Cambridge Journal of Economics 2011, 35, 705–728 doi:10.1093/cje/beq049 Advance Access publication 5 January 2011 Behavioural and experimental economics: are they really transforming economics? Ana C. Santos* Behavioural and experimental economics are part of an increasingly pluralistic mainstream economics, sharing with other recently established research pro- grammes the revision of fundamental assumptions of the previously dominant neoclassical economics research programme. The recent proliferation and consol- idation of these new approaches creates the possibility for the emergence of a new orthodoxy of economics, i.e. a new general research programme capable of replacing neoclassicism. The goal of this paper is to investigate the potential contribution of behavioural and experimental economics to help build a general research pro- gramme capable of supplanting neoclassical economics and thereby transforming economics. To this end, it focuses on two influential applied fields of behavioural and experimental economics—choice architecture and design economics. Key words: Anomalies, Choice architecture, Design economics, Market design, Recent economics JEL classifications: A12, B52, C90 1. Introduction Behavioural and experimental economics are part of recent mainstream economics, sharing with other emergent research programmes the rejection of fundamental assump- tions and commitments of the previously dominant neoclassical economics research programme. The recent proliferation and consolidation of the new approaches, with overlapping areas of research and concerns, is now raising the possibility for the emergence of a new orthodoxy of economics, i.e. a new general research programme capable of replacing neoclassicism (Davis, 2006, 2008). The goal of this paper is to investigate the potential contribution of behavioural and experimental economics to transform econom- ics. I focus, in particular, on their capacity to carry out two major transformations that are deemed as characterising the ongoing process of change in the discipline: (i) the revision of the neoclassical economics model of human action, homo economicus; and (ii) the new economic approach to market building. In order to do this, I look at two recent applications Manuscript received 11 October 2009; final version received 17 November 2010. Address for correspondence: CES, Center for Social Studies, University of Coimbra, Cole ´gio de S. Jero ´nimo, Apartado 3087, 3001-401 Coimbra, Portugal; email: [email protected]* University of Coimbra, Portugal. I acknowledge financial support from Fundac xa ˜o Calouste Gulbenkian (n° 21-107001-S). I would also like to thank the comments of Joa ˜o Rodrigues and Jose ´ Castro Caldas and those of three anonymous referees. Usual disclaimers naturally apply. Ó The Author 2011. Published by Oxford University Press on behalf of the Cambridge Political Economy Society. All rights reserved. by Robert Looney on July 2, 2011 cje.oxfordjournals.org Downloaded from
24
Embed
Economia del comportamiento transformando a la economía
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Cambridge Journal of Economics 2011, 35, 705–728doi:10.1093/cje/beq049Advance Access publication 5 January 2011
Behavioural and experimental economics:are they really transforming economics?
Ana C. Santos*
Behavioural and experimental economics are part of an increasingly pluralisticmainstream economics, sharing with other recently established research pro-grammes the revision of fundamental assumptions of the previously dominantneoclassical economics research programme. The recent proliferation and consol-idation of these new approaches creates the possibility for the emergence of a neworthodoxy of economics, i.e. a new general research programme capable of replacingneoclassicism. The goal of this paper is to investigate the potential contribution ofbehavioural and experimental economics to help build a general research pro-gramme capable of supplanting neoclassical economics and thereby transformingeconomics. To this end, it focuses on two influential applied fields of behavioural andexperimental economics—choice architecture and design economics.
Behavioural and experimental economics are part of recent mainstream economics,
sharing with other emergent research programmes the rejection of fundamental assump-
tions and commitments of the previously dominant neoclassical economics research
programme. The recent proliferation and consolidation of the new approaches, with
overlapping areas of research and concerns, is now raising the possibility for the emergence
of a new orthodoxy of economics, i.e. a new general research programme capable of
replacing neoclassicism (Davis, 2006, 2008). The goal of this paper is to investigate the
potential contribution of behavioural and experimental economics to transform econom-
ics. I focus, in particular, on their capacity to carry out two major transformations that are
deemed as characterising the ongoing process of change in the discipline: (i) the revision of
the neoclassical economics model of human action, homo economicus; and (ii) the new
economic approach to market building. In order to do this, I look at two recent applications
Manuscript received 11 October 2009; final version received 17 November 2010.Address for correspondence: CES, Center for Social Studies, University of Coimbra, Colegio de S. Jeronimo,
* University of Coimbra, Portugal. I acknowledge financial support from Fundacxao Calouste Gulbenkian(n� 21-107001-S). I would also like to thank the comments of Joao Rodrigues and Jose Castro Caldas andthose of three anonymous referees. Usual disclaimers naturally apply.
� The Author 2011. Published by Oxford University Press on behalf of the Cambridge Political Economy Society.
to help individuals make better choices, as judged by individuals themselves, or by society
as a whole (Thaler and Sunstein, 2003, 2008). Design economics is in turn devoted to the
conception of specific allocation mechanisms that aim at coordinating individual actions
for the accomplishment of the goals set by the designer (Roth, 2002). Rather than
assuming that markets emerge spontaneously and automatically generate efficient
allocations of resources, design economics puts at the forefront the complex social
engineering processes involved in the building of markets and market-like allocation
mechanisms that determine individual outcomes and the aggregate results that are
obtained by having people interacting under those mechanisms.
