Explain the Use of Strategic Management

Public Personnel Management 2016, Vol. 45(4) 335 –359

© The Author(s) 2016 Reprints and permissions:

sagepub.com/journalsPermissions.nav DOI: 10.1177/0091026016669169

ppm.sagepub.com

Article

Toward an Entrepreneurial Public Sector: Using Social Exchange Theory to Predict Public Employee Risk Perceptions

Anna Fountain Clark1

Abstract An entrepreneurial public agency pursues the implementation of innovative programs that may broaden public service choices, increase service quality, and more effectively serve citizens. Such public entrepreneurship depends on risk taking and risk tolerance; however, public servants tend to be generally risk averse in their behaviors and personal preferences, and are therefore less likely to pursue entrepreneurial approaches to public problems. Using social exchange theory as a framework to understand the reciprocal relationship between agency and employee, this study examines whether agency behaviors might alter the risk aversion of those employees and make the agency environment more conducive to entrepreneurship. Findings suggest that managers’ demonstration of risk tolerance, reward for creativity and innovation, and agency solicitation of employee input are positively related to employee perceptions of higher risk tolerance among their peers.

Keywords risk, innovation, public sector, entrepreneurship

As public managers look for new ways to operate in an environment increasingly dominated by budgetary pressures and amplified demands for program effectiveness, they have come to consider adoption of practices from the other economic sectors (Townsend, 2013). Public entrepreneurship, which implicates agency support for

1Texas Woman’s University, Denton, TX, USA

Corresponding Author: Anna Fountain Clark, Visiting Assistant Professor, Department of History and Government, Texas Woman’s University, P.O. Box 425889, Denton, TX 76204, USA. Email: [email protected]

669169PPMXXX10.1177/0091026016669169Public Personnel ManagementClark research-article2016

336 Public Personnel Management 45(4)

innovation, creativity, and risk taking, has been recommended for public sector adop- tion by scholars and practitioners over the past quarter century (Bellone & Goerl, 1992; Llewellyn & Jones, 2003). An entrepreneurial orientation in the public sector offers opportunities to provide higher quality services or expand public benefits and services (Y. Kim, 2010b) and address program inefficiencies (Llewellyn & Jones, 2003). Public entrepreneurship is about more than revenue generation or aligning with private businesses; it implicates novel approaches to governance, the ambitions of public agencies, and the professional pride of public servants (Llewellyn & Jones, 2003).

Although prior scholarship has not arrived at a definitive list of all characteristics of entrepreneurship, multiple studies indicate that risk tolerance and risk taking are fundamental to its practice (Caruana, Ewing, & Ramaseshan, 2002; Lumpkin & Dess, 1996; Morris & Jones, 1999). Y. Kim (2010a, 2010b) argues that risk taking is a fun- damental dimension of entrepreneurship upon which a multitude of scholars agree, and argues that for entrepreneurial activities to succeed in public agencies, their lead- ers and workers must be willing to embrace a modicum of risk tolerance. The present study focuses on risk taking as a primary element of public entrepreneurship because risk aversion poses a major challenge to the feasibility of entrepreneurial governance. The risk aversion inherent in the public sector, both in terms of organizational attri- butes (Bozeman & Kingsley, 1998) and employee preferences and behaviors (Bellante & Link, 1981; Dong, 2014; Roszkowski & Grable, 2009), suggests that entrepreneur- ship may encounter substantial hurdles in the public workforce that will largely pre- vent the entrepreneurial spirit from meaningfully permeating the sector. Rather than raising their collective hands and surrendering to the risk-averse natures of their work- forces, public agencies routinely implement entrepreneurial policies and procedures designed to drive success and respond to ongoing social changes.

Although public agencies cannot control the inner workings of the individuals in their employ, they might potentially engage in practices that demonstrate organiza- tional risk tolerance to enhance employees’ risk tolerances, thereby reducing a serious impediment to entrepreneurial orientation on the job. This study thus poses the follow- ing research question:

Research Question 1: Can the organizational mechanisms that encourage risk tol- erance among public employees, such as reward structures or invitations to partici- pate in decision making, alter the risk preferences of those employees and make the agency environment more conducive to entrepreneurship?

This study aims to identify the relationships between perceptions of employee risk preferences and three entrepreneurial organizational practices that may encourage a greater risk appetite among employees: top management’s demonstration of risk pref- erence, reward for innovation and creativity, and invitations for employees to provide input on agency process or program changes. Social exchange theory (SET) will be used to provide a framework for understanding the nature of these relationships. SET posits that when organizations demonstrate their support for employees via various

Clark 337

support behaviors, employees will tend to reciprocate with positive work behaviors that benefit the organization. This study will test whether such reciprocal social exchanges might extend into the realm of public sector employees’ risk tolerance. Specifically, public employees’ inherent risk aversion may potentially be overcome by the organization’s entrepreneurial orientation, as demonstrated by efforts and policies that support the risk taking that accompanies them.

Entrepreneurship and Risk in the Public Sector Agency

In developing a theory of public entrepreneurship, Klein, Mahoney, McGahan, and Pitelis (2010) argue that entrepreneurial public organizations seek to “unleash creative energy in pursuit of the public interest” (p. 12). The authors suggest that such public agencies may accomplish a number of objectives, including but not limited to, altering the organizational landscape by changing administrative rules, processes, or organiza- tional norms; establishing new agencies or departments; redefining performance objectives; reorganizing preexisting agencies; and creatively managing public needs by developing public–private partnerships. The goal of such activities is not to make the public sector look more like the business sector (Llewellyn & Jones, 2003), though it may be viewed as co-evolving with the private sphere (Klein et al., 2010). Instead, the purpose of public entrepreneurship is “to increase opportunities to take challenge- able ideas and find ways to offer more public choices and benefits, providing high- quality services to citizens” (Y. Kim, 2010b, p. 781). This may translate to improvements in internal capacity (Bozeman, 2007), organizational efficiencies (Llewellyn & Jones, 2003), and overall policy outcomes (Klein et al., 2010).

Public entrepreneurship involves “making a conscious decision to assume uncer- tainty of outcomes” when introducing changes to public services or processes (Y. Kim, 2010b, p. 785). Experimentation with social service delivery, whether tinkering with or fully overhauling public programs, implies a degree of uncertainty of outcome (Windrum, 2008), and is, thus, commonly perceived by civil servants as a risky propo- sition. Lumpkin and Dess (2001) argue that entrepreneurial risk taking entails “a ten- dency to take bold actions such as venturing into unknown new markets, committing a large portion of resources to ventures with uncertain outcomes, and/or borrowing heavily” (p. 431). Given these insights, public entrepreneurship, which entails service and/or service delivery innovations and new venture creation (Llewellyn & Jones, 2003), as well as creative management of public resources and hybridization of public and private sector practices (Klein et al., 2010), is inherently risk-laden.

Risk may be conceptualized as both the lack of predictability of an outcome and the high relative probability of either significant gains or losses associated with that even- tual outcome (Lumpkin & Dess, 2001). To pursue new ideas or directions, a person must find value in the pursuit prior to receiving evidence or justification in support of it (Burton, 2008). Taking a risk means attempting or doing that of which we are uncer- tain. The pursuit of a goal in the face of uncertainty or without prior justification requires confidence, faith, and determination; taking a risk means going out on the proverbial limb without assurance that it can bear the weight. The need for certainty,

338 Public Personnel Management 45(4)

and the preference for lower risk that it implies, influences a person’s choice patterns and affects their response to ambiguity (Mitchell, 1995; D. T. Wilson & Mathews, 1971). Simon (1979) recommends flexibility and satisficing in the face of uncertainty, but some will have more visceral reactions, as uncertainty may incite trepidation and fear in some while it may inspire bold action in others.

Risk tolerance is contextual (Ganzach, Ellis, Pazy, & Ricci-Siag, 2008; Kahneman & Tversky, 1979; March & Shapira, 1987) and predicated upon the environment in which it is contemplated (Wildavsky & Dake, 1990). As in other social or cultural environments, the degree of optimal risk tolerance or aversion in an employment set- ting will vary by organization and job type (Molina-Morales, Martínez-Fernández, & Torlò, 2011). That is, risk taking is neither an absolute good nor something to be uni- versally avoided; rather, its value is conditioned upon the environment in which it is contemplated. In environments that prioritize responsiveness, accountability, and effi- ciency, such as public service agencies, any change that may alter or threaten service quality or quantity may be met with resistance and perceived as a threat (Drennan, McConnell, & Stark, 2015). However, by expressing support for risk-laden entrepre- neurial activities, organizations may imply that they welcome employees to express varied or unpopular opinions and will tolerate failures that result from employees’ attempts to work in the agencies’ best interests. As such, risk tolerance in the name of entrepreneurship may spur higher levels of employee risk tolerance in a multitude of workplace activities.

