- Email: [email protected]
Deregulation and Nuclear Power Safety
Vicki Bier is an Associate Professor in the Departments of Industrial Engineering and Engineering Physics at the University of Wisconsin– Madison, where she is also Director of the Center for Human Performance and Risk Analysis. She also has extensive experience as a consultant in the area of risk analysis for nuclear power plants. James Joosten is a Managing Partner for Connect-USA.com, an international nuclear consulting company. He was previously Principal Administrator for nuclear development at the Nuclear Energy Agency of the Organisation for Economic Co-operation and Development in Paris. David Glyer is a Senior Economist at Laurits R. Christensen Associates, an economic consulting company specializing in creative solutions to business challenges and high-caliber economic analysis. He has held faculty appointments at Oregon State University and the University of Colorado–Denver. Jennifer Tracey is a Staff Economist at Christensen Associates and an economics instructor at Madison Area Technical College. Michael Welsh is a Senior Economist at Christensen Associates. This report was prepared as an account of work sponsored by an agency of the U.S. government. Neither the U.S. government nor any agency thereof, nor any of their employees, makes any warranty, expressed or implied, or assumes any legal liability or responsibility for any third party’s use, or the results of such use, of any information, apparatus, product, or process disclosed in this report, or represents that its use by such third party would not infringe privately owned rights. The views expressed in this paper are not necessarily those of the U.S. Nuclear Regulatory Commission.
May 2001
Deregulation and Nuclear Power Safety: What Can We Learn from Other Industries? An examination of the U.S. aviation and rail industries and the U.K. nuclear power industry provides evidence that changes associated with deregulation can be expected to create major challenges to the management of safety by the U.S. nuclear power industry and its safety regulators. Vicki Bier, James Joosten, David Glyer, Jennifer Tracey, and Michael Welsh
T
he effects of economic deregulation in electricity generation may be seen across a broad spectrum of safety-related areas in the U.S. nuclear power industry. For example, while economic deregulation could increase a company’s motivation to avoid costly safety problems, it could also result in downsizing that could increase human error rates. Since deregulation reduces the ability of licensees to automatically pass along costs to ratepayers for safety-related activities, it can also influence decisions on plant upgrades and safety improvements. Finally, economic deregulation could affect the
workload and requirements imposed on government safety regulators. This article attempts to shed light on the possible consequences of electricity deregulation for nuclear power safety. We are particularly interested in the effects of industry restructuring in response to deregulation on factors such as: financial pressures; equipment maintenance, aging, and failure rates; mergers, acquisitions, and corporate restructuring; and factors relevant to human performance (e.g., downsizing, use of contractors, and labor relations). Because deregulation of the U.S.
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
49
electricity industry is very recent, this article adopts a historical approach, focusing on the effects of economic deregulation on safety in other deregulated industries. Using both literature reviews and interviews with industry representatives, evidence is analyzed regarding the effects of deregulation in three other technologically sophisticated, safety-critical industries: the U.S. aviation and rail industries; and the U.K. nuclear power industry. In the course of the case studies, we reviewed more than 250 documents and interviewed senior industry representatives with direct experience in safety issues. This approach made it possible for us to move beyond speculation about the possible adverse safety consequences of deregulation, to identify situations in which particular adverse safety consequences actually occurred. Hopefully, the results of this study will make it possible to anticipate any adverse effects of deregulation on nuclear power safety, and develop strategies for the industry and safety regulators to detect, monitor, and mitigate those effects. efore presenting our findings, there are several important issues to note. First, like the other case study industries, the U.S. nuclear power industry is not being deregulated with respect to safety, since the U.S. Nuclear Regulatory Commission (NRC) still maintains its safety oversight and regulatory roles. Second, the U.S. electricity industry is currently in a state of flux, and the degree and timing of deregulation differ from
B
50
state to state. Therefore, we do not attempt a comprehensive analysis of the specific situation in each state, but rather assess the relevance of experiences in other industries to the U.S. nuclear industry as a whole. Finally, we emphasize that as the electricity generation industry in the United States transitions from an extended period of cost-of-service regulation1 to a less economically regulated mode of operation, the
Generating companies that cannot achieve operating costs at or below the market price for electricity will not be competitive.
incentives for nuclear generating companies will change dramatically. In particular, generating companies that cannot achieve operating costs at or below the market price for electricity in their region will not be competitive. On the other hand, companies will also have stronger incentives to achieve operating costs substantially below market prices, since any productivity improvements or cost savings will contribute directly to increased profit. With this backdrop in mind, we now review the experiences of the case study industries in safetyrelated areas to help assess how
the ongoing restructuring of the electricity industry is likely to affect future operations, safety, and financial performance of nuclear power plants. This article also discusses the favorable conditions that accompanied deregulation and/or restructuring in the three case study industries (which may have mitigated some adverse safety consequences), and attempts to assess whether similarly favorable circumstances are likely to exist in the U.S. nuclear power industry. Finally, we end with some recommendations for ensuring nuclear power safety as the process of electricity deregulation unfolds.
I. Background The airline industry was economically deregulated with passage of the Airline Deregulation Act of 1978, with the Federal Aviation Administration (FAA) maintaining continued responsibility for aviation safety. Deregulation resulted in a number of changes in the nature of the airline industry. In particular, the volume of airline transportation increased greatly. Deregulation also encouraged a large number of new entrants to the aviation industry, and triggered a large number of mergers, acquisitions, and bankruptcies. Overall accident rates in the airline industry had been steadily declining for many years, and that trend continued after deregulation.2 However, the rates of both near midair collisions and “runway incursions” increased during the mid-1980s. Detailed examination
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
The Electricity Journal
of postderegulation experience in the aviation industry also reveals a number of other changes with adverse safety impacts, as discussed below. he railroad industry was economically deregulated with passage of the Railroad Revitalization and Regulatory Reform Act in 1976, and the Staggers Act in 1980. The Staggers Act in particular gave railroads greater freedom to set rates, and enhanced their ability to abandon unprofitable lines. In addition, the Federal Railroad Administration (FRA) had been established in 1967 with responsibility for safety regulation. After the Staggers Act, the financial performance of the railroad industry improved significantly, and expenditures on track increased dramatically.3 Track-related accidents peaked in 1978 (because the poor financial health of railroads in the 1960s and 1970s had resulted in reduced expenditures on track maintenance), and had dropped dramatically by 1982. It is difficult, however, to determine whether the increased FRA emphasis on safety regulation or the improved financial health of railroads was primarily responsible for the improvements in safety observed after passage of the Staggers Act. The U.K. electricity supply industry, which had been governmentowned and -operated, was largely privatized and deregulated with passage of the Electricity Act in 1989, when all conventional power stations were assigned to two new private companies. However, due to concerns over potential nuclear liabilities, nuclear power stations
T
May 2001
remained under government ownership at that time. Based on their favorable performance records, the government subsequently privatized the newer reactors in 1996, forming a new nuclear company called British Energy (the older Magnox plants, which had worse economic prospects, remained in government ownership). The U.K. Nuclear Installations Inspectorate (NII) was actively involved in planning for and monitoring the
After deregulation, both the aviation and rail industries exhibited shifts in allocation of resources for maintenance.
privatization process, and relicensed the privatized reactors. Employment in the industry decreased significantly, beginning before deregulation and continuing afterward. The privatized nuclear power plants have done well on many measures of economic and safety performance. However, in response to observed problems, the NII audited British Energy in 1999.4 The audit findings included excessive reliance on contractor staff, and lack of definition of the skills required for license responsibilities. As a result, further reduction in staffing was restricted
pending development of a satisfactory staffing plan.
II. Equipment Maintenance and Aging After deregulation, both the aviation and rail industries exhibited significant shifts in allocation of resources for equipment maintenance. For example, one study of the aviation industry found that airlines reduced the frequency of engine overhauls after deregulation, but did not experience a higher rate of engine failures as a result,5 apparently because of improved efficacy and efficiency in maintenance programs. The rail industry undertook even more dramatic reallocation of resources, with workforce reductions of more than 50 percent (partially due to work-rule changes), and increased capital expenditures on track by nearly a factor of five. The increases in track maintenance are widely viewed as being responsible for much of the observed improvement in rail safety. o significant changes in maintenance budgets were observed in the U.K. nuclear industry. There appears to be a realization in the United Kingdom that high levels of reliability are critical to keeping nuclear units on line and producing revenue. In fact, due to the high costs associated with unplanned outages, the new competitive market has increased the incentives to avoid unplanned shutdowns, even at the expense of delayed startups if necessary to “do the job right.” Simi-
N
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
51
larly, there are stronger incentives today to comply with technical specifications and avoid regulatory shutdowns. eregulation systematically changes the incentives faced by U.S. nuclear generating companies with regard to maintenance expenditures. Some of these incentives will encourage minimization of maintenance expenditures, with possible adverse safety consequences, while others will encourage increased maintenance or capital investments in particular areas. Cost minimization will be more important after deregulation than before, so some safety-related maintenance activities may be cut back or eliminated. For example, South Texas Project (STP) managers expect to achieve significant annual savings by “optimizing maintenance frequencies, [reevaluating] how detailed post-maintenance testing needs to be and reconsidering how detailed plant documentation packages need to be” for components with little or no risk significance.6 However, the economic value of maintenance activities that do yield significant benefits in productivity, asset protection, or avoidance of regulatory sanctions is likely to be greater after deregulation than before, since any savings will contribute directly to increased profit. For example, Gupta and Thompson list “capital upgrades to decrease costs and increase the capacity factor” as a possible area for improvement in the electricity industry.7 Thus, initiatives to optimize maintenance efforts are likely to be seen in the U.S. nuclear
D
52
power industry, as in the aviation and rail industries. Although it is possible to achieve maintenance reductions without jeopardizing safety, cuts can sometimes go too far, since feedback indicating the adverse effects of deferred maintenance may be ambiguous or delayed. Thus, Carroll et al. note that some organizations in both the nuclear power and chemical processing industries “fail . . . to give due consideration
In the U.S. nuclear industry, there has been strong interest in life extension as deregulation proceeds.
