Preventing Transboundary Crises: The Management and ...
Published on: Mar 4, 2016
Transcripts - Preventing Transboundary Crises: The Management and ...
Preventing Transboundary Crises: The Management and
Regulation of Setbacks
California State University, East Bay
Setbacks—unanticipated, unwanted, and often sudden checks on moving forward—are common in
crisis prevention. Managing setbacks, however, is rarely discussed. Even less discussed is the central
role that setback management can play in preventing transboundary crises across interdependent
critical infrastructures for electricity, water, transportation, and ﬁnancial services. With a focus on
the ongoing ﬁnancial crisis and on strategies for resilience and anticipation, the article draws out
implications for crisis prevention as the ﬁrst, rather than last, line of defense in managing future
KEY WORDS: organizational setbacks, crisis prevention, high reliability management, disaster and risk
Introduction to Setback Management and Initial Distinctions
Reportedly, Mao Tse-tung’s prime minister, Chou En-lai, famously said when
asked about the signiﬁcance of the French Revolution, “It’s too soon to say.” It is
always too soon to say if today’s crisis will end up viewed later as a major setback on
the long road to success. For example, Three Mile Island was a major nuclear power
failure. Yet nuclear power and environmental groups have recast it as a setback to
nuclear energy’s signiﬁcant role in a post-global climate change world. Given long
enough and a context broad enough, almost any disaster can be a setback on the
path to some blessed prospect.
This article concerns the opposite setback. I will not be examining how to
manage today’s failure so as to become a harbinger for success in the future. Rather,
what follows is about how to manage what is considered a setback today so it does
not become a failure tomorrow or in the near future. In one case, the setback has
to be managed so it does not lead to conditions that could themselves lead to a crisis;
in the other case, the setback would actually lead to crisis, unless directly prevented
from doing so.
Setback management is crisis prevention, but with a difference.2 First, relatively
little has been written about how to manage the inevitable setbacks that occur in the
process of trying to prevent crises. It behooves us to think more about this, if only
because preventing a crisis is preferable to managing one. Second, it is important to
distinguish crisis prevention in terms of whether the setback is itself a harbinger to
something worse or whether it is being managed so as to prevent it from becoming
that harbinger. What is at issue are different mixes of resilience and anticipation
strategies, which in turn determine in what ways crisis prevention is the ﬁrst or last
line of defense against failure.
For the purposes of what follows, setback management means managing events
that are recognized to be on the way to becoming disasters for the system,
unless managed. The events can be those conditions that create a close call
Review of Policy Research, Volume 26, Number 4 (2009)
© 2009 by The Policy Studies Organization. All rights reserved.
458 Emery Roe
or near miss, if not managed, or they could be the actual close call or near miss
leading to something worse unless preventative action is taken. In either case,
the event is unwanted or unexpected—a setback—that could in the absence of
management morph into a crisis. How that management takes place depends
on whether the events are directly or indirectly a precursor to the crisis
To help understand these distinctions within crisis prevention and their role in
addressing future crises, two literatures are drawn upon, one on high reliability
management (e.g., on high reliability organizations [HROs]) and the other on crisis
management. The two literatures intersect in their grappling with the precursors to
disasters and crises. They share concern for what my colleague, Paul Schulman,
calls the precursor zone.
The precursor zone is best thought of as the set of activities of managers when
they are at the cognitive limits of the known patterns and scenarios they use in
decision making. The precursor zone is where the skills of managers to recognize
patterns and formulate scenarios are signiﬁcantly challenged, as these skills are no
longer matched to the intensiﬁed or now-novel tasks they face. Managers continue
to calculate risks and formulate worst-case scenarios, but in uncertain situations that
continually test the limits of their knowledge. We’re hitting the boundaries of
known reality, reported one fund manager concerning the Treasury bond market
during the ﬁnancial crisis (Rappaport & Tuna, 2008, p. 20). In this way, the
precursor zone lies within the domain of professional competence but at its very
edge (for more, see Roe & Schulman, 2008).
This is a dangerous area for managers to operate in and have events happen that
are unanticipated or sudden, as are setbacks. Here the consequences of a setback
and trying to manage it can be so uncertain that the risk of failing to manage—the
product of the probability of failing and the magnitude of that hazard—cannot
be calculated. As such, it is better that a setback be managed before entering
the precursor zone, a kind of risk management involving both anticipation and
resilience, as we shall see.
If, however, the setback occurs in the midst of uncertainty such that its risky
consequences cannot be estimated, let alone measured, then the inability to manage
risks arising from the setback is the hazard which must be managed and which calls
for a different kind of management strategy, one that is more reactive than resilient
and anticipatory. Since the precursor zone has no bright boundary when entering
or leaving, the preference to manage a setback before it becomes a precursor to
disaster is patent.
The literature on HROs is storied with cases of organizations and systems trying
to stay out of the precursor zone, although they sometimes ﬁnd themselves in it.3
The crisis management literature is storied with cases where organizations and
systems are already in the precursor zone, trying not to be pushed over into
outright disaster.4 Within these endpoints, setbacks in crisis prevention ﬁt some-
where between conditions that make for a close call or near miss and a full-blown
crisis, with the feature that they could arguably go either way—a close call or a
disaster—if the setbacks were not managed. Regulation, as we will see, has an
important role, albeit a different kind of regulation than commonly supposed
Managing Organizational Setbacks 459
A short example of how these initial distinctions work in practice is useful. Let us
take as our case the ﬁnancial crisis underway at the time of writing (May 2009). The
crisis has the added virtue of illuminating other distinctions developed later in the
A great deal in the preceding depends on what is taken to be the precursor zone,
and this depends on how the disaster is described, in this case the ﬁnancial crisis.
