Journal of Tort Law
Volume 2, Issue 1

2008

Article 1

Operationalizing Deterrence: Claims
Management (In Hospitals, a Large Retailer,
and Jails and Prisons)
Margo Schlanger∗

∗

Washington University School of Law, mschlanger@wulaw.wustl.edu

Copyright c 2008 The Berkeley Electronic Press. All rights reserved.

Operationalizing Deterrence: Claims
Management (In Hospitals, a Large Retailer,
and Jails and Prisons)∗
Margo Schlanger

Abstract
The theory that the prospect of liability for damages deters risky behavior has been developed
in countless articles and books. The literature is far sparser, however, on how deterrence is operationalized. And prior work slights an equally important effect of damage actions, to incentivize
claims management in addition to harm-reduction responses that are cost- rather than liabilityminimizing. This article works in the intersection of these two understudied areas, focusing on
claims management steps taken by frequently sued organizations, and opening a window into the
black box of deterrence to see how those steps may end up serving harm-reduction purposes as
well. To summarize, I observe that damage actions regulate risky enterprise by inducing organizations to develop claims management capabilities—that is, the capacity to process any resulting
disputes. I then argue that these claims management practices and personnel are sometimes used,
secondarily but importantly, to improve safety, reduce risk, and increase compliance with external legal requirements. Organizations’ internal claims management operations can, though they
need not, facilitate care-taking in four important ways: (a) promoting the gathering and analysis
of claims information; (b) requiring the hiring of specialized personnel with a mission to reduce
claim payouts; (c) encouraging bureaucratized procedures that may be harm-reducing, and (d)
increasing the salience of claims to various actors within the organization. I discuss the theory
underlying these four points, drawing on organizational economics and sociology, as well as on
psychology and behavioral law and economics. Then I discuss these four channels of influence in
particular factual settings which serve as case studies, looking at a single large retailer, and then
more generally at hospitals and hospital doctors, and jails and prisons. Because organizational
theory tells us that this kind of transformation or repurposing is quite ordinary, the preliminary evidence I canvass suggests that claims management should be included in any study of how damage
action deterrence is operationalized within large risk-creating organizations.

∗

Professor of Law, Washington University in St. Louis, and for Fall 2008, Visiting Professor,
University of Michigan Law School. Thanks to participants in the 2008 Law & Society annual
conference, the Insurance & Society Seminar, and law faculty workshops at Southern Methodist
University and the University of North Carolina for useful comments on earlier drafts. I received
helpful feedback, as well, from Ken Abraham, Pauline Kim, Kyle Logue, Michelle Mello, Bill
Sage, Kent Syverud, an anonymous reviewer for the Journal of Tort Law, and, as always, Sam
Bagenstos. All remaining errors are, of course, mine.

This article thus makes two chief scholarly contributions. It proposes and theorizes concrete operational paths by which damage actions may elicit organizational compliance with external norms.
And it describes in-house claims management, a heretofore underobserved arena in which law
influences organizational activity. In the conclusion, I propose that who performs claims management functions may matter, as well, and suggest that in future research, claims management
should be considered along with liability and loss prevention as the trio of liability-related operational areas in which firms must implement a “make-or-buy” decision.

Schlanger: Operationalizing Deterrence

INTRODUCTION
For several decades, scholars and policymakers have emphasized the regulatory
function served by claims and lawsuits seeking damages, focusing on the
incentive effects these damage actions create. The foundational insight is that our
ex post damages system translates into ex ante care because the prospect of civil
liability encourages organizations engaging in risky activities to minimize the
justiciable harm they cause.1 This deterrence theory, offered in support of both
negligence and strict liability regimes, has been developed in countless articles
and books. The literature is far sparser, however, on how deterrence is
operationalized. And prior work slights an equally important effect of damage
actions, to incentivize claims management in addition to harm-reduction
responses to litigation—that is, responses that are cost- rather than liabilityminimizing. This article works in the intersection of these two understudied areas,
focusing on claims management steps taken by frequently sued organizations, and
opening a window into the black box of deterrence to see how those steps may
end up serving harm-reduction purposes as well.
To summarize, I observe that damage actions regulate risky enterprise by
inducing organizations to develop claims management capabilities—that is, the
capacity to process any resulting disputes. I then argue that these claims
management practices and personnel are sometimes used, secondarily but
importantly, to improve safety, reduce risk, and increase compliance with external
legal requirements. Organizations’ internal claims management operations can,
though they need not, facilitate care-taking in four important ways: (a) promoting
the gathering and analysis of claims information; (b) requiring the hiring of
specialized personnel with a mission to reduce claim payouts; (c) encouraging
bureaucratized procedures that may be harm-reducing, and (d) increasing the
salience of claims to various actors within the organization. I first discuss the
theory underlying these four points, drawing on organizational economics and
1

Liability minimization is, of course, weighted against accident-prevention costs. For
canonical statements of this now commonplace approach, see, e.g., GUIDO CALABRESI, THE COSTS
OF ACCIDENTS: A LEGAL AND ECONOMIC ANALYSIS 26 (1970) (“I take it as axiomatic that the
principal function of accident law is to reduce the sum of the costs of accidents and the costs of
avoiding accidents.”); Richard Posner, A Theory of Negligence, 1 J. LEGAL STUD. 29, 33 (1972)
(“If . . . the benefits in accident avoidance exceed the costs of prevention, society is better off if
those costs are incurred and the accident averted, and so in this case the enterprise is made liable,
in the expectation that self-interest will lead it to adopt the precautions in order to avoid a greater
cost in tort judgments.”). For much older theorizing along these lines, see, e.g., Powell v. Fall, 5
Q.B. 597 (1880) (Bramwell, L.J.) (“It is just and reasonable that if a person uses a dangerous
machine, he should pay for the damage which it occasions; if the reward which he gains for the
use of the machine will not pay for the damage, it is mischievous to the public and ought to be
suppressed.”); United States v. Carroll Towing Co., 159 F.2d 169, 171-73 (2d Cir. 1947).
Published by The Berkeley Electronic Press, 2008

1

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

sociology, as well as on psychology and behavioral law and economics. Then I
discuss these four channels of influence in three particular factual settings which
serve as case studies, looking at a single large retailer, and then more generally at
hospitals and hospital doctors, and jails and prisons. Because organizational
theory tells us that this kind of transformation or repurposing is quite ordinary, the
preliminary evidence I canvass suggests that claims management should be
included in any study of how damage action deterrence is operationalized within
large risk-creating organizations.
This article thus makes two chief scholarly contributions. It proposes and
theorizes concrete operational paths by which damage actions may elicit
organizational compliance with external norms. And it describes in-house claims
management, a heretofore underobserved arena in which law influences
organizational activity. In the conclusion, I propose that who performs claims
management functions may matter, as well, and suggest that in future research,
claims management should be considered along with liability and loss prevention
as the trio of liability-related operational areas in which firms must implement a
“make-or-buy” decision.2
I. PRIOR LITERATURE
Deterrence and regulatory insurance. A great deal of the abundant scholarship
on deterrence assumes that the translation from law to action is fairly
straightforward, at least for individuals and profit-maximizing organizations.
Even government actors, long thought to maximize something other than profits,3
have been found to respond to damage action incentives.4 Of course, transactions
costs, agency costs, and information costs interfere with perfect implementation
of the incentivized conduct,5 and both individual, firm, and governmental
2

See R.H. Coase, The Nature of the Firm, 4 ECONOMICA (n.s.) 386, 393-94 (1937), reprinted
in R.H. COASE, THE FIRM, THE MARKET, AND THE LAW 33 (1988); OLIVER E. WILLIAMSON, THE
ECONOMIC INSTITUTIONS OF CAPITALISM: FIRMS, MARKETS, RELATIONAL CONTRACTING (1985).
3
See, e.g., WILLIAM A. NISKANEN, JR., BUREAUCRACY AND REPRESENTATIVE GOVERNMENT
36-42 (1971) (arguing that government bureaucrats are most interested in maximizing budgets).
4
On government response to suits for damages, see, e.g., PETER H. SCHUCK, SUING
GOVERNMENT: CITIZEN REMEDIES FOR OFFICIAL WRONGS (1983); Charles Epp, Exploring the
Costs of Administrative Legalization: City Expenditures on Legal Services, 1960-1995, 34 LAW &
SOC’Y REV. 407 (2000); Daryl Levinson, Making Government Pay: Markets, Politics, and the
Allocation of Constitutional Costs, 67 U. CHI. L. REV. 345 (2000); Margo Schlanger, Inmate
Litigation, 116 HARV. L. REV. 1555, 1664-90 (2003).
5
For foundational work on transactions costs, see Oliver E. Williamson, Transaction-Cost
Economics: The Governance of Contractual Relations, 22 J.L. & ECON. 233 (1979); see also THE
ECONOMICS OF TRANSACTION COSTS (Oliver E. Williamson & Scott E. Masten eds., 1999). On
agency costs, see Michael C. Jensen & William H. Meckling, Theory of the Firm: Managerial
Behavior, Agency Costs and Ownership Structure, 3 J. FINAN. ECON. 305 (1976); MICHAEL C.
http://www.bepress.com/jtl/vol2/iss1/art1

2

Schlanger: Operationalizing Deterrence

compliance with external regulation are subject to error and distraction. Various
social norms may undermine or augment the deterrent signal.6 Moreover, the
litigation setting, with its complex procedures and rules of proof and justiciability,
may warp the deterrent message sent by the liability and damages rules.7 And
finally, liability insurance may pose a threat to deterrence by dampening damage
actions’ regulatory signal; this is the familiar story of moral hazard.8 (Other
analysts agree that liability insurance is problematic for deterrence, but focus on
the way in which liability insurance blurs, if not softens, the deterrent message,
because insurers are forced to predict risk using actuarial groupings9 rather than
JENSEN, FOUNDATIONS OF ORGANIZATIONAL STRATEGY (1998). On information costs, see George
J. Stigler, The Economics of Information, 69 J. POL. ECON. 213 (1961); Armen A. Alchian &
Harold Demsetz, Production, Information Costs, and Economic Organization, 62 AM. ECON. REV.
777 (1972); KENNETH J. ARROW, THE LIMITS OF ORGANIZATION 33-59 (1974); Joseph E. Stiglitz,
The Contributions of the Economics of Information to Twentieth Century Economics, 115 Q.J.
ECON. 1441 (2000). More generally, on the limits on organizational capacity to optimize complex
choices, see JAMES G. MARCH & HERBERT A. SIMON, ORGANIZATIONS (1958), and on the
relationship these various factors have with the choice to conduct a particular activity within or
without the firm’s boundaries, see Coase, Nature of the Firm, supra note 2.
6
For a summary of the law and economics work on social norms, see Richard H. McAdams
& Eric B. Rasmusen, Norms and the Law, in 2 HANDBOOK OF LAW AND ECONOMICS 1573 (A.
Mitchell Polinsky & Steven Shavell eds., 2007); for a summary of the psychological literature
relating to norms and compliance and its resistance to rational actor expectations, see Robert B.
Cialdini & Melanie R. Trost, Social Influence, Social Norms, Conformity, and Compliance, in 2
HANDBOOK OF SOCIAL PSYCHOLOGY 151 (Daniel T. Gilbert, Susan T. Fiske & Gardner Lindzey
eds., 4th ed. 1998).
7
See, e.g., Margo Schlanger, Second Best Damage Action Deterrence, 55 DEPAUL L. REV.
517 (2006) (arguing that damage actions can induce potential defendants to favor more cognizable
or demonstrable care, and less cognizable or demonstrable harm—behavioral changes in the type
rather than amount of care defendants take, or the type rather than amount of harm they inflict that
do nothing to shift the probability or severity of accidents but rather limit expected liability by
lowering the probability of claims or the expected amount of damages).
8
A long list of sources making this argument is cited by Gary T. Schwartz, The Ethics and the
Economics of Tort Liability Insurance, 75 CORN. L. REV. 312, 312. For discussions of the moral
hazard argument and the sociology of insurance, see, e.g., CAROL A. HEIMER, REACTIVE RISK AND
RATIONAL ACTION: MANAGING MORAL HAZARD IN INSURANCE CONTRACTS (1985); Tom Baker,
On the Genealogy of Moral Hazard, 75 TEX. L. REV. 237 (1996). On the economics of insurance
and accident prevention, see, e.g., Steven Shavell, On Liability and Insurance, 13 BELL J. ECON.
120 (1982) (setting out parameters for interaction between insurance and efficient accident
prevention); WILLIAM M. LANDES & RICHARD A. POSNER, THE ECONOMIC STRUCTURE OF TORT
LAW 13 (1987) (conceding that liability insurance “does reduce deterrence,” though not
inefficiently); Schwartz, supra, at 336-59.
9
See, e.g., CALABRESI, COSTS OF ACCIDENTS, supra note 1, at 241 (“Since individuals are
allowed to insure against responsibility, however, the deterrence pressure resulting from the
allocation of costs affects the actuarial groups to which the party held responsible belongs. . . .
This characteristic gives rise to some of the most significant criticisms of the system.”); see also
id. at 248-49.
Published by The Berkeley Electronic Press, 2008

3

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

experience rating or feature rating.10) But all these caveats operate only at the
edges of the main point, which is that to some degree at least, damage actions
function to price and internalize to risk-creating organizations many harms—
caused by negligence, statutory noncompliance, or constitutional violations—that
would otherwise remain externalities.
In most of the work on deterrence, however, the precise ways in which the
liability signal is processed and its incentives made salient or even relevant to
actors within the affected organizations is undertheorizesd. There are, of course,
exceptions. The social norms scholarship already referenced illuminates the
social psychology of communities of risk-takers and risk-imposers. More
sociologically, a growing “self-governance” literature looks at the ways in which
organizations adapt to law’s requirements—some scholars focusing on the ways
in which organizations domesticate or even subvert those requirements,11 others
highlighting self-regulation’s positive potential for more effective and efficient
care-taking.12 I draw on some of this work below. Most relevant here, given the
comprehensive interdependence of damage actions and insurance,13 a number of
scholars examine how liability insurance, far from undermining risk reduction,
instead augments it.14 This “regulatory insurance” literature identifies three ways
in which this augmentation takes place, each facing moral hazard as a constraint
and a challenge. First, where insurance is compulsory, it tends to price very risky
(or broke and therefore less deterable15) actors out of the activity in question.16 In
10

See KENNETH S. ABRAHAM, DISTRIBUTING RISK: INSURANCE, LEGAL THEORY, AND PUBLIC
POLICY 46 (1986) (discussing experience rating and feature rating as necessary for optimal
deterrence of insured entities).
11
See, e.g., the work of Lauren Edelman and coauthors, discussed infra at text accompanying
notes 62-74.
12
For a discussion of this body of work, see, e.g., Orly Lobel, The Renew Deal: The Fall of
Regulation and the Rise of Governance in Contemporary Legal Thought, 89 MINN. L. REV. 342
(2004).
13
See, e.g., Kent D. Syverud, On the Demand for Liability Insurance, 72 TEX. L. REV. 1629,
1640 (1994) (“[Some observers] place liability insurance at the periphery of tort litigation—they
view insurance as distinct, incidental, and largely reactive to developments in the courts. This
view is wrong.”).
14
See, e.g., Deborah Stone, Beyond Moral Hazard: Insurance as Moral Opportunity, in
EMBRACING RISK: THE CHANGING CULTURE OF INSURANCE AND RESPONSIBILITY 52 (Tom Baker
& Jonathan Simon eds., 2002); Richard Ericson, Dean Barry, & Aaron Doyle, The Moral Hazards
of Neo-Liberalism: Lessons from the Private Insurance Industry, 29 J. ECON. & SOC’Y 532, 53258 (2000). Tom Baker, Insuring Morality, 29 J. ECON. & SOC’Y 559, 559-77 (2000). Of course,
this effect is in addition to what is usually considered the most important function of insurance,
spreading risk over time and over population.
15
See, e.g., Steven Shavell, The Judgment Proof Problem, 6 INT’L REV. L. & ECON. 45
(1986); Stephen G. Gilles, The Judgment-Proof Society, 63 WASH. & LEE L. REV. 603 (2006).
16
See Steven Shavell, Minimum Asset Requirements and Compulsory Liability Insurance as
Solutions to the Judgment-Proof Problem, 36 RAND J. ECON. 63 (2005) (arguing that this effect
http://www.bepress.com/jtl/vol2/iss1/art1

4

Schlanger: Operationalizing Deterrence

addition, insurers can use experience rating to reimpose accident costs on
insureds, at least in settings in which policyholders need insurance after as well as
before they incur a loss.17 And finally, having gained insight into accident
prevention by sustained research18 and by analysis of claims,19 insurers can use
various substantive contract terms related to prerequisites, pricing, and
monitoring, as well as other less formal methods, to induce policyholders to
engage in more effective harm reduction.20 “Fire insurers might, for example,
refuse to give coverage to businesses without sprinkler systems and fire drills,”21
or “[i]nsurance experts teach insureds the habits of prevention, for example,
providing them with regimes of preventive health care and of preventive security
for their property.”22
As Ken Abraham has summarized this last point, “[w]hen insurers are in fact
strategically positioned to be the cheapest cost avoiders, they will have to
emphasize preinsurance inspections, periodic regulatory compliance audits,
subjective evaluation of the applicant’s operations, and continuing involvement in
risk management.”23 Essentially, Abraham is describing insurance as a system by
which firms outsource harm prevention information gathering—when, for
example, insurers, with their larger risk bases, are better suited than policyholders
occurs, though it may sometimes be undesirable). Some observers suggest that compulsory car
insurance tends to make the uninsured drive less, rather than not at all. See Alma Cohen & Rajeev
Dehejia, The Effect of Automobile Insurance and Accident Liability Laws on Traffic Fatalities, 47
J.L. & ECON. 357 (2004).
17
See, e.g., HEIMER, REACTIVE RISK, supra note 8, at 42.
18
See, e.g., ABRAHAM, DISTRIBUTING RISK, supra note 10, at 16; Schwartz, Ethics and
Economics of Insurance, supra note 8, at 356; NATIONAL BOARD OF FIRE UNDERWRITERS,
PIONEERS OF PROGRESS: 1866-1941 (1941); JOHN BAINBRIDGE, BIOGRAPHY OF AN IDEA: THE
STORY OF MUTUAL FIRE AND CASUALTY INSURANCE (1952).
19
See, e.g., George M. Cohen, Legal Malpractice Insurance and Loss Prevention: A
Comparative Analysis of Economic Institutions, 4 CONN. INS. L.J. 305, 325 (1997); Schwartz,
Ethics and Economics of Insurance, supra note 8, at 356.
20
Accounts of this harm-reduction component of insurance operations include: ABRAHAM,
DISTRIBUTING RISK, supra note 10, at 59-60; Cohen, Legal Malpractice Insurance and Loss
Prevention, supra note 19; RICHARD V. ERICSON, AARON DOYLE, & DEAN BARRY, INSURANCE AS
GOVERNANCE (2003); HEIMER, REACTIVE RISK, supra note 8; Carol Heimer, Insuring More,
Ensuring Less: The Costs and Benefits of Private Regulation Through Insurance, in EMBRACING
RISK, supra note 14, at 116; Harris Schlesinger & Emilio Venezian, Insurance Markets with LossPrevention Activity: Profits, Market Structure, and Consumer Welfare, 17 RAND J. ECON. 227
(1986). Schlesinger and Venezian make the important point that different insurance lines tend to
be more or less focused on loss prevention. See id. at 228-29. For an account of safety-regulation
and selection in the earliest days of the American insurance industry, see BAINBRIDGE,
BIOGRAPHY OF AN IDEA, supra note 18.
21
HEIMER, REACTIVE RISK, supra note 8, at 13; see also BAINBRIDGE, BIOGRAPHY OF AN
IDEA, supra note 18.
22
ERICSON ET AL., INSURANCE AS GOVERNANCE, supra note 20, at 54.
23
ABRAHAM, DISTRIBUTING RISK, supra note 10, at 59.
Published by The Berkeley Electronic Press, 2008

