How the market gives us what we want

1
How the market gives us what we want even if we
are irrational British Academy Keynes Lecture
2010 Robert Sugden
2
As this is the Keynes Lecture in Economics, named
in honour of one of the greatest economists of
the 20th century, I ought to start with some link
between my topic and the work of John Maynard
Keynes.
This might seem difficult, because my lecture is
a defence of the market, rather in the tradition
of Friedrich von Hayek. Hayek is usually seen as
the patron saint of pro-market economics, and
Keynes the patron saint of market regulation and
planning. But
2
3
Keynes to Hayek in 1944, after reading The Road
to Serfdom We all have the greatest reason to
be grateful to you for saying so well what needs
so much to be said... Morally and
philosophically I find myself in agreement with
virtually the whole of it and not only in
agreement, but in a deeply moved agreement.
3
4
Keynes over lunch at the Bank of England, 10 days
before death in 1946 I find myself more and
more relying for a solution of our problems on
the invisible hand which I tried to eject from
economic thinking twenty years ago.
4
5
The invisible hand by directing that industry
in such a manner as may be of the greatest value,
the merchant intends only his own gain, and he
is in this, as in many other cases, led by an
invisible hand to promote an end which was no
part of his intention. Nor is it always the
worse for society that is was no part of it.
Adam Smith, Wealth of Nations, 1776, p. 456.
Or more generally the market is a spontaneous
order with unintended consequences that are
broadly beneficial, and that may be more
beneficial than those that can be produced by
deliberate planning.
5
6
I dont want to pretend that Keynes has been
misrepresented, or that in the last few days of
his life he had a conversion experience.
But these quotations tell us something about how
deeply the idea of the invisible hand is embedded
in the professional thinking of economists. This
has been a core idea of economics for over two
hundred years, and one that non-economists
continue to find counter-intuitive. Economists
now find Smiths insight so obvious that they are
often more excited about investigating exceptions
(as Keynes did in his macroeconomics), but that
doesnt make the insight less true or important.
In this lecture, I want to defend the idea of the
invisible hand against a new challenge, from
behavioural welfare economics.
6
7
Behavioural economics tries to explain economic
behaviour by using research methods and
theoretical ideas adapted from psychology.
Ive followed this approach since the early
1980s. So, I see myself as a behavioural
economist.
Until recently, behavioural economics was an
almost wholly descriptive enterprise
discovering patterns in individual and
small-group behaviour which are inconsistent with
traditional rational-choice theories, but which
psychology can explain.
However, its findings pose severe problems for
conventional forms of normative or welfare
economics (i.e. the analysis of recommendations
about how the economy should be organised, what
economic policies governments should follow,
etc.) Recently, behavioural economists have
started to think about these problems. Hence,
proposals for behavioural welfare economics.
7
8
The problem of reconciling normative and
behavioural economics
From the early 20th century, the dominant form of
economics has been neoclassical. A fundamental
assumption of neoclassical theory is that
individuals have coherent preferences over all
relevant economic outcomes, and act according to
those preferences ( maximise utility).
Coherent preferences are -- stable (i.e. not
subject to random or arbitrary variation)
-- context-independent (i.e. not affected by
arbitrary changes of framing of decision
problems)
-- internally consistent (i.e. satisfying
rationality principles such as transitivity).
Neoclassical welfare economics uses the
satisfaction of preferences as its normative
criterion.
8
9
But the findings of behavioural economics cast
doubt on whether coherent preferences really
exist. In many cases, individuals economic
behaviour reveals incoherent preferences the
incoherencies (anomalies) are systematic and
can be explained psychologically. A few
examples
-- preferences between two options depend on
which is perceived as the status quo (the
endowment effect, loss aversion)
-- preferences revealed in choices are not the
same as preferences revealed in valuations of the
same objects (preference reversal)
-- preference between A and B varies according to
whether C is in the opportunity set (decoy
effect if C is clearly inferior to B, but not
to A, adding C to the set makes B more
attractive).