Assuredly, the crisis seems everywhere conﬁrmed. The descriptors are dire: The
ﬁnancial crisis is a “catastrophe,” “meltdown,” and “hurricane,” the “worst since the
Great Depression” and one full of “contagion,” and “herd mentality.”
The facts are also unprecedented: Trillions of dollars promised in loan, credit,
and insurance guarantees to banks and ﬁnancial institutions by a host of national
governments and central banks; far more trillions lost in value of 401k’s, residential
mortgages, and stock market value; and a series of bailouts, the ﬁrst of which in the
United States exceeded what its federal government spent on Medicare, Medicaid,
and Food Stamps that year (e.g., Davies, Atkins, & Mulligan, 2008, p. 31; Marsh,
2008, WK4). Last and its own unique indicator of the crisis, economists are far less
surefooted and more confused: They talk about how markets no longer “reﬂect
fundamentals” and “rational prices,” without explaining how these could even be
markets if this were so. Certainly, the freezing up of credit markets in a ﬁnancial
sector whose raison d’être is to provide credit is nothing less than scandalous.
For our purposes, the best measures of being in the ﬁnancial crisis are not the
facts and views of ﬁnancial journalists or academic economists, but the actual
experience of the bankers, traders, and investors undergoing the crisis at its height
(so far) in September–October 2008. “It feels as if we are 15 minutes away from the
end of the world,” the head of equities at a large UK bank told the Financial Times
concerning the lead up to the ﬁrst major U.S. bailout (FT Reporters, 2008, p. 1).
“The market has changed more in the past 10 days than it had in the previous 70
years,” reports a senior executive at a European investment bank in late 2008 (Thal
Larsen & Guerrera, 2008, p. 12). “The reality is that we are not going to know what
the right price is for years,” says a bond portfolio manager at a major mutual fund
(quoted in Bijaj, 2008, p. A1). “It was just mayhem,” said the CEO of New York-
based hedge fund concerning the markets (Mollenkamp et al., 2008, p. A22);
“People were paralyzed by fear of what could erupt.” “We have no idea of the details
of our derivative exposures and neither do you,” conceded a senior Lehman
Brothers ofﬁcial at a meeting of bankers and regulators just before it collapsed
(quoted in Guerrera & Bullock, 2008, p. 16).
In the midst of one particularly bad trading day at this time, Morgan Stanley CEO
John Mack sent an urgent email to employees: “What’s happening out here? It’s
very clear to me—we’re in the midst of a market controlled by fear and rumors’
(Mollenkamp et al., 2008, p. A22). After the plunge in Morgan Stanley shares, a
senior manager at the Swiss UBS said, “It felt like there was no ground underneath
your feet. I didn’t know where it was going to end” (Sorkin et al., 2008). Nor was he
alone. George Soros (2008a, p. 11), the global investor, argued, “the ﬁnancial crisis
is spinning out of control.” The chair of Morgan Stanley Asia concluded, “We have
gone to the edge of an abyss that few thought was ever possible” (Roach, 2008, p. 26).
460 Emery Roe
Other quotes could be added, but the point remains the same: Those in the midst
of the ﬁnancial crisis—the traders, bankers, and investors—were operating outside
known patterns and scenarios in a domain of thorough-going uncertainty, if not
ignorance. For those operating within—just barely within—their knowledge and
skills base, the challenge was to manage the uncertainty that would spiral out of
control, if action were not taken.
As an example of this kind of preventative action, a major debate over fair-value
accounting of assets (including credit default swaps) also took place within the
ﬁnancial sector in the last quarter of 2008. This accounting, it was argued, created
a destructive downward cycle that had to be stopped if full-blown ﬁnancial collapse
were to be averted.
Fair-value accounting requires ﬁrms to report the majority of their holdings at
current market prices. In a plunging market, this practice of marking-to-market
meant ﬁrms and banks had to write down billions in assets values. The write-downs
in turn increased pressure to lower the ﬁrms’ credit ratings, given their weakened
balance sheets and capital reserves. Lower credit ratings meant higher collateral
requirements on the loans, which sent the ﬁrms scrambling for even more funds,
thereby weakening the ﬁrms further. Government authorities eventually modiﬁed
fair-value accounting practices in late 2008 as a way of managing to break (perhaps
only temporarily) the self-feeding spiral of asset decline (e.g., Hughes, 2008a,
2008b; Scannell, 2008). Here crisis prevention was underway well within a precur-
sor zone, where the failure to be avoided was quite visible to those concerned and
where each setback in the process of obtaining regulatory relief proved to be
What about managing in order to stay out of the precursor zone altogether? Are
there examples from the height of the current crisis, where ﬁnancial professionals
were able to manage a setback in such a way that they not only avoided conditions
that could worsen the crisis, but did so by controlling those conditions? One such
example is from the Lehman bankruptcy.
After the collapse of Lehman Brothers, it became a priority to close out its
trading positions, many of which were very complicated. Yet the transfer of almost
all trading positions held by Lehman to other banks proceeded far better than
initially expected. While a great many problems remain with unwinding Lehman
commitments, the former president of the New York Federal Reserve Gerald
Corrigan (2008, p. 26), reported:
First, in line with [earlier recommendations], the dealer community adopted a common
methodology for the execution of close-out against a defaulted counterparty. Because this
was accomplished prior to the Lehman failure, the extraordinarily complex and delicate
close-out process in the Lehman case is being managed with greater certainty and greater
safety that would otherwise have been the case, helping to contain systemic risk.
