Informal institutions, financial decisions and financial development

1
Informal institutions, financial decisions and
financial development

• Luigi Guiso
• European University Institute

2
Lecture 6-7 Social capital, Trust and Business
Behavior
3
Outline

• What are social capital and trust?
• Are they linked and how?
• Discuss different types of social capital
• How can we measure S.C.? How can we measure
  trust?
• Do S.C. and trust matter for finance and if so
  why?
• Present some theoretical foundations of S.C.
• Look at Empirical determinants of trust and
  social capital
• http//www.socialcapitalgateway.org/NV-eng-basicre
  adings.htm

4
What is trust/social capital?

• Gambetta (1988) on trust
• When we say we trust someone or that someone is
  trustworthy, we implicitly mean that the
  probability that he will perform an action that
  is beneficial is high enough for us to consider
  in engaging in some form of cooperation with
  him.
• Putnam (1993) on social capital
• features of social life - networks, norms, and
  trust - that enable participants to act together
  more effectively to pursue shared objectives".

5
What is trust/social capital? - 2

• Social capital gt trust, the opposite is not
  necessarily true
• Example in a police state I may trust others,
  but not cooperate with them.
• Social capital is a more general concept, it
  captures also
• sense of duty,
• Machiavellis civic virtue

6
The trust/social capital link

• Put differently social capital viewed as the
  set of ties that links together a community of
  individuals sustains and thus implies trust
  the reverse is not true as a high level of trust
  may stem from sources other than social ties
• We will discuss first social capital and then
  trust

7
The different types of social capital

• 1) Sociologists social capital
• the advantages and opportunities accruing to
  people through membership in certain communities
• It is a characteristic of small groups, not of
  larger communities or nations.
• It is a resource that emerges from social ties
  individuals have, not a public good
  (embeddedness)
• As such, it is important that some are excluded
  from it.
• Ex Ethnic lending in the United States
• gt we will label it network capital

8
The different types of social capital -2

• 2) Management scientists social capital
• Organizations are more than a collection of
  individuals intent on achieving their own private
  purpose.
• The difference is the stock of active
  connections among people the trust, mutual
  understanding, and shared values and behaviors
  that bind the members of human networks and
  communities and make cooperative action
  possible(Cohen and Prusak (2001) )
• Ex UPS/Goldman Sachs
• gt we will label it organizational capital

9
Different types of social capital -3

• 3) Political scientists social capital
• collection of good behavior that tend to be
  simultaneously present in certain
  communities/countries whose inhabitants vote,
  obey the law, and cooperate with each other and
  whose leaders are honest and committed to the
  public good. (Putnam, 1993 and 1995)
• Ex North vs. South of Italy
• gt I will label it civic capital

10
Summary up to this point

• Depending on the level of analysis we can
  distinguish three forms of social capital
• 1) At the level of individuals/ small groups
  network capital
• 2) At the level of firms organizational capital
• 3) At the level of larger communities/nations
  civic capital.
• These three notions are clearly related. But they
  do not always go hand-in-hand.

11
Summary up to this point -2

• A very strong network capital in small groups
  tend to be accompanied by low levels of civic
  capital
• Banfield (1958) amoral familism gt very strong
  ties within the family (a small group) but weak
  ties outside the family network
• Fukuyama (1995) strong trust within the family
  is associated with an extremely low level of
  trust toward people in general.
• Guiso, Sapienza, and Zingales (2000) family
  lending is negatively correlated with measures of
  civic capital.

12
Social capital and finance 1

• Having addressed what social capital is let us
  ask why social capital may matter for finance
• Intuition all financial contracts are promises.
  They are an exchange of money today against a
  promise of (more) money tomorrow
• But what makes sure the promise is kept? How do
  we know the borrower returns capital and
  interest?

13
Social capital and finance 2

• The normal answer is that one relies on contracts
  and contracts are enforced by courts
• Hence a lender is willing to lend because the
  contract is granted perfect legal protection
• gt in a word where this is true there is no role
  for social capital
• In practice legal protection is imperfect

14
When is legal enforcement imperfect?

• Legal enforcement is normally imperfect for a
  number of reasons
• Courts, as all institutions, are imperfect and
  cannot grant prompt enforceability in case of
  misbehavior they are slow, they are costly
• Contracts are normally incomplete law can grant
  protection for something that the two parties
  have established in a contract gt not all
  contingencies can be written, simply because
  cannot be foreseen
• Courts can enforce promises that are verifiablegt
  in many instances this is not the case (e.g. the
  output of a firm)

