Outsourcing Public Service Delivery ... - Wiley Online Library

160 downloads 2231 Views 756KB Size Report
Outsourcing Public Service Delivery: Management Responses in Noncompetitive Markets 887. Amanda M. Girth. The Ohio State University. Amir Hefetz.
Amanda M. Girth The Ohio State University Amir Hefetz University of Haifa, Israel

Jocelyn M. Johnston American University Mildred E. Warner Cornell University

Outsourcing Public Service Delivery: Management Responses in Noncompetitive Markets

Capturing the benefits of competition is a key argument for outsourcing public services, yet public service markets often lack sufficient competition. The authors use survey and interview data from U.S. local governments to explore the responses of public managers to noncompetitive markets. This research indicates that competition is weak in most local government markets (fewer than two alternative providers on average across 67 services measured), and that the relationship between competition and contracting choice varies by service type. Public managers respond to suboptimal market competition by intervening with strategies designed to create, sustain, and enhance provider markets. In monopoly service markets, managers are more likely to use intergovernmental contracting, while for-profit contracting is more common in more competitive service markets. The strategies that public managers employ to build and sustain competition for contracts often require tangible investments of administrative resources that add to the transaction costs of contracting in noncompetitive markets.

T

noncompetitive markets because governments need to find and retain vendors and exercise greater oversight given the lack of discipline instilled by competition. Under such conditions, governments might allocate scarce administrative resources to strategies designed to mitigate the disadvantages of these suboptimal markets. Public managers must respond to competing demands from policy makers who support marketbased nongovernmental service delivery (sometimes even when market conditions are less than favorable), as well as to citizen expectations for accountability and service quality, which can be more difficult to achieve in thin markets. In local governments, which are the focus of this research, the competitiveness of public service markets is likely to vary by location—that is, by rural, suburban, and urban characteristics—and by the type of service contracted out. Rural markets are most widely affected by low levels of competition (Kodrzycki 1994; Mohr and Deller 2010; Warner 2009). However, our research also finds that thin markets exist across a number of service delivery areas. We seek to better understand the dynamics of this problem in local service markets.

he economic efficiency produced by competition is a fundamental premise underlying the practice of government contracting. Competition offers government agencies the potential for improved quality and The purpose of our research reduced costs generated by We explore the relationship on these topics is twofold. We market forces (Boyne 1998; between competition and explore the relationship between Pack 1987; Savas 2000). A key contracting by focusing on the competition and contractobjective of contracting is to impact of service type and level ing by focusing on the impact correct inefficiencies by harnessof competition on the decision of service type and level of ing the market to maximize to contract and on the type of competition on the decision returns on taxpayer investment, to contract and on the type of improve government perforcontractor used. contractor used. We also offer mance, and enhance citizen detailed insights into managewelfare (Kelman 2002; Osborne ment responses to noncompetitive markets, including and Gaebler 1992). efforts to enhance public value in such environments. Yet “thin” provider markets (Weimer and Vining The article proceeds as follows: Through the local 2005) for government contracts are a common government lens, we review the theoretical foundation problem. Noncompetitive public service markets of competitive public service delivery, addressing precan compromise efficiency gains and contribute to vious literature on location and service characteristics contract transaction costs. Transaction costs rise in

Amanda M. Girth is assistant professor in the John Glenn School of Public Affairs at The Ohio State University. Her research interests include government contracting and public–private partnerships, with a focus on performance and accountability. E-mail: [email protected] Amir Hefetz is a researcher at Haifa University in Israel. His research focuses on the privatization practices of U.S. local governments. E-mail: [email protected] Jocelyn M. Johnston is associate professor of public administration and policy at American University. Her current research focuses on public management and alternative public service delivery, specifically, government contracting and collaborative service delivery networks. E-mail: [email protected] Mildred E. Warner is professor of city and regional planning at Cornell University. Her research focuses on privatization, decentralization, and economic development policy among U.S. local governments. E-mail: [email protected]

Public Administration Review, Vol. 72, Iss. 6, pp. 887–900. © 2012 by The American Society for Public Administration. DOI: 10.111/j.1540-6210.2012.02596.x.

Outsourcing Public Service Delivery: Management Responses in Noncompetitive Markets 887

and on contract management, especially as affected by market conditions. We hypothesize that local governments are less likely to contract in less competitive markets and that they tailor contractor choice to market and service needs, relying on for-profit contractors when markets are more competitive and on intergovernmental and nonprofit contracting when markets are weaker. Next, we present results from survey and interview data collected from local government officials across the United States. Finally, we assess the policy implications resulting from the findings of this study.

contract. As a rule of thumb, a set of three or more bidders is indicative of a minimal level of competition. Former Indianapolis mayor Stephen Goldsmith, an early contracting proponent, contended that three or more bidders met competitive thresholds, and in that count, he included city departments. Alternatively, two bidders yield sufficient competition according to the Reason Foundation (Van Slyke 2007, 166). However, assumptions underlying the concept of a fully functioning market typically include “many” producers and suppliers that can easily enter and exit the market.

This research contributes to our understanding of local government contract supply This research contributes to our Competition is helpful but does not guarmarkets, contracting decisions, and manageantee contracting success; careful contract understanding of local government responses to contextual conditions. We management is also required (Fernandez ment contract supply markets, offer evidence that market conditions and 2009; Hefetz and Warner 2007; Lamothe and contracting decisions, and man- Lamothe 2010; Romzek and Johnston 2005). service characteristics help explain choice of agement responses to contextual The benefits of competition are especially contracting form and that public managers are often effective assessors of their contractquestionable for some social welfare services conditions. ing environments and tailor their contract because stability and service disruptions management efforts accordingly (Stein 1993). can be especially problematic for vulnerable In thin markets, which are not unusual in the local government populations (Schlesinger, Dorward, and Pulice 1986). In fact, more markets analyzed here, managers adjust contract type and oversight monopolistic and less competitive markets may be preferable for methods and devote resources to improving supplier markets when such services (Johnston and Romzek 2008; Milward and Provan they can, with the objective of improving their position in the 1998, 2000). Similarly, for some public works (e.g., water distribupurchaser–contractor relationship and enhancing accountability and tion, waste collection, energy), managing monopoly may be preferservice quality. able to managing competition because of the natural monopoly of network infrastructure and the benefits of economies of scale Theoretical Review (Ceriani, Doronzo, and Florio 2009; Ramesh, Araral, and Wu 2010; Warner and Bel 2008). Competition and Public Service Delivery Economic theory and the market failure rationale serve as the At the same time, outsourcing costs may be higher than measured foundations for the competitive sourcing of public services, with in efficiency comparisons. As our understanding of contracting competition widely described as critical to outsourcing success has increased, more attention has been given to transaction costs (Donahue 1989; Kettl 2002; Osborne and Gaebler 1992; Savas and their impact on contracting (Bel and Fageda 2008; Brown and 2000). Essentially, market theory tells us that competition fosters Potoski 2003, 2005; Fernandez 2009; Ferris and Graddy 1991; cost and quality control because there are punitive consequences Hefetz and Warner 2012; Johnston and Girth 2012; Lamothe, to inefficient behavior—namely, the purchaser’s selection of another supplier to provide the good or service. However, empirical Lamothe, and Feiock 2008; Levin and Tadelis 2010; Nelson 1997; Sclar 2000; Whittington forthcoming). Contracting agencies may research comparing public and private delivery does not find conunderestimate the true cost of contracting because transaction costs sistent support for efficiency gains under private production (Bel, are often excluded in cost–benefit analyses comparing the two secFageda, and Warner 2010; Bel and Warner 2008; Boyne 1998; tors (Sclar 2000). By some estimates, these transaction costs can Hirsch 1995; Hodge 2000; Leland and Smirnova 2009; Lowery reach up to one-quarter of the cost of the contract (DeHoog 1990; 1982; Stein 1990). Pack 1989; Prager 1994; Whittington forthcoming). Such costs One potential explanation is that gains from the injection of market include the administrative resources needed to manage the solicitation and bidding and award processes, implement the contract, and forces rely on the assumption that markets are competitive, yet perform adequate oversight. Some of these costs are not exclusive this fundamental premise is flawed in many public service markets to outsourced services, but goal conflicts, information asymmetry, (Heinrich 2010; Hirsch 1995; Kodrzycki 1994; Sclar 2000; Warner vendor opportunism, moral hazard, and other elements of the and Bel 2008). Governments supply many public goods and contracting relationship (Carr, LeRoux, and Shrestha 2009) could services that are not efficiently produced in the private marketplace. Creating markets for these services can require considerable adminis- potentially raise elements of these costs beyond levels needed for trative resources and may not succeed (Lowery 1998).1 According to internal oversight. There is also evidence that oversight of local government contracts may be negligible, and perhaps inadequate surveys conducted by the International City/County Management (Hefetz and Warner 2004; Johnston, Romzek, and Wood 2004). Association (ICMA), the percentage of municipalities experiencing Marvel and Marvel (2007) suggest that oversight of in-house proinadequate vendor supply grew from 25 percent in 1992 to 31 perduction, while lower than under for-profit contracting, is actually cent in 2007 (Warner 2009). Even when competitive public service higher than for intergovernmental and nonprofit contracts. This markets exist, they may erode over time (Moore 2004; Sclar 2000). could result from the contract management resource constraints described in some studies of contracting (Johnston and Girth 2012; The definition of a competitive market remains elusive, with no GAO 2006).2 clear consensus on the ideal number of bidders for a government 888