Risk taking faces significant barriers in the public sector. To some extent, this can be viewed as a result of the organizational environment, which is characterized by a preference for predictability and general tendency toward risk aversion (Bellante & Link, 1981; Chen & Bozeman, 2012; Dong, 2014). As Merton (1940) famously argues, strict adherence to bureaucratic rules “induces timidity, conservatism, and technicism” among agency employees (p. 564), such that any effort to operate outside of adminis- trative norms (e.g., taking a creative problem-solving approach not previously pre- scribed by the organization) may be viewed as non-conformity and trigger antagonism from those who adhere to agency customs. Public servants who envision themselves as entrepreneurial “loose cannons and rule breakers” (Borins, 2000, p. 498) will likely find themselves cutting against the grain in agencies that are generally characterized as slow-moving, conservative, and inhospitable to creative or innovative pursuits (Bloch & Bugge, 2013; Windrum, 2008).

Public sector risk aversion has also been linked to the types of individuals who are employed within these bureaucratic structures. For instance, common mythology por- trays government employees as finger pointers who fear mistakes and accept medioc- rity rather than pursuing creativity and excellence (Goodsell, 2003). Goodsell (2003) indicates that public employees have been traditionally characterized as unable to learn, inflexible, unimaginative, rigid, and stagnant in their work behaviors; such char- acteristics are decidedly antithetical to the flexible, innovative characterizations of public sector entrepreneurs. Furthermore, unlike their counterparts in the business world, public personnel lack a market-based rationale for risk taking, in that they com- monly work in traditional merit systems, where tenure-based pay and promotions, as

Clark 339

well as other job protections, reduce their likelihood to innovate as a means to earning financial rewards or retaining their jobs (Dong, 2014; Gneezy & Rustichini, 2000). The changing landscape of public employment and the elimination of merit systems in several states, however, may suggest that public sector agencies might be actively encouraging market-oriented solutions to public service delivery and administration.

Extensive evidence indicates that public sector employees are the most risk-averse employment group in the three economic sectors. Comparisons of public and private sector employees have repeatedly confirmed that public sector employees are more risk averse than their private sector counterparts in terms of personal behaviors (Bellante & Link, 1981), investment decisions (Roszkowski & Grable, 2009), and employment choices (Dong, 2014). Similar findings have been produced in compari- sons of public and nonprofit sector employees (Chen & Bozeman, 2012). Until recently, scholarship had not produced a clear answer to the chicken-and-egg question of whether the public sector creates risk-averse employees or whether risk-averse employees seek public sector employment. Dong (2014) finds that personal risk pref- erences result in self-selection into the public sector, as risk-averse individuals are more likely to choose to work in the public sector than in the business world, and are more likely to remain employed in the public sector than are their less risk-averse counterparts.

Public employees’ risk aversion may also reflect their commitment to accountabil- ity and desire to avoid backlash or punishment for failed efforts to alter their agencies’ current portfolio of services. Public agency experimentation likely displeases the gen- eral public, who possess low tolerance for a trial-and-error approach to public policy implementation (Eckerd, 2014) and the apparent waste of public resources (Townsend, 2013). Ever watchful for failed efforts and fixated on wastefulness, agency leaders, policy makers, and the media continually scrutinize public employees’ decision mak- ing and frequently jump at the chance to identify and eradicate the source of waste and abuse of public funds (Borins, 2001). To avoid the appearance of impropriety associ- ated with implementing risky public policy, or the ostensibly wasteful, long, and daunting process associated with authorizing merely incrementally innovative changes (Townsend, 2013), a public servant might, indeed, be wise to keep his or her head down and make as few waves as possible in the organization.

The proliferation of New Public Management (NPM) and its emphasis on entrepre- neurship have sharpened the focus on risk and risk taking by public employees in the name of innovative solutions to public problems. NPM intensified the emphasis on public service quality relative to its cost and drove public agencies to explore novel solutions to problems of efficiency and effectiveness (Windrum, 2008). Recent schol- arship on entrepreneurial approaches to public service suggests that market-like behaviors, and the greater tolerance for risk that they imply, offer multiple benefits to public agencies and their constituencies, including, but not limited to, flexibility in addressing public problems (Y. Kim, 2010b), new or expanded services (Llewellyn & Jones, 2003), and better overall agency effectiveness (Borins, 2014; Y. Kim, 2010a). Despite the potential it holds for improving agency performance, the risk-averse pub- lic agency climate remains a significant barrier to entrepreneurship in the public sector

340 Public Personnel Management 45(4)

(Koch & Hauknes, 2005; Townsend, 2013). If public agencies aim to respond to the rapidly changing world in a timely and effective manner, they must develop the orga- nizational and human capital capacities to take calculated chances on unproven, inno- vative, or otherwise risky changes to public service provision.

Social Exchange and Employee Risk Tolerance

Scholars regard SET as one of the most influential paradigms for understanding and explaining behavior in the workplace (Cropanzano & Mitchell, 2005). On a broad scale, SET proposes that the agency’s treatment of its employees has consequences for employee behavior. SET characterizes workplace relationships as voluntary social and/or economic exchanges (Blau, 1964), wherein the agency and its employees engage in a “two-sided, mutually-contingent, and mutually rewarding process” (Emerson, 1976, p. 336) that reinforces pro-agency employee behaviors. That is, agency reward structures or other support for employee performance on the job will encourage employees to work in the agency’s interests by way of enhanced productiv- ity or commitment (Gould-Williams & Davies, 2005). Conversely, lack of incentives for employee behavior or overt punishment for particular employee activities would be expected to discourage employee behaviors that are opposed by the organization.

Social exchange may be classified as either negotiated or reciprocal in nature (Blau, 1964; Emerson, 1976; Molm, Takahashi, & Peterson, 2000). While negotiated exchanges require that actors engage in concerted decision-making processes to reach agreement on the exchange, reciprocal exchanges are individual efforts performed for the purpose of benefiting another party without knowledge of when or if the other party will reciprocate in the future (Molm et al., 2000). Reciprocal exchange relation- ships begin when one party takes action, and if the other follows suit, subsequent rounds of exchange may begin. Once in motion, the reciprocity cycle is not only a continuous, self-reinforcing one (Cropanzano & Mitchell, 2005) but also predicated upon some degree of risk and trust between the individual actors. In these types of social exchanges, actors may initially be unsure when or if their actions will be recip- rocated by the other, and they must therefore proceed in the face of uncertainty until the other party responds in kind. Once reciprocal exchange has been made, trust devel- ops between the parties that their efforts will be rewarded and the bond between them will strengthen (Molm et al., 2000). This form of reciprocal exchange generates inter- dependence between the involved parties, which offers the promise of a high-quality, lasting relationship between them (Cropanzano & Mitchell, 2005).

In the agency–employee relationship, reciprocal exchange implicates the employ- ees’ felt obligation to “base their concern with the organization’s welfare and work effort on how favorably they have been treated by the organization” (Eisenberger & Stinglhamber, 2011, p. 56). Eisenberger and Stinglhamber (2011) argue that employee behavior and work effort will reflect their perception of how the agency regard them, as indicated by reward structures, support for professional development, and job secu- rity. Where employees feel valued and trusted by the organization, fairly rewarded for their efforts, and empowered by the agency to use their discretion on the job, they will

Clark 341

respond with higher levels of motivation, commitment, and intentions to remain in their jobs (Gould-Williams & Davies, 2005).