If preferences are incoherent, how can
preference-satisfaction be used as a normative
standard?
9
10
In behavioural economics, a consensus seems to be
developing around a particular response to this
problem libertarian paternalism/ asymmetric
paternalism/ soft paternalism/ behavioural
welfare economics.
This has become influential both within academia
and outside -- Thaler and Sunsteins popular
book Nudge -- Thaler recently visited 10 Downing
Street to advise the behavioural insight team
or Nudge unit on how to apply his approach to
public policy.
This approach is often presented as a challenge
to the idea of the invisible hand. Ive responded
to this challenge in a series of theoretical and
philosophical papers. This lecture summarises
the arguments I have been developing.
Including American Economic Review 2004 Social
Choice and Welfare 2007 Constitutional Political
Economy 2008 Economics and Philosophy 2008
(with Bruni) Economics and Philosophy 2010.
10
11
In part, my argument will be a critique of
behavioural welfare economics.
Not of the specific policy proposals advanced by
behavioural welfare economists, which are often
quite sensible (e.g. regulations on displays of
tariffs to facilitate price comparisons), but of
the theoretical approach that they are
using. This is fair behavioural welfare
economics is presented as a whole new way of
thinking about normative economics (the real
Third Way Thaler and Sunstein).
But my argument will also be a critique of the
way that the invisible hand idea has been
understood by neoclassical economics.
11
12
Neoclassical economists have represented Smiths
insight in theoretical models in which economic
agents are rational and in which the satisfaction
of rational preferences is the standard by which
institutions are evaluated. In these models
markets are efficient in satisfying preferences.
The findings of behavioural economics do
challenge this understanding of the invisible
hand idea.
My response is to develop a different
understanding of what markets do for us, which
retains the insights of the liberal tradition of
Smith, but is compatible with behavioural
findings.
12
13
What is behavioural welfare economics?
Two (remarkably similar) manifestos appeared in
2003 Cass Sunstein and Richard Thaler.
Libertarian paternalism is not an oxymoron.
University of Chicago Law Review, 70 (2003)
1159-1202. The academic paper that was expanded
and popularised as Nudge. Colin Camerer, Samuel
Issacharoff, George Loewenstein, Ted ODonaghue
and Matthew Rabin (2003). Regulation for
conservatives behavioral economics and the case
for asymmetric paternalism. University of
Pennsylvania Law Review 151 (2003) 1211-1254.
Each paper has a legal scholar as a co-author.
Otherwise, a roll-call of the great and the good
of American behavioural economics.
Titles (libertarian paternalism, regulation
for conservatives) signal that the authors will
propose interventions in the economy that have
traditionally been opposed by pro-market thinkers
but the authors arguments will be immune to
their opponents usual criticisms.
13
14
These papers implicitly criticise traditional
welfare economics, but do not explain how, if
their authors proposals are accepted, welfare
economics should be reconstructed. Sunstein and
Thalers most concrete advice public decisions
should be based on a form of cost-benefit
analysis whose goal is to measure the full
ramifications of any design choice. (No
discussion of how to measure costs and benefits
when individuals dont have coherent
preferences.) Similarly unspecific suggestions
about cost-benefit analysis by Camerer et al.
More concrete proposals for reconstructing
welfare economics are now being made. For
example Douglas Bernheim and Antonio Rangel.
Beyond revealed preference choice-theoretic
foundations for behavioral welfare economics.
Quarterly Journal of Economics 124 (2009)
51104. Esssentially Bernheim and Rangel take
the approach advocated in the 2003 manifestos and
try to adapt theoretical welfare economics to fit
it.
14
15
In this lecture, Ill focus on the principles of
behavioural welfare economics, as advocated by
Sunstein and Thaler.
Richard Thaler
Cass Sunstein
And Ill focus on one of Sunstein and Thalers
central claims (in opposition to traditional
welfare economics) The findings of behavioural
economics force us to recognise that paternalism
is inevitable the idea that there are viable
alternatives to paternalism is a
misconception the anti-paternalist position is
incoherent, a nonstarter.