Public Administration Review • November | December 2012

Location and Markets

Contract and Market Management

The literature on the adequacy of competition suggests considerable variation in supply levels across place. Location serves as a proxy for local market conditions and other differences in citizen or managerial preferences. Market conditions have been shown to be more important in explaining variation in contracting levels across metropolitan status than managerial factors (Warner 2006; Warner and Hefetz 2002b). Rural areas, because of their small size and sparse settlement, are less attractive to private vendors (Bel and Fageda 2011; Kodrzycki 1994; Warner and Hefetz 2003). Metro core communities have more complex service requirements that limit the supply of private vendors, and large cities already enjoy internal economies of scale that can promote efficiencies in direct government delivery. Comparison of contracting levels over time shows that suburbs are consistently higher than metro core or rural municipalities in their levels of contracting (Greene 2002; Warner 2006). Suburbs represent average-sized governments, and there are enough of them in a metro region to create a market for private vendors (Joassart-Marcelli and Musso 2005; Warner and Hefetz 2002a). Yet weak public service markets exist even in urban and suburban areas, where the supply of nongovernmental service providers should be robust (Hirsch 1995; Johnston, Romzek, and Wood 2004). For instance, Savas’s (2002) examination of New York City’s homeless services contracts reveals that 70 percent attracted only one or two bids—this in a city with hundreds of nonprofits, many of which are equipped to provide these services.3

When public service markets fail to meet acceptable thresholds of provider competition, several management challenges emerge. Research on contracting reveals that public managers regularly create, shape, nurture, and expand provider markets, and these efforts intensify when markets are thin. These “market management” activities create transaction costs that are typically ignored in the contracting literature (with some exceptions; see Allen and Walker 2007; Brown and Potoski 2004; Graddy and Chen 2006; Hefetz and Warner 2007, 2012; Johnston and Girth 2012; Levin and Tadelis 2010; Whittington forthcoming). Research has shown that few contracting decisions account for the costs of preparing bids and contract oversight—let alone the less visible “market management” costs (Johnston and Girth 2012; Whittington forthcoming).

Intergovernmental contracting offers an important alternative to for-profit or nonprofit contracting. Contracting among municipalities can create a public market for service delivery when private markets are thin (Warner 2011). Managers may pursue intergovernmental contracting to increase scale or market power and to promote regional service integration (Bel and Costas 2006; Bel and Mur 2009; Hebdon and Jalette 2008; Parks and Oakerson 2000; Warner and Hefetz 2002a). Services and Markets

Service differences also help explain variations in the strength of competition and levels of contracting. The decision about production mode—whether to contract out or provide a service directly— is affected by several factors. Services that are asset specific and/or require capital-intensive network infrastructure will tend toward natural monopoly and thus make poor contracting candidates (Ceriani, Doronzo, and Florio 2009; Warner and Bel 2008). Similarly, services that require internal knowledge and control are less suitable candidates for contracting because of the high transaction costs involved (Brown and Potoski 2003; Deloitte 2005; Durant, Girth, and Johnston 2009; Thurmaier and Wood 2002). Complex services, which entail difficulty in specifying contracts or monitoring performance (e.g., when contracts will be incomplete or prone to failure), typically will be less desirable candidates for contracting (Brown, Potoski, and Van Slyke 2010), although empirical research suggests that market characteristics matter more than service characteristics in determining the level of contracting (Hefetz and Warner 2012). Other relevant factors include level of uncertainty, frequency of transactions, institutional and legal frameworks in which the organization operates, political environment, and necessity for longterm investments (Feiock and Jang 2009; Ferris and Graddy 1986; Hefetz and Warner 2007; Williamson 1991, 1999).

Market management strategies include creative contracting arrangements designed to foster competition and prevent contractor monopolization. For instance, Hansen (2003) found that contracting officials in New York City use multi-award contracts to curb monopolization by nongovernmental social service providers. Managers also adopt mixed service delivery—by retaining at least a portion of the service in house—to create competitive pressures on private and nongovernmental firms (Johnston, Romzek, and Wood 2004; Warner and Hefetz 2008). In addition, managers adjust service boundary areas, often in rural areas, to maintain “economically viable enrollment levels” (Fossett et al. 2000, 40; see also Romzek and Johnston 2005). Local governments also use reverse privatization or contract back in to address lack of competition, insufficient cost savings, or concerns about service quality (Hefetz and Warner 2004, 2007, forthcoming). Markets are sometimes purposely weakened by policies that limit competition to local or regional vendors in order to allow the municipality or region to internalize the contract’s economic benefits. In such situations, managers may use both stricter contract oversight and other market management strategies to protect the government’s contract bargaining position. The need for market management strategies does not end when the contract is signed, nor do transaction costs. In fact, many researchers argue that transaction costs begin to build at that point (DeHoog 1990; Johnston and Girth 2012; Sclar 2000; Van Slyke 2007; Whittington forthcoming). Considerable attention must be paid to markets after a contract takes effect because government outsourcing may in fact encourage provider consolidation (Sclar 2000). Schlesinger, Dorward, and Pulice (1986) described the multiple forces encouraging consolidation among contractors for local mental health services in Massachusetts, including economies of scale in both provision and bidding. For solid waste contracts in both the United States and Europe, vendor consolidation and trends toward monopoly are common (Bel and Fageda 2011; Davies 2007; Dijkgraaf and Gradus 2008). In the postcontract phase, market management includes the investment of scarce administrative resources in order to retain, nurture, and strengthen contractor organizations that threaten to leave the market or are otherwise vulnerable to failure (Johnston and Girth 2012). In some cases, scrutiny of monopoly, or “sole source” contracts, is higher, as reported to us by a county manager in Michigan. In his county, “no one wants [contracts] to show up on the sole source list” because these are subject to mandatory review by the full county governing board. So contract managers face incentives to increase competition.

Outsourcing Public Service Delivery: Management Responses in Noncompetitive Markets 889

DeHoog (1990), Van Slyke (2007), and Fernandez (2007) discuss ongoing efforts to improve contractor performance under the rubric of “relationship contracting” and “stewardship.” This nurturing and mentoring of existing contractors is costly, and it is likely to cost more when markets are noncompetitive because there are few alternative providers and the government has an incentive to protect market supply. Under such conditions, effective contract oversight may require more frequent government exchanges with the contractor in order to fine-tune incentives, mutual understanding, and performance in general (DeHoog 1990; Romzek and Johnston 2005). In the section that follows, we explore these issues in greater depth using both quantitative and qualitative data. We address the following questions: How competitive are local public service markets? Does competition vary across location? Does the level of competition affect the decision to contract out, and which type of provider to use? Do these dynamics vary across service types? How do managers respond to these conditions? Data and Analysis This study follows a multi-method approach and relies on quantitative data from two national surveys and qualitative data from interviews with government managers. The strength of this approach is that the survey data analysis, which generates aggregate results that focus on the impact of competition and service type on service delivery choice, is enriched and informed by the detailed insights gleaned from in-depth interviews with contract managers. The quantitative analysis relies on two national surveys of city managers conducted by the ICMA in 2007. The first survey—the Alternative Service Delivery Survey—asked how local governments deliver each of 67 services. Public delivery, intergovernmental contracting, for-profit contracting, and nonprofit contracting are the primary alternatives to in-house delivery. The survey was sent to the chief administrative officer in all cities with populations of more than 10,000 and all counties with populations of 25,000 or more.4 In 2007, we conducted a supplemental Service Characteristics Survey with the ICMA to determine the level of competition for each service measured in each municipality (Hefetz and Warner 2012).5 The second survey asked local managers to give the number of alternative suppliers for each of 67 commonly provided local government services.6,7 Competition, or the number of alternative suppliers, was measured on the following scale: 0 = government only, 1 = one alternative provider, 2 = two alternative providers, 3 = three alternative providers, and 4 = four or more alternative providers. In addition, interviews were conducted with 24 local government officials, most between 2008 and 2010. The objective of the interviews was to obtain contract administrators’ perspectives on managing contracts, especially in suboptimal provider markets. A convenience sample targeted public officials with experience in managing contracts. Respondents included generalist managers in general purpose local governments, purchasing officers, and program specialists with responsibilities for such services as community mental health, corrections, social welfare, public safety, parks and recreation, refuse collection, and others. Through a semistructured interview protocol, managers were asked to describe their government’s decision to contract, their criteria for making the contracting 890