This study extends SET to a new area of inquiry: explicit organizational support for risk taking on the job, and the potential it holds for public entrepreneurship. SET suggests that when organizations indicate their support for employees and their activities, employees will reciprocate with better performance and commitment (Eisenberger & Stinglhamber, 2011; Gould-Williams & Davies, 2005). Three indicators that imply the organization’s tolerance for risk—employees’ perceptions of top management’s risk tolerance, organiza- tional reward for innovativeness and creativity, and the agency’s invitation for employees to participate in organizational decision making—will be assessed for their relationships to perceptions of employees’ risk tolerance. The expectation is that the presence of these three indicators of organizational support for risk taking may send the message that the agency supports a measure of risk-taking behaviors among its employees and possesses greater tolerance for errors produced in the name of public entrepreneurship.

Top Managers’ Risk Tolerance

Top organizational managers exert considerable influence over subordinate employ- ees’ behaviors and perceptions by conveying and reinforcing organizational culture (Bozeman & Kingsley, 1998; Rainey, 2009; Schein, 2010). Employees are then encouraged to embrace and proliferate that culture for the benefit of the agency. Agency executives accomplish the construction of organizational culture through a variety of mechanisms, such as methodical role modeling or coaching, strategic human resource practices, and calculated responses to agency crises (Rainey, 2009). In gen- eral, organizational managers are cognizant of the contributions they make to the agency environment and culture; they tend to recognize that employees emulate mana- gerial behaviors on the job and will attempt to alter or improve their own performance to stimulate their employees to follow suit (de Jong & Den Hartog, 2007).

For agency employees, top managers tend to embody or represent the organization as a whole. Eisenberger and Stinglhamber (2011) offer explanation as to the extent to which top management represents the agency from the perspective of subordinate employees. The authors argue that because employees tend to view their agencies as having a personality that is embodied in executives, those executives’ actions and communications may come to characterize those of the organization itself. Thus, in addition to top managers’ involvement in the creation and dissemination of organiza- tional culture, executives contribute significantly to the establishment and endorse- ment of the organization’s support for employee behavior (Bozeman & Kingsley, 1998; Eisenberger & Stinglhamber, 2011). Eisenberger and Stinglhamber further argue that support received from one’s top managers is more indicative of overall organizational support than that which comes from other coworkers or immediate supervisors. Managers’ priorities represent the extent to which they view employees as valued human capital. Their high organizational status lends credibility to their inter- actions with lower level employees and demonstrates employees’ organizational value (Eisenberger & Stinglhamber, 2011).

342 Public Personnel Management 45(4)

Top managers are more likely than mid-level managers or professionals with mana- gerial responsibilities to self-identify as entrepreneurial and risk tolerant (Llewellyn, Lewis, & Woods, 2007), so it comes as no surprise that they are particularly important in influencing employees’ risk perceptions (Bozeman & Kingsley, 1998). When top managers articulate what types of risks will be tolerated, or imply their tolerance via cultural traditions in the organizational environment, they may dramatically affect their employees’ risk perceptions (Sitkin & Pablo, 1992). Top managers’ attitudes and behaviors, such as flexibility and the recognition of creativity, have been found to affect employees’ acceptance of risk taking in the organization (S. Kim & Yoon, 2015), as employees who sense managers’ support of such behaviors report a heightened sense of innovation-driven agency culture. Top managers’ personal characteristics, including organizational tenure and innovative attitudes, have direct effects on their agencies’ likelihood to adopt program or process innovations (Damanpour & Schneider, 2009). To spur greater tolerance for risk taking on the job, top managers can indicate their trust in employees (Neves & Eisenberger, 2014), as well as demonstrate the agency’s tolerance for errors and reluctance to punish employees for failures asso- ciated with attempted innovations (Townsend, 2013).

The exchange relationships between top management and agency employees offer the potential to generate a sense of obligation toward the organization as a whole (Wayne, Shore, & Liden, 1997) and encourage employees to behave in a manner con- sistent with agency preferences. When agency executives indicate their support for preferred employee behaviors (i.e., risk taking on the job), they demonstrate their understanding that agency outcomes rely, in part, on their support for entrepreneurial risk-taking behavior and employees’ reciprocal performance. Furthermore, based on their high position in the organizational hierarchy, top managers possess the authority to initiate social exchanges with subordinate employees and sustain social exchange cycles by communicating or rewarding desired outcomes. Therefore, based on the capacity of top management to define an entrepreneurial agency culture and affect the behaviors of subordinate employees through reciprocal exchange, it is hypothesized as follows:

Hypothesis 1: Managerial risk tolerance will be positively related to employees’ risk tolerance.

Reward for Creativity and Innovation

Public sector agencies are generally characterized as slow-moving, conservative, and bureaucratic, and viewed as inhospitable to creativity and innovation (Bloch & Bugge, 2013; Windrum, 2008). Bureaucratic processes, demands for accountability, and a ten- dency to discourage non-conformity often prevent public agencies from embracing entrepreneurial approaches to programs or processes. However, as Simon (1997) argues, it is a myth that people are their most creative when they are fully autonomous; indeed, creativity, according to Simon, can thrive within the confines of organizational structure. In concert with Simon’s assertion, public entrepreneurship proponents

Clark 343

advocate the pursuit of innovative solutions to public problems and encourage propa- gation of creativity and innovativeness within the public agency.

Creativity may be defined as “breaking away from existing rules, practices or con- cepts, or crossing or mixing two or more elements not customarily linked” (Gow, 2014, p. 15). Creativity reflects the development of new ideas that offer potential use- fulness to the organization, and occurs when employees engage in activities such as brainstorming or other means of expressing ideas not already in practice in the organi- zation. The expression of creative ideas implicates risk taking, as employees cannot fully foresee the agency’s tolerance of ideas that may be entirely novel, unpopular, or contrary to present practice. Creativity in lower level public employees may be espe- cially valuable in addressing public service inadequacies or enduring financial diffi- culties (Borins, 1998, 2001; Gow, 2014), but cannot produce tangible results unless implemented.

Innovation “is thought to be the way to harness the creative potential of the human race in order to survive, to progress, and to prosper” (Gow, 2014, p. 2), and involves “the introduction of a new idea, method or device” (Merritt & Merritt, 1985, p. 11). Public sector innovation is the manifestation of public employees’ creative efforts; it is the practice of implementing creative solutions to public problems, which is done in the name of the broader public good and to create public value, rather than to produce private value by way of monetary incentives for employees (Light, 1998). Public sec- tor agencies have traditionally avoided innovation—and the risks that are inherent in it—due to the low tolerance for public sector errors among oversight bodies, the media, and the general public (Borins, 2001, 2014).

According to SET, when organizations offer reward opportunities to employees, they indicate that they value employees’ contributions (Gaertner & Nollen, 1989; Rhoades & Eisenberger, 2002). Reward for strong employee performance indicates the organization’s commitment to employees’ self-determination, acknowledgment of employees’ autonomy from the organization, and the employees’ increased compe- tence and control over their work tasks (Eisenberger & Stinglhamber, 2011). An agen- cy’s reward or recognition of employee creativity indicates its support for similar, ongoing behavior (Dewett, 2004). When an employee perceives that creative or inno- vative contributions have been valued by important others within the agency, they may be more willing to respond by taking risks on similar efforts in the future.

The reciprocal exchange that occurs between agency and employee is itself risk- laden, in that employees cannot necessarily predict when or if their efforts will be rewarded or recognized by the agency. When agencies respond to positive employee behaviors with reward or recognition, they teach employees that the organization can be trusted to acknowledge employee efforts (Neves & Eisenberger, 2014), and employ- ees will tend to perpetuate the cycle through similar efforts (Eisenberger, Armeli, Rexwinkel, Lynch, & Rhoades, 2001). The exchange of creative and innovative ideas for reward or recognition builds interdependence between agency and employee, as the agency gains the benefits of employees’ program and process expertise, while employees receive reward for their efforts. Given public employees’ intrinsic motiva- tions in their work (Bright, 2009; Perry, Hondeghem, & Wise, 2010), rewarding their

344 Public Personnel Management 45(4)

efforts likely represents something other than private gain, such as organizational value, trust, and encouragement in their pursuits. It also tends to send a message that the agency supports employees who go out on a limb, express themselves freely in a problem-solving context, and develop innovative solutions to public problems.

