Like many aspects of governmental functions that have undergone reform over the past two decades, the traditional ways of achieving regulatory goals have given way to experimentation with a variety of innovations. These include voluntary approaches under which regulators work with industry associations in developing codes of practice (Lyon & Maxwell 2001), self-auditing that entails assessments of compliance by regulated entities or third parties (Potoski & Prakash 2004), management-based systems that entail firm responsibility for adhering to plans that limit regulated harms (Coglianese & Lazer 2003), and performance-based approaches that emphasize regulation for results rather than specification of specific actions or technologies (Coglianese et al. 2003).
None of these reforms has wholly or even widely supplanted traditional regulation that emphasizes enforcement of rules by governmental agencies and penalties for noncompliance with the rules. Nonetheless, widespread interest in the new approaches is evidenced by the adoption by various governments of one or more of these innovations as part of the regulation of air and water quality, building and fire safety, consumer product safety, energy efficiency, food safety, forest practices, nuclear power plant safety, pipeline safety, railroad safety and worker safety.
Although much of the academic commentary about these approaches has focused on their rationale and design (e.g. Gunningham & Grabosky 1998; Fiorino 2001; Gunningham & Sinclair 2002; Organization for Economic Co-operation and Development 2003), a more basic set of issues concerns the implications of regulatory innovations for governance. In particular, a central feature of the newer approaches is a shift in key regulatory responsibilities from governmental regulators to nongovernmental actors. This shift, in turn, raises fundamental issues concerning regulatory accountability. Accountability is a key concept for democratic governance broadly concerned with holding officials responsible for their actions (more generally see Behn 2001; Gormley & Balla 2004; Mashaw 2006). At issue is how that accountability is achieved when nongovernmental actors assume important roles in regulatory regimes.
This article addresses accountability issues for system- and performance-based regulatory regimes. As with prescriptive regulation, the newer regimes invoke aspects of legal, bureaucratic, professional and political accountability. However, they differ in their emphasis on different levels of accountability and in the challenges that they pose for ensuring accountability. Case studies are presented of system-based approaches to food safety and nuclear power plant safety and of performance-based approaches to building and fire safety regulations. These cases illustrate potential accountability shortfalls for the newer regulatory regimes and the potential for subtle forms of regulatory capture. Although the cases are selective in stressing different accountability issues for limited settings, they underscore the importance of finding the right fit between regulatory circumstances – in terms of types of firm, complexity of tasks, potential for harm and so on – and the design of regulatory regimes.
One can think of a regulatory regime as a means for achieving regulatory goals (see Hood et al. 2001; May 2002; Lodge 2004). A regime comprises an institutional structure and assignment of responsibilities for carrying out regulatory actions. The institutional structure is made up of rules that prescribe expected behaviors or outcomes, standards that are benchmarks against which compliance can be measured, a mechanism for determining the degree of regulatory compliance, and sanctions for failure to comply with the rules.
Given the variety of ways that rules and standards can be crafted and responsibilities for regulatory actions can be assigned, there is no definitive categorization of regulatory regimes. Comparisons of the regulatory regimes that are the foci of this article are presented in Table 1. This draws a contrast between traditional emphases on prescriptive regulation with newer forms of system-based and performance-based regulations. Each of these regimes entails a governmental role in setting forth regulations and enforcing them. They differ, however, with respect to the nature of the rules and standards and the means for gauging adherence to them.
Table 1. Comparison of regulatory regimes
|Regulatory foci||Prescribed actions||Process or system||Results or outcomes|
|Compliance determination||Adherence to prescribed actions||Acceptable production system||Achievement of desired results|
|Nature of rules and standards||Particularistic and detailed specifications||Process-oriented specifications||Goal-oriented outcome specifications|
|Basis for achieving regulatory goals||Adherence to prescriptions presumed to meet goals||Appropriate system controls are designed to meet goals||Regulatory goals are embedded in the results orientation|
|Examples||Dominant regulatory approach in the USA for environmental and social regulation||Aspects of food safety, industrial health and safety, nuclear power plant safety||Aspects of air and water quality, building and fire safety, energy efficiency, forest practices, pipeline safety|
Many regulations are highly prescriptive in telling regulated entities and individuals what to do and how to do it. For example, traditional building code provisions covering the safety of buildings run to hundreds of pages and address such things as nailing patterns in support walls, spacing of beams, and roofing and insulation materials. As with building codes, prescriptive regulation tends to be highly particularistic in specifying required actions and standards for adherence to them. Regulatory enforcement for prescriptive regulation emphasizes adherence to the prescribed rules and standards, which in turn is presumed to provide acceptable outcomes in meeting regulatory goals. Critics argue that unreasonable regulations and capricious enforcement practices impose needless burdens on regulated entities. These factors have been important sources of discontent with prescriptive regulation and key motivations for seeking regulatory reform.
One direction for reform is the use of a system-based regime that has also been labeled as a process-based or a management-based regulation. The overall logic of this approach is prescriptive regulation falls short because the production systems of firms are too complicated to be able to effectively prescribe regulatory fixes. Instead, adherents of the system-based approach argue that regulatory goals can be achieved by instituting the appropriate systems for monitoring production processes by firms. Regulatory scholars Coglianese and Lazer (2003) discuss the use of this approach for food safety and industrial safety. They describe the approach as follows:
[F]irms are expected to produce plans that comply with general criteria designed to promote the targeted social goal. Regulatory criteria for management planning specify elements that each plan should have, such as the identification of hazards, risk mitigation actions, procedures for monitoring and correcting problems, employee training policies, and measures for evaluating and refining the firm’s management with respect to the stated social objective (p. 694).