15
A role for social capital 1

• When legal enforcement is imperfect there is a
  potential role for social capital
• Where ties among individual are stronger social
  networks can provide social punishment and
  enforce contracts this way
• If social punishment is strong individuals will
  anticipate this at time of contracting and will
  trust counterparts more gt they will be more
  willing to lend (exchange promises) as they
  expect the promises to be kept
• In sum social capital can be a substitute for
  imperfect courts

16
Is this role realistic? Yes, in many countries
Italy650
Italy650
Germany154
Argentina300
UK101
Source Shleifer et. al. "Courts The Lex Mundi
Project", QJE http//www.andrei-shleifer.com/dat
a.html
17
A role for social capital 2

• This role is not limited to finance but it
  extends to businesses more generally
• The set of exchanges that involve promises is
  much broader than financial exchanges
• All exchanges that are not merely spot
  transactions involve an element of trust or need
  cooperation
• Examples
• A worker who promises to work (cannot be
  continuously monitored)
• A purchase of a good through the internet with
  future delivery at home
• A hiring today of a professional with the job
  starting one year from now (he may not show up)
• A bunch of friends who promise to attend your
  party one week from now.(May not come and your
  party fails).. And many more relevant cases

18
Is Social Capital Necessarily Good For Business?

• Network capital
• i) Good
• It facilitates cooperation and saves on
  transactions costs. Ex Jewish diamond merchants
  in New York.
• It facilitates access to lending ex ethnic
  lending, angel networks
• ii) Bad
• It inhibits innovation which is considered
  deviant (Bourgois, 1995)
• It makes adaptation more difficult (Uzzi, 1997)
• It makes transactions with outsiders more
  difficult, thus preventing entry

19
Is Social Capital Good For Business? - 2

• Organizational capital
• i) Good
• Enhances cooperation Xerox Eureka system (Cohen
  and Prusak (2001)).
• ii) Bad
• Ties that blind members of a close-knit
  organization can support one anothers
  misjudgments about the desirability of its
  products (e.g., Wang Lab).

20
Is Social Capital Good For Business? -3

• Civic capital
• By definitions it has only positive effects.
• Examples
• Knack and Kneefer (1996) positive correlation
  between a country's level of trust and its rate
  of growth.
• La Porta et al. (1997) positive correlation
  between the trust prevailing in a country and
• low levels of corruption, high level of judicial
  efficiency
• the presence of large organizations.
• Later in the course we will be focusing more on
  the role of civic capital for financial
  development

21
Measuring social capital
22
How to measure Social Capital?

• Focus on Civic Capital
• Social capital is hard to measure
• The set of ties are not observable
• The concept is not univocal in Putnams
  definition features of social life - networks,
  norms, and trust - that enable participants to
  act together more effectively..
• But if one wants to make progress and test its
  relevance it needs to be measured, albeit
  imperfectly (see Adam Roncevic, 2005)

23
Outcome based measures 1

• Idea instead of estimating the capital
  estimate its output. This is the route taken in
  practice
• One can rely on several proxies that capture one
  of the features of social capital
• Examples from Putnam (1993)
• Attitudes towards cooperation number of
  cooperatives across locales
• Measures of social behavior participation is
  voluntary associations
• Attitudes towards free riding and cheating
  compliance with taxes, littering, queuing etc.
  
• All these variables are indicators of the
  civicness that social capital produces

24
Outcome based measures 2

• But some of them may be inadequate they may
  proxy for other features that are unrelated to
  social capital (an output may be produced by
  various inputs not only by social capital)
• Tax incentives can affect the number of
  cooperatives, or cooperatives may reflect
  patterns of industrial specialization (difficult
  to run a car industry as a cooperative)
• Differences in tax compliance and littering
  across communities may result from differences in
  the quality of legal enforcement or the legal
  punishment of tax evasion which may have nothing
  to do with social capital
• One needs to be careful..

