Economics ECON 4915 Lecture 12 Andreas Kotsadam Outline Long run effects of the slave trade Nunn and Wantchekon 2009 Recap recap and recap Nunn and Wantchekon 2009 ID: 426371
Download Presentation The PPT/PDF document "Development" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Development Economics ECON 4915 Lecture 12
Andreas KotsadamSlide2
OutlineLong run effects of the slave trade (Nunn and Wantchekon 2009).Recap, recap
, and
recap
. Slide3
Nunn and Wantchekon (2009)Research question: Did the slave trade cause a persistent culture of mistrust?
Interesting
?
Yes
, long
run
effects
of
slavery
. Culture and
development
. The
origins
of
trust. Test the
causal
mechanisms
behind
the
results
in
Nunn
(2008).
Original?
Yes, their causal channel has not been credibly tested before.
Feasible
?
Yes, by collecting innovative data and using IV.Slide4
Why would slave trade affect trust today?
Early in the slave trade, slaves were primarily captured through state organized raids and
warfare
.
By the end of the
trade individuals
- even friends and family members - began to turn on one another, kidnapping, tricking, and selling each other into slavery
.
This
environment, where everyone had to constantly be on guard against being sold or tricked into slavery by those around them,
generates a culture
of
mistrust.Slide5
Table 1 from wp versionSlide6
But why do we find these effects today
?
Using ‘rules-of-thumb
’ can be an optimal
strategy
in environments where information acquisition is either costly or
imperfect.
T
hese
rules-of-thumb
evolve
according to which norms yield the highest payoff.
”
Our
view is that in areas more exposed to the slave trade, rules-of-thumb or beliefs based on the mistrust of others would have been more beneficial relative to norms of trust and therefore
would have become more prevalent over time
.
”Slide7
Cultural transmissionFor the explanation to work the culture
must be transmitted over
time
or
cultural
change
and social
learning
must
occur
slowly
.
Note
that
these
aspects
are
not
tested
in the paper and in
that
sense
cultural
transmission is still a black box. Slide8
Data: Afrobarometer contains questions on trust. Table 2 from wpSlide9
Data: Ethnicity level slave exports.The country-level estimates are constructed by combining data on the total number of slaves shipped from all ports and regions of Africa with data on the ethnic origins of slaves shipped from Africa.
The individual samples are originally from a variety of historical records, such as slave runaway notices, plantation inventories, marriage records, death records, etc. Slide10
Data: Ethnicity level slave exports.The ethnic groups in the historic documents are matched
to a common classification
scheme and linked to the map provided by Murdock (1959).Slide11Slide12Slide13
EstimationOLS:i, individualse, ethnic
groups
,
d,
districts
,
c,
countries
. Slide14
Basic results (1)Slide15
Basic results (2)Slide16
How do we know it is causal?”The negative correlation between slave exports and trust … is consistent with our hypothesis that the slave trade engendered a culture of
mistrust
.
”
“However
, the correlation could also be explained by omitted
variables that
are correlated with selection into the slave trade and with subsequent trust
.”Slide17
Example of problemLess trusting groups may have been
more
likely
to
be taken
during
the
slave
trade
and
they
continue
to
be less
trusting
today
for
other
reasons
. Slide18
3 solutionsControll for many things at the ethnicity level.Use selection on observables to say something about possible selection on unobservables.Use IVSlide19
ControlsTry to capture colonial rule and initial conditions by controlling for:Disease environment (following AJR), precolonial population density, historical location of railways, and missionary stations.Slide20
IVWe need an instrument that is correlated with slave exports, but uncorrelated with any characteristics of the ethnic groups that may affect trust today. Historic distance of each ethnic group from the coast is used as an instrument for the number of slaves taken during the slave trade.Slide21
ValidityThe critical issue is whether an ethnic group’s historic distance from the coast is correlated with factors other than the slave
trade that may affect how trusting the ethnic group is today
.
What
is the
most
obvious
problem?Slide22Slide23
ValidityControlling for current distance to the coast,
identification
will
be driven by
the effect of the slave trade on the trust of individuals that live in a
location different
from their ancestors.
