<<

. 9
( 13)



>>

lem solving at regional and national levels). Effective institutional
designs create complex, multi-tiered systems with some levels of du-
plication, overlap, and contestation. The policy analyst™s penchant for
neat, orderly hierarchical systems needs to be replaced with a recogni-
tion that complex polycentric systems are needed to cope effectively
with complex problems of modern life.

9.2 Testing the Predictions of the Standard Model of Rational
Choice

One of the great advantages of contemporary game theory and formal
models of collective-action theory is that they generate clear predic-
tions of expected behavior in speci¬c types of situations. Given precise
models of collective-action situations and clear predictions of expected
behavior, it is possible to set up experimental laboratory designs that
enable one to test the empirical veracity of the predictions. With the
substantial methodological advances in conducting experimental labo-
ratory research (Smith 1982; Plott 1979), this method has become a use-
ful tool for social scientists in the testing of theories and the replication
of ¬ndings by multiple scholars in diverse cultures. Experimental re-
search related to the theory of collective action has generated very clear
predictions that have repeatedly been challenged in the lab. Let us
brie¬‚y discuss two related sets of predictions and results.

9.2.1 Predictions and Empirical Results from Linear Public Good
Games
When individuals are in a one-shot linear public good situation, each
individual can choose between contributing nothing to the provision
of a bene¬t that all will share or contributing some portion of a given
endowment of assets. Each individual is predicted to contribute zero
assets. When the game is repeated a ¬nite number of times, each indi-
vidual would contribute zero assets in the last round, and because of
backward induction, each individual is predicted to contribute zero
assets in each and every round leading up to the ¬nal round.
256 Ostrom



Not only do we have evidence from many ¬eld settings that individ-
uals do contribute to the provision of public goods (see, for example,
Loveman 1998; Kaboolian and Nelson 1998), there is similar evidence
from a large number of carefully controlled laboratory experiments.
Between 40 to 60 percent of subjects in a one-shot linear public good
situation contribute assets to the provision of a public good (Dawes,
McTavish, and Shaklee 1977; Isaac, Walker, and Thomas 1984; Davis
and Holt 1993; Ledyard 1995; Offerman 1997). About the same per-
centage of subjects contribute tokens in the ¬rst round of a ¬nitely
repeated public good experiment. The rate of contribution, however,
decays over time, approaching but never reaching the predicted zero
level (Isaac and Walker 1988). Because of the decay toward zero contri-
butions in the experiment™s last ten rounds, an initial reaction by theo-
rists was that it took subjects ten rounds to learn the rational way to
play the game. Subsequent experiments extended the pre-announced
time horizon to 20, 40, and 60 repetitions. These showed that subjects
tended to keep cooperation levels varying in the 30 to 50 percent range
for long sequences of time and that the decay toward zero contribu-
tions did not occur a few rounds prior to the announced ¬nal round
(Isaac, Walker, and Williams 1994).

9.2.2 Predictions and Empirical Evidence Related to Second- and
Third-Level Social Dilemmas
Not only is there a clear prediction concerning the lack of provision in
public good situations, participants are viewed as helpless in getting
out of such situations. An effort to arrive at an agreement for determin-
ing how much of a public good should be provided and how the costs
of provision should be shared would take time and effort to achieve.
Once achieved, everyone would bene¬t whether or not they had con-
tributed to the design of such an agreement. Thus, the prediction is
that no one would participate in the effort to extract themselves from
the initial dilemma. Furthermore, monitoring compliance to such an
agreement and sanctioning those who did not give their agreed-upon
share would be costly for those who might think about undertaking
such an activity. Again, everyone would bene¬t from such activities
whether or not they had contributed. Thus, no one is expected to invest
any of their own resources in monitoring and sanctioning activities.
But this is not what is found in many settings.
For example, in experiments where subjects are offered an opportu-
nity to pay a fee in order to assess a ¬ne against someone else, subjects
Policies That Crowd out Reciprocity and Collective Action 257



are willing to expend their own resources to punish non-cooperators
¨
(E. Ostrom, Walker, and Gardner 1992; Fehr and Gachter 1998; Yama-
gishi 1986). Similar to ¬eld settings, subjects in a lab are rather indig-
nant and angry at others who do not do their share in protecting a
common-pool resource or providing a public good. These subjects give
up costly resources to sanction noncooperators. And, when individuals
agree upon their own sanctioning system, they do not need to use it
extensively”as the compliance rate with a self-imposed harvesting
limit and sanctioning system is extremely high (E. Ostrom, Gardner,
and Walker 1994).
Cardenas, Stranlund, and Willis (2000) report on a common-pool re-
source experiment conducted in the Colombian countryside with cam-
pesinos who frequently have to deal with resource problems in their
everyday life. In one of the experimental conditions, the campesino sub-
jects were given a choice of withdrawal levels from the resource that
would be monitored by an external observer. The externally imposed
rule was that the subjects should harvest at an optimal level for group
returns, or face a realistic but low level of monitoring and a sanction
imposed by the outside observer. The subjects in this experimental
condition actually increased their withdrawal levels. This is in marked
contrast to their own behavior in those experiments where the subjects
could talk on a face-to-face basis and no rule was imposed. What was
remarkable about this experiment was that subjects, who were simply
allowed to communicate with one another on a face-to-face basis, were
able to achieve a higher joint return than the subjects who had an opti-
mal but imperfectly enforced rule imposed on them. As the authors
conclude:
We have presented evidence that indicates that local environmental poli-
cies that are modestly enforced, but nevertheless are predicted by standard
theory to be welfare-improving, may be ineffective. In fact, such a policy can
do more harm than good, especially in comparison to allowing individuals
collectively to confront local environmental dilemmas without intervention.
We have also . . . presented evidence that the fundamental reason for the
poor performance of external control is that it crowded out group-regarding
behavior in favor of greater self-interest. (Cardenas, Stranlund, and Willis
2000, 1731)

It has also been found that individuals in both experimental and
¬eld settings are willing to invest substantial time and energy in de-
signing and adapting rules so that they can achieve collective out-
comes. In ¬eld settings, the time and effort may be substantial (Lam
258 Ostrom



1998; Tang 1992; Gibson, McKean, and Ostrom 2000; Varughese and
Ostrom 2001). When local users feel a sense of ownership and depen-
dence on a local resource, many of them invest intensively in designing
and implementing ingenious local institutions”some of which are sus-
tained for centuries (E. Ostrom 1990).
Being involved in a face-to-face discussion about solving their own
overharvesting problem may generate unexpected capabilities and
learning. Frohlich and Oppenheimer (2003) designed an experiment in
which one set of ¬ve subjects ¬rst faced an ˜˜incentive compatible de-
vice™™ (ICD) in dealing with a repeated Prisoner™s Dilemma (PD) game
that randomized who was to receive the payoffs from their decisions.
While using this device, players were indeed able to achieve high
levels of joint bene¬ts, but about the same level as a control group that
was allowed to communicate on a face-to-face basis during the initial
phase of the experiment. Frohlich and Oppenheimer had hypothesized
that the experience of playing within an incentive compatible device
should carry over to a second phase when the same subjects played
seven rounds of a PD game without the device. They expected that
subjects who had experienced this device in the ¬rst phase would
contribute more after the removal of the device. Frohlich and Oppen-
heimer ultimately had to reject their own hypothesis, however. ˜˜With-
out discussion, there was no carryover effect. With discussion, the ICD,
possibly due to the decoupling of ethical concern and behavior, led to
signi¬cantly lowered levels of contributions than were found in the
regular PD™™ (Frohlich and Oppenheimer 2003, 289). They found that
face-to-face communication had a higher level of positive carryover
than the ICD.
Frohlich and Oppenheimer concluded that the use of an incentive
compatible device”at least the one that they used in the experiment”
could be ˜˜a two-edged sword, and ought to be studied further before
being advocated widely™™ (Frohlich and Oppenheimer 2003, 289). In
reasoning about their unexpected ¬nding, they considered how the in-
stitution affected the problem-solving process of those involved. When
making a decision within the ICD, subjects ˜˜confront a situation in
which their self-interest and the interests of all others coincide exactly.
What is best for them is, by explicit design, best for the group as a
whole™™ (Frohlich and Oppenheimer 2003, 290). Thus, these individuals
could make a decision without facing any tension between the self-
interested and the ethically best strategy. Subjects did not need to take
the impact of their decisions on others into account as contrasted with
Policies That Crowd out Reciprocity and Collective Action 259



their own best interest. ˜˜They don™t have to ¬‚ex their ethical muscles™™
(Frohlich and Oppenheimer 2003, 290).