Yet it is not clear that the reform of the core principles of economics is under way. Both
choice architecture and design economics seem to retain two fundamental principles of
neoclassical economics—rationality and efficiency. Having observed that economic agents
are not always rational and that their actions in market and non-market contexts do not
always produce desirable outcomes, economists seem now to be dedicated to the
fabrication of the conditions of rationality and efficiency. And there is a clear division of
labour. While architect-economists are devoted to the creation of contexts of choice that
make rational choices viable, designer-economists are devoted to the construction of
market mechanisms that ensure socially efficient outcomes. Thus, while importing new
models of individual decision-making and new experimental technologies, choice
architecture and design economics still seem to retain the neoclassical view of rational
choice, defined in terms of the choice that confers the individual the highest net benefit
given constraints, and of the market understood as a mere allocation mechanism. This
might in fact explain the success of choice architects and design economists as policy
advisors. Notwithstanding the ongoing process of change in economics, economic
discourse and general perceptions about economics may lag far behind. Policy proposals
that appeal to the standard normative notions of rationality and efficiency might thus still
be well-received by policy-makers and the general public. However that may be, the fact of
the matter is that Richard Thaler, Cass Sunstein and Alvin Roth are all accomplished and
much sought after scholars to give advice and help reconfigure a variety of socioeconomic
institutions.2
If on the one hand, the rejection of key assumptions of neoclassical economics may entail
a transformative potential, the preservation of the neoclassical criteria of rationality and
efficiency, on the other hand, may suggest that neoclassical economics principles are still
very much present in the new approaches. The goal of this paper is to investigate this
tension so as to evaluate the prospects of change in economics and the emergence of a new
general research programme capable of replacing neoclassicism. Section 2 starts off by
briefly reviewing the recent transformations taking place in economics. Section 3 presents
and analyses choice architecture while Section 4 is given over to the presentation and
examination of design economics. Section 5 then discusses the tensions that arise from the
revision of the key assumptions of the neoclassical research programme that preserves its
normative criteria of rationality and efficiency. Based on the analysis of choice architecture
2 Thaler and Sunstein are from Chicago University and were famously advertised as campaign advisors tothe then Presidential candidate, Barack Obama. Sunstein is the current Administrator of the White HouseOffice of Information and Regulatory Affairs. Thaler was advisor to the leader of the British Conservativeparty, David Cameron, and continues to advise the institutional investment firm Fuller & Thaler AssetManagement, Inc., which he founded (see Chakrabortty, 2008; Lewis, 2008; Wilby, 2008). Roth is fromHarvard University and he too has been a consultant for various American public entities, such as TheNational Resident Matching Program and The New York City Department of Education.
norms when facing particular individual and collective decision-making problems. Design
economics builds instead upon the ‘technological’ experiments of economics that are
particularly tailored to study market institutions, namely their incentive compatibility
attributes, i.e. their capacity to induce self-interested individuals to take the actions that
achieve desirable results at the aggregate level (cf. Santos 2007; 2010, cap. 10).
Behavioural experiments have produced a substantial amount of evidence that shows
that human beings are prone to systematic error even in areas of economic relevance where
stakes are high (e.g. Thaler, 1992; Camerer, 1995). Rather than grounding individual
choice on the calculus of the costs and benefits of alternative options so as to choose the
alternative that provides the highest net benefit, individuals have recourse to a variety of
decisional rules and are influenced by various contextual factors that jeopardise the pursuit
of individuals’ best interests. The increased understanding of how people actually select
and apply rules for dealing with particular forms of decision problems and of the influence
of contexts on individual choices is the starting point of choice architecture devoted to the
study of choice setups that can curb human idiosyncrasies to good result, as judged by
individuals themselves, or by society as a whole (Thaler and Sunstein, 2003, 2008).
Technological experiments have produced a substantial amount of evidence of the
relative performance of various market mechanisms (e.g. Kagel, 1995). More recently,
they have been used as engineering tools for building new markets from scratch. That is,
they have been used for building ‘economic machines’, which ‘are supposed to work for
several years, in different contexts and without constant supervision of their manufacturer’
(Guala, 2001, p. 464) or as ‘testbeds’ of ‘a working prototype of a process that is going to be
employed in a complex environment’ (Plott, 1997, p. 605). These experiments have forced
economists to explicitly recognise that markets are the outcome of complex social
engineering processes that determine the rules under which individuals are to act and
the aggregate results that obtain by having economic agents interacting under these rules
(Roth, 2002). In the next two sections, I look in more detail into the policy proposals of
choice architecture and design economics so as to investigate the extent to which they
depart from neoclassical economics and thereby assess the potential of the recent
approaches to transforming economics.
3. The assumptions of neoclassical economic theory, behavioural anomalies
and choice architecture
The neoclassical economics model of human action, homo economicus, relies on two basic
assumptions—unbounded rationality and self-interest— which underlie the utility max-
imising analysis of human behaviour. Experimental work first carried out by cognitive
psychologists, and subsequently by economists, identified so-called ‘anomalies’,
i.e. patterns of judgment and choice that are inconsistent with utility maximisation,
challenging both assumptions.4 Behavioural economists have, then, attempted to replace
the standard assumptions of economics with more realistic descriptions of human
behaviour that could account for people’s bounded rationality and other-regarding
considerations. However, the revised conception of human action was not accompanied
by an equivalent change in economic theorising. While accommodating ‘anomalous’
4 Richard Thaler had an important role in introducing these results to economists in the column‘anomalies’ of the prestigious Journal of Economic Perspectives, between 1987 and 1990. His The Winner’sCurse: Paradoxes and Anomalies of Economic Life, published in 1992, brings together some of theseexperimental results. See also Camerer (1995).
those who do not wish to be part of the plan may always opt out. The authors consider that
the low rates of opt outs observed in the past suggest that most employees do benefit from
this measure, enrolling earlier than they would otherwise do, if they did so at all. These
measures would not be necessary in a fully rational world in which individuals always
choose the best option regardless of the default. But individuals’ inertia and tendency to
procrastinate more likely lead to no action. Default options have been shown to be
particularly effective, especially so if they also entail an explicit endorsement from the
default setter that it is the recommended course of action (Thaler and Sunstein, 2008,
p. 83).
The choice architect may also select the relevant information for decision-making and
design information disseminating devices that facilitate comparison of the alternative
options. This solution is proposed for decisions involving complex pricing schemes, such as
mortgages, cell phone calling plans and insurance policies. According to Thaler and
Sunstein, in these cases the government should regulate disclosure practices, requiring that
customers be informed of fees, in a legible format, and be given regular reports of the fees
that have been charged. These reports would provide consumers with feedback so that they
could shift their demand to the services that most adequately fit their consumption
patterns. Besides the design of defaults or information disclosure devices, choice architects
may have recourse to other tools to attenuate people’s propensity to err, say the framing of
the context of choice, which may be tamed to better capture the analytical structure of the
decision-problem; moreover, they may take advantage of people’s attributes, for instance,
people’s susceptibility to social influence, by making salient the prudent and beneficial
actions of others.