to preventive maintenance in organizational decisions,” even though “sparse allocation of resources to maintenance is not a rational strategy for the organization as a whole.”8 With respect to equipment aging, no significant evidence was found of a trend toward use of older equipment after deregulation in the rail industry. In fact, although aging and deterioration of track had been a significant problem prior to deregulation, the quality of track maintenance improved significantly after restructuring due to the improved financial health of the major railroads and
the abandonment of unprofitable lines. The aviation industry did experience a trend toward longer operating lives for aircraft. The available statistics do not show a clear link between aging aircraft and increased risk, presumably due to effective preventive maintenance. However, two interview respondents identified concerns about aging aircraft as an important safety issue in the future, especially at smaller carriers that may not have the resources to adequately maintain their aircraft as they age. In the U.S. nuclear industry, there has been strong interest in life extension as deregulation proceeds. Because the marginal costs of nuclear power are generally favorable compared to coal and oil, companies are interested in getting the most out of their nuclear plant assets by increasing their lifetimes. As of this writing, owners of 43 plants have obtained or formally expressed interest in license renewal.9 As a result, some industry observers have argued for additional research on risks associated with aging nuclear power plants. y contrast, companies with less profitable plants (and plants facing large short-term costs or large uncertainties) have in some cases decided to close those plants. Joskow notes that “The pressures of emerging competition have . . . led to the early retirement of at least four nuclear reactors in the U.S. More are likely to follow this path in the next few years.”10 Plant owners considering early shutdown may be reluctant
B
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
The Electricity Journal
to undertake discretionary safetyrelated investments during the last few years of operation. Thus, deregulation may simultaneously accelerate the rate of plant closures and also increase the interest in life extension, both of which can pose safety issues.
III. Financial Pressures Economic theory suggests a link between poor profitability and inadequate safety investments, predicting for example that firms facing possible bankruptcy are more likely to take chances,11 such as deferring spending on discretionary safety improvements. We found evidence in two of the three case studies that financial difficulties may indeed be associated with poor safety records. For the aviation industry, statistical analyses of airline data found lower profitability to be associated with higher accident and incident rates, particularly for smaller carriers.12 In fact, based on her results, Rose specifically recommended “more intense scrutiny of the safety practices of financially marginal carriers.” In the rail industry, the poor safety performance of the railroads in the period leading up to deregulation is generally believed to be associated with their poor financial health during the same period. In addition, studies of accident rates in the railroad industry prior to deregulation found that worse financial health was generally associated with higher accident rates.13 This relationship was strongest for those
May 2001
railroads that were actually unprofitable, rather than only marginally profitable. We found no evidence that financial difficulties were associated with safety problems for the U.K. electricity industry. This is because the privatized nuclear plants became very profitable following deregulation, partly due to the nature of the U.K. privatization process, which provided significant subsidies for nuclear power in
Economic theory suggests a link between poor profitability and inadequate safety investments.
the years immediately following privatization. In the U.S. electricity industry, a central issue for the competitiveness of nuclear power is the fact that the construction costs of most nuclear power stations were high, usually much higher than anticipated. Under cost-of-service regulation, capital costs (a rate of return on capital, plus depreciation) are included in regulated rates. Continued recovery of these capital costs would require electricity prices higher than could likely be attained under competition. Therefore, in most states with deregulated electricity markets, owners of
such overvalued plants have been allowed to collect “stranded cost” payments as part of approved deregulation plans, to compensate them for a portion of the returns that they would have received under cost-of-service regulation.14 The consensus seems to be that with such stranded cost recovery, “most nuclear units should be able to compete with other sources of electricity in a competitive generation business.”15 Thus, the most successful nuclear generating companies are expected to have strong operating profits after deregulation, and these profits will not be artificially limited by economic regulation. evertheless, deregulation can be expected to create financial difficulties for at least some plants during the transition period. The statistically significant relationships between financial difficulty and safety problems observed in two of the three case studies suggest that such financial stresses may contribute to declining safety margins. These financial pressures also create the possibility of further plant closures. Like impending bankruptcy, anticipated plant closures can reduce the incentives to invest in discretionary safety improvements during the last few years of a plant’s operation, since these costs can be amortized over only a short time. (It should be noted, however, that minimum safety standards would need to be maintained to satisfy the conditions of the plant’s operating license.) The nuclear power industry today also faces substantial finan-
N
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
53
cial uncertainties. Gupta and Thompson cite uncertainties about “the tax consequences (and time requirements) of transferring ownership, revaluing the asset, and transferring the decommissioning funds.”16 Similar uncertainties relate to plant life extension, nuclear waste disposal, and future attitudes to nuclear safety. Such uncertainties (and the corresponding financial risks) will generally lead companies to adopt higher discount rates. For example, Joskow states that under deregulation, electricity generators “will base decisions on higher costs of capital, shorter payback periods and more stringent cash flow requirements.”17 This can be expected to result in a trend toward shorter-term planning, with long-term investments (possibly including some safetyrelated investments) requiring high levels of benefit to be economically justified.
IV. Human Performance and Labor Issues The three case study industries differed widely with regard to the effects of deregulation on human performance and experience levels. In the aviation industry, rapid industry growth after deregulation created significant growth in total employment. Thus, several observers of the aviation industry raised concerns about the human performance implications associated with high turnover rates and hiring of less experienced personnel, especially at smaller carriers. By contrast, both the rail indus-
54
try and the U.K. electricity supply industry experienced dramatic downsizing. Rail industry employment decreased by more than half in the first decade after deregulation. Nonetheless, the rail industry was generally able to maintain high levels of employee experience, with few concerns being raised about employee inexperience. However, investigations of major railroad accidents in recent
that “Fatigue issues are much worse since deregulation.”19 Dramatic employment cuts were also observed following deregulation and privatization of the electricity industry, both in the United Kingdom and elsewhere.20 In particular, total employment in the U.K. electricity industry fell by roughly half in the five years immediately after the restructuring of the industry in the late 1980s, following a long period of significant but slower downsizing. n the United Kingdom, the NII has expressed concern that rapid downsizing, accompanied by increased use of contractors, may have left the U.K. nuclear operating companies with marginal levels of experience and expertise to safely operate their nuclear power plants. These shortfalls occurred in part because anticipated workload reductions did not materialize as rapidly as expected after workforce reduction measures had been implemented. The NII has also raised concerns about the qualifications and safety culture of the contractors, and about possibly excessive use of overtime.21 Such concerns have led the NII to begin requiring prior approval of significant changes in staffing levels, to prevent similar problems in future. In the U.S. nuclear power industry, downsizing is currently one of the most common approaches for reducing costs. For example, Fox states that “total labor requirements on average across the industry have fallen by 14 percent during the past five years.”22 Moreover, the trend to downsizing
I
years have raised concerns about excessive downsizing, resulting in understaffing, fatigue, and inadequate supervision. As a result of those findings, the FRA has increased its focus on fatigue in recent years; a representative of the FRA also indicated concern about increased use of overtime.18 Significant concerns about fatigue also arose in relation to work-rule changes (especially minimum crew sizes). In particular, rail industry labor representatives stated that the steady decrease in crew sizes had increased problems of fatigue, with one union representative specifically commenting
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
The Electricity Journal
appears to be continuing, as it did in the United Kingdom. For example, utility vice presidents surveyed by Public Utilities Fortnightly generally reported downsizing of 10 percent or more in their companies.23 equired plant staffing levels appear to vary as a function of factors such as size, age, reactor type, number of units, and utility characteristics. This creates the risk that a plant attempting to emulate the staffing levels of the low-cost performers in the industry, without correcting for such differences, could make excessive cuts in safety-critical areas. More generally, McKinley et al. note that determining appropriate staffing levels is especially difficult when “organizational units lack clear performance standards,” and when “the relationship between means and ends is ambiguous.”24 Such “gray areas” in performance standards and ambiguous relationships between means and ends may be more typical of safety-related functions than of activities necessary for day-to-day production.