9.3 Multiple Types of Players and Intrinsic Preferences

These (and closely related) empirical ¬ndings consistently challenge
predictions based on a presumption that all individuals can be char-
acterized by a single model of rational behavior when they interact
outside a highly competitive market setting. Furthermore, preferences
have not only been considered to be entirely self-interested, but also
¬xed and unchanged by the experience of being within a particular
institutional arrangement. It is thus necessary to reconstruct our basic
theories of collective action and to assume that at least some partici-
pants are not rational egoists (Sen 1977). At least some individuals in
social dilemma situations follow or can learn norms of behavior”such
as those of reciprocity, fairness, and trustworthiness”that lead them
to take actions that are directly contrary to those predicted by contem-
porary rational choice theory.
In other words, the behavior of many individuals is based on intrin-
sic preferences related to how they prefer to behave (and would like
others to behave) in situations requiring collective action to achieve
joint bene¬ts or avoid joint harm. Intrinsic preferences lead some
individuals to be conditional cooperators”willing to contribute to
collective action so long as others also contribute”as well as con-
ditional punishers”willing to sanction others who do not behave as
agreed upon or as accepted norms or rules prescribe so long as
they believe others are also conditional sanctioners. Intrinsic prefer-
ences transform some dilemmas into assurance games where there
are two equilibria and not just one (Chong 1991; Sen 1974). On the
other hand, some individuals do behave in a manner that closely
approximates the prediction for how rational egoists will behave.
Thus, one needs to assume multiple types of actors rather than only
rational egoists.
In many ongoing ¬eld situations, humans obtain considerable infor-
mation about each other and are able to engage in collective action
with those that they estimate share similar norms. In such situations,
some rational egoists will survive along with strong reciprocators so
that it is not possible to rely exclusively on the intrinsic motivation of
all participants to cooperate”especially if cooperation must be sus-
tained over time. Thus, in many cases, intrinsic motivation must be
260 Ostrom



backed up by institutions that enable those individuals motivated to
solve collective-action problems while protecting them from free-riders
and untrustworthy partners.
The rules crafted in many common-property regimes tend to in-
crease the probability of long-term interactions among participants.
Further, appropriation rights tend to be designed so that actions can
be monitored at a relatively low cost by other participants. When indi-
viduals can monitor each other, they increase the probability that those
who break rules will become known to others in the community. Thus,
the rules crafted by robust self-organized, common-property regimes
increase the probability that reciprocity will be widely practiced (Lam
1998; Tang 1992; Gibson, McKean, and Ostrom 2000; E. Ostrom 2000).
Evidence that institutions can crowd out intrinsic motivations (as
well as crowding them in) has been mounting over the past three de-
cades since Titmuss (1970) ¬rst raised this possibility. Psychological re-
search provides evidence that intrinsic motivation is diminished when
individuals feel that their own self-determination or self-esteem is ad-
versely affected (Deci and Ryan 1985; Deci, Koestner, and Ryan 1999).
In a recent review of this theory, Frey and Jegen (2001, 594“595) iden-
tify the psychological conditions when crowding out or crowding in is
likely to occur:

1. External interventions crowd out intrinsic motivation if the indi-
viduals affected perceive them to be controlling. In this case, both self-
determination and self-esteem suffer, and the individuals react by
reducing their intrinsic motivation in the activity controlled.
2. External interventions crowd in intrinsic motivation if the individu-
als concerned perceive it as supportive. In this case, self-esteem is fos-
tered, and individuals feel that they are given more freedom to act,
thus enlarging self-determination.

A recent meta-analysis of 128 laboratory studies that have explored
the effect of extrinsic rewards on intrinsic motivation found that tangi-
ble rewards tend to have a substantially negative effect on intrinsic
motivation (Deci, Koestner, and Ryan 1999). As the authors conclude:

Although rewards can control people™s behavior”indeed, that is presumably
why they are so widely advocated”the primary negative effect of rewards is
that they tend to forestall self-regulation. In other words, reward contingencies
undermine people™s taking responsibility for motivating or regulating them-
selves. When institutions”families, schools, businesses, and athletic teams, for
Policies That Crowd out Reciprocity and Collective Action 261


example”focus on the short-term and opt for controlling people™s behavior,
they may be having a substantially negative long-term effect. (Deci, Koestner,
and Ryan 1999, 659)

There are obviously many interactions where ˜˜controlling people™s be-
havior™™ is what is desirable. Individuals, in their role as citizens, are
not, however, someone else™s employees or agents. Intrinsic values are
important sources of citizens™ motivation to participate in political life
by volunteering to do community service, ¬nding solutions to commu-
nity problems, and paying taxes.
Paying taxes is obviously one of the most important acts that citi-
zens in a contemporary democratic system make. In a survey of
prior studies of taxpayer behavior, Snaveley (1990, 70) concluded that
studies repeatedly show ˜˜compliance policies which emphasize in-
creasing risk for tax evasion will not in themselves be suf¬cient to curb
cheating.™™ He urges that policymakers adopt a more comprehensive
approach to increasing the rate of taxpayer compliance by encourag-
ing the development of ˜˜taxpaying values™™ through education pro-
grams that stress service. ˜˜Taxpayer decisions are in¬‚uenced by a
combination of economic self-interest factors and noneconomic criteria,
therefore, both the coercive and service/values policy approaches are
necessary. The choice to be made is one of balancing the two types of
policies™™ (Snaveley 1990, 70).
Paying taxes is obviously one of the methods used by governments
at all levels for solving collective-action problems by requiring funds
from bene¬ciaries to provide public goods and protect common-pool
resources. When there is a clear relationship between the taxes that an
individual pays and the goods and services obtained, taxes also repre-
sent an important link between citizens and their of¬cials. In some
cases, relatively long-lived local institutions have been seriously chal-
lenged by a lack of understanding by of¬cials and by policy analysts
of how the institutions operated and why an effective and clear linkage
between resource users and their of¬cials is so important. An intrigu-
ing example is from Taiwan, where a weakening of this linkage be-
tween resource users and their of¬cials has led to a weakening of the
links between of¬cials and investment in resources, and in the amount
of time that resource users spend monitoring each other™s behavior and
the condition of the resource.
In Taiwan, a set of seventeen irrigation associations has been respon-
sible for the operation and maintenance of a large number of Taiwan™s
262 Ostrom



irrigation systems. The irrigation associations were corporations orga-
nized by farmers, who have paid fees to their local irrigation associa-
tion for many years. The irrigation association, in turn, took
substantial responsibility for the day-to-day maintenance and opera-
tion of local canals, while the Government of Taiwan has undertaken
responsibility for the construction and operation of the larger irrigation
works. The irrigation associations have repeatedly been acclaimed as
major contributors to ef¬cient irrigation in the country and thus to sub-
stantial agricultural development (Levine 1977; Moore 1989; Lam
1996).
Taiwan, like other countries whose economies are less and less de-
pendent on agriculture and more dependent on industrial and service
industries, has been trying to ¬nd ways of adjusting a variety of eco-
nomic policies. Furthermore, the rural population still has a signi¬-
cant vote and national politicians have been vying for support in
rural areas. In the early 1990s, politicians argued that farmers faced
hard times and could not make a decent living. As Wai Fung Lam
described, ˜˜The government, argued these politicians, should not bur-
den the farmers with irrigation fees. In 1993, after much political nego-
tiation, the government agreed to pay the irrigation fees on behalf of
the farmers™™ (Lam forthcoming, 7“8). As it turned out, both major na-
tional parties supported the cancellation of irrigation fees as no one
wanted to be seen as against the farmer, even though many of the of¬-
cials familiar with irrigation expressed substantial concern about the
long-term consequences.
The cancellation of the fee has had substantially adverse conse-
quences. Farmers are much less likely to volunteer work activities,
pay voluntary group fees, or pay much attention to what is happen-
ing on the canals and in the ecological environment around them as
they had done before (Wade 1995). As one irrigation association of¬-
cial expressed it: ˜˜The problem facing irrigation management at the
¬eld level is not simply a matter of ¬nding one or two farmers to serve
as [local group] leaders, the more serious challenge is that nowadays
fewer and fewer farmers have good knowledge of their own systems
and understand how to engage with one another in organizing collec-
tive action™™ (quoted in Lam forthcoming, 12). Maintenance of the sys-
tems has been declining precipitously, and the cost of water supply
has been increasing rather than decreasing. Thus, systems that have
been robust for a long period of time have largely been destroyed by
an effort ˜˜to help™™ farmers by reducing the burden on them”a ˜˜bur-
Policies That Crowd out Reciprocity and Collective Action 263