Consumption, saving and investment decisions constitute obvious areas for the
application of choice architecture. The high rates of household indebtedness in developed
economies (but not only) are to some extent explained by individuals’ failure to make wise
consumption/investment decisions, due to non-transparent and difficult-to-process in-
formation and to self-control problems, given that the temptation of immediate
gratification is too salient as compared with the costs of hasty/risky decisions that are
only suffered at a later time.
Matters are at present even more complicated by the highly sophisticated and complex
financial markets that have rendered credit and investment decisions increasingly more
difficult. Less educated individuals are in a particularly disadvantaged position because
they are more vulnerable to aggressive campaigns and marketing strategies deployed by
those who profit from consumers’ hasty decisions. This is explicitly recognised by Thaler
and Sunstein, when referring to mortgage markets:
When markets get more complicated, unsophisticated and uneducated shoppers will beespecially disadvantaged by the complexity. The unsophisticated shoppers are also more likelyto be given bad or self-interested advice by people serving in roles that appear to be helpful andpurely advisory. In this market, mortgage brokers who cater to rich clients probably havea greater incentive to establish a reputation for fair dealing. By contrast, mortgage brokers whocater to the poor are often more interested in making a quick buck. (Thaler and Sunstein, 2008,p. 134)
In this particular situation, Thaler and Sunstein’s proposal is to demand that mortgage
lenders provide the relevant information in a readable format so that all the fees and interest
rate provisions can be more easily compared. They argue that this will not only make it easier
to shop, but it will also allow independent third parties to offer much better advice, thus
making the mortgage market more competitive (Thaler and Sunstein, 2008, p. 138).
Choice architecture thus mobilises the recently acquired understanding of how
individuals process information and make decisions, to design contexts of choice that
help people make better choices. The choice architect may focus on the design of default
options that protect individuals from inertia or conceive information devices to help
improve people’s ability to select options that make them better off. At the same time the
designer should avoid, as much as possible, changing the incentive structure of the
problem-situation. The authors stress ‘[t]he central goal would be to inform costumers of
fees rather than set prices’ (Thaler and Sunstein, 2008, p. 83).
It is by now clear that the neoclassical models of rational choice and of the competitive
market loom in the background of choice architecture. The problem, as Thaler and Sunstein
see it, is that individuals fail to conform to the rational choice model due to incomplete
information, cognitive limitations and weak willpower. Prevalent incentives in opaque markets
may even exacerbate this problem because it can be highly profitable to exploit people’s
bounded rationality. The choice architect should then help people with decision-making so
that they more closely align their demand with the actual benefits derived from consumption.
In this way, choice architecture also contributes to bettering the functioning of competitive
markets by reducing the detrimental effects of asymmetric information between buyers and
sellers. The more consumers make the right choices, the more markets become competitive.
Thaler and Sunstein thus work within the larger framework of the neoclassical research
programme trying to deal with the problems of asymmetric information and decision-
makers’ bounded rationality. Even though they acknowledge that in many circumstances
choice architectures are unavoidably designed by someone else for others to make their
choices, they do not address this key issue to choice architecture. They merely identify
complex and difficult decision-making as obvious areas for the intervention of the
benevolent choice architect. Thus, they do not consider the possibility that some of
people’s ‘errors’ might in fact be the result, intended or unintended, of current choice
architectures. This means that choice architecture only addresses those problems caused
by incomplete information, bounded rationality and weak willpower, leaving out those that
stem from faulty choice architectures. This is patent when Thaler and Sunstein discuss
mortgage markets, disregarding the participation of the wider financial system in growing
household rates of overindebtedness, limiting their focus on the complexity of these kinds
of decision, the main problem to be fixed by choice architecture.
The work of Robert J. Shiller (2000, 2008) on financial markets illustrates the interplay
between choice architectures and human behaviour showing that understanding individual
decision-making requires understanding the institutional structure inside of which people
act. Shiller interprets the so-called subprime crisis not so much in terms of the complexity
of the decision-making process, but instead in terms of an ‘epidemic of irrational
enthusiasm for housing investments’ caused by the perverse incentives of the financial
sector. On Shiller’s view, the unsustainable levels of mortgage borrowing in the USA were
ultimately caused by dramatic institutional changes in financial markets over the last three
decades, through deregulation and, more recently, financial innovation, culminating in
what is known as the New Financial Architecture. The new institutional setting, together
with rising housing prices and low interest rates, rendered mortgage lending a highly
lucrative investment fed by a new securitisation industry.8 When house prices collapsed
8 The feedback mechanisms are well-known: mortgage brokers sold loans, investment bankers packagedthe loans into securities, banks and specialist institutions serviced the securities, rating agencies gave their sealof approval, and insurance companies protected holders of such securities against loss through the use ofcredit default swaps.
arise when applying theory to practical matters, for which end it mobilises the resources from
three emergent research programmes of economics: game theory, computational and
experimental economics. Game theory provides the general framework for addressing
design, in particular, for the definition of the ‘rules of the game’. The tools of computational
and experimental economics are instead used to ‘bridge the gap’ between theory and the
institutional details of particular markets. Computational methods, in particular, ‘will help us
analyze games that may be too complex to solve analytically’, while laboratory experiments
‘will help inform us about how people will behave when confronted with these environments,
both when they are inexperienced and as they gain experience’ (Roth, 2002, pp. 1373–4).
Design economics emerged in the 1990s when game theorists began to be hired as
consultants for the design of various allocation mechanisms that were soon to be
implemented in real world environments. The design and implementation of these new
mechanisms raised technical complications for which no theory or past experience could
suggest any solution, and new devices had to be designed from scratch. Two famous
exemplars of these allocations mechanisms are the design of labour clearinghouses for the
US National Resident Matching Program (NRMP) that would assign hospital positions to
young doctors, and the design of an auction mechanism for the US Federal Communi-
cations Commission (FCC) that would allocate licenses to use the electromagnetic
spectrum for various telecommunications services.
The NRMP clearinghouse called for the design of a matching algorithm that would
allocate applicants to residence programs based on preferences lists of both doctors and
hospitals, avoiding favouring one side of the market over the other, while producing
a stable matching (i.e. a matching on which both parties could agree). To this end, the
performances of alternative algorithms were assessed and compared in terms of the impact
on each side of the market, speed, stability and so forth. The FCC auction was intended to
allocate airwave spectrum rights in a transparent and efficient way. But while economists
were hired to design new allocation mechanisms the responsibility for selecting or deciding
whether to adopt the proposed design was retained by the public authority that hired them,
in consultation with its various constituencies.