5

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

to research cost-effective methods of reducing harm.24 Others put the point more
sociologically, emphasizing that policyholding firms’ attitudes about risk and its
reduction are far from rigid, as insurance companies work to refashion their
customers’ culture and priorities. For example, Carol Heimer acknowledges that
“the insurer will engage in strategic interaction to alter the incentive structure of
the policyholder to make it resemble that of the insurer (or of the ‘prudent
uninsured owner’),”25 but she emphasizes, as well, that “[i]nsurers recognize that
if one is to exert control over agents this is usually not so much a matter of
providing incentives for individual action as it is a matter of developing
organizational routines and standard operating procedures.”26
Whether regulatory insurance scholarship describes firms that have
outsourced to their insurers the task of risk and care analysis, or insurers who
work to transform both the capabilities and attitudes of policyholders, the point is
that it has traced one important way in which damage action incentives are
operationalized. When insurers are the bearers of liability risk, they often institute
practices to induce care-taking and corresponding harm reduction, managing
moral hazard and reducing harm by compelling (by contract), incentivizing,
advising, or persuading their customers to undertake various harm-reduction
methods.
Claims management. For deterrence scholars, how firms manage claims
against themselves, working to respond efficiently and effectively, is an
afterthought. Even for regulatory insurance theorists, claims processing is far
from central; the harm-reducing impact of claims management practices has not
previously received sustained analysis. That is not to say that claims management
itself is unknown to observers as a necessary function for organizations whose
activities provoke claims. Some (though not many) scholars have looked at
claims management for its own sake, not as a part of the deterrent system. Most
24

Gary Schwartz approached the issue similarly, describing “loss control services that
liability insurers might be able to provide their insureds.” Schwartz, supra note 8, at 356.
25
HEIMER, REACTIVE RISK, supra note 8, at 9; see also id. at 42 (“[S]ometimes insurers
collect a great deal of information about how to prevent losses, then agree to reclassify
policyholders into new categories with lower rates if they will make the appropriate modifications;
sometimes, when a factor makes a big difference in loss experience, the renewal of the insurance
policy at the current rate is made contingent on the policyholder making appropriate changes;
sometimes insurers charge policyholders a rate appropriate to their rate classification but agree to
return part of the premium if policyholders avoid losses (by whatever means).”); ERICSON ET AL.,
INSURANCE AS GOVERNANCE, supra note 20, at 88.
26
HEIMER, REACTIVE RISK, supra note 8, at 24; see also, e.g., ERICSON ET AL., INSURANCE AS
GOVERNANCE, supra note 20, at 88 (“A key aspect of surveillance systems for governing moral
risk at a distance is to make the insured self-governing. The ideal is to make each policyholder a
watcher as well as watched and a bearer of her own control. Self-governance is accomplished
through a number of interconnected mechanisms of creating individual responsibility for risk
control.”).
http://www.bepress.com/jtl/vol2/iss1/art1

6

Schlanger: Operationalizing Deterrence

famously, H. Laurence Ross’s classic book, Settled Out of Court,27 published in
1970, investigated insurance “claims men” and claims processes, and argued that
insurance claims management practices worked more to encourage speedy claims
resolution than low payouts. (In this book, at least, Ross was thoroughly
uninterested in deterrence, which he suggested was something of a fool’s hope,
because, he believed, car accidents were typically caused by momentary and
unavoidable lapses in attention.28) Other studies of insurers and their social
practices might also be included in this same category.29 Likewise, there is a
limited amount of work on the costs, risks, and benefits of outsourcing claims
management, either with30 or without31 bundled liability coverage. All this prior
work on claims management is concerned with what are, indeed, its two major
goals, from the perspective of damage action defendants—to minimize processing
costs by making responding to claims more efficient; and to minimize payouts,
chiefly by detecting and deterring fraud, defending against claims, and negotiating
resolutions.32
The interaction between claims management and deterrence has received
much less attention. One point that prior work has noticed is that claims
management costs money, and therefore contributes to the felt deterrent impact of
claims made. Scholars interested in deterrence have accordingly included claims
27

H. LAURENCE ROSS, SETTLED OUT OF COURT: THE SOCIAL PROCESS OF INSURANCE CLAIMS
ADJUSTMENT (1970).
28
Ross endorsed research that suggested that “run-of-the-mill accidents nearly all involve
failure to see an approaching hazard and to predict its course accurately. However, in the context
of these accidents such failure can rarely be considered faulty,” id. at 251, and concluded, more
generally, that “accidents are most likely not the result of deficient driving,” id. at 253. In fact, at
the time Ross was writing, between 10% and 20% of injurious traffic accidents and about half of
traffic fatalities had alcohol as a factor. R.K. JONES & J.H. LACEY, ALCOHOL AND HIGHWAY
SAFETY 2001: A REVIEW OF THE STATE OF KNOWLEDGE ch. 2 (U.S. Department of Transp. 2001),
available at http://www.nhtsa.dot.gov/people/injury/research/AlcoholHighway. But Ross’s book
was written before the emergence of drunk driving as a salient social problem, in the early 1980s.
(Ross himself did important work in this area, later. See H. LAURENCE ROSS, DETERRING THE
DRINKING DRIVER: LEGAL POLICY AND SOCIAL CONTROL (1982); H. LAURENCE ROSS,
CONFRONTING DRUNK DRIVING: SOCIAL POLICY FOR SAVING LIVES (1992).)
29
See, e.g., sources cited supra note 20.
30
Efficiency in claims handling has long been thought to be an impetus for corporate
insurance purchases. David Mayers & Clifford W. Smith, Jr., On the Corporate Demand for
Insurance, 55 J. BUSINESS 281 (1982).
31
On public outsourcing of claims administration, without insurance, see Peter C. Young &
John Hood, Risk and the Outsourcing of Risk Management Services: The Case of Claims
Management, PUB. BUDGETING & FIN., Fall 2003, at 109; John Hood & Peter C. Young, The Risk
Management Implications of Outsourcing Claims Management Services in Local Government,
RISK MGMT., July 2003, at 7.
32
See Mayers & Smith, Corporate Demand for Insurance, supra note 30, at 285; sources
cited supra note 31.
Published by The Berkeley Electronic Press, 2008

7

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

management costs—for example, defendants’ own attorneys’ fees—as a
component of the damage action deterrence apparatus.33 In addition, observers
have occasionally pointed out that claims management expenditures may function
as a substitute for harm-reducing investment; a profit-maximizing organization is
likely to spend money on aggressive defense or anti-fraud measures, for example,
until the marginal return on such expenditures is less than the return on accidentprevention measures.34
This article analyzes a different dynamic relationship between claims
management and accident prevention. I suggest that claims management
practices, at least in some organizations under some circumstances, produce an
important secondary effect of enabling and encouraging a variety of harmprevention or accident-avoidance measures. By secondary, I mean to describe
these effects as non-motivating—not the reason for the activity, but rather its
predictable and desirable byproducts. Nonetheless, I argue, to understand damage
actions’ regulation of care or risk requires understanding this aspect of litigation
response. In Part II, I discuss the theory—economic, sociological, and
psychological—bolstering my observations regarding claims management. In
Part III, I describe in more detail the ways in which claims management induces
care-taking in three factual settings which serve as case studies—a large “big
box” retailer, hospitals and hospital doctors, and jails and prisons. The case
studies vary somewhat in method—the first looks at a single example and is based
mostly on one in-depth interview; the second is more grounded in secondary
sources along with interviews; the third is based on interviews, secondary sources,
and my own observations. I do not offer the case studies to provide rigorous
proof of the theory; they function rather as its source, having helped to generate
the hypotheses whose theoretical plausibility is sketched out in Part II.35 Part IV
concludes the customary promise of/plea for future research that will test these
hypotheses and will expand the analysis from the what of claim management to
who, considering together all of the litigation-related “make-or-buy” decisions—
liability, loss-prevention, and claims management.

33

See, e.g., A. Mitchell Polinsky & Yeon-Koo Che, Decoupling Liability: Optimal Incentives
for Care and Litigation, 22 RAND J. ECON. 562, 564 (1991).
34
See Cohen, Legal Malpractice Insurance and Loss Prevention, supra note 19, at 315-17
(1997); Terry Thomason & Silvana Pozzebon, Determinants of Firm Workplace Health and Safety
and Claims Management Practices, 55 INDUS. & LAB. REL. REV. 286 (2002).
35
On the use of qualitative case studies, see, e.g., ALEXANDER L. GEORGE & ANDREW
BENNETT, CASE STUDIES AND THEORY DEVELOPMENT IN THE SOCIAL SCIENCES (2005).
http://www.bepress.com/jtl/vol2/iss1/art1

8

Schlanger: Operationalizing Deterrence

II. CLAIMS MANAGEMENT AND OPERATIONALIZING DETERRENCE
Damage action filings induce organizations to develop claims management
abilities—that is, the capacity to process any resulting disputes, ex post. This
capacity imposes both processing costs, including paying for whatever personnel
is involved (accountants, lawyers, paralegals, investigators, customer service
staff, operational staff as needed, etc.), and payout costs (for both settlements and
litigated judgments, and including fraudulent as well as legitimately owed
payments).36 The goal of claims management activity is to minimize the sum of
the costs of processing and payout. But a possible, though not necessary,
byproduct can be harm-reduction measures. In particular, organizations’ internal
claims management operations can, though they need not, promote care-taking in
four important ways: (a) by promoting the gathering and analysis of claims
information; (b) by requiring the hiring of specialized personnel with a mission to
reduce claim payouts; (c) by encouraging bureaucratized procedures that are
harm-reducing, and (d) by increasing the salience of claims to various actors
within the organization. In this Part, I discuss the organizational theory—
economic, sociological, and psychological—underlying each of the four channels
of influence just listed.
If organizations were perfectly informed and perfectly rational, this project
would be very different. Under perfect information and rationality, no operative
difference could exist between “primary” and “secondary” effects, between goals
and desirable byproducts. All predictable effects would be anticipated, and
included in any decisional calculus. Organizations could chose their goals and
calibrate precisely how much to invest towards them based on full information not
only about primary but also secondary costs and benefits. But of course even for
individuals, perfect information is impossible, both cognitively37 and
economically.38 And organizations, while they can improve on individual
decisionmaking in many respects,39 also face large obstacles in harnessing
36

This is all fairly straightforward, except for the complexities created by the prevalence of
fraud in some claims situations. See, e.g., Keith J. Crocker & Sharon Tennyson, Insurance Fraud
and Optimal Claims Settlement Strategies, 45 J. L. & ECON. 469 (2002); Sharon Tennyson & Pau
Salsas-Forn, Claims Auditing in Automobile Insurance: Fraud Detection and Deterrence
Objectives, 69 J. RISK & INS. 289 (2002); David S. Loughran, Deterring Fraud: The Role of
General Damage Awards in Automobile Insurance Settlements, 72 J. RISK & INS. 551 (2005).
37
For a summary of work on the limited cognitive capacity of human decisionmakers, and
how that limited capacity affects the rational choice model of human behavior, see Russell B.
Korobkin & Thomas S. Ulen, Law and Behavioral Science: Removing the Rationality Assumption
from Law and Economics, 88 CAL. L. REV. 1051, 1075-1126 (2000).
38
See sources cited on information costs, supra note 5.
39
See MAX WEBER, THE THEORY OF SOCIAL AND ECONOMIC ORGANIZATION (1947); ARROW,
LIMITS OF ORGANIZATION, supra note 5; PAUL MILGROM & JOHN ROBERTS, ECONOMICS,
ORGANIZATION AND MANAGEMENT (1992). For a summary of additional literature that
Published by The Berkeley Electronic Press, 2008

9

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

individual initiative and activity towards the collective goals, even if those goals
are properly chosen.40
There is, moreover, substantial evidence that
organizational decisionmaking frequently does not follow such an
anthropomorphic path—that more useful than analyzing the organization’s
“goals” may be examination of its structures, including personnel, capacities,
routines, and standard operating procedures.41 The four channels of influence
whose description follows operate in those areas of organizational structure.
A. Information
The case studies in Part III suggest that firms collect a great deal of information in
order to manage claims made against them. The case studies include more
specifics, but to cite one example, organizations collect statements of employee
witnesses to harmful incidents, in order to use those statements to assess and
contest liability. It is my contention that this information, once collected, may
end up being used, as well, to design interventions for harm prevention or
reduction.
If information or its processing and use were cost-free for organizations, this
suggestion would be insupportable. Any information useful for harm prevention
would of course be collected—without cost—for that very purpose. But of course
information and its use are far from free42; as with other factors of production and
management, accumulating and using information requires capital and non-capital
investment, in technology, personnel, training, and materials. The result is
bounded rather than perfect rationality in decisionmaking. (As economist Roy
demonstrates ways in which “[c]oordinating tasks through formally rational structures permits
firms to augment, rather than just aggregate, the capacities of the individual,” see Kenneth A.
Bamburger, Regulation as Delegation: Private Firms, Decisionmaking, and Accountability in The
Administrative State, 56 DUKE L.J. 377, 409-13 (2006).
40
See Bamburger, Regulation as Delegation, supra note 39, at 417-34 and sources cited; see
also sources cited supra note 5.
41
Cf. supra text accompanying note 26. There is, of course, a very large literature on
organizational behavior; I have benefited from discussions of that literature and its relation to legal
compliance in Jeb Barnes & Thomas F. Burke, The Diffusion of Rights: From Law on the Books to
Organizational Rights Practices, 40 LAW & SOC’Y REV. 493 (2006), and Bamberger, Regulation
as Delegation, supra note 39, at 414-16. For some key contributions and reviews, see PHILIP
SELZNICK, LEADERSHIP IN ADMINISTRATION (1957); MARCH & SIMON, ORGANIZATIONS, supra
note 5; RICHARD M. CYERT & JAMES G. MARCH, A BEHAVIORAL THEORY OF THE FIRM (1992)
(1963) (especially relevant here is id. at 120-34 (on standard operating procedures)); Paul J.
DiMaggio & Walter W. Powell, The Iron Cage Revisited: Institutional Isomorphism and
Collective Rationality in Organizational Fields, 48 AM. SOC. REV. 147 (1983) (examining
structures within organizations); Markus C. Becker, Organizational Routines: A Review of the
Literature, 13 INDUS & CORP. CHANGE 643 (2004).
42
See sources cited supra note 5.
http://www.bepress.com/jtl/vol2/iss1/art1

10

Schlanger: Operationalizing Deterrence

Radner has written, “[t]o study seriously the economics of managing, one must
face squarely the boundedness of rationality of economic decision makers. This
phenomenon has long been recognized by theorists, if rarely acted upon.”43)
Given high information costs, it may be that claims information is insufficiently
useful in harm prevention to justify its costly collection for that purpose alone. Or
perhaps this informational investment does not occur because the potential uses of
claims information for harm prevention are non-obvious to organizational
decisionmakers, or because those uses are too uncertain.
Once claims
management needs for information are added into the mix, however, the case
studies suggest that the economic or sociological calculus sometimes changes.
JoAnne Yates has written a fascinating account that highlights, in the context
of American manufacturing firms between 1850 and 1920, the ways in which
intra-firm supply of and demand for information may shift.44 She points out that
investments in information gathering, processing, and dissemination “are shaped
by many organizational and technological factors, as are their consequences.”
These “[i]nvestment decisions interact with issues of information needs, ideology,
and organizational power. As each element of an organization’s information
system is established, it becomes entrenched locally by virtue of the human and
nonhuman capital invested in it.” Both information investments and their
abandonment are, accordingly, lumpy and sticky; firms cannot or at least do not
“buy information in increments of any size . . . at the moment when its value
exceeds it theoretical cost,”45 and they likewise do not abandon prior information
investments without hesitation.
Foundational work by economist Kenneth Arrow46 on organizational
information acquisition similarly emphasized that information comes into
43

Roy Radner, Hierarchy: The Economics of Managing, 30 J. ECON. LIT. 1382, 1389 (1992).
Joanne Yates, Investing in Information: Supply and Demand Forces in the Use of
Information in American Firms, 1850-1920, in INSIDE THE BUSINESS ENTERPRISE: HISTORICAL
PERSPECTIVES ON THE USE OF INFORMATION (Peter Temin ed., 1991); see also JOANNE YATES,
CONTROL THROUGH COMMUNICATION: THE RISE OF SYSTEM IN AMERICAN MANAGEMENT (1989).
For other economic historical accounts of the ways in which information within firms affects
operations, see, e.g., INSIDE THE BUSINESS ENTERPRISE, supra.
45
YATES, Investing in Information, supra note 44, at 152.
46
ARROW, LIMITS OF ORGANIZATION, supra note 5. For more recent theoretical work on
information flow within organizations, see, e.g., Patrick Bolton & Mathias Dewatripont, The Firm
as a Communication Network, 109 Q. J. ECON. 809 (1994) (explaining how organizations can
minimize costs of processing and communicating information by specialization, collaboration, and
centralization); Wouter Dessein, Authority and Communication in Organizations, 69 REV. ECON.
STUD. 811 (2002) (modeling delegation as an alternative to communication); Steven D. Levitt &
Christopher M. Snyder, Is No News Bad News? Information Transmission and the Role of “Early
Warning” in the Principal-Agent Model, 28 RAND J. ECON. 641 (1997) (analyzing optimal
incentive contracts when an agent has information about the likelihood of the project’s success not
available to the principal, except by the agent’s disclosure); Roy Radner, The Organization of
44

