Entry Barriers and Nonprofit Founding Rates: An Examination of the Milwaukee Voucher School Population

Introduction

Although nonprofit entrepreneurship lacks a singular definition, a common characterization ties it closely to the formation and establishment of new nonprofit ventures (Bilodeau and Slivinski 1998; Cordes, Steuerle, and Twombly 2004; Haugh 2007). By this view, the pivotal nonprofit entrepreneurial event is the creation and operational launch of a nonprofit venture. The purpose of this paper is to examine how contextual conditions generate variations in the number of nonprofit foundings over time. Specifically, we investigate how institutional barriers to entry, primarily formal ones such as policies and regulations, have impacted the founding rate of nonprofit voucher schools in the City of Milwaukee between 1990 and 2015. Institutions, the formal and informal rules of the game, are a vital component in modern entrepreneurship research as intuitions are viewed as having a significant influence on the allocation of entrepreneurial activities (Minniti 2008).

Formal institutions play a dual role for nonprofit entrepreneurs. Some constrain nonprofit entrepreneurial activity by creating and raising various types of entry barriers. Others establish conditions that help create windows of opportunity, and thus encouraging nonprofit entrepreneurial activity, for example, by removing or lowering entry barriers. Milwaukee was the first city in America to create a publicly funded private school voucher program in 1990 (Witte 2000), and therefore offers a unique setting for investigating the relation between institutional entry barriers and founding rates of an entire population. To be clear, formal institutions are not the only factor explaining variations in nonprofit founding rates, but as posited by Aldrich (1990), “institutional forces” are most likely to have their greatest impact on founding rates when “a new form of organization is emerging” as in the case of the Milwaukee voucher school population.

Although this paper only looks at one population of nonprofits in a localized market, we believe it can contribute to, and expand, the existing nonprofit entrepreneurship literature in important ways. First, this research illuminates and reinforces the significant role played by institutional factors, in addition to the characteristics and behaviors of the individual nonprofit entrepreneur, in shaping nonprofit entrepreneurship. Even though, as observed by Frumkin (2002, 130), “[o]ne of the principal reasons the nonprofit and voluntary sector is becoming an attractive vehicle for entrepreneurship is that the barriers to entry is low,” nonprofit entrepreneurship research rarely studies population level phenomena, or how particular and/or changing institutional barriers impact entrepreneurial entry in the nonprofit sector. A better comprehension of the role and impact of such entry barriers is not just useful for nonprofit entrepreneurs, it is also valuable from a nonprofit entrepreneurship policy perspective. As highlighted by Harrison and Thornton (2014), policies are powerful tools for erecting (and/or removing) barriers to entry, and thereby having a direct impact on the supply of nonprofit entrepreneurs.

Second, our data covers all voucher schools in Milwaukee since the inception of the voucher program in 1990. Hence, rather than focusing on nonprofit entrepreneurship in the context of relatively stable and established populations this paper examines nonprofit entrepreneurship in a new and emerging population.

Third, not only does our data cover the entire Milwaukee voucher school population but also founding attempts of voucher schools in Milwaukee, which in turn allows us to analyze the actual founding rate of new nonprofit voucher schools. The founding rate can be depicted as the outcome of two underlying rates, each associated with a key stage in the start-up process of new organizations. The first is the rate at which attempts at founding are initiated; the second is the rate of success of those attempts (Delacroix and Carroll 1983). Thus, success in this second process is associated with the founding event itself and is to be distinguished from the success or failure of the organization once founded.

Because it is difficult to capture founding attempts, empirical nonprofit entrepreneurship research must typically rely on data of formally established nonprofits i. e. nonprofits that have been incorporated and/or received tax-exempt status. The problem with this approach is that these organizations represent a culmination of the operation of both processes mentioned above. In other words, this data excludes those nonprofit entrepreneurs exiting the entrepreneurial process before being formally established. Thus, by utilizing the actual founding rate this paper is able to overcome the selection bias resulting from including only nonprofit entrepreneurial efforts that actually resulted in a formal new nonprofit organization.

