Lecture 13

# Overland distance from a countrys interior to the

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: s. – 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. Overland distance from a country’s interior to the closest point on the coast. 2. Sailing distance to the closest market of the Atlantic slave trade. 3. Sailing distance to the closest market of the Indian Ocean slave trade. 4. Overland distance to the closest market of the Saharan slave trade. 5. Overland distance the closest port of the Red Sea slave trade. Distance Instruments IV Estimation Yi = α1 + β1 Si + C′ δ1 + ε1i i Si = α2 + Z′ βz + C′ δ2 + ε2i i i Y – Log real per capita GDP in 1998 S – Log of slaves taken per km2 of land D – Indicator variables for identity of colonizer Z – Distance instruments Table 8: IV Regressions. Instrumenting slave exports. OLS IV IV OLS IV IV Second Stage. Dependent variable is log income in 1998 ln(exports/area) F-stat Number obs. −.12 −.18 −.12 −.15 −.21 (−5.37) Colonial dummies −.12 (−3.79) −(4.56) (−5.68) (−4.97) (−5.06) No No No Yes Yes Yes 28.9 13.8 20.0 7.1 5.4 4.8 50 50 50 50 50 50 First Stage. Dependent variable is ln(exports/area) Interior distance −.004 (−2.79) Atlantic distance −.004 (−2.80) −.002 −.001 −.002 −.001 (−4.86) (−3.78) (−4.68) (−3.50) −.001 −.001 −.001 −.001 (−3.58) (−2.78) (−3.25) (−2.42) −.003 −.002 −.003 −.002 (−3.55) (−2.92) (−3.09) (−2.41) −.002 −.001 −.002 −.001 (−2.14) (−1.18) (−2.34) (−1.33) 7.37 6.33 3.34 2.46 Over-id test (p-value) .01 .16 .04 .55 Hausman test (p-value) .90 .08 .98 .49 Indian distance Saharan distance Red Sea distance F-stat Channels: A Small First Step • What are slave exports correlated with? 1.1 Current Ethnic Fractionalization and Past Slave Exports (beta coef = .60, t−stat = 5.21) UGA LBR MDG ZAR COG CMR TCD NGA CIV SOM SLE GNB DJI BEN AGO GMB GAB GIN BFA ETH TZA SDN TGO SEN M MLI OZ MWI GHA NER KEN CAF LBY ZMB Ethnic Fractionalization .5 ZAF NAM MUS CPV BWA RWA BDI LSO SYC MRT MAR ZWE DZA GNQ EGY SWZ TUN 0 COM −4 0 5 Log Total Slave Exports per 1,000 km sq of Land Area 11 1.1 Civil War and Past Slave Exports Fraction of years with civil war &gt; 25 deaths since indep .5 (beta coef = .14, t−stat = .97) AGO TCD SDN ETH UGA MOZ LBY ZWE SOM ZAF BDI RWA NAM ZAR MAR DZA EGY DJI MRT CAF KEN ZMB GNQ SEN COG NGA GIN NER COMGNB GHA CIV TZA TGO MLI GMB CMR BFA GABMDG BEN MWI −.1 BWA LSO TUN CPV SWZ SYC STP MUS SLE LBR −4 0 5 Log Total Slave Exports per 1,000 km sq of Land Area 11 1.1 19th Century State Centralization and Past Slave Exports (beta coef = −.46, t−stat = 3.43) MUS SWZ SYC LSO BDI RWA COM ZWE BWA MWI MOZ ETH 19th Century State Cent...
View Full Document

Ask a homework question - tutors are online