Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
MIT OpenCourseWarehttp://ocw.mit.edu
14.772 Development Economics: Macroeconomics Spring 2009
For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.
The Long-Term Effects of Africa’s Slave Trades
Nathan Nunn
Table 1: Africa and the Rest of the World.
Mean std. dev. Min Max N
Real per capita GDP in 2000
Rest of World 8,809 7,591 521 37,471 124
Africa 1,833 2,002 218 10,652 52
Growth in real per capita GDP, 1960–2000
Rest of World 2.00 1.71 −3.54 6.41 94
Africa 0.70 1.65 −3.11 5.94 52
Historic Causes of Africa’s Underdevelopment?
• Colonialism
There is no doubt that a large number of negative
structural features of the process of economic
underdevelopment have historical roots going back to
European colonization. – Paul Bairoch (1993).
• The slave trade
Slavery was corruption: it involved theft, bribery, and
exercise of brute forces as well as ruses. Slavery thus
may be seen as one source of precolonial origins for
modern corruption. – Patrick Manning (1990)
Empirical Questions
• Can Africa’s poor performance be partially explained by the
slave trade?
• If so, what are the channels?
Summary of Paper
• History of Africa’s slave trades.
• Construct country-level slave export figures.
• Estimate the relationship between slave exports and economic
development across countries within Africa.
• Deal with the econometric problems: causality and
measurement error.
• Examine the potential channels of causality.
Africa’s slave trades Figure by MIT OpenCourseWare.
Magnitude of the Slave Trades
Slave Trade 1400–1599 1600–1699 1700–1799 1800–1900 1400–1900
trans-Atlantic 230,516 861,936 5,687,051 3,528,694 10,308,197
trans-Saharan 675,000 450,000 900,000 1,099,400 3,124,400
Red Sea 400,000 200,000 200,000 505,400 1,305,400
Indian Ocean 200,000 100,000 260,000 379,500 939,500
Total 1,505,516 1,611,936 7,047,051 5,512,994 15,677,497
Channels of causality
• Ethnic fractionalization
– Ties between village federations deteriorate
– Villages raid one another
• Political fractionalization and weakening of states
– Intra-community slave raiding and kidnapping
– Kingdoms experience civil wars
• Break-down of traditional judicial systems
– Judicial system comes to be used as tool for enslavement
Data Sources: Shipping Records
• Atlantic slave trade.
– Know port of embarkation.
• Indian Ocean slave trade.
– Know region of embarkation.
• Saharan slave trade.
– Know slaves’ destinations.
– Know which caravan slaves arrived on.
• Red Sea slave trade.
– Know port of embarkation.
Data Sources: Ethnicity Data
• Atlantic slave trade.
– 53 samples, 80,656 slaves, 229 ethnicities
• Indian Ocean slave trade.
– 6 samples, 21,048 slaves, 80 ethnicities
• Saharan slave trade.
– 2 samples, 5,385 slaves, 23 ethnicities
• Red Sea slave trade.
– 2 samples, 67 slaves, 32 ethnicities
Table 2: Slave Ethnicity Data: Trans-Atlantic Slave Trade, 1450–1799
Num. Num. Region Years Ethnic. Obs. Record Type
Valencia, Spain 1482–1516 77 2,675 Crown Records Puebla, Mexico 1540-1556 14 115 Notarial Records Dominican Republic 1547–1591 26 22 Records of Sale Peru 1548–1560 16 202 Records of Sale Mexico 1549 12 80 Plantation Accounts Peru 1560–1650 30 6,754 Notarial Records Lima, Peru 1583–1589 15 288 Baptism Records Colombia 1589–1607 9 19 Various Records Mexico 1600–1699 28 102 Records of Sale Dominican Republic 1610–1696 33 55 Government Records Chile 1615 6 141 Sales Records Lima, Peru 1630–1702 33 411 Parish Records Peru (Rural) 1632 25 307 Parish Records Lima, Peru 1640–1680 33 936 Marriage Records Colombia 1635–1695 6 17 Slave Inventories Guyane (French Guiana) 1690 12 69 Plantation Records Colombia 1716–1725 33 59 Government