R
V. Influences on Corporate Culture: Mergers, Acquisitions, and New Entrants Issues of corporate culture were identified as being critical to safety in all three case studies. In both the U.S. aviation and rail industries, these concerns were primarily associated with mergers and acquisitions, in which the rapid pace of change sometimes resulted in periods of confusion that com-
May 2001
promised safety. Such mergers and acquisitions were widespread in the post-deregulation period; in fact, Cudahy goes so far as to suggest that “Nothing is certain about deregulation except the mergers that follow.”25 In the rail industry, Passell described the acquisition of the Southern Pacific and Chicago and North Western railroads by Union Pacific as “the most spectacular
merger fiasco of modern times,” resulting in several fatal accidents as well as significant freight delays.26 In particular, he noted, “Union Pacific’s by-the-book culture clashed badly with Southern Pacific’s, where managers had long been accustomed to making do with chewing gum and bailing wire.” Concern about possible negative effects of mergers prompted the Federal Railroad Administration to propose regulations requiring advance approval of consolidations, based on a satisfactory plan for safety management during restructuring. Several interview respondents urged caution regard-
ing the potential safety problems associated with mergers and acquisitions in the nuclear power industry. Rail industry interviewees also expressed widespread concern about underreporting of safety problems. One former railroad employee stated that “Deregulation came with low-grade fear to keep people from reporting violations.”27 In the aviation industry, the U.S. Office of Technology Assessment similarly noted that mergers had a potentially adverse impact on corporate culture, and hence on aviation safety.28 In fact, post-merger lack of coordination has been associated with an increased rate of pilot deviations.29 Several respondents in the aviation industry interviews also mentioned this issue, one of them describing issues of corporate culture as the biggest single problem associated with mergers and acquisitions. (However, another respondent with significant experience in both safety regulation and airline management commented that overall, “consolidation has led to the creation of stronger companies with more resources at their disposal,” suggesting that the safety problems associated with consolidation may be largely short-term in nature.) n addition, new entrant airlines have generally been associated with increased risk; for example, Barnett and Higgins found that the fatality risk for new entrants was 12 times higher than that for “trunklines.”30 Corporate culture issues appear to have played a part in this; for example, Gray stated
I
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
55
that “Many new carriers have little notion of or experience in how an airline must be run.”31 In the United Kingdom, the nuclear power industry is already heavily consolidated due to decisions made during the privatization process. There also, however, further consolidation is taking place, primarily in the form of alliances and partnerships, and the NII has raised concern about this trend. In particular, Allars has emphasized the “need for partners to fully understand the hazard/risks and to share in safety culture initiatives and goals set by the licensee.”32 any observers expect to see significant mergers and acquisitions within the U.S. electricity and nuclear power industries as deregulation proceeds. New ownership, mergers and acquisitions, changes in management teams, and even major changes in management philosophy or shareholder expectations (all of which can be expected to take place in the U.S. nuclear industry) can all affect corporate culture. In particular, Anderson, a utility industry consultant, discusses a number of problems that can arise from inadequate planning or implementation of mergers and acquisitions. He cites “poor cultural integration” as one possible pitfall, noting also that “risk management, regulatory compliance, environmental affairs, and internal audit functions” may not be implemented uniformly across the resulting merged company.33 Note, too, that divestiture and unbundling can also pose risks. For
example, Raughley cites possible adverse effects of divestiture on offsite power availability, observing that “Licensees are selling their generating facilities that supply offsite power to the nuclear plants.”34 He observes that such arrangements can reduce coordination, and thereby extend the time needed to recover from events such as loss of offsite power. The increase in power flows over the
M
56
transmission network due to increased electricity competition can also degrade grid reliability, posing a risk for the safe operation of nuclear power stations.35 Despite the possibility of significant short-term problems, however, many observers expect consolidation to be good for the industry in the long run. For example, Entergy’s Hintz anticipates that consolidation will be beneficial, arguing that “there is an advantage to having all the plants operated by operators with good performance records—companies that have a real focus on nuclear”;36 Gupta
and Thompson make similar observations.37
VI. Experiences of Safety Regulators In all three case study industries, deregulation had significant implications for safety regulators. In the aviation industry, deregulation significantly increased the workload of the FAA at a time when agency budget and staffing levels were already under pressure. Part of the increase in workload was due to dramatic increases in both sales volume and number of airlines after aviation deregulation, changes that are not expected to occur in the nuclear power industry. However, other factors (such as increased reliance on contractor maintenance, increased pilot turnover, and an aging fleet of aircraft) increased the burden on FAA inspectors. FAA staffing levels were also cut; for example, the number of FAA inspectors was reduced by roughly 20 percent in the years immediately following deregulation. As a result, the FAA has acknowledged that it was unprepared for the effects of deregulation. Moreover, inadequate FAA oversight has occasionally been cited as having contributed to particular accidents, such as the Air Florida crash in Washington, DC, in 1982. ome interview respondents from the aviation industry specifically recommended that the NRC attempt to learn from the FAA’s experiences after deregulation, and emphasized the importance of maintaining a highly
S
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
The Electricity Journal
trained and skilled regulatory workforce. In particular, one individual who had formerly worked for the FAA noted that the agency today has a significantly larger oversight infrastructure than at the time of deregulation. He suggested that the NRC might similarly need to strengthen its staffing to ensure effective oversight of facility operations, monitoring of regulatory compliance, and maintenance of good working relationships with regulated companies. nlike the FAA, the Federal Railroad Administration had significantly increased both the number of safety regulations and the level of inspection effort around the time of rail deregulation. These changes are often cited as contributing to the post-deregulation improvement in rail safety. Nevertheless, the FRA has sometimes been criticized for being inflexible and unresponsive to changes in technology, and for relying too heavily on punitive sanctions as the primary means to obtain compliance. As a result, in 1993 the FRA instituted a more cooperative approach among regulators, railroads, and labor in addressing safety problems; several rail industry interviewees stressed the benefits of such cooperation. The FRA has also recently increased its emphasis on human performance issues such as fatigue, based on the results of accident investigations over the last few years. In the United Kingdom, the Nuclear Installations Inspectorate takes a less prescriptive approach than the NRC, and generally has
more cooperative relationships with its licensees. It is also noteworthy that the NII was actively involved in planning for and monitoring the privatization process in the United Kingdom, and actually increased its staffing levels modestly in anticipation of increased regulatory workloads. In the United States, staffing and budget cuts of roughly 20 percent have already taken place at the
U
May 2001
NRC over the past seven years.38 As a result, current staffing levels are comparable to those in 1978, before the accident at Three Mile Island (TMI). Recent budgets are also at historically low levels, 40 percent below the post-TMI peak in the early 1980s. These cuts are of the same order of magnitude as those at the FAA in the years immediately following airline deregulation, which were indicated as a cause of concern in the airline industry. As noted by industry representatives, some reductions in staffing and budgets at the NRC may be appropriate, given the “safety record of the commercial nuclear
industry [and] the maturity of the technology.”39 However, deregulation can be expected to cause workload increases for the NRC, e.g., due to the need for approval of license transfers and review of decommissioning funds associated with mergers and acquisitions, and the large number of anticipated requests for life extension. In addition, if reduced safety margins occur at some companies, this might increase workloads for NRC safety regulators. As at the FAA, therefore, more effective targeting of scarce agency resources is likely to be required after deregulation. One way to achieve that may be through the NRC’s recent adoption of new risk-informed, performance-based approaches to safety regulation, which some observers see as a promising way to target resources at the most risksignificant problems. If successful in achieving that goal, riskinformed and performance-based regulation could allow both the NRC and nuclear plants to become more cost-effective without compromising safety. Unlike safety regulators in the rail and U.K. nuclear power industries, which have recently implemented requirements for prior approval of major organizational changes, the regulatory philosophy adopted by the NRC generally involves “inferring licensee performance from . . . plant inspections and other routine assessments.”40 The General Accounting Office criticized this approach as not paying adequate attention to management issues. The revised (perfor-
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
57
mance-based) regulatory process currently being implemented at the NRC recognizes the importance of safety culture and other cross-cutting issues.41 However, it remains to be seen how effective performance-based regulation will actually be at addressing such cross-cutting issues.
VII. Conditions Favorable to Safety In all three case studies, many aspects of safety appear to have been maintained, or continued pre-deregulation improving trends. In particular, both the air and rail industries in the U.S. had better safety records after deregulation than before, despite problems in particular industry segments, such as new-entrant airlines or companies that had recently undergone mergers and acquisitions. The safety of the U.K. nuclear power industry has also shown improvement in several areas, although concerns over the management of change have surfaced. Thus, if managed properly (by both the industry and safety regulators), deregulation need not be detrimental to safety. owever, it would be a mistake to conclude that deregulation does not create challenges to safety. By its very nature, deregulation stimulates reorganization and restructuring, and as we have seen, the resulting management changes pose a threat to maintaining former levels of safety, at least in the short run. In addition, in all three case study industries, dereg-
H
58
ulation occurred in parallel with other changes that were beneficial for safety, which may have mitigated or counteracted any safety problems due to deregulation and restructuring. Therefore, it is not possible to conclude that deregulation was necessarily the cause of the observed safety improvements. In the aviation industry, the primary condition favorable to safety was the decades-long
some adverse safety consequences of deregulation, such as disruptions due to mergers and acquisitions. In the rail industry, one major factor conducive to safety was simply the improved financial performance of the industry after deregulation (largely due to eased restrictions on abandonment of unprofitable routes). Moreover, rail deregulation took place at a time when the FRA was becoming more active with regard to safety regulation, which may well have had a beneficial effect. inally, in the United Kingdom, the years immediately following nuclear power privatization were accompanied by extensive subsidies that protected the financial health of the industry. Downsizing and other challenges to safety might well have been greater in the absence of these subsidies. The proactive role of the NII in planning for and monitoring the safety-significant impacts of privatization may also have reduced or mitigated some safety problems. The extent to which the U.S. nuclear power industry will benefit from similar changes remains to be seen. The nuclear power industry may well see continued technological improvements, more effective safety regulation, improved financial profitability, and/or favorable economic treatment such as recovery of stranded assets. If such favorable conditions do not materialize, however, deregulation could have more detrimental impacts on safety in the U.S. nuclear power industry than
F
trend toward safety and technology improvements leading up to, and continuing after, deregulation. Among the changes observed in the decade or so immediately after deregulation were: improved training (due to both high-fidelity flight simulators and crew resource management techniques); increased automation in the cockpit; improved equipment reliability (especially engine reliability); and a better understanding of effective preventive maintenance practices. The enhanced levels of safety due to these technological improvements may well have masked
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
The Electricity Journal
was observed in the case study industries.