The FCC auction provides a very useful illustration of design economics, which is worth
looking at more closely.9 The FCC auctions have been greeted as the biggest engineering
success of economics. In 1994, the FCC implemented what was to be known as the
simultaneous–multiple-round–independent auction, which would soon be praised as ‘the
greatest auction in history’ (McAfee and McMillan, 1996, p. 159). This auction launched
a market for thousands of spectrum licenses, and its success in raising billions of dollars for
the public treasury has been taken as evidence of the successful accomplishment of the
goals set by the regulator.
According to the official version, as recounted by the game theorists involved in their
design, the auctions aimed at creating a transparent and efficient market that would
allocate the airwave spectrum rights to highest value users—those who most valued and
made best use of them.10 Until 1982, spectrum licenses were assigned by an administrative
9 See Guala (2001) for a more detailed account of the engineering work involved in the construction of theFCC auctions.
10 This account is based on game theorists’ reports of events, after efficiency had been set as the main goalof the auction to the detriment of welfare goals defined by Congress, such as the expansion of public access tonew technologies, products and services, and the decentralisation of the licenses awarded to include smallbusinesses, rural telephone companies and minority groups. For a more complete account of the politicalprocess involving the FCC auctions see Nik-Khah (2008).
hearing process (recognisably slow and non-transparent), which allocated licenses for free.
After 1982, licenses were sold and allocated via a lottery system, which significantly
improved the speed and transparency of the allocation mechanism, but it did not prevent
opportunistic behaviour. Licenses could be bought and resold by individuals who did not
want to use them and thus undeservedly appropriated revenue raised by the commercial
use of the public spectrum.
The auction mechanism seemed to offer a tremendous advantage over the alternatives. It
offered the possibility of identifying the firms with the highest use-values for the spectrum,
which would be in a position to pay the highest prices for using it and, as a result, maximise
the FCC’s revenue. This in turn required the design of an auction mechanism that
encouraged bidders to reveal their true valuations, while preventing opportunistic
behaviour on their part.
The building of the FCC auctions was a complex endeavor, best depicted as a patchwork
of various and partial solutions to the particular issues that arise when building new
markets. Auction design resembled ‘a kind of engineering activity’ that had recourse to all
sorts of resources ranging from ‘practical judgments, guided by theory and all available
evidence’ to ‘ad hoc methods to resolve issues about which theory is silent’ (Milgrom, 2000,
p. 271). Game theory merely assisted in ‘developing intuition’, in particular in ‘show[ing]
how people behave in various circumstances and [. . .] identify[ing] the tradeoffs involved in
altering those circumstances’ (McAfee and McMillan, 1996, p. 171).
The auction had to tackle three major technical issues. First, to ensure that the highest-
value users bought and paid for the licenses at their value; second, to allow the composition
of favoured combinations of licenses, which had to take into account licenses’ comple-
mentarities and substitutability; and, third, to prevent opportunistic behaviour on the part
of bidders, which would jeopardise the competitive gains obtained from instituting the
market. Theory would help look at the strategic structure of the decision-making problem
and anticipate ‘how bidders choose their bids, not knowing the value of the item for sale
and not knowing what their rivals know; and what the seller can do to stimulate the bidding
competition, not knowing how much any of the bidders is willing to pay’ (McMillan, 1994,
p. 146).
Based on advice from the game theorists, the FCC opted for the simultaneous–multiple-
round–independent auction, which gave bidders the possibility of operating in several
markets at the same time and thus of composing desirable aggregations of items or
adjusting their aggregation to a last-resort composition if their first-choice aggregations
became unattainable. The licenses would then be allocated to the highest bidder that paid
his/her bid price. Many detailed rules were then defined to organise the running of the
auction, namely how bidders were to engage in the transactions while attempting to
prevent the opportunistic exploiting of any gap.11
The next step consisted of gluing together these partial solutions and evaluating whether
they could be implemented in an operational environment. Laboratory experiments were
crucial in order to put the various pieces together into a workable mechanism and solve the
11 For example, an activity rule required the payment of deposits on the total number of desired licenses atthe beginning of the auction to ensure that market participants had actually intended to own and use thelicenses. Given the high stakes at play, the government was also concerned with simplifying procedures inorder to reduce the incidence of mistakes. To avoid the ‘winner’s curse’, i.e. selling of licenses to traders whooverestimated their value, or to avoid the extra-cautionary behaviour of risk-averse bidders, the bids wereannounced at every round so that traders could make better estimates of the licenses’ values. The incidence ofunpredictable mistakes was further taken into account by allowing bid withdrawal, though with a penalty.The auction rules in the end amounted to a 130 page document.
complications that emerged while trying to do so (Guala, 2001; Nik-Khah, 2008). The
building of the FCC auctions thus followed a division of labour in which game theorists
proposed the auction form and the rules that would organise the functioning of the market;
and experimental economists implemented these rules in an electronic market. After
stabilising the auction rules, the experimenters subsequently tested the auction under
conditions that closely resembled the market to be implemented and thereby assessed the
combined effect of the auction’s rules, which could not possibly be predicted by non-
experimental means.
Even though the accounts of the game theorists make us believe otherwise, the success
story of the FCC auctions is not uncontroversial. In an evaluation of the results, Peter
Cramton (1998, p. 735), a successful design economist, states that ‘any auction would look
good relative to the FCC’s past experience with comparative hearings and lotteries’. At the
same time, he concedes that ‘[s]ince I do not observe the bidders’ actual valuations, it is
impossible to say exactly how efficient the auctions were’ and retreats to the more vague
claim that the auctions were successful for the government, as judged by the revenues
raised, and for bidders, as judged by the stability of license compositions (pp. 728–9).
Edward Nik-Khah, based on the archives of the FCC, tells a different story, concluding:
‘[o]verall, the allocation of licenses produced by the auctions proved to be unstable, as the
industry has gone through a spate of mergers, acquisitions, and bankruptcies, ultimately
leading to a high degree of license concentration’ (Nik-Khah, 2008, p. 90). Based on the
analysis of 10 years of FCC auctions, comprising the data of 58 auctions, Gregory F. Rose
and Mark Lloyd (2006) conclude that the auctions failed to maximise receipts and
promote efficiency in the sector because bidders managed to carry out manipulative
strategies (e.g. tacit collusion and pre-emptive bidding), which resulted in the auctioning of
licenses at significantly lower prices.