25
Outcome based measures 3

• Examples of good measures of social capital from
  Guiso, Sapienza Zingales (2004)
• Blood donation across Italian provinces
• only one voluntary collector (Avis)
• is not subsidized
• Is well defined
• Does not depend on formal enforcement mechanisms
  
• Participation in referenda across provinces
• Voting is a right and a (moral) duty, but no
  legal punishment from non participation
• Voting at referenda is free from possible
  contamination in the incentives to vote from
  exchange votes (is not a political election)
• It only reflects moral duties from a purely
  rational viewpoint one should not vote and free
  ride

26
Social capital across provinces

• "

• "

• "

• "

• "

• "

• "

• "

• "

Blood Donation

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

Italian Provinces

• "

• "

• da donation

• "

• "

• 47

• a

• 106

• (21)

• 34

• a

• 47

• (16)

• "

• "

• "

• "

• 17

• a

• 34

• (18)

• "

• "

• 8

• a

• 17

• (20)

• "

• "

• 0

• a

• 8

• (20)

• "

• "

27
Social capital across provinces
Participation in referenda

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

• "

Italian Provinces

• "

• da trust1

• "

• "

• "

• 87

• .6

• a

• 91

• .6

• (19)

• "

• 84

• .2

• a

• 87

• .6

• (18)

• "

• "

• "

• 79

• .1

• a

• 84

• .2

• (19)

• "

• "

• "

• 72

• a

• 79

• .1

• (19)

• "

• 62

• a

• 72

• (20)

• "

• "

28
Features on these measures 1

• There is considerable variation across local
  communities
• There is a clear north-south divide with the
  North leading the South in terms in terms of
  social capital intensity
• the two measures are highly correlated, areas
  were people donate more blood are also areas
  where people participate more in referenda
  (donate their time to the benefit of all)
• High correlation across outcome based measures is
  a general feature

29
Features on these measures 2

• These measures refer to the 1980s and 1990s. They
  are highly correlated with measures of social
  capital from the late 1870sgt social capital
  tends to depreciate very slowly (hence it
  deserves the name capital)
• Since Social Capital evolves very slowly at each
  point in time it is predetermined and thus acts
  as a constraint on the economy over the short
  run (decades) most of the causation is from
  social capital to the economy

30
Determinants of social capital
31
Pending Questions

• Why does social capital persist so much?
• What can explain the differences in the level of
  social capital across areas belonging to the
  same country that over the past 160 years has the
  same laws, tax rules, formal institutions etc.?
  
• Ultimately, what originates social and how does
  it get accumulated ?

32
Where Does Social Capital Come from ?

• These are difficult questions to answer and we
  are not going to answer them fully. But in order
  to be able to address them we need a better
  understanding of the sources of different types
  of social capital.
• For this reason we are going to review
• 1) Theoretical foundations of different types of
  social capital
• 2) Empirical evidence on them

33
Theoretical Foundations of Social Capital1

• Sources of social capital can be grouped into two
  main categories
• 1) Behavioral
• People act in a certain way because of strong
  internalized norms (what sociologists call
  consummatory behavior)
• among Jews, fathers must send sons to school to
  learn the Torah
• Unclear where these norms come from but several
  factors may matter religion, past history, etc.
• How do they persist? Are they efficient? Were
  they efficient when they were set up? Why were
  they set up?

34
Theoretical Foundations of Social Capital 2

• 2) Rational
• a) Multiple equilibria it is optimal to
  cooperate if you expect others to cooperate.
• b)Social capital enhances the level of social
  punishment in a society, making it easier to
  support cooperative equilibria.
• The larger the number of networks a person
  belongs to, the easier it is to punish him for
  deviant behavior.
• c) Once a (Nash) equilibrium is established it
  tends to persist as nobody has an incentive to
  deviate
• d) Behavior tends to crystallize in norms to
  which individuals subscribe

35
Persistence of Social Capital

• The Rational explanationgt natural mechanism that
  can explain persistence if cooperation is a Nash
  equilibrium nobody has an incentive to deviate
  from it similarly if non-cooperation is an
  equilibrium it naturally persists
• Example driving in Rome at a crossgt people
  expect others to move to the middle of the cross
  even when the traffic is stuck optimal response
  is to move there. This is a non-cooperative
  equilibrium
• These beliefs become part of the local culture

36
Persistence of Social Capital

• What about the Behavioral explanation?
• Most natural mechanism is education within the
  family
• Parents endorse certain norms of behavior and
  tend to teach these same norms to their kids
  which in turns teach them to their kids and so on
  from generation to generation
• Tabellini (2007)gt on norms
• GSZ (2008, JEEA) gt on beliefs

37
Persistence of Social Capital

• Dohmen et. al (2006) use data from the German SEP
  on attitudes of two linked generations (parents
  and sons)
• Childrens attitudes towards trust and risk
  aversion are highly correlated with their parents
  attitudes
• the attitudes of the mother matter the most,
  consistent with attitudes being transmitted with
  education at home