VIII. Summary and Recommendations Overall, the experience of the case study industries indicates that economic deregulation need not be incompatible with a reasonable safety record. However, this case study analysis also identified a number of safety problems associated with deregulation in the case study industries, which are potentially relevant to the U.S. nuclear power industry. Specifically, the magnitude and speed of the changes associated with deregulation in the case study industries sometimes proved difficult for both companies and safety regulators to manage effectively and safely. Since such safety problems were observed in each of the three industries we studied, safety clearly should not be taken for granted after deregulation. One of our most robust findings is that deregulation has significant effects on corporate culture, management, and organizational issues, which were found to be critical to safety after deregulation in all three case study industries. Therefore, both the nuclear power industry and its safety regulators may wish to pay particular attention to these issues in the transition to a deregulated industry. For example, the experience of the case study industries points out the importance of safety and change management in planning
May 2001
downsizing efforts, mergers, and acquisitions. We also note the importance of the NRC’s new oversight system. The switch to risk-informed, performance-based safety regulation has the potential to be beneficial for both safety and cost. However, its implementation should be monitored to determine whether the revised system adequately captures any observed safety culture
ance, and NRC financial qualifications assurance. Overall, the magnitude and speed of the changes associated with deregulation can be expected to create major challenges to the management of safety by the U.S. nuclear power industry and its safety regulators, as in all three case study industries. We hope that careful review and study of past problems will make it possible for industry and regulators to identify proactive ways to minimize similar safety problems in the U.S. nuclear power industry, where their consequences are potentially severe. j Endnotes:
problems, is sufficiently proactive given the speed of changes associated with deregulation, and effectively targets scarce agency resources at the most significant safety problems. Finally, further research on the effects of electricity deregulation may be worthwhile, for example, to evaluate whether there are links between financial pressures and safety in the nuclear power industry, as in the air and rail industries. Ideally, such research should also address the effects of factors such as stranded asset recovery, decommissioning funding assur-
1. In cost-of-service regulation, which has been the standard in the United States throughout the 1900s, companies are allowed to recover those operating expenses that are deemed “prudent” by the regulator, in addition to a reasonable rate of return on their capital investments. This is accomplished by regulatorapproved price increases, which are possible because the companies are not competing with other providers on the basis of price. 2. Clinton V. Oster, Jr., John S. Strong, and C. Kurt Zorn, Why Airplanes Crash: Aviation Safety in a Changing World (New York: Oxford University Press, 1992). 3. Ian Savage, The Economics of Railroad Safety (Boston: Kluwer Academic Publishers, 1998). 4. U.K. Health and Safety Executive, Safety Management Audit of British Energy Generation Limited and British Energy Generation (UK) Limited, http:// www.hse.gov.uk/nsd/beaudit.htm, (April 26, 2001). Balliol, U.K., Nuclear Safety Directorate, 2000. 5. D. Mark Kennet, Did Deregulation Affect Aircraft Engine Maintenance? An
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
59
Empirical Policy Analysis, 24(4) Rand J. Econ., 1993, at 542–58.
their incremental or marginal operating costs.
6. David Stellfox, STP Finds Risk Ranking Components Pays, Even Before NRC Changes, Inside N.R.C., Dec. 20, 1999, at 3–4.
15. Nuclear Energy Institute, Nuclear Energy: 2000 and Beyond—1999 Update to A Strategic Direction for Nuclear Energy in the 21st Century (NEI 98-W1), 1999; see also Moody’s Investors Service, Nuclear Update: A Buyer’s Market for Nuclear Plants (New York, 1999).
7. Nainish K. Gupta and Herbert G. Thompson, Jr., The Market Value of Nuclear Power, Elec. J., Oct. 1999, at 38–45. 8. John S. Carroll, John Sterman, and Alfred A. Marcus, Playing the Maintenance Game: How Mental Models Drive Organizational Decisions, in Debating Rationality: Nonrational Aspects of Organizational Decision Making ( Jennifer J. Halpern and Robert N. Stern, eds., Ithaca, NY: ILR Press, 1998). 9. U.S. Nuclear Regulatory Commission, License Renewal, 2001, http:// www.nrc.gov/NRC/REACTOR/LR/ index.html (April 26, 2001). 10. Paul L. Joskow, Electricity Sectors in Transition, 19(2) Energy J., 1998, at 25–52. 11. Devra Golbe, Risk Taking by Firms Near Bankruptcy, 28 Econ. Letters, 1988, at 75–79. 12. Nancy L. Rose, Financial Influences on Airline Safety, in Transportation Safety in an Age of Deregulation (Leon N. Moses and Ian Savage, eds., Oxford: Oxford University Press, 1989).
16. Supra note 7. 17. Supra note 10. 18. Private conversation. 19. Private conversation. 20. Supra note 10. 21. U.K. Health and Safety Executive, supra note 4. 22. Michael R. Fox, Nuke Plant Staffing: How Low Can You Go?, Pub. Util. Fortnightly, Mar. 15, 1999, at 22–25. 23. Joseph F. Schuler, Jr., Workforce Management: Human Resources VPs Speak, Pub. Util. Fortnightly, June 15, 1999, at 56–65. 24. William McKinley, Carol M. Sanchez, and Allen G. Shick, Organizational Downsizing: Constraining, Cloning, Learning, 9(3) Academy of Mgmt. Exec., 1995, at 32–42. 25. Richard D. Cudahy, The Folklore of Deregulation, 15 Yale J. on Dereg., 1998, at 27–42.
13. Devra Golbe, Product Safety in a Regulated Industry, 21 Econ. Inquiry, 1983, at 39–52.
26. Peter Passell, When Mega-Mergers are Mega-Busts, N.Y. Times, May 17, 1998, Sec. 4, at 18.
14. Stranded cost recovery provides plant owners the opportunity to recover their capital costs, and thereby avoid going into competition with large debts that they cannot readily repay from future revenues. Plant sales significantly below book value accomplish the same effect (as did privatization in the United Kingdom), since new owners are unencumbered by high initial construction costs. In such cases, the competitiveness of nuclear power stations will be determined primarily by the difference between the revenues that they can produce and
27. Private conversation.
60
28. U.S. Office of Technology Assessment, Safe Skies for Tomorrow: Aviation Safety in a Competitive Environment, Washington, DC, U.S. Government Printing Office, 1988. 29. Ron Westrum, Human Factors Experts Beginning to Focus on Organizational Factors in Safety, ICAO J., Oct. 1996, at 6–8, 26. 30. Arnold Barnett and Mary Higgins, Airline Safety: The Last Decade, 35 Mgmt. Sci., 1989, at 1–21.
31. Robert R. Gray, Aviation Safety: Fact or Fiction, Technology Rev., Aug./Sept. 1987, at 33–40. 32. Kevin Allars, The Regulatory Perspective of Alliancing and Partnering in the Nuclear Supply Chain, presentation to British Nuclear Industry Forum Symposium, March 1999. 33. James Earl Anderson, Making Operational Sense of Mergers and Acquisitions, Elec. J., Aug./Sept.1999, at 49–59. 34. William S. Raughley, The Effects of Deregulation of the Electric Power Industry on the Nuclear Plant Offsite Power System: An Evaluation, U.S. Nuclear Regulatory Commission, Washington, DC, 1999. 35. Kenneth C. Rogers, NRC’s Concerns About Electricity Restructuring, Elec. J., Aug./Sept. 1997, at 22–26. 36. One on One: Entergy Corp.’s Don Hintz Sees Competitive Nuclear Power Plants in America’s Energy Future, Nuclear Energy Insight, Aug. 1999, at 3–4. 37. Supra note 7. 38. U.S. Nuclear Regulatory Commission, Information Digest (NUREG-1350), Washington, DC, Vol. 11, 1999. The NRC Office of the Chief Financial Officer provided data for budget year 2000. 39. Nuclear Energy Institute, A New Regulatory Oversight Process: Toward RiskInformed, Performance-Based Assessment, Inspection and Enforcement, Washington, DC, 1998. 40. General Accounting Office, Nuclear Regulatory Commission: Preventing Problem Plants Requires More Effective Action by NRC, testimony GAO/T-RCED-98252 (1998), Washington, DC, U.S. Government Printing Office 41. Patrick Baranowsky et al., Technical Framework For Licensee Performance Assessment, Washington, DC, U.S. Nuclear Regulatory Commission, 1999, http://www.nrc.gov/NRC/ COMMISSION/SECYS/secy1999-007/ 1999-007scy_attach.pdf (April 26, 2001).