den™™ that they had earlier placed on themselves. Thus, understanding
the difference between an internally constituted rule and method for
¬nancing services and an externally imposed inducement or sanction
is crucial in efforts to enhance the capabilities of citizens to engage in
collective action.
In a fascinating study of citizens™ willingness to accept a nuclear
waste repository in their community”an example of a classic NIMBY
(not in my back yard) problem”Frey and Oberholzer-Gee (1997) con-
ducted a survey of citizens in a region of Switzerland where of¬cials
were attempting to ¬nd a location for such a facility. Respondents
were initially asked if they were willing to accept a facility in their
community. About half (50.8 percent) of the respondents indicated a
willingness to have a nuclear waste facility in their community. When
the same respondents were asked their willingness to accept such a fa-
cility if the Swiss parliament offered substantial compensation to all
residents of a community that accepted the facility, the level of willing-
ness dropped dramatically to 24.6 percent. Being offered a ¬nancial
reward to accept a NIMBY-type project thus led one-quarter of the
respondents to change their minds and oppose the placement of the fa-
cility in their community.
Some scholars have accepted the possibility that material incentives
may crowd out intrinsic motivations in some settings, but are skeptical
that the overall effect is negative (Lazear 2000). In other words, if exter-
nal incentives generate suf¬cient effort, this may be a more ef¬cient
way to motivate citizens and employees than relying as much on in-
trinsic motivation.
Frey and Benz (2001) report on a recent experiment conducted at
¨
the University of Zurich that directly addresses this question. They
designed an experiment based on the trust game in which a ¬rst player
(the principal) may send part of an endowment to a second player (the
agent) that is then tripled in value. In the second stage, after being in-
formed of the amount sent by the principal, the agent selects a costly
but complementary combination of variables referred to as work quan-
tity and work quality that will return funds to the principal if positive
values are chosen. In the base experiment, a pro¬t-maximizing subject
should choose the minimal level of work quantity and work quality,
since there is no opportunity for the principal to punish the agent for
choosing the minimal level. Knowing that a rational egoist will keep
all of the funds sent by the principal, the principal should offer
the minimum feasible contract in the ¬rst place. Prior experiments,
264 Ostrom



however, had already shown that a substantial proportion of subjects
in both roles contributed substantially higher-than-minimum levels to
the other player (see, for example, Berg, Dickhaut, and McCabe 1995;
Kirchler, Fehr, and Evans 1996; E. Ostrom and Walker 2003).
In this experimental design, Frey and Benz added an experimental
condition whereby the principal has a possibility to punish the agent
for shirking. They found that this material incentive had a dramatic
effect on the agent™s contribution to the quantity dimension (which
could be monitored and punished), but also in regard to the second
dimension that could not be monitored and punished. The agents
who showed at least a potential propensity to cooperate voluntarily
responded to the threatened monitoring and sanctioning of their con-
tribution by strongly reducing their effort in the work dimension that
could not be monitored, while modestly increasing the work dimen-
sion that could be monitored. Further, the introduction of ˜˜perfor-
mance incentives™™ led the intrinsically motivated agents to exert
signi¬cantly lower overall efforts when both dimensions were consid-
ered. As Frey and Benz (2001, 19“20) conclude:
The economic incentive in itself changes the frame of the exchange relationship
across treatments. Over and above the relative price effect they produce, incen-
tives undermine part of the underlying intrinsic motivation by transforming a
relational contract into a purely transactional contract. . . . Individuals lower
their intrinsic efforts in the dimensions where they have the leeway to do so,
i.e., in those areas where they do not face the countervailing relative price effect
provided by the incentive mechanism.

Ernst Fehr and Bettina Rockenbach (2003) have also conducted a fas-
cinating and relevant experiment using a one-shot trust game. In their
experiment, there were three conditions under which a principal could
allocate an amount (up to the ten monetary units) to a trustee:
(1) the principal could simply record a ˜˜desired back-transfer™™ that the
principal would like to receive back from the trustee,
(2) the principal could indicate that he or she planned to impose a ¬ne
of four monetary units if the trustee did not return the desired level, or
(3) the principal could indicate a plan not to waive the imposition of
a ¬ne.
Fehr and Rockenbach found that trustees paid back substantial funds
under all conditions and used reciprocity in determining the amount
they paid back”the more the principal transferred, the more the
Policies That Crowd out Reciprocity and Collective Action 265



trustee paid back. Across all investment levels by a principal, however,
the trustees sent back more funds when the principal refrained from
imposing the ¬ne and returned the lowest amount when the principal
imposed the ¬ne. Fehr and Rockenbach conclude that ˜˜strong reciproc-
ity™™ can help explain this paradoxical behavior.
First, refraining from the threat of ¬ning, although the threat is available, could
itself be perceived as a fair act, which induces the trustees to increase their co-
operation. Second, attempts to use the sanction to enforce an unfair distribu-
tion of income may be perceived as hostile acts, inducing the trustees to
reduce cooperation. (Fehr and Rockenbach 2003, 139“140)

Prior research in regard to the performance of police of¬cers in ¬eld
settings has found similar patterns without being able to precisely
measure the relative investment. In regard to policing, researchers
have argued that performance rewards based on crime ¬ghting pri-
marily reduce the contribution that of¬cers are willing to make in re-
gard to noncrime service to citizens. In light of his extensive research
on police, Herman Goldstein concluded that the ˜˜traditional methods
for measuring the rewarding of ef¬ciency of both individual of¬cers
and organizational units place no positive value on the quality of the
police response in other than crime-related situations™™ (cited in Brown
1977, 91). In other words, if an of¬cer is ˜˜only rewarded for ˜crook
catching,™ and given no recognition for treating people decently, he
simply does not have any incentive to be a public servant™™ (Brown
1977, 91).
Three important lessons can be derived from these research efforts
and other recent theoretical and empirical research based on an as-
sumption of multiple types of players including rational egoists and
conditional cooperators who have adopted norms of fairness, reciproc-
ity, and trust. The ¬rst lesson is that many individuals are motivated
by social norms that affect intrinsic motivation or are at least capable
of learning social norms and using them to guide some of these deci-
sions. Second, it is possible for individuals who adopt these norms to
survive in repeated situations where they face rational egoists as well
as others who share similar norms. So long as they can identify one
another, trustworthy fair reciprocators actually achieve higher mate-
rial rewards over time than rational egoists! In other words, they can
¬‚ourish. The third lesson is that achieving some reliable information
about the trustworthiness of others is crucial to this accomplishment.
Consequently, institutions that enhance the level of information that
266 Ostrom



participants obtain about one another is essential to increase the capac-
ity of individuals to solve collective-action problems. Information rules
are as important (or more important) in solving collective-action prob-
lems than payoff rules, but payoff rules have been the primary focus of
a considerable percentage of public policy initiatives. This is not the
only problem, however, with the types of public policies that have
been recommended”and in some cases implemented”based on cur-
rently accepted theory.

Public Policies Based on the Extant Theory of Collective Inaction
9.4

The theory of collective inaction articulated in 1965 by Mancur Olson
was reinforced by the powerful metaphor of the ˜˜tragedy of the com-
mons™™ articulated by Garrett Hardin in 1968 and by considerable work
in non-cooperative game theory examining various collective-action
problems related both to public goods and to common-pool resource
problems. The 1960s and 1970s were an era in which considerable faith
existed in the capacity of strong national governments to solve both so-
cial and environmental problems through the application of rational
planning and the design of incentives to induce positive and deter
negative behavior. Many national policies”especially in developing
countries”were adopted on the presumption that local users of natu-
ral resources were unable to cope effectively with the governance and
management of local forests, water resources, wildlife, and ¬sheries
(Gibson 1999; Arnold 1998). In many countries, control over natural
resources was turned over to a national bureaucracy charged with the
responsibility of devising ef¬cient and effective ways of utilizing these
resources and sustaining their long-term productivity (Bromley et al.
1992; Agrawal 1999).
In many settings where individuals have managed small- to
medium-sized resources for centuries by drawing on local knowledge
and locally crafted institutions, their disempowerment led to a worsen-
ing of environmental problems rather than to their betterment (Finlay-
son and McCay 1998; Wunsch and Olowu 1995; Shivakoti and Ostrom
2002). Weak and frequently corrupt bureaucratic agencies were not
able to monitor use of these resources effectively, let alone devise effec-
tive plans for their long-term sustainability (Repetto 1986). What had
been de facto community property became de jure government prop-
erty. In reality, it then became de facto open access and unregulated
property (Bromley et al. 1992).
Policies That Crowd out Reciprocity and Collective Action 267