Published by The Berkeley Electronic Press, 2008

11

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

organizations by way of specific “structures” or “channels.”47 In a discussion of
information and its relation to pricing and risk-bearing, Arrow noted:
It follows that the information structure of individual economic
agents powerfully conditions the possibilities of allocating riskbearing through the market. . . . Thus the possibility of using the
price system to allocate uncertainty, to insure against risks, is
limited by the structure of the information channels in existence.
Put the other way, the value of nonmarket decision-making, the
desirability of creating organizations of a scope more limited than
the market as a whole, is partially determined by the characteristics
of the network of information flows.48
The point, which Arrow did not limit to pricing and risk but rather made
generally for any organizational use of information, is that information structures
may well predate a firm’s calls upon those structures, and determine whether and
how well those calls will be answered. As he said some pages later, “history
matters.”49 Yet, Arrow continued:
[T]he presence or absence of information channels is not
prescribed exogenously to the economic system. Channels can be
created or abandoned, and their capacities and the types of signals
to be transmitted over them are subject to choice, a choice based
on a comparison of benefits and costs.50
Arrow proceeded to elaborate the costs of information gathering and
decoding, and then made three different points relevant here. First, he noted, “the
demand for investment in information is less than it would be if the value of the
information were more certain.”51 Second and, Arrow surmised, likely most
important, information investments are path-dependent. Or, as he put it, “random
accidents of history will play a bigger role in the final equilibrium.”52 The reason
is that “[o]nce the investment has been made and an information channel
acquired, it will be cheaper to keep on using it than to invest in new channels.”53
Third, the price or effort needed to develop new information structures or
Decentralized Information Processing, 61 ECONOMETRICA 1109 (1993); Timothy Van Zandt,
Real-Time Decentralized Information Processing as a Model of Organizations with Boundedly
Rational Agents, 66 REV. ECON. STUD. 633 (1999).
47
ARROW, LIMITS OF ORGANIZATION, supra note 5, at 37.
48
Id.
49
Id. at 56.
50
Id. at 37.
51
Id. at 41.
52
Id.
53
Id. For more general discussions of the path-dependent results of firm activities, see
Kenneth J. Arrow, The Economic Implications of Learning by Doing, 29 REV. ECON. STUD. 155
(1962); Alwyn Young, Invention and Bounded Learning by Doing, 101 J. POL. ECON. 443 (1993).
http://www.bepress.com/jtl/vol2/iss1/art1

12

Schlanger: Operationalizing Deterrence

channels varies based on the organization’s prior activities, both related to
information and to production. In particular, Arrow suggested, expertise transfers
better across small conceptual gaps than large ones: “Learning generalizes
naturally and cheaply in some directions, with much greater difficult in others.”54
Likewise, often information can be relatively cheap because of ongoing related
productive activities: “There is a complementary between a productive activity
and some kinds of information. An individual cannot help making observations
while working at some task.”55
My suggestion that claims management can aid in care-taking by promoting
the gathering and analysis of claims information, which information may then be
used to monitor and improve harm prevention, harvests insights from both Yates’
and Arrow’s work. Organizations make investments in information gathering and
processing for the purpose of efficient and effective processing of claims. That
investment, once made, will tend to become somewhat entrenched. And it may
well encompass personnel who have sufficient extra capacity to shift or expand
their mission, if that is useful to them (a point I develop in the next section).
Moreover, “an information channel used primarily for one purpose”—here,
claims management—“may turn up a signal with implications for taking action in
a hitherto passive area”56—here, harm prevention. As Gary Schwartz put the
point in the context of insurance, it makes sense that insurers sell loss prevention
services bundled with claims management services; the result economizes on
information costs, because “the insurer acquires its [loss-prevention] information
in a natural low-cost way, as an incident to its normal activities of underwriting
and claims evaluation.”57
B. Professional Personnel
The case studies below suggest that personnel assigned to claims management
tasks sometimes take on the additional job of preventing harm or promoting legal
compliance. This shift is not unexpected, in light of sociological findings about
how professionals function within organizations.
Organizational sociologists suggest that law’s impact on organizational
practice is mediated by the content of the structures chosen by organizational
participants. The relevant choices are in part rationally and instrumentally made.
As Philip Selznick described in his early work, organizations are designed as
54

ARROW, LIMITS OF ORGANIZATION, supra note 5, at 41-42; see also H.B. Malmgren,
Information, Expectations and the Theory of the Firm, 75 Q. J. ECON. 399 (1961) (noting ways in
which the costs of information acquisition may vary based on prior firm investments or activities).
55
ARROW, LIMITS OF ORGANIZATION, supra note 5, at 42.
56
Id. at 58.
57
Schwartz, Ethics and Economics of Insurance, supra note 8, at 356.
Published by The Berkeley Electronic Press, 2008

13

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

“technical instrument[s] for mobilizing human energies and directing them toward
set aims,” and “conceived as an exercise in engineering, . . . governed by the
related ideals of rationality and discipline.”58 But, Selznick continued, over time
organizations become “institutionalized,” a process “reflecting the organization’s
own distinctive history, the people who have been in it, the groups it embodies
and the vested interests they have created, and the way it has adapted to its
environment.”59 As he defined it, “‘to institutionalize’ is to infuse with value
beyond the technical requirements of the task at hand.”60 The interaction of
professionals of different stripes and their organizations, and the development of
the relevant professional orientations towards claims and risk management, are a
key site for inquiry into the what Selznick would call institutionalization.
DiMaggio and Powell, for example, point to the role of professional training,
career paths, and networks in molding organizational structures and routines,
which they argue tend to evolve “isomorphically,” to resemble those in other
similar firms.61 As described by Edelman, Fuller, and Mara-Drita: “Professionals
carry ideas as they move among organizations, and through participation in
professional networks: conferences, workshops, and the professional personnel
literature all offer forums for the exposition and diffusion of new ideas within
professions. . . . [C]ertain ideas, practices, routines, and scripts become
institutionalized as the professions offer normative solutions to perceived
managerial problem[s].”62
The work of Lauren Edelman and her many coauthors is particularly
enlightening on the topic of professional personnel and legal compliance.63 As
Edelman et al. summarize the findings of a long series of investigations into the
58

SELZNICK, LEADERSHIP IN ADMINISTRATION, supra note 41, at 5.
Id. at 16.
60
Id. at 17 (emphasis in original); see also, e.g., Stewart Macaulay, Private Government, in
LAW AND THE SOCIAL SCIENCES 445 (Leon Lipson & Stanton Wheeler eds., 1986).
61
See DiMaggio & Powell, The Iron Cage Revisited, supra note 41, at 152-54.
62
See Lauren B. Edelman, Sally Riggs Fuller, & Iona Mara-Drita, Diversity Rhetoric and the
Managerialization of Law, 106 AM. J. SOC. 1589, 1596 (2001).
63
Id.; Lauren Edelman, Legal Environments and Organizational Governance: The Expansion
of Due Process in the American Workplace, 95 AM. J. SOC’Y 1401 (1990); Lauren B. Edelman,
Stephen Petterson, Elizabeth Chambliss & Howard S. Erlanger, Legal Ambiguity and the Politics
of Compliance: Affirmative Action Officers’ Dilemma, 13 LAW & POL’Y 73 (1991); Lauren
Edelman, Legal Ambiguity and Symbolic Structures: Organizational Mediation of Civil Rights
Law, 97 AM. J. SOC’Y 1531 (1992); Lauren B. Edelman, Stephen E. Abraham & Howard S.
Erlanger, Professional Construction of Law: The Inflated Threat of Wrongful Discharge, 26 LAW
& SOC’Y REV. 47, 48-49 (1992); Lauren Edelman & Mark C. Suchman, The Legal Environments
of Organizations, 23 ANNUAL REV. SOC’Y 479 (1997); Lauren B. Edelman & Mark C. Suchman,
When the “Haves” Hold Court: Speculations on the Organizational Internalization of Law, 33
LAW & SOC’Y REV. 941, 963 (1999); Lauren Edelman et al., The Endogeneity of Legal
Regulation: Grievance Procedures as Rationalized Myth, 105 AM. J. SOC’Y 406 (1999).
59

http://www.bepress.com/jtl/vol2/iss1/art1

14

Schlanger: Operationalizing Deterrence

law governing and large organization personnel offices, those investigations
support three important theoretical points. First, “legal ambiguity amplifies the
opportunities for professionals to identify management problems and to propose
new ideas to remedy those problems.”64 Second, this mediation of law by
professionals is consequential: “[I]deas about law and compliance that originate
with the professions tend to become institutionalized in organizational fields and,
over time, generate a diffusion of new organizational practices.”65 And third,
“[a]s law is communicated by and among professions, it is filtered through a
variety of lenses, and colored by different professional backgrounds, training, and
interests.”66
Edelman’s work, and that of others, demonstrates several of the different
results of professional filtering of the law’s message. For example, in some
circumstances, “to inflate their own status within organizations and to expand the
markets for their services, the professional [may] create[] the impression of a
much greater threat than . . . doctrine actually pose[s],”67 thus “amplify[ing] the
threat of law.”68 Alternatively (though it may look much the same), personnel
assigned or hired to ensure compliance with legal norms may “tend to become
internal advocates for the values that the practices symbolize.”69 In still other
circumstances, professionals within firms may “transform[] legal threats into
nonlegal problems”70 when “complaint handlers tend[] to recast discrimination
complaints as managerial or interpersonal problems, thus deemphasizing the legal
aspects of these claims.”71 In yet other settings, professionals may “construct[]
the nature of the response rather than the legal threat itself,” promulgating the
notion that “grievance procedures offer[] organizations substantial protection
64

Edelman et al., Diversity Rhetoric, supra note 62, at 1596.
Id.
66
Id.
67
Id.
68
Id. at 1597.
69
Edelman & Suchman, When the “Haves” Hold Court, supra note 63, at 941, 963. Serge
Taylor points to a similar situation when environmental analysts are hired by government
agencies, because they have the technical skills to produce environmental impact statements. He
observes:
Environmental analysts . . . tend to have distinctive personal values. Being strongly
committed to professional standards of “truth,” they want their agency to study and reveal all
potentially significant environmental problems. Being strongly committed to environmental
policy values, they often disagree with their organization’s decisions. For both reasons, they
are potentially disloyal to their organization. And disloyalty can have can have significant
political and legal consequences, by providing environmentalists on the outside with potent
ammunition against their agency’s projects.
SERGE TAYLOR, MAKING BUREAUCRACIES THINK: THE ENVIRONMENTAL IMPACT STATEMENT
STRATEGY OF ADMINISTRATIVE REFORM 6 (1984).
70
Edelman et al., Diversity Rhetoric, supra note 62, at 1597.
71
Id. at 1596-97.
65

Published by The Berkeley Electronic Press, 2008

15

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

from discrimination lawsuits,” despite the dearth of evidence to support that
claim.72 The organizational environments themselves may further mold the effect
of professional orientation; my own prior work has pointed to the ways a context
can influence professional conceptualization of legal compliance. I pointed out
that in the oppositional milieu of jail and prison management, compliance
personnel may need to “ensure that they are not too deeply identified with the
inmates by their colleagues,” and may therefore “develop[] a finely honed
derision for inmate complaints.”73
What all these factually specific discussions have in common is that they
support the theoretical point that while “[t]he precise impact of the professions
varies,” “professionals tend to interject their own interests and training into how
they understand law.”74 Scholars in other disciplines and traditions agree: They,
too, point out that different actors within organizations bring to bear different
mind-sets.75 The case studies below develop the ways in which professionals
assigned to claims management tasks sometimes expand their mission to
encompass, as well, harm prevention or legal compliance.
C. Bureaucracy: Written Records and Standard Operating Procedures
Two of the case studies (hospitals and prisons/jails) suggest that claims
management needs promote the use of contemporaneous written records. Such
records are useful for claim defense because compared to oral testimony they can
be produced more easily and earlier in litigation, and are often more persuasive.
In addition, the prison and jail case study in particular suggests that claims
management priorities can push organizations to use standard operating
procedures; these can help reduce liability because they can be characterized in
court as rational and even expert and therefore worthy of deference, and because
they provide circumstantial evidence of due care or compliance even when no
defense witness has a specific recollection relating to the harm-causing event.
Both of these organizational strategies, geared towards claims management, turn
out to have potential for harm reduction, as well.
72

Id. at 1597.
Schlanger, Inmate Litigation, supra note 4, at 1671.
74
Edelman et al., Diversity Rhetoric, supra note 62, at 1597.
75
See, e.g., W. Bernard Carlson, The Coordination of Business Organization and
Technological Innovation within the Firm: A Case Study of the Thomson-Houston Electric
Company in the 1880s, in COORDINATION AND INFORMATION: HISTORICAL PERSPECTIVES ON THE
ORGANIZATION OF ENTERPRISE 55, 59 (Naomi R. Lamoreaux & Daniel M.G. Raff eds., 1995) (“I
find it useful to think of the firm as a collection of interest groups, each with its own mind-set.”);
id. at 60 (citing as a precursor SAMUEL B. BACHARACH & EDWARD J. LAWLER, POWER AND
POLITICS IN ORGANIZATIONS: THE SOCIAL PSYCHOLOGY OF CONFLICT, COALITIONS, AND
BARGAINING (1980)).
73

http://www.bepress.com/jtl/vol2/iss1/art1

16

Schlanger: Operationalizing Deterrence

Note first that use of standard operating procedures and of written records are
both core components of bureaucratization. Indeed, the core elements of
bureaucracies, according to Weber, bureaucracy’s great expositor, are: their
dependence on rules and officially designated duties; their hierarchical
organization, under which higher bureaus supervise lower ones; their adherence to
principles of expertise and training, and, therefore, standardization of tasks along
the lines developed by experts; and their use of written records.76 It should be
expected, then, that claims management should, in turn, at least sometimes drive
bureaucratization.
How might standardization and written records prevent harm and encourage
legal compliance? Standardization of operations, in particular, has familiar costs
and benefits. Standard operating procedures tend to reduce line-level discretion,
which can be extremely beneficial (to the task, if not the autonomy of the affected
workers77) if the task in question is one that can be done in what Frederic
Winslow Taylor labeled the “one best way.”78 Thoughtful and informed
standardization can promote decisionmaking and conduct that is more expert,
more effective, less harmful.79 Of course, in other, more complex or ambiguous,
circumstances, line-level discretion can be more beneficial than standardization to
the ultimate values sought to be implemented. It is well beyond the scope of this
article to develop the circumstances in which discretion is better promoted or
76

See, e.g., MAX WEBER, THE THEORY OF SOCIAL AND ECONOMIC ORGANIZATION 329-36
(A.M. Henderson & Talcott Parsons trans., The Free Press 1947) (1925). Weber thought these
features key to the effectiveness of bureaucracy, which he famously described as “superior to any
other form in precision, in stability, in the stringency of its discipline, and in its reliability. It thus
makes possible a particularly high degree of calculability of results for the heads of the
organization and for those acting in relation to it. It is finally superior both in intensive efficiency
and in the scope of its operations.” Id. See also GARY J. MILLER, MANAGERIAL DILEMMAS: THE
POLITICAL ECONOMY OF HIERARCHY (1992). Miller agrees that hierarchical organizations can be
extremely effective, but he argues that hierarchies succeed not because of their advantages with
respect to incentives and supervision, but because hierarchy “can be a means for creating common
knowledge and cooperative work norms.” Id. at 217.
77
For accounts sympathetic to the complaint of lost autonomy, see, e.g., Karl E. Klare, The
Labor-Management Cooperation Debate: A Workplace Democracy Perspective, 23 HARV. C.R.C.L. L. REV. 39 (1988); Marion Crain, The Transformation of the Professional Workforce, 79
CHI.-KENT L. REV. 543 (2004). For a discussion of autonomy as an obstacle to appropriate caretaking, because it interferes with more rational, standardized operations, see David A. Hyman &
Charles Silver, The Poor State of Health Care Quality in the U.S.: Is Malpractice Liability Part of
the Problem or Part of the Solution?, 90 CORN. L. REV. 893, 921, 952-56 (2005).
78
See ROBERT KANIGEL, THE ONE BEST WAY: FREDERICK WINSLOW TAYLOR AND THE
ENIGMA OF EFFICIENCY (1997); FREDERICK WINSLOW TAYLOR, PRINCIPLES OF SCIENTIFIC
MANAGEMENT (1911).
79
See, e.g., KENNETH CULP DAVIS, POLICE DISCRETION (1975); Brigette M. Hales & Peter J.
Pronovost, The Checklist—A Tool for Error Management and Performance Improvement, 21 J.
CRITICAL CARE 231 (2006).
Published by The Berkeley Electronic Press, 2008

17

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

restrained;80 suffice it to say that if claims management encourages organizations
to develop and implement standard operating procedures—as the hospital and
prison/jail case studies suggest that it does—there will be occasions in which that
rein on discretion is helpful for harm prevention.
Likewise, when claims management imperatives support contemporaneous
recordkeeping because of the usefulness of contemporaneous records in court,
those records are sometimes also useful for harm prevention; contemporaneous
records, first, enable closer review of line-level staff decisions by their
hierarchical superiors, and second, themselves assist in beneficial standardization.
The first point—that written records are more efficient than other methods in
transferring information within large organizations81—may be more intuitive than
the second. But the second point is equally clear. Generation of written records
such as logs, reports, charts, and the like is regulatory as well as communicative,
because the form in which the record is kept can structure job performance and
decisionmaking, as with a checklist82 or a log with a space for observations once
each specified period. Moreover, substantive reporting requirements of various
kinds are well known to foster the report-makers’ consideration of issues that
might otherwise receive less notice.83
80