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
The Electricity Journal
May 2001
Deregulation and Nuclear Power Safety: What Can We Learn from Other Industries? An examination of the U.S. aviation and rail industries and the U.K. nuclear power industry provides evidence that changes associated with deregulation can be expected to create major challenges to the management of safety by the U.S. nuclear power industry and its safety regulators. Vicki Bier, James Joosten, David Glyer, Jennifer Tracey, and Michael Welsh
T
he effects of economic deregulation in electricity generation may be seen across a broad spectrum of safety-related areas in the U.S. nuclear power industry. For example, while economic deregulation could increase a company’s motivation to avoid costly safety problems, it could also result in downsizing that could increase human error rates. Since deregulation reduces the ability of licensees to automatically pass along costs to ratepayers for safety-related activities, it can also influence decisions on plant upgrades and safety improvements. Finally, economic deregulation could affect the
workload and requirements imposed on government safety regulators. This article attempts to shed light on the possible consequences of electricity deregulation for nuclear power safety. We are particularly interested in the effects of industry restructuring in response to deregulation on factors such as: financial pressures; equipment maintenance, aging, and failure rates; mergers, acquisitions, and corporate restructuring; and factors relevant to human performance (e.g., downsizing, use of contractors, and labor relations). Because deregulation of the U.S.
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
49
electricity industry is very recent, this article adopts a historical approach, focusing on the effects of economic deregulation on safety in other deregulated industries. Using both literature reviews and interviews with industry representatives, evidence is analyzed regarding the effects of deregulation in three other technologically sophisticated, safety-critical industries: the U.S. aviation and rail industries; and the U.K. nuclear power industry. In the course of the case studies, we reviewed more than 250 documents and interviewed senior industry representatives with direct experience in safety issues. This approach made it possible for us to move beyond speculation about the possible adverse safety consequences of deregulation, to identify situations in which particular adverse safety consequences actually occurred. Hopefully, the results of this study will make it possible to anticipate any adverse effects of deregulation on nuclear power safety, and develop strategies for the industry and safety regulators to detect, monitor, and mitigate those effects. efore presenting our findings, there are several important issues to note. First, like the other case study industries, the U.S. nuclear power industry is not being deregulated with respect to safety, since the U.S. Nuclear Regulatory Commission (NRC) still maintains its safety oversight and regulatory roles. Second, the U.S. electricity industry is currently in a state of flux, and the degree and timing of deregulation differ from
B
50
state to state. Therefore, we do not attempt a comprehensive analysis of the specific situation in each state, but rather assess the relevance of experiences in other industries to the U.S. nuclear industry as a whole. Finally, we emphasize that as the electricity generation industry in the United States transitions from an extended period of cost-of-service regulation1 to a less economically regulated mode of operation, the
Generating companies that cannot achieve operating costs at or below the market price for electricity will not be competitive.
incentives for nuclear generating companies will change dramatically. In particular, generating companies that cannot achieve operating costs at or below the market price for electricity in their region will not be competitive. On the other hand, companies will also have stronger incentives to achieve operating costs substantially below market prices, since any productivity improvements or cost savings will contribute directly to increased profit. With this backdrop in mind, we now review the experiences of the case study industries in safetyrelated areas to help assess how
the ongoing restructuring of the electricity industry is likely to affect future operations, safety, and financial performance of nuclear power plants. This article also discusses the favorable conditions that accompanied deregulation and/or restructuring in the three case study industries (which may have mitigated some adverse safety consequences), and attempts to assess whether similarly favorable circumstances are likely to exist in the U.S. nuclear power industry. Finally, we end with some recommendations for ensuring nuclear power safety as the process of electricity deregulation unfolds.
I. Background The airline industry was economically deregulated with passage of the Airline Deregulation Act of 1978, with the Federal Aviation Administration (FAA) maintaining continued responsibility for aviation safety. Deregulation resulted in a number of changes in the nature of the airline industry. In particular, the volume of airline transportation increased greatly. Deregulation also encouraged a large number of new entrants to the aviation industry, and triggered a large number of mergers, acquisitions, and bankruptcies. Overall accident rates in the airline industry had been steadily declining for many years, and that trend continued after deregulation.2 However, the rates of both near midair collisions and “runway incursions” increased during the mid-1980s. Detailed examination
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
The Electricity Journal
of postderegulation experience in the aviation industry also reveals a number of other changes with adverse safety impacts, as discussed below. he railroad industry was economically deregulated with passage of the Railroad Revitalization and Regulatory Reform Act in 1976, and the Staggers Act in 1980. The Staggers Act in particular gave railroads greater freedom to set rates, and enhanced their ability to abandon unprofitable lines. In addition, the Federal Railroad Administration (FRA) had been established in 1967 with responsibility for safety regulation. After the Staggers Act, the financial performance of the railroad industry improved significantly, and expenditures on track increased dramatically.3 Track-related accidents peaked in 1978 (because the poor financial health of railroads in the 1960s and 1970s had resulted in reduced expenditures on track maintenance), and had dropped dramatically by 1982. It is difficult, however, to determine whether the increased FRA emphasis on safety regulation or the improved financial health of railroads was primarily responsible for the improvements in safety observed after passage of the Staggers Act. The U.K. electricity supply industry, which had been governmentowned and -operated, was largely privatized and deregulated with passage of the Electricity Act in 1989, when all conventional power stations were assigned to two new private companies. However, due to concerns over potential nuclear liabilities, nuclear power stations
T
May 2001
remained under government ownership at that time. Based on their favorable performance records, the government subsequently privatized the newer reactors in 1996, forming a new nuclear company called British Energy (the older Magnox plants, which had worse economic prospects, remained in government ownership). The U.K. Nuclear Installations Inspectorate (NII) was actively involved in planning for and monitoring the
After deregulation, both the aviation and rail industries exhibited shifts in allocation of resources for maintenance.
privatization process, and relicensed the privatized reactors. Employment in the industry decreased significantly, beginning before deregulation and continuing afterward. The privatized nuclear power plants have done well on many measures of economic and safety performance. However, in response to observed problems, the NII audited British Energy in 1999.4 The audit findings included excessive reliance on contractor staff, and lack of definition of the skills required for license responsibilities. As a result, further reduction in staffing was restricted
pending development of a satisfactory staffing plan.
II. Equipment Maintenance and Aging After deregulation, both the aviation and rail industries exhibited significant shifts in allocation of resources for equipment maintenance. For example, one study of the aviation industry found that airlines reduced the frequency of engine overhauls after deregulation, but did not experience a higher rate of engine failures as a result,5 apparently because of improved efficacy and efficiency in maintenance programs. The rail industry undertook even more dramatic reallocation of resources, with workforce reductions of more than 50 percent (partially due to work-rule changes), and increased capital expenditures on track by nearly a factor of five. The increases in track maintenance are widely viewed as being responsible for much of the observed improvement in rail safety. o significant changes in maintenance budgets were observed in the U.K. nuclear industry. There appears to be a realization in the United Kingdom that high levels of reliability are critical to keeping nuclear units on line and producing revenue. In fact, due to the high costs associated with unplanned outages, the new competitive market has increased the incentives to avoid unplanned shutdowns, even at the expense of delayed startups if necessary to “do the job right.” Simi-
N
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
51
larly, there are stronger incentives today to comply with technical specifications and avoid regulatory shutdowns. eregulation systematically changes the incentives faced by U.S. nuclear generating companies with regard to maintenance expenditures. Some of these incentives will encourage minimization of maintenance expenditures, with possible adverse safety consequences, while others will encourage increased maintenance or capital investments in particular areas. Cost minimization will be more important after deregulation than before, so some safety-related maintenance activities may be cut back or eliminated. For example, South Texas Project (STP) managers expect to achieve significant annual savings by “optimizing maintenance frequencies, [reevaluating] how detailed post-maintenance testing needs to be and reconsidering how detailed plant documentation packages need to be” for components with little or no risk significance.6 However, the economic value of maintenance activities that do yield significant benefits in productivity, asset protection, or avoidance of regulatory sanctions is likely to be greater after deregulation than before, since any savings will contribute directly to increased profit. For example, Gupta and Thompson list “capital upgrades to decrease costs and increase the capacity factor” as a possible area for improvement in the electricity industry.7 Thus, initiatives to optimize maintenance efforts are likely to be seen in the U.S. nuclear
D
52
power industry, as in the aviation and rail industries. Although it is possible to achieve maintenance reductions without jeopardizing safety, cuts can sometimes go too far, since feedback indicating the adverse effects of deferred maintenance may be ambiguous or delayed. Thus, Carroll et al. note that some organizations in both the nuclear power and chemical processing industries “fail . . . to give due consideration
In the U.S. nuclear industry, there has been strong interest in life extension as deregulation proceeds.
