human evolution and economic developmentmedia.virbcdn.com/...zeuthenlectures-lectureiii.pdf ·...

Human Evolution and Economic Development

Oded Galor

Zeuthen Lectures - Lecture III

May 20, 2016

Oded Galor Evolution and Growth May 20, 2016 1 / 5

Evolutionary Growth Theory

Captures the coevolution of human traits and the economic development overthe course of human history

The effect of the economic environment on the evolutionary processesin the composition of human traits

The impact of the evolution in the composition of human traits on thegrowth process

Main Hypothesis

The coevolution of human traits and the development process is critical forthe understanding of:

The transition from stagnation to growthComparative economic development

During the Malthusian epoch the composition of human traits that werecritical for the growth process had not been stationary

The Malthusian pressure affected the size and the composition of thepopulationHereditary physical and cognitive traits that generated higher income

Higher reproductive successBecome more prevalent in the population

The forces of natural selection

Increased the representation of traits that were complementary to thegrowth processReinforced the growth processstimulated the take-off from an epoch of stagnation to sustained growth

Main Hypothesis

The transition from stagnation to growth

Comparative economic development

Main Hypothesis

The Malthusian pressure affected the size and the composition of thepopulation

Hereditary physical and cognitive traits that generated higher income

Main Hypothesis

Higher reproductive success

Become more prevalent in the population

Main Hypothesis

The forces of natural selectionIncreased the representation of traits that were complementary to thegrowth process

Reinforced the growth processstimulated the take-off from an epoch of stagnation to sustained growth

Main Hypothesis

The forces of natural selectionIncreased the representation of traits that were complementary to thegrowth processReinforced the growth process

stimulated the take-off from an epoch of stagnation to sustained growth

Main Hypothesis

The forces of natural selectionIncreased the representation of traits that were complementary to thegrowth processReinforced the growth processstimulated the take-off from an epoch of stagnation to sustained growth

Selection of Predisposition Towards Child Quality

The rise of the reward for human capital has increased the evolutionary optimalinvestment in offspring’s quality

The evolution of the human brain and the complementarity betweenbrain capacity and investment in human capitalIncreased economic complexity in the course of the Neolithic Revolution

The Malthusian pressure increased the representation of human traits thatwere complementary to investment in human capital

Preference for child quality (Galor-Moav, QJE 2002)

Higher life expectancy (Galor-Moav, 2005, 2007; Franck-Galor-Ozak, 2016)

Entrepreneurial spirit (Galor-Michalopoulos, JET 2012)Moderate fecundity (Galor-Klemp, 2015)

The evolution of the human brain and the complementarity betweenbrain capacity and investment in human capital

Increased economic complexity in the course of the Neolithic Revolution

Entrepreneurial spirit (Galor-Michalopoulos, JET 2012)

Moderate fecundity (Galor-Klemp, 2015)

Evolutionary Changes in Humans in the Past 10,000 Years

Lactose Tolerance

Triggered by the domestication of animals in the course of the NeolithicTransition

Genetic immunity to malaria - Sickle Cell Trait

Triggered by the engagement in slash-and-burn agriculture

700 regions of the human genome

Reshaped by natural selection within the past 5,000 to 15,000 years (Voightet al., 2006)

Lactose Tolerance

The Biocultural Origins of Human Capital Formation

Oded Galor and Marc Klemp

19 May, 2016

Galor and Klemp Biocultural Origins of Human Capital 19 May, 2016 1 / 72

Introduction Motivation

Research Questions

The evolutionary foundations of human capital formation

Have the forces of natural selection favored individuals with a largerpredisposition towards child quality?

Have the forces of natural selection contributed to:

Human capital formation?

The demographic transition?

The transition from stagnation to growth?

Research Questions

Existing Theories

The Life-History Theory (Lack, 1954)

Reproductive strategies of any organisms have evolved to maximize theirlong-run reproductive success

Clutch size (e.g., number of eggs) maximizes long-run reproduction

Evolutionary Growth Theory (Galor-Moav, 2002)

Natural Selection is the origin of economic growth

Traits associated with pre-disposition towards child quality were selected,triggering the transition from stagnation to growth

Existing Theories

Fundamental Empirical Questions

Is there an intergenerational trade-off in reproductive success?

Is higher fecundity (e.g., higher sperm count) conducive for reproductivesuccess in the long run?