15
16
Why (according to Sunstein and Thaler) is
paternalism inevitable?
Traditionally, economics has assumed that
individuals have coherent preferences which are
prior to the decision situations in which they
are revealed. If that assumption were true, it
would make sense to ask whether those preferences
should be respected. The anti-paternalist would
say Yes, the paternalist would say Perhaps
not.
But in fact, individuals often form their
preferences only when confronting specific
decision problems. Those preferences are
sensitive to apparently arbitrary details of
framing (e.g. endowment effect, preference
reversal, decoy effect). So, the
anti-paternalists principle (respect
preferences) cant be applied the objection to
paternalism fails.
That in itself doesnt make paternalism
inevitable, only unobjectionable. But
16
17
Sunstein and Thaler conceive of themselves as
advising a planner (later choice architect)
who is responsible for designing the presentation
of options to individuals.
If individuals preferences are sensitive to
framing, the planners choice of frame can affect
the preferences that individuals reveal there is
no way of simply standing back and respecting
preferences. So the planner cannot avoid a
decision about the direction in which to steer
the individual, and the only reasonable criterion
for making this decision is the planners
judgement about the individuals best
interests. This is the sense in which there are
no viable alternatives to paternalism.
But if the planners only intervention is to set
the framing, individuals remain free to ignore
that framing. Thus, this form of paternalism can
still be called libertarian.
17
18
Sunstein and Thalers favourite example of
libertarian paternalism the cafeteria. Consider
the cafeteria at some organisation
(Lets say the cafeteria at the University of
East Anglia...)
18
19
Customers proceed along a line, choosing food
items from a display, until they reach the
checkout. The cafeteria director notices that
given items are more likely to be chosen if they
are placed earlier in the line. On the basis of
current medical knowledge, she judges that most
customers would be better off if they ate fewer
sweet desserts and more fruit. Which should she
display first, the fruits or the desserts?
19
20
Sunstein and Thaler draw up a shortlist of four
rules the director might follow in deciding how
to display food items. Two really are
non-starters choose at random and make the
customers as obese as possible. This leaves two
plausible contenders
20
21

1. She could give customers what she thinks they
   would choose on their own notice give or

(2) She could make choices that she thinks
would make the customers best off, all things
considered.
An anti-paternalist would favour (1), but this
rule is meaningful only if what the customer
would choose can be defined independently of the
directors choice. But it cant consumers
lack well-formed preferences, in the sense of
preferences that are firmly held and preexist the
directors own choices about how to order the
relevant items. If the arrangement of the
alternatives has a significant effect on the
selections the customers make, then their true
preferences do not formally exist.
So the cafeteria director has to use (2).
21
22
And in the UEA cafeteria ...
22
23
Another view ...
23
24
The shortlist of rules for the director didnt
include Choose the display that maximises
profit.
The (later) Nudge version of the story (now set
in a school) does, but This rule has some
appeal, especially if Carolyn the director
thinks that the best cafeteria is the one that
makes the most money. But should Carolyn really
try to maximise profits if the result is to make
children less healthy, especially since she works
for the school district?
The reference to children is a diversionary
tactic (ST are proposing paternalism for
adults). The thought seems to be that a cafeteria
manager should be concerned with the welfare of
her customers, and not just making money . If
coherent preferences existed, they might be
treated as indicators of customers welfare (and
so making money would produce welfare) but they
dont exist.
Implication if individuals preferences are
incoherent, the usual argument for the market
(i.e. that it satisfies preferences) is
invalidated then, paternalism is the only
defensible response.
24
25
Are Sunstein and Thaler being too neoclassical?
ST instruct the planner (or cafeteria director)
to nudge individuals towards the choices that are
in their best interests. But how is best
interest judged?
ST spend much more time discussing how people
can be nudged than on how the planner decides in
which direction to nudge them. But their official
position is to use an informed desire or true
preference criterion. Individuals are treated
as not acting in their own best interests if
their decisions are ones they would change if
they had complete information, unlimited
cognitive abilities, and no lack of willpower.