This close-out was undertaken primarily by LCH.Clearnet, Europe’s largest
clearer, who “in the space of just over a week, managed down the main risks
associated with the Lehman default without any disruption to the markets and
without having to resort to using the clearer’s main default fund” (Grant & Hughes,
2008, p. 20). Similar success was achieved on the American side by The Depository
Trust & Clearing Corporation in closing out market participants’ exposure in the
Managing Organizational Setbacks 461
United States as a result of the Lehman collapse (Grant, 2008, p. 29). Although
complications with the Lehman bankruptcy will persist for years, what is important
is that the close-outs took place in a context and format where risks could be
managed. Thus, another setback—say, it took two weeks to undertake the close-out
of major risks when one week looked doable—became much less unsettling.5
The difference in crisis prevention depending on whether managers are already
in the precursor zone having to manage setbacks there or they are managing
setbacks so as to stay out of that danger zone takes us directly to issues of resilience
and anticipation in setback management.
Anticipation and Resilience in Managing Setbacks in Crisis Prevention
There is no better place to continue this discussion than with the dictionary deﬁ-
nition of a setback: an unanticipated or sudden check in progress.
The feature of being unanticipated has at least three implications for managing
setbacks in crisis prevention. First, the pressure is to catch setbacks early on. This
means trying to better address the conditions that make for their being unantici-
pated. Second, because setbacks are unanticipated (or sudden, even if anticipated),
managers will ﬁnd them surprising. Third, because uncertainty is higher in the
precursor zone, setbacks and surprises will be more common there.
These factors, in turn, imply that because we cannot anticipate the contingencies
ahead in crisis prevention, we must have the resilience to bounce back from that
which we cannot anticipate (this is a shared concern of both the crisis management
and high reliability management literatures). However, the implication also is that
one cannot get away from a considerable degree of planning ahead when trying to
manage setbacks, because we cannot expect that lack of front-end anticipatory
planning will be compensated, one-for-one, by greater resilience later on after a
crisis hits. Accordingly, planning ahead for setbacks has a much greater chance for
success in high reliability management settings where the goal is to stay out of
unplanned situations that could lead to a crisis.
Here it is productive to engage an important difference between the HRO and
crisis management literatures when it comes to anticipation and resilience strategies
for disaster management. It turns out the respective mix of strategies are opposites,
and never more so than when disaster strikes.
Start with the HRO literature (e.g., Roe & Schulman, 2008). For control room
operators Paul Schulman and I have studied, anticipating the next step ahead is an
instrumental part of bouncing back from a shock that has just happened. Anticipa-
tion is important because bouncing back to the original position may not be the best
position to which to return. The last thing operators would want to do in some cases
is absorb shocks so they can get back to where they started.
Operators are able to be anticipatory and resilient in the same instant because of
their skills in recognizing system-wide patterns and formulating contingency-
speciﬁc scenarios. Translating the patterns recognized and the contingency sce-
narios for the case at hand into the provision of reliable services means operators
both plan the next steps ahead and respond to unanticipated events during the
current or just preceding steps undertaken. That is why managers are so keen to
stay out of their precursor zone: Inside, their ability to bounce back and plan ahead
462 Emery Roe
is compromised. Instead of being resilient and anticipatory, they end up reactive to
the degree they cannot plan ahead and know what to bounce back to. Indeed, a
widening reactivity and with it the desperate search to start all over again becomes
very much a part of deﬁning a crisis.6
Turn now to the wider literature on crisis prevention and see how it discusses the
relationship between anticipation and resilience. A major point in crisis preparation
is to plan ahead in order to be agile and quick (aka resilient) as the crisis hits. You
plan in order not to be reactive. To see how this works, take an example from
another major disaster. A report to the President, The Federal Response to Hurricane
Katrina: Lessons Learned, identiﬁed 11 “critical actions” that had to be completed
before the next hurricane season as key to a “national preparedness system” and a
“culture of preparedness.” A sampling of the actions is provided below:
• Colocate relevant Federal, State, and local decision makers, including leaders
of State National Guards, to enhance unity of effort;
• Pre-position a fully resourced and integrated interagency Federal Joint Field
Ofﬁce to coordinate and, if necessary, direct Federal support to the disaster;
• Ensure situational awareness by establishing rapid deployable communica-
• Designate locations throughout the country for receiving, staging, moving,
and integrating them;
• Identify and develop rosters of Federal, State, and local government personnel
who are prepared to assist in disaster relief;
• Employ all available twenty-ﬁrst-century technologies both to update and
utilize the national Emergency Alert System in order to provide the general
public with advanced notiﬁcation of and instruction for disasters and
• Encourage States to pre-contract with service providers for key disaster relief
• Enhance ongoing review of State evacuation plans; and incorporate planning
for Continuity of Government to ensure essential and emergency services.
(accessed online on May 9, 2008 at http://www.georgebush-whitehouse.