Furthermore, citizens are effectively told two rather devastating
messages in regard to the long-term development and sustenance of
a democratic society. First, public pronouncements stress that only
short-term sel¬sh actions are expected from ˜˜the common people.™™
When this is the case, solving collective-action problems requires
public policies that are based on externally designed and monitored
inducements. What we know from social psychological research, how-
ever, is that external inducements tend to ˜˜crowd out™™ intrinsic moti-
vations when individuals feel like they have lost control. Or, as Bruno
Frey (1997, 44) has stressed, ˜˜a constitution designed for knaves tends
to drive out civic virtues.™™ When intrinsic motivations are crowded
out, substantially more material resources are required to induce
effort than when incentives support a sense of control and reliance
on intrinsic as well as material incentives. When citizens feel a
moral obligation to pay their taxes, it is possible to design a tax col-
lection service that keeps collection costs at a low level. An effective
tax system, however, requires that most citizens accept the norm that
they should pay taxes. To achieve this objective over the long run, the
tax system must function in a fair manner and citizens must be able
to trust that others are also contributing their fair share (Rothstein
1998).
The second message contained in the policy literature is that citizens
do not have the knowledge or skills needed to design appropriate
institutions to overcome collective-action problems. Professional plan-
ners are, on the other hand, assumed to have the skills to analyze com-
plex problems, design optimal policies, and implement these policies.
Citizens are effectively told that they should be passive observers in
the process of design and implementation of effective public policy.
The role of citizenship is reduced to voting every few years between
competing teams of political leaders. Then, citizens are supposed to sit
back and leave the ˜˜driving™™ of the political system to the experts hired
by these political leaders. Let us brie¬‚y examine whether the assump-
tion that national of¬cials can actually select optimal policies for the
regulation of natural resources is realistic.
Over the last ¬fteen years, colleagues associated with the Workshop
in Political Theory and Policy Analysis have collected thousands of
written cases of resources managed by local users of inshore ¬sheries,
irrigation systems, and grazing lands (Schlager 1994; Schlager and
Ostrom 1992; Blomquist 1992; Agrawal 1994; E. Ostrom, Gardner, and
268 Ostrom



Walker 1994; Tang 1992; Lam 1998; Hess 1999). In Nepal, for example,
we have now collected data about the rules and general management
strategies used to govern and manage over 200 irrigation systems
including systems that are managed by government agencies (agency
managed irrigation systems, or AMIS) as well as those managed by
the farmers themselves (farmer managed irrigation systems, or FMIS).
We have consistently found that FMIS are able to achieve a higher ag-
ricultural yield than AMIS, that water is distributed more equitably in
FMIS than AMIS, and that the irrigation systems are better maintained
by FMIS than by AMIS (see Lam 1998; Joshi et al. 2000; Shivakoti and
Ostrom 2002).
What is striking, moreover, is the difference in how rules are
enforced by the farmers themselves on their own systems versus gov-
ernment of¬cials on the government systems. On 23 percent of the
AMIS systems, farmers report that government of¬cials are likely to
record of¬cial infractions. In contrast, farmers on 58 percent of the
FMIS report that their own farmer-monitors record infractions ob-
served ( Joshi et al. 2000, 76). In addition, ¬nes are more likely to actu-
ally be imposed within FMIS than within AMIS. Furthermore, farmers
also report that rules are highly likely to be followed 65 percent of the
time in FMIS and only 35 percent of the time in AMIS ( Joshi et al. 2000,
76). Thus, rules and sanctions devised by the farmers themselves”and
monitored by individuals who are responsible to the farmers”are
more likely to be enforced and lead to higher levels of rule compli-
ance than rules and sanctions imposed by an external agency. Rules
enforced by FMIS crowd in cooperation levels rather than crowding
out cooperation.
We are now engaged in a massive ten-country comparative, over-
time study of diverse forest institutions (Gibson, McKean, and Ostrom
2000; Poteete and Ostrom 2004). We have again found that resource
systems where local users have considerable authority to make their
own rules and enforce them are able to increase the level of coopera-
tion achieved as contrasted to systems where rules are imposed from
the outside. We are also paying particular attention to the speci¬c
rules that individuals use to regulate entry and allocate uses of local
resources.
What one learns from this research is the huge variety of rules that
are used in practice”many combinations of which are successful. For
example, we have identi¬ed twenty-seven different types of boundary
rules used by self-organized resource regimes (for speci¬cs, see E.
Policies That Crowd out Reciprocity and Collective Action 269



Ostrom 1999). Many of these rules enhance the likelihood that individ-
uals know each other and will be engaged over the long-term with one
another. In other words, the endogenously designed rules enhance the
conditions needed to solve collective-action problems. We have also
identi¬ed over 100 authority rules used to allocate resource users™
rights to the ¬‚ow from a resource system (E. Ostrom 1999). Many of
these rules focus on time, space, and technology rather than on the
quantity of resource ¬‚ow allocated. Consequently, these rules increase
the information that individuals obtain about the actions taken by
others at a low cost. Compliance rates increase when individuals feel
that others are also following the rules.
The policy of assigning all authority to a central agency to design
rules is based on a false conception that there are only a few rules that
need to be considered and that only experts know these options and
can design optimal policies. Our empirical research strongly challenges
this assumption. There are thousands of individual rules that can be
used to manage resources. No one, including a scienti¬cally trained profes-
sional staff, can do a complete analysis of any particular situation.
All policies need to be viewed as experiments (Campbell 1969). The
possibility of errors is always present given human limitations. Thus,
creating some redundancy in the design of rules for well-bounded lo-
cal resources (or communities) encourages considerable experimenta-
tion essential to discover some of the more successful combinations of
rule systems (Low et al. 2001). Further, ecological systems vary from
one place to another and from one mix of species to another. The com-
bination of rules that works well for lobster ¬sheries may be a disaster
for deep-sea ¬sheries (and vice versa) (Wilson et al. 2001). A good
combination of rules for a river system that has multiple regulatory
devices, such as dams, may be a disaster for a run-of-the-river system,
and vice versa.
Thus, instead of proposing highly centralized governance systems,
the best empirical evidence we can bring to bear on the question of
building sustainable democratic systems for sustainable resource use
is to design polycentric systems (V. Ostrom 1987, 1997). A polycentric
system has multiple semiautonomous units of governance located at
small, regional, national, and now international scales of organization
(Keohane and Ostrom 1995). Some of these governance units may be
organized in the private sector while others are organized in the public
sector. Government is not the only form of governance that humans
have devised over the centuries. The essential elements of dynamic
270 Ostrom



polycentric systems are mechanisms for generating information about
patterns of interactions and outcomes and mechanisms for oversight
and self-correction. A completely decentralized system of small local
units without overlap is as incapable of learning and self-correction
as a fully centralized system. Large-scale, overlapping units are an
essential part of a modern democratic system. However, smaller- to
medium-scale units are also a necessary part of an overall polycentric
system.
Modern policy analysis needs to catch up with contemporary empir-
ical and theoretical research. The two implicit messages contained in
much of contemporary public policy analysis are not only inef¬cient
and ineffective, they are dangerous for the long-term sustainability of
democratic systems of governance. The ¬rst message undermines the
normative foundations of a free society. It basically says that it is okay
to be narrowly self-interested and to wait for externally imposed
inducements or sanctions before voluntarily contributing to collective
action. The second message undermines the positive foundations of a
free society by destroying the capacity of citizens to experiment with
diverse ways of coping with multiple problems and to learn from this
experimentation over time. This message basically says that there is
one best way of solving all collective-action problems and it is only
knowable to experts. Citizens are viewed as having little to contribute
to the design of public policies.
Thus, much of contemporary policy analysis and the policies
adopted in many modern democracies crowd out citizenship and vol-
untary levels of cooperation. They do this by crowding out norms of
trust and reciprocity, by crowding out the knowledge of local circum-
stances, by crowding out the discussion of ethical issues with others
who are affected, and by crowding out the experimentation needed
to design effective institutions. Crowding out reciprocity, coopera-
tion, and citizenship is a waste of human and material resources and
presents a serious challenge to the sustainability of democratic institu-
tions over time.