Consistent with this argument, in a study of public entrepreneurship in state gov- ernments, Y. Kim (2010b) finds a positive, though statistically insignificant, relation- ship between organizational risk taking and performance-based rewards. The rewards analyzed in Y. Kim’s (2010b) study were offered based on employees’ general job performance, however, and did not specifically assess the relationship between employee risk tolerance and incentives for creative or innovative work behaviors. With an eye toward Y. Kim’s (2010b) prior work, this study argues that the relationship between reward for innovative or creative endeavors and perceptions of employee risk tolerance will be revealed as even stronger than the relationship between reward for general job performance and perceptions of employee risk tolerance. Therefore, it is hypothesized as follows:

Hypothesis 2: Reward for employee innovation and creativity will be positively related to employees’ risk tolerance.

Agency Solicitation of Employee Input

Entrepreneurial public agencies work to identify and remedy the gaps between actual and potential program or service outcomes, and will actively seek out opportunities for internal improvement when possible (Klein et al., 2010). One such opportunity lies with the agency’s existing workforce. Street-level, supervisory, and managerial employees of public agencies, at any level of government, experience daily the effects their agencies’ programs and services have on clientele (Goodsell, 2003, 2015). Such employees become experts in their job functions, often implementing in real life the programs that have been abstractly conceived of in offices far removed from the reali- ties of daily organizational operations (J. Q. Wilson, 1991). They know what works and what does not, which processes ought to be used and which should be avoided, and how to keep operations running smoothly in the face of financial or manpower challenges that constantly threaten to gum up the works (Goodsell, 2003, 2015; J. Q. Wilson, 1991).

By acknowledging the expertise of their rank-and-file employees via entreaties to provide input on the development of new organizational processes or programs, entre- preneurial public agencies empower their employees and increase the likelihood that they will contribute to agency innovation (Y. Kim, 2010b). Light (1998) argues that such organizations assume that good ideas can come from anyone and will “invite ideas upward” (p. 113). There is nothing more important to such an organization, according to Light, than demonstration that it welcomes employee input. Consistent with this argument, in studies of public agencies that have received national recogni- tion for their innovativeness, Borins (2001, 2014) finds that a substantial number of innovations emanate from lower organizational levels. Borins asserts that employees

Clark 345

are ready and willing to offer input; they merely need to be asked for their suggestions. While that position may somewhat overstate reality, the agency’s invitation to provide input on program or process development tends to send the message that employee input benefits the organization, and the agency values the opinions and contributions of its employees. When workers feel their voices are being heard by those within the organization, they are likely to believe their opinions are important to the agency, and they will reciprocate by taking a greater personal stake in the organization’s success (de Jong & Den Hartog, 2007; Light, 1998). Therefore, it is hypothesized as follows:

Hypothesis 3: Organizational solicitation of employee ideas will be positively related to employee risk tolerance.

Data and Method

Sample

The present study examines the relationship between employee risk tolerance and organizational support for innovation via a cross-sectional analysis of quantitative data from the National Administrative Studies Project–Decision-Making (NASP-DM) data set. Researchers at the NASP at the University of Georgia randomly selected a sample pool of 8,000 public employees from Nevada and Indiana (6,000 and 2,000, respec- tively) and queried them in a multi-stage, web-based invitation and survey administra- tion process during the period of October to December 2012. In total, 833 individuals (618 from Nevada and 215 from Indiana) completed or nearly completed the survey, resulting in a total response rate of 21.49%. After listwise deletion of cases with miss- ing data, the final sample size for the present study is 712.

The survey queried employees of myriad state agencies and departments, with sizable portions of the sample coming from Health and Human Services (41.65% of Nevada sample), Child Services (35.25% of Indiana sample), Family and Social Services (12% of Indiana sample), and Transportation Services (16.25% of Nevada and 10% of Indiana samples). Because, the data have been completely de-identi- fied, each individual respondent’s agency or department is unknown. However, because NASP researchers wished to collect perspectives on decision making, they intentionally oversampled managers in Indiana, where such titles were pub- licly available; managerial titles were not available for the Nevada subsample. Pretests of the survey instrument indicated that managers from both states were twice as likely to respond to the instrument as were non-managers; thus, findings from the study may be less representative of rank-and-file workers than they are of managers. Furthermore, although respondents’ demographic characteristics, such as gender, age, and career tenure, are consistent with those found in earlier phases of the NASP that studied state-level managers and staff (see, for example, Moynihan & Pandey, 2007), like any study based on a particular group, general- izations based on the findings ought to be made with a degree caution (Moynihan & Pandey, 2007).

346 Public Personnel Management 45(4)

Measures

To examine the relationship between employee risk orientation and agency support for entrepreneurship, the study does not utilize empirical measures of risk taking or entre- preneurial agency behaviors. Rather, the dependent variable (DV) and primary inde- pendent variables (IVs) are perceptions reported by respondents about top management, organizational practices, and other employees. Moon (1999) argues that perceptual measures are appropriate when they are logically linked with reality within the agency, and capture attitudes and characteristics that may be compared across organizations. Bozeman and Kingsley (1998) defend reports of individuals’ perceptions as valid for measuring risk taking and risk tolerance, as individuals who perceive higher tolerance for risk taking among organizational superiors and other employees are likely to believe that risk taking on the job is acceptable and less likely to be punished by the agency. Employee perceptions effectively define the environmental reality of the agency and create a sense of organizational culture, as employee “perceptions provide cues to acceptable behavior” (Bozeman & Kingsley 1998, p. 111). From this perspec- tive, respondents’ perceptions of managerial and organizational behaviors are expected to reflect whether agencies that support entrepreneurship also encourage higher levels of employee risk tolerance.

Dependent variable. The dependent variable, employee risk tolerance, is measured as the response to the survey item, “Employees here are not afraid to take risks,” and reports respondents’ perceptions about fellow employees’ risk tolerance, rather than their personal tolerance levels. Prior studies (Bozeman & Kingsley, 1998; Y. Kim, 2010a) have utilized similar measures for the purpose of evaluating employee percep- tions of risk-taking propensity among public sector workers. In the data, the survey item is measured on a 5-point scale (strongly disagree, disagree, neither agree nor disagree, agree, strongly agree), which measures two discrete characteristics: (a) the nature of the agreement, and (b) the magnitude of the agreement. The dependent vari- able, employee risk tolerance, has been collapsed to a 3-point categorical variable (disagree/strongly disagree, neither agree nor disagree, agree/strongly agree), which means that the analysis focuses on the nature of respondents’ agreement with the sur- vey item. The use of a collapsed, 3-point dependent variable is appropriate, given that it groups categories logically and accounts for respondents’ willingness to commit to either agreeing or disagreeing with the survey item, while also including space for those who could commit to neither.

The DV is ordinal in nature, and it is advisable to avoid the assumption that the distance between each response category of the DV is equal (Long & Freese, 2014). Therefore, ordered logistic regression models are used to analyze the relationship between the DV and the various IVs. Preliminary analyses indicated that the DV in its original 5-point form caused a violation of the parallel lines assumption for the ordered logistic regression model because too few cases fell into the strongly agree category to allow for proper model fit. The use of the collapsed, 3-point form of the DV, therefore, finds additional support.

Clark 347

Independent variables. The first key independent variable, top management’s risk toler- ance, is measured by the extent of respondents’ agreement with the statement, “Top management here is not afraid to take risks.” The item is measured on a 5-point scale (strongly disagree, disagree, neither agree nor disagree, agree, strongly agree). Boze- man and Kingsley (1998) previously utilized a similar measure as an element of their organizational risk culture construct, and argued that top management plays a signifi- cant role in establishing and maintaining organizational culture. The behaviors and preferences of top management send messages to employees about appropriate work efforts (de Jong & Den Hartog, 2007) and the acceptability of employee behaviors (Sitkin & Pablo, 1992), such as risk taking, in the course of their daily work.

The second key independent variable, employee perceptions of reward for creativ- ity and innovation, is measured by the extent of respondents’ agreement with the fol- lowing survey item: “Over the past year, this organization has rewarded employees explicitly for creativity and innovation.” Like the first IV, this second IV is measured on a 5-point scale (strongly disagree, disagree, neither agree nor disagree, agree, strongly agree). As previously noted, creativity and innovation are typically perceived as distinct constructs, but for the purposes of this study are captured by the same mea- sure, given the survey question wording.