Public Administration Review • November | December 2012

decision, their contract selection practices, and their oversight strategies. They were also asked to elaborate on how they responded to noncompetitive markets and whether and how they tried to address weak supply markets.8 Location and Service Effects on Competition and Contracting

Responses from the two ICMA surveys were merged to form a sample of 118 municipalities that answered both surveys (30 metro, 66 suburban, 22 rural).9 This combined data set enables us to link data on competition levels with data on how each individual service is provided. On average, the local governments in our sample provide about 38 of the total 67 listed services (the complete list of services is provided in figures 1–3). The combined sample yielded 4,364 cases of known service delivery and competition levels across the 118 municipalities. The ICMA Service Characteristics Survey results indicate that, on average, there are fewer than two alternatives to in-house provision for most local services. Rural communities reported, on average, only one alternative provider (1.1) for most services. Suburban and metro governments, on average, face fewer than two alternative providers (1.8), but even this is not a high level of competition (Hefetz and Warner 2012). Only a third of the services measured have more than two alternative providers; this would be considered minimally competitive by most definitions. Note that respondents were asked about the number of suppliers in the market, as opposed to the actual supply of bidders; therefore, these results may overestimate the availability of real alternatives to current service delivery. Prior research has used this combined survey data to assess how competition, measures of transactions costs, and metro status are related to service delivery choice in a multivariate framework (Hefetz and Warner 2012). Our interest here is in assessing the general relationship between the form of service delivery and the level of competition. The particular focus is whether and how that relationship differs across service types; we look closely at each of the 67 services and conclude that the variation is considerable. To supplement this result, we provide context through subsequent qualitative analysis of management strategies in thin markets. Table 1 provides detail on our cases. About half (51 percent) of the individual service cases (2,241) face monopoly conditions in the local market, meaning that there is only the government (37 percent) or one alternative provider (14 percent) for the service. The other half of the cases reported markets of two or more alternate providers: 728 service cases (17 percent) reported only two alternative providers, 371 (9 percent) cases reported three alternative providers, and 1,024 (24 percent) cases reported four or more alternative providers. In terms of the form of delivery, 51 percent of the 4,364 local service cases are delivered directly by government (public delivery). In all, 28 percent (1,214) are delivered totally as contracts (for-profit, nonprofit, or with other governments). Mixed delivery accounts for 21 percent of all service cases in our combined sample; this delivery form, which involves a combination of direct government delivery and some form of contracting, is becoming common among local governments (Lamothe, Lamothe, and Feiock 2008; Warner and

Table 1 Selected Descriptive Statistics Level of Competition 0 No. service cases 1,622 Percent 37%

1

2

3

4+

Total

619 14%

728 17%

371 9%

1,024 24%

4,364 100%

Extent of Contracting Public No. service cases 2,223 Percent 51%

Mixed Delivery

Totally Contracted Out

Total

927 21%

1,214 28%

4,364 100%

Form of Delivery Public No. service cases 2,223 Percent 51%

Intergovernmental

For-Profit

Nonprofit Comb.

666 15%

875 20%

209 5%

391 9%

Privatization using for-profit firms is reserved for cases with higher competition (0.57**). In the nonprofit category, the correlation is low and insignificant (0.22), suggesting that the choice of nonprofit delivery is not correlated with competition levels. Table 2 also provides a breakdown of these correlation results by metro status. We note similar results as overall, except that the correlations with mixed and public delivery are higher for metro core than suburb or rural, and the correlations for intergovernmental contracting are higher for suburbs and rural areas than metro core places. This is consistent with other studies of metro differences in contracting (Bel and Fageda 2011; Greene 2002; Hefetz, Warner, and Vigoda-Gadot, forthcoming; Joassart-Marcelli and Musso 2005; Kodrzycki 1994; Warner 2006; Warner and Hefetz 2003).

Total

4,364 100%

Source: Author analysis of ICMA Alternative Service Delivery Survey, 2007, and ICMA Service Characteristics Survey, 2007.

Hefetz 2008; Hefetz, Warner, and Vigoda-Gadot 2012). Further detail on form of delivery is provided in table 1.10 Because our interest is in differences by service, we tested correlations between the average competition level of each service (N = 67) across all places versus the frequency of each delivery choice (public, intergovernmental contracting, for-profit contracting, and nonprofit contracting) for each service. In table 2, we present these by-service correlations, aggregated across all responding governments. The correlations are consistent with our hypothesis that local governments are less likely to contract as market competition decreases. We see that direct public delivery is negatively and significantly correlated with level of competition (–0.36**). This makes sense; when competition is higher, public delivery is lower.11 Next, we see that mixed delivery is positively correlated with competition (0.45**). Although mixed delivery is used in part to create competition (between public and private providers), we would have expected to see a negative correlation—that is, more mixed delivery when competition levels are lower. Finally, the correlation between competition and totally contracted out is insignificant (0.11). This comes as a surprise and is contrary to our hypothesis that contracting would be higher when competition is higher. To better understand these counterintuitive results for mixed delivery and totally contracted out, we divided the service cases in table 2 into two categories—intergovernmental contracting and for-profit contracting. Both involve contracts, but intergovernmental contracting uses a public market of competing (or cooperating) governments, whereas for-profit contracting relies on a private market of alternative suppliers. For both mixed delivery and totally contracted out, we see that intergovernmental contracting is correlated with low competition levels (–0.47** mixed and –0.41** total out), whereas for-profit contracting is correlated with higher competition (0.47** mixed and 0.35** total out). These correlations fit with our hypothesized directions and are statistically significant. When governments face low levels of competition, they rely more on public markets and intergovernmental contracting (–0.46**).

Average levels of competition do not tell the whole story. We are especially interested in assessing differences across individual services. We divided each service into two categories, based on supply levels—monopoly (zero or one provider, e.g., government or only one alternative provider) and competition (two, three, or four or more providers). For each service, we look at the delivery choices made by all local governments that provide the service. In each figure (1–3), the length of each bar represents 100 percent of service delivery for that service. The portion of the bar above the center line represents cases in which there is a monopoly situation, and the proportion of the bar below the line represents cases in which there is market competition. We order the services from least competitive (figure 1) to most competitive (figure 3). For each service, we present the frequencies of delivery form (public, intergovernmental, for-profit firm, or nonprofit), as found under monopoly conditions (zero or one alternative providers), and under more competitive conditions (two or more alternative providers) (see figures 1–3). The frequency of service delivery clearly varies by service. We report the number of cases for each service in the service label. The maximum possible number of cases is 118, the total number of places in the combined surveys. The highest service delivery frequency is 103 Table 2 Correlations: Choice of Delivery Form and Level of Competition Overall

Metro Core

Suburb

Rural

% Public entirely

0.36**

–0.42**

–0.29**

–0.30**

% Mixed delivery % MIX intergovernmental cooperation

0.45**

0.57**

0.29**

0.28**

0.09

0.11

0.13

% MIX for-profit % Totally contracted out % OUT intergovernmental cooperation % OUT for-profit % Intergovernmental cooperation (MIX and OUT) % For-profit contracting (MIX and OUT) % Nonprofit contracting (MIX and OUT)

–0.47** 0.47** 0.11 –0.41** 0.35** –0.46**

–0.29**

–0.33**

–0.36**

0.57**

0.53**

0.54**

0.58**

0.22

Source: Author analysis of ICMA Alternative Service Delivery Survey, 2007, and ICMA Service Characteristics Survey, 2007. By service correlations, N = 67 services based on average across all cases, overall and by metro status. The average competition level for the metro status analyses is computed just for those cases. Correlations could not be run by metro status for subcategories of mixed and totally contracted or for nonprofit because of zero frequency for some services in those subcategories. ** Significant at p < .01.