The literature attempting to do just that is large and very interesting. In law, it often
concerns the debate between judicially enunciated rules and standards. See, e.g., Duncan
Kennedy, Form and Substance in Private Law Adjudication, 89 HARV. L. REV. 1685 (1976);
Pierre Schlag, Rules and Standards, 33 UCLA L. REV. 379 (1985); Carol M. Rose, Crystals and
Mud in Property Law, 40 STAN. L. REV. 577 (1988); Antonin Scalia, The Rule of Law as a Law of
Rules, 56 U. CHI. L. REV. 1175 (1989); FREDERICK SCHAUER, PLAYING BY THE RULES: A
PHILOSOPHICAL EXAMINATION OF RULE-BASED DECISION-MAKING IN LAW AND IN LIFE (1991);
Louis Kaplow, Rules versus Standards: An Economic Analysis, 42 DUKE L.J. 557 (1992);
Kathleen Sullivan, Foreword: The Justices of Rules and Standards, 106 HARV. L. REV. 22 (1992);
Cass R. Sunstein, Problems with Rules, 83 CAL. L. REV. 953 (1995); Russell B. Korobkin,
Behavioral Analysis and Legal Form: Rules vs. Standards Revisited, 79 OR. L. REV. 23 (2000);
Steven M. Shavell, Optimal Discretion in the Application of Rules, 9 AM. L. & ECON. REV. 175
(2007).
81
JoAnne Yates’ historical account of information technology from 1850 to 1920, discussed
above, demonstrates this point well in its description of the importance of document copying
technology to hierarchical control within large firms such as Sears and the Scovill Manufacturing
Company. YATES, Investing in Information, supra note 44.
82
See MICHAEL SCRIVEN, THE LOGIC AND METHODOLOGY OF CHECKLISTS (2000, revised
Dec.
2007)),
available
at
http://www.wmich.edu/evalctr/checklists/papers/
logic&methodology_dec07.pdf; Hales & Pronovost, The Checklist, supra note 79.
83
See, e.g., TAYLOR, MAKING BUREAUCRACIES THINK, supra note 69; Bradley C.
Karkkainen, Information as Environmental Regulation: TRI and Performance Benchmarking,
Precursor to a New Paradigm?, 89 GEO. L.J. 257 (2001); cf. Bengt Holmstrom & Paul Milgrom,
Multitask Principal-Agent Analyses: Incentive Contracts, Asset Ownership, and Job Design, 7 J.L.
ECON. & ORG. 24, 33-38 (1991) (suggesting that easily measured tasks are preferred by multitasking agents, at least if their incentives are tied to performance).
http://www.bepress.com/jtl/vol2/iss1/art1

18

Schlanger: Operationalizing Deterrence

D. Salience
Finally, various claims management techniques may remind organizational actors
of both the possibility of mishap and the threat of litigation that might result. This
reminder makes those possibilities more “available” or cognitively salient. As
Amos Tversky and Daniel Kahneman have explained, “people assess the
frequency of a class or the probability of an event by the ease with which
instances or occurrences can be brought to mind.”84 Because people tend to be
overconfident about the threat to them of bad outcomes,85 deterrence may depend
on such countervailing availability.86 This point is analytically simple, so its
theory does not need much elaboration—but the hospital and prison case studies
suggest it is no less important for its simplicity.
In short, organizational theory, sociology, economics, and psychology all
suggest that certain strategies that organizations use, ex post, to respond
efficiently and effectively to tort claims may also generate the byproduct of harm
reduction and legal compliance. The next part examines claims management in
three specific organizational settings, to put some meat on the theoretical bones
just described.
III. CASE STUDIES
I chose three settings in which to investigate claims management by looking for
two common features—frequent claiming and a significant degree of in-house
processing—along with enough variation to surface a range of possible responses
to claims management imperatives. Thus I picked organizations of different
sizes, facing different liability environments and threats, operated by governments
and private entities, profit-seeking and not. The case studies are the factual
source rather than the proof of this article’s theory; they help with hypothesis
generating rather than hypothesis testing.
Table 1 summarizes some important characteristics of the case study
organizations.
84

Amos Tversky & Daniel Kahneman, Judgment under Uncertainty: Heuristics and Biases,
185 SCIENCE 1124, 1127 (1974); see also Korobkin & Ulen, Law and Behavioral Science, supra
note 37, at 1085-90 (citing additional studies on availability bias).
85
Christine Jolls, Behavioral Economics Analysis of Redistributive Legal Rules, 51 VAND. L.
REV. 1653, 1659 & n.22 (1998) (noting the existence of nearly two hundred studies demonstrating
optimism bias).
86
Eric A. Posner, Probability Errors: Some Positive and Normative Implications for Tort and
Contract Law, 11 SUP. CT. ECON. REV. 125 (2004) (describing how excess optimism might
undermine tort deterrence). Posner emphasizes, as well, that potential defendants’ mistakes with
respect to the change in the probability of harm caused by their care may induce them to take too
much care, but this point seems to me not pertinent here.
Published by The Berkeley Electronic Press, 2008

19

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

Table 1: Case study comparisons

Type of
organization

Size of enterprise

Annual Annual lawsuits
claims (subset of claims)
(N)
(N)
Annual payouts

Retailer: “Acme” $6 billion in sales;
Private, for-profit $100 million in net
(2004)
income

500

15

Large academic ~1000 staffed beds;
~ $1 billion gross
hospital
Private, non-profit
revenues
(2006)

100150

50-100

4000

400

Ratio—
Payouts:
budget

$3 million

1: 2000

$14 million
(insurance
premiums +
defense costs +
payouts)

1:70

Corrections
Public, non-profit
Federal Bureau $2.6 billion annual
of Prisons
budget
(1992-1998)

$150,000 (claims) 1:1200
$2 million (suits)

Cal. Dept. of
$4.6 billion annual
?
585
$6.2 million
1:750
Corr. (1999)
budget
Sources: Acme: 2004 Annual Report and interview with Acme director of risk management.
See infra note 87.
Hospital: Interviews with medical school director of risk management; academic
hospital director of risk management.
Federal BOP: Results of FOIA request, see infra note 174.
Cal. Dep’t of Corrections: California documents on file with author, see infra notes
184-185.

A. Acme
I gathered information about the publicly held corporation I will call “Acme” in
2005. It is a large retailer87—in 2004 it had over $6 billion in annual sales and
over $100 million in annual net income, over 150 stores, and about 20,000
employees. Damages claims made against Acme are significant but not
overwhelming; As of 2005, Acme’s stores saw about 500 potentially compensable
accidents a year, mostly categorized by its risk management office as “slips, trips,
and falls,” and “struck-by’s” (when customers are hit by merchandise falling from

87

Information about “Acme” comes from its 2004 annual report, on file with the author, and
from an extensive telephone interview (Feb. 15, 2005) of the head of its risk management
department. I was given full information on the condition that I shield the identity of the
corporation.
http://www.bepress.com/jtl/vol2/iss1/art1

20

Schlanger: Operationalizing Deterrence

store shelves).88 The company paid about $3 million each year to its claimants.
Most were settled without any lawsuit or even lawyer; only about 50 had
attorneys involved, and only 15 got to actual litigation. This large base of claims
and small point of litigated cases appears to create a fairly typical dispute
pyramid.89 The vast majority of the payout amount was due to just a few large
claims. Acme self-insured for all claims below $500,000.
To understand the impact of claiming costs, an intra-firm comparison may be
useful. Acme’s claims payouts in 2006 were lower than its losses to shop-lifting:
“inventory shrinkage” accounted, Acme’s annual report states, to less than 0.2%
of net sales, which amounted to something under $12 million, compared to about
$3 million in claims (.05% of net sales). Still, three million dollars is not an
insignificant amount of money, and 500 claims annually was likewise a number
large enough to require an infrastructure.
Because Acme needs to be able to process the claims its customers predictably
bring, it has developed a procedure that store management follows whenever there
is an accident. Each and every time a customer complains about an accident,
store management is directed to assemble a file, complete with statements and
pictures. The file is then forwarded to a central risk management office. Acme’s
head of risk management (I’ll call him A., for ease of reference) explains that
claims are then processed entirely in-house for all but the few in which claimants
are represented by counsel. Small cases dominate Acme’s claims docket. In fact,
because Acme’s average customer spends several thousand dollars annually at
Acme, the loss of an average customer costs more, for Acme, than an average
claim. Accordingly, A. considers the most important quality for a good claims
manager to be strong orientation to customer service. Great customer service
people (“people people,” A. calls them) can typically settle claims by customers
for store coupons, discounts, merchandise, or membership fee waivers. These are
attractive settlement items for Acme because they have a higher value to
customers than their cost to Acme, and because they assist in preserving the
customer’s connection to the firm.
Acme is just one firm, and I have no information on how typical its claims
management features are. But its situation is at least suggestive. In any event, at
Acme, claims management is linked to harm prevention in three of the four ways
88

These categories are common in similar operations. One source reports that Wal-Mart, for
example, had by 2001 recorded “about 30,000 falling-merchandise incidents resulting in injuries
to customers.” Jeffrey A. Hyman & Molly E. Homan, Falling Merchandise, TRIAL, January 2001,
at 44. See also Lea S. VanderVelde, Wal-Mart as a Phenomenon in the Legal World: Matters of
Scale, Scale Matters, in WAL-MART WORLD: THE WORLD’S BIGGEST CORPORATION IN THE
GLOBAL ECONOMY 115, 130 (Stanley D. Brunn ed., 2006) (“Claims against Wal-Mart typically
involve injuries resulting from slips and falls and from falling merchandise.”).
89
William Felstiner, Richard Abel & Austin Sarat, The Emergence and Transformation of
Disputes: Naming, Blaming, Claiming, 15 LAW & SOC’Y REV. 631 (1981).
Published by The Berkeley Electronic Press, 2008

21

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

theorized above. Claims management imperatives have promoted information
production and use, the hiring of risk management staff, and the salience of
accident-related losses:
Information production. I have already mentioned the accident report and
claims file that store managers, by policy, assemble for each customer claim. A.
conceptualizes this policy as serving a core claims management purpose. He
describes the policy as requiring store managers to do “everything that can be
done early on after an event to solidify the file” so that it can be used to respond
to subsequent litigation. Obviously, there is a large liability defense advantage in
assembling such a file right away. Whatever conditions caused the accident are
still present to be described or photographed; witnesses can be identified and their
statements can be memorialized. The file can therefore be both more
comprehensive and more accurate, less costly to assemble, and available both for
immediate use in negotiations and eventual use in litigation. The rationale for
assembling the files, then, is a claims management one.
In addition to assembling individual files, Acme tracks claims statistically, in
order to monitor costs. Here, Acme’s primary design seems split between claims
management and harm reduction. The claims management purpose is that part of
managing a claim is accounting for losses, which are paid centrally but attributed
back to the store that occasioned them; when a claim occurs, the involved store’s
profit and loss statement is “taxed” $20,000. This is quite rough justice, of
course—most of the losses are a great deal less costly to Acme. That
approximation also serves a risk-reduction purpose, however. The accounting
treatment of claims payouts is an important component of Acme’s loss-prevention
program, because of its incentive effects. The $20,000 tax on the profit and loss
statement works out to about $300-500 less for its manager’s annual bonus. In
addition, a collective fund for employees at each store (for parties and the like) is
credited with $250 for each quarter that is accident free, and an extra $1000 for
each year that is accident free. The point is to give all of the relevant
organizational actors some limited but real reason to notice and wish to reduce
claims.
These basic statistics are not, however, the end of the analysis. Once the files
and tracking information are assembled, they are used not only for claims
management but for accident prevention as well. In particular, they are used to
generate strategic information that central management analyzes in order to
instruct store managers how to reduce accidents. For example, A. explains that
files for claims coming out of Florida alerted Acme to the problem that stores
were not changing floor mats quickly enough after rainstorms, so that customers
were tracking in water, which caused falls. The problem has, he says, been
corrected now, and subsequent tracking demonstrates that claims are down in the
relevant categories. The information generated to satisfy Acme’s need to respond

http://www.bepress.com/jtl/vol2/iss1/art1

22

Schlanger: Operationalizing Deterrence

and process litigation efficiently, and keep track of its costs in that arena of
operations, is thus harnessed in service of safety improvements. This is not an
isolated example; such interventions happen frequently, A. says.
Personnel. The strategic use of claims information described above is far
from inevitable. Logically, it seems to depend most crucially on the second
issue—the presence of claims personnel with time and inclination to take on a
broader risk-management mission. Until a few years prior to my interview of
Acme’s risk manager, A., Acme had contracted with a third-party administrator to
process claims. A. explained that this approach turned out not to be a good one;
“Their thought process was deny, deny, deny, fight, fight, fight.” The third-party
combatants spent far more of Acme’s money than its in-house people. A.
explains that because Acme’s own personnel come to the task with customer
service rather than claims skills, they are interested in preserving customer
satisfaction, and that works out cheaper. In other words, Acme’s current “makeor-buy” decision, choosing in-house over third-party claims administration, was
reversed for claims management, not harm prevention, reasons.
Nonetheless, the decision to go in-house has clear harm prevention effects, as
well. In-house claims managers like A. conceptualize their role as reducing the
impact of claims on Acme’s bottom line. They can achieve that role by
economizing on claims processing costs, or claims settlement costs, or by
reducing the number of harmful incidents that cause claims altogether. And they
often have a long-standing background with and commitment to the firm, with
concomitant insight into its operations. At the time I interviewed him, A. had
spent nearly 15 years working at Acme. Lacking similar inside information or
access, third-party claims managers have much less ability than their internal
counterparts to take on a harm-reduction role. Moreover, ordinarily they have
much less incentive to do so, because they are paid based on claims processing,
not claims reduction. (Here they are unlike insurers, which have an incentive to
decrease payouts, and therefore take the kind of regulatory measures described in
the introduction. But I am not aware of any literature describing situations in
which insurers perform the kind of firm-specific risk analysis that is ordinary
within Acme’s claims management operation.)
Salience. At Acme, claims management techniques can and often do serve as
reminders of safety imperatives. Because the file is assembled at a low level
within the organization, the very act of putting it together—the interviews and
other investigation—joins up with the accounting techniques used to bring home
to Acme’s managers and lower level employees the lesson that accidents do
happen. As explained in Part II, this makes the need for harm prevention more
available, cognitively.

Published by The Berkeley Electronic Press, 2008

23

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

B. Hospital Medical Malpractice
In almost every way, risk and claims management look very different for
hospitals than for retailers. The most important difference is that medical injuries
are both more common and more serious than the injuries Acme processes.
Iatrogenic injury—injury caused by medical treatment—is a leading cause of
death in the United States; the much-publicized Institute of Medicine report, To
Err is Human, estimated that preventable medical errors, which cause a bit over
half the iatrogenic injuries experienced by patients, kill between 44,000 and
98,000 Americans each year—more than car accidents, breast cancer, or AIDS.90
Another key difference is that causation is far more opaque for medical injuries
than for Acme’s “slips, trips, and falls” and “struck-by’s.” After all, it is in the
nature of hospital patients that they are, in large part, sick. It can be very difficult
to figure out whether a patient’s bad outcome is attributable to treatment at all,
much less to error or negligence rather than misfortune in that treatment.91 Also
important is the extremely low rate of claiming.92 Only a very small percentage
of those who suffer even serious avoidable harm in hospital settings actually make
claims, whether because they do not know if they were the victims of malpractice
or other breaches; because those same informational and proof difficulties
(coupled with the expense of litigating) make their cases unenticing to lawyers; or
because they like and trust their doctors.93 To repeat, causation, by contrast, is
90

INSTITUTE OF MEDICINE, TO ERR IS HUMAN: BUILDING A SAFER HEALTH SYSTEM 26
(2000). For sources making a range of estimates on iatrogenic injury in American hospitals, see
Hyman & Silver, Problem or Solution, supra note 77, at 901-04 & nn.23-41.
91
The most extensive study of adverse events was known as the Harvard Medical Practice
Study; it used an elaborate protocol to identify first treatment-caused, and then more narrowly
tortiously caused injuries. For a description of the protocol, and results, see HARVARD MEDICAL
PRACTICE STUDY, PATIENTS, DOCTORS, AND LAWYERS: MEDICAL INJURY, MALPRACTICE
LITIGATION, AND PATIENT COMPENSATION IN NEW YORK: THE REPORT OF THE HARVARD
MEDICAL PRACTICE STUDY TO THE STATE OF NEW YORK ch. 5 (1990); Troyen Brennan et al.,
Incidence of Adverse Events and Negligence in Hospitalized Patents: Results of the Harvard
Medical Practice Study I., 324 N. ENG. J. MED. 370 (1991); Lucian Leape et al., The Nature of
Adverse Events in Hospitalized Patients: Results of the Harvard Medical Practice Study II, 324 N.
ENG. J. MED. 377 (1991).
92
For summaries of the evidence on claiming rates, see TOM BAKER, THE MEDICAL
MALPRACTICE MYTH 22-38 (2005), and sources there cited; David A. Hyman & Charles Silver,
Medical Malpractice Litigation and Tort Reform: It’s the Incentives, Stupid, 59 VAND. L. REV.
1085, 1113-15 (2006). Interesting additional evidence, comparing claiming rates to observed
clinical discussions of errors in patient care is presented in Lori Andrews, Studying Medical Error
in Situ: Implications for Malpractice Law and Policy, 54 DEPAUL L. REV. 357 (2005), and Lori B.
Andrews et al., An Alternative Strategy for Studying Adverse Events in Medical Care, 349 LANCET
309 (1997).
93
See sources cited supra note 92; Marlynn L. May & Daniel B. Stengel, Who Sues Their
Doctors? How Patients Handle Medical Grievances, 24 LAW & SOC’Y REV. 105 (1990).
http://www.bepress.com/jtl/vol2/iss1/art1

24

Schlanger: Operationalizing Deterrence

typically cartoonishly obvious in Acme’s cases, and presumably there is no
similar bond that discourages claiming against retailers.
Organizationally, retailers and hospitals are also extremely different.
Hospitals have the significant complication that often different risk management
and claims management systems serve the hospital’s attending physicians, who
may be, technically, independent contractors (and for whose torts the hospital
itself may therefore not be vicariously liable) and hospital employees like nurses,
custodial staff, technicians of various kinds. Interns and resident physicians may
be employed by the hospital itself or by some (related or affiliated) organization.94
A medical school, might, for example, run risk management and claims
management for non-resident physicians at its teaching hospitals. In addition,
hospitals themselves do risk and claims management for the parts of operations
for which they are legally responsible. (Some hospitals, however, merge these
two offices. For example, risk management for the Harvard Medical School
affiliated hospitals covers both the hospitals and their employees and also the
independent contractor physicians.95)
As with Acme, the components of hospitals that process claims are referred to
as “risk management” rather than “claims management” offices. The risk referred
to was, at least in hospital risk management’s infancy, conceptualized as
monetary rather than as health-care related. Thus the American Hospital
Association, which “helped to foster the acceptance of risk management in health
care,” defined risk management as the “science for the identification, evaluation,
and treatment of the risk of financial loss.”96 And claims have been the central
preoccupation of hospital risk management, which was born in the medical
malpractice insurance crisis of the mid-1970s.97 As one observer has commented
(with regret):
[T]he dollar impact of claims management and insurance
purchasing activities is easily, if deceptively, measured upon
94