Notes

An earlier version of this paper was presented at a conference on ˜˜Intrinsic Motivation in
Law and Business™™ in Gerzensee, Switzerland, June 18“22, 2001. The author thanks the
MacArthur Foundation and the Ford Foundation for ¬nancial support; Bruno Frey, Her-
bert Gintis, and two anonymous reviewers for very useful comments; and Patty Lezotte
for her excellent editing of multiple versions of this paper.
Policies That Crowd out Reciprocity and Collective Action 271



References

Agrawal, Arun. 1994. ˜˜Rules, Rule Making, and Rule Breaking: Examining the Fit be-
tween Rule Systems and Resource Use.™™ In Rules, Games, and Common-Pool Resources.
Eds. Elinor Ostrom, Roy Gardner, and James M. Walker. Ann Arbor: University of Mich-
igan Press, 267“282.
”””. 1999. Greener Pastures: Politics, Markets, and Community among a Migrant Pastoral
People. Durham, NC: Duke University Press.
Arnold, J. E. M. 1998. Managing Forests as Common Property. FAO Forestry Paper #136.
Rome, Italy: Food and Agriculture Organization of the United Nations.
Berg, Joyce, John W. Dickhaut, and Kevin McCabe. 1995. ˜˜Trust, Reciprocity, and Social
History.™™ Games and Economic Behavior 10(1):122“142.
Blomquist, William. 1992. Dividing the Waters: Governing Groundwater in Southern Califor-
nia. Oakland, CA: ICS Press.
Bowles, Samuel. 1998. ˜˜Endogenous Preferences: The Cultural Consequences of Markets
and Other Economic Institutions.™™ Journal of Economic Literature 36:76“111.
Bromley, Daniel W., David Feeny, Margaret McKean, Pauline Peters, Jere Gilles, Ronald
Oakerson, C. Ford Runge, and James Thomson, eds. 1992. Making the Commons Work:
Theory, Practice, and Policy. Oakland, CA: ICS Press.
Brown, Lee P. 1977. ˜˜Bridges over Troubled Waters: A Perspective on Policing in the
Black Community.™™ In Black Perspectives on Crime and the Criminal Justice System: A Sympo-
sium. Ed. Robert L. Woodson. Boston, MA: G. K. Hall, 79“105.
Campbell, Donald T. 1969. ˜˜Reforms as Experiments.™™ American Psychologist 24:409“429.
Cardenas, Juan-Camilo, John K. Stranlund, and Cleve E. Willis. 2000. ˜˜Local Envi-
ronmental Control and Institutional Crowding-out.™™ World Development 28(10):1719“
1733.
Chong, D. 1991. Collective Action and the Civil Rights Movement. Chicago: University of
Chicago Press.

Davis, Douglas D., and Charles A. Holt. 1993. Experimental Economics. Princeton, NJ:
Princeton University Press.

Dawes, Robyn M., Jeanne McTavish, and Harriet Shaklee. 1977. ˜˜Behavior, Communica-
tion, and Assumptions about Other People™s Behavior in a Commons Dilemma Situa-
tion.™™ Journal of Personality and Social Psychology 35:1“11.
Deci, E. L., R. Koestner, and R. M. Ryan. 1999. ˜˜A Meta-analytic Review of Experiments
Examining the Effects of Extrinsic Rewards on Intrinsic Motivation.™™ Psychological Bulletin
125.
Deci, E. L., and R. M. Ryan. 1985. Intrinsic Motivation and Self-Determination in Human
Behavior. New York: Plenum Press.
¨ ¨
Fehr, Ernst, and S. Gachter. 1998. ˜˜Cooperation and Punishment.™™ Working paper. Zu-
¨
rich, Switzerland: University of Zurich.

Fehr, Ernst, and Bettina Rockenbach. 2003. ˜˜Detrimental Effects of Sanctions on Human
Altruism.™™ Nature 442 (March 13): 137“140.
272 Ostrom


Finlayson, A. C., and Bonnie J. McCay. 1998. ˜˜Crossing the Threshold of Ecosystem Resil-
ience: The Commercial Extinction of Northern Cod.™™ In Linking Social and Ecological Sys-
tems: Management Practices and Social Mechanisms for Building Resilience. Eds. Fikret Berkes
and Carl Folke. Cambridge: Cambridge University Press.
Frey, Bruno S. 1994. ˜˜How Intrinsic Motivation Is Crowded Out and In.™™ Rationality and
Society 6:334“352.
”””. 1997. Not Just for the Money: An Economic Theory of Personal Motivation. Chelten-
ham, England: Edward Elgar.
Frey, Bruno S., and Matthias Benz. 2001. ˜˜Motivation Transfer Effect.™™ Working paper.
¨ ¨
Zurich: University of Zurich, Institute for Empirical Research in Economics.

Frey, Bruno S., and Reto Jegen. 2001. ˜˜Motivation Crowding Theory: A Survey of Empir-
ical Evidence.™™ Journal of Economic Surveys 15:589“611.
Frey, Bruno S., and F. Oberholzer-Gee. 1997. ˜˜The Cost of Price Incentives: An Empirical
Analysis of Motivation Crowding Out.™™ American Economic Review 87:746“755.
Frohlich, Norman, and Joe A. Oppenheimer. 2003. ˜˜Optimal Policies and Socially Ori-
ented Behavior: Some Problematic Effects of an Incentive Compatible Device.™™ Public
Choice 117:273“293.
Gibson, Clark C. 1999. Politicians and Poachers: The Political Economy of Wildlife Policy in
Africa. New York: Cambridge University Press.
Gibson, Clark C., Margaret McKean, and Elinor Ostrom, eds. 2000. People and Forests:
Communities, Institutions, and Governance. Cambridge, MA: MIT Press.
Hardin, Garrett. 1968. ˜˜The Tragedy of the Commons.™™ Science 162:1243“1248.
Hess, Charlotte. 1999. A Comprehensive Bibliography of Common Pool Resources. (CD-ROM)
Bloomington: Indiana University, Workshop in Political Theory and Policy Analysis.
Isaac, R. Mark, and James M. Walker. 1988. ˜˜Group Size Effects in Public Goods Provi-
sion: The Voluntary Contributions Mechanism.™™ Quarterly Journal of Economics 103:179“
199.

Isaac, R. Mark, James M. Walker, and Susan Thomas. 1984. ˜˜Divergent Evidence on Free
Riding: An Experimental Examination of Some Possible Explanations.™™ Public Choice
43:113“149.
Isaac, R. Mark, James M. Walker, and Arlington Williams. 1994. ˜˜Group Size and the Vol-
untary Provision of Public Goods: Experimental Evidence Utilizing Large Groups.™™ Jour-
nal of Public Economics 54:1“36.
Joshi, Neeraj N., Elinor Ostrom, Ganesh P. Shivakoti, and Wai Fung Lam. 2000. ˜˜Institu-
tional Opportunities and Constraints in the Performance of Farmer-Managed Irrigation
Systems in Nepal.™™ Asia-Paci¬c Journal of Rural Development (APJORD) 10:67“92.
Kaboolian, L., and B. J. Nelson. 1998. ˜˜Creating Organizations of Concord: Lessons for
Collective Goods Theory.™™ Working paper. Berkeley: University of California, School of
Public Policy and Social Research.
Keohane, Robert O., and Elinor Ostrom, eds. 1995. Local Commons and Global Interdepen-
dence: Heterogeneity and Cooperation in Two Domains. London: Sage.
Policies That Crowd out Reciprocity and Collective Action 273


Kirchler, E., Ernst Fehr, and R. Evans. 1996. ˜˜Social Exchange in the Labor Market: Reci-
procity and Trust versus Egoistic Money Maximization.™™ Journal of Economic Psychology
17:313“341.
Lam, Wai Fung. 1996. ˜˜Institutional Design of Public Agencies and Coproduction: A
Study of Irrigation Associations in Taiwan.™™ World Development 24(6):1039“1054.
”””. 1998. Governing Irrigation Systems in Nepal: Institutions, Infrastructures, and Collec-
tive Action. Oakland, CA: ICS Press.
”””. Forthcoming. ˜˜Reforming Taiwan™s Irrigation Associations: Getting the Nesting
of Institutions Right.™™ In Asian Irrigation in Transition: Responding to Challenges. Eds.
Ganesh Shivakoti, Douglas Vermillion, Wai Fung Lam, Elinor Ostrom, Ujjwal Pradhan,
and Robert Yoder. New Delhi: Sage Publications India.
Lazear, Edward P. 2000. ˜˜Performance Pay and Productivity.™™ American Economic Review
90(5):1346“1361.
Ledyard, John O. 1995. ˜˜Is There a Problem with Public Goods Provision?™™ In The Hand-
book of Experimental Economics. Eds. John Kagel and Alvin Roth. Princeton, NJ: Princeton
University Press, 111“194.
Levine, Gilbert. 1977. ˜˜Management Components in Irrigation System Design and Oper-
ation.™™ Agricultural Administration 4:37“48.
Loveman, M. 1998. ˜˜High-Risk Collective Action: Defending Human Rights in Chile,
Uruguay, and Argentina.™™ American Journal of Sociology 104:477“525.

Low, Bobbi S., Elinor Ostrom, Robert Costanza, and James Wilson. 2001. ˜˜Human-
Ecosystems Interactions: A Basic Dynamic Integrated Model.™™ In Institutions, Ecosystems,
and Sustainability. Eds. Robert Costanza, Bobbi S. Low, Elinor Ostrom, and James Wilson.
New York: Lewis Publishers, 33“57.
McGinnis, Michael, ed. 1999a. Polycentric Governance and Development: Readings from the
Workshop in Political Theory and Policy Analysis. Ann Arbor: University of Michigan Press.
”””, ed. 1999b. Polycentricity and Local Public Economies: Readings from the Workshop in
Political Theory and Policy Analysis. Ann Arbor: University of Michigan Press.
”””, ed. 2000. Polycentric Games and Institutions: Readings from the Workshop in Political
Theory and Policy Analysis. Ann Arbor: University of Michigan Press.
Moore, Mike. 1989. ˜˜The Fruits and Fallacies of Neoliberalism: The Case of Irrigation Pol-
icy.™™ World Politics 17(1):733“750.