Does a larger number of children generate a larger number of offspringin the long run?

Have the forces of natural selection favored moderately fertile indivi-duals (i.e., the quality strategy) during the Malthusian epoch?

Conditions for the Optimality of Moderate Fertility

Necessary Condition: Quantity-Quality Trade-off

Quantity of children come on the account of their quality (survival,education, marriageability)

Lower quality reduces the child’s earning capacity and therefore fertility

Sufficient Condition:

The product of the following elasticities is greater than 1:

The elasticity of child quality with respect to to quantity

The elasticity of grandchildren with respect to child quality

Evidence for a Quantity-Quality Trade-off

Plants

Trade-off between the number and size of seeds

Animals

Trade-off between the number of offspring and their size

Humans

Trade-off between the number of children and their human capital

Plants

Animals

Humans

Plants

Animals

Humans

Plants

Animals

Humans

Plants

Animals

Humans

Plants

Animals

Humans

Empirics of Intertemporal Trade-off in Reproductive Success in Humans

Poorly identified

Unobserved heterogeneity

Reverse causality

Small samples

Often post-demographic transition populations

No intertemporal trade-off in reproductive success:

Kenyan Kipsigis (Borgerhoff-Mulder, 2000)

82 men and 64 women in 4 income groups

No trade-off for men and for 3 of the 4 income groups

New Mexican men (Kaplan et al., 1995)

Post-demographic transition

Poorly identified

Reverse causality

Small samples

Poorly identified

Reverse causality

Small samples

Poorly identified

Reverse causality

Small samples

Poorly identified

Reverse causality

Small samples

Poorly identified

Reverse causality

Small samples

Poorly identified

Reverse causality

Small samples

Poorly identified

Reverse causality

Small samples

Poorly identified

Reverse causality

Small samples

Poorly identified

Reverse causality

Small samples

Historical Lab: Quebec During its Demographic Explosion

Quebec During its Demographic Explosion

Quebec 1608–1800

Founder population order of magnitude smaller than carrying capacity

Reproductive histories of 3,798 lineages

Genealogy of nearly 500,000 individuals

Quebec 1608–1800

Findings

First comprehensive evidence for presence of an intertemporal trade-offin reproductive success (across any species)

An intermediate level of fecundity is associated with maximal long-runreproductive success.

Confirmation of selection of a quality strategy over the pre-industrial,pre-demographic transition era

The optimal level of fecundity is below the population mean

Findings

Main Hypothesis: Moderate Fertility Maximizes LR Reproductive Success

LR Reproductive Success

Children

Main Hypothesis: Moderate Fertility Maximizes LR Reproductive Success

LR Reproductive Success

Children

Empirical challenges: Unobserved Heterogeneity

Income and education

Transmitted intergenerationally

Positively affects fertility

Biases the results towards the absence of an intergenerational trade-off

Challenge: Unobserved heterogeneity

Grandchildren

Children

Grandchildren

Children

Ordinary

Grandchildren

Children

Grandchildren

Children

Empirical Challenge: Reverse causality

Effect of offspring quality on quantity

Lower quality of children that results in higher child mortality

Higher fertility via the replacement motive

Introduction Empirical strategy

Empirical strategy

Exploits random variation in fecundity (i.e., the physiological ability toreproduce) to establish the optimality of moderate fecundity for long-run reproductive success.

Employ the time interval between the date of marriage and the first birth(TFB) as a source of variation in fecundity, .

TFB is affected by:

Genetic predisposition

Socio-environmental conditions

Random elements that affect conception

Identification will be based on the random variation in TFB

Empirical strategy

TFB is affected by:

Empirical strategy

TFB is affected by:

Empirical strategy

TFB is affected by:

Empirical strategy

TFB is affected by:

Empirical strategy

TFB is affected by:

Empirical strategy

TFB is affected by:

The Distribution of Time to First Birth (TFB)

0 38 142

Weeks from marriage to first birth

Time to First Birth (TFB) and the Number of Children

40 60 80 100 120Weeks from marriage to first birth

Show unconditional relationship

Isolating the Random Variation in TFB

Control for socioeconomic factors

Literacy

Gender

Birth year, birth parish and death parish

Month of marriage and month of birth of firstborn

Marriage and stoppage age

Birth order

Lineage-specific fixed effects

Control for genetic predisposition

Identification based on variation across siblings as opposed to thepopulation as a whole