Note that a setback in executing one or more of the action items ahead of time
can determine why a subsequent crisis is the crisis it is. If you do not work out the
colocation plans or assistance rosters or evacuation plans well before the disaster,
you will be worse off after the disaster. A setback in anticipation leads to the inability
to respond as adequately as would have been the case had the action item
been implemented beforehand. Here, setback in anticipation leads to setback in
Much the same complaint has been made with respect to this ﬁnancial crisis. It
has been argued that the subprime mortgage crisis was anticipated, these warnings
were not heeded, and consequently, the crisis has turned out much worse.7 Since
Managing Organizational Setbacks 463
the mortgage crisis morphed into the ﬁnancial crisis, reactive measures such as
“one-off events,” “special dispensations,” seriatim bailouts, and U-turn manage-
ment reversals, have become common. At the time of writing, it is difﬁcult to see
many managers who are bouncing back in ways that they can plan the next steps
The role of setbacks in anticipation and resilience in conventional crisis preven-
tion is thus orthogonal to the role of successful resilience and anticipation in high
reliability management. In the former, the failure to anticipate reduces the ability to
be resilient later on. In the latter, being resilient and anticipatory occurs in the same
moment, i.e., when operators respond to a shock they are already planning the next
step ahead. In the former, setbacks in anticipation precede resilience; in the latter,
successful resilience entails thinking ahead, and in virtually the same moment. This
then is the chief virtue of preventing a crisis by managing to undertake activities
that do not precipitate events that could themselves lead to a crisis: You get both
resilience and anticipation jointly, not disjointly.
This suggests that for there to be few or no setbacks in crisis prevention, high
reliability management should be in place from the outset. Sounds obvious? Not if
you believe, pace crisis prevention, that more and better anticipation is the answer
to more and better resilience later on. Not if you believe, pace high reliability man-
agement, that the lack of resilience early on is one major reason accounting for the
lack of resilience later on after disaster has struck.
In this view of high reliability management as a special case of crisis prevention,
disaster management in response to Hurricane Katrina may well have not been
better had there been more pre-Katrina reports pointing out levee deﬁciencies,
more newspaper articles on the pending disaster, and more “Hurricane Pam”
simulations showing the range of likely disaster in the region. So too crisis man-
agement in the current ﬁnancial crisis may not have been appreciably better had
there been more Ned Gramlichs and Warren Buffetts sounding the early alarms.
This ﬁnancial crisis, and Hurricane Katrina, were long predicted and soundly so
(Alden, 2005; Authers, 2008, respectively). From a high reliability management
perspective, the primary problem was not anticipation; it is centered on what was
happening to resilience before and after the crises.
The high reliability management perspective would be asking among other
things the following: Given some 60 percent of the U.S. grain exports went through
New Orleans and over 25 percent of domestic oil production—most offshore oil
drilling—came from the Gulf of Mexico, how did grain companies and oil
reﬁneries—HROs in presumption if not fact—build in resilience in their operations
before the disaster and how did that capacity affect their rebounding back
Parallel questions would be raised as a matter of priority in the ﬁnancial crisis:
Not all banks and investment ﬁrms, in the United States or abroad, have weathered
the ﬁnancial crisis in the same way with the same effect, and how like banks have
been able to avoid like problems is of keen interest, if only to compare their better
or best practices to what reactive decision makers are now undertaking by way of
crisis management. Or in the terminology of this article, how did those oil and grain
companies or these banks and investment ﬁrms—which like many public sector
agencies have long-linked supply chains—handle the inevitable setbacks in their
464 Emery Roe
resilience capacity building and in planning their next steps ahead, and what can
crisis prevention learn from this?8
To summarize, crisis prevention in the precursor zone may be our last line of
defense, but high reliability management is crisis prevention’s ﬁrst line of defense.
Crisis Prevention, Setback Management, and Regulation
It is scarcely surprising that what policy-makers say they learn from a crisis is that
better regulation would prevent further ones. More stringent levee construction
designs and capital reserve requirements have predictably surfaced as regulatory
priorities in our examples. Yet the focus on managing setbacks in crisis
prevention—be it in the precursor zone when things are already bad or before that
when high reliability management is required—suggests a different role for regu-
lation: regulation as the guardian of setback learning.
Government regulation has three interrelated functions when it comes to service
provision related to society’s core infrastructures. First, regulators seek to prevent
crises in or affecting the critical infrastructures because the knock-on impacts could
be catastrophic. Regulation of nuclear power is the exemplary case. What operates
here is, in Paul Schulman’s terms, a precluded events reliability: Some events are so
dreadful that society avoids even the precursors that might lead to them. Regulators
ensure our critical infrastructures operate in ways that seek to avoid such circum-
Second, regulators can seek to increase the options and reduce the task environ-
ment volatility faced by the critical infrastructures that cannot anticipate every
accident waiting to happen. In such a world, regulators have a role in increasing the
anticipation and resilience capabilities of the frontline managers and operators in
the infrastructures. Third, regulators can seek to set rule-based regulations that
ensure the ﬁrst and second functions appear as speciﬁc operations in the critical
infrastructure, i.e., these procedures and activities reﬂect those options and that task
environment, which in turn reﬂect this precluded events reliability, where speciﬁ-
cation of “this, that, these, and those” is ensured by the regulators.
The three functions entail an increasing knowledge intensiveness on the part
of the regulators as to what infrastructure managers and operators actually do,
the third being more intensive than the ﬁrst, while the second recognizes that
the managers and operators have real-time operational requirements that
regulators cannot foresee and regulate in advance. In all three cases, regulators
are meant to have a high degree of knowledge concerning what operators
actually do—if only to ensure regulators do not get in the way and make things
It is important to understand the manner in which the third regulatory function
is knowledge intensive. Not only do the regulators need to know what is going on
operationally in the critical infrastructures, they must know how to transform dread
and precluded events reliability into operational activities in longer-term ways that
the infrastructure’s operators and managers who are caught up in real-time opera-
tions may not appreciate or recognize. Ideally, regulators would have the skills of
real-time pattern recognition and scenario formulation that operators in critical
infrastructures have, especially when it comes to managing unforeseen setbacks and
Managing Organizational Setbacks 465
redesigning operations in light of them.9 Having those skills, however, would mean
that regulators not only could take on the jobs of the professionals they are regu-
lating. They could also do those jobs better because they, the regulators, have more
prevention-oriented patterns and scenarios at their call, namely, they have longer-
term cause-and-effect knowledge that operators do not have. Yet our regulatory
regimes—be they principle- or rule-based—avoid fostering that kind of operational
experience among regulators.