Background

In his seminal article, Baumol (1990) makes the argument that while the supply of entrepreneurial agents are more or less constant, it is the nature of the institutions in which these agents are embedded that end up dictating the outcomes of entrepreneurial actions. Hence, by putting in place and/or attempt to alter institutions, policy makers can play a significant role in influencing various elements related to entrepreneurship. This idea is supported by Minniti (2008), who suggests that institutions, and the policies that shape them, are essential determinants of entrepreneurial behavior as “[g]overnment policies mold institutional structures for entrepreneurial action, encouraging some activities and discouraging others.”

Nonprofit entrepreneurship scholars have indeed long emphasized the importance of institutional factors, such as policies and normative pressures, playing a potentially influential role for nonprofit entrepreneurial entry. (e. g., Young 1980; Schiff 1986; Bilodeau and Slivinski 1998; Cordes, Steuerle, and Twombly 2004; Steinberg 2006). In addition, one can find plenty of empirical literature looking into various aspects of nonprofit market entry (e. g., Corbin 1999; Saxton and Benson 2005; Harrison and Thornton 2014), yet relatively few inquiries have explicitly examined the relation between formal institutions and nonprofit entrepreneurship. One exception is Twombly (2003), who investigated the relation between the implementation of a federal welfare reform (the AFDC waiver program) and new nonprofit entry. He found the introduction of the reform initiative to be a highly significant determinant in explaining new nonprofit entry, and concluded that his study (p. 232) “[…] illustrates that policy initiatives such as welfare reform can substantially affect the composition of local human service sectors.” Another example is Stephan, Uhlaner and Stride’s (2015) interesting paper that examined the relationship between the institutional context and social enterprise start-ups (includes nonprofits as well as hybrid legal forms) in twenty-six countries. They find a positive relationship between government activism through the use of various forms of formal institutional support and social enterprise start-up activity, and conclude that “[o]ur findings on institutional configurations suggests that policymakers need to take formal and informal institutions into account when pondering policy decisions.”

The Milwaukee Parental Choice Program (MPCP)

The call for educational reform has often involved a significant institutional component. When Milton Friedman (1955) first formulated his idea of voucher schools as a way to improve education it was premised on creating an institutional framework that would stimulate and increase the supply of education entrepreneurs. Hess (2008) posits a stable and hospitable policy environment represents the foundation for establishing viable markets for educational entrepreneurship. Moreover, the focus of such policies must be to incentivize the creation of new schools (214): “In sector after sector, solving new problems–or more effectively addressing stubborn ones–has been the province of new entrants. Thus, the challenge is […] to summon new ventures to build on fresh turf.”

As the oldest and largest, and by some deemed most important (Epple, Romano, and Urquiola 2015), urban private school voucher program in the United States, the MPCP offers an outstanding case for exploring the role of institutions and nonprofit entrepreneurship. Not only is the Milwaukee voucher case unique in that it is well-established via its 26-year history, it has also been populated almost exclusively by non-profit organizations (Andersson and Ford 2015), and has generated large numbers of new entrants and large numbers of failed providers. Furthermore, there is growing recognition of the inconsistent performance of school choice programs, which has placed more emphasis on the need to study organizations providing alternative publicly funded educations as opposed to the mechanism by which a student ends up in a non-traditional school. In previous work, Ford and Andersson (2016) argue forcefully that the performance of hollow state activities built on social and/or nonprofit entrepreneurship is dependent on the strength or the organizations involved.

Since its start, the MPCP has been embedded in a range of formal institutional arrangements (Witte 2000). Over the life of the program the regulations governing MPCP schools evolved steadily. This policy evolution, detailed fully by Ford (2015b), is the foundation for our case analysis of the relationship between entry barriers and nonprofit entrepreneurship.

A sizable body of research has examined the impacts of the MPCP over its 26-year history (see Andersson and Ford 2016 for a review), but little research has targeted the relationship between new entrants and the evolving policy environment. To set the stage for our analysis, the next section outlines how a nonprofit entrepreneur would enter the Milwaukee education market.