to preventive maintenance in organizational decisions,” even though “sparse allocation of resources to maintenance is not a rational strategy for the organization as a whole.”8 With respect to equipment aging, no significant evidence was found of a trend toward use of older equipment after deregulation in the rail industry. In fact, although aging and deterioration of track had been a significant problem prior to deregulation, the quality of track maintenance improved significantly after restructuring due to the improved financial health of the major railroads and
the abandonment of unprofitable lines. The aviation industry did experience a trend toward longer operating lives for aircraft. The available statistics do not show a clear link between aging aircraft and increased risk, presumably due to effective preventive maintenance. However, two interview respondents identified concerns about aging aircraft as an important safety issue in the future, especially at smaller carriers that may not have the resources to adequately maintain their aircraft as they age. In the U.S. nuclear industry, there has been strong interest in life extension as deregulation proceeds. Because the marginal costs of nuclear power are generally favorable compared to coal and oil, companies are interested in getting the most out of their nuclear plant assets by increasing their lifetimes. As of this writing, owners of 43 plants have obtained or formally expressed interest in license renewal.9 As a result, some industry observers have argued for additional research on risks associated with aging nuclear power plants. y contrast, companies with less profitable plants (and plants facing large short-term costs or large uncertainties) have in some cases decided to close those plants. Joskow notes that “The pressures of emerging competition have . . . led to the early retirement of at least four nuclear reactors in the U.S. More are likely to follow this path in the next few years.”10 Plant owners considering early shutdown may be reluctant
B
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
The Electricity Journal
to undertake discretionary safetyrelated investments during the last few years of operation. Thus, deregulation may simultaneously accelerate the rate of plant closures and also increase the interest in life extension, both of which can pose safety issues.
III. Financial Pressures Economic theory suggests a link between poor profitability and inadequate safety investments, predicting for example that firms facing possible bankruptcy are more likely to take chances,11 such as deferring spending on discretionary safety improvements. We found evidence in two of the three case studies that financial difficulties may indeed be associated with poor safety records. For the aviation industry, statistical analyses of airline data found lower profitability to be associated with higher accident and incident rates, particularly for smaller carriers.12 In fact, based on her results, Rose specifically recommended “more intense scrutiny of the safety practices of financially marginal carriers.” In the rail industry, the poor safety performance of the railroads in the period leading up to deregulation is generally believed to be associated with their poor financial health during the same period. In addition, studies of accident rates in the railroad industry prior to deregulation found that worse financial health was generally associated with higher accident rates.13 This relationship was strongest for those
May 2001
railroads that were actually unprofitable, rather than only marginally profitable. We found no evidence that financial difficulties were associated with safety problems for the U.K. electricity industry. This is because the privatized nuclear plants became very profitable following deregulation, partly due to the nature of the U.K. privatization process, which provided significant subsidies for nuclear power in
Economic theory suggests a link between poor profitability and inadequate safety investments.
the years immediately following privatization. In the U.S. electricity industry, a central issue for the competitiveness of nuclear power is the fact that the construction costs of most nuclear power stations were high, usually much higher than anticipated. Under cost-of-service regulation, capital costs (a rate of return on capital, plus depreciation) are included in regulated rates. Continued recovery of these capital costs would require electricity prices higher than could likely be attained under competition. Therefore, in most states with deregulated electricity markets, owners of
such overvalued plants have been allowed to collect “stranded cost” payments as part of approved deregulation plans, to compensate them for a portion of the returns that they would have received under cost-of-service regulation.14 The consensus seems to be that with such stranded cost recovery, “most nuclear units should be able to compete with other sources of electricity in a competitive generation business.”15 Thus, the most successful nuclear generating companies are expected to have strong operating profits after deregulation, and these profits will not be artificially limited by economic regulation. evertheless, deregulation can be expected to create financial difficulties for at least some plants during the transition period. The statistically significant relationships between financial difficulty and safety problems observed in two of the three case studies suggest that such financial stresses may contribute to declining safety margins. These financial pressures also create the possibility of further plant closures. Like impending bankruptcy, anticipated plant closures can reduce the incentives to invest in discretionary safety improvements during the last few years of a plant’s operation, since these costs can be amortized over only a short time. (It should be noted, however, that minimum safety standards would need to be maintained to satisfy the conditions of the plant’s operating license.) The nuclear power industry today also faces substantial finan-
N
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
53
cial uncertainties. Gupta and Thompson cite uncertainties about “the tax consequences (and time requirements) of transferring ownership, revaluing the asset, and transferring the decommissioning funds.”16 Similar uncertainties relate to plant life extension, nuclear waste disposal, and future attitudes to nuclear safety. Such uncertainties (and the corresponding financial risks) will generally lead companies to adopt higher discount rates. For example, Joskow states that under deregulation, electricity generators “will base decisions on higher costs of capital, shorter payback periods and more stringent cash flow requirements.”17 This can be expected to result in a trend toward shorter-term planning, with long-term investments (possibly including some safetyrelated investments) requiring high levels of benefit to be economically justified.
IV. Human Performance and Labor Issues The three case study industries differed widely with regard to the effects of deregulation on human performance and experience levels. In the aviation industry, rapid industry growth after deregulation created significant growth in total employment. Thus, several observers of the aviation industry raised concerns about the human performance implications associated with high turnover rates and hiring of less experienced personnel, especially at smaller carriers. By contrast, both the rail indus-
54
try and the U.K. electricity supply industry experienced dramatic downsizing. Rail industry employment decreased by more than half in the first decade after deregulation. Nonetheless, the rail industry was generally able to maintain high levels of employee experience, with few concerns being raised about employee inexperience. However, investigations of major railroad accidents in recent
that “Fatigue issues are much worse since deregulation.”19 Dramatic employment cuts were also observed following deregulation and privatization of the electricity industry, both in the United Kingdom and elsewhere.20 In particular, total employment in the U.K. electricity industry fell by roughly half in the five years immediately after the restructuring of the industry in the late 1980s, following a long period of significant but slower downsizing. n the United Kingdom, the NII has expressed concern that rapid downsizing, accompanied by increased use of contractors, may have left the U.K. nuclear operating companies with marginal levels of experience and expertise to safely operate their nuclear power plants. These shortfalls occurred in part because anticipated workload reductions did not materialize as rapidly as expected after workforce reduction measures had been implemented. The NII has also raised concerns about the qualifications and safety culture of the contractors, and about possibly excessive use of overtime.21 Such concerns have led the NII to begin requiring prior approval of significant changes in staffing levels, to prevent similar problems in future. In the U.S. nuclear power industry, downsizing is currently one of the most common approaches for reducing costs. For example, Fox states that “total labor requirements on average across the industry have fallen by 14 percent during the past five years.”22 Moreover, the trend to downsizing
I
years have raised concerns about excessive downsizing, resulting in understaffing, fatigue, and inadequate supervision. As a result of those findings, the FRA has increased its focus on fatigue in recent years; a representative of the FRA also indicated concern about increased use of overtime.18 Significant concerns about fatigue also arose in relation to work-rule changes (especially minimum crew sizes). In particular, rail industry labor representatives stated that the steady decrease in crew sizes had increased problems of fatigue, with one union representative specifically commenting
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
The Electricity Journal
appears to be continuing, as it did in the United Kingdom. For example, utility vice presidents surveyed by Public Utilities Fortnightly generally reported downsizing of 10 percent or more in their companies.23 equired plant staffing levels appear to vary as a function of factors such as size, age, reactor type, number of units, and utility characteristics. This creates the risk that a plant attempting to emulate the staffing levels of the low-cost performers in the industry, without correcting for such differences, could make excessive cuts in safety-critical areas. More generally, McKinley et al. note that determining appropriate staffing levels is especially difficult when “organizational units lack clear performance standards,” and when “the relationship between means and ends is ambiguous.”24 Such “gray areas” in performance standards and ambiguous relationships between means and ends may be more typical of safety-related functions than of activities necessary for day-to-day production.
