Managing Local Government Service Delivery: Harnessing the Benefits of Contracting

 

 

Trevor L Brown

School of Public Policy and Management

Ohio State University

Columbus, OH 43210

brown.2296@osu.edu

(614) 292-4533

 

Matthew Potoski

Department of Political Science

Iowa State University

Ames, IA 50011

potoski@iastate.edu

(515) 294-2935

 

 

Contracting and Contract Management

Under pressure to do more with less, governments across the country have increasingly moved from direct service provision to providing services via contract.  Citizens today receive public goods and services not only from their general service local governments, but also from a variety of vendors working under contract, including for-profits, non-profits and government agencies from other jurisdictions.  Of course, there are many accounts of successful contracting.  Advocates of such alternative service delivery arrangements promote competitive contracting with promises of efficiency, cost savings, and improved effectiveness (Ferris and Graddy, 1991; Ostrom and Ostrom, 1977; Stein, 1990).  At the same time, the academic literature and popular press are fraught with accounts of contracting disasters where governments have been badly burned by unscrupulous vendors.  Citing a growing number of incomplete, failed, and corrupt contract arrangements, critics argue that contracting creates numerous accountability problems, sacrifices service quality for efficiency and cost savings, often does not result in improved efficiency and cost savings, and ultimately “hollows” the state (deLeon and Denhardt, 2000; Milward, 1996; Milward, Provan, and Else, 1994).

Missing in this debate is rigorous analysis of the capacity of both governments and contract vendors to manage the contracting process. Contracting is not a “one size fits all” proposition. The success or failure of any alternative service delivery arrangement likely depends on how well governments are able to manage the entire process, from assessing the feasibility of contracting for particular services, through implementing contract arrangements, to monitoring and evaluating service outcomes and vendor performance – activities that require strong government management capacity.  When governments can easily and quickly write very detailed contracts describing exactly what actions the vendor should take and what outcomes the vendor should achieve, the risks of contract failure are reduced, and consequently the costs inherent in negotiating, implementing and monitoring a contract relationship are low. However, real world complexities and uncertainties in social interaction often exceed governments’ ability to predict future events, specify contract provisions for all circumstances, and ensure that actual outcomes match defined objectives. 

A fundamental decision confronting all organizations is choosing whether to internalize production (i.e. to make it) or to externalize production (i.e. to buy it through contracting). Transaction costs scholarship argues that organizations’ production choices reflect the relative costs of traditional production factors – the costs of fixed assets, labor, and capital – and transaction costs – the “…comparative costs of planning, adapting, and monitoring task completion under alternative governing structures (Williamson 1981: 552-553).”  Transaction costs are essentially the management costs associated with either internally producing the service or buying it through contracting (Coase, 1937; Williamson, 1981, 1996).  Transaction costs are often higher under contracted service provision. For example, for many public services it is difficult to measure service outcomes (i.e. fire prevention) and therefore specify how contract vendors will be evaluated. According to transaction costs logic, governments should internalize high transaction costs services; if governments decide to contract in pursuit of efficiencies and cost savings, they will need to implement high transaction costs oversight and management procedures to mitigate associated contracting risks.  Transaction costs theory provides a framework for analyzing which services should be contracted and how to manage contracts to improve contract success.

In our previous work (Brown and Potoski 2003a, 2003b, 2003c), we began to address these issues.  We synthesized transaction costs principles with insights from theories of government organization and urban development scholarship.  Theories of government organization frame relations among government agencies, elected officials and constituents (Bawn 1995, Epstein and O’Halloran 2000).  The urban development literature (i.e. Stein, 1990) argues that governments have more choices about how to structure contracts when the market of potential service providers is robust. For example, governments located in large metropolitan areas are more likely to contract for refuse collection simply because they have more opportunities to do so.  First, we investigated the factors that influence governments’ service delivery choices (Brown and Potoski, 2003a).  Our analysis of the service delivery practices of over 1,500 local governments for 64 different services shows that governments select service production options to address the variable transaction costs risks inherent in the service marketplace and in the particular service to be delivered.  When governments contract for highly asset specific services, those that tend to be offered through monopoly-dominated markets, they choose mechanisms that lower the risk of vendor opportunism – namely contracting with other governments.  When service outcomes are difficult to measure, governments decrease their reliance on contracts with private firms and increase their use of joint contracting and internal service production. 

Second, we examined how municipal governments manage their capacity to deliver services under different production arrangements (Brown and Potoski, 2003b).  Effective contract management requires mitigating specific problems that can plague the contract process.  These problems arise from dissatisfaction with prior contracting experiences; transaction costs; characteristics of the government’s structure and operation; and characteristics of the governments’ external environment.  Third, we analyzed how governments use four contract oversight procedures – monitoring citizen complaints, implementing citizen satisfaction surveys, analyzing vendor performance data, and auditing vendor activities in the field – to help better monitor vendor performance and thus mitigate the risk of contract failure (Brown and Potoski, 2003c). 