Yet game theorists were eager to wrap their contribution in the allure of science,
emphasising that ‘the auction design process was driven not by politics, but by economics’
(McMillan, 1994, p. 147). But the process of building the auction was marked by the
interests of the constituencies in place, namely those of the large telecommunications
corporations. Large corporations hired game theorists to help them to position themselves
in the policy-making process, first by lobbying for the most favourable architectures for the
auctions and then by assisting in defining their clients’ bidding strategies (Nik-Khah,
2008).12 As Charles Plott (1997, p. 606), one prominent consultant, candidly acknowl-
edged: ‘Business understood that the rules and form of the auction could influence who
acquired what and how much was paid’. Thus, the building of the FCC auction was not
a mere engineering exercise. It was a complex political process, where big companies
actively exercised their influence before and during its operation. Because the FCC failed
to prevent collusion and other anti-trust strategies, taking place both inside and outside the
market, the market became more concentrated in the hands of a few large corporations.
The FCC auctions thus make it plain that markets, and non-market allocation
mechanisms for that matter, are complex institutional arrangements that require the
engineering efforts of various social scientists and social actors, and that market outcomes
depend on the particular configurations of these arrangements. Rather than assuming at
the outset that markets ensure efficiency, via the symbiotic conjunction of agents’
12 To give just a few prominent names, Paul Milgrom, Robert Wilson and Charles Plott were hired byPacific Bell, Preston McAfee by Airtouch Communications, Peter Cramton by CI, John Ledyard and DavidPorter by the National Telecommunications and Information Administration and, finally, John McMillan bythe FCC.
rationality and the information disseminated through prices, design economists are
devoted to the study of ‘the rules of the game’. In Roth’s wording:
The largest lesson in all this is that design is important because markets don’t always grow likeweeds—some of them are hothouse orchids. Time and place have to be established, related goodsneed to be assembled, or related markets linked so that complementarities can be handled,incentive problems have to be overcome, etc. (Roth, 2002, pp. 1373–4)
It is also clear that design economics is to be devoted to the ‘technical issues’ that arise in
market building. That allocation mechanisms are the outcome of difficult and complex
political processes is not adequately taken into account, especially the impact of these processes
on the ‘technical’ goals set beforehand. Design economists, qua specialised technicians, deal
with the technical complications that arise in social engineering, specifically with those that
stem from the strategic environment and the opportunistic behaviour of economic agents,
who will try to outwit the regulator, and from the cognitive limitations of real economic
agents that may compromise the goals set by the regulator. This means that design
economists also take into account the actual characteristics of economic agents. The
effectiveness of market mechanisms requires that the design setter be able to predict how
economic agents will behave under the new institutional setup. But in contrast to choice
architecture, design economists focus on the structure of incentives. The new arrange-
ments must be incentive compatible devices. They should align individual and collective
interests in such a way that individuals’ incentives correspond to what is needed to achieve
group optima, while making sure that economic agents understand the incentive structure
so that they behave predictably, like homo economicus. In the FCC auctions this meant that
the auction mechanism had to succeed in eliciting the subjective values of spectrum
licenses to ensure telecom companies acquired the desired licenses, while contributing to
an efficient allocation of the licenses, thus defined, and the maximisation of FCC revenue.
In sum, design economics is devoted to the (re)design of complex markets and other
economic institutions to be implemented in context-specific environments, to which end the
opportunistic behaviour of economic agents and their propensity to err must be taken into
account. The ultimate goal is to conceive a structure of incentives such that individual actions
can generate desirable social states. Insofar as it overlooks the political process involved in the
(re)building of new economic institutions, design economics risks failing on their own terms,
that is, it fails to pursue its narrowly defined goals of efficiency. The higher the stakes, the
more the (re)creation of a new market gives rise to an intense struggle for influence over the
collective definition of ‘the new rules of the game’. The outcome will contain a high degree of
uncertainty and it will depend on the correlation of power of those involved and their
capability to bring forward their favoured solutions. This means that market building frames
and shapes the interactions of individuals for the attainment of rather elusive goals, say the
allocation of resources in an operational way, while attempting to curb opportunistic
behaviour on their part. To put it in another way, the efficacy of design economics ultimately
hinges on determining the extent to which economists are able to implement their models in
the real world and make reality conform to their theoretical constructs, that is, on
determining the performativity of economics (to be further discussed in Section 5).13
13 This view is also shared by Vernon Smith (2008, ch. 6). While generally favourable to market design, hethinks that the ‘ecological fitness’ of ‘rational constructivist designs’ is only achieved, if at all, after a longprocess of trial-and-error that corrects behavioural incentives and strategic problems not considered indesign. Smith clearly does not think that the FCC auctions have already stabilised and achieved theirecological fitness.
This exclusive concentration on ‘technical’ issues is even more salient when design
economists propose the extension of market relations to non-market domains of social life.
The introduction of market-like arrangements that set up contractual forms of exchange
involving the transfer of money and property rights has encountered severe resistance,
since this new form of exchange becomes ‘repugnant’, as Roth puts it, when money is
added to the transaction. This ‘distaste for certain kinds of transactions can be a real
constraint on markets and how they are designed, every bit as real as the constraints
imposed by technology or by the requirements of incentives and efficiency’ (Roth, 2007, p.
38). The challenge, from the point of view of the design economist, is to learn how to deal
with these constraints, which are perceived as part of a technical problem that needs to be
tackled. Roth then urges economists to understand better and engage more with the
phenomena of repugnant transactions. This is so because ‘attitudes about the repugnance
(or other kinds of inappropriateness) of transactions shape whole markets, and therefore
shape what choices people face’ (Roth, 2007, p. 38). When referring to the debate over the
creation of a market for the sale of kidneys, Roth (2007, pp. 53–4) makes this clearer by
contrasting the irrational reactions of its opponents to the frustration of economists ‘at the
failure to adopt what they see as a feasible solution that could be implemented quickly’. In
Roth’s view, design economists, qua rational technicians, should take on ‘the important
educational role of pointing to inefficiencies and tradeoffs, and costs and benefits’ (Roth,
2007, p. 53). They should nonetheless be aware that the sources of repugnance (such as
other-regarding motivations) may affect the efficacy of incentive schemes due to the crowding
out of altruistic motivations (cf. Frey, 1997; Frey and Jegen, 2001).14 In order to be effective,
new markets have to deal with all sorts of complications that arise in market building.