These
may
be different
types
of
people
,
but
they
show
that
,
if
anything
,
are
more
trusting
.Slide24
ResultsThe first stage shows that the instrument is relevant and that
slave
exports is
primarily
explained
by
historic
distance
rather
than
current
distance
from the
coast
.
The second
stage
estimates
show
similar
magnitudes
as OLS.
Hence
, OLS is
likely
not
biased
.Slide25
Falsification tests“As is generally the case with instruments, it is possible that despite our second stage controls, our instrument still does not satisfy the necessary exclusion restriction. For this reason, we also perform a number of falsification exercises to assess the validity of our identification strategy.”Slide26
Falsification testsIf the assumption is satisfied, then we should not observe a similar positive relationship between distance from the coast and trust in the parts of the world where the slave trade did not occur.They
find
no relationship
between distance from the coast and trust outside of Africa
.
Or in
African
countries
not
affected
by the
slave
trade
.Slide27
MechanismsAre people less trusting today via the evolution of vertically transmitted norms?Or because the slave trade
damaged
the institutions so
that
people
are
less
trusty
due
to
e.g
.
weak
rule
of
law
?Slide28
TestThey create a measure quantifies the number of slaves that were taken from the geographic location where the individual is
living today
.
If the slave trade affects trust
through
internal factors, then
mistrust
should be correlated with the extent to which
their ancestors
were heavily impacted by the slave trade.
If the slave trade affects trust
through
its deterioration of domestic
institutions,
then mistrust should be correlated with whether the external environment that the individual
is living in today
was heavily impacted by the slave trade.Slide29
ResultBoth factors are significant but
the magnitude of the internal channel is
approximately twice
the magnitude of the external channel.Slide30
External validity“it is important to keep in mind that all of the results in the paper apply only to the 17 sub-Sahara African countries included in our sample. The effects of the slave trade for the countries not included in our analysis may be different from the effects we estimate here.”What
about
other
shocks
? Slave
trade
in
Norway
.
What
about
the
mechanisms
?Slide31
RecapEmail me if you want something
specific
covered
during
next
lecture
.Slide32
Rural credit marketsYou should have some thoughts
about
:
Why
the
poor
often
cannot
borrow
on the formal market.
Why
the
poor
pay
so
much
interest
on
their
loans
,
if
they
are
able
to
borrow
.
The
role
of
institutions.
What
can
be
done
to
improve
the situation.Slide33
Why is credit important?Credit is needed for efficient production as well as smoothing out consumption. Production requires investments.
Income streams often fluctuate.Slide34
There are two basic (and related) problemsMoral hazard: Lenders cannot monitor the actions of the borrowers.Adverse selection: Lenders cannot distinguish between borrowers with different characteristics. Slide35
These problems are severe for formal lenders They don´t have personal knowledge regarding the clients.They cannot monitor how the loans are used.Limited liability implies that borrowers take to much risk or default voluntarily.Collaterals may solve this problem, but this is infeasible for the poorest. Slide36
Characteristics of rural credit marketsInformation problems (also for informal lenders) leading to:High interest rates.Segmentation.Interlinkage.Interest rate variation.
Rationing.
Exclusivity.Slide37
Lender’s risk hypothesis for informal lendingAssume perfect competition.Let L= Loan amount,
r= Lender’s opportunity cost,
p= Fraction of loans repaid, and
i= Interest rate.Slide38
The expected profit of the lender is therefore:
Setting expected profits equal to zero
and
solving for the interest rate gives
:Slide39
Main lesson of the modelHence, even under competition informal sector interest rates are very sensitive to the default risk.
But
is it
true
?Slide40
True with an important twistLooking at data it is obvious that defaults are quite rare in rural credit markets.
So, this mechanism of potential default is largely circumvented but this is costly.
This cost is basically what drives the observed high interest rates.
And
since
some
of
these
costs
are
fixed
, small
loans
demand
a
higher
interest
rate.Slide41
Information assymetries and rationingInformation assymetries may cause credit rationing as
lenders
are
not
able
to
fully
observe
if
a
borrower
is
of
high
or
low
risk.
Too
high
interest
rates
may
drive
away
the
low
risk
type
of
borrowers
.