Case Study Summaries: Police Service and Refuse Collection in Columbus Ohio

Building on this empirical research, we have begun to explore service delivery practices up close through a series of case studies examining relations among citizens, governments, and service vendors.  Over the past year, we have been studying two service areas – refuse collection and policing – in nine communities in the Columbus, Ohio metropolitan area.  These communities are:  Columbus, Dublin, Gahanna, Grove City, Hilliard, Reynoldsburg, Upper Arlington, Westerville, and Worthington.  Our case studies are based on semi-structured interviews with key informants (mainly police chiefs and public service directors) supplemented with information collected from publicly available sources (e.g. copies of all contracts).  In our interviews, we asked these respondents questions about their service delivery practices, including whether they contract for particular functions and what steps they take to manage those contracts. 

In the logic of transaction costs scholarship, refuse collection with relatively easily measured outcomes and outputs and assets that are relatively unspecific, is a prime candidate for successful contracting, particularly in a larger metropolitan area where vendors compete for contracts. Our case analyses so far indicate that this is the case for most suburban governments. Only two governments in our case sample – Columbus and Upper Arlington – do not contract for refuse collection.  Interviews with Public Service directors in the other cities indicate that the primary reason governments contract for refuse collection is cost savings. The fixed costs of maintaining a fleet of refuse trucks are prohibitive for these seven suburban governments (all have populations around 30,000).  Columbus, on the other hand, as a city of over 750,000 can achieve of economies of scale and consequently fund their own service provision.  Analysis of the contracts reveals similar practices among the contracting governments.  In fact, many of these governments have worked in collaboration through the Franklin County Solid Waste Authority to craft essentially the same contract.  Where the communities differ in is the degree to which they engage in oversight.  The wealthier communities typically have a full time staff member who dedicated a substantive portion of their working week to overseeing the refuse collection process (i.e. tracking citizen complaints; communicating with the vendor), while the less well off communities lacked the resources and the capacity to engage in this level of oversight.  This jives with our empirical results about the role that government capacity plays in the degree to which governments engage in monitoring and oversight.  The interviews also revealed that public service directors – particularly those who had been in their position for awhile and in a community that had been through several rounds of contracting – were fairly savvy about reading market conditions and understanding the transaction costs risks associated with contracting.  Many public service directors not only kept close watch of the number and economic health of local refuse collectors, but also followed the national refuse collection market.  When probed, several service directors understood that if the market became monopolized they would have to court vendors from outside the Columbus metropolitan area to compete.

Policing poses more severe management challenges with difficult to measure outcomes and assets that are relatively specific.  None of the police departments in the case sample contract for core policing services (i.e. traffic control, investigation), although many outsource some of their peripheral functions (i.e. forensic analysis, animal control).  When asked about whether financial pressures existed that might stimulate the expansion of contracting for policing tasks, the interviewed police chiefs typically indicated that they were largely buffered from budgetary issues.  This is not to say that police chiefs were not sensitive to cost concerns; on the contrary all could list several steps they had taken to reduce costs.  Rather, they all indicated that policing was such a central concern in each community that most departments had been spared severe budget reductions relative to other city services.  In addition, most noted that there was not really a market for the services they performed in the metropolitan area.  When they did acknowledge areas where they lacked capacity, the default means of addressing the deficit is to enter into a mutual aid agreement with a neighboring police department.  For example, many suburban police departments have mutual aid agreements with the Columbus police department for use of their SWAT services.

Taken together, our interviews with public service directors regarding refuse collection – a service with fewer transaction costs risks – and with police chiefs regarding core policing services – a service with considerable transaction costs risks – suggest that a quartet of factors influence whether governments decide to contract, and if they do, what types of steps they take to manage the contract process.  These factors are similar to those that we discovered in our empirical analysis and are described below.

Fiscal Imperatives

The genesis for most contracting arrangements is the drive to reduce costs.  The combination of expensive service related costs, economies of scale, and community cost pressures often fuel the drive to seek alternatives to direct service provision.

Service-Specific Contracting Risks

            Some services are more risky contract targets than others.  Services that require highly asset specific investments, such as buying specialized equipment or developing service specific technologies, are better delivered in-house by governments than by contract.  Likewise, services whose outcomes and outputs are more difficult to measure are also riskier contract targets. 

Market Conditions

            Successful contracting requires a robust market of competitors.  Competition need not come only from private firms, but may also include non-profits and other governments.  In such contexts, governments can compare prices and practices across vendors while competitive pressures discipline contract winners. 

Government Capacity

            Managing contracts requires attention across the service delivery process, from negotiating and developing contracts, to monitoring service delivery outputs, to ensuring outcomes match specified objectives.  To mitigate the risk of contract failure, governments can adopt procedures to improve contract management, such as mechanisms for reviewing and correcting vendors’ performance during the course of service delivery. 

Next Steps: Listening to Vendors

            Previous research on alternative service delivery has tended to neglect the perspective of “street-level” managers who actually produce the service.  We believe this perspective is vital for a complete understanding of the service delivery process.  For example, in our first round of interviews, we noticed that several managers reported consistent themes in their contracting experience. At first, these managers were quite dissatisfied with their vendor’s performance.  According to these managers, the vendors’ performance did not improve, even in the face of corrective measures, until they threatened terminating the contract and replacing the vendor.  An obvious omission from this account is the vendors’ perspective: how did they view their own performance, their relations with the government, and how these changed over time?  Consequently, our next step is to interview service vendors.  This means interviewing service vendors in the event of contracting, and comparing their experiences with those of managers engaged with internal service production in other jurisdictions. 