Records French Louisiana 1717–1769 23 223 Notarial Records Dominican Republic 1717–1827 11 15 Government Records South Carolina 1732–1775 35 681 Runaway Notices Colombia 1738–1778 11 100 Various Records Spanish Louisiana 1770–1803 79 6,615 Notarial Records St. Dominique (Haiti) 1771–1791 25 5,413 Sugar Plantations Bahia, Brazil 1775–1815 14 581 Slave Lists St. Dominique (Haiti) 1778–1791 36 1,280 Coffee Plantations Guadeloupe 1788 8 45 Newspaper Reports St. Dominique (Haiti) 1788–1790 21 1,297 Fugitive Slave Lists Cuba 1791–1840 59 3,093 Slave Registers St. Dominique (Haiti) 1796–1797 56 5,632 Plantation Inventories
Table 3: Slave Ethnicity Data: Trans-Atlantic Slave Trade, 1800–1900
Num. Num. Region Years Ethnic. Obs. Record Type
American Louisiana 1804–1820 62 223 Notarial Records Salvador, Brazil 1808–1842 6 456 Records of Manumission Trinidad 1813 100 12,460 Slave Registers St. Lucia 1815 62 2,333 Slave Registers Bahia, Brazil 1816–1850 27 2,666 Slave Lists St. Kitts 1817 48 2,887 Slave Registers Senegal 1818 17 80 Captured Slave Ship Berbice (Guyana) 1819 66 1,127 Slave Registers Salvador, Brazil 1819–1836 12 871 Manumission Certificates Salvador, Brazil 1820–1835 11 1,106 Probate Records Sierra Leone 1821–1824 68 605 Child Registers Rio de Janeiro, Brazil 1826–1837 31 772 Prison Records Anguilla 1827 7 51 Slave Registers Rio de Janeiro, Brazil 1830–1852 190 2,921 Free Africans’ Records Rio de Janeiro, Brazil 1833–1849 35 476 Death Certificates Salvador, Brazil 1835 13 275 Court Records Salvador, Brazil 1838–1848 7 202 Slave Registers St. Louis/Goree, Senegal 1843–1848 21 189 Emancipated Slaves Bakel, Senegal 1846 16 73 Sales Records d’Agoue, Benin 1846–1885 11 70 Church Records Sierra Leone 1848 132 12,425 Linguistic and British Census Salvador, Brazil 1851–1884 8 363 Records of Manumission Salvador, Brazil 1852–1888 7 269 Slave Registers Cape Verde 1856 32 314 Slave Census Kikoneh Island, Sierra Leone 1896–1897 11 185 Fugitive Slave Records Total 80,656
Constructing Estimates
↑ AFRICA
Country A Country B
Country C
Country D
Country E
NAtlantic
Ocean 100, 000 ⇐
250, 000 ⇐
From the ethnicity data, I calculate:
A : B = 4 : 1
C : D : E = 3 : 1 : 1
Calculations
Atlantic
Ocean
100, 000 ⇐ 250, 000 ⇐
AFRICA
80,000 20,000
150,000
50,000
50,000
↑
N
A= 100, 000 × 4/5 = 80, 000
B= 100, 000 × 1/5 = 20, 000
C= 250, 000 × 3/5 = 150, 000
D= 250, 000 × 1/5 = 50, 000
E= 250, 000 × 1/5 = 50, 000
Complications
• Movement between coastal countries or ‘diagonal’ movements
from inland to coastal countries.
– Estimated upper-bound is 15%.
• Under-sampling of interior slaves.
– Direction of bias is known.
– Can use IV.
• Countries are not laid out on a grid.
Total Slave Exports from 1400 to 1900 by Country.
Trans- Indian Trans- Red All
Atlantic Ocean Saharan Sea Slave Share
Country Trade Trade Trade Trade Trades of Total
Angola 3,616,027 0 0 0 3,616,027 23.1%
Nigeria 1,411,758 0 555,796 59,337 2,026,891 12.9%
Ghana 1,603,335 0 0 0 1,603,335 10.2%
Ethiopia 0 0 813,899 633,357 1,447,256 9.2%
Mali 524,102 0 509,950 1,034,052 6.6%
Sudan 615 0 408,260 454,913 863,788 5.5%
Dem. Rep. of Congo 752,828 0 0 0 752,828 4.8%
Mozambique 382,337 274,024 0 0 656,402 4.2%
Chad 823 0 409,367 118,673 528,863 3.4%
Tanzania 10,834 507,595 0 0 518,429 3.3%
Benin 461,782 0 0 0 461,782 2.9%
Senegal 222,359 0 98,732 0 321,091 2.0%
Togo 280,842 0 0 0 280,842 1.8%
Guinea 242,691 0 0 0 242,691 1.5%
Baseline Estimating Equation
ln yi = β0 + β1 ln(exportsi/areai) + C ′ δ + Xi′ γ + εii
• ln yi is the natural log of real per capita GDP in country i in
2000
• ln(exportsi/areai) is the natural log of the total number of
slaves exported between 1400 and 1900 normalized by land
area.