Implication a persons best interests correspond
with the preferences he would reveal if he had
complete information, unlimited cognitive
abilities, and no lack of willpower.
Notice the implicit assumption that true
preferences exist, and are coherent. What
justifies this true preferences assumption?
25
26
Sunstein and Thaler are effectively assuming
that, inside every behavioural human being,
there is a neoclassical rational agent. The
rational agents optimal choices are frustrated
by imperfections that are external to it,
(imperfect information, imperfect cognition,
imperfect self-control).
This idea doesnt fit well with the methodology
of behavioural economics the concept of true
preferences seems to have come from a priori
rational choice theory, not empirical psychology.

Information, cognition and willpower are all
inert without desires to act on. Superhuman
agents who had perfect information, perfect
cognitive powers and perfect willpower would
still have to deal with their actual desires,
which are matters of psychology, not rationality.
Anomalies in human decision-making (i.e.
deviations from rational-choice theory) may
reflect the structure of desires (e.g. loss
aversion), not error.
So perhaps ST are being too neoclassical. (I
think my approach is more behavioural as well as
more liberal!)
26
27
Behavioural welfare economics the fundamental
logic Think about the structure of Sunstein and
Thalers argument
1. Welfare economics is addressed to a planner,
who decides what individuals should be given (not
to citizens who decide whether they want to have
a planner, or to choose for themselves how to
spend their own money).
2. The planner seeks to maximise the welfare of
each individual (other things equal). There is a
presumption that preference is an indicator of
welfare. So, if the individual has coherent
preferences, the planner gives him what he
prefers.
3. The market may be a convenient mechanism for
doing this (the invisible hand argument) but the
justification of the market is that it gives each
of us what the ideal planner would give us.
4. If individuals choices are influenced by
arbitrary factors, the data provided by
revealed preferences (and on which the market
operates) are corrupted. So the invisible hand
argument for the market fails.
27
28
Two concerns about this form of argument
1. Mismatch between the liberal tradition to
which Adam Smith belongs and the idea of
defending the market by taking the viewpoint of a
planner, for whom individuals choices are just a
source of data.
2. If the argument in support of the market
really applies only when individuals have
consistent preferences, its power is very
limited. We all know that our preferences are
affected by arbitrary factors (e.g. tendency to
favour status quo, sensitivity to the way goods
are displayed, influence of other people ...).
Given preferences is a modelling assumption,
not a fact about the world.
These concerns are not specific to behavioural
welfare economics they apply to neoclassical
welfare economics too.
28
29
An alternative approach
Behavioural and neoclassical welfare economics
interpret the invisible hand argument as claiming
that markets are effective in satisfying given
preferences. (First we specify preferences, then
we ask whether the market satisfies them.)
I propose that we reformulate the invisible hand
argument as claiming that markets allow
individuals to satisfy their preferences,
whatever those preferences turn out to be. (We
evaluate the market from the perspective of
individuals who do not yet know what their
preferences will be.)
Preference inconsistencies disable the first
approach, but not the second.
29
30
Reconstructing the first fundamental theorem of
welfare economics
This theorem is generally seen as the core
neoclassical statement of the invisible hand
argument. Roughly, it says that competitive
markets are efficient in satisfying
preferences. Ill show how the theorem can be
reconstructed so that it doesnt refer to given
preferences. (Just the first step in a bigger
project ...)
First, Ill state the theorem more precisely, as
applied to a very simple exchange economy.
30
31
In this economy, there are many
individuals. There are many goods, all of which
are private. There is a fixed stock of each
good, so the only economic problem is to divide
these stocks between individuals. Any division
of goods between individuals is an allocation.
We start with an initial allocation (individuals
endowments). Individuals are then able to
exchange goods by mutual consent.
Competitive equilibrium is a list of prices, one
for each good, such that all markets clear (i.e.
for each good, total amount offered for sale
total amount demanded). Trade at these prices
brings about a new allocation.