See John D. Hodson, Annotation, Liability of Hospital or Sanitarium for Negligence of
Physician or Surgeon, 51 A.L.R. 4TH 235 (1987).
95
At the Harvard hospitals, risk management and claims management is run by a unit of
Controlled Risk Insurance Company (CRICO), the “captive” insurance company owned by the
medical institutions it insures. See http://www.rmf.harvard.edu/company/about-us.aspx.
96
Deborah Korleski, The Emergence of a Profession, in ESSENTIALS OF HOSPITAL RISK
MANAGEMENT 1, 4 (Barbara J. Youngberg ed., 1990), citing Todd Dankmyer & James Groves,
Taking Steps for Safety’s Sake, HOSPITALS, May 16, 1977, at 60; Sally T. Holloway & Arline B.
Sax, AHA Urges, Aids Hospitals to Adopt Effective Risk Management Plans, HOSPITALS, May 16,
1977, at 57.
97
Deborah Korleski, Emergence of a Profession, supra note 96, at 1; JAMES E. ORLIKOFF
WITH AUDRONE M. VANAGUNAS, MALPRACTICE PREVENTION AND LIABILITY CONTROL FOR
HOSPITALS 32-34 (2d ed. 1988). Indeed, some hospitals hired risk managers only in order to
comply with a requirement of their insurance contract. Korleski, supra, at 3, ORLIKOFF WITH
VANAGUNAS, supra, at 34; cf. supra text accompanying notes 8-22.
Published by The Berkeley Electronic Press, 2008

25

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

superficial examination.
Consequently, patient injury and
malpractice loss prevention activities are often accorded secondclass status in hospitals when compared with the more financial,
and therefore more apparently measurable, areas of risk
management.98
Another discussion notes that “the most common function associated with risk
management programs” is its response “to limit the liability associated with
incidents that have occurred.”99 Hospital risk managers interact with insurance
personnel and lawyers, with the assigned role of making litigation response more
efficient and effective. A sample job description for a risk manager published in
the mid-1980s prioritized legal and insurance rather than clinical or management
expertise, emphasizing that the risk manager should “[s]erve as legal affairs
person in conjunction with hospital defense counsel” and possess a “B.A. degree
with basic knowledge of insurance industry,” and “[i]nterest in or in pursuit of
law courses related to risk management.”100
However, this professional tilt towards claims, law, and insurance has
changed substantially over the past decade or so, as hospital risk managers’
orientation has leaned increasingly towards safety and patient care. I describe this
shift below, as part of the discussion of personnel, but to foreshadow that
discussion here, many hospital risk managers have worked hard to shift the
central preoccupation of risk management towards patient safety improvement.
Just as with Acme, hospitals have processes that are supposed to inform risk
management offices about potential claims (Of course, these processes often fail
to work the way they are supposed to, because of inertia or because medical staff
may be reluctant to admit problems have occurred. But I deal here with design.)
First, there is “incident” or “occurrence” reporting. Because of its claims
management origin,101 hospital incident reporting has typically covered injuries
for which hospitals themselves might be liable—those caused by hospital
employees (nurses and custodial staff) rather than by the independent-contractor
doctors.102 Thus, at least by accounts in the 1980s and earlier, the main topics for
incident reporting tended to be patient or visitor slips-and-falls, lost patient
property, and medication errors.103 More recent evidence essentially confirms the
98

ORLIKOFF WITH VANAGUNAS, MALPRACTICE PREVENTION, supra note 97, at 37.
Florence Kavaler & Allen D. Spiegel, Risk Management Dynamics, in RISK MANAGEMENT
IN HEALTH CARE INSTITUTIONS: A STRATEGIC APPROACH 3, 5 (Florence Kavaler & Allen D.
Spiegel eds., 2d ed. 2003).
100
INTERQUAL, CLINICAL RISK MANAGEMENT—A PRACTICAL APPROACH 41-42 (1986).
101
Orley H. Lindgren, Ronald Christensen & Don Harper Mills, Medical Malpractice Risk
Management Early Warning Systems, 54 LAW & CONTEMP. PROBS. 23, 24 (1991).
102
Id. at 24.
103
See OFFICE OF TECHNOLOGY ASSESSMENT, THE QUALITY OF MEDICAL CARE:
INFORMATION FOR CONSUMERS 101 (1988); UNITED STATES GENERAL ACCOUNTING OFFICE,
99

http://www.bepress.com/jtl/vol2/iss1/art1

26

Schlanger: Operationalizing Deterrence

point: A recent discussion based on intense ethnographic observation of surgical
units of a large tertiary care urban teaching hospital explains that incident
reporting “did not adequately capture the range and number of errors,” and indeed
“generally focused on the most mundane errors”—those involving “medications
and complications,” rather than “diagnosis, surgery, and treatment.”104 Lori
Andrews, the lead author of the ethnographic study, attributes incident reporting’s
failure to “provide an effective early warning system about claims” to various
methods by which reporting was discouraged. She mentions, for example, that
“new medical residents were actually told by more senior doctors not to fill out
occurrence reports.”105
The result of low levels of incident reporting is to compromise incident
reporting’s claims management coverage. One report explains:
The incident report has been shown to be an effective tool in
identifying physical and environmental hazards and those
situations of a clinical nature that result in a minor or transient
injury. However, incident reports have not been successful in
identifying those episodes of clinical care that eventually generate
a lawsuit.106
On the other hand, in some institutions, incident reporting is quite effective in
capturing incidents that lead to claims. I spoke with one medical school’s head of
risk management, whose school handles risk management for faculty physicians
but not hospitals or residents. Whether for that or another reason, she reports that
about 85% of claims and lawsuits brought against the physicians involved an
issue previously reported to her office by an incident report or telephone call.107
Whether they are filed nearly all the time or less consistently, hospital incident
reports that are filed serve the same kind of claim defense purpose as Acme’s
incident reporting, by solidifying the file before memories and evidence can fade.
One 1977 discussion that urged safety-related analysis of incident reports
complained about this very emphasis: “Too often in the past, the incident report
form has been used solely to help establish the hospital’s defense in a lawsuit
resulting from a claim.”108
Different hospitals supplement incident reporting with a variety of other
claims-related reports. The Andrews study, for example, examines “potential
claims files,” and finds that these reports “did a better job of capturing problems
INITIATIVES IN HOSPITAL RISK MANAGEMENT (1989); Grace Duran, Positive Use of Incident
Reports, 53 HOSPITALS 60-68 (1979).
104
See Andrews et al., Alternative Strategy, supra note 92, at 310; Andrews, Error in Situ,
supra note 92, at 368-69.
105
Andrews, Error in Situ, supra note 92, at 368-69.
106
INTERQUAL, CLINICAL RISK MANAGEMENT, supra note 100, at 52.
107
Interview with medical school director of risk management (Apr. 2005).
108
Dankmyer & Groves, Taking Steps for Safety’s Sake, supra note 96, at 60, 61.
Published by The Berkeley Electronic Press, 2008

27

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

in diagnosis, surgery, and treatment,” although they “dramatically underreported
the full range of serious errors” observed by the study.109 Other observers discuss
“clinical screening” or “occurrence screening,”110 which use chart-review to
locate adverse events that might result in claims. Similarly, in 1991, for example,
the heads of one risk management operation, the “Professional Risk Management
Group,” described a system in place at nearly all the university teaching hospitals
in California, explaining that it was “designed to capture significant medical
injuries before claimants’ lawyers do so,” by using both written and phoned
reports to initiate claims investigations.111
Non-claims-related alert systems coexist with the ones just described. The
oldest such system is the hospital institution of the mortality and morbidity
conference. M&M conferences are run by and for doctors, to discuss bad
outcomes and their sources. Surgical M&M conferences regularly deal with error
(here they seem to differ from internal medicine M&M conferences112). In some
hospitals, risk managers may attend occasional M&M conferences and start a
potential claims file as a result,113 but the system is really aimed at professional
acculturation and peer-review more than risk or claims management.
A more recent error-review innovation, instituted at the direction of the Joint
Commission (previously known as the Joint Commission on Accreditation of
Healthcare Organizations), whose seal of approval is necessary for a hospital to
remain open, is that every hospital has in place a policy for the reporting,
investigation (“root cause analysis”), and systemic response (“action plan”) to
every “sentinel event.”114 The Joint Commission explains that a sentinel event is
an “unexpected occurrence involving death or serious physical or psychological
injury, or the risk thereof. Serious injury specifically includes loss of limb or
function. The phrase, ‘or the risk thereof’ includes any process variation for
which a recurrence would carry a significant chance of a serious adverse
outcome.” The sentinel event system, which started in 1996, was not initiated as
109

Andrews, Error in Situ, supra note 92, at 369.
See ORLIKOFF WITH VANAGUNAS, MALPRACTICE PREVENTION, supra note 97, at 60-64;
INTERQUAL, CLINICAL RISK MANAGEMENT, supra note 100, at 71-100.
111
Lindgren et al., supra note 101, at 26-27.
112
Edgar Pierluissi et al., Discussion of Medical Errors in Morbidity and Mortality
Conferences, 290 JAMA, 2838 (2003); see also Jay D. Orlander, Thomas W. Barber & B. Graeme
Fincke, The Morbidity and Mortality Conference: The Delicate Nature of Learning from Error, 77
ACAD. MED. 1001 (2002).
113
Cf. Joel C. Rosenfeld, Using the Morbidity and Mortality Conference to Teach and Assess
the ACGME General Competencies, 62 CURRENT SURGERY 664, 664 (2005) (describing risk
management as one of the goals of the M&M conference).
114
JOINT COMMISSION, SENTINEL EVENT POLICY AND PROCEDURES (Updated: July 2007),
available at http://www.jointcommission.org/SentinelEvents/PolicyandProcedures/. Many states
enforce impose a parallel adverse event reporting requirement. See Lucian L. Leape, Reporting of
Adverse Events, 347 NEW ENG. J. MED. 1633 (2002).
110

http://www.bepress.com/jtl/vol2/iss1/art1

28

Schlanger: Operationalizing Deterrence

a claims management system.115 Indeed, critics have charged that it threatens to
increase hospital liability exposure.116 But it may serve, like other types of
reporting, to notify risk management personnel of the possibility of a future claim.
Whatever kinds of reports are used in a given hospital, if a report of a serious
incident arrives at a risk management office, risk management personnel decide
whether or not to open a claims file and conduct a preliminary investigation. An
initial claims file contains medical information and statements by involved care
providers. In at least some hospitals, it is carefully constructed so as not to assist
plaintiffs. One published list of “Do’s and Don’ts” includes the following:
• Don’t include names and addresses of witnesses, even if
the form requests such information. Such data make it
easier for attorneys to sue the institution. Check with the
supervisor on supplying this information. . . .
• Don’t admit liability or blame or identify others as
responsible.
Obviously, this incrimination could be
harmful to the agency if a lawsuit ensues.117
Once a claim file is opened and the preliminary investigation is complete, the
file most typically awaits the filing of a claim or complaint, which may or may
not occur. In recent years, however, a number of large hospital systems have
begun programs to conduct a more complete investigation even if no claims is
filed. In California’s public and teaching hospitals, even before a claim is made,
“immediate investigation and, in selected cases, active intervention [are]
undertaken to assist the patient, reduce liability potential, and extract information
for purposes of preventing future mishaps.”118 And similarly, the Veterans’
Administration has a program by which it investigates, discloses, and apologizes
for adverse events,119 and then attempts restitution by settling any resulting
claims.120 Some states even require disclosure.121 But most typically, the
115

JOINT COMMISSION, SENTINEL EVENTS: EVALUATING CAUSE AND PLANNING
IMPROVEMENT (2d ed. 1998); Rebecca Voelker, “Treat Systems, Not Errors,” Experts Say, 276
JAMA 1537, 1538 (1996) (quoting Dennis O’Leary, President of the Joint Commission for the
Accreditation of Health Care Organizations, on the new sentinel event requirements).
116
Paula Krapf, ISMS, AMA Oppose Sentinel Event Policy, ILL. MED., Jan. 22, 1999, at 1.
117
Kavaler et al., Risk Management Dynamics, supra note 99, at 12.
118
Lindgren et al., supra note 111, at 26-27.
119
Dep’t of Veterans Affairs, Veterans Health Admin., Disclosure of Adverse Events to
Patients,
VHA
Directive
No.
2008-002
(Jan.
18,
2008),
available
at
http://www.ethics.va.gov/ETHICS/docs/policy/VHA_Directive_2008002_Disclosure_of_Adverse_Events_20080118 .pdf.
120
See Steven S. Kraman & Ginny Hamm, Risk Management: Extreme Honesty May Be the
Best Policy, 131 ANNALS INTERNAL MED. 963 (1999); Lee Taft, Apology and Medical Mistake:
Opportunity or Foil?, 14 ANNALS HEALTH L. 55 (2005).
121
See William M. Sage, Joshua Graff Zivin & Nathaniel B. Chase, Bridging the RelationalRegulatory Gap: A Pragmatic Information Policy for Patient Safety and Medical Malpractice, 59
Published by The Berkeley Electronic Press, 2008

29

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

preliminary file, relatively undeveloped, awaits a claim made directly by a patient
or a patient’s family member, or by a lawyer on behalf of a patient. Many
hospital risk managers work hard to settle claims before a lawyer gets involved.
If a lawsuit does ensue, the next step is detailed investigation by hospital
employees. By the time claims files are complete, they are very thick. (For those
inclined to secrecy, there’s less need to worry about helping the plaintiff after the
claim is filed, because the documents generally are likely to constitute privileged
litigation work-product.122) A claims file for a claim that does not settle right
away is likely to contain medical information, statements, depositions, the views
of witness and non-witness expert consultants, and more. Their purpose is to be
fodder for lawyers, who use the file information to settle or try the cases. The file
detail is necessitated by the same two dominating features of these claims already
mentioned—their causal opacity and their seriousness: The majority of injuries to
hospital patients are non-compensable to plaintiffs (because they were either not
iatrogenic at all, or not preventable); and determining whether anyone associated
with the organization violated the standard of care—that is, separating out
compensable from non-compensable injuries—is, even for the hospitals
themselves, extremely difficult. When liability exists, however, the damages can
be very large; a permanently disabled person with ongoing medical and assistive
needs might have a potential recovery of many millions of dollars. The result is
that for hospitals, liability exposure is of an entirely different magnitude than for
retailers, or, indeed, most other types of firms.
A large hospital and its associated physicians (employed, for example, by a
university or other central employer), say one with about 1000 staffed beds, might
have gross revenues of close to a billion dollars, one-seventh the size of Acme’s
operation, and might budget $14 million dollars annually for insurance premiums
and malpractice payouts.123 This is close to 1.5% of total budget. Considering
the entire system, tort liability as a fraction of total hospital revenues is nearly 30
times what it is for Acme.
VAND. L. REV. 1263, 1283-84 (2006) (describing four states with mandatory disclosure of adverse
events to patients, under Fla. Stat. Ann. § 395.1051 (West Supp. 2006); Nev. Rev. Stat. § 439.855
(2006); N.J. Stat. Ann. § 26:2H-12.25 (West Supp. 2006); and 40 Pa. Cons. Stat. Ann. § 1303.308
(West Supp. 2006)).
122
See Fed. R. Civ. P. 26(b)(3); Hickman v. Taylor, 329 U.S. 495 (1947).
123
Interviews with medical school director of risk management; academic hospital director of
risk management. Some sources indicate that this malpractice insurance/payout amount may often
be far higher. One informed observer estimates annual insurance premiums and malpractice
payouts for a typical hospital of 1000 beds at over $30 million. Telephone Interview with Marian
Dwyer, Director, Risk Management Services, RMF Strategies (Feb. 16, 2005). And a recent
profile of Maimonides Medical Center, with 705 patient beds and $626 million in revenues, put its
malpractice insurance bill at $17.7 million. JULIE SALAMON, HOSPITAL: MAN, WOMAN, BIRTH,
DEATH, INFINITY, PLUS RED TAPE, BAD BEHAVIOR, MONEY, GOD, AND DIVERSITY ON STEROIDS 3
(2008) (the figures appear to be for 2003).
http://www.bepress.com/jtl/vol2/iss1/art1

30

Schlanger: Operationalizing Deterrence

It is important to note that the difference between hospitals and the other
organizations described here lies in the liability exposure per lawsuit rather than
the amount of litigation. Hospitals do not see more lawsuits than other kinds of
entities; if anything, they are subjected to fewer claims. The 1000-bed academic
hospital just described opens only 100-150 claims annually, of which only 50-100
are lawsuits.124 This is not an aberrant number. For example, between them,
CRICO’s member hospitals (Massachusetts General Hospital, Brigham and
Women’s Hospital, Beth Israel Deaconess Medical Center, McLean Hospital, and
many others) and their doctors open fewer than 300 claims and lawsuits each
year.125 In both examples, this is fewer than the annual number of claims made
against Acme, notwithstanding Acme’s much larger scope.
These relatively small numbers of claims coupled with the large magnitude of
payouts confirm the point that, on average, claims against hospitals are worth far
more than claims against Acme. Vastly more than Acme, then, what hospitals
and those who deal with risk management for hospital doctors should be after is
not easy response to litigation, but effective—that is, cost-reducing—response.
This is, in fact, what the system just described delivers. The detail of the
investigation and the depth of the resulting file would be far too expensive for
low-damages claims.
But even though effective claims management is extremely important for
hospitals, so too is harm prevention. Over the past two to three decades, the
system of hospital claims management just described has become intimately
associated with harm prevention, as claims management imperatives have
promoted information production and use, the hiring and activity of risk
management staff, the use of bureaucratic strategies that themselves reduce harm,
and the salience of accident-related losses:
Information production. As with Acme’s accident files, hospital incident
reports and claims files are produced in order to facilitate tracking, settlement, and
defense of claims. And as with Acme, claims information then gets harnessed in
the service not only of claims management but of harm reduction. For hospitals,
this occurs in two different ways: Claims files are used by researchers, and both
claims files and incident reports are used within hospitals to assess safety and
quality of care problems at the particular facility, in order to design useful
interventions.