Outsourcing Public Service Delivery: Management Responses in Noncompetitive Markets 891

cases, reported for payroll services, of which 64 cases were reported operating within a competitive market and 39 operating under monopoly. The lowest frequency is 14 cases, reported for hospital services, of which half were reported operating within a competitive market and half operating under monopoly. We split the cases into three groups—monopoly, low competition, and moderate competition—using the ratio of the number of cases in a monopoly market over the total number of cases for each service. Figure 1 (monopoly) presents services with a monopoly percentage ranging from 95 percent to 66 percent. Figure 2 (low competition) presents services with a monopoly percentage ranging from 65 percent to 44 percent. Figure 3 (moderate competition) presents services with a monopoly percentage ranging from 39 percent to zero (there are no cases of monopoly in vehicle towing).

programs for the elderly, and so on. Intergovernmental contracting is much lower in this group than in the monopoly group (figure 1), but it is most common in mental health programs, paratransit operation, and hazardous materials disposal. It is interesting to note that there is no direct public delivery in mental health or hospital operation among these cases. Overall, this group of low-competition services exhibits a wide variety of delivery forms. Competitive markets. Figure 3 shows the services with the highest levels of competition (below the line). Direct government delivery (white) remains the primary mode of service delivery for several administrative and support services. Although many of these are back-office services with little public interaction, governments may wish to maintain internal control and expertise. As services exhibit more competitive markets, we see the rate of for-profit delivery rise. The highest levels of for-profit delivery are found in commercial and residential waste collection, street repair, vehicle towing, legal services, and tree trimming. These historically have been the local government services with the highest levels of outsourcing. Intergovernmental contracting is not common among this set of more competitive market services (except in job training). Nonprofit delivery is used for cultural programs, day care facilities, and some social services and culture/art programs, among others. In administrative operations and maintenance services, direct public delivery still dominates. Overall, as competition increases, for-profit delivery becomes more common.

Monopoly markets. Figure 1 shows the services most prone to monopoly markets. As expected, direct government delivery (white) is the most common delivery choice for the services at this end of the continuum. Direct government delivery is especially high in crime prevention and patrol, fire prevention, traffic control, police and fire communication, water distribution and treatment, and code enforcement. For these—services with few alternative providers and high impact on the public—governments tend to choose to maintain internal control. When contracting does occur among these monopoly services, the typical alternative form is intergovernmental contracting, which keeps service delivery public. Intergovernmental These data suggest that local government These data suggest that local contracting is most common in complex managers consider both market competigovernment managers conmonopoly human services such as child tion and service context when they choose sider both market competition welfare and public health programs. It is also public delivery, intergovernmental contractand service context when they very common in jail operation. We also see ing, or contracts with for-profit or nonprofit intergovernmental contracting in services for organizations. This is why intergovernmental choose public delivery, interwhich cooperation may be needed to contracting is most common in this group of governmental contracting, or promote regional coordination—airport and services. contracts with for-profit or library operation and police and fire nonprofit organizations. communication, for instance. Finally, we see In the next section, we use interview data substantial levels of intergovernmental to provide further detail on why managers contracting for administrative services—tax assessing, title records, respond to noncompetitive markets as they do, looking beyond the and tax bill processing. Overall, figure 1 indicates that for-profit choice of service delivery form. While our aggregate data provide and nonprofit contracting are minimal in this monopoly group. insights into the relationship between market competition and Governments are unlikely to contract out in monopoly markets, contract type, the interview data offer a more nuanced view of and if they do, contracting will be to other governments—thereby managers’ perceptions and choices. Public managers are clearly retaining public control. cognizant of transaction costs and seek to minimize them. Managers in noncompetitive markets try to build competition, in part to Low competition. The next group of services (figure 2) faces mixed reduce contracting transaction costs. They may also purposely avoid markets. Monopoly is still dominant among these services (above for-profit contracting when the advantages of partnering with other the line), but a growing percentage of municipalities benefit from governments—such as trust, common service cultures, and willingsome competition. Direct government delivery (white) is the most ness to cooperate to maximize scale economies—offer the potential prevalent form of delivery for half of the services in this group, for reduced transaction costs and higher net benefits to the partners which include emergency medical, recreation, and street cleaning, (Brown and Potoski 2003; Thurmaier and Wood 2002; Warner among others. However, for-profit delivery becomes more common 2011). in this group of services; for example, utilities and waste management are areas with high asset specificity and in which the Managing Markets and Contracts under Noncompetitive private sector has well-known expertise—even if under monopoly Conditions conditions. The only other services that use for-profit delivery Interviews with local contract managers indicate that noncompetiheavily in these low competitive markets are hospitals12 and tive markets present special administrative challenges. Contract ambulance services. Nonprofit delivery is the primary delivery administrators seek to minimize transaction costs in their conchoice for most social services in this group—mental health, tracting decisions and subsequent contract management efforts. 892

Public Administration Review • November | December 2012





snow plowing/sanding; n=68

electric utility management; n=25 emergency medical service; n=68 gas utility management; n=20

animal shelter operation; n=62 paratransit system maintenance; n=32 utility meter reading; n=57

sludge disposal; n=60

hazardous materials disposal; n=45

parking meter maintenance; n=27

recreation facilities maintenance; n=88

utility billing; n=63

cemeteries maintenance; n=42

mental health programs operation; n=21

waste disposal; n=71

elderly programs; n=50

delinquent tax collection; n=64 hospital operation/management; convention centers/auditoriums;

ambulance service; n=62

personnel services; n=88 traffic sign maintenance; n=73

street/lot cleaning; n=84

bus system maintenance; n=41

Figure 1 Service Delivery Choices under Monopoly

museums operation; n=27

Source: Author analysis of ICMA Alternative Service Delivery Survey, 2007, and ICMA Service Characteristics Survey, 2007.

tax bill processing; n=70

inspection/code enforcement; n=99

animal control; n=79

water treatment; n=65

water distribution; n=70

sanitary inspection; n=66

public health programs; n=38 title records/plat map maintenance; n=50

tax assessing; n=61 sewage collection/treatment; n=74 police/fire communications; n=99 child welfare programs; n=26 traffic control/parking enforcement; n=85 fire prevention/suppression; n=87 prisons/jails; n=42

libraries operation; n=63

welfare eligibility determination; n=26

airport operation; n=36

crime prevention/patrol; n=101

Non-Profit

For-Profit

Inter-Govt.

Government

Non-Profit For-Profit Inter-Govt. Government

Source: Author analysis of ICMA Alternative Service Delivery Survey, 2007, and ICMA Service Characteristics Survey, 2007.

Figure 2 Service Delivery Choices under Low Competition

Outsourcing Public Service Delivery: Management Responses in Noncompetitive Markets 893



vehicle towing and storage; n=45 daycare facilities operation; n=20

legal services; n=80

tree trimming/planting; n=67 drug/alcohol treatment programs; n=22 commercial waste collection; n=55 street repair; n=68

residential waste collection; n=71 secretarial services; n=91

buildings/grounds maintenance; n=88 all other vehicles maintenance; n=85 cultural/arts programs operation; n=33 parks landscaping/maintenance; job training programs; n=27 homeless shelters operation; n=18 data processing; n=84

insect/rodent control; n=45

building security; n=73

lots/garages operation; n=44 public relations/public information; n=86 payroll; n=103

heavy equipment maintenance; n=81 emergency vehicles maintenance; n=82

Government

Inter-Govt.

For-Profit

Non-Profit

Source: Author analysis of ICMA Alternative Service Delivery Survey, 2007, and ICMA Service Characteristics Survey, 2007.

Figure 3 Service Delivery Choices under Moderate Competition Managers also respond to weak markets in part by invoking strategies designed to build and sustain provider competition when necessary and to improve their position as purchasers of essential services. These interview data are not based on a random sample of managers; rather, our purpose here is to provide context for the survey results, to better understand management responses to thin markets, and to derive insights from managers as to why markets may be thin even when location and service type would suggest otherwise. While the aggregate results from our survey data indicate correlations between service delivery form and provider supply, the qualitative data suggest that some contract management challenges—such as reluctant bidders and vendor consolidation— can persist across markets and services, driving up oversight and transaction costs. The aggregate data focus on choice of service delivery mode, while the qualitative data offer more detail on the execution of contracts after the mode choice is made. Some respondents recognize clearly that the choice to contract may not be optimal, given the level of market competition, but they must nonetheless implement the contract if the decision was made elsewhere in the jurisdiction’s hierarchy. In effect, we are shedding light on the implementation phase, subsequent to service mode choice, but it seems clear that these contract management experiences can affect service delivery choice in the future. 894

