independent variable, we ran separate models for the public and nonprofit groups. When we control for sector, with a single independent variable, all the variance between nonprofit and public sector respondents falls into the intercept. By splitting the sample, we allow each independent variable and the outcomes to vary by sector and are able to assess how these relationships vary for individuals selecting into the public sector as compared to those selecting the nonprofit sector. The separate models for public and nonprofit managers can provide significantly different beta estimates on the independent variables, giving a more detailed understanding of how the relationships between all of the variables differ by sector.

Results
We examine perceptions of red tape and the flexibility of personnel structures in the public and nonprofit sectors discussing the results for all of the models in order of the four hypotheses, followed by a discussion of the control variables. The results are presented in tables 5, 6 and 7.
[Tables 5, 6 and 7 about here] A first step in this analysis examines whether respondents in the two sectors differ significantly in their perceptions of red tape and personnel flexibility. The model predicting the perception of Organizational Red Tape confirms H1a and H1b (table 6). Respondents who work in the nonprofit sector report significantly lower levels of organizational red tape compared to public sector workers. Working in the nonprofit sector is related to a nearly three point lower ranking of red tape, as compared to working in the public sector.
To test our second hypotheses on personnel flexibility (H2a and H2b), we compared the mean scores by sector for the entire sample on the three rules items (table 5). Respondents working in the public sector are significantly more likely than nonprofit respondents to agree that formal rules make it hard to remove a poorly performing manager and to reward a good manager with higher pay. Nonprofit respondents on average report significantly more agreement that because of organization rules promotions are based mainly on performance. The comparison of means in table 5 confirms H2a and H2b; perceptions of personnel flexibility vary between employees in the public and nonprofit sectors and nonprofit managers, compared to public managers, report more flexibility to remove and reward based on performance. Table 7 further indicates that the public versus nonprofit distinction relates significantly to the personnel flexibility index, with nonprofit respondents reporting more personnel flexibility. Respondents from public organizations perceive that the rules place stronger constraints on dismissal, tying pay to good performance, and tying promotion to performance.
By testing the outcomes for the full sample and the split models, we are able to see the strength of the variable indicating whether the respondent works in the public or nonprofit sector.
For example, in the model predicting organizational red tape for the full sample the adjusted R squared is .337. When we split the model the adjusted R squared for the nonprofit sector is .176 and for the public sector 0.078. In practical terms, the full model, with a variable indicating sector explains nearly 34% of the variance in red tape perceptions, whereas the variables in the nonprofit model explain 17.6% and in the public sector model a mere 7.8%. We find similar explanatory power in the models predicting personnel flexibility, with the split models predicting 15.4% and 17% of the variance in the nonprofit and public samples, respectively, but the full model (including a variable for sector) explaining 47% of the variance in the dependent variable.
Thus, sector is a strong predictor of variance in red tape perceptions and personnel flexibility.
The OLS models enable us to investigate the remaining hypotheses about the relationships between perceptions of red tape, personnel flexibilities, work motivations, and state (Illinois versus Georgia). First, we find some support for hypothesis 3a, about security motivation and career advancement motivation and red tape perceptions. Respondents who report higher desire for job security as a motivation for taking their current job report higher perceptions of red tape in their organizations. Conversely, those who took their current job with a desire for career advancement report lower perceptions of red tape.
When we split the model by sector, we see that nonprofit respondents with a stronger desire for job security report significantly higher perceptions of red tape as compared to nonprofit respondents with lower desires for job security. Meanwhile, public sector respondents who report that career advancement was an important motivator for taking the job have significantly lower perceptions of red tape. In the public sector model, we find no significant relationships between a desire to serve the public and salary as a motivation for taking the current position and perceptions of red tape. With regard to perceptions of personnel flexibility, among public managers career advancement and job security motivation are not significantly related to perceived personnel constraints. However, public managers who report a lower desire to serve the public report fewer personnel constraints. Thus, we find mixed support for our second hypothesis that individual motivation for selecting a public or nonprofit sector position shape managers' perceptions of personnel flexibility in their respective sectors. We confirm a relationship between a desire for job security and for promotion and perceptions of personnel flexibility.
Our models confirm hypotheses 4a and 4b and indicate that working in Georgia is significantly and negatively related to perceptions of red tape and personnel constraints.
Respondents in Georgia report significantly lower perceptions of red tape compared to respondents in Illinois. This finding holds for the split models as well. Being a nonprofit manager in Georgia, as compared to Illinois, is significantly related to a 1.181 lower rating of perceived organizational red tape. Similarly, respondents in the public sector in Georgia report significantly lower organizational red tape than their Illinois counterparts. Respondents in Georgia, as compared to Illinois, also report higher levels of perceived personnel flexibility.
Respondents from Illinois, in both the public and nonprofit sectors, report higher personnel constraints than their respective Georgia counterparts. This finding supports the conclusion that external institutional context can influence the dependent variables, and confirms the need to take such factors into account.
Next, we turn to the control variables in these models. Being a manager and reporting that the current job was a promotion are both negatively related to perceptions of personnel constraints in the work place but not significantly related to general perceptions of red tape in the full models. By splitting the nonprofit and public sector respondents, we see that the negative relationship of being a manager and being promoted to perceptions of personnel constraints holds for public sector employees, but not for nonprofit respondents. Respondents who are public managers, compared to professionals, technicians, and other higher-ranking non-managers in the public sector, are significantly more likely to report lower personnel flexibilities.