124

Interview with medical school director of risk management; academic hospital director of
risk management.
125
Marian Dwyer, Director of Risk Management Services for RMF Strategies (a subsidiary of
CRICO), thinks this number is low, compared to many other hospital operations. But the point in
text holds even given what she reports is more typical, 350-400 annual claims. Interview with
Marian Dwyer, supra note 123.
Published by The Berkeley Electronic Press, 2008

31

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

The most developed (and familiar) effort to used closed claim research for
clinical insights is sponsored by the American Society of Anesthesiologists. The
project includes data from 1975 on; currently over 7000 observations.126 In each,
reviewing anesthesiologists complete a detailed data form and narrative summary,
based on the claims file.127 The use of these data to reform anesthesiology
practice and routines is the great success story of the American medical
malpractice system.128 The ASA’s closed claims data have been the field in
which anesthesiologists have harvested ideas for systematic clinical
improvement,129 and the results are spectacular. Anesthesia-related deaths have
dropped from their 1970s and 1980s rate of between 1 in 2,000 and 1 in 10,000 to
1 in 200,000.130
Especially in the past ten years, other investigators have similarly, if less
dramatically, used closed claims as a research base for important patient safety
proposals.131 The research methodology is attractive because it allows a “critical
incident” approach, but avoids the enormous expense of generating detailed

126

For current information, see http://www.asaclosedclaims.org.
Frederick W. Cheney, The American Society of Anesthesiologists Closed Claims Project:
What Have We Learned, How Has It Affected Practice, and How Will It Affect Practice in the
Future, 91 ANESTHESIOLOGY 552 (1999).
128
For an account of this history, see Hyman & Silver, Problem or Solution, supra note 77, at
917-23.
129
See ANALYSIS OF ANESTHETIC MISHAPS (Ellison C. Pierce & Jeffrey B. Cooper eds.,
1984).
130
See Hyman & Silver, Problem or Solution, supra note 77, at 918, citing Alexander
Goldstein & Arthur S. Keats, The Risk of Anesthesia, 33 ANESTHESIOLOGY 130, 133 tbl.1 (1970),
and Lucian L. Leape, Error in Medicine, in MARGIN OF ERROR: THE ETHICS OF MISTAKES IN THE
PRACTICE OF MEDICINE 95, 107 (Susan B. Rubin & Laurie Zoloth eds., 2000); Ellison C. Pierce,
Jr., Anesthesia: Standards of Care and Liability, 262 JAMA 773, 773 (1989).
131
James E. Ludlam, Using Malpractice Claims Data to Successfully Attack the Problem of
Infant Brain Damage, in LEGAL MEDICINE 227 (1991); Lorraine M. Jordan, Michael Kremer,
Karen Crawforth & Susan Shott, Data-driven Practice Improvement: The AANA Foundation
Closed Malpractice Claims Study, 69 AANA J. 301 (2001); David M. Benjamin & Robert F.
Pendrak, Medication Errors: An Analysis Comparing PHICO’s Closed Claims Data and PHICO’s
Event Reporting Trending System (PERTS), 43 J. CLIN. PHARMACOL. 754 (2003); Barry M.
Manuel & Linda M. Greenwald, Using Medical Malpractice Closed Claims Data To Reduce
Surgical Risk and Improve Patient Safety, BULL. OF AM. C. SURG., March 2007, at 27; F. Dean
Griffen et al., The American College of Surgeons’ Closed Claims Study: New Insights for
Improving Care, 204 J. AM. COLL. SURG. 561 (2007); Harry A. Quigley, David S. Friedman, &
Steven R. Hahn, Evaluation of Practice Patterns for the Care of Open-Angle Glaucoma
Compared with Claims Data: The Glaucoma Adherence and Persistency Study, 114
OPHTHALMOLOGY 1599 (2007); Paul D. Hain et al., Using Risk Management Files to Identify and
Address Causative Factors Associated with Adverse Events in Pediatrics, 3 THERAPEUTICS &
CLIN. RISK MGMT. 625 (2007).
127

http://www.bepress.com/jtl/vol2/iss1/art1

32

Schlanger: Operationalizing Deterrence

information about past events,132 and because it tends to turn up iatrogenic
injuries. (Of course, the methodology is non-random, which raises its own set of
statistical issues, but generally researchers have found the advantages outweigh
these concerns.133)
The types of closed claim studies just described are not facility-specific, and it
may therefore be unusual to consider them in this study of “deterrence,” usually
thought to be a matter of firm or individual behavioral change. But I think it
would, in fact, be inappropriate to omit them. The object of this article’s inquiry
is how damage actions induce claims management, and how claims management
activities, in turn, induce harm reduction. Just as insurers’ safety prescriptions
belong in a study of how deterrence is operationalized, so too do care
improvements that occur by way of broad research, when that research is made
possible because of the existence and depth of claim files.
Moving from broad systemic research to inquiry focused on one facility,
closed claims have also emerged as a vital data source for efforts to improve
patient safety and quality of care, one hospital at a time. RMF Strategies’ director
of risk management explained to me how the company structures the consulting
services it provides to hospitals. When her staff works with a facility, she said,
closed claims are crucial. Indeed, the first thing they do is to ask for data on all
medical malpractice claims: the date of the loss, the date the claim was brought,
when it was closed, what the hospital “reserved” for it (that is, how much money
a self-insured facility set aside to pay it), what the payout was, information on the
injury including severity, nature, and so on, and finally information on the nature
of the alleged error, the relevant service (e.g., obstetrics, anesthesiology, nursing).
From this, supplemented by interviews, RMF can draw the broad outlines of
strengths and weaknesses in the hospital. Next, they benchmark the claim results
against other facilities in the RMF database. Finally, they pick the problem areas
and sample them, in depth, using the closed case files. She explained to me that
“There’s nothing better than a litigation file, because it has so much investigative
information.”134 Published sources, as well, emphasize and support the similar
use of the documentation in closed claim files as guides to loss prevention.135
132

Two important studies that did undertake this kind of expense are Jeffrey B. Cooper,
Ronald S. Newbower, Charlene D. Long & Bucknam McPeek, Preventable Anesthesia Mishaps:
A Study of Human Factors, 49 ANESTHESIOLOGY 399 (1978), a study that was important in
prompting the American Society of Anesthesiologists to undertake its Closed Claims Project, and
the Harvard Medical Practice Study, supra note 91.
133
See, e.g., To See or Not to See, 105 ANESTHESIOLOGY 1071 (2006) (editorial view); cf.
Frederick K. Orkin, Statistical Indiscretions in Papers from the American Society of
Anesthesiologists Closed Claims Project, 104 ANESTHESIOLOGY 615 (2006) (correspondence).
134
Interview with Marian Dwyer, supra note 123.
135
See, e.g., Richard L. Kravitz, John E. Rolph & Kimberly A. McGuigan, Malpractice
Claims Data As a Quality Improvement Tool I: Epidemiology of Error in Four Specialties, 266
Published by The Berkeley Electronic Press, 2008

33

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

Other hospitals use closed claim files in other ways. For example, some
hospitals use them to conduct credentialing reviews, in order to avoid awarding
practice privileges to incompetent doctors.136 One medical school’s risk
management office, which deals with over 1000 physicians at several dozen inand out-patient facilities, works in conjunction with department chairs to analyze
closed claims; the chairs can then do followup training on grand rounds, if it
seems useful.137 (Physician risk management may, however, be less invested than
hospital risk management in the systems-focused patient safety approach.) The
point is a simple one: In many hospitals, the information gathered so
painstakingly and at such great expense in order to manage and defend claims is
used, as well, for risk minimization.
Not all risk management and patient safety experts emphasize closed case
files quite so heavily. In some hospitals, patient safety improvement efforts rely
more on incident reports. Eric Larson, the Medical Director of the University of
Washington’s Medical Center, described his hospital’s use of incident reports in
2002 as “critical to risk reduction and risk management”: “[B]y building on a
continuous improvement approach and a system approach to error prevention, an
effective incident-reporting system can have great value as a mechanism to
identify risk for the purpose of reducing risk and improving safety for patients.138
The theoretical point holds, however; Larson, like those who use claims files,
repurposes for harm-reduction documents generated for claims management
purposes.139
JAMA 2087 (1991); John E. Rolph, Richard L. Kravitz & Kimberly A. McGuigan, Malpractice
Claims Data As a Quality Improvement Tool, II: Is Targeting Effective? 266 JAMA 2093 (1991);
Paul R. Frisch, Sara C. Charles, Robert D. Gibbons & Donald Hedeker, Role of Previous Claims
and Specialty on the Effectiveness of Risk-Management Education for Office-Based Physicians,
163 WESTERN J. MED. 346, 346 (1995) (“Risk-management education courses are increasingly
based on a review of closed claims.”); Robert Stanyon, Total Quality Management, Continuous
Quality/Process Improvement, and Evaluation of the Risk Management Program, in RISK
MANAGEMENT IN HEALTH CARE INSTITUTIONS: A STRATEGIC APPROACH, supra note 99, at 133,
152-54.
136
See Kathleen M. Shostek & Robert F. Pendrak, Use of Malpractice Data in Medical Staff
Credentialing, J. HEALTHCARE RISK MGMT., Summer 1996, at 16.
137
Interview with medical school director of risk management.
138
Eric B. Larson, Measuring, Monitoring, and Reducing Medical Harm from a Systems
Perspective: A Medical Director’s Personal Reflections, 77 ACAD. MED. 993, 996 (2002)
(emphasis in original).
139
Other risk managers or patient safety reformers prefer to rely on sentinel event
documentation. The Director of Risk Management at the Brigham and Women’s Hospital, in
Boston, told me that a team that already does root cause analysis of sentinel events has much less
to learn from litigation files. They are detailed, she explained, and full of good information—but
they are too old to be much use. But she conceded that in the absence of non-litigation detailed
investigation, litigation files could be extremely helpful for accident avoidance. And she, too,
uses aggregate closed claims data to track problems that needed solution. Telephone Interview
http://www.bepress.com/jtl/vol2/iss1/art1

34

Schlanger: Operationalizing Deterrence

Personnel. As already described, hospital risk managers were brought into
hospitals’ organizational structures in order to minimize the cost and maximize
the effectiveness of claims processing. Over the past twenty years, however,
hospital risk managers have shifted their focus in significant part, though far from
entirely, to ex ante patient safety and harm prevention. The transition from claims
management to risk management has, it seems, been extremely professionally
attractive for risk managers, perhaps because within health care institutions,
patient care takes pride of place. The result of risk managers’ growing
professional orientation towards patient safety is an increasing likelihood that they
will find time to concern themselves with harm prevention, instead and in
addition to claims management; risk managers accordingly have become likely to
use the tools of their claims management tasks for their newer harm prevention
goal.
Hospital risk managers’ professional focus on claims had substantial staying
power, but the shift was urged (if not taken) as early as the 1970s in professional
journal articles and other professional sources. For example, in an article entitled
“Taking Steps for Safety’s Sake,” in the Journal of the American Hospital
Association, the authors explained: “Risk management is not solely reactive, as
some mistakenly believe. The potential for incidents needs to be dealt with if the
hospital’s risk management program is to stand any chance of success. Claims
handling, after the incident has taken place, is only a small part of the entire risk
management process.”140 The article urged risk managers to conduct systematic
review and analysis of incident reports. (Another article in the same issue of the
same journal noted that any such effort would face the inertia of physicians, who
were, the article implied, somewhat uninterested in a risk-management approach
to patient safety.141)
Hospital risk managers’ professional association is the American Society for
Healthcare Risk Management (ASHRM), founded in 1980. Every year since,
ASHRM has pushed for the expansion and acknowledgement of its members’
contribution to patient safety, not just to hospitals’ bottom line. A retrospective
description of the organization’s first 25 years, for example, labeled 1984 to 1988
“[t]he ‘quality’ era.”142 ASHRM’s journal documents the continuing effort by
with Janet Barnes, Director of Risk Management, Brigham & Women’s Hospital (Boston, MA)
(Feb. 24, 2005).
140
Dankmyer & Groves, Taking Steps for Safety’s Sake, supra note 108, at 60, 61.
141
See J. AM. HOSP. ASS’N, May 16, 1977, at 63, 66 (“the big claims are medically related
and, with few exceptions, far too little has been done because of an inability to coordinate the
physician component with the hospital component. . . . Little interest in taking action, however,
has been evinced by the American Medical Association to date”).
142
ASHRM, CELEBRATING 25 YEARS: A BRIEF HISTORY OF THE AMERICAN SOCIETY FOR
HEALTHCARE
RISK MANAGEMENT,
1980-2006,
at
10
(2006),
available
at
http://www.ashrm.org/ashrm/aboutus/pdf/ASHRM25Book.pdf.
Published by The Berkeley Electronic Press, 2008

35

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

professional leaders to prioritize health care quality. For example, a 1995 article
stated: “Early warning systems should be designed to identify adverse events and
poor quality of care, not potential lawsuits. The goal of the risk manager, which
is sometimes forgotten in day-to-day activities, is to improve the care of patients.
If that can be achieved, litigation will be reduced to minor and frivolous
claims.”143
The transition in the profession’s leaders’ conceptualization of hospital risk
management took many years, but it appears at this point to be complete. Where
insurance or legal training were once the prerequisite for risk management jobs,144
nursing has became the more typical professional background.145 This change is
reflected in ASHRM’s presidents’ backgrounds, as well.146 The organizational
results are marked. In 2002, seizing the opportunity presented by the growing
patient safety movement,147 ASHRM pronounced its “vision” to be “safe and
trusted health care,” and promulgated a new version of its mission statement
focusing on care, rather than finance: “[T]o advance safe and trusted patientcentered health care delivery, ASHRM promotes proactive and innovative
management of organization-wide risk.”148 In 2004, it published a report on a
“new health care role,” the patient safety officer, urging its members to take every
“opportunity to highlight their current contributions to patient safety, develop
additional skills and expand their profile.”149 After highlighting the ways in
which risk managers were suited to become patient safety officers, because they
already “possess the breadth of experience in identifying, managing and reducing
risk,” the publication warned that the opportunity of professional expansion itself
came with a risk of professional eclipse: “Whether the health care industry turns
to risk managers to fulfill that role [of patient safety officer] . . . depends in part

143

William C. Quinlan, How do Risk Managers Really Learn About Potential Claims, J.
HEALTHCARE RISK MGMT., Spring 1995, at 3.
144
Id.
145
See advertisements, ASHRM Forum, available from 2001 to present at
http://www.ashrm.org/ashrm/news/publications2.html.
146
ASHRM, CELEBRATING 25 YEARS, supra note 142, at 30-43.
147
For foundational documents in the patient safety movement, see Lucian L. Leape, Error in
Medicine, 272 JAMA 1851 (1994); INSTITUTE OF MEDICINE, TO ERR IS HUMAN, supra note 90;
INSTITUTE OF MEDICINE, CROSSING THE QUALITY CHASM: A NEW HEALTH SYSTEM FOR THE 21ST
CENTURY (2001). For background, see Agency for Healthcare Research and Quality, Web M&M,
In Conversation with… Lucian Leape, MD, http://www.webmm.ahrq.gov/perspective.aspx?
perspectiveID=28. See also ASHRM, CELEBRATING 25 YEARS, supra note 142, at 23
(“Meanwhile [in 2001], patient safety arrived as a distinct discipline.”).
148
See ASHRM, CELEBRATING 25 YEARS, supra note 142, at 25.
149
ASHRM, THE GROWING ROLE OF THE PATIENT SAFETY OFFICER: IMPLICATIONS FOR RISK
MANAGERS, June 2004, at 3, available at http://www.ashrm.org/ashrm/resources/files/
Monograph.PSO.pdf.
http://www.bepress.com/jtl/vol2/iss1/art1

36

Schlanger: Operationalizing Deterrence

on risk managers’ ability to leverage their existing skill sets and enhance them
with the skills required to lead a patient safety program.”150
Thus at the current moment in the evolution of the profession of hospital risk
management, safety is not justified serving the monetary goal of liability
reduction but rather the reverse: Risk management’s focus on money is itself
likely to be justified with reference to the resulting impact on safety. For
example, another ASHRM publication explains: “[A]n organization’s ability to
deliver safe, quality care depends in large part upon its financial strength. Every
dollar spent on professional liability losses is a dollar that might be spent on
patient care.”151
In summary, it has been a major effort of many hospital risk managers to try
to move the core of risk management away from claims management and towards
patient safety improvement. This is a professional project still in progress, but it
is evidently high on the agenda of the kinds of risk managers who run
professional societies, put on conferences, and write academic and quasiacademic articles. The result is that members of the profession act to use for
patient safety the materials and techniques developed for claims management
efficiencies.
Bureaucratization. Hospital claims management imperatives—the desire to
respond efficiently and effectively to claims, ex post—have a notable
bureaucratizing effect. A major organizational strategy chosen for its claims
management usefulness is documentation, which can prevent harm as well as
assist in rebutting claims.
It has long been observed that the existence of malpractice lawsuits sometimes
induces doctors and nurses to spend more time and effort on documentation more
readily (and cheaply) introduced in court—for example, devoting more time to
charting observations than they otherwise might. As one guide to health
institution risk management states, “The absence of . . . documentation makes
defending a medical malpractice action against the hospital difficult. In the words
of one plaintiffs’ attorney: ‘If it’s not in the chart, it never happened.’”152 Another
publication advises doctors, “if litigation occurs,” “[i]ncomplete, illegible and
inconsistent drug documentation . . . suggest[s] careless, negligent prescribing
practices when none occurred.”153 Similarly, an article on risk management in
150

Id.
ASHRM, DIFFERENT ROLES, SAME GOAL: RISK AND QUALITY MANAGEMENT PARTNERING
FOR PATIENT SAFETY, June 2007, available at http://www.ashrm.org/ashrm/resources/files/
Monograph.07RiskQuality.pdf; see also Kavaler & Spiegel, Introduction, in RISK MANAGEMENT
IN HEALTH CARE INSTITUTIONS, supra note 99, at xxvi (“[M]onetary awards from litigation
drastically reduce the resources available to render high-quality health care to those in need.”).
152
Kavaler et al., Risk Management Dynamics, supra note 99, at 25.
153
Kathi Burton, Prescribing Medications and Malpractice Suits: Clear, Consistent
Documentation is the Key, 94 W.VA. MED. J. 272, 272 (1998).
151