Offerman, T. 1997. Beliefs and Decision Rules in Public Goods Games: Theory and Experiments.
Dordrecht, the Netherlands: Kluwer.

Olson, Mancur. 1965. The Logic of Collective Action: Public Goods and the Theory of Groups.
Cambridge, MA: Harvard University Press.
Ostrom, Elinor. 1990. Governing the Commons: The Evolution of Institutions for Collective
Action. New York: Cambridge University Press.
”””. 1998. ˜˜Self-Governance of Common-Pool Resources.™™ In The New Palgrave Dictio-
nary of Economics and the Law, vol. 3. Ed. Peter Newman. London: Macmillan Press, 424“
433.
274 Ostrom


”””. 1999. ˜˜Coping with Tragedies of the Commons.™™ Annual Review of Political Science
2:493“535.
”””. 2000. ˜˜Collective Action and the Evolution of Social Norms.™™ Journal of Economic
Perspectives 14(3):137“158.
Ostrom, Elinor, and James Walker, eds. 2003. Trust and Reciprocity: Interdisciplinary
Lessons from Experimental Research. New York: Russell Sage Foundation.
Ostrom, Elinor, Roy Gardner, and James M. Walker. 1994. Rules, Games, and Common-Pool
Resources. Ann Arbor: University of Michigan Press.
Ostrom, Elinor, James M. Walker, and Roy Gardner. 1992. ˜˜Covenants with and without
a Sword: Self-Governance is Possible.™™ American Political Science Review 86:404“417.
Ostrom, Vincent. 1987. The Political Theory of a Compound Republic: Designing the American
Experiment. 2d rev. ed. Oakland, CA: ICS Press.
”””. 1997. The Meaning of Democracy and the Vulnerability of Democracies: A Response to
Tocqueville™s Challenge. Ann Arbor: University of Michigan Press.
Plott, Charles R. 1979. ˜˜The Application of Laboratory Experimental Methods to Public
Choice.™™ In Collective Decision Making: Applications from Public Choice Theory. Ed. Clifford
S. Russell. Baltimore, MD: Johns Hopkins University Press, 137“160.
Poteete, Amy, and Elinor Ostrom. 2004. ˜˜Heterogeneity, Group Size, and Collective
Action: The Role of Institutions in Forest Management.™™ Development and Change 35:435“
461.
Repetto, Robert. 1986. Skimming the Water: Rent-seeking and the Performance of Public Irriga-
tion Systems. Research report no. 4. Washington, DC: World Resources Institute.
Rothstein, B. 1998. Just Institutions Matter: The Moral and Political Logic of the Universal Wel-
fare State. Cambridge: Cambridge University Press.
Schlager, Edella. 1994. ˜˜Fishers™ Institutional Responses to Common-Pool Resource
Dilemmas.™™ In Rules, Games, and Common-Pool Resources. Eds. Elinor Ostrom, Roy
Gardner, and James M. Walker. Ann Arbor: University of Michigan Press, 247“265.

Schlager, Edella, and Elinor Ostrom. 1992. ˜˜Property-Rights Regimes and Natural
Resources: A Conceptual Analysis.™™ Land Economics 68:249“262.
Sen, Amartya K. 1974. ˜˜Choice, Orderings and Morality.™™ In Practical Reason: Papers and
¨
Discussions. Ed. S. Korner. Oxford: Blackwell.
”””. 1977. ˜˜Rational Fools: A Critique of the Behavioral Foundations of Economic
Theory.™™ Philosophy and Public Affairs 6:317“344.
Shivakoti, Ganesh P., and Elinor Ostrom, eds. 2002. Improving Irrigation Governance and
Management in Nepal. Oakland, CA: ICS Press.
Smith, Vernon L. 1982. ˜˜Microeconomic Systems as an Experimental Science.™™ American
Economic Review 72:923“955.
Snaveley, Keith. 1990. ˜˜Governmental Policies to Reduce Tax Evasion: Coerced Behavior
versus Services and Values Development.™™ Policy Sciences 23:57“72.
Tang, Shui Yan. 1992. Institutions and Collective Action: Self Governance in Irrigation. Oak-
land, CA: ICS Press.
Policies That Crowd out Reciprocity and Collective Action 275


Titmuss, R. M. 1970. The Gift Relationship. London: Allen and Unwin.
Varughese, George, and Elinor Ostrom. 2001. ˜˜The Contested Role of Heterogeneity in
Collective Action: Some Evidence from Community Forestry in Nepal.™™ World Develop-
ment 29(5):747“765.
Wade, Robert. 1995. ˜˜The Ecological Basis of Irrigation Institutions: East and South Asia.™™
World Development 23(12):2041“2049.
Wilson, James, Robert Costanza, Bobbi S. Low, and Elinor Ostrom. 2001. ˜˜Scale Misper-
ceptions and the Spatial Dynamics of a Social-Ecological System.™™ In Institutions, Ecosys-
tems, and Sustainability. Eds. Robert Costanza, Bobbi S. Low, Elinor Ostrom, and James
Wilson. New York: Lewis Publishers, 59“75.

Wunsch, James S., and Dele Olowu, eds. 1995. The Failure of the Centralized State: Institu-
tions and Self-Governance in Africa. 2d ed. Oakland, CA: ICS Press.
Yamagishi, Toshio. 1986. ˜˜The Provision of a Sanctioning System as a Public Good.™™ Jour-
nal of Personality and Social Psychology 51:110“116.
10 Reciprocity and the
Welfare State

Christina M. Fong, Samuel
Bowles, and Herbert Gintis




A man ought to be a friend to his friend and repay gift with gift. People should meet
smiles with smiles and lies with treachery.
From the Edda, a thirteenth-century collection of Norse epic verse



10.1 Introduction

The modern welfare state is a remarkable human achievement. In the
world™s advanced economies, a substantial fraction of total income is
regularly transferred from the better off to the less well off, and the
governments that preside over these transfers are regularly endorsed
by publics (Atkinson 1999). The modern welfare state is thus the most
signi¬cant case in human history of a voluntary egalitarian redistribu-
tion of income among total strangers. What accounts for its popular
support?
We suggest below that a compelling case can be made that people
support the modern welfare state because it conforms to deeply held
norms of reciprocity and conditional obligations to others. Economists
have for the most part offered an alternative (empirically implausible)
theory of self-regarding human motivation to explain who votes for re-
distribution. The most widely accepted model of the demand for redis-
tribution in economics is the median voter model, which holds that each
voter desires a personal wealth-maximizing level of redistribution. Un-
der appropriate assumptions, it follows that the redistribution imple-
mented by a government elected under a majority rule system is that
preferred by the median-income voter. Because the distribution of in-
come is generally skewed to the right (there are a few very rich indi-
viduals), the median voter is poorer than the mean voter and will
therefore demand a positive level of redistribution.
278 Fong, Bowles, and Gintis



An important implication of this model is that demand for redis-
tribution decreases as personal income increases (Roberts 1977). But
personal income is a surprisingly poor predictor of support for redistri-
bution (Gilens 1999; Fong 2001). A large fraction of the poor oppose in-
come redistribution and a large fraction of the rich support it. Among
respondents of a nationally representative American survey (Gallup
Organization 1998) who have annual household incomes of at least
$150,000 and expect their lives to improve in the next ¬ve years, 24
percent responded that the government should ˜˜redistribute wealth by
heavy taxes on the rich,™™ and 67 percent respond that the ˜˜government
in Washington, DC, should make every possible effort to improve the
social and economic position of the poor.™™ Equally striking is the fact
that among the respondents with annual family incomes of less than
$10,000 who did not expect to be better off in ¬ve years, 32 percent re-
port that the government should not redistribute wealth by heavy taxes
on the rich, and 23 percent say that the poor should help themselves
rather than having the government ˜˜make every possible effort to im-
prove the . . . position of the poor.™™1
Thus, while self-interest is an important human motive, and income
does explain some of the variance in redistributive attitudes, other
motives appear to be at work. Abundant evidence from across the so-
cial sciences”much of it focusing on the United States with similar
¬ndings in smaller quantities from other countries around the world”
has shown that when people blame the poor for their poverty, they
support less redistribution than when they believe that the poor are
poor through no fault of their own. That is, generosity toward the
poor is conditional on the belief that the poor work hard (Williamson
1974; Heclo 1986; Farkas and Robinson 1996; Gilens 1999; Miller 1999).
For instance, in a 1972 sample of white women in Boston, the per-
ceived work ethic of the poor was a far better predictor of support for
aid to the poor than one™s family income, religion, education, and a
host of other demographic and social background variables (William-
son 1974). Indeed in predicting support for such aid, the addition of a
single variable measuring beliefs about work motivation tripled the
explanatory power of all the above background variables together.
Mof¬tt, Ribar, and Wilhelm (1998) were among the ¬rst economists to
report ¬ndings on this relationship. They used the General Social Sur-
vey, a large nationally representative data set with observations in
nearly every year since 1972 to show that those who believe that peo-
ple get ahead by ˜˜lucky breaks or help from others™™ rather than hard
Reciprocity and the Welfare State 279