Literacy

Gender

Birth order

Literacy

Gender

Birth order

Literacy

Gender

Birth order

Literacy

Gender

Birth order

Literacy

Gender

Birth order

Literacy

Gender

Birth order

Literacy

Gender

Birth order

Literacy

Gender

Birth order

Literacy

Gender

Birth order

Literacy

Gender

Birth order

Reproductive Success of Head of Dynasties

Focus on head of dynasties born prior to 1685 allows for:

Reproductive histories of 3,798 head of dynasties

Tracing reproductive histories over 4 generations

Account for lineage-specific fixed effects

Head of Dynasty

The Lineage of Head of Dynasties

Lineages of Head of Dynasties

Number of Children

1 5 10 15 20 25 30Number of children

Number of Grandchildren

1 50 100 150 200Number of grandchildren

Data & Analysis Data

Summary Statistics

Mean Median S.D. Count(1) (2) (3) (4)

Children 9.70 10 3.99 3,798Surviving childrena 4.63 4 2.59 3,798Grandchildren 47.35 44 28.07 3,798Great-grandchildren 187.59 159 140.17 3,798Great-great-grandchildrenb 294.01 171 363.58 3,798Years from marriage to first birth (TFB) 1.20 1.02 0.47 3,798Literate 0.66 1 0.47 2,222Fraction of literate children 0.74 1 0.35 3,448Fraction of surviving childrena 0.49 0.50 0.21 3,798Fraction of surviving children with known literacya 0.76 0.67 0.56 3,727Age at first marriage 22.67 22.2 5.46 3,798Age at last delivery 41.95 42.1 8.61 3,798c Survival is recorded at age 40.b The moderate increase in the mean and median number of descendants from the third to the fourthgeneration (i.e. from great-grandchildren to great-great-grandchildren) reflects the fact that these cohortsare less fully observed. Furthermore, since men produce children at later ages than women, this effect ismore pronounced among men.

Children vs. TFB: restricted cubic splines

38 142

Grandchildren vs. TFB: restricted cubic splines

38 142

Great-Grandchildren vs. TFB: restricted cubic splines

38 142

Great-Great-Grandchildren vs. TFB: restricted cubic splines

38 142

Econometric Specification

The effect of TFB on the number of children:

lnDi ,1 = β0,1 + β1,1TFBi + Ziβ2,1 + εi ,1,

The effect of TFB on the number of offspring in generation t = 2, 3, 4:

lnDi ,t = β0,t + β1,tTFBi + β2,tTFB2i + Ziβ3,t + εi ,t ,

Econometric Specification

The effect of TFB on the number of children:

lnDi ,1 = β0,1 + β1,1TFBi + Ziβ2,1 + εi ,1,

The effect of TFB on the number of offspring in generation t = 2, 3, 4:

lnDi ,t = β0,t + β1,tTFBi + β2,tTFB2i + Ziβ3,t + εi ,t ,

TFB and Number of Descendants for Head of Lineages Born before 1685(Accounting for Maternal Founder FE)

Log number of descendants in:Gen. 1 Gen. 2 Gen. 3 Gen. 4 Gen. 1 Gen. 2 Gen. 3 Gen. 4 Gen. 1 Gen. 4

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

TFB -.052** .167 .505** .783*** -.062*** .140 .463** .773*** -.077*** .810***(.024) (.163) (.205) (.264) (.024) (.162) (.204) (.264) (.011) (.258)

TFB2 -.068 -.193*** -.310*** -.063 -.183*** -.309*** -.325***(.053) (.067) (.087) (.053) (.067) (.087) (.084)

Literate -.006 .063 .148*** .138** -.027* .109*(.030) (.040) (.051) (.066) (.014) (.066)

Male .220*** .254*** .299*** .131** -.028* .036(.031) (.039) (.047) (.060) (.015) (.063)

Stoppage age fixed effects No No No No No No No No Yes Yes

Number of observations 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798Adjusted R2 .015 .016 .038 .306 .032 .029 .052 .307 .799 .355Joint sign.-level of TFB & TFB2 .031 .196 .002 .000 .010 .130 .002 .000 .000 .000Maximizing TFB 1.224 1.307 1.261 1.113 1.263 1.249 1.247Lower limit of 90% CI - .961 .999 - .827 .976 1.002Upper limit of 90% CI - 1.467 1.398 - 1.435 1.389 1.376This table presents the results of a series of fixed-effects regressions of the number of descendants in generation t on time to first birth measured in years, i.e. TFBand TFB2. All regressions account for Maternal Founder fixed effects. Birth year and marriage age dummies are included as controls. Furthermore, stoppage agedummies are included in columns 9–10. A dummy indicating unknown literacy is included in the regressions underlying column 5–10. Standard errors clustered atthe level of the firstborn are reported in parentheses. * p <0.10, ** p <0.05, *** p <0.01.