How a School Participates in the MPCP

In this analysis, we exploit a unique program feature that allows us to capture data on attempted organizational entries into the MPCP, in addition to actual entries into the program i. e. successful attempts. Though the regulation of new schools has varied throughout the MPCP’s history, the first required step for joining the program has remained consistent. Schools seeking to participate in the program are, under state law, required to file an “Intent to Participate Form” with the Wisconsin DPI no later than February 1st in the school year prior to their participation in the MPCP (Pugh 2015). For example, if a school wished to open in Fall of 2017 they would need to file their form by February 1st 2017. The form requires potential school operators to share the following information: (i) the name of their school; (ii) the address of their school, if known; (iii) the name and contact information of the school administrator; (iv) the months in which the school will accept application; and (v) the number of available voucher seats by grade.

The inclusion of the data on attempts allows us to examine two crucial features of nonprofit entrepreneurship. First, the filing of the “Intent to Participate Form” represents one of the earliest formal actions taken by a MPCP entrepreneur. Although filing the form is no guarantee the school will actually open, or enroll voucher pupils, it visibly and distinctly demonstrates intent on the part of the entrepreneur. Intent indicates more than mere desire, so while many individuals may indeed yearn to create a new voucher school, many never take further action and move beyond merely playing with the idea of starting a new organization. As a consequence, entrepreneurship scholars consider intentionality to signify one of the key properties of emerging organizations (Katz and Gartner 1988) i. e., an activity that takes place before an organization becomes an organization during the so-called nascent of gestation stage (Reynolds and Miller 1992). Thus, the data on attempts permits us to say something about the fundamental, yet scarcely researched, nascent stage of nonprofit entrepreneurship. Second, by capturing the number of attempted organizational entries into the MPCP we can also calculate and investigate the founding rate of new voucher schools. As outlined in the introduction, the founding rate captures two entrepreneurial processes, the rate at which individuals initiate attempts to start new schools, and a second rate depicting the success of those attempts, in the sense that a voucher school actually gets opened. According to Carroll and Khessina (2005), empirical research capturing both rates remains quite rare, and entrepreneurship scholars often assume that the two rates are closely connected, suggesting the rates move together in a consistent manner. But as Hannan and Carroll (1992) have pointed out, from an ecological perspective it is just as plausible to imagine the influence of environmental factors, for example formal institutions, that might propel one rate in one direction while carrying opposing implications for the other rate.

Ultimately, being able to observe both attempts and successful attempts to start new voucher schools brings an intriguing dimension to this research. Having highlighted how a nonprofit entrepreneur will participate in the MPCP we now move to examine the regulatory environment surrounding voucher schools in Milwaukee. In previous work (Andersson and Ford 2014; Ford and Andersson 2016) we have argued and demonstrated the key role environmental factors plays in the determining entry as well as existence of Milwaukee’s voucher schools. In the subsequent section we turn our focus to variations over time in the founding rate of new voucher schools, and link these variations to the implementation of the various policy changes shaping the MPCP. By considering formal barriers to actually opening a new school, and how these barriers have evolved over the MPCP’s history, we seek to discover policy relevant factors pertaining to the influence of these barriers on attempts as well as successful attempts to found new voucher schools.

Data and Methodology

As stated, the purpose of this article is to use the Milwaukee school voucher case to illustrate the relationship between barriers to entry, institutional factors, and organizational founding rates. We present our analysis as a case study meant to illustrate the nexus between public policy and nonprofit entrepreneurship. A case study approach offers many advantages. First, because we have 25 years of population data, inferential statistics are not appropriate. We are not attempting to use a sample to make conclusions about a population, but rather present a comprehensive case to demonstrate how public policies influence nonprofit entrepreneurship (Gerring 2004). Second, Gerring (2004) concludes that a case study is appropriate when one case can offer theoretical and/or practical insights to other cases. The MPCP is the rare example of a mature private school voucher program whose lessons can be applied to younger school choice programs in other locales, as well as other public policy initiatives designed to improve public performance through the encouragement of nonprofit entrepreneurship. Third, the organizational churn present in the Milwaukee voucher case presents an opportunity to test the previously discussed theories of nonprofit entrepreneurship.

