Deep Roots of Comparative Development€¦ · Domesticable species, Orientation of continents (Diamond, 1997) Migratory distance from the cradle of humankind in East Africa to

Deep Roots of Comparative Development

Oded Galor

AEA Continuing Education Program

Lecture III - AEA 2014

Oded Galor Roots of Comparative Development Lecture III - AEA 2014 1 / 41


The Origins of Inequality in Income per Capita across the Globe in 2010

0-1000

1000-3

000

3000-5

000

5000-1

0000

10000-

15000

15000-

20000

20000-

30000

30000-

40000

40000+



Deeply-Rooted Factors of Comparative Development

Persistent effects of:

Geographical characteristics:

Climate, Soil quality, Latitude, Isolation

Biogeographical conditions that triggered the Neolithic Revolution (NR)

Domesticable species, Orientation of continents (Diamond, 1997)

Migratory distance from the cradle of humankind in East Africa toindigenous settlements across the globe:

Variation in genetic diversity across countries (Ashraf-Galor, AER 2013)


Deep Roots of Comparative Development The Neolithic Revolution

The Neolithic Origins of Comparative Development —Diamond’s Hypothesis

The transition from hunter-gatherer tribes to agricultural communities:

Emergence of non-food-producing class:

=⇒ Knowledge creation (science, technology & written languages)

Technological head start and its persistent effect via:

Urbanization, nation states, colonization

Variations in biogeographical characteristics conducive for the NR:

=⇒ Origins of the observed patterns of comparative development



Variation in the Onset of the Neolithic Revolution

0-2000

2000-4

000

4000-6

000

6000-8

000

8000-1

0000

10000+



Independent Origins

Source: Diamond (Nature 2002)



Biogeographical Origins of the Onset of the Neolithic Revolution

Geographical factors that maximized biodiversity (climate, latitude, landmass)

Availability of domesticable species of plants and animals

=⇒ Onset of domestication

Orientation of continents:

=⇒ Diffusion of agricultural practices along similar latitudes



Orientation of Continents

Source: Diamond (Nature 2002)


The Diamond Hypothesis Testable Predictions

The Diamond Hypothesis

The domination of Euro-Asia in the pre-colonial era reflects:

Larger number of domesticable species of plants and animals

East-West orientation

=⇒ Technological head start

The economic domination of Europeans and their offshoots in the contempo-rary world reflects:

Persistence of head start

=⇒ Guns, Germs and Steel


The Diamond Hypothesis Testable Predictions

Testable Predictions

Earlier onset of the Neolithic Revolution yields:

Over the entire course of development

Technological superiority

During the Malthusian epoch

Higher population densityNegligible effect on income per capita

Over the contemporary era

Higher income per capita (accounting for the time elapsed sincethe ancestral populations of a country experienced the NR)


The Diamond Hypothesis Persistence of Technological Head Start

The Neolithic Revolution (NR) & Technological Level: 1000 BCE—2000 CE

(1) (2) (3) (4)

Dependent Variable: Log Technological Level

1000 BCE 1500 CE 2000 2000

Log Years since NR 4.34*** 1.05*** -0.12

(1.24) (0.07) (0.92)

Log Years since NR 0.128

(Ancestry Adjusted) (0.08)

Continent Dummies Yes Yes Yes Yes

Observations 112 143 132 132

R2 0.24 0.51 0.48 0.46

Robust standard errors in parentheses; *** p<0.01, ** p<0.05, * p<0.1.

Source: Ashraf-Galor (AER 2011); Ashraf-Galor (2014)


The Diamond Hypothesis Association with Population Density

The Neolithic Revolution on Population Density in 1500

OLS OLS OLS OLS IV

(1) (2) (3) (4) (5)

Dependent Variable: Log Population Density in 1500

Log years since Neolithic 0.833*** 1.025*** 1.087*** 1.389*** 2.077***(0.298) (0.233) (0.184) (0.224) (0.391)

Log land productivity 0.641*** 0.576*** 0.573*** 0.571***(0.059) (0.052) (0.095) (0.082)

Log absolute latitude -0.353*** -0.314*** -0.278** -0.248**(0.104) (0.103) (0.131) (0.117)

Mean distance to nearest -0.392*** 0.220 0.250coast or river (0.142) (0.346) (0.333)

% land within 100 km 0.899*** 1.185*** 1.350***of coast or river (0.282) (0.377) (0.380)

Continental dummies Yes Yes Yes Yes Yes

Observations 147 147 147 96 96

R2 0.40 0.66 0.73 0.73 0.70

First-stage F-statistic 14.65

Overidentification p-value 0.44

Instrumental variables: number of prehistorical domesticable species of plants and animals.