The third key independent variable, solicitation of employee ideas, is measured by the extent to which respondents agree with the survey item, “Over the past year, this organization has solicited employee ideas for new programs and processes.” As with the other key IVs, this variable is measured on a 5-point scale (strongly disagree, dis- agree, neither agree nor disagree, agree, strongly agree). Prior measures of employee participation in decision making on the job include respondents’ reports of opportuni- ties to provide input on programmatic and budgetary decision-making processes (Y. Kim, 2010b). The measure in the present study assesses whether agencies solicit employee input and does not specify the extent to which that input affects the deci- sions being made relative to key organizational goals. It is expected that all three key IVs will be positively related to respondents’ perceptions of employee risk tolerance on the job.

Control variables. Prior studies posit that individuals’ personal characteristics have influence over their perceptions of and sensitivity to risks in the work environment. For instance, Chen and Bozeman (2012) argue that individuals’ personal characteris- tics influence their perceptions of risk in the work environment. Because individual differences may bias their perceptions, the authors contend that it is important to con- trol for respondents’ personal characteristics. Prior studies have found consistently higher risk tolerance among men (Croson & Gneezy, 2009; Eckel & Grossman, 2008; Roszkowski & Grable, 2009), younger people (Harbaugh, Krause, & Vesterlund, 2002; Roszkowski & Grable, 2009), more educated people (Dong, 2014), and unmar- ried individuals (Jianakoplos & Bernasek, 1998). Therefore, control variables are included to account for respondents’ sex, age (in years), education attainment (high school, some college, 4-year degree, or more than a 4-year degree), and marital status (dummy variable). When using a similar set of demographic control variables as used

348 Public Personnel Management 45(4)

here, Chen and Bozeman (2012) find significant relationships between perceptions of employees’ risk aversion and respondents’ age, marital status, and gender. In addition, the model includes a 6-point ordinal variable for personal risk tolerance, which reflects respondents’ performance on hypothetical income risk gambles as utilized by Kimball, Sahm, and Shapiro (2008) and Dong (2014).

Respondents’ job-related characteristics have also been controlled for in the models. Because survey respondents reside in two different states—Nevada and Indiana—a dummy variable for state has been included in an effort to control for state-specific artifacts of employment system and culture. Specifically, the pres- ence of a more traditional merit system in Nevada coupled with inherently risk- laden legalized gambling in the state may produce risk perceptions that differ from those in Indiana, where state employment is at-will and gambling is illegal. In addition, variables measuring job-related characteristics that may influence one’s perceptions or lead to altered levels of risk tolerance are included. Tenure in the organization (in years), supervisory status (dummy variable), and overall job satis- faction (strongly disagree, disagree, neither agree nor disagree, agree, strongly agree that respondent is satisfied with his or her job) have therefore been included in the analysis.

Findings

Descriptive Statistics

Table 1 provides a summary of the survey questions utilized in the present study and their descriptive statistics. As to the dependent and main independent vari- ables, nearly half of all respondents (49.44%) indicate that employees in their agency are afraid to take risks, while less than one in five (18.54%) perceive employees in the agency to be risk tolerant. Top management is viewed as slightly more risk tolerant than employees, with approximately 30% of respondents agree- ing or strongly agreeing that those in the highest organizational positions are not afraid to take risks. Conversely, more than 41% perceive top managers to be risk intolerant. In the year preceding the survey, approximately 29% of respondents report that their agencies rewarded creative or innovative efforts, while 45% respond that their agencies did not do so. However, agencies appear to have been more active in soliciting employee ideas for program and process alterations, as nearly 58% of respondents agree or strongly agree that their agencies had requested employee input in program or process development during the previous year, while 27% report that their agencies did not solicit their input in program or pro- cess development.

As to the demographic variables, the sample is evenly split between men and women, nearly three quarters of the sample are married or living with a domestic part- ner, and respondents have earned, on average, a 4-year college degree. In general, respondents are moderately risk averse as it relates to their personal income, half have supervisory responsibilities in their agencies, and the majority (62.64%) are satisfied with their jobs.

Clark 349

Table 2 presents the Spearman correlation analysis for the variables used in the regression models. Results of the correlation analysis, including strength and direction

Table 1. Descriptive Statistics (n = 712).

Dependent variable % D/SD % A/SA M SD Minimum Maximum

Employee risk tolerance

49.44 18.54 1.691 0.765 1 3

Independent variables % SD % D % A % SA M SD Minimum Maximum

Managerial risk tolerance

13.97 27.94 26.08 4.33 2.788 1.109 1 5

Reward for creativity/ innovation

18.05 28.43 23.61 5.32 2.697 1.168 1 5

Solicitation of employee ideas

8.88 18.00 43.65 14.18 3.363 1.187 1 5

Control variables % SD % D % A % SA M SD Minimum Maximum

Sex 0.500 0.500 0 1 Marital status

(1 = married) 0.744 0.437 0 1

State (0 = Nevada; 1 = Indiana)

0.246 0.431 0 1

Supervisory status (1 = supervisor)

0.501 0.500 0 1

Education attainment (1 = high school; 2 = some college; 3 = 4-year degree; more than 4-year degree)

2.898 0.882 1 4

Personal risk tolerance (% of annual income risked in 50/50 gamble: 1 = none; 2 = 10%; 3 = 20%; 4 = 33%; 5 = 50%; 6 = 75%)

2.350 1.118 1 6

Age (in years) 49.37 10.25 24 77 Organizational

tenure (in years) 11.466 8.763 0.5 44

Job satisfaction (1 = SD; 5 = SA)

4.97 14.29 49.19 13.54 3.510 1.045 1 5

Note. D = disagree; SD = strongly disagree; A = agree; SA = strongly agree.

350

T a b

le 2

. Sp

ea rm

an C

o rr

el at

io n

M at

ri x.

V ar

ia bl

e na

m e

1 2

3 4

5 6

7 8

9 10

11 12

13

1 .

Em pl

o ye

e ri

sk t

o le

ra nc

e 1.

00 0

2

. M

an ag

er ia

l r is

k to

le ra

nc e

.5 19

** 1.

00 0

3

. R

ew ar

d fo

r cr

ea ti vi

ty /in

no va

ti o n

.3 75

** .4

26 **

1. 00

0

4 .

So lic

it at

io n

o f em

pl o ye

e id

ea s

.3 47

** .4

06 **

.5 77

** 1.

00 0

5

. Se

x (1

= m

al e)

.0 08

− .0

57 .0

24 −

.0 77

1. 00

0

6 .

A ge

− .0

13 .0

42 −

.0 23

.0 00

.0 58

1. 00

0

7 .

Ed uc

at io

n at

ta in

m en

t .0

17 −

.0 34

− .0

19 −

.0 01

− .0

21 .0

06 1.

00 0

8

. M

ar it al

s ta

tu s

(1 =

m ar

ri ed

) .0

19 .0

09 .0

39 .0

12 .1

35 **

.0 41

− .0

91 1.

00 0

9

. Pe

rs o na

l i nc

o m

e ri

sk −

.0 98

** −

.1 00

** −

.1 06

** −

.1 00

** .1

92 **

.0 63

.1 92

** −

.0 04

1. 00

0

10 .

St at

e (1

= I nd

ia na

) .0

04 .0

94 .1

66 **

.0 24

− .0

49 −

.1 06

** .0

86 −

.0 32

− .0

28 1.

00 0

11

. O

rg an

iz at

io na

l t en

ur e

− .0

14 .0

06 .0

45 .0

20 .1

77 **

.3 14

** −

.2 25

** .0

44 −

.0 32

− .0

22 1.

00 0

12

. Su

pe rv

is o ry

s ta

tu s

(1 =

s up

er vi

so r)

.0 64

.1 05

** .1

36 **

.1 77

** .1

24 **

.1 00

** −

.0 09

.1 13

** .0

65 −

.0 46

.2 77

** 1.

00 0

13

. Jo

b sa

ti sf

ac ti o n

.2 51

** .3

08 **

.3 75

** .3

73 **

− .0

87 .0

06 −

.0 03

− .0

06 −

.1 11

** .0

22 .0

49 .1

11 **

1. 00

0

** Si

gn ifi

ca nt

a t

p <

.0 1

le ve

l.

Clark 351

of the relationships between the DV and key IVs, are consistent with SET. That is, the significant, positive correlations between the DV and main IVs suggest that there exists a reciprocal relationship between organizational practices and perceptions of employee risk tolerance in public agencies. However, prior to drawing substantive conclusions based on the correlation results, regression analysis will determine the relative effect of the three independent variables on employee risk tolerance.