Regulatory rules detail the elements of the plan. Compliance is determined by whether a firm has an acceptable plan, not on the basis of detailed adherence to prescriptions or outcomes of the production process. Deficiencies in regulatory outcomes are viewed as potential indicators of defects in the management plan.
A second direction for regulatory reform is to emphasize regulatory outcomes under what has been labeled performance-based regulation. This approach emphasizes regulation for results while leaving it to regulated entities to determine how best to achieve the desired results. The central change under the performance-based approach is the attention to outcomes as part of regulatory rules and standards. The regulatory focus is on whether the desired level of performance has been obtained as defined by the relevant performance-oriented rules and standards. As discussed by Coglianese et al. (2003), performance-based regulations differ in the generality of the performance objective, the extent to which they quantify performance, the extent to which they specify detailed outcomes, and the mechanisms for monitoring or predicting performance.
The notion that elected officials and their appointees are accountable to citizens for governmental performance is clearly a cornerstone of democratic governance. As noted by a number of public administration scholars (Mulgan 2000; Behn 2001; Gormley & Balla 2004), the concept of accountability has taken on a broader meaning as new forms of governance have arisen that involve increased reliance on nongovernmental actors as part of service delivery. The quest in these reforms is for better performance of public programs. But, accountability by itself cannot ensure that regulatory or other programs are effective because performance also depends on such things as the appropriateness of the policy design, the mix instruments that are used and quality of implementation. As articulated by Aucoin and Heintzman (2000), accountability concerns control of abuse of public authority, assurance that public resources are being appropriately used, and learning that facilitates pursuit of improvements. Stated differently, ensuring regulatory accountability might be considered a necessary but insufficient condition for increasing regulatory effectiveness.
Relatively little scholarship specifically addresses accountability in regulatory contexts. Scott (2000) applies a broader conceptualization of accountability to regulatory settings in observing that it must be considered a multilevel concept within a devolved regulatory state. Lodge (2004) further expands the notion of multiple forms of regulatory accountability in discussing different instruments for improving regulatory transparency and accountability. Mashaw (2006) discusses different institutional designs for enhancing accountability in drawing a distinction among accountability based on governance, markets and social accountability. These notions, along with distinctions in types of accountability suggested by Barbara Romzek and her colleagues (Romzek & Dubnick 1987; Romzek & Ingraham 2000), provide a foundation for discussion of different levels of and issues for regulatory accountability.
Levels of accountability
The most basic level is the accountability of those who promulgate regulations with respect to the content of the regulatory provisions. This level of accountability is labeled legal accountability in what follows. The relevant issues are the fairness and appropriateness of rules and standards. The potential for regulatory capture in which some firms or industries benefit from regulations at the expense of the broader public interest is a primary concern in the design of regulations (see Wilson 1980; Ayres & Braithwaite 1992). Another concern, as popularized in Philip Howard’s book, The Death of Common Sense (Howard 1994), is the reasonableness of rules and standards (see also Bardach & Kagan 1982).
The presumption is that regulatory processes that are more transparent help to avert these problems. In the USA, primary responsibility for rule-making is delegated by elected officials to regulatory agencies who are held accountable for their decisions in designing regulations through administrative legal processes. Lodge (2004) cites a variety of mechanisms for increasing the transparency of regulatory rule making that include various forms of citizen empowerment, public representatives in regulatory rule making and enhanced oversight of regulatory processes.
A second level of accountability is answerability in the implementation of regulatory provisions. This entails accountability of both regulators and of regulated entities. A key concern is that regulators do not abuse their authority by inspectors enforcing regulations in a capricious way or by other misuses of power. Another concern is that regulated entities comply with regulatory requirements and are held accountable for noncompliance. The traditional accountability mechanism for achieving both ends is a set of controls that are labeled here as bureaucratic accountability (see Behn 2001; Gormley & Balla 2004). Checklists for enforcement actions, reviews of enforcement decisions, and ratings of inspectors are bureaucratic controls that limit discretion of inspectors and guide their actions. The checklists, detailed reporting requirements, and other enforcement procedures are also bureaucratic controls for helping ensure that regulated entities have complied with regulatory provisions.
Recognition of the limits of bureaucratic controls has led regulatory scholars such as Bardach and Kagan (1982) and Sparrow (2000) to emphasize the importance of greater reliance on professional accountability for averting regulatory abuse and achieving regulatory ends. This shifts emphasis from elaborate controls over actions of regulators and regulated entities to appropriate exercise of professional judgment. Standards of conduct for police, fire and other professions along with codes of practice for industry groups are examples of mechanisms for enhancing professional accountability. Bureaucratic accountability operates on external controls whereas professional accountability rests more on internalized norms and peer pressure.
A third level of accountability is the responsiveness of elected officials to shortfalls in regulatory regimes. This is labeled political accountability in what follows. Lodge (2004) suggests that feedback about performance of regulatory systems is important for regulatory reform. The reverberations from regulatory effects are typically numerous, difficult to decipher, and involve a combination of oversight by elected officials, external review by public interest, trade groups and the media, and occasional political involvement. At issue is the ability and willingness of elected officials to learn about shortfalls in regulatory regimes and to make necessary adjustments in regulatory systems.