The Diamond Hypothesis Association with Income per Capita

The Neolithic Revolution and Income per Capita: 1—1500 CE

OLS OLS OLS OLS OLS OLS

(1) (2) (3) (4) (5) (6)

Log Income per Capita Log Population Density

1500 1000 1 1500 1000 1

Log years since Neolithic 0.159 0.073 0.109 1.337*** 0.832*** 1.006***(0.136) (0.045) (0.072) (0.594) (0.363) (0.483)

Log land productivity 0.041 -0.021 -0.001 0.584*** 0.364*** 0.681***(0.025) (0.025) (0.027) (0.159) (0.110) (0.255)

Log absolute latitude -0.041 0.060 -0.175 0.050 -2.140** -2.163**(0.073) (0.147) (0.175) (0.463) (0.801) (0.979)

Mean distance to nearest 0.215 -0.111 0.043 -0.429 -0.237 0.118coast or river (0.198) (0.138) (0.159) (1.237) (0.751) (0.883)

% land within 100 km of 0.124 -0.150 0.042 1.855** 1.326** 0.228coast or river (0.145) (0.121) (0.127) (0.820) (0.615) (0.919)

Continental dummies Yes Yes Yes Yes Yes Yes

Observations 31 26 29 31 26 29

R2 0.66 0.68 0.33 0.88 0.95 0.89

Robust standard errors in parentheses; *** p<0.01; ** p<0.05; * p<0.1.



The Neolithic Revolution and Population Density in 1500

Conditional on land productivity, geographical factors, and continental fixed effects.

Source: Ashraf-Galor (AER 2011)



The Neolithic Revolution and Population Density in 1500 (Income Sample)





The Neolithic Revolution and Income per Capita in 1500




The Diamond Hypothesis Association with Current Income per Capita


OLS OLS OLS OLS

(1) (2) (3) (4)

Log Income per Capita in 2005

Log years since Neolithic 0.525* 0.010 -0.336(0.276) (0.216) (0.272)

Log years since Neolithic 0.211(ancestry adjusted) (0.322)

Log land productivity -0.379*** -0.515*** -0.494***(0.080) (0.077) (0.078)

Log absolute latitude 0.765*** 0.353** 0.375***(0.149) (0.138) (0.142)

Mean distance to nearest -0.462 -0.465* -0.425coast or river (0.302) (0.261) (0.260)

% land within 100 km 1.624*** 0.686* 0.768**of coast or river (0.364) (0.367) (0.361)

Continental dummies No No Yes Yes


R2 0.04 0.47 0.62 0.62



The Diamond Hypothesis Association with Current Income per Capita


AGO

MOZ

ZWE

BWA

LSOFIN

SWZ

HTI

IDN

MWILTUEST

JPN

KHM

KOR

LVA

NAM

MNG

MDGBLR

NOR

PHL

ZMB

BTN

LKA

PNG

RUSGTM

IRLSWE

NPLBGD

TZA

LAOBOL

THAMYSKAZ

VNM

GEO

CHE

COL

KGZ

PER

UZB

TCD

GBRRWABELECUMEX

BLZ

DNK

HND

ZAF

NGA

PAN

NIC

ARE

OMN

SAU

TJK

NLDCOG

POLDOMBFA

MLI

SLV

YEM

VEN

GAB

UKR

CMR

CRI

NZL

PRY

BRAAUT

ZAR

PRT

GUY

CZE

GIN

BENCHLTKMSEN

MRT

QAT

USA

GMB

KWT

CUB

SLE

TGO

LBR

AZEARM

CAN

GNB

SVN

BIH

AUS

ESP

MDA

HRV

MKDNER

FRA

CAFGHACIV

ROM

AFG

BGR

ALB

HUNIND

KEN

BDI

UGA

PAK

CHN

ISR

DZAMARIRN

ITA

ARG

DEU

URY

IRQ

ETHJOR

TUR

GRC

SYR

TUNLBN

SDN

LBY

EGY

42

02

4In

com

e pe

r Cap

ita in

200

5

1 .5 0 .5 1Years Elapsed since the Neolithic Revolution (Ancestry Adjusted)

coef = .21117707, (robust) se = .32236237, t = .66


Source: Ashraf-Galor (2014)