Regression Results

Results of the primary ordered logistic regression model are reported in Table 3. The likelihood (chi-square) ratios for each of the models (Model 1: χ2 = 282.10; Model 2: χ2 = 266.51) indicate that they are, as a whole, statistically significant (p < .001) and demonstrate that the results are very unlikely due to chance. Table 3 reports the odds ratios for the variables in each model and permits interpretation of the relative impact each IV has on the DV. The odds ratio may be understood as the degree to which an independent variable increases or decreases the likelihood of an employee’s report of perceived risk tolerance among his or her peers. An odds ratio of less than1 indicates that it is less likely that a respondent will report higher risk tolerance among fellow employees, while an odds ratio of greater than 1 indicates that it is more likely that a respondent will report higher risk tolerance among other employees.

Models 1 and 2 report results of ordered logistic regression and test the concurrent relationship between perceived employee risk tolerance and the three main independent

Table 3. Ordered Logistic Regression Results.

Independent variables Model 1 Model 2

Managerial risk tolerance 2.554*** [2.167, 3.012] 2.597*** [2.171, 3.106] Reward for creativity/innovation 1.292*** [1.107, 1.508] 1.338*** [1.125, 1.593] Solicitation of employee ideas 1.266** [1.078, 1.488] 1.238* [1.036, 1.500] Sex (1 = male) 1.245 [0.904, 1.714] Age 0.994 [0.978, 1.010] Education attainment 1.11 [0.923, 1.334] Marital status (1 = married) 1.055 [0.737, 1.510] Personal income risk tolerance 0.92 [0.792, 1.068] State (1 = Indiana) 0.656* [0.451, 0.955] Organizational tenure 0.995 [0.976, 1.016] Supervisory status (1 = supervisor) 0.872 [0.628, 1.211] Job satisfaction 1.125 [0.950, 1.333]

n 712 712 χ2 282.10*** 266.51*** Pseudo R2 .171 .183

Note. Listwise deletion of cases with missing values. Odds ratios with confidence intervals in brackets. The DV is employee risk tolerance. DV = dependent variable. *p < .05. **p < .01. ***p < .001.

352 Public Personnel Management 45(4)

variables. In Model 1, all three key independent variables possess strong, significant, positive relationships to the dependent variable, and Model 1 has a pseudo r-squared value of .171. After incorporating control variables in Model 2, the coefficient of each key IV is similar in size and significance to that found in Model 1, and the pseudo r-squared value increases in Model 2 to .183. According to the Model 2, respondents who report that they perceive higher risk tolerance among top management are 2.6 times more likely to report perceptions of higher risk tolerance among agency employ- ees (p < .001). Likewise, those who report that their organizations have rewarded cre- ativity and innovation and solicited employee input over the past year are 1.34 (p < .001) and 1.24 (p < .05) times more likely, respectively, to report that they perceive higher risk tolerance among their fellow employees.

Synthesis of the descriptive and inferential findings provides additional insight into the relationships between the DV and key IVs. The descriptive results indicate that, from respondents’ perspectives, on the whole, their managers do not demonstrate a willingness to take risks, nor do the majority of respondents’ agencies appear to have rewarded innovation and creativity in the previous year. Such perceptions suggest that respondents’ agencies are decidedly not risk-seeking in nature, and cast doubt on their entrepreneurial attitude. However, as demonstrated by the regression results, increases in perception of managerial risk tolerance or agency reward for innovation or creativ- ity are accompanied by an increase in the likelihood that respondents will also report their perception of greater risk tolerance among agency employees. That is, as respon- dents perceptions of risk-tolerant agency behaviors increases, so too do the odds that respondents will perceive risk tolerance in their fellow workers.

In addition to the findings on the primary IVs, one control variable, respondent’s employment state, is statistically significant in Model 2. Respondents from Nevada are 52.3% more likely to report higher employee risk tolerance than are Indiana employ- ees (p < .05). Despite prior findings that individual characteristics likely affect percep- tions of others’ risk tolerance (Chen & Bozeman, 2012), no other control variables are significant in the present analysis. To determine whether the models meet the parallel lines assumption for ordered logistic regression, a Brant test was performed on each (Long & Freese, 2014). Brant test results indicate no violation of the parallel lines assumption for the ordered logistic regression models.

Discussion and Conclusion

The strong, significant, positive relationships between perceptions of employee risk toler- ance (DV) and indicators of organizational risk tolerance (main IVs) conform to expecta- tions of reciprocity described by SET. The study confirms a reciprocal, or at least concurrent, relationship between perceptions of risk-tolerant agency behaviors, such as the emphasis on innovation and a demonstrated interest in employee input, and perceptions of risk tolerance among agency employees. In general, the study’s findings suggest that, at a minimum, the presence of risk-tolerant organizational practices and preferences will coincide with reports of higher risk tolerance among state agency employees. Where top managers indicate their own risk tolerance and public agencies have implemented risk-tolerant practices, such as

Clark 353

rewarding innovation and creativity on the job and requesting employee input when devel- oping new programs or processes, employee risk tolerance levels likely increase.

In the present study, the effect of respondents’ perceptions of top managers’ risk tolerance on perceived employee risk tolerance is quite large. This finding conforms to and reinforces the positions espoused by Schein (2010), Eisenberger and Stinglhamber (2011), and Rainey (2009) in relation to top managers’ sizable role in defining and promoting organizational culture. These results suggest that when top managers, who are more likely to describe themselves as risk tolerant and entrepreneurial than are lower level managers (Llewellyn et al., 2007), demonstrate their preference for risk- taking behavior to employees, employees may respond with a higher tolerance for risk. This may prove important for public agencies that frequently become mired under their own weight and inflexible to the ever-changing world around them, as they may find value in strategic initiatives designed to hire and promote executives with entrepreneurial attitudes, and particularly higher risk tolerance. These efforts may include recruitment of managers from outside the public sector to infuse more enter- prising approaches to developing and running public service delivery programs, as individuals in both the private and nonprofit sectors have been found to be more risk tolerant than their public sector counterparts (Bellante & Link, 1981; Chen & Bozeman, 2012; Dong, 2014).

In addition to top managers’ demonstration of their own risk tolerance, two organi- zational practices have been identified by this study as having a positive relationship to perceptions of employee risk tolerance. A public agency’s reward for creativity and innovation accomplishes a number of items, including indicating that it values its employees’ contributions (Gaertner & Nollen, 1989; Rhoades & Eisenberger, 2002) and supporting ongoing creative behavior (Dewett, 2004). Based on the findings of the present study, we may add a new item to that list, as the exchange of creative or inno- vative activities at work for organizational reward produces an increase in the likeli- hood that employees will tolerate risk on the job. Public employees’ presumed public service motivation (Bright, 2009; Perry et al., 2010) does not crowd out their incentive to take risks in the name of innovation and creativity.

The agency’s invitation for employees to provide input on program or process development serves as a second practice that may be used by the agency to signal its own risk tolerance, and thus encourage increased risk tolerance among public ser- vants. This finding is consistent with SET, as the agency’s recognition that ideas can “bubble up” from anywhere in the hierarchy (Borins, 2001) encourages input from employees, and employees respond by offering solutions that might have otherwise been withheld due to timidity, conservatism, and general risk aversion. Increased risk tolerance among public employees resulting from invitations to provide input on agency programs may result in more than just the development of a more entrepre- neurial orientation. Because the social exchange that occurs between agency and employee demonstrates their value to the organization and engenders trust in the agency (Eisenberger et al., 2001; Eisenberger & Stinglhamber, 2011; Neves & Eisenberger, 2014), they may be more confident in expressing their own views and also more resistant to pressures to engage in conformism and groupthink.

354 Public Personnel Management 45(4)

Regression results indicate that both organizational practices for enhancing employee risk tolerance—reward for creativity and innovation and solicitation of employee ideas—operate independent of managerial risk tolerance as a means to diminishing employees’ fear of risk taking. Thus, even in the absence of managerial behavior that suggests that risk taking is permitted or appropriate in the organizational setting, other practices, such as providing some form of remuneration for innovative activities or encouraging employee input in agency decisions, may be utilized to encourage higher levels of risk tolerance among employees. The increasing pace of social change and the amplified demand it places on the public sector suggests that public agencies, officials, and service providers will discourage creative, innovative solutions to public policy problems at their peril. Public sector employers must empower their employees to take risks, either by speaking up to propose creative solu- tions or by rewarding the development and implementation of those innovative ideas, when attempting to address work process or program initiatives. For the agency priori- tizes entrepreneurship and innovative approaches to public service delivery, the dem- onstration of organizational risk tolerance may go a long way to onboard employees in these pursuits.