We now see why the three conventional functions of regulation are problematic,
because each assumes regulators have operational knowledge they do not have nor
could have, given the fast-changing sectors they regulate. Let me propose, then, a
fourth regulatory function that makes sense in a world where setback management
as a form of crisis prevention is expected—namely, regulation as the custodian of
setback learning across our critical infrastructures.
Regulation in this fourth view would focus on interdependent critical infrastruc-
tures with transboundary operations. This is a regulatory world in which electricity
depends on hydropower and telecommunications depend on electricity while ﬁnan-
cial services depend on telecommunications and electricity. Infrastructures are
vulnerable to what happens in other infrastructures—in brief, it is a world where
setbacks and their consequences circulate and amplify, when not managed. It is a
world in which preventing a crisis has often meant preventing others.
We are already seeing the effect of the ﬁnancial crisis on some electricity, ship-
ping, and public transit (respectively, Jackson, 2008, p. 17; McNulty, 2008, p. 23;
Politi, 2008, p. 2), and it is easy to see how it can affect the operation of other critical
infrastructures and their services. “Since the international system depends on the
free, reliable and orderly ﬂow of ﬁnancial resources,” writes the head of the
Brookings Institution, Strobe Talbott (2008, p. 11), “failure to solve the current
mess will stymie progress in those other areas—trade, climate and proliferation.” A
cross-infrastructure crisis that extends from ﬁnance through electricity, telecommu-
nications, and transportation would be its own economic calamity, and one calling
for a kind of inter-infrastructure regulation we do not now have.
More formally, this regulation would seek to ensure that setbacks in intercon-
nected critical infrastructures are managed in such ways that (1) a setback in one
infrastructure does not pull it into its precursor zone nor others into theirs,
although if that should happen, (2) the infrastructures together would manage so
as not to be pulled across their respective edges into joint disaster. The role of
regulation here is to be the guardian of setback management and promoter of its
evolution across infrastructures. Regulators are the protectors of the evolutionary
advantage of learning from managing inevitable setbacks that have inter-
infrastructure, transboundary implications. This is especially pertinent when ofﬁ-
cials in the respective infrastructures in which the setbacks are occurring may be
denying they are occurring, or if setbacks are recognized, the respective organiza-
tions are so seized up in reactive response that they do have the cross-scale,
networked perspective to transform the setbacks into learning.
In this management world, regulation is about aiding mutual adjustments when
the respective edges of infrastructures are interdependent. Note that in this
approach, setbacks with transboundary consequences frequently have to happen
before they trigger this kind of regulation. This is because the setbacks, when they
466 Emery Roe
occur, substantiate the magnitude of the hazard that should be managed, which if
not, could lead to an even worse situation involving more systems. In these ways,
the “long run” is deﬁned and introduced into our critical infrastructures via regu-
lation, albeit through learning from setbacks that are taken to compel operational
changes across infrastructures continually preoccupied with real time.
Before describing just what this “learning” is about, an example of the fourth
regulatory function is useful. One caution before proceeding: Nothing in the above
argues that all regulation should be with respect to guarding and promoting
learning from setback management across infrastructures, nor does anything below
argue that all critical infrastructures have setbacks that can be managed by regula-
tors as in the following example. We will once again see that, when it comes to the
regulators, trying to manage a setback within the precursor zone is more difﬁcult
and ambiguous than managing it so as to prevent entering that zone.
Try to put yourself in the position of the secretary of the U.S. Treasury or the
head of the Federal Reserve in February 2008, before the collapse of Lehman
Brothers, the bailout of Freddie Mac and Fannie Mae, and the further bailouts of
AIG and Citigroup. You know nothing about the future ahead, but you are trying
to prevent the subprime mortgage crisis from metastasizing into the kind of free-fall
crisis it became.
What you do know in early 2008 is that underlying the subprime mortgage crisis
are novel ﬁnancial instruments that operated under no—or virtually no—mandates
to ensure the high reliability management of those instruments. Credit derivatives,
credit default swaps, collateralized debt obligations, special conduit vehicles, and
other securitized ﬁnancial assets enabled banks to shift off their balance sheet “a
dizzying array of mortgages, credit-card receivables and corporate loans” (Econo-
mist, 2007, p. 28), without similar reliability arrangements found at those banks.
The securitized ﬁnancial instruments were also exceptionally complex and opaque.
The supposedly good news that risk was to be dispersed through these instruments
had, by the end of 2007, been more than offset by the bad news that the risk was
concealed in the process, only to end up back on the banks’ balance sheets and
under the purview of their reliability mandates.10 “Imagine NASA sending men to
the moon before it had ﬁgured out how to get them back home,” writes back-page
columnist in the Financial Times at the end of 2007. “That is sort of what happened
in the world of securitisation this year, as complex structures such as collateralized
debt obligations blew up” (Lex Column, 2007).
