Is Africa Caught in a Poverty Trap

1
Is Africa Caught in a Poverty Trap?
Preliminary and not for citation

• NBER April 21, 2006
• Andrew Berg and Carlos Leite
• Policy Development and Review Department
• IMF
• Loosely based on What Makes Growth Sustained?,
  by Andrew Berg, Carlos Leite, Jonathon Ostry, and
  Jeromin Zettelmeyer.

This presentation should not be reported as
representing the views of the IMF. The views
expressed in this presentation are those of the
author(s) and do not necessarily represent those
of the IMF or IMF policy. This describes research
in progress by the author(s) and is intended to
elicit comments and to further debate.
2
Introduction

• Poverty trap idea is now very influential
• What does it mean?
• Narrow definition feedback from poverty to
  growth leading to low-level trap
• Broad definition Any feedback mechanism that
  results in persistent poverty
• Mechanism is key

3
There are lots of proposed mechanisms, but only
some imply big push

• Savings
• Increasing returns to scale
• Human capital accumulation/credit market failures
• Health
• Conflict
• Corruption
• Etc.

4
Evidence on Poverty Trap

• Divergence Big Time
• Gap between rich and poor increasing
• Between 1870 and 1990 ratio of rich to poor
  increases from 91 to 451
• Average income gap between the richest and the
  rest grew from 1,286 to 12,000
• Poor countries stagnating at low incomes.
• Even in 1820, Europeans had per capita incomes of
  1,000 to 1,500
• Africa persistently falling behind, particularly
  since 1980

5
(No Transcript)
6
In Africa, not even coastal economies have
performed well ...
7
Basic facts of growth and poverty dont support
notion of persistent low-level equilibrium

• Poor countries are not a persistently
  well-defined group
• Easterly (2005) growth rates are not
  statistically lower in poor countries income
  levels are not stationary.
• There is lots of movement across quintiles of
  countries, including growth successes and growth
  disasters.

8
Growth Successes ...
9
(No Transcript)
10
And growth disasters ...
11
(No Transcript)
12
What about Growth Accelerations?

• Accelerations are common, even in Latin America
  and Africa (e.g. Berg, Leite, Ostry, and
  Zettelmeyer (2006?)).
• Most of these accelerations do not lead to
  take-offs
• Africas problem is more duration of growth
  spells and what happens after rather than taking
  off.
• This doesnt sound like a poverty trap in narrow
  terms.

13
Frequency of Growth Accelerations by Region,
1960-2002

14
Including spells initiated prior to the beginning
of the sample period ...
Growth Accelerations (p0.25) and Duration by
Region
15
(No Transcript)
16
Savings traps and increasing returns to scale
dont seem to fit the evidence

• Kraay and Raddatz (2005) try to see if increasing
  returns to production or increasing savings rates
  with income can generate poverty trap. In fact,
  savings rates dont behave as required.
• Nor does total factor productivity increase with
  per capital income in a way that would generate a
  poverty trap.
• Other micro evidence of poverty traps scarce.

17
Is That the Last Word? No, of course

• Maybe there are traps for a subset of countries,
  e.g. tropical landlocked countries
• Maybe the trap is about bouncing along the
  bottom

18
(No Transcript)
19
(No Transcript)
20
Do the Duration Regressions Shed Light on the
Trap?

• No evident role for aid, savings, or investment
  in driving duration (or incidence of
  acceleration)
• Terms-of-trade plays a role in duration. Maybe
  for these countries, accelerations are
  terms-of-trade related and so dont lead to
  take-off?
• A political-economy/income distribution story?

21

22

23
Where does this leave the Big Push?

• Remains an attractive story
• Little evidence for a resource-type story, either
  in role of aid or investment, or in pattern of
  accelerations.
• Maybe non-poverty-related sources of persistence?
• Institutions/corruption? Evidence is weak on
  causality from poverty to governance.
• There are take-offs, though. Maybe about
  export-led growth (Rajan and Subramanian Ostry,
  Subramanian, and Johnson)
• Republic of Congo

24
(No Transcript)
25
Where does this leave Africa now?

• Some increase in frequency of accelerations in
  Africa, and increase in average growth
• Prospects for persistence?
• Some seem tot-led
• Macro/inflation
• Income distribution
• Heavy aid?
• End on a positive note

26
(No Transcript)
27
Ongoing Growth Accelerations in SSA
28
(No Transcript)
29
(No Transcript)