• Ci is a vector of dummy variables that indicate the origin of
the colonizer prior to independence.
• Xi is a vector of control variables that are meant to capture
differences in countries’ geography and climate.
Angola
Burundi
Benin
Burkina Faso
Botswana
Central African Republic
Ivory Coast
Cameroon
Congo
Comoros
Cape Verde Islands
Djibouti
AlgeriaEgypt
Ethiopia
Gabon
Ghana
Guinea
Gambia
Guinea−Bissau
Equatorial Guinea
Kenya
Liberia
Libya
Lesotho
Morocco
Madagascar
Mali
Mozambique
Mauritania
Mauritius
Malawi
Namibia
Niger
Nigeria
Rwanda
Sudan
Senegal
Sierra Leone
Somalia
Sao Tome & Principe
Swaziland
Seychelles
Chad
Togo
Tunisia
Tanzania
Uganda
South Africa
Democratic Republic of Congo
Zambia
Zimbabwe
5
7.5
1
0
Av
erag
e in
com
e p
er p
erso
n i
n 2
00
0
−4 0 5 11 Slave exports normalized by land area
Relationship Between Current Income and Past Slave Exports
Angola
Burundi
Benin
Burkina Faso
Botswana
Central African Republic
Ivory Coast
Cameroon
Congo
Comoros
Cape Verde Islands
Djibouti
AlgeriaEgypt
Ethiopia
Gabon
Ghana
Guinea
Gambia
Guinea−Bissau
Equatorial Guinea
Kenya
Liberia
Libya
Lesotho
Morocco
Madagascar
Mali
Mozambique
Mauritania
Mauritius
Malawi
Namibia
Niger
Nigeria
Rwanda
Sudan
Senegal
Sierra Leone
Somalia
Sao Tome & Principe
Swaziland
Seychelles
Chad
Togo
Tunisia
Tanzania
Uganda
South Africa
Democratic Republic of Congo
Zambia
Zimbabwe
5
7.5
1
0
ln 2
00
0 r
eal
per
cap
ita
GD
P
2 9 16 Log Total Slave Exports Normalized by average Population from 1400−−1900
Dependent variable is log real per capita GDP in 2000, ln y
(1) (2) (3) (4) (5) (6)
ln(exports/area) −.112∗∗∗
−.076∗∗∗
−.108∗∗∗
−.085∗∗
−.103∗∗∗
−.128∗∗∗
(.024) (.029) (.037) (.035) (.034) (.034)
Distance from equator .016 −.005 .019 .023 .006
(.017) (.020) (.018) (.017) (.017)
Longitude .001 −.007 −.004 −.004 −.009
(.005) (.006) (.006) (.005) (.006)
Lowest monthly rainfall −.001 .008 .0001 −.001 −.002
(.007) (.008) (.007) (.006) (.008)
Avg max humidity .009 .008 .009 .015 .013
(.012) (.012) (.012) (.011) (.010)
Avg min temperature −.019 −.039 −.005 −.015 −.037
(.028) (.028) (.027) (.026) (.025)
ln(coastline/area) .085∗∗ .092∗∗ .095∗∗ .082∗∗ .083∗∗
(.039) (.042) (.042) (.040) (.037)
Island indicator −.398 −.150
(.529) (.516)
Percent Islamic −.008∗∗∗
−.006∗
−.003
(.003) (.003) (.003)
French legal origin .755 .643 −.141
(.503) (.470) (.734)
North Africa indicator .382 −.304
(.484) (.517)
ln(gold prod/pop) .011 .014
(.017) (.015)
ln(oil prod/pop) .078∗∗∗ .088∗∗∗
(.027) (.025)
ln(diamond prod/pop) −.039 −.048
(.043) (.041)
Colonizer fixed effects Yes Yes Yes Yes Yes Yes
Number obs. 52 52 42 52 52 42
R2 .51 .60 .63 .71 .77 .80
Sao Tome & PrincipeCape Verde Islands
Djibouti
Rwanda
Central African Republic Morocco
Liberia
Zimbabwe
Tunisia
Niger
Gabon
Uganda
Botswana
Congo
Mauritius
Lesotho
Egypt
Swaziland
Cameroon
Zambia
Comoros
Somalia
Burundi
Sudan
Seychelles
Equatorial GuineaNamibiaLibya
Burkina Faso
Mali
Mauritania
South Africa
Ivory CoastNigeria
Sierra Leone
KenyaAlgeria
Tanzania
Chad
Guinea
Benin
Angola
Gambia
Ghana
Senegal
Ethiopia
Guinea−Bissau
Malawi
Togo
Democratic Republic of Congo
Mozambique
Madagascar
−1
.5
0
1.1
A
ver
age
inco
me
per
per
son
in
20
00
−7 0 Slave exports normalized by land area
9
Robustness of Results
• Changes in the construction of the slave export estimates.