This review of levels of regulatory accountability helps to distinguish key issues for regulatory accountability as they apply to different levels of accountability. The illustrative instruments for achieving accountability at different levels noted here cannot be considered an exhaustive cataloging of means for enhancing accountability. Suffice it to note that public administration and regulatory scholars have examined a variety of accountability tools that include consideration of exit and voice options as feedback mechanisms (Lodge 2004), of institutional designs for regulatory oversight (Scott 2000), and of democratic accountability mechanisms for controlling actions of subordinate agents or trustees (Behn 2001; Gormley & Balla 2004).
The preceding discussion of differences in regulatory regimes and of accountability issues provides the foundation for considering how prescriptive, system-based and performance-based regimes differ in their accountability structures. The accountability structures for each regime are summarized in Table 2.
Table 2. Accountability structures
|Legal||Transparency in setting rules and standards||Transparency in establishing features of desired systems||Transparency in establishing performance goals|
|Bureaucratic||Monitoring for adherence to prescribed rules||Monitoring for adequacy of management system||Monitoring for adherence to performance goals|
|Professional||Enforcement decisions by regulatory inspectors||System design decisions by regulated entities||Adherence to performance goals by regulated entities|
|Political||Triggered by complaints about regulatory process||Triggered by multiple system breakdowns||Triggered by systemic undesired outcomes|
As discussed above, legal accountability concerns the transparency with which rules and standards are established in an effort to ensure their fairness and appropriateness. Transparency in administrative processes for setting rules and standards seeks to ensure these properties (more generally see Scott 2000). Cheit (1990, pp. 214–218) shows how the transparency of prescriptive regulation breaks down when standards developed by consensus-based private organizations (“reference standards”) are used instead of those developed through processes that are more open. The newer forms of regulation have much less reliance on private standard-setting and have generally been more open in their promulgation of desired systems under system-based regulation and for the establishment of performance goals and standards under performance-based regulation.
A different aspect of legal accountability is the appropriateness of rules and standards. Critics of prescriptive regulation like Howard (1994) argue that many prescriptive rules are unnecessary in that they have little to do with regulatory goals. System-based regulation seeks to avoid prescriptive details to allow greater flexibility in how regulated entities address potential harms. By focusing on regulatory goals, the performance-based approach seeks to avoid the regulatory pitfalls of inappropriate and unnecessary requirements.
The key issues in bureaucratic accountability are the answerability of regulators and regulated entities for the implementation of regulatory provisions. The myriad of requirements poses major enforcement and compliance challenges under prescriptive regulation (see Bardach & Kagan 1982; Sparrow 2000). Inspectors must decide what to enforce for which abuse of that discretion can lead to overly zealous enforcement (“going by the book”) and nitpicky enforcement. Regulated entities must decide what to comply with for which there is always the possibility of token compliance by meeting the minimum requirements on a checklist.
The roles and challenges for the front lines are very different under the other regulatory regimes. Inspectors no longer look for particular items to check off boxes that indicate compliance with prescriptions. Instead, they are charged with certifying the adequacy of systems or the adherence to regulatory goals. This requires a different type of expertise and different interactions with regulated entities and as such necessitates a cultural transformation of enforcement. One issue is the ability of enforcement personnel to gauge the quality of systems or the adherence to desired performance goals.
The transformation of enforcement roles stresses the role of professional accountability in the newer regulatory regimes. Systems-based and performance-based regulatory regimes give regulated entities primary responsibility for regulatory functions and to show adequate compliance. Traditional, prescriptive regulation emphasizes accountability through bureaucratic controls that seek compliance with prescribed actions. System-based and performance-based regulatory regimes relax these controls in favor of greater reliance on professional accountability. This is intentional as the newer regulatory regimes are designed to shift key regulatory functions from governmental regulators to regulated entities.
The final entries in Table 2 concern the ability and willingness of elected officials to learn about shortfalls in regulatory regimes. Regulatory programs typically only engender political attention when scandals arise or when there are major regulatory failures as reflective of the “fire alarm” perspective for legislative oversight (McCubbins & Schwartz 1984). This type of political accountability is potentially important in keeping regulatory excesses in check or in addressing gaps in regulatory provisions. The failures of a system-based approach are likely to come from repeated system breakdowns, and the failures of performance-based approaches are likely to come from systemic undesired outcomes.
Experiences with newer regulatory regimes
Four experiences with variants of system-based and performance-based regulations are considered in what follows with attention to accountability issues. As depicted in Table 3, the cases reflect different stages of development of system-based and performance-based regimes. The cases were selected to consider accountability challenges that arise at the formative stages (i.e. partial regime) as well as at the more mature stage (i.e. extensive regime) development of regulatory regimes. The cases are necessarily selective as they have been chosen from regulatory situations for which accountability issues were notable. Although the cases illustrate the potential for accountability shortfalls in system-based and performance-based regimes, they clearly do not suggest that such shortfalls are inevitable; a topic that is addressed more fully in the conclusions. Except for the discussion of performance-based approaches to building regulation in New Zealand, each of the experiences is within the USA. The New Zealand experience is especially noteworthy as it is the only case of a fully implemented performance-based regime that spans a whole sector of regulation.