To sum up, design economics is devoted to the (re)design of complex market institutions
that deal with varied technical complications in context-specific environments. The
ultimate goal is to (re)align the structure of incentives so that individual actions can bring
about desirable social states. Design economics is not circumscribed to the market domain.
The expansion of markets or market-like forms of exchange to other domains of social life,
however, requires that the economist qua social engineer take on the pedagogic role of
pointing to inefficiencies and tradeoffs, as well as costs and benefits, in discussions
informed by taboo or visceral reactions.15 This brings into the analysis the impact of other-
regarding considerations on the efficacy of public policy. The underlying political processes
of market design and the struggle for political influence are left out, however.
5. Choice architectures and market designs are political constructs
Choice architecture and design economics convey two major transformations deemed as
characterising the ongoing process of change in the discipline: (i) the revision of the
neoclassical model of human action, homo economicus; and (ii) the recent economics
approach to market building. But while fully acknowledging that individuals are not always
rational and that markets are complex socially engineered institutions, these recent
developments retain the view of economics as the science of choice, which ought to
conform to the rational choice model, and select the market as the most adequate
institution to coordinate individual actions in various realms of social life. With minor or
14 To be further explained below.15 Roth’s account provides an unflattering caricature of critical reactions to the expansion of market forms
of social interaction. See the Journal of Economic Perspectives, vol. 21, no. 3, for the debate around the creationof a market for the sale of human kidneys, where both positions are more reasonably presented and discussed.
change alters the distribution of resources, power and opportunities. Those most affected
by the new institutional setup tend to depict the new policy as an unnatural ‘intervention’
and will struggle hard to prevent it. But, as the FCC auctions illustrate, markets are the
outcome of complex political processes requiring the involvement of an organised power
capable of imposing a new set of rules defining who can participate and how. As Ha-Joon
Chang put it:
Emphasizing the institutional nature of the market [. . .] requires that we have to bring politicsexplicitly into the analysis of the market (and not just into the analysis of the state) and stoppretending that markets need to be, and can be, ‘de-politicised’. Markets are in the end politicalconstructs in the sense that they are defined by a range of formal and informal institutions thatembody certain rights and obligations, whose legitimacy (and therefore whose contestability) isultimately determined in the realm of politics (Chang, 2002, p. 553)
It is now clearer that choice architecture and design economics are not mere technical
solutions. Their proposals may affect the distribution of economic and political power and,
insofar as they do, they may see their legitimacy questioned on grounds that they violate
freedom of choice or introduce distortions in the market. That is, they may be faced with
the liberal rejection that they are unwarranted ‘interventions’ to solve problems that are
best corrected by the ‘free’ functioning of the market.17
The expectation of this kind of reaction may indeed explain the overly defensive tone of
Thaler and Sunstein and their insistence on the principle of ‘freedom of choice’, defined
negatively as the protection from interference by others. As we have seen, they have
repeatedly stressed that choice architectures are to leave the menu of choices unaltered and
grant individuals the exclusive power to use and dispose of things and services as they wish.
They thereby accept the status quo ante as legitimate and avoid, as much as possible,
altering it. By so doing they focus on people’s choices, leaving the analysis of the context
and the circumstances of the individual in the background, as is typical of neoclassical
economic theory (cf. Peter, 2004). This means that choice architecture is circumscribed to
problems of choice that stem from people’s bounded rationality. It cannot deal with those
problems that stem from the wider institutional setup where individuals act and interact.
Design economics, in contrast, proposes major institutional change, which may
substantially alter the relative position of different social groups. The proposal of the
expansion of markets to new spheres of social life has, moreover, forced design economists
to acknowledge that market relations can be coercive. The discussion around the creation
of a market for the sale of kidneys made the coercive power of pecuniary incentives more
salient for economists, acknowledging that the introduction of a monetary payment may
force the most deprived groups of the population to engage in unwanted transactions due
to extreme economic necessity (Roth, 2007). Design economists have therefore, if only
implicitly, recognised that the government does not have the prerogative of the use of
power to control people’s behaviour. Power can be exerted by various means and by
various actors.18 Nonetheless, design economists still advocate market-based solutions to
various kinds of social problem. Even when the legitimacy of these solutions is questioned,
bringing about ‘reactions of repugnance’, Roth, as we have seen, pushes economists not to
give up their traditional role of pointing to trade-offs and the costs and benefits of the
alternatives at hand, proposing that the coercive power of pecuniary incentives be tackled
17 See Sugden (2008) for a critique of choice architecture along these lines.18 See Grant (2006) for a comparative exercise on the various forms of power, including coercion,
by tuning in to their optimum level and by replacing part of monetary compensations with
in-kind rewards (Roth, 2007, p. 50).
An ample literature, however, exists in the area of political philosophy, which discusses
the need to block the expansion of markets so as to prevent so-called desperate transactions
(Walzer, 1983, 1984; Anderson, 1990, 1993). In this view, and unlike neoclassical
economics, restricting market expansion may, in fact, enhance individual freedom.
Keeping the various spheres of social life separated may be the most effective way to
ensure a range of options large enough actually to allow people to act in conformity with
their preferences and values. For, as Elizabeth Anderson (1990, p. 201) puts it, ‘the
realization of some values demands that certain goods be produced, exchanged, and
enjoyed outside market relations, in accordance with nonmarket norms’.19
On this view, not only may the introduction of market relations in realms of social life
traditionally protected from monetised exchanges create an unacceptable strain on
individuals and impede the realisation and expression of values important to human life,
but it may also have a detrimental impact on the values fostered. This is so because
preferences are endogenous (Bowles, 1998). They are affected by the institutional
arrangements that delineate the patterns of social interaction amongst the people who
make up society. The social norms and relations of the market, by conveying particular
ideals and nurturing specific values, may crowd out other fundamental values (Frey, 1997;
Frey and Jegen, 2001). Anderson’s (1993), for example, depicts market social relations as
impersonal and selfish, where individuals perceive their relations with others as a means to
satisfy their own ends, feeling free to pursue their personal advantage without consider-
ation for that of others.
