Transcript Reforming Distorted Political Economies

Conflicts, Security and Development: Capturing potential
complementarities from econometric and case study approaches
Richard Auty (Lancaster University)
Thesis
Closer links with case studies can benefit econometric analysis because case
studies:
1. Help set tightly focused econometric studies in their broader developmental
context
2. Generate insights to suggest sharper proxy variables for econometric
analysis
1
Growth Collapses Often Precede Strife: What Successful Cases
Reveal
Mansoob identifies institutions as key transmission mechanism,
May risk making institutions latest in Easterly’s line of ‘failed’ silver bullets
because analysis of growth collapses and recovery point to stronger roles
for factors other than institutions.
Assumes point linkages adversely affect development and associated with
mines (true) + plantations (dubious) whereas diffuse linkages of peasant
cash crop production are beneficial for growth (true), but:
a) If so, SSA countries like Ghana should outstrip Asian countries like
Malaysia
b) Conversion of diffuse linkages to point linkages appeared sanctioned
by interventionist economic policies, notably ISI that became
fashionable during 1950s + 1960s.
2
Natural resource endowment and growth collapses
Median GDP per capita (constant 1995 US$)
resource-rich and resource-poor developing countries
1200
1000
800
600
400
200
0
1960
1965
1970
1975
Resource-rich
1980
1985
1990
1995
Resource-Poor
3
Growth Collapses 1973-85 of Economies Weakened by
Two Decades of Policy-Induced Cumulative Distortion
Triggered New Policies to Shrink Role of Predatory
Government, Which Worked More Effectively In:
i) Large economies than small ones
ii) Resource-poor economies than resource-rich
iii) Countries near dynamic de-industrialising advanced economy >
countries in remote regions/ near large moribund economies
iv) Market reform incremental + endogenous, rather than rapid +
externally imposed.
Institutions don’t show up here, let alone dominate
4
Natural Resources, Growth Collapses and Incentives for
Civil Strife
Growth collapses may not be the ‘trigger’ for civil strife but provide
conditions in which such triggers easily emerge.
Consistent with Collier (2000), civil strife strong where
a. Primary exports,
b. Economic decline (i.e. a growth collapse), which leaves a
relatively large young male population (also linked to growth
collapse because collapse retards demographic cycle), with little
education so conflict offers immediate financial gain.
In this broader context, civil strife exacerbates degradation of capital
that is the legacy of a growth collapse and injects a desire for revenge
as an added complication in seeking post-war reconstruction.
5
Factoring in Relationship Between Civil Strife + Scale of
Rent
Literature suggests positive relationship between civil conflict + drugs:
Expansion of illicit drug trade draws domestic economic activity out of the
formal economy and thereby diminishes government revenue while
increasing public expenditure on security = undermines government
legitimacy.
Results suggest:
i) Both share of taxes in GDP + share of income taxes within total taxes
are lower in high-drug economies.
ii) Corruption + civil strife are higher in drug-producing countries.
iii) Indirect link: both lower government expenditure + perceived
government corruption feed civil strife, but perceived corruption seems
more influential (cuts government moral authority).
6
Drug-Fuelled Civil Strife as an Extreme Case of Rent
Capture (whether natural resource rent, geopolitical rent
(foreign aid) or contrived rent (change relative prices))
Rent is > 40% of GDP in most LDCs.
a) High rent encourages governments to redistribute at expense of
incentives for wealth creation. Scale of LDC rent makes little wonder
many governments corrupted when invited to override markets
b) Drugs = extreme on continuum of rent capture + deployment, with
resource-poor countries at other extreme (e..g. Mauritius + East Asian
four).
c) Corrective policy = practical strategy to re-channel rent from
redistribution, which can spiral into violence, -> broad-based wealth
creation.
7
Share of Rents in GDP 1994 and GDP Growth
1985-97 (World Bank 1999)
Resource endowment
PCGDP
growth 198597
Total rent
(%GDP)
Pasture &
Cropland rent
(% GDP)
Mineral
rent (%
GDP)
Large
4.7
10.56
7.34
3.22
Small
2.4
9.86
5.41
4.45
Large
1.9
12.65
5.83
6.86
Small, non-mineral
0.9
15.42
12.89
2.53
Small, hard mineral
-0.4
17.51
9.62
7.89
Small, oil
-0.7
21.22
2.18
19.0
15.03
8.78
6.25
Resource Poor
Resource Rich
All Countries
8
Can More Nuanced Classifications Shed More Insight into
Ethnicity, Civil Strife + Public Expenditure?
Findings clear and refute literature: Factionalism does not increase
militarization and so cannot cause the observed lower public spending.
Heterogeneity may depress long-term growth because sweeteners (rent)
may be key part of process by which ethnic groups resolve tensions in
heterogeneous societies, at the expense of public expenditure.
May be able to say more about ethnic heterogeneity by deploying more
nuanced classification of LDC political state, beyond democracy +
autocracy
9
Evolution of Political Accountability Under High Rent
Autonomy of
State
Basic Aims of
State
Critical
Features
Rent Pattern
Predatory Autocracy
Maximise elite rent
siphoning, through
force if necessary
High rent, violent
predation, staple trap
trajectory
Point rent extraction
by elite depresses
GDP growth
Concentrated
Oligarchy
Dominant faction
captures policy to
sustain rent + power
High rent, unequal
asset shares, staple
trap trajectory
Point extraction, but
some public goods
benefit mainly elite
Democracy polarised
so policy swings
retard GDP growth
Rent extraction +
distribution to clients
> PCGDP growth
Polarised Democracy Capture rent for
clients, even if slows
GDP growth
10
Evolution of Political Accountability Under Low Rent
Autonomy of
State
Basic Aims of
State
Critical
Features
Rent Pattern
Benevolent
Autocracy (Nation
Builder)
Secure rapid GDP
growth to sustain
elite + build social
unity
Low rent, external
threat, poor have low
opportunity cost
Low rent siphoning,
efficient diffuse rent
raising + dispersal
Diffusing Oligarchy
Co-opt new rich into
elite to deter policy
capture + sustain
rapid GDP growth
Low rent, intra-elite
rivals; rapid low gini
PCGDP growth
Low diffuse wealth
extraction for public
goods + broadened
wealth creation
Consensual
Democracy
Growth then equity
by providing basic
social entitlements
Low rent, middle
class growth dilutes
elite + shrinks poor
Diffuse extraction +
dispersal for growth
with equity
11
Conclusion
It might help while flying at 40,000 feet to have more input from ground
control.
For example:
1. Civil strife appears to grow out of growth collapses and the cumulative
degradation of all forms of capital which that process entails
2. Patterns of rent flows derived from case studies reveal a lot about the
process of economic development.
3. A more elaborate classification of political states should remove the
blurring that arises from a simple dual classification, for example
12