Table 3. Comparative regulatory experiences
|Extensive||Food-safety regulation – HACCP management system||Building-safety regulation – performance-based building requirements in New Zealand|
|Partial||Nuclear power plant safety regulation –“risk-informed” safety management||Fire-safety regulation – performance-based code alternatives|
Several methods were used to document and validate the depictions of these experiences. A search of the secondary published work was undertaken to identify the context for each regulatory reform. Governmental reports, external reviews and the websites of relevant regulatory agencies provided a basis for describing the regulatory approaches and depicting implementation issues. Given the constraints of space and the continuing evolution of each regime, the depictions that follow can only be considered limited snapshots.
Food safety: Changed roles under system-based regulation
The issue of food safety was propelled onto the American governmental agenda with the 1993 Escherichia coli bacterial outbreak in Jack-in-the-Box restaurants in the state of Washington. Four children died and another 400 people became ill. This was not an isolated case relating to food safety. From the early 1990s to the present there have been major recalls of contaminated meats in the USA, a temporary ban by European countries on the import of British beef over fear of “mad cow” disease, similar bans on importing suspect tainted beef from Alberta, and a temporary ban on import of poultry from Belgium because of dioxin-contaminated feed. That these problems could arise in countries with well-developed systems for regulating food safety was all the more shocking to an unaware public.
Largely in response to the sensation created by the E.coli scare, the Clinton administration initiated an overhaul of the way in which meats and poultry are inspected in the USA. This resulted in a new state-of-the-art, science-based inspection system. This regulatory approach, Hazard Analysis and Critical Control Point (deemed HACCP), requires meat and poultry processors to identify potential sources of contamination within processing plants, to monitor those critical control points, to institute additional controls that are aimed at preventing contamination, and to test for the presence of E.coli and of Salmonella. The “poke and sniff” inspection regime dating to the Meat Inspection Act of 1906 did not adequately target and reduce microbial pathogens and consequently was literally hit or miss. After several years of rule making and commentary, the HACCP regulatory system for meat and poultry was introduced in 1997 with administration by the Food Safety and Inspection Service (FSIS) of the US Department of Agriculture (USDA).
As discussed by Coglianese and Lazer (2003), the HACCP systems approach exemplifies the system-based approach to regulation. The cornerstone of this is the identification by firms of potential food-safety hazards and critical control points in meat and poultry production and processing. A critical control point is a step or procedure where controls can be used to prevent, reduce to an acceptable level, or eliminate food-safety hazards. As part of the HACCP plan, plants must establish critical limits of a hazard for each critical control point. The HACCP regulatory approach transforms the burden of showing adequate food-safety controls to plant operators and substantially changes the role of inspectors.
Two key accountability issues are raised by the experience to date. One issue is the legal basis for assessing the quality of the system. Under HACCP, the adequacy of the system is certified by the firm as verified by FSIS review and through ongoing monitoring of system performance. Testing for the presence of pathogens by plant personnel and others by the FSIS is a primary bureaucratic control mechanism. The uncertain legality of these tests as indicators of HACCP system performance creates a potential gap in bureaucratic accountability. In one key ruling concerning Salmonella testing, the Fifth US Circuit Court of Appeals ruled that testing for the presence of Salmonella bacteria in raw meat could not be used to close plants that fail the tests (Supreme Beef Processors Inc. v. United States Department of Agriculture 2001). The Court found that such testing did not necessarily assess plant conditions or performance and thus the USDA could not use the results as a basis for enforcement actions against meat producers. This legal logic was also used in a preliminary injunction in 2003 by a Federal district court to prevent the closure of a Nebraska beef processor. Fearing an adverse final decision, the Department of Agriculture decided to settle the processor and agreed to let the processor continue operations with increased oversight.
As a consequence of these actions, the FSIS is left with weakened bureaucratic control mechanisms for monitoring systems performance that in turn increases reliance on certification and monitoring of their systems. FSIS regulators have taken the position that the legal rulings simply put more emphasis on identifying defects in plant systems as a basis for corrective actions, rather than on outcomes of those systems. However, as noted in a report by the US General Accountability Office (USGAO 2002), FSIS enforcement personnel have not consistently identified and failures of meat and poultry plant HACCP system failures.
The latter stresses the second accountability issue involving the shift from bureaucratic to professional accountability. This is most evident from the changed role of FSIS inspectors from emphasizing bureaucratic accountability through detailed inspection to emphasizing professional accountability of firms. Not surprisingly, the shift in roles engendered confusion as is evident with the remarks made by the FSIS administration in a 1998 memorandum to field in-plant personnel:
It is the responsibility of industry to identify potential hazards, to develop a HACCP plan containing controls to prevent, eliminate, or reduce hazards to an acceptable level, to monitor the performance of controls, and to maintain required records. The primary responsibility of in-plant inspectors is to evaluate the implementation and maintenance of a HACCP plan’s process controls. It is not the responsibility of in-plant inspectors to determine whether the form and content of a HACCP plan is adequate… (FSIS Memorandum dated 13 February 1998).
This memo clearly emphasizes the change in accountability structure under HACCP. A key issue is the ability to provide the requisite oversight for which a key impediment is the fact that many FSIS inspectors do not have the technical training in microbiological aspects of food safety to assess HACCP plant controls.