In markets people are less compelled to follow non-market norms and values. By
aligning self-interest with the interests of others, market mechanisms moreover obviates
the need for ethical reasoning; as a result, individuals no longer have the opportunity, as
Steve Turnbull puts it, to ‘flex their ethical muscles’ (Frohlich and Oppenheimer, 2003,
p. 290). Individuals’ ability to behave in accordance with non-market norms and values,
then, will be seriously compromised. On the contrary, living with the tension between the
best strategy from a rational, self-interested point of view and the ethically best strategy
keeps the ethic imperative active.20
From this it follows that choice architecture and design economics are to be subjected to
ethical evaluation. Insofar as choice architecture and design economics alter the balance of
economic opportunities among people in order to control or influence their actions toward
the goals set by choice architects, by economic designers or by those who hire them, the
legitimacy of their proposals must be subject to ethical evaluation, like any other political
instrument or proposal. The market rhetoric of freedom of choice and economic efficiency
does not obviate the need for such ethical assessment. Indeed, if one takes Ruth Grant’s
(2006) criteria—legitimacy of purpose, voluntary response and the effect on the character
of the parties involved—one clearly comprehends that expanding the range of choice does
not ensure that the available options are equally legitimate, that individuals are capable of
pursuing their most preferred actions, or that the outcomes of individual actions are
19 See Rodrigues (2008) for a review on this literature.20 Frohlich and Oppenheimer (2003) describe a prisoner’s dilemma experiment where there is a conflict
between the best individual strategy and the best social strategy. The removal of this conflict through theintroduction of an incentive compatible device is effective in promoting the desirable behaviour. But when thedevice is removed, the level of cooperative behaviour reaches its minimum level, lower than the level observedin groups accustomed to the moral dilemma.
Anderson, E. 1993. Value in Ethics and Economics, Cambridge, MA, Harvard University PressBinmore, K. 1999. Why experiment in economics? The Economic Journal, vol. 109, F16–24Bromley, D. 2006. Sufficient Reason: Volitional Pragmatism and the Meaning of EconomicInstitutions, Princeton, NJ, Princeton University Press
Bromley, D. 2008. Beyond market failure: volitional pragmatism as a new theory of public policy,Economia Polıtica, vol. 25, no. 2, 219–41
Bowles, S. 1998. Endogenous preferences: the cultural consequences of markets and othereconomic institutions, Journal of Economic Literature, vol. 36, no. 2, 75–111
Callon, M. 2007. What does it mean to say that economics is performative? pp. 311–57 inMacKenzie, D., Muniesa, F., and Siu, L. (eds), Do Economists Make Markets? On thePerformativity of Economics, Princeton, NJ, Princeton University Press
Camerer, C. F. 1995. Individual decision making, pp. 587–703 in Kagel, J. H. and Roth, A. E.(eds), The Handbook of Experimental Economics, Princeton, NJ, Princeton University Press
Camerer, C. F. and Fehr, E. 2004. Measuring social norms and preferences usingexperimental games: a guide for social scientists, pp. 55–95 in Henrich, J., Boyd, R.,Bowles, S., Camerer, C., Fehr, E., and Gintis, H. (eds), Foundations of Human Sociality:Economic Experiments and Ethnographic Evidence from Fifteen Small-Scale Societies, Oxford,Oxford University Press
Camerer, C. F. and Hogarth, R. M. 1999. The effects of financial incentives in experiments:a review and capital–labor–production framework, Journal of Risk andUncertainty, vol. 19, 7–42
Camerer, C. F. and Loewenstein, G. 2004. Behavioral economics: past, present, future, pp. 3–51in Camerer, C. F., Loewenstein, G., and Rabin, M. (eds), Advances in Behavioral Economics,Princeton, NJ, Princeton University Press
Camerer, C. F., Issacharof, S., Loewenstein, G., O’Donoghue, T. and Rabin, M. 2003.Regulation for conservatives: behavioral economics and the case for asymmetric paternalism,University of Pennsylvania Law Review, vol. 151, 1211–54
Carvalho, L. F. and Rodrigues, J. 2008. Are markets everywhere? Understanding contemporaryprocesses of commodification, pp. 267–86 in Davis, J. B. and Dolfsma, W. (eds), The ElgarCompanion to Social Economics, Massachusetts, USA, Edward Elgar
Chakrabortty, A. 2008. From Obama to Cameron, why do so many politicians want a piece ofRichard Thaler? The Guardian, July 12, available at: http://www.guardian.co.uk/politics/2008/jul/12/economy.conservatives [date last accessed June 20, 2011]
Chang, H.-J. 2002. Breaking the mould: as institutionalist political economy alternative tothe neo-liberal theory of the market and the state, Cambridge Journal of Economics, vol. 26,539–59
Commons, J. R. 1931. Institutional economics, American Economic Review, vol. 21, 648–57Cramton, P. 1998. The efficiency of the FCC spectrum auctions, Journal of Law and Economics,
vol. 41, 727–36Crotty, J. 2009. Structural causes of the global financial crisis: a critical assessment of the ‘New
Financial Architecture’, Cambridge Journal of Economics, vol. 33, 563–80Davis, J. 2006. The turn in economics: neoclassical dominance to mainstream pluralism? Journalof Institutional Economics, vol. 2, no. 1, 1–20
Davis, J. 2008. The turn in recent economics and return of orthodoxy, Cambridge Journal ofEconomics, vol. 32, 349–66
Frey, B. S. 1997. Not Just For The Money – An Economic Theory of Personal Motivation,Cheltenham, UK and Lyme, USA, Edward Elgar
Frey, B. S. and Benz, M. 2004. From imperialism to inspiration: a survey of economics andpsychology, pp. 61–83 in Davis, J., Marciano, A., and Runde, J. (eds), The Elgar Companion toEconomics and Philosophy, Aldershot, UK, Edward Elgar