It
may
therefore
be optimal
to
have
a
lower
interest
rate and a
higher
probability
of
receiving
the
money
back.Slide42
Now we know the theory behind: Why interest rates are high.Why there is credit rationing.It all has to do with information asymmetries in the following way:Slide43
Adverse selection and moral hazardAdverse selection: If banks raise interest rates the project mix will become riskier.Moral hazard: If interest rates increase, borrowers themselves choose more risky projects and/or put in less effort to repay.Slide44
PoliciesA nice (but not so simple) solution would be to build
up good institutions and eliminate
poverty.
In the
meantime
,
we
have
covered
:
Government intervention to expand credit (Burgess and
Pande
2005)
.
Microfinance (
Banarjee
and
Duflo
2010)
.Slide45
Typical exam question2a) Give some arguments for and against the idea that a state led expansion of rural banks should reduce poverty (2 points). 2b) If we are interested in the effects of rural banks on poverty, why is it a bad idea to draw conclusions by simply comparing poverty in areas that have banks to poverty in areas that do not have banks? (1 point) Slide46
Typical exam question2c) Burgess and Pande (2005) instead use a policy rule in India between 1977 and 1990 that forced banks who wanted to open in a location that already had banks to open banks in four areas that had no banks. In particular, they exploit the trend reversals between 1977 and 1990 and between 1990 and 2000 (relative to the 1961- 1977 trend) in the relationship between a state's initial financial development and rural branch expansion as instruments for branch openings in rural unbanked locations. What arguments are provided for using these instruments? (4 points) Slide47
Typical exam question2d) What are their conclusion and how can it be criticized? (3 points)Slide48
Their conclusion“We provide robust evidence that opening branches in rural unbanked locations in India was associated with reduction in rural poverty.” Slide49
Critical questionsHave they really showed
that
rural banks
matter
or just
that
this
policy
had
effects
?
Does it
matter
that
the bank
openings
were
not
randomly
assigned
?
Is the
result
generalizable
to
other
contexts
?
Do
we
know
why
the reform
had
an
effect
?
What
about
the long term
effects
?
Was
it
cost
effective
? Slide50
Typical exam question3a) Banarjee and Duflo (2010) define microcredit as innovations that lower the administrative cost of making small loans. Describe these innovations and discuss their advantages and disadvantages (5 points). Slide51
Some innovations and mechanismsDynamic incentives.Group liability.Repayment frequency and social interactions.Simplified collection technology.Temptation and self-control.Slide52
Dynamic incentives.Default implies a lost opportunity of
larger
loans
in the
future
.
Theoretically
,
dynamic
incentives
cannot
work
alone
…
… and
competition
may
undermine
them
.Slide53
Group liability.Default by one member hurts the other members.This should make clients invest in screening and monitoring. But the drawback may be too little risk-taking.Empirical evidence suggests joint liability is not the driving factor. Slide54
Repayment frequency and social interactions.Weekly repayment is the typical time period.Evidence suggest that longer time periods are better for investment…
…but worse for default.
Compatible with a social capital story, which actually recieves empirical support.Slide55
Simplified collection technology.The costs of collecting
the
loans
are
very
low
.
Loan
officers
are
able
to
collect
payments
from
many
people
each
day
and it
becomes
very
easy
.Slide56
Temptation and self-control.What if
borrowers
have
self
control
problems?
Wouldn’t
easy
credit
make
this
worse
?
It
actually
seems
to
be the
other
way
around
:
m
icrocredit
helps
people
commit
to
a
savings
plan.
But
is it the
best
way
to
achieve
commitment
?Slide57
InsuranceThe problem of risk.Why doesn’t insurance markets work well?What can be done? Slide58
To reduce risk, smoothing of consumption is necessary
How
smooth
consumption
?
One
way
is via
credit
.
We
have
already
seen
that
this
is not
very
easy
.
Another
way
is via
self-insurance
.Slide59
Typical exam question4b) If we have data on individuals in a village and we observe that the change in consumption perfectly follows the change in village income and is completely unrelated to changes in income at the household level. Is this evidence of perfect insurance? (3 points)