In sum, by February 2008, you know enough to understand that the ﬁnancial
instruments have increased tight coupling and complex interactivity within the
ﬁnancial sector and that this happened without having the more strenuous reliabil-
ity mandates that banks conventionally have. You know, in short, you are in the
Speciﬁcally, the debt these instruments represented to the banking sector con-
stituted “an accident waiting to happen” (Norris, 2008; see also Wolf, 2007). Here
the management challenge was, as one UK regulator put it earlier, to turn “an
accident waiting to happen into a near miss” (quoted in Davies, Tett, & Scholtes,
2006). That is what Treasury and Federal Reserve tried to do—to turn a disaster
waiting to happen into a close call—when March 2008 rolled into view, and with it
the challenge that surfaced with Bear Stearns, the investment bank.
Managing Organizational Setbacks 467
In early March 2008, the U.S. Federal Reserve arranged with J.P. Morgan a
last-minute bailout for Bear Stearns, as the latter moved to bankruptcy. The reason
was that if Bear Stearns collapsed, the ﬁnancial sector was at considerable risk
of collapsing as well (e.g., Stoddard, 2008). Why was Bear Stearns vulnerable?
Because, although comparatively small, it was a central counterparty to transactions
involving a very large number of credit default swaps and related ﬁnancial instru-
ments. Clearly, having to bail out Bear Stearns posed a setback to Treasury and Fed
managers. A Bear Stearns failure would mean it could not meet its obligations for
contracts involving these instruments, and the knock-on effects were thought too
dire to risk.
How that bailout was managed is instructive:
In an action almost unprecedented in takeover history, JPMorgan bought 39.5 percent of
Bear on the spot to ensure that it would have close to a majority of the votes to approve
the deal. That agreement completely disregards New York Stock Exchange’s rules that
prevent anyone from buying more than 20 percent of [a] company without a shareholder
vote. Other parts of the new agreement either stretch the rules or disregard years of
precedent in Delaware, where both banks are incorporated. Of course, all this rule-
bending was done with the tacit, if not outright, approval of the federal government.
(Sorkin, 2008, p. C10)
Paul Volcker, former head of the Federal Reserve, summarizes this just-for-now
behavior as having taken the Fed to the “very edge of [its] lawful and implied
powers” (Scholtes, 2008, p. 4), while Paul Krugman (2008), the economist and
journalist, calls it “barely legal.” This article calls it setback management through
ofﬁcial and unofﬁcial channels taken under increasing urgency within the ﬁnancial
sector’s precursor zone.11
In hindsight, the Bear Stearns bailout looks to have failed in preventing the
ﬁnancial crisis from spreading. But did it fail? Having to answer that question is
the problem of managing setbacks in the precursor zone: You cannot ascertain the
counterfactual in all that uncertainty. What would have happened had not Bear
Stearns been bailed out will be the topic of the years ahead. “All one can say is, ‘It’s
not as bad as it would have been,’ ” Alice Rivlin, former director of the Congres-
sional Budget Ofﬁce concludes (quoted in Bendavid, 2008, p. 9). However, can you
even say that with any degree of assurance?
Thus, from this article’s perspective, it is better to regulate in order to ensure
(1)—a setback in one infrastructure does not pull it into its precursor zone nor
others into theirs—because by the time (2) comes around conditions have become
incalculably—repeat, immeasurably—worse. When infrastructures are already in
their precursor zone trying not to be pulled across into joint disaster, it is always
unclear to what extent setback management there is effective. All I can say—and it
is important when it can be said—is that those in the midst of such circumstances
feel they are managing in ways that forestall their setbacks from spinning out of
control, however temporary that respite may prove.
What Then Is to Be Learned from Setback Management?
If regulation is to be the guarantor of lessons learned from jointly managing
setbacks affecting multiple critical infrastructures, what kind of learning are the
468 Emery Roe
regulators looking for? Here the challenge is to learn from setbacks that cannot be
prevented so as to better prevent the crises those setbacks could or would otherwise
First and foremost, we cannot assume learning will be there nor can we assume
even the conditions for learning. All too often, the preconditions for learning—
stability in preferences, goals, and tasks—are precisely those that are missing in a
Still, it seems appropriate to ask what speciﬁc kinds of learning from setbacks
would be useful to managers trying to forestall failures across their interconnected
infrastructures. Presumably, their answer would be: We want the same things we
value from better managing setbacks within our own infrastructures. My reading
and research suggest at least four types of setback management that cast setbacks in
a positive role for those who manage them. Best known is the argument that
organizations do not transition from one stage to another in their life cycles without
overcoming the obstacles characteristic of the stage in which the organization ﬁnds
itself (Harrison & Shirom, 1999). This could hold for crisis preventing organizations
Other setbacks, in contrast, serve as a test bed of developing (more) anticipatory
or resilient strategies in the crisis preventing organization. Some setbacks are better
thought of as design probes as to whether that organization is broadly on the “right
track,” or if not, what track it should be on. In yet other cases, setbacks serve to
remind crisis prevention managers that other things matter for what they are doing.
Setbacks unsettle what had been settled knowledge to that point. Yet the setback
does this in a way that does not question the premise of continuity without which
any interruption would make no sense.
In these four ways, setbacks are used for learning, and it is a learning that can
connect today’s crisis management with tomorrow’s crisis prevention. For example,
how effective setback management is in the ﬁnancial crisis depends on if and how
the regulators and their networks manage to transform those setbacks into positive
interruptions, probes, tests, and obstacles to be surmounted. Will the crisis serve as
a timely interruption that reminds us just how central the regulators are to the
continuity of the ﬁnancial and credit systems? Will the crisis end up as a much-
needed probe of how well the ﬁnancial and credit institutions themselves are
keeping their sectors on track and under mandate? Will the crisis be the test bed for
more resilient and anticipatory strategies in credit lending and ﬁnancial investing?