Published by The Berkeley Electronic Press, 2008

37

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

obstetrics counsels “since legal action may be initiated as long as 21 years
following the delivery in the case of an injured newborn, it is important that
medical record documentation be accurate, objective, and complete, and provide
the rationale to support all patient management decisions, including the decision
not to intervene.”154 In addition, the potential for litigation (and its concomitant
necessity of proving up adequate care) may sometimes lead medical providers to
prefer the types of care that produce a written record—giving an EKG instead of a
particularly thorough chest exam,155 for example.
There are those who call choices like these, usually disparagingly, “defensive
medicine”156 But as one recent study of the reported magnitude of defensive
medicine explained, actually these could well be very good treatment decisions.
Its authors noted: “Defensive medicine may reduce or improve quality,
depending on the circumstances.”157 EKGs are valuable diagnostic tools, for
example, and ease consultation where it is useful. The article on medication
documentation quoted above also notes that the types of errors sometimes
observed—ambiguous or incomplete notes, “[i]llegible handwriting and carelessly
written decimal points and numbers,” and the like, “can cause . . . errors and
injury” and “compromise quality patient care.”158 And, more generally, good
(though not excessive) charting is essential for high quality healthcare.159 The
reason for this last observation is particularly pertinent. Good medical charting
requires the charter to engage in certain inquiries that are necessary for good
154

Kavaler et al., Risk Management Dynamics, supra note 99, at 25.
See Jerry R. Green, Medical Malpractice and the Propensity to Litigate, in THE
ECONOMICS OF MEDICAL MALPRACTICE 193, 197 (Simon Rottenberg ed., 1978) (noting a
tendency among health care personnel to prioritize “[t]he attributes of care readily observable in
court”—“typically . . . those procedures for which records are kept: tests ordered, X-rays,
frequency of reexamination, length of hospital stay”); David M. Studdert et al., Defensive
Medicine Among High-Risk Specialist Physicians in a Volatile Malpractice Environment, 293
JAMA 2609, 2612 (2005) (identifying extra imaging studies among the most common “defensive
acts” by physicians).
156
See Green, Medical Malpractice, supra note 155, at 197 (“Focusing on these aspects of
care when others that would be more efficient are bypassed is defensive medicine.”). Oddly,
given the frequent repetition of the idea that malpractice liability risk encourages documentation,
such documentation is not usually included in less abstract definitions of defensive medicine. See,
e.g., Studdert et al., Defensive Medicine, supra note 155, at 2609. This fact does not alter the
point in text.
157
Studdert et al., Defensive Medicine, supra note 155, at 2616.
158
Burton, Prescribing Medications, supra note 153, at 272.
159
See, e.g., Douglas L. Wood, Documentation Guidelines: Evolution, Future Direction, and
Compliance, 110 AM. J. MED. 332 (2001). The point generalizes to non-hospital settings as well.
See, e.g., Susan Keane Baker, Minimizing Litigation Risk: Documentation Strategies in the
Occupational Health Setting, 48 AAOHN J. 100, 100 (2000) (“Effective documentation strategies
have the dual benefit of improving care and decreasing the risk of malpractice litigation in the
occupational health setting.”).
155

http://www.bepress.com/jtl/vol2/iss1/art1

38

Schlanger: Operationalizing Deterrence

medical care—charts often follow the SOAP method, for example (subjective,
objective, assessment, and plan),160 which reminds providers to think about and
investigate each of these topics. And good medical charting is vital, too, for
continuity of care.161 Here, as in so many areas, written recordkeeping can
structure conduct in a very productive way. So practices engaged in as claims
management practices—in this case, generation of contemporaneous written
patient and treatement records—turn out to have a harm-reduction payoff.
Using a still broader perspective, one would expect that the kinds of evidence
that potential defendants believe lend themselves to error-free adjudication in the
event of litigation-provoking accidents would, for exactly the same reasons, lend
themselves to other quality control systems. It is for precisely this reason that
written records are a standard bureaucratizing strategy. Development of the
systems perspective that currently seems so promising for patient safety depends
on bringing into hospitals quality assurance techniques that have worked in other
arenas. These include document creation and auditing.162 Records created
because of they make ex post response to litigation more efficient and effective
can also make systems analysis and improvement more efficient and effective.
Salience. Finally, dealing with actual claims and the claims management
apparatus—taking part in incident reviews, depositions, and the like—may
reinforce doctors’ and others’ awareness that accidents do happen, and their desire
to avoid litigation, countering optimism bias and bolstering deterrence. Note,
however, that only a fraction of the limited empirical evidence available supports
this point: Some studies have found that doctors who have been sued report
changed practices due to the malpractice environment at a higher rate than doctors
who have not been sued; more such studies find no significant difference.163 As
160

Lawrence L. Weed, Medical Records that Guide and Teach, 278 NEW ENG. J. MED. 593 &
652 (1968); KATHLEEN WATERS & GRETCHEN A. MURPHY, MEDICAL RECORDS IN HEALTH
INFORMATION 74 (1979).
161
See, e.g., Alice Epstein & Gary H. Harding, Risk Management in Selected High-Risk
Hospital Departments, in RISK MANAGEMENT IN HEALTH CARE INSTITUTIONS, supra note 99, at
325, 346 (“Physicians and risk managers alike have found [diligent charting necessary] to ensure
the best continuum of care.”).
162
See Donald M. Berwick, Continuous Improvement as an Ideal in Health Care, 320 NEW
ENG. J. MED. 53 (1989); AL ENDRES, IMPLEMENTING JURAN’S ROAD MAP FOR QUALITY
LEADERSHIP: BENCHMARKS AND RESULTS (2000); Larson, Measuring, Monitoring, and Reducing
Medical Harm, supra note 138.
163
Compare, e.g., Carol S. Weisman, Laura L. Morlock, Martha Ann Teitelbaum, Ann C.
Klassen & David D. Celentano, Practice Changes in Response to the Malpractice Litigation
Climate: Results of a Maryland Physician Survey, 27 MED. CARE 16, 21-23 (1989) (finding
differences in reported response of “providing more services”—increased tests or monitoring,
consultation, preventive services), with David M. Studdert et al., Defensive Medicine, supra note
155, at 2615 (finding that litigation history had no significant effect on reported tendency to
practice defensive medicine); Peter A. Glassman, John E. Rolph, Laura P. Petersen, Melissa A.
Published by The Berkeley Electronic Press, 2008

39

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

one study suggested, “the signal to practice defensively may have been broadcast
so widely that individual experience is overshadowed by collective anxiety.”164
C. Prisoner litigation
For this article’s third and final claims management case study, I look at prisons
and jails.165 The key feature of the claims docket in prisons and jails is the
combination of high volume and low plaintiffs’ success rate. More types of
injuries are federally actionable for prisoners than for people whose relationships
with the state are less all-embracing.166 And, in any area of law in which
prisoners retain legal rights similar to those of nonprisoners, those rights tend to
run not against many different persons, firms, or agencies, but against one
litigating opponent—the prison or jail that holds them. As the Supreme Court has
noted:
For state prisoners, eating, sleeping, dressing, washing, working, and
playing are all done under the watchful eye of the State, and so the
possibilities for litigation under the Fourteenth Amendment are boundless.
What for a private citizen would be a dispute with his landlord, with his
employer, with his tailor, with his neighbor, or with his banker becomes,
for the prisoner, a dispute with the State.167
At prisoners’ civil rights litigation’s peak in the federal court system, in 1995,
nearly 1.6 million jail and prison inmates168 filed about 40,000 lawsuits in federal
Bradley & Richard L. Kravitz, Physicians’ Personal Malpractice Experiences Are Not Related to
Defensive Clinical Practices, 21 J. HEALTH POL. POL’Y & L. 219 (1996).
164
Glassman et al., Physicians’ Personal Malpractice Experiences, supra note 163, at 235.
165
By prison, I mean a state or federally operated facility that houses convicted felons; by jail,
I mean a county or city (or very occasionally federally) operated facility that houses some
combination of pretrial detainees, felony convicts awaiting sentencing or transfer to prison, and
misdemeanant and felony convicts serving relatively short terms. For a fuller discussion of the
operational and litigation differences between these types of facilities, see Schlanger, Inmate
Litigation, supra note 4, at 1579 n.76 & 1686-89; Anne Morrison Piehl & Margo Schlanger,
Determinants of Civil Rights Filings in Federal District Court by Jail and Prison Inmates, 1 J.
EMPIRICAL LEG. STUD. 79 (2004).
166
Compare Estelle v. Gamble, 429 U.S. 97, 104 (1976) (holding that “deliberate
indifference to serious medical needs of prisoners” violates the Eighth Amendment), and
Youngberg v. Romeo, 457 U.S. 307, 324 (1982) (holding that a mentally retarded person
involuntarily committed to a government institution has “constitutionally protected interests in
conditions of reasonable care and safety”), with DeShaney v. Winnebago County Dep’t of Soc.
Servs., 489 U.S. 189, 201 (1989) (distinguishing Estelle and Youngberg, and holding that,
ordinarily, state and local governments have no constitutional obligation to protect citizens from
harm by private actors).
167
Preiser v. Rodriguez, 411 U.S. 411, 492 (1973).
168
BUREAU OF JUSTICE STATISTICS, CORRECTIONAL POPULATIONS 2006, available at
http://www.ojp.usdoj.gov/bjs/glance/tables/corr2tab.htm.
http://www.bepress.com/jtl/vol2/iss1/art1

40

Schlanger: Operationalizing Deterrence

court169; even in 2006, after a huge decline caused by prisoner-specific federal tort
reform enacted in 1996,170 over 2.2 million prisoners171 filed over 20,000 federal
cases.172 As Table 1 shows, in fiscal year 1998, the last year for which I have the
necessary data, prisoners confined by the Federal Bureau of Prisons (then the
third-largest of the nation’s prison systems, housing 120,000 prisoners173), filed
nearly 4000 administrative tort claims, and over 400 tort and civil rights
actions.174 State and local prisoners add to their federal lawsuits many more cases
in state court. In 1999 in California—then the largest prison system, with over
160,000 inmates175—the Department of Corrections reported that 585 new cases
were filed176 (the Department’s entire active docket was over 2000 cases, about
one-fifth of them federal).177 This litigation rate is far far higher than for the other
organizations described in this article.
But although prisoners’ lawsuits are frequent, payouts are far less so. Even
under the loosest definition of plaintiff success, prisoner plaintiffs are rarely
successful. As I have previously demonstrated, prior to 1996:

169

See ADMINISTRATIVE OFFICE OF THE U.S. COURTS, JUDICIAL BUSINESS OF THE UNITED
STATES COURTS 1997, at 132 tbl. C-2a, Prisoner Petitions: Civil Rights, and Prisoner Petitions:
Prison Conditions, available at http://www.uscourts.gov/judicial_business/c2asep97.pdf (note that
these are cases filed, remanded, or transferred; filings alone are slightly smaller in number).
170
Prison Litigation Reform Act, Pub. L. No. 104-134, §§ 801-810, 110 Stat. 1321, 1321–66
to –77 (1996) (codified at 11 U.S.C. § 523; 18 U.S.C. §§ 3624, 3626; 28 U.S.C. §§ 1346, 1915,
1915A; and 42 U.S.C. §§ 1997-1997h).
171
BUREAU OF JUSTICE STATISTICS, CORRECTIONAL POPULATIONS, supra note 168.
172
See ADMINISTRATIVE OFFICE OF THE U.S. COURTS, JUDICIAL BUSINESS OF THE UNITED
STATES COURTS 2007, at 149 tbl. C-2a, Prisoner Petitions: Civil Rights, and Prisoner Petitions:
Prison Conditions, available at http://www.uscourts.gov/judbus2007/appendices/C02ASep07.pdf
(note that these are cases filed, remanded, or transferred; filings alone are slightly smaller in
number).
173
ALLEN J. BECK & CHRISTOPHER J. MUMOLA, BUREAU OF JUSTICE STATISTICS, PRISONERS
IN 1998, Aug. 1999, NCJ 175687, available at http://www.ojp.usdoj.gov/bjs/pub/pdf/p98.pdf, at 5
tbl. 5.
174
Beth Mellen Harrison, Legal Claims Initiated by Federal Prisoners, Fiscal Years 19922001 (May 2003) (unpublished manuscript, on file with author), at 62 tbl. 4.4, 69 tbl. 4.7. This
paper is based on data provided to me by the Bureau of Prisons in response to a Freedom of
Information Act request.
175
ALLEN J. BECK, BUREAU OF JUSTICE STATISTICS, PRISONERS IN 1999, Aug. 2000, NCJ
183476, available at http://www.ojp.usdoj.gov/bjs/pub/pdf/p99.pdf at 3 tbl. 5.
176
Letter from C.A. Terhune, Director, California Department of Corrections, to Sen. Richard
Polanco, Chair, California Joint Legislative Committee on Prison Construction and Operations
(August 7, 2000) (on file with author).
177
Inmate Litigation Log, spreadsheet accompanying letter from C.A. Terhune, Director,
California Department of Corrections, to Sen. Richard G. Polanco, Chair, California Joint
Legislative Committee on Prison Construction and Operations (June 30, 2000) (on file with
author).
Published by The Berkeley Electronic Press, 2008

41

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

[I]nmates typically won some relief in about one percent of their
federal civil rights cases; they received something worth settling
for in another six to seven percent; and they either simply gave up
and decided to quit, or received something justifying the
withdrawal of the lawsuit, in another six to eight percent of
cases.178
These low success rates make prisoners by far the least successful group of
federal plaintiffs,179 and subsequent data (though still incomplete) show even
lower success rates in more recent years.180 Not only are payouts few and far
between, the damages paid tend to be extremely low; the median in 1993 federal
trial verdicts for prisoner plaintiffs was just $1000.181 (Note that the expense of
complying with injunctive orders can, by contrast, be extraordinarily high. But
injunctive litigation is beyond the scope of this article.182)
When prisoners’ very high claiming rate is balanced with their low success
rate, the result is that litigated or settled damages are, in prisons and jails (as in
retailing but not hospitals), a tiny piece of the very large correctional budgets.
Between 1992 and 1998, for example, the Federal Bureau of Prisons (total
average budget over those years, $2.6 billion183) paid out an annual average of
less than $150,000 in administrative settlements, and only a little more than $2
million, on average, in litigation settlements and judgments.184 Litigation payouts
formed a similarly small proportion of California’s prison outlay; in 1999, the
state (total prison budget, $4.6 billion) paid out just $6.2 million in settlements
and judgments, a sum that includes both payments to prisoners and attorneys fees
for successful plaintiffs.185 In both these two very large correctional jurisdictions,
then, damages litigation payouts are on the order of one-tenth of one percent of
total budget, an order of magnitude less than for hospitals.
Table 1’s comparison of Acme’s sales revenues to a large prison system’s
budget demonstrates that Acme is an enterprise similar in size to the federal
178

Schlanger, Inmate Litigation, supra note 4, at 1597.
Id. at 1598.
180
Id. at 1658-64, sets out figures up through fiscal year 2002. The most recent data, on file
with the author, confirm the same trends through 2004.
181
Id. at 1602-03; see id. at 1622-23 for a discussion of the causes of these low damages.
182
For discussion of prisons and jails and injunctive civil rights cases, see Margo Schlanger,
Civil Rights Injunctions Over Time: A Case Study of Jail and Prison Court Orders, 81 N.Y.U. L.
REV. 550 (2006).
183
Data derived from U.S. Department of Justice, Budget Staff, Justice Management
Division, Budget Trend Data: 1975 Through the President’s 2003 Request to the Congress: Total
Federal
Prison
System
(Spring
2002),
available
at
http://www.usdoj.gov/jmd/budgetsummary/btd/1975_2002/2002/html/page109-112.htm.
184
Beth Mellen Harrison, supra note 174, at 67 tbl. 4.6, 78.
185
Letter from C.A. Terhune to Richard Polanco, supra note 176. California also paid out
another $2.3 million in special master fees in some large injunctive cases. Id.
179

http://www.bepress.com/jtl/vol2/iss1/art1

42

Schlanger: Operationalizing Deterrence

Bureau of Prisons or the California Department of Corrections. Moreover,
Acme’s payouts are not too different from those by correctional agencies:
Acme’s 2005 experience of $3 million in total payouts for torts is roughly
comparable to California’s 1999 experience of $6 million in total payouts to
prisoners and their counsel, or to the Bureau of Prisons’ average payout of $2+
million. The crucial difference is that Acme’s payouts occur on vastly fewer
disputes, and on much less formal litigation.
In addition to having a larger litigation docket, jails and prisons have a more
formal non-litigation complaint system than hospitals or retailers. In jails and
prisons, litigation coexists with a formal administrative grievance system;
prisoners’ use of available grievance procedures is currently mandatory if they
wish to litigate a claim.186 Grievance systems can undertake any degree of review
of prisoners’ complaints, from basically none to a full-fledged investigation. As
Dora Schriro, currently head of corrections in Arizona, has written,
Prisons routinely rely on formal problem solving processes notably
the inmate grievance procedure and lawsuits to address prisoner
complaints. Grievance procedures and lawsuits are not the first
steps citizens ordinarily take. Moreover, cumbersome and time
consuming, they don’t work very well. All too often issues are
dismissed for technical reasons, not on their merits and rarely do
inmates prevail. Far less frequently the underlying conditions
causing or contributing to the problems are addressed.187
But sometimes, grievances are investigated quite comprehensively and
institutional changes made. A New York State amicus brief before the U.S.
Supreme Court recently characterized grievance systems as crucial for selfregulation:
For decades, prison grievance procedures have played an important
role in prison administration. Inmate grievances provide timely
feedback to state officials about problems that arise in correctional
186

See Prison Litigation Reform Act, 42 U.S.C. 1997e. For the case in which the Supreme
Court established this interpretation of the PLRA, see Woodford v. Ngo, 548 U.S. 81 (2006). For
copies of the grievance policies in nearly every state, see Prison and Jail Grievance Policies,
available at http://law.wustl.edu/Faculty/index.asp?id=6430. See also Brief for the Jerome N.
Frank Legal Services Organization of the Yale Law School Supporting Respondent, Woodford v.
Ngo, 548 U.S. 81 (2006) (No. 05-416), available at http://www.law.yale.edu/documents/pdf/
Woodford_Amicus_brief.pdf.
187
Dora Schriro, Correcting Corrections: The Arizona Plan: Creating Conditions for Positive
Change in Corrections, statement given before the Commission on Safety and Abuse in America’s
Prisons at 17 (Feb. 9, 2006), available at http://www.prisoncommission.org/statements/
schriro_dora.pdf; see also Dora Schriro, Correcting Corrections: Missouri’s Parallel Universe,
SENTENCING & CORRECTIONS: ISSUES FOR THE 21ST CENTURY, May 2000, at 1, 6, available at
http://www.ncjrs.gov/pdffiles1/nij/181414.pdf (“Problem solving in prison is often adversarial.
Riddled with formalities, inmate grievance mechanisms short-circuit many complaints.”).
Published by The Berkeley Electronic Press, 2008