The Neolithic Revolution Life Expectancy

The Neolithic Origins of Contemporary Variation in Life Expectancy

The Neolithic Revolution increased the exposure and the vulnerability ofhumans to infectious diseases via:

the rise in population densitythe domestication of animalsthe increase in work effort

Natural selection of individuals who were genetically pre-disposed towardsresistance to infectious diseases (Galor-Moav, 2005)

Variation in the timing of the Neolithic Revolution among the ancestral pop-ulations of each county contributed to the variation in life expectancy acrosscountires (Galor-Moav, 2007)



The Neolithic Origins of Variation in Life Expectancy in 2004

OLS OLS OLS IV

(1) (2) (3) (4)

Life Expectancy in 2004

Log years since Neolithic 1.559*** 1.463*** 1.635*** 1.909**(ancestry adjusted) (0.461) (0.355) (0.444) (0.889)

Log land productivity -1.386*** -1.636*** -1.371***(0.350) (0.444) (0..519)

Log absolute latitude 0.195*** 0.124* 0.055(0.053) (0.068) (0.048)

Mean distance to nearest —5.030*** —7.130*** -5.121***coast or river (1.084) (1.937) (1.668)

Island 9.365*** 8.083 9.154***(1.990) (2.723) (2.670)

Continental dummies Yes Yes Yes Yes


R2 0.71 0.80 0.86


Instrumental variable: number of prehistorical domesticable animals.




The Neolithic Origins of Variation in Life Expectancy in 2004

FIN

IDN

PRK

KOR

LTUESTKHM

MNG

HTI

LVABWAMOZ

MMRAGO

LSO

BLRBTN

SWZZWE

NOR

JPN

PNG

BGDNAMIRL

MWI

SWE

NPL

GEO

RUS

MYSTHAVNM

KAZ

BOL

LAOPHL

GBR

GTM

ZMB

DNKBELUZBPER

KGZCOL

ECU

LKA

ZAF

TZA

CHE

BLZ

TCD

NLD

POL

MEX

NGAMDG

VEN

HNDARETJKCOG

PRYMLI

RWAGAB

PANCHL

MRT

GMBSENBFADOMCMR

GNB

GIN

NIC

UKR

OMNLBR

SAU

SLE

SLVBRABEN

SOM

ZAR

YEM

NZL

GUY

TGO

CZE

AUT

CRI

PRT

MAR

QATNERCIV

GHA

DZA

USA

TKMKWT

CAF

KENSVNCAN

AZE

MDA

ARMBDIUGA

BIHHRV

ETH

TUN

ESPFRA

URYHUN

AUS

ARGROM

ALB

BGRMKD

ISR

INDCUB

AFG

DEU

PAK

ITASDN

CHN

LBY

IRNTUR

GRC

IRQ JORLBN

SYR

EGY

20

10

010

20LI

fe E

xpec

tanc

y in

200

4

4 2 0 2 4Time Elapsed Since the Neolithic Revolution

coef = 1.4625621, (robust) se = .35539057, t = 4.12


Source: Galor-Moav (2007)



The Neolithic Origins of Mortality from Infectious Diseases in 2004

OLS OLS OLS IV

(1) (2) (3) (4)

Years of Life Lost to Infectious Disease 2004

Log years since Neolithic -3.818*** -3.593*** -3.504*** -8.664**(ancestry adjusted) (1.160) (0.918) (1.286) (2.918)

Log land productivity 3.000*** 2.016*** 2.150(0.899) (1.468) (1.572)

Log absolute latitude -0.571*** -0.407*** -0.380(0.124) (0.164) (0.192)

Mean distance to nearest 8.026*** 10.07*** 9.133*coast or river (3.032) (4.583) (5.514)

Island -25.39*** -18.24* -28.56**(7.977) (10.90) (13.29)

Continental dummies Yes Yes Yes Yes


R2 0.77 0.84 0.87


Instrumental variable: number of prehistorical domesticable animals.