TFB and Number of Descendants for Head of Lineages Born 16851660–1685 (accounting for Maternal Founder FE)

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

TFB -.065** .237 .544** .830*** -.075*** .193 .487** .802*** -.079*** .773***(.025) (.175) (.216) (.280) (.025) (.174) (.216) (.282) (.012) (.277)

TFB2 -.096* -.214*** -.337*** -.086 -.199*** -.330*** -.323***(.057) (.071) (.092) (.057) (.071) (.093) (.091)

Literate -.031 .023 .120** .116* -.035** .087(.033) (.042) (.054) (.070) (.015) (.070)

Male .219*** .279*** .315*** .177*** -.035** .115*(.032) (.040) (.049) (.063) (.015) (.066)

Number of observations 3,376 3,376 3,376 3,376 3,376 3,376 3,376 3,376 3,376 3,376Adjusted R2 .019 .020 .042 .335 .036 .036 .058 .337 .804 .379Joint sign.-level of TFB & TFB2 .010 .066 .001 .000 .003 .042 .000 .000 .000 .000Maximizing TFB 1.230 1.270 1.232 1.124 1.219 1.214 1.196Lower limit of 90% CI -11.448 .923 .961 - .761 .919 .886Upper limit of 90% CI 1.505 1.427 1.37 - 1.392 1.357 1.343This table presents the results of a series of fixed-effects OLS regressions of the number of descendants in generation t on time to first birth measured in years, i.e.TFB and TFB2 for heads of lineages born in the period 1660–1685. All regressions account for Maternal Founder fixed effects. Birth year and marriage age dummiesare included as controls. Furthermore, stoppage age dummies are included in columns 9–10. A dummy indicating unknown literacy is included in the regressionsunderlying column 5–10. Standard errors clustered at the level of the firstborn are reported in parentheses. * p <0.10, ** p <0.05, *** p <0.01.

Robustness to Alternative Cohorts: No Restrictions on Cohorts (Accountingfor Maternal Founder FE)

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

TFB -.064*** .055 .319** .409** -.071*** .031 .293** .393** -.090*** .459**(.017) (.120) (.141) (.199) (.017) (.119) (.140) (.199) (.008) (.198)

TFB2 -.039 -.133*** -.192*** -.033 -.126*** -.187*** -.212***(.040) (.046) (.065) (.039) (.046) (.065) (.065)

Literate .007 .078*** .133*** .134*** -.014 .121**(.023) (.030) (.037) (.049) (.011) (.048)

Male .222*** .220*** .191*** .088** -.013 .019(.018) (.023) (.027) (.036) (.009) (.038)

Number of observations 7,664 7,664 7,664 7,664 7,664 7,664 7,664 7,664 7,664 7,664Adjusted R2 .026 .022 .067 .413 .049 .036 .077 .414 .789 .431Joint sign.-level of TFB & TFB2 .000 .015 .000 .000 .000 .008 .000 .000 .000 .000Maximizing TFB .694 1.200 1.065 .457 1.166 1.049 1.079Lower limit of 90% CI - .741 .466 - .604 .396 .610Upper limit of 90% CI - 1.376 1.269 - 1.355 1.261 1.264This table presents the results of a series of fixed-effects OLS regressions of the number of descendants in generation t on time to first birth measured in years,i.e. TFB and TFB2 for heads of lineages born in the entire sample period. All regressions account for Maternal Founder fixed effects. Birth year and marriageage dummies are included as controls. Furthermore, stoppage age dummies are included in columns 9–10. A dummy indicating unknown literacy is included in theregressions underlying column 5–10. Standard errors clustered at the level of the firstborn are reported in parentheses. * p <0.10, ** p <0.05, *** p <0.01.