Its assumed that each individual has a given
preference ranking over all bundles of goods, and
acts on this. An allocation is Pareto-optimal if
no feasible reallocation of goods between
individuals would make some individual better off
and no one worse off (in terms of their
preferences).
The theorem tells us every competitive
equilibrium is Pareto-optimal.
32
A sketch of a proof of the first fundamental
theorem
Deliberately, the proof is not quite standard.
It proves as much as possible using only the most
minimal assumptions about preferences, and
introduces rationality assumptions at the very
end. This allows us to identify desirable
properties of the market that dont depend on
rationality assumptions.
Preliminary. Competitive equilibrium can be
defined without using the concept of preference
(or utility). Conventionally, demand and
supply (and hence market-clearing) are defined
in terms of utility-maximising choices. But all
we need to assume is that individuals make
decisions about how much to buy and sell at the
prices that are on offer. These decisions need
not reveal consistent preferences (e.g. desired
holdings of goods may depend on endowments,
and/or on arbitrary framing features). All I
assume about preferences is that one good
(money) is always desired.
32
33
A possible doubt Competitive equilibrium is
defined as a list of market-clearing prices. In
neoclassical welfare economics, this is treated
as an idealised representation of the outcome of
real markets. For this idea to be plausible, do
we need to assume that individuals have
consistent preferences?
No. Its sufficient that trades are mediated by
profit-seeking professional traders (who are
rational in their professional activities). No
one needs to have consistent preferences over
bundles of goods. I show this in a paper in
American Economic Review 2004.
33
34
Step 1 of proof In competitive equilibrium, all
opportunities for voluntary transactions have
been made available to individuals (severally).
Let Z be the initial allocation. Let X be the
allocation reached in a competitive equilibrium,
by trades from Z made at the price list P. What
would it mean to say that some opportunity for
voluntary transaction had not been made available?
Suppose that were so. Then there would be (a)
a feasible allocation Y (not the same as X),
reachable from X by some composite transaction
and (b) no party to that composite transaction
(i.e. no individual whose bundle in Y is not the
same as his bundle in X) has been offered his
part of the transaction and has rejected it.
But in fact, for any Y satisfying (a) at least
one party to that transaction was offered his
part of it through the market, and chose not to
take it.
34
35
How do we know this?
X (the outcome of the market) and Y (the outcome
of the composite transaction) contain exactly the
same goods. (In this economy, the only possible
transactions are exchanges.) So value of Y at
market prices value of X at market prices.
For each party to the transaction we can ask
whether his Y-bundle is worth more or less (at
market prices) than his X-bundle. Clearly, it
cant be the case that every party gains value
in the transaction.
So there must be at least one individual i, who
is a party to the transaction, and whose Y-bundle
is worth no more than his X-bundle.
But i had the opportunity to buy his Y-bundle (or
one unambiguously better) at the market prices,
and chose not to do so.
Which proves In competitive equilibrium, all
opportunities for voluntary transactions have
been made available to individuals.
35
36
Step 2 of proof If each individual acts in
accordance with a stable preference ordering,
then competitive equilibrium is Pareto-optimal.
For simplicity, I assume that preferences are
strictly convex (i.e. there is a uniquely optimal
choice from every budget constraint) this isnt
essential.
From Step 1, we know that if X is the outcome of
the market, then for every other feasible
allocation Y, there is at least one individual i
who chose not to take his Y-bundle in exchange
for his X-bundle. So, i prefers his X-bundle.
So, every feasible reallocation of X makes at
least one individual worse off. QED.
36
37
The significance of this proof
Behavioural welfare economists claim that the
invisible hand argument is undermined if
individuals lack coherent preferences. But
coherent preferences are needed only for Step 2.
Step 2 converts In competitive equilibrium, all
opportunities for voluntary transactions are made
available into Competitive equilibrium is
Pareto-optimal. This step switches from
individuals perspectives to the planners
perspective. The planner is trying to maximise
each individuals welfare she treats given
preferences as indicators of welfare so she
needs to show that opportunities for voluntary
transactions translate into the satisfaction of
given preferences.