Public Administration Review • November | December 2012

One pattern that emerges in the qualitative data is that when markets are thin, managers recognize the need to alter their oversight strategies. In many cases, they resort to the “relational contracting” dynamics associated with incomplete contracts (DeHoog 1990; Sclar 2000). They tend to devote more time to helping contractors with performance issues because if they do not, they could lose the only vendor available to provide the service. One waste disposal manager explained this dynamic in the context of a noncompetitive market for his city’s trash collection contract. He noted that in his limited market, contract oversight is more time consuming: “[I]f I have a poor [performing] contractor, I’m going to try to work with him as much as possible because it’s an essential service … I can’t kick him off the job—he’s got you. That’s why it’s important to have continuous, relentless inspection and take immediate action.” In other words, transaction costs are higher when competition is lower. Although residential waste collection falls into the “moderate competition” group of services found in figure 3, the average number of alternative providers is only 2.6, and 23 percent of survey respondent jurisdictions still face monopoly. This manager’s case, therefore, is illustrative of the costs and challenge of oversight in thin markets. Nurturing and mentoring contractors—activities that are less necessary in competitive markets—therefore can be pivotal to contract success in thin markets. Without the alternative vendors

or contractor discipline present in competitive markets, oversight efforts and relationships with contractors must be adjusted and typically will require more administrative resources (DeHoog 1990), as this waste disposal manager revealed. The manager must work to compensate for the government’s weakness in the purchaser-contractor relationship. Compare his situation to that of an Illinois city official, who noted that mentoring/nurturing is unnecessary in his city because “[we] don’t have that situation where you have a single vendor or limited number of vendors. So [we] haven’t had to [get involved in performance] because [we’ve] always been able to find other vendors to do the work.” Thin markets also drive managers to intervene in or “manage” the market in order to increase competition. Market management strategies that emerged in these interviews vary but may include scanning the environment through electronic means, tapping into professional networks of government contract managers to find vendors, piggybacking on contracts with other governments, designing mixed public and private delivery systems, creating brand new vendors for the market, using intergovernmental contracts, and otherwise encouraging vendor participation in markets for their government contracts. Our aggregate analysis suggests that such strategies will be more prevalent for the services found in figures 1 and 2 because competition in these service areas is quite low.

not want to contract out in a monopoly market, but he must do so to comply with policy.14 Another key theme that emerged from our interview data is that the decision to contract out to a nongovernmental entity may be made in a robust vendor market that quickly falls apart when formal bids are due. All too often, available vendors simply do not bid on public service contracts. To illustrate, a contract manager for information technology services in a California city reported that an original market of 23 vendors for one contract was reduced to six bids; another that began with eight vendors received only one bid. In other cases, robust markets do not reflect viable bids; a South Carolina local manager received 30 bids on an automobile technology project, but after initial review, the pool “narrowed to two.” Thus, assessment of market strength can have a very limited shelf life. Some managers reported that they follow a standard strategy designed to combat this problem—namely, a prebid meeting with interested vendors in which a detailed set of expectations is laid out to minimize surprises. But such strategies offer no guarantee. Even for more conventional and competitive market services, vendors may not show up at bid time. Washington, D.C. officials told us that four vendors were identified as potential bidders for a recent city school food service contract, but only two vendors ultimately bid. Thus, our survey respondents may in fact overestimate the number of alternative providers who would actually submit bids, as noted by Savas (2002).

These market management strategies can be viewed as generating transaction costs that are directly related to thin markets; such Vendor consolidation is another common costs supplement the more familiar transacconcern, pushing contract managers to tion costs associated with the procurement Markets that were once comdevote more resources to stimulating provider process (e.g., designing contracts, soliciting markets (Sclar 2000). Markets that were once bids, etc.). Indeed, some contract specialpetitive can quickly change, competitive can quickly change, becoming ists noted in our interviews that they devote becoming noncompetitive noncompetitive merely through the loss of more time to “chasing” competition than to merely through the loss of one one or more vendors. Consolidation plagues a traditional components of contract manageor more vendors. wide range of local public services, and there ment such as monitoring. Remarks about is a natural propensity of local markets to “building an industry”—that is, creating a consolidate over time as they mature (Heinrich 2010; Sclar 2000). supplier market—and “growing the competition” are not unusual, Our interviews indicate that managers are aware of this problem and and they provide evidence of the time costs associated with market its impact on contract administration. These forces also help explain management.13 the low levels of competition reported by our survey respondents. A contract manager from a city in California experienced this In these less favorable markets, public managers should, and many do, take measures to protect the power of the government purchaser problem with job training contracts. While the city once had three agencies soliciting bids, the failure of one agency left only two venin the contracting relationship and to ensure the best possible outdors in the market. (Job training services are found in the “moderate come in terms of cost-effectiveness and service quality without the competition” figure in our aggregate service analysis, but the avernatural performance incentives embedded in robust provider marage number of alternative providers is only two and 33 percent of kets. In some cases, they must manage the market despite their own responding jurisdictions report monopoly.) City contract managers best judgment against outsourcing. For example, a local procurein Minnesota and Missouri described the impact of consolidation on ment manager in Colorado noted that he was “seeing a trend. I’ve residential refuse markets in which one or two providers dominate got a list from city council for things they’d like to privatize … [like the market. Although smaller companies emerge periodically in these the] print shop.” He has not been able to find a vendor to beat the markets, such firms can only instill competitive pressures for a short public print shop cost. Yet “outsourcing is being pushed,” regardtime before they are bought out or merged. Similarly, a Colorado less of the manager’s judgment about its efficacy. Printing services manager noted that vendor consolidation among office suppliers and should be readily available in many local markets but are not in providers of office machinery, through mergers and buyouts among the case of this manager’s jurisdiction. His experience typifies the competitors, has diminished the available supplier market. situation of a contract manager who might opt against outsourcing because of market conditions but must do so nonetheless because Market choice also can be constrained by contracting regulations of prevailing policy. The irony here is that in some cases, the policy that include preference for traditionally disadvantaged businesses may be adopted in pursuit of cost savings but in fact requires more and, in some cases, dedicated contract set-asides for these types of resources than direct public delivery. This Colorado manager does Outsourcing Public Service Delivery: Management Responses in Noncompetitive Markets 895

firms. While arguably meritorious for social equity and distributional reasons, such policies artificially constrain the competitiveness of public service markets and can require more intense contract oversight, raising transaction costs. A Washington, D.C. official noted that her department “must spend 50 percent of expenditures in the local certified market” and that this policy posed considerable managerial barriers to finding qualified bidders. While contracts are active, managers continually scan for new vendors for the next cycle, in part to address consolidation issues and in part to create some purchasing advantage over existing contractors in thin markets.15 Attempts to improve government purchasing power do not always succeed. For example, after 14 years of contracting with one firm for ambulance services, city officials in Kansas City tried to build competition for its contract. They reported that, ultimately, the incumbent firm (a nonprofit that had been created by the city to oversee the service but had exhibited diminishing performance) was the only bidder, and it used the opportunity to increase the cost of the contract, raising its price by 20 percent. Low levels of competition would not be unusual for ambulance services, as suggested by our aggregate analysis; ambulance services are in the “low competition” category. A former Chautauqua County, New York, manager described the need for “reverse privatization” to deal with this kind of cost creep among contractors; he contracted back in as a means to increase pressure on private contractors to contain their costs (Warner and Hebdon 2001). To further illustrate, a California city contract manager described efforts to “rebid [contracts] to test the market” for parks management to assess competition and subsequently decided, after receiving only one bid, to “talk to other cities and see who they’re using,” recognizing that service termination or insourcing may be the only remaining options.

still have it and … can do it ourselves.” Equipment, or capacity retention, helped this manager maintain some level of power over the city’s contractors. He intervened with a mixed delivery strategy designed to improve the government’s bargaining position in the purchasing relationship, to reduce contract transaction costs, and to better safeguard the public interest and increase returns to citizens. As seen in our aggregate data, reduction of transaction costs is also sought through partnering with other governments, either through intergovernmental contracts or cooperative purchasing agreements (Thurmaier and Wood 2002). One Florida local government manager explained that “contracting with another government … was a very efficient way for us to work … [because] private providers … don’t appreciate the regulatory environment … they have different attitudes about things than we do.” This manager begins the process for new contracts as follows: “First, I look at other [neighboring] government contracts and piggyback on them [whenever possible] … as long as it was originally entered into with a competitive process.” Similarly, a contract manager in Virginia described contracts with a county in another part of the state, with the state itself, and with a nearby university for cooperative purchasing. These contract administrators are seeking to reduce, or at least share, transaction costs by tapping into the values and incentives common to governments. It is likely that the transaction costs associated with past contract management in noncompetitive markets are taken into consideration in subsequent service delivery decisions, and therefore we should expect to see the types of choices that emerge in our survey results. Because of the high costs of contract management in thin markets, jurisdictions are responding rationally by modifying their service delivery choice to reflect those costs. If they must contract in a noncompetitive market, they engage in market management activity to improve the contracting environment.