Frey, B. S. and Jegen, R. 2001. Motivation Crowding Theory: a survey of the empirical evidence,Journal of Economic Surveys, vol. 15, 589–611
Frohlich, N. and Oppenheimer, J. 2003. Optimal policies and socially oriented behaviour: someproblematic effects of an incentive compatible device, Public Choice, vol. 117, 273–93
Grant, R. 2006. Ethics and incentives: a political approach, American Political Science Review,vol. 100, no. 1, 29–39
Guala, F. 2001. Building economic machines: the FCC auctions, Studies in History and Philosophyof Science, vol. 32, 453–77
Hodgson, G. M. 2008. Markets, pp. 251–66 in Davis, J. B. and Dolfsma, W. (eds), The ElgarCompanion to Social Economics, Massachusetts, USA, Edward Elgar
Kahneman, D. and Tversky, A. (eds) 2000. Choices, Values, and Frames, Cambridge, UK,Cambridge University Press
Kagel, J. H. 1995. Auctions: a survey of experimental research, pp. 501–85 in Kagel, J. H. andRoth, A. E. (eds), The Handbook of Experimental Economics, Princeton, NJ, PrincetonUniversity Press
Lewis, C. 2008. Why Barack Obama and David Cameron are keen to ‘nudge’ you, Times Online,July 14 [date last accessed June 2, 2010]
Loewenstein, G., O’Donoghue, T. and Rabin, M. 2003. Projection bias in predicting futureutility, The Quarterly Journal of Economics, vol. 118, no. 4, 1209–48
Maki, U. 2009. Economics imperialism: concepts and constraints, Philosophy of the SocialSciences, vol. 39, no. 3, 351–80
MacKenzie, D. 2006. Is economics performative? Option theory and the construction ofderivative markets, Journal of the History of Economic Thought, vol. 28, no. 1, 29–55
MacKenzie, D., Muniesa, F. and Siu, L. (eds) 2007. Do Economists Make Markets? On ThePerformativity of Economics, Princeton, NJ, Princeton University Press
McAfee, R. P. and McMillan, J. 1996. Analysing the airwaves auction, Journal of EconomicPerspectives, vol. 10, 159–75
McMillan, J. 1994. Selling spectrum rights, Journal of Economic Perspectives, vol. 8, no. 3,145–62
Milgrom, P. 2000. Putting auction theory to work: the simultaneous ascending auction, Journalof Political Economy, vol. 108, no. 2, 245–72
Mirowski, P. 2007. Markets come to bits: Markomata and the future of computationalevolutionary economics, Journal of Economic Behaviour and Organization, vol. 63, 209–42
Nik-Khah, E. 2008. A tale of two auctions, Journal of Institutional Economics, vol. 4, no. 1, 73–97O’Donoghue, T. and Rabin, M. 1999. Doing It Now or Later, The American Economic Review,
vol. 103, 118–20Peter, F. 2004. Choice, consent, and the legitimacy of market transactions, Economics andPhilosophy, vol. 20, 1–18
Plott, C. 1997. Laboratory experimental testbeds: application to the PCS auction, Journal ofEconomics and Management Strategy, vol. 6, 605–38
Ritov, I. and Baron, J. 1992. Status-quo and omission bias, Journal of Risk and Uncertainty, vol. 5,49–61
Rodrigues, J. 2008. Boundaries, values and the contested nature of market expansion, NewPolitical Economy, vol. 13, 315–33
Rose, G. F. and Lloyd, M. 2006. The Failure of FCC Spectrum Auctions, Washington, DC, Centerfor American Progress. http://www.americanprogress.org/kf/spectrum_auctions_may06.pdf
Roth, A. E. 2002. The economist as engineer: game theory, experimentation, and computationas tools for design economics, Econometrica, vol. 70, 1341–78
Roth, A. E. 2007. Repugnance as a constraint on markets, Journal of Economic Perspectives,vol. 21, no. 3, 37–58
Samuels, W. J. 1989. Some fundamentals on the economic role of the government, Journal ofEconomic Issues, vol. 23, no. 2, 427–33
Samuels, W. J. 1992. Welfare economics, power and property, pp. 56–138 in Samuels, W. J. (ed.),Essays on the Economic Role of the Government, Vol. I, New York, New York University Press
Samuelson, W. and Zeckhauser, R. J. 1998. Status quo bias in decision making, Journal of Riskand Uncertainty, vol. 1, no. 1, 7–59
Santos, A. C. 2007. The ‘Materials’ of Experimental Economics: technological versus behavioralexperiments, Journal of Economic Methodology, 14, 311–337
Santos, A. C. 2010. The Social Epistemology of Experimental Economics, London, RoutldegeSantos, A. C. and Rodrigues, J. 2009. Economics as Social Engineering? Questioning the
Performativity Thesis, Cambridge Journal of Economics, 33, 985–1000Shiller, R. J. 2000. Irrational Exuberance, Princeton, NJ, Princeton University PressShiller, R. J. 2008. The Subprime Solution, Princeton, NJ, Princeton University PressSmith, V. L. 2008. Rationality in Economics: Constructivist and Ecological Forms, Cambridge,
Starmer, C. 2000. Developments in Non-Expected Utility Theory: The Hunt for a DescriptiveTheory of Choice under Risk, Journal of Economic Literature, 38, 332–82
Sugden, R. 2008. Why incoherent preferences do not justify paternalism, Constitutional PoliticalEconomy, 19, 226–248
Thaler, R. 1992. The Winner’s Curse: Paradoxes and Anomalies of Economic Life, Princeton, NJ,Princeton University Press
Thaler, R. H. and Sunstein, C. 2003. Libertarian paternalism, The American Economics Review,vol. 93, no. 2, 175–9
Thaler, R. H. and Sunstein, C. R. 2008. Nudge: Improving Decisions About Health, Wealth, andHappiness, New Haven, CT and London, UK, Yale University Press
Tversky, A. and Kahneman, D. 1974. Judgement and uncertainty: heuristics and biases, Science,vol. 185, 1124–31
Tversky, A. and Kahneman, D. 1981. The framing of decision and the psychology of choice,Science, vol. 211, 453–8
Walzer, M. 1983. Spheres of Justice, New York, NY, Basic BooksWalzer, M. 1984. Liberalism and the art of separation, Political Theory, vol. 12, 315–30Wilby, P. 2008. Cameron’s free-market guru, New Statesman, July 24. http://
www.newstatesman.com/society/2008/07/thaler-friedman-cameron-social [date last accessedJune 20, 2011]