What did Structural
Adjustment Adjust?
Easterly, JDE 76(1): = 1-22 2005<= o:p>
S=
tructural
Adjustment loans dating from 1980, with the McNamara era at the World Bank.=
‘The loans would
provide finance over a period of several years in return for reforms in tra=
de protection
and price incentives for efficient resource use’ (p. 2).
Came against the back=
drop of
the second oil shock – it was ‘intended as a preventative
instrument so that the “current account deficits of many developing
countries do not become so large as to jeopardize seriously the implementat=
ion
of current investment programs”’ p. (2).
Maintain growth, faci=
litate
balance of payments adjustment.
C=
omplementarity
between adjustment and growth.
Idea was a 3-5 year l=
oan to
help through a transition period.
F=
iscal
adjustment, getting prices right, trade liberalization, and, in general, a
movement towards free markets and away from state intervention. (=
p. 3).
IMF loans (often a
precondition for getting WB loans) stressed macroeconomic stabilization.
Issue with evaluating=
these
loans – they got them because there were problems. Selection based on observed outcom=
es may
lead to misleading inference.
Selection model (firs=
t stage
is selection in or not, second stage looks at impact of independent variabl=
es
on the dependent variable with a weight from the first stage in there). Sometimes call=
ed a
switching model.
M=
ixed,
unclear evidence. Did adjustment lending lead to
growth? Did adjustment lendin=
g have
an impact on policies?
Easterly notes that t=
hese
studies have treated adjustment loans as independent events. However, many =
loans
over time to a given country.
Table One: The 20 countries that got the most=
loans
and some evidence on conditions in the country. Compared to the developing country=
average
of 7 loans in a 20 year period, these countries had an average of 19.
Note at the bottom th=
at there
is not a lot to distinguish these countries from the developing country sam=
ple
for the macro indicators.
Sure does not say
convincingly that adjustment with growth was the outcome.
Figure two illustrate=
s the
conditional probability of getting another loan based on how many received =
to
date.
Half of all outstandi=
ng IMF
loans were to ‘prolonged users’ of IMF resources (7 of 10 year
period).
‘These results =
do not
prove that adjustment lending was ineffective in promoting good macroeconom=
ic
policies and good growth outcomes.
It may be that performance would have been even worse without intens=
ive
adjustment lending” (p. 9)
That said,
it is still important to ask:
=
1)&n=
bsp; Why did it not lead to better growth outcomes?
=
2)&n=
bsp; Why did they have to keep repeating the process?=
Bad shocks not convin=
cing
(pages 9-10 / table 2)
Table three illustrat=
es the
high adjustment lending are almost all (excepting
Transition cases (tab=
le 4).
“Again, this [d=
oes] not
prove that adjustment lending was ineffective…[b]ut
it places a bound on our intuition about the counterfactual – it is
necessary to believe that Ukraine would have had a worse outcome than -8.4%=
per
capita growth and 215% annual inflation without 10 adjustment loans. (p 11-12).
S=
ection 3.
Measure of distortion=
s
Inflation is greater =
than 40%
Black market premium =
is
greater than 40%
Real exchange is more=
than
40% overvalued
Real interest rate is=
less
than -5%.
Arguably, any of thes=
e mean
you are in trouble.
Figure 3: % of developing countries with any=
one
of these distortions.
While figure 3 shows =
some
decline over time (maybe it did work) figure four suggests not. The fraction h=
aving a
distortion as a function of the # of adjustment loans.
There was adjustment,=
but not
related to structural adjustment loans.
Table five: Macro distortion dummy on cumulati=
ve
number of loans and cumulative time spent under IMF programs and a time
trend.
[=
six
indicators adds in fiscal deficits and current account deficits bigger than=
5%
as other indicators that things are off track] By the way, the
Countries have adjust=
ed over
time, but not in a way related to the adjustment programs.
S=
ection 4.
Identification / reve=
rse
causality again.
Instruments: friends of donor variables.
Basically, it does no=
t alter
the overall impression in section 3, and not much evidence of reverse causa=
lity
(section three has policy outcomes conditional on loans, but what if loans =
are
conditional on policy outcome as in Dollar and Levin?).
Structural adjustment=
did not
succeed in adjusting macroeconomic policy and growth outcomes very much.
Counterpart to the fo=
reign
aid findings – aid did not clearly lead to growth, and here structural
adjustment did not clearly lead to growth directly or indirectly through po=
licy
change.
See bottom of page 20=
.
E=
conomic
Origins of Dictatorship and Democracy.