There is little question that industry favors the HACCP approach and the flexibility in process controls that it permits. At the same time, there have been spectacular lapses in the quality of meat production since the HACCP system has been implemented that have led to massive food recalls. One notable case was the inability of FSIS inspectors to take action against a ConAgra ground beef plant that had repeated problems from January 2001 until summer 2002 at which point ConAgra issued a recall notice for 19 million pounds of meat linked to an E.coli outbreak (see Petersen & Drew 2003).
This and other notable lapses have led critics to suggest that there are serious weaknesses in the accountability for food safety under the HACCP system. As documented by the USGAO (2002), FSIS inspectors have clearly been hampered by their lack of technical expertise to provide oversight of HACCP systems and by legal restrictions in their ability to take actions when defects are identified. Whether these limitations contributed to the noteworthy lapses in meat production is debatable given that FSIS personnel were able to identify the problems, but were hampered by legal restrictions on their ability to close the offending plants. Regardless of the specifics, the net result is diminished confidence in the integrity of food-safety regulation.
Building safety: Leaky buildings and performance-based regulation
The regulation of the safety of buildings is an important function that has perhaps the oldest history of all protective regulations. The earliest building codes in the USA date to the regulation of chimneys and roofing material in New Amsterdam in 1648 with comprehensive building regulations introduced in the 1850s. Until the past decade, the regulation of building safety has developed throughout the world as one of the more rule-bound and prescriptive aspects of protective regulation. Since the initial model building code in the USA was promulgated in 1927, revisions and additions have resulted in hundreds of provisions that as of the 1997 version comprised approximately 1000 pages.
A trickle of efforts that began in the 1970s and gathered momentum in a variety of forums since then has led to incorporation of performance-based concepts as part of building regulation in more than a half-dozen countries. The applicability of the performance-based provisions differs among these settings for which the most comprehensive regime was introduced in New Zealand in 1991. The problems that arose in New Zealand with the emergence of what has become known as the “leaky building crisis” is instructive with respect to accountability issues for performance-based regulation.
New Zealand’s Building Act of 1991 provided broad objectives of protecting people, their health and safety, and the environment with detailed sub-objectives that identified desirable building performance. These and other provisions embraced the New Right faith in the market and limited governmental intervention that were themes of a variety of reforms in New Zealand at the time. In the case of building regulation, market-like mechanisms were introduced by encouraging innovation in development of building materials and by allowing certification of compliance with desired building performance characteristics to be undertaken either by private certifiers or by local authorities.
As the New Zealand Herald labeled the problem in a series of two dozen articles appearing in 2002 and 2003, the “leaky building crisis” was not the typical story of shoddy construction or localized failures in building inspection that move the mundane aspects of building regulation into the public consciousness. Rather, the problems were pervasive, leading to a crisis for many homebuyers, the central government and the building industry. As happened in New Zealand, moisture entering into the membrane of a structure can lead to cracking and eventual partial or total collapse of a building. Beginning in the mid-1990s, condominiums built with a particular type of exterior cladding and high-priced residences built with similar types of synthetic stucco sheathing began to show these problems. Various investigations and media coverage suggest up to 18 000 homes and hundreds of multi-unit buildings were affected, many of which had to be abandoned as uninhabitable.
The fallout of the publicity surrounding the crisis was noteworthy. The central government was deeply involved in responding to the issue, including a commitment by the Prime Minister in her 2003 annual address to the nation that the crisis would be resolved by the government. A number of major construction firms in the Auckland area were forced into receivership because of the anticipated costs of repairing damage to structures they had built. The insurance market for building certifiers dried up, effectively putting many certifiers out of work, including the second largest firm in Auckland. Numerous lawsuits were brought against builders and local councils by owners of damaged buildings, many of which still have not been resolved.
Accountability issues loomed greatly as part of this crisis. Political accountability was invoked through “fire alarm” mechanisms following the extensive newspaper coverage of the issue. This, in turn, led to the range of governmental reactions and investigations. The political responses to the fallout, developed after extensive investigations and public commentary, were a reorganization and strengthening of the oversight of building regulatory functions by central government and the enactment of a new building regulatory regime under the Building Act of 2004.
Shortfalls at each of three levels of accountability entered into the New Zealand “leaky building” case. One was a shortfall in legal accountability resulting from the specification of goals for which the goal of “durability of structures” was insufficiently precise (New Zealand Building Industry Authority “Hunn Report” 2002). This led to inconsistencies in interpretation of the provisions by local building authorities especially as they applied to new building materials that were the source of many of the leaks when improperly applied.
A second shortfall was a lapse in bureaucratic accountability that resulted from lessening reliance on the bureaucratic controls as a means for ensuring adequate construction of buildings. Indeed, the 1991 Act did not require inspections of building during construction although local governments could require them. As put by one review of the situation:
[H]and-in-hand with the service or product provider being given the ability to determine and provide design and construction solutions must go a responsibility and accountability to guarantee their performance against the Building Code’s requirements. This has not happened (New Zealand Building Industry Authority, “Hunn Report” 2002: 11).