• Omitting influential observations.
• Changes in the sample of countries considered.
• Slave exports disaggregated by time period and by slave trade.
Robustness and Sensitivity Checks.
Specification: Coef Std Err N R2
Normalizing slave exports by average −.103∗∗∗ (.035) 52 .77
population from 1400 to 1900
Omitting zero slave export countries −.104∗∗ (.041) 41 .84
Omitting N. Africa, islands, −.140∗∗∗ (.040) 38 .70
GNQ, LSO, SWZ, and ZAF
Including five region fixed effects −.099∗∗ (.036) 52 .80
Omitting influential observations −.091∗∗∗ (.031) 42 .90
Econometric Issues
1. Unobserved/omitted country characteristics
• Areas that initially had poor institutions or domestic
slavery may have selected into the slave trade and these
characteristics persist today.
• The effect of slave exports may be overestimated: βols will
be biased away from zero.
2. Measurement error in slave export estimates
• Overall noise in the data, i.e. attenuation bias: βols will be
biased towards zero.
• Under-representation of slaves from the interior: βols will be
biased towards zero.
Strategies
• Control for observable characteristics.
• Examine the historical evidence on selection during the slave
trades.
• Instrumental variables.
Selection During the Slave Trades
• Initial prevalence of domestic slavery:
– Did internal slavery predate the external slave trades?
• Initial prosperity:
– Were the initially least prosperous societies targeted?
Angola
Burundi
Benin
Burkina Faso
Botswana
Central African Republic
Ivory Coast Cameroon
Congo
ComorosCape Verde Islands
Djibouti
Algeria
Egypt
Ethiopia
Gabon
Ghana
Guinea Gambia
Guinea−Bissau
Equatorial Guinea
Kenya
Liberia
Libya
Lesotho Morocco
Madagascar
MaliMozambique
Mauritania
Mauritius
Malawi
Namibia
Niger
Nigeria
Rwanda
Sudan
Senegal
Sierra Leone
Somalia
Sao Tome & Principe SwazilandSeychelles
Chad
Togo
Tunisia
Tanzania
Uganda
South Africa
Democratic Republic of Congo
Zambia
Zimbabwe
0
5
10
S
lav
e ex
po
rts
no
rmal
ized
by
lan
d a
rea
−2.5 0 Population density in 1400
3.5
Angola
Burundi
Benin
Burkina Faso
Botswana
Central African Republic
Ivory Coast
Cameroon
Congo
ComorosCape Verde Islands
Djibouti
Algeria
Egypt
Ethiopia
Gabon
Ghana
Guinea
Gambia
Guinea−Bissau
Equatorial Guinea
Kenya
Liberia
Libya
Lesotho Morocco
Madagascar
Mali
Mozambique Mauritania
Mauritius
Malawi
Namibia
Niger
Nigeria
Rwanda
Sudan
Senegal
Sierra Leone
Somalia
Sao Tome & Principe SwazilandSeychelles
Chad
Togo
Tunisia
Tanzania
Uganda
South Africa
Democratic Republic of Congo
Zambia
Zimbabwe
4
15
S
lav
e ex
po
rts
no
rmal
ized
by
av
erag
e h
isto
ric
po
pu
lati
on
−2.5 0 Population density in 1400
3.25
Instrumental Variables
• Instruments
– Must be uncorrelated with unobservable country
characteristics ε, but correlated with slave exports S.
– I use the distance from each country’s interior to the closest
major slave market in each of the four slave trades.
• Crucial Assumption
– Location of demand influenced the location of supply.
– Location of supply did not influence the location of demand.
What determined the location of demand?
• Climate and soil conditions suitable for plantation agriculture.
– West Indies and Mauritius
• Existence of natural resources.
– Gold and silver mines in Brazil
– Salt mines in the Northern Sahara, Arabia and Persia
– Pearls divers in the Red Sea
• Religion.
– In muslim societies slaves were used as servants eunuchs,
concubines, soldiers and government officials: Middle East
and North Africa
Instruments
1. Sailing distance to the closest market of the Atlantic slave
trade.
2. Sailing distance to the closest market of the Indian Ocean slave
trade.
3. Overland distance to the closest market of the Saharan slave
trade.