Furthermore, will the crisis in effect be an obstacle, whose surmounting is necessary
to promote the operational redesign of the ﬁnancial and credit sectors in more
These are the questions that matter to those seeking to prevent such crises in the
future and, unlike Chou En-lai, we need answers sooner than later.
1 The author wishes to thank Arjen Boin, Paul Schulman, and Ian Mitroff for their comments.
2 For those in the building and zoning arena, setback management is about regulating the distance
between boundary and facility. This article concerns the obstacles and hurdles organizations encoun-
ter along the way that must be managed in order to prevent a crisis.
Managing Organizational Setbacks 469
3 For the early work on HROs, see LaPorte and Consolini (1991), Roberts (1990), Rochlin (1993),
Schulman (1993) and Weick (1987).
4 Other ﬁelds reinforce attention on precursors. Engineering risk assessment, for instance, examines
risks of failure from both crisis management and high reliability management perspectives.
5 Some countries, as another example of trying to stay out of the precursor zone, resisted calls for a
major conference to rework global ﬁnance. Japanese leaders echoed the U.S. hesitation, worried
what signals such a conference would send to markets. “Our honest feeling is that we want to prevent
a situation where we need to hold such a summit meeting,” Japanese Prime Minister Taro Aso said
in parliament’ (McKinnon, MacDonald, & Galloni, 2008, p. A7). To hold the meeting was to admit
being then in a position that could lead to something altogether worse.
6 Some policy-makers assume an unprecedented crisis requires unprecedented reaction. The chair
of the UK’s Financial Services Authority argued for example: “When you’ve been through a [ﬁnan-
cial] crisis like this, it’s rather sensible to wipe the slate clean in terms of your previous assumptions
rather than say: ‘because this has previously been my policy line, I still stick to my policy line’ ”
(quoted in Thal Larsen, 2008, p. 20). This is hazardous thinking if it means moving to policies that
have no competent managers.
7 Ned Gramlich, former Fed Governor, warned early and often concerning the dangers of the
subprime mortgage “boom and bust,” as he called it (Andrews, 2007; for other Cassandras, see
Blackburn, 2008, pp. 81–4; and Giles, 2008).
8 Or, to take another example, how have derivative exchanges with central clearing been able to
survive the shocks of the ﬁnancial crisis in ways that noncleared, over-the-counter credit default
swaps clearly have not (e.g., Steil, 2008, p. 11)?
9 The two factors have been summarized by Alan Greenspan (2008, p. 9) for the banking sector:
Bank loan ofﬁcers, in my experience, know far more about risks and workings of their counter-
parties than do bank regulators. Regulators, to be effective, have to be forward-looking to
anticipate the next ﬁnancial malfunction. This has not proved feasible. Regulators confronting
real-time uncertainty have rarely, if ever, been able to achieve the level of future clarity required
to act pre-emptively.
10 To consider the potential magnitude of crisis, consider only credit default swaps (CDS):
The notional amount of CDS contracts outstanding is roughly $45,000 billion [$45 trillion]. To put
it into perspective, that is about equal to half the total US household wealth and about ﬁve times
the national debt. The market is totally unregulated . . . (Soros, 2008b; for a counterargument, see
At the time of writing, credit default swaps had increased to what the Economist’s special report on
international banking calls the “insanely large notional value of $62 trillion” (Palmer, 2008, p. 4).
11 Legal reservations have also been made concerning later developments, such as Treasury Secretary
Paulsen’s requirement that the CEOs of nine major ﬁnancial institutions sign a document enabling
government equity participation in their institutions. William Poole (2008, p. A13), the former CEO
of the Federal Reserve Bank of St. Louis, points out,
To my knowledge, there is no statute that permits the U.S. government to require that a
corporation see stock to the government . . . One risk posed by Treasury’s less-than-voluntary
approach: What if a courageous board of directors of one of the nine large banks doesn’t agree
with the CEO’s promise?
If such a board did not agree, then Treasury ﬁnds itself once again pushed from the precursor zone
well into the crisis.
About the Author
Emery Roe is author, along with Paul R. Schulman, of High Reliability Management: Operating
On the Edge, published by Stanford University Press, and also a member of the Collaborative
for Catastrophic Risk Management, University of California, Berkeley. He is a practicing
policy analyst and has published widely on science, technology, and environmental policy
issues in the United States and overseas.
470 Emery Roe
Alden, E. (2005, September 6). For a presidency deﬁned by tragedy, this time there is no external enemy to
blame. Financial Times, p. 5.
Andrews, E. (2007, December 18). Fed and regulators shrugged as the subprime crisis spread: Analysts ﬁnd trail
of warnings on loans. New York Times, pp. A1, A28.
Authers, J. (2008, November 24). Review of James Grant, Mr Market Miscalculates: The bubble years and
beyond. Financial Times, p. 8.
Bendavid, N. (2008, December 22). Democrats offer a reality check. Wall Street Journal, international edition,
Bijaj, V. (2008, September 25). Rescue plan’s basic mystery: What’s all this stuff worth? New York Times,
pp. A1, A25.
Blackburn, R. (2008). The subprime crisis. New Left Review, 50(March/April), 63–106.