The shortfall in bureaucratic oversight would not have been as problematic if builders were fulfilling their professional obligations. But, reviews of the crisis noted the lack of licensing requirements for builders exacerbated the weaknesses in regulatory oversight. As put by a different review of the situation:
[A]lthough the framework for building work in New Zealand may, in part, be adequately designed, a wide range of participants have not complied with it. The system of procedural and technical controls also appears, in part, to be faulty in design and therefore inadequate in preventing undesirable outcomes such as the leaky buildings crisis (New Zealand House of Representatives “Yates Report” 2003, p. 15).
Not surprisingly, the 2004 Building Act revisions tightened bureaucratic accountability with emphasis on greater specification of performance standards, stronger monitoring of building inspection practices and tighter licensing provisions for building certifiers.
Nuclear safety: Seeking a safety culture under risk-informed regulation
The 1979 nuclear power plant accident at the Three Mile Island Unit-2 reactor near Harrisburg, Pennsylvania, USA, brought the issue of safety of nuclear power plants onto the public and policy agenda. This was not a new issue given that the Nuclear Regulatory Commission (NRC) was created to provide a greater focus on safety than that provided by the Atomic Agency Commission.
The traditional regulatory approach has been the use of prescriptive regulations governing licensing and operation of nuclear power plants that one expert characterizes as “a long, fragmented checklist of requirements that safety-related systems in a plant must satisfy” for which “[t]he consistency of this checklist and its ability to promote uniform levels of safety among different power stations is questionable” (Golay 2000, p. 221). That approach is being transformed with what the NRC labels as risk-informed regulation as a basis for setting priorities for regulatory standards and activities. This approach is an outgrowth of efforts that began in the 1970s and was endorsed by NRC leadership in a series of steps in the late-1990s. At present, the approach is better considered as a desired regulatory philosophy rather than as actual practice.
The risk-informed approach is in essence a system-based regulatory approach. It evolved from the efforts to develop and employ probability-based risk analyses in setting standards and evaluating nuclear power plant performance. The system-like regulatory philosophy is evident from the NRC description of risk-informed regulation:
A risk-informed approach enhances the traditional approach by: (a) explicitly considering a broader range of safety challenges; (b) prioritizing these challenges on the basis of risk significance, operating experience, and/or engineering judgment; (c) considering a broader range of counter measures against these challenges; (d) explicitly identifying and quantifying uncertainties in analyses; and (e) testing the sensitivity of the results to key assumptions (NRC 2001, part 1-1).
The approach was characterized by the chairman of the NRC in 2002 as “perhaps the most significant change occurring at the NRC today and [is] a theme central to the NRC’s activities. This effort represents a significant shift away from our traditional approach” (Meserve 2002).
The NRC risk-informed approach shifts the emphasis in accountability from bureaucratic accountability involving detailed inspections of nuclear power plants to greater emphasis on professional accountability of plant operators for adequate safety systems as overseen by NRC inspectors and staff. Under traditional, prescriptive approaches the long checklist of safety requirements and the thousands of hours of monitoring individual plants leads to power plant owners, according to one experienced nuclear engineering expert, to “commonly treat satisfaction of the NRC’s requirements as being a sufficient effort for accident prevention and mitigation” (Golay 2000, p. 221) thereby placing emphasis on NRC bureaucratic controls in determining the adequacy of safety. The risk-informed approach attempts to shift this balance toward greater responsibility of nuclear power plant operators for identifying potential safety issues and for NRC inspectors to focus on noteworthy potential risks.
A concern for overall accountability is the extent to which the shift in emphasis from bureaucratic to professional accountability undermines the degree of overall safety. The USGAO (2000, p. 5) found that 60% of NRC staff responding to questions about the oversight process thought that the risk-informed approach would “reduce the margins of safety” at nuclear power plants and 75% of the staff thought utility and industry groups had “too much input in developing the processes.”
The difficulty of increased professional accountability for nuclear power plant safety is illustrated with the challenges of creating a culture of safety. There seems to be little doubt that a culture of safety is critical to instill to avoid potentially catastrophic consequences. As noted in comments by a former NRC chair, obtaining this is a leap of faith:
We believe that it is unnecessary to assess a licensee’s safety culture as a distinct component because the concept of safety culture is similar, if not integral, to the licensee’s more specific responsibilities. If a licensee has a poor safety culture, problems and events will continue to occur at that facility either causing various performance indicators to exceed their thresholds, or surfacing during the NRC’s baseline inspection activities (Meserve 2002: 9).
At issue is the extent to which system-based approaches and regulatory mechanisms that emphasize professional accountability can help bring about the desired safety culture. These are central issues for security and emergency preparedness programs that have been major contentions in plant licensing. Skeptics of the risk-informed approach suggest that the lack of established professional standards for such things as emergency preparedness provisions and a history of safety lapses on the part of nuclear power plant operators undermine the ability to instill a safety culture.
Fire safety: Engineering performance-based regulation
The regulation of structures for fire safety has historically evolved in response to devastating fires. Saving lives from fires became a prominent concern after fires such as the 1903 Iroquois Theater fire in Chicago, the 1911 Triangle Shirtwaist Factory fire in New York City and the 1942 Coconut Grove nightclub fire in Boston. Protecting lives has become the focal objective of fire regulations.