R
V. Influences on Corporate Culture: Mergers, Acquisitions, and New Entrants Issues of corporate culture were identified as being critical to safety in all three case studies. In both the U.S. aviation and rail industries, these concerns were primarily associated with mergers and acquisitions, in which the rapid pace of change sometimes resulted in periods of confusion that com-
May 2001
promised safety. Such mergers and acquisitions were widespread in the post-deregulation period; in fact, Cudahy goes so far as to suggest that “Nothing is certain about deregulation except the mergers that follow.”25 In the rail industry, Passell described the acquisition of the Southern Pacific and Chicago and North Western railroads by Union Pacific as “the most spectacular
merger fiasco of modern times,” resulting in several fatal accidents as well as significant freight delays.26 In particular, he noted, “Union Pacific’s by-the-book culture clashed badly with Southern Pacific’s, where managers had long been accustomed to making do with chewing gum and bailing wire.” Concern about possible negative effects of mergers prompted the Federal Railroad Administration to propose regulations requiring advance approval of consolidations, based on a satisfactory plan for safety management during restructuring. Several interview respondents urged caution regard-
ing the potential safety problems associated with mergers and acquisitions in the nuclear power industry. Rail industry interviewees also expressed widespread concern about underreporting of safety problems. One former railroad employee stated that “Deregulation came with low-grade fear to keep people from reporting violations.”27 In the aviation industry, the U.S. Office of Technology Assessment similarly noted that mergers had a potentially adverse impact on corporate culture, and hence on aviation safety.28 In fact, post-merger lack of coordination has been associated with an increased rate of pilot deviations.29 Several respondents in the aviation industry interviews also mentioned this issue, one of them describing issues of corporate culture as the biggest single problem associated with mergers and acquisitions. (However, another respondent with significant experience in both safety regulation and airline management commented that overall, “consolidation has led to the creation of stronger companies with more resources at their disposal,” suggesting that the safety problems associated with consolidation may be largely short-term in nature.) n addition, new entrant airlines have generally been associated with increased risk; for example, Barnett and Higgins found that the fatality risk for new entrants was 12 times higher than that for “trunklines.”30 Corporate culture issues appear to have played a part in this; for example, Gray stated
I
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
55
that “Many new carriers have little notion of or experience in how an airline must be run.”31 In the United Kingdom, the nuclear power industry is already heavily consolidated due to decisions made during the privatization process. There also, however, further consolidation is taking place, primarily in the form of alliances and partnerships, and the NII has raised concern about this trend. In particular, Allars has emphasized the “need for partners to fully understand the hazard/risks and to share in safety culture initiatives and goals set by the licensee.”32 any observers expect to see significant mergers and acquisitions within the U.S. electricity and nuclear power industries as deregulation proceeds. New ownership, mergers and acquisitions, changes in management teams, and even major changes in management philosophy or shareholder expectations (all of which can be expected to take place in the U.S. nuclear industry) can all affect corporate culture. In particular, Anderson, a utility industry consultant, discusses a number of problems that can arise from inadequate planning or implementation of mergers and acquisitions. He cites “poor cultural integration” as one possible pitfall, noting also that “risk management, regulatory compliance, environmental affairs, and internal audit functions” may not be implemented uniformly across the resulting merged company.33 Note, too, that divestiture and unbundling can also pose risks. For
example, Raughley cites possible adverse effects of divestiture on offsite power availability, observing that “Licensees are selling their generating facilities that supply offsite power to the nuclear plants.”34 He observes that such arrangements can reduce coordination, and thereby extend the time needed to recover from events such as loss of offsite power. The increase in power flows over the
M
56
transmission network due to increased electricity competition can also degrade grid reliability, posing a risk for the safe operation of nuclear power stations.35 Despite the possibility of significant short-term problems, however, many observers expect consolidation to be good for the industry in the long run. For example, Entergy’s Hintz anticipates that consolidation will be beneficial, arguing that “there is an advantage to having all the plants operated by operators with good performance records—companies that have a real focus on nuclear”;36 Gupta
and Thompson make similar observations.37
VI. Experiences of Safety Regulators In all three case study industries, deregulation had significant implications for safety regulators. In the aviation industry, deregulation significantly increased the workload of the FAA at a time when agency budget and staffing levels were already under pressure. Part of the increase in workload was due to dramatic increases in both sales volume and number of airlines after aviation deregulation, changes that are not expected to occur in the nuclear power industry. However, other factors (such as increased reliance on contractor maintenance, increased pilot turnover, and an aging fleet of aircraft) increased the burden on FAA inspectors. FAA staffing levels were also cut; for example, the number of FAA inspectors was reduced by roughly 20 percent in the years immediately following deregulation. As a result, the FAA has acknowledged that it was unprepared for the effects of deregulation. Moreover, inadequate FAA oversight has occasionally been cited as having contributed to particular accidents, such as the Air Florida crash in Washington, DC, in 1982. ome interview respondents from the aviation industry specifically recommended that the NRC attempt to learn from the FAA’s experiences after deregulation, and emphasized the importance of maintaining a highly
S
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
The Electricity Journal
trained and skilled regulatory workforce. In particular, one individual who had formerly worked for the FAA noted that the agency today has a significantly larger oversight infrastructure than at the time of deregulation. He suggested that the NRC might similarly need to strengthen its staffing to ensure effective oversight of facility operations, monitoring of regulatory compliance, and maintenance of good working relationships with regulated companies. nlike the FAA, the Federal Railroad Administration had significantly increased both the number of safety regulations and the level of inspection effort around the time of rail deregulation. These changes are often cited as contributing to the post-deregulation improvement in rail safety. Nevertheless, the FRA has sometimes been criticized for being inflexible and unresponsive to changes in technology, and for relying too heavily on punitive sanctions as the primary means to obtain compliance. As a result, in 1993 the FRA instituted a more cooperative approach among regulators, railroads, and labor in addressing safety problems; several rail industry interviewees stressed the benefits of such cooperation. The FRA has also recently increased its emphasis on human performance issues such as fatigue, based on the results of accident investigations over the last few years. In the United Kingdom, the Nuclear Installations Inspectorate takes a less prescriptive approach than the NRC, and generally has
more cooperative relationships with its licensees. It is also noteworthy that the NII was actively involved in planning for and monitoring the privatization process in the United Kingdom, and actually increased its staffing levels modestly in anticipation of increased regulatory workloads. In the United States, staffing and budget cuts of roughly 20 percent have already taken place at the
U
May 2001
NRC over the past seven years.38 As a result, current staffing levels are comparable to those in 1978, before the accident at Three Mile Island (TMI). Recent budgets are also at historically low levels, 40 percent below the post-TMI peak in the early 1980s. These cuts are of the same order of magnitude as those at the FAA in the years immediately following airline deregulation, which were indicated as a cause of concern in the airline industry. As noted by industry representatives, some reductions in staffing and budgets at the NRC may be appropriate, given the “safety record of the commercial nuclear
industry [and] the maturity of the technology.”39 However, deregulation can be expected to cause workload increases for the NRC, e.g., due to the need for approval of license transfers and review of decommissioning funds associated with mergers and acquisitions, and the large number of anticipated requests for life extension. In addition, if reduced safety margins occur at some companies, this might increase workloads for NRC safety regulators. As at the FAA, therefore, more effective targeting of scarce agency resources is likely to be required after deregulation. One way to achieve that may be through the NRC’s recent adoption of new risk-informed, performance-based approaches to safety regulation, which some observers see as a promising way to target resources at the most risksignificant problems. If successful in achieving that goal, riskinformed and performance-based regulation could allow both the NRC and nuclear plants to become more cost-effective without compromising safety. Unlike safety regulators in the rail and U.K. nuclear power industries, which have recently implemented requirements for prior approval of major organizational changes, the regulatory philosophy adopted by the NRC generally involves “inferring licensee performance from . . . plant inspections and other routine assessments.”40 The General Accounting Office criticized this approach as not paying adequate attention to management issues. The revised (perfor-
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
57
mance-based) regulatory process currently being implemented at the NRC recognizes the importance of safety culture and other cross-cutting issues.41 However, it remains to be seen how effective performance-based regulation will actually be at addressing such cross-cutting issues.
VII. Conditions Favorable to Safety In all three case studies, many aspects of safety appear to have been maintained, or continued pre-deregulation improving trends. In particular, both the air and rail industries in the U.S. had better safety records after deregulation than before, despite problems in particular industry segments, such as new-entrant airlines or companies that had recently undergone mergers and acquisitions. The safety of the U.K. nuclear power industry has also shown improvement in several areas, although concerns over the management of change have surfaced. Thus, if managed properly (by both the industry and safety regulators), deregulation need not be detrimental to safety. owever, it would be a mistake to conclude that deregulation does not create challenges to safety. By its very nature, deregulation stimulates reorganization and restructuring, and as we have seen, the resulting management changes pose a threat to maintaining former levels of safety, at least in the short run. In addition, in all three case study industries, dereg-
H
58
ulation occurred in parallel with other changes that were beneficial for safety, which may have mitigated or counteracted any safety problems due to deregulation and restructuring. Therefore, it is not possible to conclude that deregulation was necessarily the cause of the observed safety improvements. In the aviation industry, the primary condition favorable to safety was the decades-long
some adverse safety consequences of deregulation, such as disruptions due to mergers and acquisitions. In the rail industry, one major factor conducive to safety was simply the improved financial performance of the industry after deregulation (largely due to eased restrictions on abandonment of unprofitable routes). Moreover, rail deregulation took place at a time when the FRA was becoming more active with regard to safety regulation, which may well have had a beneficial effect. inally, in the United Kingdom, the years immediately following nuclear power privatization were accompanied by extensive subsidies that protected the financial health of the industry. Downsizing and other challenges to safety might well have been greater in the absence of these subsidies. The proactive role of the NII in planning for and monitoring the safety-significant impacts of privatization may also have reduced or mitigated some safety problems. The extent to which the U.S. nuclear power industry will benefit from similar changes remains to be seen. The nuclear power industry may well see continued technological improvements, more effective safety regulation, improved financial profitability, and/or favorable economic treatment such as recovery of stranded assets. If such favorable conditions do not materialize, however, deregulation could have more detrimental impacts on safety in the U.S. nuclear power industry than
F
trend toward safety and technology improvements leading up to, and continuing after, deregulation. Among the changes observed in the decade or so immediately after deregulation were: improved training (due to both high-fidelity flight simulators and crew resource management techniques); increased automation in the cockpit; improved equipment reliability (especially engine reliability); and a better understanding of effective preventive maintenance practices. The enhanced levels of safety due to these technological improvements may well have masked
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
The Electricity Journal
was observed in the case study industries.