4. Overland distance the closest port of the Red Sea slave trade.
Virginia, USA
Havana, CubaKingston, Jamaica
Haiti
Martinique
Dominica
Salvador, Brazil
Djibouti
Massawa
Mascat, Oman
Suakin
Rio de Janeiro, Brazil
Guyana
Algires
Tripoli
Benghazi
Tunis
Cairo
Burkina Faso
Figure by MIT OpenCourseWare.
Instrumental Variables Estimates.
(1) (2) (3) (4)
Second Stage. Dependent variable is log income in 2000, ln y
ln(exports/area) −.208∗∗∗
−.201∗∗∗
−.286∗
−.248∗∗∗
(.053) (.047) (.153) (.071)
[−.51, −.14] [−.42, −.13] [−∞, +∞] [−.62, −.12]
Colonizer fixed effects No Yes Yes Yes
Geography controls No No Yes Yes
Restricted sample No No No Yes
F -stat 15.4 4.32 1.73 2.17
Number obs. 52 52 52 42
First Stage. Dependent variable is slave exports, ln(exports/area)
Atlantic distance −1.31∗∗∗
−1.74∗∗∗
−1.32∗
−1.69∗∗
(.357) (.425) (.761) (.680)
Indian distance −1.10∗∗∗ −1.43∗∗∗
−1.08 −1.57∗
(.380) (.531) (.697) (.801)
Saharan distance −2.43∗∗∗
−3.00∗∗∗
−1.14 −4.08∗∗
(.823) (1.05) (1.59) (1.55)
Red Sea distance −.002 −.152 −1.22 2.13
(.710) (.813) (1.82) (2.40)
F -stat 4.55 2.38 1.82 4.01
Colonizer fixed effects No Yes Yes Yes
Geography controls No No Yes Yes
Restricted sample No No No Yes
Hausman test (p-value) .02 .01 .02 .04
Sargan test (p-value) .18 .30 .65 .51
Channels
• What are slave exports correlated with?
Angola
Burundi
Benin
Burkina Faso
Botswana
Central African Republic Ivory Coast
CameroonCongo
Comoros
Cape Verde Islands
Djibouti
Algeria
Egypt
Ethiopia
Gabon
Ghana
Guinea
Gambia Guinea−Bissau
Equatorial Guinea
Kenya
Liberia
Libya
Lesotho
Morocco
Madagascar
MaliMozambique
Mauritania
Mauritius
Malawi
Namibia Niger
Nigeria
Rwanda
Sudan Senegal
Sierra LeoneSomalia
Swaziland
Seychelles
Chad
Togo
Tunisia
Tanzania
Uganda
South Africa
Democratic Republic of Congo
Zambia
Zimbabwe
−.1
.5
1
.1
Eth
nic
Div
ersi
ty
−4 0 5 11 Slave exports normalized by land area
Angola
Burundi
Benin
Burkina Faso
Botswana
Central African Republic
Ivory Coast
Cameroon
Congo
Comoros
Djibouti
AlgeriaEgypt
Ethiopia
Gabon
Ghana
Guinea Gambia
Guinea−BissauEquatorial Guinea
Kenya
Liberia
Libya
Lesotho
Morocco
Madagascar
Mali
MozambiqueMauritania Malawi
Namibia
Niger
Nigeria
Rwanda
Sudan
Senegal
Sierra Leone Somalia
Swaziland
Chad
Togo
Tunisia
Tanzania Uganda Democratic Republic of Congo
Zambia
Zimbabwe
−.1
.5
1
.1
19
th c
entu
ry s
tate
dev
elo
pm
ent
−3.5 0 5 11 ln(exports/area)
(beta coef = −.37, t−stat = −2.63, N = 47, R2 = .13)
Table 4: Testing potential channels of causality.
Without colonizer fixed effects With colonizer fixed effects
OLS IV OLS IV
coef s.e. coef s.e. coef s.e. coef s.e.
Dependent variable
Pre-colonial state dev. −.026∗∗ (.013) −.069∗∗∗ (.025) −.026∗ (.015) −.064∗∗ (.025)
Rule of law −.073∗∗∗ (.020) −.118∗∗∗ (.039) −.078∗∗∗ (.018) −.095∗∗∗ (.032)
Ethnic fractionalization .040∗∗∗ (.009) .065∗∗∗ (.018) .041∗∗∗ (.010) .066∗∗∗ (.019)
Cultural diversity .029∗∗∗ (.007) .050∗∗∗ (.014) .031∗∗∗ (.008) .051∗∗∗ (.014)
Conclusions
• Constructed country-level slave export estimates.
• Estimated a robust relationship between slave exports and
subsequent economic development.
• Used IV to establish causality and to correct for measurement
error.
• Initial evidence suggests that the relationship between the slave
trade and current development are through the same channels
that have been highlighted by historians.