The Neolithic Origins of Mortality from Infectious Diseases in 2004

FIN

IDN

PRK

KOR

LTUEST

KHM

MNG

HTI

LVA

BWA

MOZMMR

AGO

LSO

BLR

BTN

SWZZWENOR

JPN

PNGBGDNAM

IRLMWISWE

NPL

GEORUS

MYS

THAVNM

KAZ

BOL

LAOPHL

GBR

GTM

ZMBDNKBEL

UZBPER

KGZ

COL

ECU

LKA

ZAF

TZACHEBLZTCDNLD

POL

MEX

NGA

MDG

VEN

HND

ARE

TJK

COG

PRYMLI

RWAGAB

PANCHL

MRT

GMBSENBFADOM

CMR

GNB

GINNICUKR

OMN

LBR

SAU

SLE

SLVBRABEN

SOM

ZAR

YEM

NZL

GUY

TGOCZEAUT

CRI

PRT

MAR

QAT

NER

CIVGHA

DZA

USA

TKM

KWTCAFKENSVN

CAN

AZEMDA

ARM

BDIUGA

BIHHRV

ETH

TUN

ESPFRA

URYHUN

AUS

ARGROMALB

BGRMKD

ISR

IND

CUB

AFG

DEU

PAK

ITA

SDNCHN

LBY

IRNTURGRC

IRQ

JOR

LBN

SYR

EGY

40

20

020

40M

orta

lity

from

Infe

ctio

us D

isea

ses

in 2

004

4 2 0 2 4Years Elapsed Since the Neolithic Revolution

coef = 3.5926452, (robust) se = .91783991, t = 3.91


Source: Galor-Moav (2007)


Geographical Factors Persistence and Reversals

Persistence and Reversals in the Role of Geographical Factors

Dynamics of the geographical patterns of comparative development

Reversal in the role of

Land ProductivityDistance from the equator

Persistence in the role of

Disease environmentEcological diversityGeographical IsolationRange of land qualityLand suitable for large plantations


Geographical Factors Reversals

Reversal in the Role of Land Productivity & Absolute Latitude

Land productivity

1-1500 CE

Positive association with population density

2000s

Negative association with income per capita

Absolute latitude

1-1500 CE

Negative association with population density

2000s

Positive association with income per capita



Reversal in the Role of Land Productivity

World Non-Colony Ex-Colony World Non-Colony Ex-Colonysample sample sample sample sample sample

(1) (2) (3) (4) (5) (6)

Log Population Density Log Income per Capita

1500 2005

Log years since Neolithic 1.111*** 0.769* 1.383***(0.188) (0.447) (0.267)

Log years since Neolithic 0.211 -0.100 0.083(ancestry adjusted) (0.322) (0.559) (0.382)

Log land productivity 0.568*** 0.550*** 0.585*** -0.494*** -0.518*** -0.456***(0.053) (0.057) (0.115) (0.078) 0.087 0.141

Log absolute latitude -0.330*** -0.491*** -0.302** 0.375*** 0.891** 0.139(0.106) (0.136) (0.123) (0.142) (0.432) (0.149)


Observations 143 68 75 143 68 75

R2 0.73 0.72 0.70 0.62 0.64 0.57

Additional Controls: mean distance to nearest coast & river and % land within 100 km of coast & river.




Origins of the Reversal in the Role of Land Productivity

The effect is nearly identical in the:

World sample

Former colonies sample (Acemoglu-Johnson-Robinson, QJE 2002)

Non-former colonies sample

=⇒ Reversal in the role of land productivity is largely independent ofthe forces of colonialism