VIII. Summary and Recommendations Overall, the experience of the case study industries indicates that economic deregulation need not be incompatible with a reasonable safety record. However, this case study analysis also identified a number of safety problems associated with deregulation in the case study industries, which are potentially relevant to the U.S. nuclear power industry. Specifically, the magnitude and speed of the changes associated with deregulation in the case study industries sometimes proved difficult for both companies and safety regulators to manage effectively and safely. Since such safety problems were observed in each of the three industries we studied, safety clearly should not be taken for granted after deregulation. One of our most robust findings is that deregulation has significant effects on corporate culture, management, and organizational issues, which were found to be critical to safety after deregulation in all three case study industries. Therefore, both the nuclear power industry and its safety regulators may wish to pay particular attention to these issues in the transition to a deregulated industry. For example, the experience of the case study industries points out the importance of safety and change management in planning
May 2001
downsizing efforts, mergers, and acquisitions. We also note the importance of the NRC’s new oversight system. The switch to risk-informed, performance-based safety regulation has the potential to be beneficial for both safety and cost. However, its implementation should be monitored to determine whether the revised system adequately captures any observed safety culture
ance, and NRC financial qualifications assurance. Overall, the magnitude and speed of the changes associated with deregulation can be expected to create major challenges to the management of safety by the U.S. nuclear power industry and its safety regulators, as in all three case study industries. We hope that careful review and study of past problems will make it possible for industry and regulators to identify proactive ways to minimize similar safety problems in the U.S. nuclear power industry, where their consequences are potentially severe. j Endnotes:
problems, is sufficiently proactive given the speed of changes associated with deregulation, and effectively targets scarce agency resources at the most significant safety problems. Finally, further research on the effects of electricity deregulation may be worthwhile, for example, to evaluate whether there are links between financial pressures and safety in the nuclear power industry, as in the air and rail industries. Ideally, such research should also address the effects of factors such as stranded asset recovery, decommissioning funding assur-
1. In cost-of-service regulation, which has been the standard in the United States throughout the 1900s, companies are allowed to recover those operating expenses that are deemed “prudent” by the regulator, in addition to a reasonable rate of return on their capital investments. This is accomplished by regulatorapproved price increases, which are possible because the companies are not competing with other providers on the basis of price. 2. Clinton V. Oster, Jr., John S. Strong, and C. Kurt Zorn, Why Airplanes Crash: Aviation Safety in a Changing World (New York: Oxford University Press, 1992). 3. Ian Savage, The Economics of Railroad Safety (Boston: Kluwer Academic Publishers, 1998). 4. U.K. Health and Safety Executive, Safety Management Audit of British Energy Generation Limited and British Energy Generation (UK) Limited, http:// www.hse.gov.uk/nsd/beaudit.htm, (April 26, 2001). Balliol, U.K., Nuclear Safety Directorate, 2000. 5. D. Mark Kennet, Did Deregulation Affect Aircraft Engine Maintenance? An
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
59
Empirical Policy Analysis, 24(4) Rand J. Econ., 1993, at 542–58.
their incremental or marginal operating costs.
6. David Stellfox, STP Finds Risk Ranking Components Pays, Even Before NRC Changes, Inside N.R.C., Dec. 20, 1999, at 3–4.
15. Nuclear Energy Institute, Nuclear Energy: 2000 and Beyond—1999 Update to A Strategic Direction for Nuclear Energy in the 21st Century (NEI 98-W1), 1999; see also Moody’s Investors Service, Nuclear Update: A Buyer’s Market for Nuclear Plants (New York, 1999).
7. Nainish K. Gupta and Herbert G. Thompson, Jr., The Market Value of Nuclear Power, Elec. J., Oct. 1999, at 38–45. 8. John S. Carroll, John Sterman, and Alfred A. Marcus, Playing the Maintenance Game: How Mental Models Drive Organizational Decisions, in Debating Rationality: Nonrational Aspects of Organizational Decision Making ( Jennifer J. Halpern and Robert N. Stern, eds., Ithaca, NY: ILR Press, 1998). 9. U.S. Nuclear Regulatory Commission, License Renewal, 2001, http:// www.nrc.gov/NRC/REACTOR/LR/ index.html (April 26, 2001). 10. Paul L. Joskow, Electricity Sectors in Transition, 19(2) Energy J., 1998, at 25–52. 11. Devra Golbe, Risk Taking by Firms Near Bankruptcy, 28 Econ. Letters, 1988, at 75–79. 12. Nancy L. Rose, Financial Influences on Airline Safety, in Transportation Safety in an Age of Deregulation (Leon N. Moses and Ian Savage, eds., Oxford: Oxford University Press, 1989).
16. Supra note 7. 17. Supra note 10. 18. Private conversation. 19. Private conversation. 20. Supra note 10. 21. U.K. Health and Safety Executive, supra note 4. 22. Michael R. Fox, Nuke Plant Staffing: How Low Can You Go?, Pub. Util. Fortnightly, Mar. 15, 1999, at 22–25. 23. Joseph F. Schuler, Jr., Workforce Management: Human Resources VPs Speak, Pub. Util. Fortnightly, June 15, 1999, at 56–65. 24. William McKinley, Carol M. Sanchez, and Allen G. Shick, Organizational Downsizing: Constraining, Cloning, Learning, 9(3) Academy of Mgmt. Exec., 1995, at 32–42. 25. Richard D. Cudahy, The Folklore of Deregulation, 15 Yale J. on Dereg., 1998, at 27–42.
13. Devra Golbe, Product Safety in a Regulated Industry, 21 Econ. Inquiry, 1983, at 39–52.
26. Peter Passell, When Mega-Mergers are Mega-Busts, N.Y. Times, May 17, 1998, Sec. 4, at 18.
14. Stranded cost recovery provides plant owners the opportunity to recover their capital costs, and thereby avoid going into competition with large debts that they cannot readily repay from future revenues. Plant sales significantly below book value accomplish the same effect (as did privatization in the United Kingdom), since new owners are unencumbered by high initial construction costs. In such cases, the competitiveness of nuclear power stations will be determined primarily by the difference between the revenues that they can produce and
27. Private conversation.
60
28. U.S. Office of Technology Assessment, Safe Skies for Tomorrow: Aviation Safety in a Competitive Environment, Washington, DC, U.S. Government Printing Office, 1988. 29. Ron Westrum, Human Factors Experts Beginning to Focus on Organizational Factors in Safety, ICAO J., Oct. 1996, at 6–8, 26. 30. Arnold Barnett and Mary Higgins, Airline Safety: The Last Decade, 35 Mgmt. Sci., 1989, at 1–21.
31. Robert R. Gray, Aviation Safety: Fact or Fiction, Technology Rev., Aug./Sept. 1987, at 33–40. 32. Kevin Allars, The Regulatory Perspective of Alliancing and Partnering in the Nuclear Supply Chain, presentation to British Nuclear Industry Forum Symposium, March 1999. 33. James Earl Anderson, Making Operational Sense of Mergers and Acquisitions, Elec. J., Aug./Sept.1999, at 49–59. 34. William S. Raughley, The Effects of Deregulation of the Electric Power Industry on the Nuclear Plant Offsite Power System: An Evaluation, U.S. Nuclear Regulatory Commission, Washington, DC, 1999. 35. Kenneth C. Rogers, NRC’s Concerns About Electricity Restructuring, Elec. J., Aug./Sept. 1997, at 22–26. 36. One on One: Entergy Corp.’s Don Hintz Sees Competitive Nuclear Power Plants in America’s Energy Future, Nuclear Energy Insight, Aug. 1999, at 3–4. 37. Supra note 7. 38. U.S. Nuclear Regulatory Commission, Information Digest (NUREG-1350), Washington, DC, Vol. 11, 1999. The NRC Office of the Chief Financial Officer provided data for budget year 2000. 39. Nuclear Energy Institute, A New Regulatory Oversight Process: Toward RiskInformed, Performance-Based Assessment, Inspection and Enforcement, Washington, DC, 1998. 40. General Accounting Office, Nuclear Regulatory Commission: Preventing Problem Plants Requires More Effective Action by NRC, testimony GAO/T-RCED-98252 (1998), Washington, DC, U.S. Government Printing Office 41. Patrick Baranowsky et al., Technical Framework For Licensee Performance Assessment, Washington, DC, U.S. Nuclear Regulatory Commission, 1999, http://www.nrc.gov/NRC/ COMMISSION/SECYS/secy1999-007/ 1999-007scy_attach.pdf (April 26, 2001).
Published 2001, Elsevier Science Inc., 1040-6190/01/$–see front matter PII S1040-6190(01)00197-X
The Electricity Journal