work prefer more spending on welfare. Fong (2001) used nationally
representative data from a 1998 Gallup Social Audit to show that the
effects of beliefs about the causes of income on demands for redistribu-
tion are surprisingly large and cannot be explained by missing mea-
sures of self-interest. Alesina, Glaeser, and Sacerdote (2001) have
reported related ¬ndings from the World Values Survey. Americans
have much stronger beliefs that poverty is caused by laziness than
Europeans: 60 percent of Americans say the poor are lazy compared to
just 27 percent of Europeans. The authors argue that this could be an
important explanation for the small size of the American welfare state
compared to the average European welfare state.
Our interpretation of these ¬ndings is that people are willing to help
the poor, but they withdraw support when they perceive that the poor
cheat or fail to cooperate by not trying hard enough to be self-suf¬cient
and morally upstanding. Within economics, our view is most similar to
the taxpayer resentment view of the demand for redistribution mod-
eled by Besley and Coate (1992) and the effect of reciprocity sentiments
on redistributive public ¬nance by Serge Kolm (1984).2 Our view is
also consistent with interpretations by Heclo (1986) and Gilens (1999),
who cite evidence that Americans support a wide array of bene¬ts for
the poor and are primarily opposed to ˜˜welfare,™™ presumably because
˜˜welfare™™ refers to means-tested cash assistance, which may be per-
ceived as a program that bene¬ts able-bodied adults who choose to
have children out of wedlock and prefer not to work. Our interpreta-
tion is also compatible with equity theory and attribution theory.
According to equity theory, people should receive resources from a
system that are proportional to their contributions (Walster, Walster,
and Bersheid 1978; Deutsch 1985; Miller 1999). Attribution theorists ar-
gue that people are less likely to help someone if they determine that
the person is individually responsible for his or her outcome (Weiner
1995; Skitka and Tetlock 1993).
Economists have been skeptical of non-sel¬sh models for several
reasons. First, there could be unmeasured self-interest variables that
explain the support for redistribution. In particular, those with low-
mean, high-variance incomes may be more likely to think that poverty
is due to bad luck and also more likely to demand redistribution out of
self-interest for insurance against a low income. We soundly reject this
hypothesis in section 10.4.
Second, people who think that effort plays a major role in income
generation may be concerned about the incentive effects of taxation or
280 Fong, Bowles, and Gintis



transfers rather than the ˜˜worthiness™™ of recipients (Piketty 1995). We
do have two pieces of evidence, however, that incentive costs cannot
fully explain attitudes towards redistribution. One is that were incen-
tive costs of taxation the problem, those who believe that effort is im-
portant should support less government spending in general. Yet, as
we show in section 10.4, the belief that effort is important to getting
ahead in life is negatively correlated with support for redistribution
and positively correlated with support for military spending. Another
is that, as described in section 10.3, subjects in a behavioral experi-
ment on charitable giving to welfare recipients gave signi¬cantly more
money when they were randomly paired with a welfare recipient
who said she would like to work than when randomly paired with a
welfare recipient who said she would not like to work. There were no
disincentive costs in this experiment, so some other interpretation is
necessary.
This experimental result also addresses a third concern that econo-
mists have raised: People who do not want to give to the poor may
say that the poor are lazy to justify their sel¬shness. This cannot
explain why randomly assigned treatment conditions in the charity
experiment just described had signi¬cant effects on giving to welfare
recipients.
Concern about the ˜˜undeserving poor™™ is pronounced in the United
States, but is far from absent in Europe. In ¬gure 10.1 we show that in
twelve European countries, those who say that poverty is the result of
laziness support less government redistribution and are less concerned
about unemployment, poverty, and inequality than those who do not.
The data are from a Eurobarometer survey conducted in 1989 (Reif
and Melich 1993), representative of the population aged ¬fteen and
over in the twelve European Union countries of that time. Of the data
set™s 11,819 respondents, we use the 8,239 who answered all of the
questions included in our analysis. Our dependent variable is the sum
of responses to four questions about

(1) the importance of ¬ghting unemployment;
(2) the importance of ¬ghting poverty;
(3) the importance of reducing differences between regions within the
country by helping regions that are less developed or in dif¬culties; and
(4) whether the public authorities in the country do all that they
should for poor people.
Reciprocity and the Welfare State 281




(-0.034**)
Income
(-0.067***)
-0.1 Quartile
Male



-0.2
Regression Coefficient




-0.3




-0.4

(-0.42***)
Believes Poverty is
Due to Laziness
-0.5

Figure 10.1
Explaining concern about poverty using data from twelve European countries.
Note: Bars represent ordinary least squares coef¬cients (value of the estimated coef¬cient
is in parentheses) predicting concern about poverty. The dependent variable is standar-
dized so that the estimated coef¬cient represents the effect of the variable indicated on
concern about poverty measured in standard deviation units. The equation also includes
age and country dummy variables. Signi¬cance levels are based on robust standard
errors that allow for clustered errors within countries. This regression uses sample
weights, although the results are not sensitive to them. There are 8239 observations,
R 2 ¼ :161. *** Signi¬cant at the 1 percent level. ** Signi¬cant at the 5 percent level.


The measure increases in concern about poverty, unemployment, in-
equality, and the belief that the public authorities do not ˜˜do enough
for poor people.™™ For simplicity, we refer to this composite measure as
˜˜concern about poverty.™™
Our independent variable of primary interest is the belief that pov-
erty is caused by laziness rather than being caused by bad luck, in-
justice, or no reason at all, or that poverty is inevitable.3 The other
variables included in the regression are family income quartiles, sex,
and age. Note that item 4 in our dependent variable is explicitly coun-
try speci¬c. Cross-country comparisons of a question like this are of lit-
tle value because people in a country with a generous redistribution
system may care very much about poverty but believe that their own
government is doing a good job of addressing it. The other three items
used to construct our dependent measure are subject to the same con-
cern, albeit to a lesser extent. To account for the effects of unmeasured
282 Fong, Bowles, and Gintis



differences between countries, we use ¬xed effects to allow for country
differences in mean responses.
The results, presented in ¬gure 10.1, show that those who say that
poverty is caused by laziness are less concerned about poverty than
the rest of the respondents by a 0.42 standard deviation. In contrast,
family income has a very modest effect.4 The difference in concern
about poverty between the richest and poorest quartiles is less than a
quarter as great as the difference between those who think that poverty
is due to laziness and those who do not. The respondent™s sex has a
signi¬cant effect on concern about poverty (independent of income
and the other regressors), with men being less concerned than women.
We do not doubt that self-regarding motives often underpin appar-
ently generous actions. Rather, we suggest that they do not always do
so. Understanding egalitarian politics today requires a reconsideration
of Homo economicus, the unremittingly self-regarding actor of economic
theory. We do not wish to replace the textbook self-regarding actor,
however, with an equally one-dimensional altruistic actor willing to
make unconditional and personally costly contributions to the less
well off. Rather, we believe that strong reciprocity better explains the
motivations behind support for the welfare state. By ˜˜strong reciproc-
ity,™™ we mean a propensity to cooperate and share with others simi-
larly disposed, even at personal cost, and a willingness to punish those
who violate cooperative and other social norms, even when punishing
is personally costly and cannot be expected to entail net personal gains
in the future (see chapter 1).
Strong reciprocity goes considerably beyond self-interested forms
of cooperation, which we term ˜˜weak reciprocity™™ and which include
market exchange and tit-for-tat behavior”what biologists call ˜˜recip-
rocal altruism.™™
As we will see, all three of our personae”Homo economicus, the
strong reciprocator, and even the pure altruist”are represented in
most groups of any size. For this reason, egalitarian policymaking, no
less than the grand projects of constitutional design, risks irrelevance
if it ignores the irreducible heterogeneity of human motivations. The
problem of institutional design is not, as the classical economists
thought, that uniformly self-regarding individuals be induced to inter-
act in ways producing desirable aggregate outcomes. Instead, it is that
a mix of motives”self-regarding, reciprocal, and altruistic”interact in
ways that prevent self-regarding individuals from exploiting generous
individuals and unraveling cooperation when it is bene¬cial.
Reciprocity and the Welfare State 283