We present the Milwaukee case using data on the attempted and successful openings of new voucher schools during different MPCP regulatory eras. The authors obtained all data from the Wisconsin Department of Public Instruction (DPI), Wisconsin’s state education agency. Specifically, in spring of 2016, the authors requested copies of the annual DPI press release detailing the names and contact information for all schools that filed an Intent to Participate Form in each year of the MPCP’s history. The authors then created a panel dataset where each observation was a new intended participant in the MPCP. Overall, DPI’s releases indicate that between 1990 and 2015, 579 schools filed an initial Intent to Participate Form with the state of Wisconsin. However, not all of those schools were entrepreneurial efforts. The authors cross-checked all observations against previous data collected on MPCP schools (see: Anderson and Ford 2014) and determined that 128 of the 571 intended new MPCP participants were existing private schools (mostly Catholic and Lutheran). All existing schools were removed from the dataset.

Next, the authors used publicly available program enrollment data to determine which entrepreneurial attempts were successful. A successful attempt is defined as a school that filed an Intent to Participate Form and subsequently enrolled at least one MPCP pupil in the following school year. Overall, we find that 132 of the 443 entrepreneurial attempts, 29.8 percent, were successful. In addition, the data indicate that a sizable number of schools made multiple attempts; 43 schools filed forms twice, 12 filed forms three times, four filed forms four times, and three schools filed forms on five separate occasions. Importantly, schools that are successful at opening are not required to file an Intent to Participate form in subsequent school years. A majority of schools, 290, filed an Intent to Participate Form only once. Entrepreneurial attempts, successes, and success rates by year are displayed in Table 1, and Figures 1 and 2.

Figure 1:

Entrepreneurial attempts and openings.

Figure 2:

Entrepreneurial success rates.

Although school choice and school vouchers are often described as a free-market approach to education, the term free does not imply there is no role or need for formal institutions to play. In fact, as discussed by Ford (2015b), the MPCP has, since its beginning, been embedded in a web of formal institutions, although the scope and nature of these institutions have changed several times during the quarter century the program has been operational. The changing regulatory climate was a direct result of public policy decisions enacted in response to perceived deficiencies in the MPCP. As can be seen in Table 1 and Figures 1 and 2, the number of entrepreneurial attempts and successes also fluctuated over the course of the program’s history. In the following section we divide the MPCP into five distinct time periods, each of which represents a unique policy environment, and highlight the most significant institutional factors and/or changes present during each period. Finally, we demonstrate how the level of entrepreneurial activity fluctuated during each regulatory era in ways consistent with theory.

Entrepreneurship and the Changing MPCP Policy Environment

Conclusion and Discussion

The research presented in this paper represents one of few known attempts to trace and analyze the founding rates of a specific nonprofit population over time. As commented by Carroll and Kessina (2005), many scholars tend to rely on either the rate of initiation of attempts or the rate of success of those attempts but seldom examine both at the same time. By combining data on attempts to start new voucher schools, and data on the number of successful start-ups emanating from these attempts, we offer a rare glimpse into the actual founding rate in a nonprofit population.

The key question examined in this paper is how formal barriers to entry correlate with levels and changes in the foundation rate of new voucher schools in Milwaukee. We examined this relation and are able to reveal how formal institutions regulating entrepreneurial efforts impact both entrepreneurial attempts, and success rates. Not surprisingly, we find that regulation that removes or lowers barriers to entry led to more entrepreneurial attempts to form new voucher schools. For example, by removing the non-sectarian school requirement, more space for entrepreneurs wanting to open religious schools opened up, and subsequently led to an increase in entrepreneurial attempts. Likewise, we find that erecting of institutional barriers, for example in the form of a formal third party approval process, proved to have an impact on the founding success rate of new voucher schools.

The size of the barrier presented by the formal third party approval process is, in many ways, an indictment of the qualifications of many of the entrepreneurs drawn to the voucher program prior to 2010. The initial review board located at the Institute for the Transformation of Learning at Marquette University, an organization headed by prominent voucher advocate Howard Fuller, denied every start-up school during its first year of operation (Witte 2000; Borsuk 2009). Fuller told the Milwaukee Journal Sentinel, “No matter how difficult it is and no matter how much disagreement it brings, we must focus on quality … Dealing with barriers to entry is a very important part of that process” (Borsuk 2009). It was, in fact, concerns over school quality and need for barriers to entry that spurred a schism among long time voucher advocates in Milwaukee (Ford 2015b). Importantly, not much was known about the quality of voucher schools prior to the requirement for adding standardized tests, so the quality concerns were mostly anecdotal in nature.