We now turn to the relationships between previous work experience in the private sector and perceptions of red tape and personnel restrictions and flexibilities. We find no relationships between previous work experience in the private sector and perceived red tape and personnel flexibilities. The split models (table 7) show that nonprofit respondents whose previous position was in the private sector are significantly more likely to report higher perceived personnel constraints in the current position, as compared to nonprofit respondents who had not previously worked in the private sector.
Finally, although we find no significant relationships between gender and perceptions of red tape and personnel flexibility, we find that nonwhite respondents compared to white respondents report significantly lower perceptions of red tape and among nonprofit respondents, white respondents report significantly higher personnel constraints than nonwhite respondents.

Discussion and Conclusions
In the last few decades, empirical research investigating public and private sector distinctions has indicated that the public sector has a distinct set of constraints, rules, and red tape related to personnel procedures (Baldwin 1990;Bozeman and Bretschneider 1994;Walker 2005, 2006;Kalleberg et al. 2006;Lan and Rainey 1992;. The research that we report here extends our understanding of sector distinctions by comparing public and nonprofit managers' perceptions of red tape and personnel flexibilities. We find that state government managers, compared to nonprofit managers, are significantly more likely to report higher levels or red tape and less flexible personnel rules and regulations.
We find that the state where the respondent is employed, Georgia compared to Illinois, is significantly related to perceptions of red tape and personnel flexibilities. Respondents working in Georgia perceive significantly lower levels of red tape and higher levels of personnel flexibility compared to respondents in Illinois. These differences in perceptions of rules and constraints appear to be due to personnel reforms in Georgia or the presence of unions in Illinois, or both. Factors that can influence such perceptions, including strength of unions in different states and differences in administrative culture need attention in future research.
We find that when controlling for sector, there remains significant variation in red tape perceptions related to reporting that the current job was chosen due to a motivation for career advancement (negative) and security (positive). This finding makes sense, if we think of red tape and personnel constraints as providing job security and also possibly rewarding job tenure. Those individuals seeking career advancement will seek positions in organizations with less red tape and personnel constraints. Similarly, it makes sense that people seeking job security will choose positions in organizations with more personnel rules and regulations protecting jobs.
We find no significant relationships between previous private sector work experience and general red tape. Nonprofit respondents, whose previous job was in the private sector, report increased perceived personnel constraints. It is possible that these "sector switchers" are selfselected individuals who move from the private sector to the public and nonprofit sectors for mission reasons and are not significantly disturbed by organizational red tape.
Given the growth of the nonprofit sector in the last decades and the increasing competition between the public and nonprofit sectors for talented public service-minded managers, understanding the distinctions between these sectors will continue to be important to public management scholars. Moreover, this research extends findings from previous surveys (table 1) about personnel rules and constraints by comparing public and nonprofit managers.
Sector based variation in perceived red tape and the personnel flexibility items are important because they point to the personnel distinctions between these mission-driven sectors.
Previous research indicates that perceptions of red tape dampen risk-taking among city-level public managers (Feeney & DeHart-Davis 2009) and negatively affect public managers' organizational commitment, job satisfaction, and psychological attachment to their workplace and morale (DeHart-Davis and Pandey 2005). Furthermore, research focused on personnel flexibility finds that perceptions about the ability to reward a good manager is significantly related to governmental performance and that burdensome rules and regulations lower individual and organizational performance in government (Pandey and Moynihan 2006;Brewer and Walker 2005). At the same time, previous studies have reported evidence that some public managers who perceive personnel constraints in their organizations actually regard those constraints as a necessary aspect of administration in the public sector, valuable as defenses against political intrusions, among other justifications (Rainey 1983). As public service professionals choose between service through government or private not-for-profit organizations, it is important to understand variation caused by personnel constraints and efforts to ensure institutional and political accountability.
Although one should be cautious in generalizing from one study, the evidence provided here can contribute to theory development for public and nonprofit organizations. It adds to a stream of studies finding differences in respondents' perceptions of personnel systems in public and private organizations, and now private nonprofit organizations. These differences -even in a state that has reformed its personnel system to try to diminish or remove them -suggest that government organizations inherently tend towards more formalized and constrained personnel systems as compared to both private and nonprofit organizations. This theoretical proposition can be further developed and tested in future research that should include more analysis of relations between managers' perceptions of personnel system characteristics and other matters such as the appropriateness of constraints. Future research needs also to strive for more evidence of the relations between personnel system characteristics and individual and organizational performance. Policy implications of the evidence here include the indication that public officials seeking to contract with the nonprofit sector in pursuit of greater personnel flexibilities will find evidence in the present study that such an objective is justified. One might argue that the nonprofitization of the public sector will free government from personnel constraints. On the other hand, outsourcing to nonprofits may remove public programs and services from important rules and regulations that ensure accountability and protect employees. The evidence presented here indicates that even though both types of organizations are not-for profit and provide public services, public and nonprofit organizations nevertheless show sharp distinctions in red tape and personnel flexibility. Thus, while the evidence here comes from one study with obvious limitations, it can contribute valuably to policy decisions and to theory development for public and nonprofit organizations.  Adapted from Rainey (1983) and Rainey, Pandey, & Bozeman. (1995). Note: The differences remained large even after introduction of numerous control variables.