Corrigan, G. (2008, October 1). Top priorities to address key points of stress in the markets. Financial Times,
Davies, P., Atkins, R., & Mulligan, M. (2008, October 30). Outside risk advice seen as best for central banks.
Financial Times, p. 31.
Davies, P., Tett, G., & Scholtes, S. (2006, September 25). Derivative dealers’ tough match. Financial Times, p. 22.
Economist. (2007, October 20). Only human: A special report on central banks and the world economy.
Retrieved May 11, 2009, from http://www.economist.com
Financial Times. (2008, September 27–28). Bail-out talks regain momentum. Financial Times, p. 1.
Giles, C. (2008, November 26). The vision thing: Economic forecasters. Financial Times, p. 9.
Grant, J. (2008, October 31). DTCC clears Lehman hurdle. Financial Times, p. 29.
Grant, J., & Hughes, J. (2008, September 24). Lehman clean-up passes clearing milestone. Financial Times,
Greenspan, A. (2008, April 7). The Fed is blameless on the property bubble. Financial Times, p. 9.
Guerrera, F., & Bullock, N. (2008, October 31). Struggle to unearth quake’s epicentre. Financial Times, p. 16.
Harrison, M., & Shirom, A. (1999). Organizational diagnosis and assessment: Bridging theory and practice. Thousand
Oaks, CA: Sage Publications.
Hughes, J. (2008a, July 24). Fair value can ﬂatter to deceive on your own debt. Financial Times, p. 17.
Hughes, J. (2008b, October 14). Fair value accounting rules eased. Financial Times, p. 18.
Jackson, T. (2008, November 10). Time to pay close attention to the credit markets. Financial Times, p. 17.
Krugman, P. (2008, May 5). Success breeds failure. Retrieved May 11, 2009, from http://www.newyorktimes.com
LaPorte, T. R., & Consolini, P. (1991). Working in practice but not in theory: Theoretical challenges of high
reliability organizations. Public Administration Research and Theory, 1(1), 19–47.
Lex Column. (2007, December 21). Securitisation. Financial Times, p. 12.
Marsh, B. (2008, September 28). A tally of Federal rescues. New York Times, WK4.
McKinnon, J., MacDonald, A., & Galloni, A. (2008, October 18–19). Global bank rules draw resistance. Wall
Street Journal, p. A7.
McNulty, S. (2008, November 3). Energy industry is running out of steam. Financial Times, p. 22.
Mollenkamp, C., Whitehouse, M., Hilsenrath, J., & Dugan, I. (2008, September 29). Lehman’s demise triggered
cash crunch around globe. Wall Street Journal, pp. A1, A22.
Norris, F. (2008, April 4). The patient is healthy but dead. New York Times, pp. C1, C5.
Palmer, A. (2008, May 17–23). Paradise lost: A special report on international banking. The Economist, pp. 1–24.
Pickel, R. (2008, January 29). Net exposure is the best guide to derivatives’ market impact. Financial Times,
Politi, J. (2008, November 10). Default threat puts skits under transport authorities. Financial Times, p. 2.
Poole, W. (2008, October 17). Treasury has no authority to coerce banks. Wall Street Journal, p. A13.
Rappaport, L., & Tuna, C. (2008, December12–14). Why accept 0% for the balance sheet? Wall Street Journal,
international edition, p. 20.
Roach, S. (2008, October 9). Time is ripe for authorities to be decisive and opt for overkill. Financial Times,
Roberts, K. (1990). Some characteristics of one type of high reliability organization. Organization Science, 1(2),
Rochlin, G. (1993). Deﬁning “high reliability” organizations in practice: A taxonomic prologue. In K. Roberts
(Ed.), New challenges to understanding organizations (pp. 11–32). New York: MacMillan.
Roe, E., & Schulman, P. (2008). High reliability management. Stanford, CA: Stanford University Press.
Scannell, K. (2008, October 1). Regulators ease securities-valuation rules. Wall Street Journal, p. C2.
Scholtes, S. (2008, April 9). Volcker urges extended role for the Fed. Financial Times, p. 4.
Schulman, P. (1993). The negotiated order of organizational reliability. Administration and Society, 25, 353–372.
Managing Organizational Setbacks 471
Sorkin, A. (2008, March 25). The hand behind the deal. New York Times, pp. C1, C10.
Sorkin, A., Henriques, D., Andrews, E., & Nocera, J. (2008, October 2). 36 hours of alarm and action as crisis
spiraled. New York Times, pp. A1, A26–A27.
Soros, G. (2008a, October 29). America must lead a rescue of emerging economies. Financial Times, p. 11.
Soros, G. (2008b, April 3). The false belief at the heart of the ﬁnancial turmoil. Financial Times, p. 9.
Steil, B. (2008, November 21). We need a “safe-fail” approach to avert new crises. Financial Times, p. 11.
Stoddard, S. (2008, May 7). Fed, Bernanke get high marks for response to credit crisis. Investor’s Business Daily,
pp. A1, A2.
Talbott, S. (2008, November 3). A pressing need to master the tyranny of the urgent. Financial Times, p. 11.
Thal Larsen, P. (2008, October 17). British watchdog calls for overhaul of rules. Financial Times, p. 20.
Thal Larsen, P., & Guerrera, F. (2008, September 27–28). Never glad conﬁdent morning again? Fears for
ﬁnance as investment banks implode. Financial Times, p. 12.
Weick, K. (1987). Organizational culture as a source of high reliability. California Management Review, 29,
Wolf, M. (2007, November 27). Why banking is an accident waiting to happen. Retrieved May 11, 2009, from