38
Determinants of social capital

• i) There is no general theory of what determines
  social capital. Typically, norms of behavior,
  civic virtues, beliefs and attitudes that belong
  to a community are traced back to
• Religion/culture
• History and Democracy
• Communication levels
• ii) If the endowment of social capital in viewed
  as an attribute of an individual, features that
  matter are
• Diversity
• Mobility
• Age

39
Determinants of civic capital History

• Putnam (1993) traces the difference in social
  capital in the North and South of Italy, to
  history around year 1,000
• The North solved the disorder of the middle age
  by inventing the city-state
• Horizontal linkages and political independence
  educated individuals to social and civic
  participation, building social capital
• The South was dominated by the Normans, in a
  highly hierarchical regime inimical of horizontal
  linkages and thus of cooperation among
  individuals

40
Is Putnam Right?
Fraction of people who trust
42
25
3.41
Blood donation (bags per 000 people)
1.04
41
Is Putnam Right?

• Use data on the history of Italian towns located
  in North
• Whether it has been an independent town during
  the 11-14th century
• Whether it has been an independent Signoria
• Correlate these measures with today measures of
  Social Capital in the city
• Number of non-profit organizations
• Number of voluntary associations

42
Is Putnam Right? Apparently yes!
43
Is independent city proxing for something else?

• Could be proxing for some unobserved varibel that
  affects both SC and history
• If so it must be a very persistent variable e.g.
  geography, that is not picked up by controls
• Deal with this with instruments for independent
  cities. What sort of instruments?
• Rely on history

44
In search for instruments for independent cities

• History suggests three potential instruments
• Whether a city was had a archbishop
• Easier to obtain independence from the Emperor as
  an authority to coordinate with was already
  available
• Whether it was founded by the Etruscans
• Etruscans were organized in city states that
  valued and enjoyed independence and
  self-administration
• Whether it was on the track of the Emperors
  route on his trips to Rome
• If on the track of the Emperors trips likely to
  be a rebellion" city since Emperor was defeated
  (as we know) cities with a stronger desire for
  independence had a greter chance to get it

45
Is Putnam Right? Yes even using IV!
46
Social Capital and religiosity

• Another dimension of social capital are social
  norms and beliefs
• Guiso, Sapienza and Zingales (2003) look at how
  religion shapes individuals beliefs and norms of
  behavior
• Data from World Values Survey
• A sample of about 1,000 individuals from a set of
  over 60 countries
• Information on demographics and, beliefs and
  attitudes, religiosity etc.

47
How do they define religiosity?

• 4 categories
• Raised religiously at home (54)
• Exogenous component, not linked to individual
  characteristics.
• Currently religious attend religious service at
  least once a year (59)
• Actively religious attend religious service at
  least once a week (24)
• Does not believe in God (15)
• They control for health, education, gender,
  income, and social status.

48
Effects of religion on civic attitudes
49
Determinants of social capital as an individual
attribute

• Glaeser et al. (2000) model network capital as a
  standard investment problem.
• Individuals have incentives to invest in
  networks they benefit from them but they are
  costly (e.g. dining out with others). They show
  that
• SC (number of organization memberships) decreases
  with expected mobility
• increases with home ownership
• increases and then decreases with age (hump
  shaped)
• it is higher in occupations that require more
  contacts
• falls with physical distance
• Is a complement to human capital (those who
  invest more in SC invest more in HC)

50
Trust
51
Trust and social capital

• In Putnam trust is just one component of social
  capital
• This is intuitive if social capital is a dense
  set of relations within a community, this set of
  relations can sustain a high level of reciprocal
  trust gt social ties sustain trust
• Hence social capital and trust should be highly
  correlated

52
Generalized trust and social capital (Italian
provinces)
Correlation 0.7
53
What is Trust?

• But trust is also distinct from social capital
• Social capital implies trust but the reverse is
  not necessarily true
• Definition trust is a belief about the
  action(s) of another person, in particular a high
  level of trust is a belief that a person in a
  relation with another has that the other person
  will not choose actions that are detrimental for
  him/her
• Trust is a prior about what you are going to do
  that has a bearing on me

54
Features of Trust

• Trust is an expectation about actions performed
  by others
• When trust is expressed (the expectation is
  formed) the action is not yet observed
• These actions must have an impact on the person
  forming the expectation
• Why do we care?
• Its presence or absence can affect what we choose
  to do and what we can do