Acquired comparative advantage in agriculture & delayed industrial transition


Geographical Factors Reversal: Land Productivity

Land Productivity and Population Density in 1500

OMNARE

KWTQAT

MRTSAUNOR

LBY

EGY

BWA

NAM

FIN

YEM

SUR

COGMNGSWE

GAB

JOR

CHL

CAN

DZA

RUS

AGO

ZAR

LBRCOL

VENNER

MYS

BLZIRL

PER

SVNTCDTKMMLI

CRI

GUY

MDG

PANPNG

CHE

EST

CAF

AUT

MOZ

BRA

NLDBOL

BLR

LVA

GBR

IDN

ECUPRT

LAO

BEL

BIH

KEN

NZL

IRN

AUS

ZWE

ALB

SLE

LTU

IRQKGZTZA

GRC

SEN

DEU

ZAF

GIN

GNBHRV

HND

TJK

ZMB

CUB

PRYUZB

DNKCIV

SDN

LKA

CMR

CZEITA

ESP

ISRPOL

HTI

SWZ

FRANIC

PAK

DOM

ARG

TUN

JPN

BGR

MKD

MEX

MDAURY

AFG

VNM

USAPHL

GMB

GEO

UKR

MAR

HUNROM

KOR

ETH

GTM

LSO

KHM

BFA

SLV

SYRGHA

LBN

KAZ

CHN

THA

UGA

NPL

MWI

BENNGA

ARM

RWA

BGD

BDI

AZETGOTUR

IND

21

01

23

Log

Pop

ulat

ion

Den

sity

in 1

500

4 2 0 2Log Land Productivity

coef = .56769725, (robust) se = .05303816, t = 10.7

Conditional on years since Neolithic transition, geographical factors, and continental fixed effects.



Geographical Factors Reversal: Land Productivity

Land Productivity and Income per Capita in 2005

OMN

KWT

AREQAT

MRT

NORSAULBY

EGY

BWA

NAMFIN

YEM

COG

MNG

SWE

GAB

JOR

CHL

PNG

DZA

RUS

AGOCOL

ZAR

VEN

CAN

PERBLZ

NER

LBR

IRL

GUY

TCDSVN

MLI

MYS

TKMCRI

MDGBOL

PAN

BTN

BRA

CHE

EST

MOZ

AUTECU

CAF

NLD

BLR

GBR

LVAIDNPRT

LAO

BEL

BIH

ZWE

KEN

SLE

LTU

IRN

ALB

TZA

KGZIRQPRYHND

NZL

SEN

GIN

GRCDEU

GNB

HRV

TJK

UZB

MEX

ZMB

NIC

DNKISR

CIVCMRSWZ

SDNLKACZE

ZAF

POL

ESPITAFRA

ARG

HTI

PAK

JPN

TUN

BGR

AUS

MKD

GTM

USA

URY

MDA

AFG

DOM

VNM

PHL

GMB

UKR

GEOHUNLSO

ROM

KOR

MAR

CUB

ETH

SLV

KHM

BFAGHA

KAZ

SYR

LBNUGA

CHN

NPL

THA

MWI

BENNGA

ARM

RWA

BGD

BDI

AZETGO

TUR

IND

42

02

4Lo

g In

com

e pe

r Cap

ita in

200

5

4 2 0 2Log Land Productivity

coef = .49436427, (robust) se = .07843498, t = 6.3




Geographical Factors Reversal: Distance from the Equator

Reversal in the Role of Distance from the Equator

World Non-Colony Ex-Colony World Non-Colony Ex-Colonysample sample sample sample sample sample

(1) (2) (3) (4) (5) (6)

Log Population Density Log Income per Capita

1500 2005

Log years since Neolithic 1.111*** 0.769* 1.383***(0.188) (0.447) (0.267)

Log years since Neolithic 0.211 -0.100 0.083(ancestry adjusted) (0.322) (0.559) (0.382)

Log land productivity 0.568*** 0.550*** 0.585*** -0.494*** -0.518*** -0.456***(0.053) (0.057) (0.115) (0.078) 0.087 0.141

Log absolute latitude -0.330*** -0.491*** -0.302** 0.375*** 0.891** 0.139(0.106) (0.136) (0.123) (0.142) (0.432) (0.149)


Observations 143 68 75 143 68 75

R2 0.73 0.72 0.70 0.62 0.64 0.57

Additional Controls: mean distance to nearest coast & river and % land within 100 km of coast & river.