The concept of a performance-based regulatory approach has been widely embraced by the fire-safety community. A major impetus for this has been the efforts of the Society of Fire Protection Engineers, a professional association of engineering specialists in fire safety. This group sponsored a series of workshops beginning in 1991 that became important forums for identifying relevant issues and advocating regulatory changes. Closely related were the efforts of the National Fire Protection Association (NFPA), an international nonprofit association dedicated to fire prevention, in incorporating the performance-based concepts into the development of a new set of consensus code documents published in 2000 (NFPA 2000).
The use of the performance-based approach to fire safety has been mostly limited to assessments of the fire safety of nontraditional structures like the Luxor Hotel in Las Vegas. Public officials have been reluctant to discuss potential loss of life from fires, particularly after the fire-related life losses at the World Trade Center Towers on 11 September 2001. That experience also underscores the lack of reliable methods for predicting the performance of protective systems for potential fire situations. Although there are a number of computer programs for modeling the ignition and spread of fire and guidelines have been produced by the Society of Fire Protection Engineers for carrying out such evaluations, much of the commentary in technical forums about predictive modeling underscores the difficulties and inherent limits. The prediction difficulties in part stem from the complexity of potential ignition sources, spread, and other physical and engineering factors. One complicating factor, particularly evident in the World Trade Center Towers “9/11” experience, is the unpredictability of human behavior in responding to fire.
As with the other cases considered here, the performance-based approach to fire safety shifts the emphasis for accountability from bureaucratic to professional-based accountability mechanisms. In the case of fire safety, the shift is more one of emphasis than wholesale change. Fire-protection engineers have for a long time been involved in analyzing and evaluating fire protection for nontraditional structures. The performance-based approach emphasizes their role and the state of practice in evaluating whether a given structure provides adequate protection from fires. This places these engineers in a similar role of that of building certifiers in New Zealand, who turned out to be a very weak link in the accountability structure for performance-based approaches to building regulation. A notable difference, however, is the greater degree of professionalization of fire-protection engineers.
Comparing accountability shortfalls
A variety of potential accountability shortfalls are suggested by the case studies of system-based and performance-based regulatory regimes. Table 4 depicts these with respect to the different levels of regulatory accountability while illustrating key accountability shortfalls illustrated by cases. Potential accountability problems for prescriptive regulation are also indicated for comparative purposes. The shortfalls are purposely labeled as potential given the limitations of diagnosing them from the brief case materials and the fact that they are not inevitable.
Table 4. Potential accountability problems
|Legal accountability shortfalls||Rules or standards that favor private interests over public interests||Restrictions on regulatory actions when system defects are found||Inability to prescribe measurable goals and standards|
| Case examples||–||Court restrictions on ability to close meat plants with HACCP system defects||New Zealand’s lack of specification of meaningful “durability” goals and standards|
|Bureaucratic accountability||Nitpicky and capricious enforcement||Inability to gauge quality of management system;||Inability to assess or predict performance;|
| Case examples||–||inspectors’ lack of expertise for assessing HACCP systems||limits of fire prediction models for evaluating adequacy of fire-safety performance|
|Professional accountability||Poorly trained or unqualified inspectors; corruption||Regulated entities’ professionalism in designing best management system||Regulated entities’ adherence to performance standards|
| Case examples||–||Nuclear power plant “safety culture” and deficiencies in security and emergency management plans||New Zealand shortfalls in building practices and use of substandard materials|
|Political accountability||Inability to decipher what complaints mean for regulatory reform||Notable undesired outcomes that the systems could not prevent||Notable unexpected outcomes that were not part of performance goals|
| Case examples||–||Notable lapses in meat production that HACCP system could not address||“Leaky building” crisis in New Zealand’s performance-based approach to building safety|
Potential shortfalls in legal accountability concern the applicability of rules and standards. As elaborated upon in the concluding section, a key issue is the potential for some form of regulatory capture for which the traditional form is noted under prescriptive regulation. The key limitation identified for system-based approaches is the limit to the use of system defects as a basis for taking regulatory action, as illustrated by the HACCP inspection process limits. The key limitation identified for performance-based regulation is the difficulty of prescribing goals and standards, as illustrated by the failure in New Zealand to provide adequate building durability goals and standards.
Potential shortfalls in bureaucratic accountability concern the application of regulations by regulators and shortfalls in respective compliance by regulated entities. A key issue for prescriptive regulation is nitpicky and capricious enforcement that can result in missing larger compliance issues. System-based regulation places emphasis on monitoring adequacy of regulated firms’ systems that, as illustrated by the HACCP case, is undermined if inspectors do not have the expertise to fulfill their monitoring roles. Potential shortfalls in bureaucratic accountability for performance-based regimes stem from the inability to assess or predict performance, as illustrated by the difficulties of predicting the fire-safety performance of individual structures.
Potential shortfalls in professional accountability concern the lack of well-established professional norms and abuse of professional responsibilities. Both are sometimes a problem with prescriptive regulation, although corruption for regulatory programs is more of a problem in developing countries than in advanced countries (see Ogus 2004). These shortfalls are also evidenced by the brief case illustrations of system-based and performance-based regulations. The difficulties of establishing a professional “culture of safety” for security and emergency planning for nuclear power plants illustrates the lack of established safety norms and potential for subterfuge on the part of some power plant operators. Shortfalls in New Zealand building practices were directly attributed to lack of licensing requirements for builders and substandard building practices on the part of a large number of builders.