43

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

facilities. In individual cases, grievance procedures enable prison
administrators to take prompt remedial action that may satisfy the
inmate and obviate the need for litigation. From a systemic
perspective, such procedures allow prison officials to monitor
trends in prisoner complaints before unwise institutional policies or
patterns of inappropriate conduct by correctional officers lead to
frustration among the inmate population, potentially triggering
prisoner unrest or disturbances. Grievance procedures also allow
state officials to create factual records of any disputes that may
eventually be litigated in court, and to filter out meritless inmate
claims, thereby reducing the States’ costs of defending against
frivolous prisoner litigation.188
In their litigation environment of many cases but few awards, jails and
prisons’ most pressing need in claims management is simply to respond to what is
for some agencies a barrage of litigation. Once they are sued, jail organizations
and prison departments of corrections need to be able to file an answer; they need
to be able to produce, in court, their version of events, including the relevant
policies, documents, and witnesses. In addition, they need to be able to provide
more than inevitably self-interested and therefore discountable testimony by
individual officers about their own conduct. These priorities drive the
connections of claims management to harm reduction or constitutional
compliance behind bars, which occur chiefly by way of bureaucratization,
personnel, and the increased salience of litigation:
Information Production. In the other case studies, I have discussed at some
length the ways in which the information produced for claims processing can be
used, as well, for harm prevention. This is not unheard of in jails and prisons.
Sometimes litigation exposes a problem previously hidden to senior officials, and
they then act to ameliorate it. For example, it was jail strip-search litigation that
brought to senior officials attention the (plainly unconstitutional) policy in the
Miami jail of performing visual body cavity examinations of all women arrested,
no matter the charge or circumstances.189 And as New York’s brief, quoted
above, explains, there are correctional administrators who use grievance filings in
ways similar to those I have been investigating: to track trends, understand
problems, and so on. But the grievance system is not really a claims management
system—it is more of a claims substitute. And overall, it is rare in corrections
188

Brief of the States of New York et al. as Amici Curiae Supporting Petitioner, Woodford v.
Ngo, 548 U.S. 81 (2006) (No. 05-416).
189
Telephone Interview with Randall Berg, Director, Florida Justice Institute (June 5, 2007).
For a full account of this case, Haney v. Miami-Dade County, No. 1:04-CV-20516 (S.D. Fla. filed
Mar. 5, 2004), see Margo Schlanger, Jail Strip Search Cases: Patterns and Participants, 71 LAW
& CONTEMP. PROB. 65, 67-73 (2008).
http://www.bepress.com/jtl/vol2/iss1/art1

44

Schlanger: Operationalizing Deterrence

that claims management information is used to strategize harm reduction. Claims
management does not have an important impact on harm reduction by this path.
Bureaucratization. Only over the past thirty-five years have prisons (and to a
lesser extent jails) entered the modern bureaucratic state. Jim Jacobs explained in
1980:
Until recently, prisons operated as traditional, nonbureaucratic
institutions. There were no written rules and regulations, and daily
operating procedures were passed down from one generation to the
next. Wardens spoke of prison administration as an “art”; they
operated by intuition. The ability of the administration to act as it
pleased reinforced its almost total dominance of the inmates.190
Litigation has been an important driver of this profound transformation.191
Two of the methods by which litigation has had an effect are related to claims
management. First, knowing that they will be sued dozens or even hundreds of
times each year, correctional agencies have developed practices that make
responding to those lawsuits easier and more routine as well as more effective.
They now produce contemporaneous records: They write incident reports,
videotape cell extractions, keep easily copied shift logs and the like. And they
develop written policies and procedures easier to present in pleadings and
testimony.192
In addition, particularly in the 1970s and 1980s, an interest in easing defense
against claims pushed jails and prisons to use standard operating procedures.
Written policies and procedures could be offered in court proceedings as
deserving of deference, because they were at least rational, and more
aggressively, expert. In addition, they could provide circumstantial evidence of
due care or compliance. Prison and jail accreditation norms echo this approach—
the American Correctional Association’s accreditation standards frequently
require the existence of a written policy on some given topic, without specifying
what that policy should say or examining whether it is followed in practice.193 As
the ACA itself explains, accreditation provides “a defense against lawsuits

190

JAMES B. JACOBS, NEW PERSPECTIVES ON PRISON AND IMPRISONMENT 54 (1983)
(reprinting chapter originally published in 1980). For an account of the transformation in one
prison, see JAMES B. JACOBS, STATEVILLE: THE PENITENTIARY IN MASS SOCIETY (1977).
191
See generally id. at 54-54; Malcolm F. Feeley & Roger A. Hanson, The Impact of Judicial
Intervention on Prisons and Jails: A Framework for Analysis and a Review of the Literature, in
COURTS, CORRECTIONS, AND THE CONSTITUTION: THE IMPACT OF JUDICIAL INTERVENTION ON
PRISONS AND JAILS 25-28 (John J. DiIulio, Jr., ed. 1990).
192
The federal Bureau of Prisons is perhaps the most markedly bureaucratic system. For
many of its policies, see http://www.bop.gov/DataSource/execute/dsPolicyLoc.
193
See Elizabeth Alexander, What’s Wrong with the ACA?, 15 NAT’L PRISON PROJECT J. 1
(2001).
Published by The Berkeley Electronic Press, 2008

45

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

through documentation and the demonstration of a ‘good faith’ effort to improve
conditions of confinement.”194
In total, then, prison and jail organizations’ interest in minimizing the ex post
cost of claims management and defense has been an important component in the
bureaucratization of American corrections. And bureaucratization is particularly
useful in corrections, because both inadequate control and the abuse of
discretionary power can be so harmful to prisoners. Non-bureaucratic prison
regimes, to quote John DiIulio, “bounced between the poles of anarchy and
tyranny; between the Hobbesian state of inmate predators and the autocratic,
arbitrary regime of iron-fisted wardens.”195 Bureaucratic jails and prisons can be
harsh and inhumane, but it is uncontroversial that prison and jail inmates are
better off when their incarcerating facilities have, for example, written policies,
stated rules of conduct for their staff, and the variety of practices and procedures
that allow supervisors to monitor line officers.196 The point that prison litigation
has encouraged use of these minimal bureaucratic features is not novel. The
contribution of my account197 is the observation that these salutary changes were
provoked by claims management imperatives—the need to respond to litigation,
as much as anything substantive about the litigation.
Personnel. In addition to developing a set of institutional strategies for
facilitating organizational processing of and response to lawsuits, prison and, to a
more limited extent, jail systems have devoted staff to claims management. In
fact, every state department of corrections has both low- and high-level personnel
who spend significant portions of their time dealing with prisoner litigation.
There are litigation officers, compliance officers, risk assessment personnel, and
lawyers and paralegals in corrections departments and in offices of attorneys
general. The higher ranked among this group have their own professional
networks, as part of the National Association of Attorneys General and the
American Corrections Association. Even though most jails are far smaller than
most prisons, let alone prison systems, and small jails in particular are far less
likely to employ readily available lawyers with expertise in inmate litigation, jails,
too, often institutionalize some lower-priced staffing arrangement to deal with
194

AMERICAN CORRECTIONAL ASSOCIATION, STANDARDS FOR ADULT CORRECTIONAL
INSTITUTIONS xvi (4th ed. 2003); see also, e.g., William J. Rold, The Legal Context of
Correctional Health Care, in NATIONAL COMMISSION ON CORRECTIONAL HEALTH CARE,
STANDARDS FOR HEALTH SERVICES IN PRISONS 137-47 (2003).
195
JOHN J. DIIULIO, JR., GOVERNING PRISONS: A COMPARATIVE STUDY OF
CORRECTIONAL MANAGEMENT 236 (1987).
196
The most prominent and unambivalent supporter of correctional bureaucratization is John
DiIulio. See id. at 236-41. For a brief discussion of the potential downsides of bureaucratization,
see Schlanger, Inmate Litigation, supra note 4.
197
I have, however, made this point before, in Schlanger, Inmate Litigation, supra note 4, at
1667.
http://www.bepress.com/jtl/vol2/iss1/art1

46

Schlanger: Operationalizing Deterrence

prisoners’ cases, for example assigning an officer to be the “litigation officer” (in
addition to other tasks) in charge of coordinating responses to filed cases.198
Brought inside correctional organizations’ routines to respond to litigation,
staff in these categories begin to play other roles as well; it is these roles that are
important to understand in an analysis of claims managements impact on damage
action deterrence. First and foremost, litigation personnel in corrections
organizations act as law transmitters: Given the nearly omnipresent ambiguity of
legal requirements, staff inevitably must partially construct the law in order to
create a coherent account of its regulatory demands. The content of that account
is as much about organizational and interorganizational politics as it is about what
courts or legislatures say. As mentioned in Part II, in the deeply oppositional
world of corrections, claims management personnel may seek to ensure that they
are not too deeply identified with prisoners by their colleagues. But over the past
15 years, I have observed and talked to dozens of officers and lawyers assigned to
manage corrections litigation. It is obvious to me that often though not always
correctional litigation staff echo Edelman and Suchman’s description of corporate
officers in charge of implementing “[p]ractices designed to promote (or merely to
symbolize) workplace safety, or equal employment opportunity, or environmental
protection,” by “becom[ing] internal advocates for the values that the practices
symbolize.”199 Perhaps this is because jobs predictably attract people who think
the job is important, or because litigation staff learn from experience that
compliance is easier than corrections officials had feared.200 Or perhaps litigation
staff exaggerate, consciously or unconsciously, the degree of liability exposure in
order to underscore their own vital role within the organization and enhance their
professional standing.201 Regardless of the cause, the personnel hired to do the

198

See id. at 1665.
Edelman & Suchman, When the “Haves” Hold Court, supra note 63, at 962-63.
200
I am thinking here, for example, of the limited due process rights constitutionally owed
prisoners, which were strenuously resisted when first announced, but which prisons and jails now
tend to meet without difficulty and provide even when the Constitution does not so require. The
head of the Federal Bureau of Prisons, James Bennett, reported in 1974 that recently imposed due
process requirements “have not only watered down measurably the authority of the wardens but
have imposed burdens almost impossible to implement within present appropriations and
available legal talent. . . . The erosion of official authority and need for speedy trial and action
could have unforeseeable consequences if efforts to achieve full due process are pressed too far[,]
but be prepared.” James V. Bennett, Who Wants To Be a Warden?, 1 NEW ENG. J. PRISON L.
69, 72 (1974). But after the Supreme Court declared twenty years later in Sandin v. Connor, 515
U.S. 472 (1995), that prisoners have no due process rights with respect to ordinary disciplinary
hearings, prisons have not reverted to pre-due-process practice, and in fact very few have changed
how those hearings are conducted, at all.
201
See, e.g., Edelman et al., Legal Ambiguity and the Politics of Compliance, supra note 63,
at 77.
199

Published by The Berkeley Electronic Press, 2008

47

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

tasks needed for responding to litigation frequently simultaneously contribute to
reducing unconstitutional misbehavior within prison and jail systems.
Salience. Finally, it would be difficult to overstate the salience of litigation in
correctional settings. Litigation is an almost obsessive interest of correctional
staff and administrators. It is the subject of training, policy, gossip, and anxiety.
This is a puzzle, for two reasons. The first is that in nearly all prison and jail
litigation, individual officers do not pay the litigation costs or any damages, and
face few or no consequences related to either settlements or judgments.202 The
second, already stated, is the low payout risk, even for the agency itself. One
would think that hardly anyone would even notice the many lawsuits—that they
would be considered an annoyance, rather than a serious problem. The contrary
state of affairs suggests that it may be the claims management issues, rather than
the liability risks, that makes prisoner litigation loom so large in the minds of
corrections officials and staff. Significant money rarely changes hands in prison
and jail cases, but because there are so many lawsuits, prison and jail
administrators very often need to deal with them—they need to prepare affidavits,
locate records, investigate claims, sit for depositions, and the like. This likely
makes the threat of liability far more cognitively available.
CONCLUSION
This article contributes to the project of opening up the black box of damage
action deterrence, of understanding how litigation incentivizes care within large
risk-creating organizations. The case studies presented suggest that the structures,
policies, and practices that potential defendants adopt in order to become cheaper
and more effective respondents to lawsuits—the claims management
infrastructure they build—can also have consequences for risk and care. The
three case studies illustrate four ways in which organizations’ claims management
routines and structures can promote care-taking. In all three—the retailer Acme,
202

The typical on-the-books arrangement, usually statutorily enacted, is that the
correctional agency indemnifies its officers unless the act on which a lawsuit is predicated was
outside the “scope of employment” or was intentional or malicious. See SCHUCK, SUING
GOVERNMENT, supra note 4, at 85-88.
Agency-provided defense and near-universal
indemnification are the rule in practice. See, e.g., Theodore Eisenberg & Stewart Schwab, The
Reality of Constitutional Tort Litigation, 72 CORNELL L. REV. 641, 686 (1987); John Jeffries, Jr.,
In Praise of the Eleventh Amendment and Section 1983, 84 VA. L. REV. 47, 49-50 (1998);
Cornelia T.L. Pillard, Taking Fiction Seriously: The Strange Results of Public Officials’ Individual
Liability under Bivens, 88 GEO. L.J. 65, 74-80 (1999) (citing evidence that the federal government
agrees to 98% of the requests for defense of constitutional tort actions against federal employees,
and almost without exception pays any damages assessed once it has provided representation).
For additional discussion and authority, see Schlanger, Inmate Litigation, supra note 4, at 1671-72
& nn.389-391.
http://www.bepress.com/jtl/vol2/iss1/art1

48

Schlanger: Operationalizing Deterrence

hospitals, and prisons—I have suggested that the claims management personnel
and the increased salience of litigation threat caused by various claims
management endeavors contribute to harm prevention. In Acme and in hospitals
but not much in prisons, the information generated for claims management is then
used, secondarily, for harm prevention. And in hospitals and prisons but not in
Acme, claims management imperatives have promoted various bureaucratizing
strategies like documentation and implementation of standard operating
procedures, which strategies likely contribute to harm prevention. These four
connections between claims management and harm prevention—information,
personnel, bureaucratization, and salience—find support in sociological,
economic, and psychological theory. Much work remains, however. How
common are the types of interventions I have just described? And are they
effective? These and many other questions require more systematic empirical
investigation than I have attempted.
In addition, this article has dealt only implicitly with the question of who does
claims management. As I stated Part III, I chose the three case studies by looking
for organizations that handle their claims management internally, rather than
outsourcing this function. My thought was that a dynamic relationship between
claims management and harm prevention is most likely to exist when firm
operations and claims are handled within one organizational structure, shrinking
the organizational distance that must be traversed by any proposed harm
prevention techniques with a claims management source. If claims management
and loss prevention are ever interdependent,203 internal claims management
seemed the most likely field in which to find an effect, and therefore most
appropriate for this article’s project of hypothesis generation.
Although I did not choose them for this characteristic, the organizations
featured in Part III share another important institutional attribute—for none of
them does insurance relieve the pressure of most liability. Acme self-insures up
to $500,000, more than nearly all its claims. Likewise, large teaching hospitals
very often use a form of self-insurance known as captive insurers,204 in which the
hospital owns the primary insurer and therefore retains all but catastrophic risk.
For prisons, states pay any damages owed, without insurance. (Counties and
203

Cf. Young & Hood, Risk and Outsourcing Risk Management, supra note 31, at 118
(suggesting that “interdependency” between claims management and core functions of a
governmental organization might push against outsourcing claims management).
204
See Michelle M. Mello, Carly N. Kelly, David M. Studdert, Troyen A. Brennan & William
M. Sage, Hospitals’ Behavior In A Tort Crisis: Observations From Pennsylvania, 22 HEALTH
AFF. 229 (2003). For an introduction to captive insurance companies, see, e.g., International
Association of Insurance Supervisors, Issues Paper on the Regulation and Supervision of Captive
Insurance
Companies
(Oct.
2006),
available
at
http://www.iaisweb.org/__temp/Issues_Paper_on_Regulation_and_Supervision_of_Captive_Insur
ance_Companies__October_2006_.pdf.
Published by The Berkeley Electronic Press, 2008

49

Journal of Tort Law, Vol. 2 [2008], Iss. 1, Art. 1

cities sometimes directly pay damages assessed against jail officials; other times,
they pool the relevant risks in various ways.205) Just as for insurers, contracted
liability risk incentivizes efforts at loss prevention (as the regulatory insurance
scholarship described in this article’s introduction discusses), presumably this
shared risk retention contributes to these organizations’ observed readiness to
adapt operations to reduce the risk of harm they cause.
For each case study organization, then, there is no outsourcing of any of the
three areas of operations related to damage actions. Each retains in-house
liability, claims management, and loss-prevention. One would expect in such
circumstances that the connections between the three would be at their tightest.
Organizational theory and work on the economics of information suggest that
within organizations it is quite ordinary to see the kinds of transformations and
synergies discussed above. Repurposing seems most likely to happen if the firm
managing claims is also the firm managing risk-creating operations, and the firm
bearing liability risk, and least likely if all three are separated. But it would be
valuable to test this theory. Can it be demonstrated, for example, that firms with
in-house claims management offices reach systematically different results with
respect to harm prevention than those that use third-party claims managers? Does
the answer depend on any particular features of the organizations in question?
(For example, perhaps the benefit of in-house claims management is less for
highly decentralized firms.) In addition, among the issues future work ought to
examine a the various combinations among the three.206
The operationalization of deterrence is a crucial subject of research. To
understand the regulatory effects of the tort system or of other damage-action
regimes we need to analyze more thoroughly just how it is that the threat of
litigation and liability operates on risk causing organizations. I hope I have
demonstrated that study of claims management is a key part of that project.

205

See Schlanger, Jail Strip Search Cases, supra note 189.
George Cohen provides an example of the kind of insight that might result, in his
investigation of legal malpractice insurance and loss prevention. He suggests that law firms tend
to buy loss-prevention advice bundled with their liability insurance because transaction costs—
related to both contracting and information—are minimized by the use of one contractor for two
services, and because insurers can bond their information by liability indemnification. Cohen,
Legal Malpractice Insurance and Loss Prevention, supra note 19.
206

http://www.bepress.com/jtl/vol2/iss1/art1

50