But For each of us as individuals, assessing
what the market does for us, isnt All
opportunities for voluntary transactions are made
available sufficient? Do we need the planners
viewpoint?
37
38
How the market gives us what we want
My representation of the invisible hand The
market gives each of us what we want and are
willing to pay for, when we want it and are
willing to pay for it.
38
39
The market gives each of us what we want and are
willing to pay for, when we want it and are
willing to pay for it. Willing to pay willing
to give up what would induce others to take part
in the transaction.
When we want the market responds to our
preferences, whatever they may be, at the moment
at which we transact (and so, whatever framing is
in place at that moment). It takes no notice of
whether our preferences are consistent or not.
If trading takes place over time, the market
responds to our preferences at all trading
moments. If I change my mind from one moment to
another (e.g. today I want to sell my Nissan and
buy a VW, next week I want to sell the VW and buy
the Nissan back), the market gives me what I want
and am willing to pay for at both moments (in
effect, facilitating a trade between me now and
me next week).
39
40
Is this property of markets good for each of us
(from our own viewpoints)? Let me try to persuade
you that, all things considered, this is good for
us, even if our preferences dont meet the
standards of neoclassical economics.
40
41
An analogy Suppose you need to go shopping to
buy food for your next meal. You havent decided
what you want to eat. You expect that you will
decide only when you are in the shop, seeing what
is on offer (and perhaps being influenced by the
displays). At the moment, all you have to decide
is which shop to go to.
There are two grocery stores, roughly equally
distant...
41
42
A small shop, with a limited range of products...
42
43
and a large supermarket, with a much wider
range...
Which shop would you go to?
43
44
I would go to the big supermarket. (From trends
in the grocery market, I infer that many people
would make the same choice.) Why?
Not because it is more effective in satisfying my
existing preferences for meals (these preferences
are not yet defined) ...
... but because it will give me more opportunity
to buy what I want, when the time comes to
choose. Whatever my preferences turn out to be,
I will be better able to satisfy them at the
larger store. This is true, however different my
preferences then are from my preferences now.
To see this as good for me, I dont need to ask
what future choices would be best for my welfare.
I just have to be willing to delegate future
decisions to my future self.
44
45
The point of this analogy opportunity can be
valued by someone who cant predict how he will
use it, and who recognises that his preferences
may be subject to arbitrary variation.
To recognise this value, you have to take a
particular attitude to your future selves
responsibility.
You (i.e. you-now, reflecting about your
continuing interests) have to identify with your
future selves. You have to see their actions
as yours, whether or not these are the actions
that you-now would choose.
Taking this attitude, you-now want it to be the
case that you-then are able to get what you-then
want. So, you see all opportunities, present or
future, as good for you.
46
Economists find this way of thinking about future
selves surprisingly difficult to grasp.
Because economists like using rational-choice
models, their inclination is to model the
relationship between selves as a game between
agents with conflicting interests. In these
models, the present self thinks of the future
self as if it were a different person, liable to
frustrate the present selfs intentions.
So, in cases like the supermarkets, they
immediately think about self-control problems...
What if I-now am worried that the supermarket
displays will tempt me-then to buy something that
I-now think would not be good for me-then?
Can I-now use the choice of the smaller
supermarket as a way of constraining me-then?
And so on...
46
47
Of course, genuine self-control problems (i.e.
cases where people want to impose constraints on
themselves) do sometimes occur.
But I suggest that these problems are far, far
less common than cases in which people cant
fully predict their future preferences, have no
expectation that their preferences will be
consistent, but have no desire to impose their
present preferences on their future selves.
For most of us, most of the time, it is entirely
sensible to delegate future choices to future
selves. So, if markets tend to give us what we
want, when we want it, thats not such a bad deal.
47
48
Back to Sunstein and Thalers cafeteria...
In a competitive market, what displays will
cafeteria managers choose?
Obviously the displays that maximise profit.