In addition, we found that the creation and implementation of a modest fee for new schools dramatically decreased the level of attempted entrepreneurial activity, suggesting that pocket book issues can play a major role in determining the seriousness of entrepreneur intent. We were struck by the wide variation in the founding rate of voucher schools. Clearly, formal institutions are not the only factor impacting in the founding rate, but as noted in the introduction, institutions are considered as particularly pertinent when a new population is born and begins to develop. For example, although 2010 was a year when a considerable number of nonprofit entrepreneurs (53) attempted to create new voucher schools, not a single of these entrepreneurs were able to successfully launch a school. This can be compared to the year before when the success rate was approximately 40 percent. We can see few other explanations for this dramatic change in success rate than the erecting of entry barriers, such as the requirement obtain approval from an outside entity to start a school. It is important to remember that nonprofit voucher entrepreneurs, by seeking to enter a brand new industry, are different from those that can follow in the footsteps and tradition pioneered by numerous predecessors in the same industry. Thus, we agree with Aldrich’s (1990) assertion that as new populations emerge and evolve, institutions will be “constraining and imprinting the new form in distinctive ways.” Consequently, we suspect formal institutions are very much a relevant and potent source for explaining the, from time to time, notable changes in founding rates. Moreover, this research shows how formal institutions indeed can serve as powerful instruments for policy-makers to impact the allocation of nonprofit entrepreneurial action. While we have mainly looked at the creation and/or removal/lowering of entry barriers there are other ways for policy makers to influence this allocation process. For example, Stephan, Uhlaner, and Stride (2015) show how formal institutions sometimes take on an active and direct role by providing tangible (e. g., direct funding or subsidies) and intangible (e. g., administrative assistance) resource support, which end up affecting the supply of social entrepreneurs.

The demonstrated chilling effects of entry barriers and blunt nature of policy begs the question, are there ways to create and implement barriers to entry that prevent likely poor performing organizations from entering a marketplace without stymieing innovation? Put differently, can policy makers craft entry barriers screening prospective nonprofit entrepreneurs without discouraging the promising ones? The presented Milwaukee voucher case suggests it is impossible to totally eliminate the possibility of a poor-performing entrepreneurial school. Policy makers (as well as other stakeholder groups) can thus never ensure high performance, yet we believe there are policy steps to be taken to help reduce the chance for failure. First, new voucher school entrepreneurs could be required to demonstrate diverse revenue streams prior to be granted access to government revenues. This may include private fundraising dollars, evidence of private-paying customers, or reserves equal to a percentage of a school’s budgeted revenues. These requirements would force a school to show evidence of viability beyond government revenues, and help ensure that a nascent school would not fail at the first sign of fiscal stress. Second, new entries could be required to find an existing organization to sponsor their entry into the marketplace. For example, if an entrepreneur wanted to open a new voucher school, he or she would first have to have to convince an existing provider of the soundness of their plan, and have that existing provider agree to enroll their pupils if the new school is not able to open. The risk, of course, is that existing providers would be fearful of competition and refuse to sponsor new schools. This concern could be addressed by offering a financial incentive for the sponsoring school based on the performance of the new school, or allowing third party accreditation agencies to serve as an alternative sponsor. Third, new entrepreneurs could be forced to submit to a state entity (in the voucher case the state board of education or its equivalent), and obtain approval from said entity, of a transition plan to ensure the organization’s customers (i. e. students) and records are accounted for if the nascent organization fails to open, or opens and fails. While this would not prevent a school failure, it would minimize the externalities of a school failure without stopping entrepreneurship.