55
Why do we care about trust?

• It is a basic ingredient in almost all
  transactions
• In Arrow (1972) words

"Virtually every commercial transaction has
within itself an element of trust, certainly any
transaction conducted over a period of time. It
can be plausibly argued that much of the economic
backwardness in the world can be explained by the
lack of mutual confidence."
56
Why do we care about trust? A simple framework

• Consider two parties A and B who can engage in
  some profitable trade.
• A has to spend cost c to find the value of the
  project
• Project can give Vhgt0 with probability p, and
  Vllt0 with probability (1-p)
• Surplus will be split equally with probability
  (1-?), but with probability ? B will extract all
  the surplus (cheat) trust (1-?).
• The ex-ante payoff of A is

57
Implication

• Investigate and trade if
• If the level of trust (1-?) is too small the
  trade opportunity will never be investigated and
  hence undertaken
• Example, Steve Jobs (Apple CEO)
• In his memoirs, when in 1980 IBM was desperately
  looking for an operating system for PCs, it
  looked at Apple and invited him to a meeting.
• Steve Jobs, fearing that IBM would extract all
  the surplus from any possible negotiation,
  declined to go and, in so doing, missed the
  opportunity to become a Microsoft.
• Hence, lack of trust may lead to first order
  losses

58
How are priors about (1-?) formed?

• This is the same as asking what determines trust
• Available information about type Bs attitudes
  and law enforcement surely matter.
• But as we will see other variables matter such as
  
• The social behavior of the area where
  individuals were born
• religious beliefs of the trusting person affect
  how much a person trusts another
• social psychology shows that individuals trust
  more people with similar ethnic background

59
Two concepts of trust

• Personalized trust is the trust that a person
  has about another well identified individual,
  e.g. the trust Giuseppe has about Filippo, his
  class mate
• Generalized trust is the trust that a given
  person has about a generic and unknown (randomly
  drawn) member of a well identified community,
  such as your classmates, the other Italians or
  the British
• It is the second to which political scientist
  refer and that is linked to social capital
• We will mostly be focusing on generalized trust

60
Measuring Trust
61
How to measure trust

• Trust is much easier to measure than other
  concepts, partly because it is a relatively well
  defined belief about some action
• Two general approaches
• Rely on some qualitative answer than aims at
  eliciting an individual attitude to trust others
  in general
• Rely on experiments that involve some choice that
  can be identified with the amount of trust one
  has

62
Qualitative measure of trust

• Well suited to measure generalized trust
• An well established question is asked in the
  World Values Survey
• Generally speaking, would you say that most
  people can be trusted or that you can't be too
  careful in dealing with people?
• The variable is equal to 1 if participants report
  that most people can be trusted and zero
  otherwise
• Useful to compute the fraction of individuals
  that trust in each country
• European Social Survey allows answers between 0
  and 10gt useful to get a measure of trust
  intensity

63
Trust across the world in the WVS fraction who
trust
64
Qualitative measure of trust

• An alternative, similar measure is in
  Eurobarometer In each EU country it asks
• How much trust you have in people from various
  countries. For each, please tell me whether you
  have
• 1 no trust at all,
• 2 not very much trust,
• 3 some trust,
• 4 a lot of trust
• Nationalities EU countries, but also Bulgaria,
  China, Czech Republic, Japan, Romania, Russia,
  Slovakia, Hungary, Poland, Slovenia, Switzerland,
  Turkey, United States
• Benefit one gets information not only about
  trust vis-à-vis own citizen but also about
  citizens of other countries gt a matrix of trust
  with also the trustworthiness of a country

65
(No Transcript)
66
Observations

• Considerable variation in the level of trust
• Home bias in trust
• Systematic differences in trust (citizens of some
  countries trust more than others, citizens of
  some countries are trusted more than others).
• Countries that have higher trust tend also to
  receive trust, that is are believed to be
  trustworthy

67
(No Transcript)
68
(No Transcript)
69
Measuring trust with trust games

• An alternative is to rely on a trust game
• There are two types of players, senders (S) and
  receivers ( R )
• Senders and receivers do not know each other and
  cannot communicate depending on the game one can
  specify some characteristics of R and S
• Each sender S is given a certain amount of money,
  say EUR 100
• S can send a fraction t between 0 and 1 of this
  money to R
• The amount sent becomes productive, and say
  doubles. Thus the receiver receives 2t100