Sidetrack – Soc=
ial
Origins of Dictatorship and Democracy.&nbs=
p;
=
1)&n=
bsp; Bourgeois revolutions linked to capitalist path:=
=
2)&n=
bsp; Facist outcome from capitalist path:
=
3)&n=
bsp; Communist path:=
=
4)&n=
bsp; Weak impulse to modernization though democracy is
present:
Why are we looking at=
this?
Well, we just have se=
en that
democracy / freedom / rule of law …all these institutional things
‘matter’ for outcomes. <=
/span>
We have a good bit of
evidence that the ability to change policy from outside (structural adjustm=
ent
as an example) is limited.
Not anything clear (t=
hat I
know about) for governance, but my guess is about the same story will hold.=
Go back to Grief R=
11; how
do institutions evolve?
H=
istory,
case studies.
That is what we are d=
oing
now. Introducing case studies,
first to look at what makes democratic outcomes more likely.
Then country case stu=
dies to
develop our understanding of how reform does or does not happen.
Then return to lookin=
g at
market development and (time permitting) land tenure as ways the ‘rul=
e of
law’ develops.
OK, back to our regul=
arly
scheduled program.
A=
cemoglu
and Robinson.
P=
aths that
political institutions take over time.
4 stylized paths:
1)&n=
bsp;
Path that l=
eads
from nondemocracy gradually but inexorably to
democracy. Once created, never
threatened, and it endures and consolidates.
2)&n=
bsp;
Path leadin=
g to
democracy, but once created, it quickly collapses. The cycle repeats.
3)&n=
bsp;
The country
remains undemocratic, but is relatively egalitarian and prosperous, thus
stable.
4)&n=
bsp;
The country
remains undemocratic, but unequal and exploitative, so that the threat of
democracy challenges the interests of the elites, so they repress.
C=
reation of
regular Parliaments that were a forum for the aristocracy to negotiate taxes
and discuss politics with the king.
Civil War of 1642, Gl=
orious
Revolution of 1688. Monarchy versus Parliament.
Look at chapter to ge=
t the
idea, but the overall pattern is things building up to a crisis point, refo=
rm
to head off crises / unrest, reform was only partial so there is a repeat of
the process. …the threa=
t of
social disorder was the driving force behind the creation of democracy in <=
st1:country-region
w:st=3D"on">
1=
810
independence. Civil wars, conflict, 1860s things
settle down.
PAN stays in power un=
til
1916.
C=
hange of
party in 1916, new reforms.<=
span
style=3D'mso-spacerun:yes'> The old elite were not able to tra=
nsform
into a viable conservative alternative party.
Since the Radical par=
ty is in
power and the old elite are not able to mount a challenge within the politi=
cal
sphere, military coup in 1930 and then fraudulent elections. Conservatives =
in
control.
W=
hich leads
to another coup. Peron. This administr=
ation
more in lines with the radical / pro-labor approach than the former one.
W=
hich leads
to a coup in 1955.
W=
hich leads
to a coup in 1966.
Peron returns from ex=
ile in
1973.
W=
hich leads
to a coup in 1976.
H=
mmm. A=
pattern
is emerging, which is the point.
1819 Raffles gets it =
from a
local leader for the British East India Company.
6=
22 square
miles, a few hundred inhabitants.
E=
xpanded as
the
Post second
world war (and occupation) move toward independence.
L=
ack of an
aristocracy given the history.
1=
00%
urbanization rate.
75% Chinese, 15% Malay, 8% from the Indi=
an
subcontinent.
Business class is
predominantly foreign.
D=
utch East
India Company in 1652.
British seize control=
in
1806.
Boers move inland in =
1850s
/1860s.
Diamonds and gold dis=
covered
in 1870s. British move to ass=
ert control, wars ensue, by 1910 form Union of South
Africa.
ANC formed in 1912.
A=
partheid
after 1948, passes,
Ongoing rebellion and
violence –
S=
tate of
emergency in 1986, troops into the townships.
US
sanctions in late 1986.
These are illustrative
examples. What is underlying =
model
that can explain these divergent outcomes?
C=
hapter
two.
Begin by conceiving of
society as consisting of two groups:
T=
he elite
and the citizens. Citizens are more numerous.
[=
wording
– are the elite not also citizens?&n=
bsp;
And what happened to bourgeois and proletariat / peasants?]
Model is inherently b=
ased on
conflict between these groups.
Elites are rich and t=
hus
opposed to taxation.
Citizens are poor and=
thus
pro-taxation.
Politics and social c=
hoices
will impact these groups differently.
Of course, elites are=
not
always rich to start with. Bo=
ers
not always rich, military not always rich,… But overall no=
t a bad
approximation.