External events and supplier market cycles can also affect contract and market management. A public works manager in Illinois received a contract price that was “21% higher [than expected] and only one bid” in a recent competition. All of his typical bidders were Policy and Management Implications This analysis indicates that levels of competition vary considerably unavailable because the “feds have pumped dollars into infrastructure projects” in the area in response to the “great recession.” A savvy across local government services. Our survey data tell us that most of the 67 services analyzed are characterized New Jersey public works manager stressed by monopolistic or noncompetitive markets. that if bids are sought at the beginning of the Our survey data tell us that This finding raises concerns about the impacts year rather than in spring or fall, prices are most of the 67 services analyzed of policies that dictate contracting and sugbetter because contractors are “hungry for are characterized by monopolis- gests that delivery form choice should be work in winter.” shaped by contextual factors such as service tic or noncompetitive markets. type, level of competition, and overall transacDetails about these managers’ responses to tion costs. Competition is key. A California respondent described market conditions help put a face on the additional transaction costs associated with contracting in thin markets, especially in terms social service delivery in his community: “We contract out all our work. Nothing is in-house … [There is] never an expectation [for of staff time and energy. The bottom line is that when markets are the city] to provide those services … We’re well located with a lot noncompetitive, administrative resources are stretched to address of very long-standing” providers, in a strongly competitive contract competition failures. Regardless of the “causes” of noncompetitive market. In this particular environment, contracting is a good option markets—whether due to the failure of available vendors to bid for service delivery. But our analysis shows that such robust compeon contracts or provider consolidation—contract managers often tition is not common in most local service markers. Sound contexrespond strategically in an effort to mitigate the shortcomings of tual analysis in a different market may yield a different decision. thin provider markets. One local manager in Kansas illustrated the effort to build up the government’s purchasing power in thin markets. He reported that “we haven’t given up our equipment [for maintenance services that are now outsourced] … [the contractors are] smart and know we 896

Public Administration Review • November | December 2012

In competitive markets, for-profit contracting is more common— consistent with our hypothesis. Managers are assured that they have alternatives in the event of performance problems and have the luxury of competition to help reduce oversight and other transaction

costs. But for-profit contracting is found in some noncompetitive markets, especially for utilities, health care, and waste disposal. Economies of scale and technical requirements in these service areas cause delivery to tend toward monopoly. When contracting is pursued for these services, care must be taken to ensure public oversight. Public commissions provide this function for utilities, but for health care and waste disposal, such public oversight does not exist and may be necessary. Regardless, contracting out to monopoly without adequate oversight is risky and managers recognize this risk. Intergovernmental contracting is more common in service areas that lack competition. Airports, libraries, jails, public welfare programs, tax assessing, and transit tend to be monopoly services and, if contracted, are most likely to be with other governments. However, intergovernmental contracting is also common in service areas that benefit from coordination across jurisdictions such as public and mental health programs and job training, and it may be preferred in robust markets in order to reduce transactions costs. Nonprofit delivery of local services is a small portion of overall local contracting, but it is concentrated in key health and human services, culture and art, and day care services. In services for which nonprofit contracting is high, for-profit outsourcing is low—except in the case of hospitals. This suggests that some markets for public services may be differentiated by the dominance of for-profit providers, nonprofit providers, and intergovernmental providers. Overall, our survey results tell us that with the exception of utilities, hospitals, and waste management, less competitive markets are more likely to be served by intergovernmental contracts and nonprofit providers. Evidence from our interviews suggests that governments “trust” nonprofit organizations more than for-profit firms. For instance, a human services contract manager in California told us that contracts with nonprofits are given the benefit of the doubt and a relatively long leash; each agency “theoretically has three years as long as they [don’t] do anything major to jeopardize … renewal.” But the preference for nonprofit outsourcing under noncompetitive conditions is not without risk. When partnering with nonprofits, the common service mission may help reduce transaction costs, yet competition is still salient. The manager of a mid-sized local government in the state of Washington explained that a contracted nonprofit organization “was the sole provider of overnight emergency shelter—a vital community service … no one else was able or willing to provide the service. [But] they struggled with internal capacity and unfortunately there wasn’t another vendor,” and the service had to be terminated. So despite the benefits of contracting with nonprofits, some level of competition is still helpful to ensure stable service delivery.

buttress provider markets. There has been little scholarly attention directed at the transaction costs associated with chasing competition and otherwise creating, enhancing, and sustaining provider markets. Conclusion Competition is a key rationale for cost savings under government contracting, yet, as this research demonstrates, local public service markets are often noncompetitive. As a result, public managers allocate considerable administrative resources to chasing and maintaining providers and to other interventions designed to improve their bargaining positions in weak vendor markets. Across service types, levels of competition vary considerably, making some services better candidates for contracting than others. This suggests that a one-size-fits-all approach to municipal contracting is ill advised. Context is crucial (Ferris and Graddy 1986). Managers clearly understand these contexts. They recognize that market forces, opportunities for vendor opportunism, service type, contract management costs, and capacity for contract management— including market management strategies—should be considered in the decision to contract out public services. When market conditions are less than favorable, public managers may expend resources to manage the market, experimenting with strategies to attract and retain vendors. Further, if the contracted service is complex, administrative capacity may be stretched too thin to provide adequate oversight in weak markets, threatening accountability. It seems clear that the transactions costs related to managing noncompetitive markets need to be considered when governments decide how to best deliver services to citizens. Acknowledgments This research was supported in part by U.S. Department of Agriculture, National Institute for Food and Agriculture Grant no 2011-68006-30793. Notes 1. Indeed, this is one reason that governments have traditionally provided those services for which markets fail. 2. While data on contract management resources at the local level are scarce, the U.S. Government Accountability Office (2006) reported that despite the drastic increase in acquisition spending by the federal government (from $219 billion in 2000 to $347 billion in 2006), the size of the acquisitions workforce remained constant over the same time period. Similarly, at the U.S. General Services Administration, contract spending increased by more than 300 percent from 1996 to 2007, but the number of contract officers grew by only 51 (O’Harrow and Higham 2007). 3. Additional organizations expressed interest in the contracts but lacked requisite qualifications (see Savas 2002, 85, table 3). 4. The response rate for the 2007 Alternative Service Delivery Survey was 26.2 percent (6,095 surveys sent, 1,599 responses, 1,474 usable responses).

The survey results imply, and our qualitative data further support the notion, that public managers assess the contracting environment and respond with strategies that they feel will increase returns to citizens (Johnston and Girth 2012). If possible, managers address thin provider markets by using mixed delivery or intergovernmental solutions. These options help retain internal capacity and control and reduce transaction costs because of the mutual trust and public service cultures shared among governments (Thurmaier and Wood 2002; Warner 2011; Warner and Hefetz 2008). In other cases, managers respond by intervening with market management strategies to

5. The response rate for the supplemental survey was 7.4 percent (2,207 surveys sent, 164 responses). These responses were distributed similarly to the full 2007 ICMA service delivery survey (53 percent from municipalities with populations under 25,000; 22 percent from municipalities with populations of 25,000– 50,000; 12 percent from municipalities with populations of 50,000–100,000; 12 percent from municipalities with populations of 100,000–500,000; and 1 percent from municipalities with populations of 500,000). 6. The relevant survey question asked respondents to indicate the “Number of Alternative Suppliers: 0 = government only, 1 = one alternative provider, 2 = two alternative providers, 3 = three alternative providers, and 4+ = four or more

Outsourcing Public Service Delivery: Management Responses in Noncompetitive Markets 897

alternative providers.” These are managers’ estimates of number of alternative providers in the market, not necessarily the number of bidders. Questions covered each of 67 services, but not subcomponents of services, though we recognize that there could be more providers if services were broken into subcomponents, as is common in human service contracting. 7. There were 164 respondents to the supplemental survey, the majority of which were from suburban municipalities (53 percent), and the rest were from metro core (25 percent) and rural independent municipalities (22 percent). In the full

Bel, Germà, and Antón Costas. 2006. Do Public Sector Reforms Get Rusty? Local Privatization in Spain. Journal of Policy Reform 9(1): 1–24. Bel, Germà, and Xavier Fageda. 2008. Reforming the Local Public Sector: Economics and Politics in Privatization of Water and Solid Waste. Journal of Economic Policy Reform 11(1): 45–65. ———. 2011. Big Guys Eat Big Cakes: Firm Size and Contracting in Urban and Rural Areas. International Public Management Journal 14(1): 4–26. Bel, Germà, Xavier Fageda, and Mildred E. Warner. 2010. Is Private Production of