To answer are our own question, though it is impossible to prevent failed entrepreneurship, there are reasonable policies that can reduce the chances of failure, and mitigate the negative public impact of failure without eliminating entrepreneurial activity. It is important for policy makers to consider the context, however, when addressing failure risk. Policy makers should consider several questions. First, what is the public good being provided? The consequences of a failed school, for example, are greater than the consequences of a failed private provider of Department of Motor Vehicles services. Second, what is the public’s tolerance for failure? This will depend on the political attitudes and culture of different policy environments. Third, what is the state’s monetary investment in the entrepreneurial activity? In the voucher context the high level of public support creates significant financial consequences for the state when institutions fail. An entrepreneurial activity with minimal state investment may not need as much scrutiny from policy makers because the consequences of failure are less impactful on public coffers.

In this paper we have devoted little attention to such active formal institutions, nor have we taken into account the type of informal institutional factors with the power to shape and influence the founding rate e. g., prevailing norms and values, media coverage, or lobbying efforts. Future research is needed to examine a fuller spectrum of institutions and their impacts on the founding rate. A related question is how nonprofit entrepreneurs and other stakeholders are able to build legitimacy for voucher schools, and how this building process impacts the attempts as well as the founding success rate. The legitimacy question is particularly interesting because, as Aldrich and Fiol (1994) comments, entrepreneurs willing to venture in the formative years of a new industry are not just taking great risks, they may even be considered foolish. While the creation of formal institutions targeting the new industry will have some legitimizing properties, we still need to know more about what factors affects an emerging industry’s legitimacy, and more on the specific legitimating strategies pursued by nonprofit voucher entrepreneurs.

Though this research only looks at one longstanding voucher program, we believe our analysis also illuminates a noteworthy facet of nascent stage nonprofit entrepreneurship, another seldom investigated topic in the nonprofit literature.

Anecdotal evidence suggests taking the step from being a nascent nonprofit entrepreneur to having a fledgling nonprofit is far from easy, and that the idea and intent of the nascent nonprofit entrepreneur peter out because their intentions were misguided or because they were unable to acquire a sizeable enough resource base to create a new organization. Thus, the nascent stage often end up being messy and hard to control due to various selection forces putting pressure on nascent nonprofit entrepreneur. Consequently, many attempts are believed to end in near misses, and fail before the formal nonprofit is launched. Because of the difficulty identifying and sampling nascent nonprofits we have little empirical understanding of nascent stage success and failure. The research presented in this paper demonstrates that a large majority of nonprofit entrepreneurial attempts, about 70 percent in the case of the MPCP, fail somewhere between the stage of entrepreneur intent and actual establishment of the organization. This finding thus lends support to the notion that, at any given time, we only observe only a surviving fraction of a much larger pool of nonprofit entrepreneurial efforts begun but abandoned by nascent nonprofit entrepreneurs. Clearly, nonprofit scholars need to pay more attention to the nascent stage to more fully comprehend the process of nonprofit entrepreneurship. As noted by Aldrich and Martinez (2001, 43):

Failure to appreciate the level of turnover and turbulence in populations has blinded social scientists and public policy experts to the organizational fermentation simmering just below the surface in modem societies. Past research has focused mainly on the prominent exceptions of organizations that manage to survive and grow, rather than the ubiquitous efforts that fail.

The finding that seven out of ten attempts to create new voucher schools fail tells us that a new formal nonprofit organization is just one possible outcome of nonprofit entrepreneurial activity. Furthermore, not only do most attempts to start new voucher schools in Milwaukee fail, of those succeeding in creating a new organization many will not succeed in sustaining the new school once formally created (Ford and Andersson 2016).

In summary, in this paper we have examined how the nonprofit entrepreneurial process associated with the creation of new voucher schools in the City of Milwaukee has been influenced and shaped by formal institutional arrangements. We find that formal institutions have impacted the founding rate of new voucher schools, thus, shaping the entry patterns of new voucher schools and those attempting to start new schools. However, this research has only scratched the surface, and more research is required to better understand the effects of formal as well as informal institutions and how they interact with the nonprofit entrepreneurship process. Research on other vouchers program, the entrepreneurs themselves, school performance, and the interactions and capital flows between start-up and existing voucher schools would help address the limits of our study.