70
Measuring trust with trust games

• R has to decide the fraction (between 0 and
  1) that he returns to S of the money received he
  can keep the whole 2t100 and return nothing
• One can get two measurers trust and
  trustworthiness.
• Trust concerns sending behavior and is measures
  by t . It refers to the propensity of a player to
  contact another player and to make high
  transfers, which can be interpreted as a
  propensity to trust the receiver to reciprocate.
• Trustworthiness concerns the action of
  reciprocating and is measured by

71
Features of trust in trust games

• The more one trusts the more he is willing to
  send, the more R receives and can potentially
  send back
• Lack of trust results in a lost opportunity
• The ideal setting is one where S sends 100 so
  that the sum doubles and R gets 200. But S is
  willing to send the whole sum if he believes that
  R returns at least 100 for this to be the case R
  trustworthiness must be at least 0.5
• In other words trust is S belief about the
  trustworthiness of R

72
Which method is better?

• Hard to say the qualitative one recommends for
  its simplicity and can be easily asked in a
  general questionnaire
• Trust game has the advantage that one obtains a
  more rigorous measure of trust but is more
  suitable for experiments and less for general
  surveys

73
Which method is better?

• The Dutch National Survey asks the World Values
  Survey question to a large sample of individuals
  as well as question similar to the game
• Suppose that a random Dutch person that you do
  not know personally receives by mistake a sum of
  1,000 Euros that belongs to you. He or she is
  aware that the money belongs to you and knows
  your name and address. According to you, what is
  the probability he or she will return the money
  to you? (Report a number between 0 and 100)
• The two measures are highly correlated

74
Which method is better?

• Glaeser et al. (2000b) ask the standard WVS
  question (do you trust others) to a set of
  Harvard undergraduates
• then they ask to the same set of people to play a
  trust game
• they show that students who claim to trust others
  are not more trusting, but more trustworthy
• This sheds some doubts on the reliability of
  questionnaires answers but they have only a
  small sample of Harvard students
• In fact Fehr et al. (2003) find then opposite
  WVS questions are related to sender behavior not
  to trustworthiness
• See Sapienza, Toldra and Zingales (207) for a
  reconciliation

75
Determinants of Trust
76
What determines trust?

• No general theory of where trust comes from and
  how it arises. Alesina and La Ferrara (2001)
  isolate 5 broad factors
• individual culture, traditions and religion
• how long an individual has lived in a community
• recent personal history of misfortune
• the perception of being part of a discriminated
  group
• characteristics of ones community, including how
  much trust there is

77
What determines trust?

• They use the General Social Survey to analyze
  the level of generalized trust within the US.
  They find that
• bad shocks lead to lower trust
• no difference across religions
• more diversity at the local level, less trust
• Guiso, Sapienza and Zingales (2003) use the WVS
  and find that religiosity and religion
  denomination matters
• Christians tend to trust more, no difference
  between Catholics and Protestants, Muslin do not
  trust

78
What determines trust?

• In so far as social capital sustains trust all
  the determinants of social capital, including
  history, affect the level of trust of the members
  of a community
• Tabellini finds that todays trust is affected by
  political institution in place over the 19th
  century across regions of Europe gt similar to
  Putnams story about the determinants of social
  capital in Italy

79
What determines trust?

• What about the bilateral trust that individuals
  of a given country have vi-a-vis individuals of
  another?
• Ichino et. al. run a trust game among a sample of
  students from various European countries where
  the sender can choose to send money to
  individuals of different nationalities
• They find that individuals from Southern
  countries trust less and are less trusted

80
Determinants of Relative Trust

• Guiso, Sapienza and Zingales (2004) use the
  Eurobarometer trust matrix. Focus on three types
  of variables
• Information
• Distance (in kilometers between the two largest
  cities)
• Common border
• Press coverage
• Cultural
• Common religion (probability)
• Number of years the countries have been at
  war(1000-1815, 1815-1970)
• Genetic distance between two countries/ distance
  in anthropometric measures
• Information or cultural?
• Common origin of the law , Common language

81
Determinants of Trust EU countries
82
Summary

• Results consistent with relative trust being
  driven (at least in part) by cultural stereotypes
• Information (as measured by press coverage) does
  not play a role common language has an impact.

83
Take outs

• In spite of a recent boom, the literature on
  social capital and trust is still in its infancy.
  
• There is little agreement on what social capital
  is, and how is it formed.
• The empirical evidence is still scant but
  progressing
• This makes it difficult yet to provide a definite
  view of the interaction between business behavior
  and social capital.
• Nevertheless, we do know that social capital and
  trust are very important for business and
  particularly for finance