So, the more costly forms of display will be used
for the products for which display has the
largest impact on customers willingness to
pay. I conjecture chocolate- and cream-rich
desserts offer more scope for this form of added
value than apples and bananas.
48
49
Compare a supermarket-style display of cakes ...
49
50
... with a coffee-shop display.
The coffee-shop display is more costly, but also
more tempting.
50
51
If competitive cafeterias choose prominent
displays of their less healthy options, does that
count against the invisible hand?
Its easy to say that other people should be
steered away from buying what we think isnt good
for them. So, academic economists, philosophers
and lawyers who would never dream of entering a
McDonalds are often happy to recommend nudges
directed at people who do.
But when you think about the restaurants that you
go to, and the delicious but unhealthy dishes
that you like, do you really want the management
to use deliberately untempting displays (e.g.
hygienic but cracked plates, tasteless and
harmless but unattractive colourings)? Or do you
want the restaurant owner to try to offer you
dishes that you will want to order?
51
52
So, when the cafeteria director chooses a
prominent display for the desserts, that isnt
necessarily a matter for shame on her part or
disapproval on ours.
What she is doing isnt just making money (as
Sunstein and Thaler say). She is seeking out
voluntary transactions with customers. She is
trying to offer them what, when they see it, they
will want to buy.
That cafeterias try to do this isnt such a bad
thing for me as a potential customer.
52
53
3 min or 1 min Mutual advantage
I have said that that the manager of the
competitive cafeteria isnt just making money,
she is seeking out voluntary transactions with
potential customers. What does this mean?
I want to suggest a way of thinking about market
relationships that is slightly different from
Adam Smiths.
Adam Smith on the motivation of shopkeepers
54
It is not from the benevolence of the butcher,
the brewer, or the baker, that we expect our
dinner, but from their regard to their own
interest. We address ourselves, not to their
humanity but to their self-love, and never talk
to them of our own necessities but of their
advantages. Nobody but a beggar chuses to depend
chiefly upon the benevolence of his
fellow-citizens. Adam Smith, Wealth of Nations,
1776, pp. 26-27.
Notice contrast between self-love and
benevolence. Smith is saying that the shopkeeper
isnt benevolent, he is self-interested (making
money). The invisible hand converts private
self-interest into public benefit.
54
55
But there is another possible motivation mutual
benefit, i.e. acting on the intention to benefit
oneself and ones trading partners together. This
is neither self-interest nor benevolence.
Compare playing ones part in cooperative
arrangements.
So the cafeteria director who tries to supply
what her customers want to buy can be motivated
by mutual benefit just as participants in other
cooperative practices (e.g. voluntary supply of
public goods, trust) can be.
56
A text from an Italian economist, a contemporary
of Adam Smiths
Antonio Genovesi, Lectures on Commerce, or on
Civil Economy, 1765-67. The concluding paragraph
of his series of lectures on economics Here is
the idea of the present work. If we fix our eyes
at such beautiful and useful truths, we will
study not for stupid vanity ... but to go along
with the law of the moderator of the world, which
commands us to do our best to be useful to one
another.
Genovesi is telling his students that economics
teaches us how to be useful to one another, i.e.
how to achieve mutual benefit through
cooperation.
57
We can understand the market as an institutional
framework that facilitates mutually beneficial
transactions. That is why it tends to give us
what we want and are willing to pay for, when we
want it and are willing to pay for it.
Well-motivated market participants construe their
transactions as relationships of mutual benefit.
We should be cautious about disabling the
mechanisms by which markets tend to facilitate
voluntary transactions. Even when our
preferences are inconsistent, the invisible hand
may be at work.
58
Thank you for listening.
2
3
4
58
59
offcut
But Sunstein and Thaler are not completely
opposed to restricting choices How much choice
should people be given? Libertarian paternalists
want to promote freedom of choice, but they need
not seek to provide bad options, and among the
set of reasonable ones, they need not argue that
more is necessarily better (2003, p. 1196).
Notice suggestion that people are given choice
options by a benevolent planner. Not people
propose transactions to one another.
59