10.2 The Origins of Reciprocity

Earlier chapters of this book presented the experimental evidence for
strong reciprocity. Historical evidence also supports the notion that
support for redistribution is often based on strong reciprocity motives.
In his Injustice: The Social Bases of Obedience and Revolt, Barrington
Moore Jr. (1978, 21) sought the common motivational bases”˜˜general
conceptions of unfair and unjust behavior™™”for the moral outrage
fueling struggles for justice that have recurred throughout human
history. ˜˜There are grounds,™™ he concludes from his wide-ranging
investigation
For suspecting that the welter of moral codes may conceal a certain unity of
original form . . . a general ground plan, a conception of what social relation-
ships ought to be. It is a conception that by no means excludes hierarchy and
authority, where exceptional qualities and defects can be the source of enor-
mous admiration and awe. At the same time, it is one where services and
favors, trust and affection, in the course of mutual exchanges, are ideally
expected to ¬nd some rough balancing out. (Moore 1978, 4“5, 509)

Moore termed the general ground plan he uncovered ˜˜the concept of
reciprocity”or better, mutual obligation, a term that does not imply
equality of burdens or obligations . . .™™ (Moore 1978, 506) In similar
manner, James Scott (1976) analyzed agrarian revolts, identifying vio-
lations of the ˜˜norm of reciprocity™™ as one the essential triggers of in-
surrectionary motivations.
Casual observation of everyday life, ethnographic and paleoanthro-
pological accounts of hunter-gatherer foraging bands from the late
Pleistocene to the present, and historical narratives of collective strug-
gles have combined to convince us that strong reciprocity is a powerful
and ubiquitous motive.

10.3 Experimental Evidence on Unilateral Income Transfers

Behavioral experiments with human subjects provide overwhelming
evidence against Homo economicus. Many of these experiments have
been described in chapters 1 and 5 of this book, and will not be re-
peated here. However, there is additional evidence dealing more di-
rectly with charitable redistribution. Consider, for instance, the dictator
game. In this game, one of two mutually anonymous players, the ˜˜pro-
poser,™™ is given a sum of money (typically $10), asked to choose any
part of the sum to give to the second player, and permitted to keep the
284 Fong, Bowles, and Gintis



rest. Homo economicus gives nothing in this situation, whereas in actual
experimental situations, a majority of proposers give positive amounts,
typically ranging from 20 percent to 60 percent of the total (Forsythe
et al. 1994).
Using dictator games, researchers have shown that people are more
generous to individuals they perceive to be worthy recipients and bar-
gaining partners. For example, Eckel and Grossman (1996) found that
subjects in dictator games gave roughly three times as much when the
recipient was the American Red Cross than when it was an anony-
mous subject. More recently, Fong (2004) conducted charity games (n-
donor dictator games) in which several dictators were paired with a
single real-life welfare recipient. The treatment conditions were ran-
domly assigned and differed according to whether the welfare recipi-
ent expressed strong or weak work preferences on a survey that she
completed. Dictators read the welfare recipients™ surveys just prior to
making their offers. Dictators who were randomly assigned to welfare
recipients who expressed strong work preferences gave signi¬cantly
more than dictators who were assigned to recipients that expressed
weak work preferences. These experiments provide evidence for our
view that strong reciprocity is a common motivation.
Another result that is consistent with reciprocity is that cooperating
and punishing behavior are very sensitive to the situation framing the
interaction. In early research on what is known as inequality aversion,
Loewenstein, Thompson, and Bazerman (1989) found that distribu-
tional preferences are sensitive to social context. They asked subjects to
imagine themselves in various hypothetical situations. In one, the sub-
ject and another college student share the gains and losses from a joint-
ly produced product. In another, the subject and a neighbor split the
pro¬t from selling a vacant lot between their homes. In a third, the sub-
ject is a customer dividing the proceeds from an expired rebate or the
cost of repairs with a salesperson.
These scholars found that subjects care about relative payoffs even
more than they care about their absolute payoffs. They also found
that, controlling for the subjects™ own payoffs, earning less than the
other person had a strong negative effect on utility in all situations
and relationship types. However, an effect on utility of earning more
than the other person (referred to as advantageous inequality) was
also present and depended on the relationship and the situation.
Subjects disliked advantageous inequality if the relationship was
friendly. However, if the relationship was unfriendly, advantageous
Reciprocity and the Welfare State 285



inequality had little effect on their satisfaction level. Interestingly, these
researchers found that subjects preferred advantageous inequality in
the customer/salesperson scenario, but disliked it in the other two
scenarios (producing a product and splitting the proceeds from an
empty lot).
Although there may be many additional factors contributing to the
context dependence of behavior, the ¬nding that subjects are more
averse to advantageous inequality (or, equivalently, desire higher
relative payoffs for the other subject) in friendly relationships than in
unfriendly relationships is fully consistent with our interpretation of
reciprocity. In another example, fraternity brothers at the University
of California“Los Angeles (UCLA) were asked to rank outcomes in a
Prisoner™s Dilemma situation under ¬ve different scenarios: interacting
with a fellow fraternity brother, a member of another (unnamed) fra-
ternity, a non-fraternity student at UCLA, a student from the nearby ri-
val University of Southern California, and an of¬cer from the UCLA
Police Department. They showed a strong preference for mutual coop-
eration over defection against one™s partner when playing with fra-
ternity brothers, with the rankings reversing with increasing social
distance”they were as willing to exploit the University of Southern
California students as the UCLA police (Kollock 1997)!

10.4 Survey Evidence

These results support our interpretation of attitudinal survey results,
which show that people support more government redistribution to
the poor if they think that poverty is caused by bad luck rather than
laziness. Our interpretation of this is that because of strong reciprocity,
people wish to help those who try to make it on their own, but for rea-
sons beyond their own control cannot. People wish to punish”or
withhold assistance from”those who are able but unwilling to work
hard. However, there are several alternative explanations for the effect
of beliefs about the worthiness of the poor that are consistent with pure
self-interest. In this section, we test these alternative explanations and
¬nd that self-interest alone cannot explain the relationship between
beliefs about the worthiness of the poor and support for redistribution.
These results are based on Fong (2001).
We use the 1998 Gallup Poll Social Audit Survey, ˜˜Haves and Have-
Nots: Perceptions of Fairness and Opportunity,™™ a randomly selected
national sample of 5,001 respondents. In each test, we use the set of all
286 Fong, Bowles, and Gintis



individuals who responded to all of the questions used in the regres-
sion, unless noted otherwise.5
Relative to other commonly used surveys, the Gallup survey has a
large sample size for a large number of questions on inequality and
distribution. The sample size permits running regressions with full
controls on narrow segments of the sample”namely, high income and
low income sub-samples. There are several self-interest measures that
include not only the usual objective socioeconomic variables, but also
subjective measures of economic well-being and future expectations.
These may widen the net intended to capture self-interest.
To construct our dependent variable, we added the responses to the
¬ve questions below, signing the responses so that the measure in-
creases in support for redistribution.

1. People feel differently about how far a government should go. Here
is a phrase which some people believe in and some don™t. Do you think
our government should or should not redistribute wealth by heavy
taxes on the rich? (response categories: should, should not)
2. Some people feel that the government in Washington, DC, should
make every possible effort to improve the social and economic position
of the poor. Others feel that the government should not make any spe-
cial effort to help the poor, because they should help themselves. How
do you feel about this? (response categories: government should help
the poor, the poor should help themselves)
3. Which one of the following groups do you think has the greatest
responsibility for helping the poor: churches, private charities, the gov-
ernment, the families and relatives of poor people, the poor them-
selves, or someone else? (response categories: groups other than the
poor, the poor themselves)
4. Do you feel that the distribution of money and wealth in this coun-
try today is fair, or do you feel that the money and wealth in this coun-
try should be more evenly distributed among a larger percentage of
the people? (response categories: distribution is fair, should be more
evenly distributed)
5. Do you think that the fact that some people in the United States are
rich and others are poor (1) represents a problem that needs to be ¬xed,
or (2) is an acceptable part of our economic system? (response catego-
ries: problem, acceptable)

Two sets of measures of the causes of income are used in this study.
The ¬rst contains two questions concerning the importance of effort
Reciprocity and the Welfare State 287



and luck in causing wealth and poverty and one question on whether
or not there is plenty of opportunity to work hard and get ahead in
America today. The second set is a series of questions about the impor-
tance of various factors, including race and sex, for getting ahead in
life.
Self-interest is measured by income and other variables likely to
predict current and future tax obligations and current and future reli-
ance on social insurance or redistribution programs. In ¬gures 10.2
and 10.3, we control for self-interest by including income, race, sex,


Bad Luck
Causes
Poverty Luck
0.6 Luck
Causes Lack of
(0.50) and
Luck and
Wealth Effort
Effort
(0.39) Cause
0.4 Cause
Wealth
Poverty
(0.27)
(0.23)
0.2
Regression Coefficient




0.0



-0.2
(-0.18)
(-0.25)
Never (-0.27)

<<

. 9
( 13)



>>