Alternative Service Delivery 2007 survey, the majority of the 1,474 respondents

Public Services Cheaper than Public Production? A Meta-Regression Analysis of

were from suburbs (53 percent), but more respondents were from rural munici-

Solid Waste and Water Services. Journal of Policy Analysis and Management 29(3):

palities (30 percent) than from central cities (17 percent) because of a deliberate oversample of rural municipalities. The ICMA survey respondents tended to be from larger governments, but this is also the subset that is more likely to pursue contracting. 8. The qualitative data are drawn from interviews with officials in local governments with contract management responsibilities. Respondents represented governments in 16 states and the District of Columbia; according to census designations, 22 were from metropolitan areas, and one was from a nonmetropolitan area; the regional distribution was as follows: Northeast, 12.5 percent; Midwest, 25 percent; West, 25 percent, South, 37.5 percent. Indeed, our purpose was not to seek a representative sample, but rather to identify response patterns across varying service areas and locations focusing on suburban and metro areas with greater concentrations of contracting. Two researchers conducted the interviews—some separately and some together—and used identical protocols (although the semistructured nature of the protocol allowed for some tailoring to the respondent). Independent assessments of thematic and conceptual content were conducted and then compared; some minor discrepancies were then resolved. 9. This merged data set has a similar breakdown to the full samples from which it is drawn: metro, 25 percent; suburb, 56 percent; and rural, 19 percent. 10. Because we are most interested in the relationship between competition levels and contracting choice, when a service involved mixed delivery (public delivery and contracting) we counted it with the contracting option. However, 9

553–77. Bel, Germà, and Melania Mur. 2009. Intermunicipal Cooperation, Privatization and Waste Management Costs: Evidence from Rural Municipalities. Waste Management 29(10): 2772–78. Bel, Germà, and Mildred E. Warner. 2008. Does Privatization of Solid Waste and Water Services Reduce Costs? A Review of Empirical Studies. Resources, Conservation and Recycling 52(12): 1337–48. Boyne, George A. 1998. Bureaucratic Theory Meets Reality: Public Choice and Service Contracting in U.S. Local Government. Public Administration Review 58(6): 474–84. Brown, Trevor L., and Matthew Potoski. 2003. Transaction Costs and Institutional Explanations for Government Service Production Decisions. Journal of Public Administration Research and Theory 13 (4): 441–68. ———. 2004. Managing the Public Service Market. Public Administration Review 64(6): 656–68. ———. 2005. Transaction Costs and Contracting: The Practitioner Perspective. Public Performance and Management Review 28(3): 326–51. Brown, Trevor L., Matthew Potoski, and David Van M. Slyke. 2010. Contracting for Complex Products. Supplement, Journal of Public Administration Research and Theory 20: 41–58. Carr, Jered B., Kelly LeRoux, and Manoj Shrestha. 2009. Institutional Ties, Transaction Costs, and External Service Production. Urban Affairs Review 44(3): 403–27. Ceriani, Lidia, Raffaele Doronzo, and Massimo Florio. 2009. Privatization,

percent of the cases involved combinations of two or more forms of contracting

Unbundling, and Liberalization of Network Industries: A Discussion of the

(combo). Because we could not distinguish forms of contracting in these cases,

Dominant Policy Paradigm in the EU. Department of Economics, Management,

they were dropped from subsequent analysis, leaving us with a final sample of 3,973 cases of service delivery. 11. This has been recently tested in a multivariate framework; see Hefetz and Warner (2012). 12. Hospitals are an interesting case. Half of the cases are competitive and half are monopoly, but for-profit delivery is more common in monopoly markets, whereas competitive markets are more dominated by nonprofit delivery. 13. If managers are “chasing competition,” and if their jurisdiction does not have adequate contract management capacity, one implication is that opportunity costs may be created by the diversion of scarce administrative resources to market management and away from oversight. 14. Similar dynamics have emerged recently in the area of prison management, with decisions to continue outsourcing despite higher costs (see Oppel 2011). 15. Johnston and Girth (2011) found that 53 percent of surveyed public managers reported that they seek additional vendors during active contracts for the next outsourcing cycle. Some 15 percent indicated that bidders buy each other out postcontract, and 21 percent reported that bidder supply for specific services decreases over time.

and Quantitative Methods Working Paper no. 09-2009, University of Milan. Davies, Steve. 2007. Politics and Markets: The Case of UK Municipal Waste Management. Working Paper no. 95, School of Social Sciences, Cardiff University. DeHoog, Ruth Hoogland. 1990. Competition, Negotiation, or Cooperation: Three Models for Service Contracting. Administration & Society 22(3): 317–40. Deloitte. 2005. Calling a Change in the Outsourcing Market: The Realities for the World’s Largest Organizations. http://www.syllogos.gr/links/docs/ OutsourcingStudyRV1.pdf [accessed July 26, 2012]. Dijkgraaf, Elbert, and Raymond Gradus. 2008. Institutional Developments in the Dutch Waste Collection Market. Government and Policy 26(1): 110–26. Donahue, John D. 1989. The Privatization Decision: Public Ends, Private Means. New York: Basic Books. Durant, Robert F., Amanda M. Girth, and Jocelyn M. Johnston. 2009. American Exceptionalism, Human Resource Management, and the Contract State. Review of Public Personnel Administration 29(3): 207–29. Feiock, Richard C., and Hee Soon Jang. 2009. Nonprofits as Local Government Service Contractors. Public Administration Review 69(4): 668–80. Fernandez, Sergio. 2007. What Works Best When Contracting for Services? An Analysis of Contracting Performance and the Local Level in the U.S. Public

References Allen, Barbara Ann C., and J. Bruce Walker. 2007. Evolution of Local Government Supply Markets: Evidence from an Evaluation. Paper presented at the Biannual Meeting of the Public Management Research Association, Tucson, AZ, October 25. 898

Public Administration Review • November | December 2012

Administration 85(4): 1119–41. ———. 2009. Understanding Contracting Performance: An Empirical Analysis. Administration & Society 41(1): 67–100. Ferris, James M., and Elizabeth Graddy. 1986. Contracting Out: For What? With Whom? Public Administration Review 46(4): 332–44.

———. 1991. Production Costs, Transaction Costs, and Local Government Contractor Choice. Economic Inquiry 29(3): 541–54. Fossett, James W., Malcolm Goggin, John S. Hall, Jocelyn Johnston, L. Christopher Plein, Richard Roper, and Carol Weissert. 2000. Medicaid Managed Care: Are States Becoming Prudent Purchasers? Health Affairs 19(4): 39–49. Graddy, Elizabeth A., and Bin Chen. 2006. Influences on the Size and Scope of Networks for Social Service Delivery. Journal of Public Administration Research and Theory 16(4): 533–52. Greene, Jeffrey D. 2002. Cities and Privatization: Prospects for the New Century. Upper Saddle River, NJ: Prentice Hall. Hansen, Janna J. 2003. Limits of Competition: Accountability in Government Contracting. Yale Law Journal 112(8): 2465–2507. Hebdon, Robert, and Patrice Jalette. 2008. The Restructuring of Municipal Services: A Canada–United States Comparison. Environment and Planning C: Government and Policy 26(1): 144–58. Hefetz, Amir, and Mildred E. Warner. 2004. Privatization and Its Reverse: Explaining the Dynamics of the Government Contracting Process. Journal of Public Administration Research and Theory 14(2): 171–90. ———. 2007. Beyond the Market vs. Planning Dichotomy: Understanding Privatization and Its Reverse in U.S. Cities. Local Government Studies 33(4): 555–72. ———. 2012. Contracting or Public Delivery? The Importance of Service, Market and Management Characteristics. Journal of Public Administration Research and Theory 22(2): 289–317. Hefetz, Amir, Mildred E. Warner, and Eran Vigoda-Gadot. Forthcoming. Privatization and Inter-Municipal Contracting: U.S. Local Government Experience 1992–2007. Environment and Planning C: Government and Policy 30(4): 675–692. ———. 2012. Concurrent Sourcing in the Public Sector: Contracting and Agent Strategies to Manage Risk. Paper presented at workshop, Neither Public Nor Neither Public nor Private: Mixed Forms of Service Delivery around the Globe, Barcelona, Spain, May 17–18, 2012. Heinrich, Carolyn J. 2010. Third-Party Governance under No Child Left Behind: Accountability and Performance Management Challenges. Supplement, Journal of Public Administration Research and Theory 20: 59–80. Hirsch, Werner Z. 1995. Contracting Out by Urban Governments: A Review. Urban Affairs Review 30(3): 458–72. Hodge, Graeme. 2000. Privatization: An International Review of Performance. Boulder, CO: Westview Press. Joassart-Marcelli, Pascale, and Juliet Musso. 2005. Municipal Service Provision Choices within a Metropolitan Area. Urban Affairs Review 40(4): 492–519. Johnston, Jocelyn M., and Amanda M. Girth. 2011. Assessing the Costs of Contracting in Noncompetitive Markets. Paper presented at the Association for Public Policy and Management Fall Research Conference, Washington, DC, November 3. ———. 2012. Government Contracts and “Managing the Market”: The Implications of Strategic Management Responses to Weak Vendor Competition. Administration & Society 44(1): 3–29. Johnston, Jocelyn M., and Barbara S. Romzek. 2008. Social Welfare Contracts as Networks: The Impact of Networks Stability on Management and Performance. Administration & Society 40(2): 115–46. Johnston, Jocelyn M., Barbara S. Romzek, and Curtis H. Wood. 2004. The Challenges of Contracting and Accountability across the Federal System: From Ambulances to Space Shuttles. Publius: The Journal of Federalism 34(3): 155–82. Kelman, Steven J. 2002. Contracting. In The Tools of Government: A Guide to the New Governance, edited by Lester M. Salamon, 282–318. New York: Oxford University Press. Kettl, Donald F. 2002. The Transformation of Governance: Public Administration for Twenty-First Century America. Baltimore: Johns Hopkins University Press.