Political accountability is concerned with the responsiveness of officials to regulatory problems. The HACCP and New Zealand cases illustrate visible regulatory shortfalls that arose under system-based and performance-based regulatory regimes. The latter became a crisis that engaged top levels of New Zealand government while the former lingered for years as regulators and elected officials tried to grapple with legal restrictions on food recalls.
The discussion of regulatory regimes and accountability in this article stresses different levels of accountability and the accountability issues that arise for prescriptive, system-based and performance-based regulatory regimes. Each regulatory regime has potential for accountability shortfalls at one or more levels. However, they differ in the specifics and what they suggest about consequences for regulatory outcomes.
Regulatory capture revisited
A long-standing concern of regulatory scholars is the potential for regulatory capture (see Wilson 1980; Ayres & Braithwaite 1992). The classic forms under prescriptive regulation are when some private interests win out if they are able to enact standards that promote particular products or technologies and when entire industries are favored if they gain from exclusionary practices (e.g. market limitations, higher prices) that are embedded in regulations. The former is a local form of regulatory capture and the latter is a more global form.
An appealing aspect of system-based and performance-based approaches is that they provide a more level playing field by not prescribing particular methods or materials. Consequently, particular producers are not favored over others or at the expense of the public interest. Indeed, the newer forms of regulation are aimed at promoting competition to provide better and more cost-effective ways of complying with regulations. But, the cases considered here show that the more global form of regulatory capture does not necessarily disappear under the newer regulatory regimes. It simply appears in a more subtle form.
Industry, or experts hired by industry, have varying abilities to determine adequacy of mandated systems or desired performance. Under the HACCP food-safety program, plant personnel have substantial authority and arguably greater expertise when it comes to deciding whether a food production process provides adequate food safety. In the New Zealand building regulation case, de facto standards for performance of cladding systems were established by building certifiers with those standards falling short of what was intended by the performance-based code. The situation for the regulation of fire safety is similar to that of building safety given the reliance on third-party professionals for certifying the adequacy of fire protection for any structure.
Accountability and regulatory performance
Any regulatory regime must confront a fundamental issue of how tight controls should be in seeking consistency versus how much discretion should be granted in promoting flexibility and innovation. Traditional, prescriptive regulation emphasizes consistency through bureaucratic controls that seek compliance with prescribed actions. System-based and performance-based regulatory regimes relax these controls in favor of greater reliance on professional accountability. At issue is the degree to which this substitution of professional accountability for bureaucratic accountability mechanisms affects the performance of a given regulatory regime.
The case evidence suggests how accountability shortfalls can undermine regulatory performance. The HACCP system for the regulation of food safety raises issues about the role of regulated entities in devising and maintaining adequate systems. Regulatory performance is undermined by insufficient legal authority to use microbiological testing to monitor safety of plant processes and the limited technical abilities of regulatory personnel to monitor plant systems. In New Zealand, the failure to invest the necessary resources for adequate regulatory oversight contributed to overreliance on poorly trained third-party inspectors and to lax review of alternative building products. The result was the crisis that emerged for which a leaky regulatory regime failed to provide meaningful accountability that, in turn, limited regulatory performance.
Implications for design of regulatory regimes
These concerns about accountability underscore the importance of finding the right fit between the regulatory circumstances – in terms of types of firm, complexity of tasks, potential for harm and so on – and the design of regulatory regimes (see Coglianese & Lazer 2003). Finding the right fit can make a big difference in reducing confusion over regulatory requirements and the incentives for regulated entities to shirk their responsibilities.
The key accountability considerations for the system-based and performance-based regimes considered in the cases described in this article are professional judgment and the exercise of professional responsibility. Each reduces the role of traditional regulators in favor of increased reliance on the professional expertise of industry personnel and third-party experts. When such expertise is lacking, as was the case in New Zealand’s performance-based building safety regime, there is a clear mismatch in regulatory design. When regulators do not have the skills to evaluate complex systems, as was the case in the HACCP approach to food safety, there is also a mismatch in regulatory design.
The challenge for regulatory designers is how to compensate for “bad apples” and the lack of the requisite professional expertise under system-based or performance-based approaches. In some instances, there may be little choice but to revert to a largely prescriptive regime. New Zealand officials added prescriptive requirements as part of their reform of the performance-based approach to building regulations without giving up on the basic performance notions. In other instances, it may be possible to strengthen a sense of professional accountability through education programs or through increased legal liability for failure to meet codes of conduct. The encouraging news is that regulatory scholars are identifying an expanding toolkit of potential accountability mechanisms that could come into play for newer regulatory regimes (see, in particular, Lodge 2004).
The discouraging news for advocates of newer regulatory regimes is that the experiences reviewed here show potential shortfalls in different aspects of accountability and a potential for subtle forms of regulatory capture. These problems reinforce the dilemma of allowing for increased regulatory flexibility without sufficient accountability structures.
Invaluable assistance with this research was provided by Chris Koski and Melissa Poague. Revisions to earlier drafts were stimulated by comments provided by Cary Coglianese, George Hoberg, Neil Gunningham, and reviewers and the editors of this journal. Support for this work was provided by funding to the University of Washington by the Pacific Earthquake Engineering Research Center (PEER) as part of the Earthquake Engineering Research Centers Program of the National Science Foundation under Award Number EEC-9701568. The findings are not necessarily endorsed by the National Science Foundation, the PEER Center or the University of Washington.