In the area of refuse collection we will interview the area supervisors and contract managers for each of the vendors to investigate how they have responded to steps service directors have taken to improve contract performance. In particular, the interviews illicit information about whether different contract arrangements provide refuse vendors with more flexibility over other arrangements. For example, in latest round of contracting, some of the public service directors have added penalty clauses to their town’s contracts.  Consequently, we will ask refuse vendors about whether they manage their contract relations with communities that impose penalty clauses differently than those than do not impose penalties. 

For police services, interviews with the suburban police chiefs indicate that there are two areas where they outsource – forensics and jailing.  All of the governments send offenders in their communities to the county jail of the county in which the illegal activity takes place. The county jails charge each community a per diem rate for prisoners.  Most of the suburban communities maintain small holding facilities, although by law they are not jails. However, because of rising per diem costs, some have considered converting their holding facilities into legal jails.  Interviews with representatives from the respective county jails will provide insight into how the per diem rates are calculated and whether increases in these rates are designed to stimulate suburban communities to handle more of their own jailing responsibilities.  For forensics, all of the suburban communities rely on either the City of Columbus’ crime lab or the Ohio State Bureau of Criminal Investigation (BCI) for forensic analysis (i.e. DNA analysis, finger print analysis, ballistics).  Suburban governments pay a fee for Columbus’ services, but receive BCI’s services free of charge.  This allows us to compare the delivery of a free service versus a paid service.  Interviews with Columbus crime lab and BCI employees will focus on whether there are differences in how the two organizations perceive the forensic services they deliver and whether there are any differences in the quality of forensic analysis. Do Columbus crime lab employees see the forensic analysis they provide as a commodity and revenue-generating instrument?  Do Columbus crime lab employees focus more on the quality of the service provided relative to BCI employees because police departments pay for these services? 

Taken together these interviews will map out the service marketplace in each policy area more completely.  At the moment, our primary view is through the lens of the purchaser – the government procuring the service.  In order to get a better picture of the service marketplace and how it affects the delivery of these two public services, we require corresponding views from the vendors.  Our overall objective is to build the case studies and then integrate them with our survey based research, thus focusing multiple analytic lenses on the full service delivery process, from feasibility, through implementation, to outcome assessment.

 

References

Bawn, Kathleen. 1995. Political Control Versus Agency Expertise: Congressional Choices about Administrative Procedures. American Political Science Review 89: 62-73

Brown, Trevor and Matthew Potoski. 2003a. The Influence of Transactions Costs on Municipal and County Government Choices of Alternative Modes of Service Provision. Forthcoming in the Journal of Public Administration Research & Theory.

Brown, Trevor and Matthew Potoski.  2003b. Contract Management Capacity in Municipal and County Governments. Public Administration Review 63 (2): 136-147.

Brown, Trevor and Matthew Potoski.  2003c. “Managing Contract Performance: A Transaction Costs Approach,” Journal of Policy Analysis and Management 22 (2): 275-297. 

Coase, Ronald.  1937 “The Nature of the Firm.”  Economica, 4: 386-405.

DeLeon, Linda and Robert A. Denhardt.  2000.  The Political Theory of Reinvention.  Public Administration Review 60 (2): 89-97.

Epstein, David and Sharyn O’Halloran 2000. Delegating Powers: A Transaction Cost Politics Approach to Policymaking Under Separate Powers. Cambridge: Cambridge University Press.

Ferris, James M. and Elizabeth Graddy. 1991.  Production Costs, Transaction Costs, and Local Government Contractor Choice.  Economic Inquiry 24 (July): 541-554.

Huber, John D. and Charles R. Shipan. 2002.  Deliberate Discretion: The Institutional Foundations of Bureaucratic Autonomy.  New York: Cambridge University Press.

Williamson, Oliver. 1981. The Economics of Organization. American Journal of Sociology 87: 548-577.

Williamson, Oliver. 1996. The Mechanisms of Governance. New York: Oxford University Press.

Milward, H. Brinton. 1996.  Symposium on the Hollow State: Capacity, Control and Performance in Interorganizational Settings. Journal of Public Administration Research and Theory 6 (2):193-95.

Milward, H. Brinton, Keith G. Provan, and Barbara A. Else.  1993.  What Does the Hollow State Look Like? In Public Management: The State of the Field, edited by Barry Bozenman.  San Francisco: Jossey-Bass.

Ostrom, Elinor and Vincent Ostrom. 1977. Public Goods and Public Choices. In Alternatives for Delivering Public Services, edited by E.S. Savas. Boulder, CO: Westview Press.

Stein, Robert M. 1990.  Urban Alternatives: Public and Private Markets in the Provision of Local Services.  Pittsburgh: University of Pittsburgh Press.