Distance from the Equator and Population Density in 1500

ZAR

MYS

COGUGA

GABKEN

ECU

RWA

IDNSUR

LKA

COL

BDI

GUY

PNGKHM

PHL

TZA

VEN

YEM

THA

PANCAF

CMR

VNM

CRI

BRA

LBRRUS

LAO

PER

OMNPRTESP

GRCALB

CIVAREMKD

ETH

CHE

GHATGO

IND

BIH

SAU

BGR

ITA

SLESVN

NIC

AUT

HRVQAT

AGO

BLR

LVA

FRA

ROM

IRL

LTU

BEL

FIN

MDA

NOR

BGD

HUNEST

UKR

CZE

NGA

SLV

BEN

NLD

AUS

GBR

POL

HTI

HNDDEU

DNKSWE

CUBNPL

KWT

DOMGTMGIN

KGZ

BOL

ZMB

PAK

TCDMNG

MWI

BLZ

UZBTJK

BFA

AFGJOR

GNBTKMKAZIRN

JPN

IRQ

CHN

BWA

KOR

NER

ISRMLI

GMB

SENMOZ

SDN

SYR

LBN

NAM

ZWE

AZEGEOARMPRY

MEX

MRT

TURMDG

NZLCHL

SWZ

URY

ARG

LBYCANZAFLSO

USA

DZA

EGY

MARTUN

21

01

2Lo

g P

opul

atio

n D

ensi

ty in

150

0

2 1 0 1 2Log Absolute Latitude

coef = .33068066, (robust) se = .10617177, t = 3.11





Distance from the Equator and Income per Capita in 2005

ZAR

COGUGA

GAB

KEN

MYS

ECURWA

IDNCOL

LKA

PNG

GUY

BDI

VEN

YEM

KHMPHLCAFPANTZA

CMRBRA

PER

THA

CRI

LBR

VNMOMN

LAO

RUS

GRCPRT

ETH

ESP

ALB

CIV

MKD

GHAIND

ITAARE

BGR

BIH

TGO

NICSLE

SVN

SAU

HRV

FRACHEQAT

ROM

AUT

HUN

MDA

SLVBENCZE

UKR

DEU

HND

BELNGANLDIRL

POL

KWT

BOL

NOR

BLR

GBR

GTM

BGDGINPAKBLZ

DNK

BTNLVALTU

SWE

JOR

AFG

EST

NPL

FINIRN

KGZ

TCD

TJK

IRQBFA

AGO

UZB

TKM

GNB

SDN

NER

ISR

CHNDOM

AUSKAZ

MNG

HTI

ZMBSYR

LBN

MWIMLIGMB

SENCUB

PRY

MEXAZE

JPN

ARM

KOR

TUR

MRT

GEOCHL

BWA

MOZ

ZWE

NAMLBY

MDG

NZLARGURY

EGY

USA

DZA

CAN

SWZ

ZAF

LSOMARTUN

42

02

4Lo

g In

com

e pe

r Cap

ita in

200

5

3 2 1 0 1Log Absolute Latitude

coef = .37452849, (robust) se = .14185733, t = 2.64





Origins of the Reversal in the Role of Distance from the Equator

The effect is qualitatively simialr in the:

World sample

Non-former colonies sample

=⇒ Reversal in the role of distance from the equator is largely inde-pendent of the forces of colonialism

Temperate drift hypothesis: Advanced technologies gradually complementedproduction in temperate zones