Parsimony again raise=
s its
head.
What is a democracy?<= span style=3D'mso-spacerun:yes'> Many different institutional forms. <= o:p>
Then ask the question=
about nondemocracy.
What types of forms can be put in that bin?
The commonality there=
is that
the government represents the preferences of a subgroup of the people, the
‘elite’.
Democracies approxima=
te a
situation of political equality.
So one way of thinkin=
g about
this is what moves a system towards or away from political equality.
=
3)&n=
bsp; Building blocks.
First:
E=
conomic,
rational choice model.
Individuals have well=
defined
preferences over outcomes or the consequences of their actions.
People behave strateg=
ically,
decide strategically.
Game theory is the to=
ol to
use then, as people decide individually on what is best for them, these
decisions interact.
M=
ultiple
decision makers acting strategically.
“Most important=
, we are
getting a license to focus on the consequences of the regimes” (p. 19)
This is in contrast t=
o saying
people have a taste for or against democracy.
While there may be an
ideological preference, the approach here is to place more emphasis on what=
is
delivered – the outcome, not the direct consumption.
Second:
I=
nherently conflictual.
Policy choices create
distributional conflict. Poli=
cy
choices create winners and losers. <=
/span>
While the focus is on
individuals making decisions, clearly there can be groups will common
interest. Collective ac=
tion
may be possible, and may characterize this conflict.
If so, a decision wil=
l favor
one group over another. Which=
group
will be favored? The
one with political power.
Political power is the
capacity of a group to obtain its favorite policies against the resistance =
of
other groups (p. 21).
Two types can be
considered: de facto and de jure.
De facto revolves
around the use of force.
De jure
relies on power allocated by the political system as an outcome of a politi=
cal
process.
Third:
The social and politi=
cal
arrangements that allocate de jure<=
/i>
political power are political institutions.
The most important
institutions are those which determine who gets to take part in political
decision making.
3.&n=
bsp;
Towards our=
basic
story
Building
from there, the conflict is often between wealthy elites who do not favor
democracy and citizens who do.
The
majority of citizens wants democratic institutio=
ns
because they benefit from them.
Page
23 “=
;However,
if the citizens have sufficient political power, why don’t they use t=
his
power to simply obtain the social choices and policies that they prefer rat=
her
than first fight for democracy and then wait for it to deliver the policies=
for
them?”
There
is today and tomorrow in the model.
Policies
can change between today and tomorrow.&nbs=
p;
However,
institutions are durable. They
persist over time. Political
institutions allocate political power now and in the future.
Transition
to democracy emerges as a way of regulating future allocation of political
power.
Securing the right to things, rather than the th=
ings
themselves.
De
facto political power it transitory – you may have it today, but
uncertainty about tomorrow.
For
the citizens, the nature of their de facto power lies with their ability to
rise up. To do so, they have =
to
solve collective action problems.
But
why would the elites (who control de jure power) agree to institutional ref=
orms
that will reduce their benefits over time?=
Why not give in on other things?
First,
giving in on policies is not a credible response. In response to revolutionary press=
ure,
give in on desired policy, pressure dissipates, then rescind policies.
A
credible reform has to change future political power.
Why
don’t they revolt anyhow? A sense that even if it succeeds, revolution is costly for a
society.
They
argue that the history of transitions to democracy centers around
significant social turmoil and revolutionary threats.
Quotes
on 27-28.
“Democracy is usually not given by the elite because its values have changed.” <= o:p>
“Democracy aris=
es when
concessions are not credible and repression is not attractive because it is=
too
costly.”
What leads to democra=
tic
consolidation?
A set of institutions=
that
characterize democracy endure through time.
S=
ome
elements that influence democratization and consolidation.
C=
ivil
society.
S=
hocks and
crises.
S=
ources of
income and composition of wealth.
La=
nd
easier to tax than other forms of capital.
Landowners more insula=
ted
from turbulence.
Agriculture allows for
some kinds of institutions.
Why revolution may be=
more
likely in agrarian societies, and concessions more likely in industrial soc=
ieties.
Form of political
institutions.
Inter-group inequality
(inverted U shape argument).
R=
ole of the
middle class.
“Almost all
revolutionary movements were led by middle class actors=
’ (p. 39)
Also has a role as a =
buffer –
they are aligned with the citizens in wanting more political power, but by
relative advantage have in interest in preserving existing outcomes.
G=
lobalization.
Financial security
Returns to labor / cap=
ital
modified
Gains to trade create
interest in stability
In=
ternational
pressure.
I=
dentities. As
long as interests are tied to identities, the basic model holds.
Look through graphs in
chapter three.