Limitations of the Study

The present study is not without its limitations, and its findings should be interpreted with a degree of caution. The primary challenge stems from the use of a single, percep- tion-based data set as the source for both dependent and independent variables utilized in the regression model. The lack of empirical measures of risk-taking behavior in the public sector, as well as the general dearth of data about the risk perceptions of public employees, necessitates the use of available data. Use of measures from a single sur- vey for both dependent and independent variables could introduce common source bias and threaten the veracity of the findings, as it creates potential for correlated errors among the independent and dependent variables and may result in significant inflation or deflation of self-reported relationships (Meier & O’Toole, 2013). Unfortunately, short of using independent sources of data for a study’s variable set, no viable solution exists to deal with the problem of common source bias (Favero & Bullock, 2015), and scholars have yet to coalesce around any particular post hoc sta- tistical tests for its presence or effects (Conway & Lance, 2010; Meier & O’Toole, 2013).

Although concerns regarding inflation of relationships due to single survey mea- surement error are worthy of scholars’ attention, they do not, in and of themselves, invalidate a study’s findings (Conway & Lance, 2010). In the present study, reports of personal observations are an especially valuable, rather than costly, aspect of the data analysis because they report the perceptions that define the respondents’ organiza- tional and cultural realities (Bozeman & Kingsley, 1998)—aspects that would not be ascertainable otherwise, as empirical measures of risk culture in public agencies are not presently available. Furthermore, as noted by Meier and O’Toole (2013), survey questions about observable behaviors in others, rather than self-reports of one’s own

Clark 355

behaviors or attitudes, appear to be less affected by common source bias than other item types. Thus, it is possible that NASP-DM survey items that query respondents about the attitudes and performance within their agency that they have personally witnessed may suffer less severely from common source bias, as they report respon- dents’ direct observations of organizational and employee behaviors. Importantly, the findings do confirm the nature and direction of the relationships proposed by extant literature on public entrepreneurship and explained via SET. If nothing else, the poten- tial for common source bias ought to inspire public administration scholars to pursue stronger, observable metrics of public organization performance (Meier & O’Toole, 2013) and risk propensities of their managers and employees.

Also of concern is the inclusion of single-item variables, rather than multi-item constructs, as measures of the dependent and primary independent variables. Although multi-item constructs can offer improvements to validity and depth to a phenomenon under study, single-item measures can offer significant insight into the concept under study and may be acceptable for use. For instance, Ganzach and colleagues’ (2008) examination of single- versus multi-item measures of risk perception reveals that a single-item measure of risk perception better measures risk perception than prior, multi-item scales. Furthermore, the authors find that the risk judgment of laypeople who do not research risk and its measures, such as those surveyed by the NASP-DM instrument, may correspond as well as or better to existing theories of risk taking than do those of experts in risk assessment. Thus, individual respondents’ understanding of risk in the survey instrument, specifically in the measures of employee and managerial risk tolerance, likely reflects a position that is consistent with risk perception theory and corresponds well to the conditions they perceive in their environments. Furthermore, the main IVs for reward and solicitation of employee ideas report respon- dents’ perceptions on the frequency of agency activities and have been included at face value rather than as part of some larger construct. Conclusions about the relationship between these variables relative and the DV have been narrowly tailored to reflect the limitations of the IVs’ single-item construction.

Due to the potential bias presented by the use of a single survey for all variables, as well as the limitations imposed by the single-item constructs mentioned above and the cross-sectional nature of the analysis, this study makes no causal claims about the relationships discussed herein. Rather, the analysis suggests that causal mechanisms for risk tolerance may be present, but recommends further research prior to making such a claim.

Future Research

This study represents a small but important step toward understanding the complexi- ties of public sector employees’ risk tolerance and public entrepreneurship. It relies on perceptual measures and does not include measures of actual occurrences of risk-tak- ing behavior. Future research may use this study as a starting point for developing empirical, observable measures of both employee and organizational risk tolerance. In addition to developing new measures, future work may also compare particular types

356 Public Personnel Management 45(4)

or sizes of public agencies to determine whether all are uniformly subject to the reci- procity implicated among state agency employees in this sample. Based upon their diverse missions and client orientations, different types of public organizations, for example, human service, financial, or police agencies, may possess very different atti- tudes toward workplace risk tolerance. Alternatively, different types of agencies may experience reciprocal exchange, or emphasize reward or the provision of employee input to varying degrees. These features may also vary depending on the level of gov- ernment, where smaller, local governments may experience differences in risk toler- ance, entrepreneurial behaviors, and reciprocal exchange between the organization and its employees.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Bellante, D., & Link, A. N. (1981). Are public sector workers more risk averse than private sec- tor workers? Industrial & Labor Relations Review, 34, 408-412.

Bellone, C. J., & Goerl, G. F. (1992). Reconciling public entrepreneurship and democracy. Public Administration Review, 52, 130-134.

Blau, P. M. (1964). Exchange and power in social life. New York, NY: John Wiley. Bloch, C., & Bugge, M. M. (2013). Public sector innovation—From theory to measurement.

Structural Change and Economic Dynamics, 27, 133-145. Borins, S. (1998). Innovating with integrity: How local heroes are transforming American gov-

ernment. Washington, DC: Georgetown University Press. Borins, S. (2000). Loose cannons and rule breakers, or enterprising leaders? Some evidence

about innovative public managers. Public Administration Review, 60, 498-507. Borins, S. (2001). The challenge of innovating in government. Arlington, VA: The Price

Waterhouse Coopers Endowment for the Business of Government. Borins, S. (2014). The persistence of innovation in government. Washington, DC: Brookings. Bozeman, B. (2007). Public values and public interest: Counterbalancing economic individual-

ism. Washington, DC: Georgetown University Press. Bozeman, B., & Kingsley, G. (1998). Risk culture in public and private organizations. Public

Administration Review, 58, 109-118. Bright, L. (2009). Why do public employees desire intrinsic nonmonetary opportunities? Public

Personnel Management, 38(3), 15-37. Burton, R. A. (2008). On being certain: Believing you are right, even when you’re not. New

York, NY: St. Martin’s Griffin. Caruana, A., Ewing, M. T., & Ramaseshan, B. (2002). Effects of some environmental chal-

lenges and centralization on the entrepreneurial orientation and performance of public sec- tor entities. Service Industries Journal, 22(2), 43-58.

Clark 357

Chen, C., & Bozeman, B. (2012). Organizational risk aversion: Comparing the public and non- profit sectors. Public Management Review, 14, 377-402.

Conway, J. M., & Lance, C. E. (2010). What reviewers should expect from authors regarding common method bias in organizational research. Journal of Business Psychology, 25, 325- 334.

Cropanzano, R., & Mitchell, M. S. (2005). Social exchange theory: An interdisciplinary review. Journal of Management, 31, 874-900.

Croson, R., & Gneezy, U. (2009). Gender differences in preferences. Journal of Economic Literature, 47, 448-474.

Damanpour, F., & Schneider, M. (2009). Characteristics of innovation and innovation adoption in public organizations: Assessing the role of managers. Journal of Public Administration Research and Theory, 19, 495-522.

de Jong, J. P. J., & Den Hartog, D. N. (2007). How leaders influence employees’ innovative behaviour. European Journal of Innovation Management, 10, 41-64.

Dewett, T. (2004). Employee creativity and the role of risk. European Journal of Innovation Management, 7, 257-266.

Dong, H. D. (2014). Individual risk preference and sector choice: Are risk-averse individuals more likely to choose careers in the public sector? Administration & Society. Advance online publication. doi:10.1177/0095399714556500

Drennan, L. T., McConnell, A., & Stark, A. (2015). Risk and crisis management in the public sector (2nd ed.). New York, NY: Routledge.

Eckel, C. C., & Grossman, P. J. (2008). Men, women and risk aversion: Experimental evidence. Handbook of Experimental Economics, 1, 1061-1073.