Kodrzycki, Yolanda K. 1994. Privatization of Local Public Services: Lessons for New England. New England Economic Review, May/June, 31–46. Lamothe, Meeyoung, and Scott Lamothe. 2010. Competing for What? Linking Competition to Performance in Social Service Contracting. American Review of Public Administration 40(3): 326–50. Lamothe, Scott, Meeyoung Lamothe, and Richard C. Feiock. 2008. Examining Local Government Service Delivery Arrangements over Time. Urban Affairs Review 44(1): 27–56. Leland, Suzanne, and Olga Smirnova. 2009. Reassessing Privatization Strategies 25 Years Later: Revisiting Perry and Babitsky’s Comparative Performance Study of Urban Bus Transit Services. Public Administration Review 69(5): 855–67. Levin, Jonathan, and Steve Tadelis. 2010. Contracting for Government Services: Theory and Evidence from U.S. Cities. Journal of Industrial Economics 58(3): 508–41. Lowery, David. 1982. The Political Incentives of Government Contracting. Social Science Quarterly 63(3): 517–29. ———. 1998. Consumer Sovereignty and Quasi-Market Failure. Journal of Public Administration Research and Theory 8(2): 137–72. Marvel, Mary K., and Howard P. Marvel 2007. Outsourcing Oversight: A Comparison for In-House and Contracted Services. Public Administration Review 67(3): 521–30. Milward, H. Brinton, and Keith G. Provan. 1998. Principles for Controlling Agents: The Political Economy of Network Structure. Journal of Public Administration Research and Theory 8(2): 203–21. ———. 2000. Governing the Hollow State. Journal of Public Administration Research and Theory 10(2): 359–79. Mohr, Robert, and Steven C. Deller. 2010. Alternative Methods of Service Delivery in Small and Rural Municipalities. Public Administration Review 70(6): 894–905. Moore, Adrian T. 2004. Don’t Believe The Hype: Successful Water Privatization Is the Norm. Privatization Watch 28(2): 6–13. Nelson, Michael A. 1997. Municipal Government Approaches to Service Delivery: An Analysis from a Transactions Cost Perspective. Economic Inquiry 35(1): 82–96. O’Harrow, Robert, Jr., and Scott Higham. 2007. Changes Spurred Buying, Abuses: Taxpayers Overcharged Millions in Sun Deal, Auditor Says. Washington Post, May 23. Oppel, Richard A., Jr. 2011. Private Prisons Found to Offer Little in Savings. New York Times, May 18. Osborne, David, and Ted Gaebler. 1992. Reinventing Government: How the Entrepreneurial Spirit Is Transforming the Public Sector. New York: Penguin. Pack, Janet Rothberg. 1987. Privatization of Public-Sector Services in Theory, Practice. Journal of Policy Analysis and Management 6(4): 523–40. ———. 1989. Privatization and Cost Reduction. Policy Sciences 22(1): 1–25. Parks, Roger B., and Ronald J. Oakerson. 2000. Regionalism, Localism and Metropolitan Governance: Suggestions from the Research Program on Local Public Economies. State and Local Government Review 32(3): 169–79. Prager, Jonas. 1994. Contracting out Government Services: Lessons from the Private Sector. Public Administration Review 54(2): 176–84. Ramesh, M., Eduardo Araral, Jr., and Xun Wu, eds. 2010. Reasserting the Public in Public Services: New Public Management Reforms. New York: Routledge. Romzek, Barbara S., and Jocelyn M. Johnston. 2005. State Social Services Contracting: Exploring the Determinants of Effective Contract Accountability. Public Administration Review 65(4): 436–47. Savas, Emanuel S. 2000. Privatization and Public–Private Partnerships. New York: Seven Bridges Press. ———. 2002. Competition and Choice in New York City Social Services. Public Administration Review 62(1): 82–91. Schlesinger, Mark, Robert Dorward, and Richard Pulice. 1986. Competitive Bidding and States’ Purchase of Services: The Case of Mental Health Care in Massachusetts. Journal of Policy Analysis and Management 5(2): 245–63.

Outsourcing Public Service Delivery: Management Responses in Noncompetitive Markets 899

Sclar, Elliott D. 2000. You Don’t Always Get What You Pay For: The Economics of Privatization. Ithaca, NY: Cornell University Press. Stein, Robert M. 1990. Urban Alternatives: Public and Private Markets in the Provision of Local Services. Pittsburgh, PA: University of Pittsburgh Press. ———. 1993. Arranging City Services. Journal of Public Administration Research and Theory 3(1): 66–92. Thurmaier, Kurt, and Curtis Wood. 2002. Interlocal Agreements as Overlapping Social Networks: Picket-Fence Regionalism in Metropolitan Kansas City. Public Administration Review 62(5): 585–98. U.S. Government Accountability Office (GAO). 2006. Highlights of a U.S. GAO Forum: Federal Acquisition Challenges and Opportunities in the 21st Century. Washington, DC: U.S. Government Printing Office. GAO-07-45SP. Van Slyke, David M. 2007. Agents or Stewards: Using Theory to Understand the Government-Nonprofit Social service Contracting Relationship. Journal of Public Administration Research and Theory 17(1): 157–87. Warner, Mildred E. 2006. Market-Based Governance and the Challenge for Rural Governments: U.S. Trends. Social Policy and Administration: An International Journal of Policy and Research 40(6): 612–31. ———. 2009. Civic Government or Market-Based Governance? The Limits of Privatization for Rural Local Governments. Agriculture and Human Values 26(1): 133–43. ———. 2011. Competition or Cooperation in Urban Service Delivery? Annals of Public and Cooperative Economics 82(4): 421–35. Warner, Mildred E., and Germà Bel. 2008. Competition or Monopoly? Comparing Privatization of Local Public

900

Public Administration Review • November | December 2012

Services in the U.S. and Spain. Public Administration 86(3): 723–35. Warner, Mildred E., and Robert Hebdon. 2001. Local Government Restructuring: Privatization and Its Alternatives. Journal of Policy Analysis and Management 20(2): 315–36. Warner, Mildred E., and Amir Hefetz. 2002a. Applying Market Solutions to Public Services: An Assessment of Efficiency, Equity and Voice. Urban Affairs Review 38(1): 70–89. ———. 2002b. The Uneven Distribution of Market Solutions for Public Goods. Journal of Urban Affairs 24(4): 445–59. ———. 2003. Rural–Urban Differences in Privatization: Limits to the Competitive State. Environment and Planning C: Government and Policy 21(5): 703–18. ———. 2008. Managing Markets for Public Service: The Role of Mixed Public/Private Delivery of City Services. Public Administration Review 68(1): 155–66. ———. Forthcoming. Insourcing and Outsourcing: The Dynamics of Privatization among US Municipalities 2002–2007. Journal of the American Planning Association. Weimer, David L., and Aidan R. Vining. 2005. Policy Analysis. 4th ed. Upper Saddle River, NJ: Pearson. Whittington, Jan. Forthcoming. When to Partner for Public Infrastructure? Transaction Cost Evaluation of Design-Build Delivery. Journal of the American Planning Association. Williamson, Oliver E. 1991. Comparative Economic Organization: The Analysis of Discrete Structural Alternatives. Administrative Science Quarterly 36(2): 269–96. ———. 1999. Public and Private Bureaucracies: A Transaction Cost Economics Perspective. Journal of Law, Economics, and Organization 15(1): 306–42.