Geographical Factors Persistent: Institutions & Human Diversity

Population Density in 1500 and Income per Capita in 2005 —World Sample

CANAUSUSA

ARGURYBRABWANAM

GUY

LBY

FINNZL

MNG

VEN

LSO

SWZ

ZAF

PRY

MLIMRT

NER

TCD

DZA

OMN

RUSCHL

MWI

ZMBZWE

BOL

ARE

BHR

KWTQAT

KAZ

KGZTJK

TKM

UZB

SAU

COL

MDG

MYS

MOZ

NOR

CUBDOM

HTI

AGO

CAF

CMR

COG

GAB

ZAR

PNG

PERBLZCRI

GTMHNDNIC

PAN

SLV

PHL

SDNLAO

ETH

SWE

TZA

ECUJOR

IRQ

IRNBLR

ESTLTULVA

MDA

UKR

MEX

KEN

ARMAZEGEO

AFG

MARTHA

EGYYEMIDN

BENBFA

CIV

GHA

GINGMBGNB

LBR

NGASEN

SLETGO

TUN

VNM

BGRALB

UGA

LBN

SYR

GRC

TUR

KHM

ROMBIH

HRV

MKD

SVN

IRL

CHN

ESP

POLHUN

PRTISR

NPL

DNK

LKA

GBR

KOR

CHE

CZE

AUT

BDI

RWA

DEU

BGD

INDPAK

NLDFRAITABELJPN

42

02

4Lo

g In

com

e pe

r Cap

ita in

200

5

4 2 0 2 4Log Population Density in 1500

coef = .04305199, (robust) se = .11347758, t = .38



Population Density in 1500 & Income per Capita in 2005 —Ex-ColoniesSample

CANAUSUSA

ARGURYBRABWA

NAM

GUY

NZL

VEN

LSO

SWZ

ZAF

PRY

MLI

MRT

NER

TCD

DZA

CHL

MWI

ZMBZWE

BOL

COL

MDG

MYS

MOZ

DOM

HTI

AGO

CAF

CMR

COG

GAB

PERBLZCRI

GTM

HNDNIC

PAN

SLV

PHL

SDNLAO

ETH

TZA

ECU

MEX

KEN

MAR

EGYIDN

BENBFA

CIV

GHA

GINGMB

NGASEN

SLETGO

TUN

VNM

UGA NPL

LKA

BDI

RWA

BGD

INDPAK

20

24

Log

Inco

me

per C

apita

in 2

005

4 2 0 2 4Log Population Density in 1500

coef = .54395783, (robust) se = .07224269, t = 7.53



Reversal of Fortune

This reversal in the relative performance of countries is:

Absent in the world sample

Present in the former colonies sample

=⇒ Reversal of Fortune is largely triggered by colonialism (Engerman-

Sokoloff, 1997; Acemoglu et al., AER 2001, QJE 2002)



Origins of Reversal of Fortune among Ex-Colonies

Persistent effect of institutions implemented by colonial powers (Engerman-Sokoloff,1997; Acemoglu et al., AER 2001, QJE 2002)

Exclusive (growth retarding) institutions imposed in densely populatedareasInclusive (growth enhancing) institutions implemented in sparsely popu-lated areas

Persistent effect of the human capital and diversity brought by the colonists

Larger effect of colonizers in sparsely populated areas (Glaeser et al., JEG 2004;Easterly-Levine, 2012; Ashraf-Galor, 2014)



Population Density in 1500 & Subsequent Changes in Genetic Diversity


Geographical Factors Persistence Effect of Geography

Persistent Effects of Some Geographical Factors

Disease environment

Persistent effect on labor productivity & investment in human capital(Gallup-Sachs, 2001; Andersen-Dalgaard-Selaya, 2012)

Geographical isolation

Reduced trade and technological diffusion (Gallup-Mellinger-Sachs, 1999)

Persistence of culture conducive for innovations (Ashraf-Galor-Ozak, JEEA 2010)



Persistent Effects of Geographical Factors

Land suitable for large plantations

Inequality:

Extractive institutions (Engerman-Sokoloff, 1997)

Concentration of landownership:

Suboptimal investment in public education (Galor-Moav-Vollrath, RES 2009)

Soil quality conducive for agriculture

Specialization in unskilled-intensive goods

Reduces human capital formation & increases fertility & slows thetransition to modern growth (Galor-Mountford, RES 2008)



Persistent Effects of Geographical Factors

Range of soil quality

Emergence of geographical specific human capital =⇒ reduced mobility=⇒ ethnic fractionalization (Michalopoulos, AER 2012)

Persistent effect of ethnic fractionalization (Easterly-Levine, QJE 1997)

Ecological diversity & storable crops

Emergence & persistence of state capacity (Fenske, JEEA 2014; Mayshar-Moav-Neeman,2013)

Geographical determinants of body size

Determined fertility & income per capita in the Malthusian epoch andthe timing of the take-off (Dalgaard-Strulik, 2013)


Documents

Deep Roots of Comparative Development€¦ · Domesticable species, Orientation of continents (Diamond, 1997) Migratory distance from the cradle of humankind in East Africa to