Eckerd, A. (2014). Risk management and risk avoidance in agency decision making. Public Administration Review, 74, 616-629.

Eisenberger, R., Armeli, S., Rexwinkel, B., Lynch, P. D., & Rhoades, L. (2001). Reciprocation of perceived organizational support. Journal of Applied Psychology, 86, 42-51.

Eisenberger, R., & Stinglhamber, F. (2011). Perceived organizational support: Fostering enthu- siastic and productive employees. Washington, DC: American Psychological Association.

Emerson, R. M. (1976). Social exchange theory. Annual Review of Sociology, 2, 335-362. Favero, N., & Bullock, J. B. (2015). How (not) to solve the problem: An evaluation of scholarly

responses to common source bias. Journal of Public Administration Research and Theory, 25, 285-308.

Gaertner, K. N., & Nollen, S. D. (1989). Career experiences, perceptions of employment prac- tices, and psychological commitment to the organization. Human Relations, 42, 975-991.

Ganzach, Y., Ellis, S., Pazy, A., & Ricci-Siag, T. (2008). On the perception and operationaliza- tion of risk perception. Judgment and Decision Making, 3, 317-324.

Gneezy, U., & Rustichini, A. (2000). Pay enough or don’t pay at all. Quarterly Journal of Economics, 115, 791-810.

Goodsell, C. T. (2003). The case for bureaucracy: A public administration polemic (4th ed.). Washington, DC: CQ Press.

Goodsell, C. T. (2015). The new case for bureaucracy. Los Angeles, CA: SAGE. Gould-Williams, J., & Davies, F. (2005). Using social exchange theory to predict the effects of

HRM practice on employee outcomes. Public Management Review, 7(1), 1-24. Gow, J. I. (2014). Public sector innovation theory revisited. The Innovation Journal: The Public

Sector Innovation Journal, 19(2), 1-15.

358 Public Personnel Management 45(4)

Harbaugh, W. T., Krause, K., & Vesterlund, L. (2002). Risk attitudes of children and adults: Choices over small and large probability gains and losses. Experimental Economics, 5, 53-84.

Jianakoplos, N. A., & Bernasek, A. (1998). Are women more risk averse? Economic Inquiry, 36, 620-630.

Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47, 263-292.

Kim, S., & Yoon, G. (2015). An innovation-driven culture in local government: Do senior man- ager’s transformational leadership and the climate for creativity matter? Public Personnel Management, 44, 147-168.

Kim, Y. (2010a). Improving performance in U.S. state governments. Public Performance & Management Review, 34, 104-129.

Kim, Y. (2010b). Stimulating entrepreneurial practices in the public sector: The roles of organi- zational characteristics. Administration & Society, 42, 780-814.

Kimball, M. S., Sahm, C. R., & Shapiro, M. D. (2008). Imputing risk tolerance from survey responses. Journal of the American Statistical Association, 103, 1023-1038.

Klein, P. G., Mahoney, J. T., McGahan, A. M., & Pitelis, C. N. (2010). Toward a theory of public entrepreneurship. European Management Review, 7, 1-15.

Koch, P., & Hauknes, J. (2005). On innovation in the public sector—Today and beyond. Oslo, Norway: NIFU. Retrieved from http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1. 104.3988&rep=rep1&type=pdf

Light, P. C. (1998). Sustaining innovation: Creating nonprofit and government organizations that innovate naturally. San Francisco, CA: Jossey-Bass.

Llewellyn, N., & Jones, G. (2003). Controversies and conceptual development: Examining pub- lic entrepreneurship. Public Management Review, 5, 245-266.

Llewellyn, N., Lewis, P., & Woods, A. (2007). Public management and the expansion of an entrepreneurial ethos? Public Management Review, 9, 253-267.

Long, J.S., & Freese, J. (2014). Regression models for categorical dependent variables using Stata, 3rd ed. College Station, TX: Stata Press.

Lumpkin, G. T., & Dess, G. G. (1996). Clarifying the entrepreneurial orientation construct and linking it to performance. Academy of Management Review, 21, 135-172.

Lumpkin, G. T., & Dess, G. G. (2001). Linking two dimensions of entrepreneurial orientation to firm performance: The moderating role of environment and industry life cycle. Journal of Business Venturing, 16, 429-451.

March, J. G., & Shapira, Z. (1987). Managerial perspectives on risk and risk taking. Management Science, 33, 1404-1418.

Meier, K. J., & O’Toole, L. J. (2013). Subjective organizational performance and measurement error: Common source bias and spurious relationships. Journal of Public Administration Research and Theory, 23, 429-456.

Merritt, R., & Merritt, A. J. (Eds.). (1985). Innovation in the public sector. Beverley Hills, CA: SAGE.

Merton, R. K. (1940). Bureaucratic structure and personality. Social Forces, 18, 560-568. Mitchell, V. (1995). Organizational risk perception and reduction: A literature review. British

Journal of Management, 6, 115-133. Molina-Morales, F. X., Martínez-Fernández, M. T., & Torlò, V. J. (2011). The dark side of trust:

The benefits, costs and optimal levels of trust for innovation performance. Long Range Planning, 44, 118-133.

Clark 359

Molm, L. D., Takahashi, N., & Peterson, G. (2000). Risk and trust in social exchange: An exper- imental test of a classical proposition. American Journal of Sociology, 105, 1396-1427.

Moon, M. J. (1999). The pursuit of managerial entrepreneurship: Does organization matter? Public Administration Review, 59, 31-43.

Morris, M. H., & Jones, F. F. (1999). Entrepreneurship in established organizations: The case of the public sector. Entrepreneurship Theory and Practice, 24, 71-91.

Moynihan, D. P., & Pandey, S. K. (2007). The role of organizations in fostering public service motivation. Public Administration Review, 67, 40-53.

Neves, P., & Eisenberger, R. (2014). Perceived organizational support and risk taking. Journal of Managerial Psychology, 29, 187-205.

Perry, J. L., Hondeghem, A., & Wise, L. R. (2010). Revisiting the motivational bases of pub- lic service: Twenty years of research and an agenda for the future. Public Administration Review, 70, 681-690.

Rainey, H. G. (2009). Understanding and managing public organizations (4th ed.). San Francisco, CA: John Wiley.

Rhoades, L., & Eisenberger, R. (2002). Perceived organizational support: A review of the litera- ture. Journal of Applied Psychology, 87, 698-714.

Roszkowski, M. J., & Grable, J. E. (2009). Evidence of lower risk tolerance among public sector employees in their personal financial matters. Journal of Occupational and Organizational Psychology, 82, 453-463.

Schein, E. H. (2010). Organizational culture and leadership (4th ed.). San Francisco, CA: Jossey-Bass.

Simon, H. A. (1979). Rational decision making in business organizations. The American Economic Review, 69, 493-513.

Simon, H. A. (1997). Administrative behavior (4th ed.). New York, NY: Free Press. Sitkin, S. B., & Pablo, A. L. (1992). Reconceptualizing the determinants of risk behavior.

Academy of Management Review, 17(1), 9-38. Townsend, W. (2013). Innovation and the perception of risk in the public sector. The

International Journal of Organizational Innovation, 5(3), 21-34. Wayne, S. J., Shore, L. M., & Liden, R. C. (1997). Perceived organizational support and leader-

member exchange: A social exchange perspective. Academy of Management Journal, 40, 82-111.

Wildavsky, A., & Dake, K. (1990). Theories of risk perception: Who fears what and why? Daedalus. Journal of the American Academy of Arts and Sciences, 119(4), 41-60.

Wilson, D. T., & Mathews, H. L. (1971). Impact of management information systems upon purchasing decision-making. Journal of Purchasing, 7, 48-56.

Wilson, J. Q. (1991). Bureaucracy: What government agencies do and why they do it. New York, NY: Basic Books.

Windrum, P. (2008). Innovation and entrepreneurship in public services. In P. Windrum & P. Koch (Eds.), Innovation in public sector services: Entrepreneurship, creativity and man- agement (pp. 3-20). Northampton, MA: Edward Elgar.

Author Biography

Anna Fountain Clark, PhD, is Visiting Assistant Professor of Government at Texas Woman’s University. Dr. Clark’s research interests include organization theory and behavior, public and nonprofit management, and the application of technological and policy innovations in public and nonprofit organizations.