Data & Analysis Parametric analysis

Mechanisms

The observed patterns may reflect the positive effect of reduced fertilityand thus higher child quality on the reproductive success of each child

Child quality enhances the likelihood that a child would

Reach the reproductive age and marry – a preconditions for reproductivesuccess

Marry early – a preconditions for large number of children

Be educated and thus would have higher earning capacity and repro-ductive success

Mechanisms

The Effect of TFB on the Fraction of Children Surviving to Age 40 andMarry

Fraction of children survivingto age 40 that got married

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

TFB .299*** .256** .235** .232**(.113) (.112) (.111) (.110)

Literate .770*** .763*** .779***(.110) (.110) (.109)

Male .396*** .365***(.112) (.120)

Stoppage age fixed effects No No No Yes

Number of observations 3,727 3,727 3,727 3,727This table presents the results of a series of fractional logit regressions of the fractionof children, i.e., individuals in the first generation, surviving to age 40 that gotmarried on time to first birth measured in years, i.e., TFB, for heads of lineageswith at least one child surviving to age 40. Birth year and marriage age dummiesare included as controls. Furthermore, stoppage age dummies are included in column4. A dummy indicating unknown literacy is included in the regressions underlyingcolumn 2–4. Standard errors clustered at the level of the firstborn are reported inparentheses. * p <0.10, ** p <0.05, *** p <0.01.

The Effect of TFB on the on the Average Marriage Age

Average marriage age of children

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

TFB -.430*** -.400*** -.376** -.339**(.005) (.008) (.013) (.023)

Literate -.629*** -.621*** -.705***(.001) (.001) (.000)

Male -.406*** -.720***(.009) (.000)

Number of observations 3,796 3,796 3,796 3,796Adjusted R2 .006 .010 .011 .036This table presents the results of a series of OLS regressions of the average marriageage of childre,, i.e., individuals in the first generation, on time to first birth measuredin years, i.e., TFB. Birth year and marriage age dummies are included as controls.Furthermore, stoppage age dummies are included in column 4. A dummy indicatingunknown literacy is included in the regressions underlying column 2–4. Standard errorsclustered at the level of the firstborn are reported in parentheses. * p <0.10, ** p <0.05,*** p <0.01.

The Effect of TFB on the on Share of Literate Offspring

Fraction of literate children

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

TFB .401*** .351*** .322*** .337***(.090) (.090) (.091) (.091)

Literate 1.308*** 1.307*** 1.305***(.094) (.094) (.095)

Male .563*** .407***(.090) (.098)

Number of observations 3,448 3,448 3,448 3,448This table presents the results of a series of fractional logit regressions of the shareof children, i.e., individuals in the first generation, obtaining literacy on time to firstbirth measured in years, i.e., TFB, for heads of lineages with at least one survivingchild with observed literacy status. Birth year and marriage age dummies are includedas controls. Furthermore, stoppage age dummies are included in column 4. A dummyindicating unknown literacy is included in the regressions underlying column 2–4. Standarderrors clustered at the level of the firstborn are reported in parentheses. * p <0.10, **p <0.05, *** p <0.01.

Findings

Maximal reproductive success is attained by couples with a moderatelevel of TFB

Lower TFB maximizes the number of children

An intermediate TFB maximizes long-run reproductive success

In comparison to highly fertile couples whose first child is born withoutany delays (i.e., 38 weeks after the marriage) they have:

0.3 fewer children

0.4 more grandchildren

8.4 more great-grandchildren

15.7 more great-great-grandchildren

Selection of a quality strategy

The optimal level of TFB and fertility is below the population median

Findings

0.3 fewer children

Findings

0.3 fewer children

Findings

0.3 fewer children

Findings

0.3 fewer children

Findings

0.3 fewer children

Findings

0.3 fewer children

Findings

0.3 fewer children

Findings

0.3 fewer children

Findings

0.3 fewer children

TFB and number of descendants for head of lineages born –1685(accounting for Maternal Founder FE)

Log number of descendants in:Gen. 1 Gen. 2 Gen. 3 Gen. 4 Gen. 1 Gen. 2 Gen. 3 Gen. 4 Gen. 1 Gen. 2 Gen. 3 Gen. 4 Gen. 1 Gen. 2 Gen. 3 Gen. 4

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)

TFB -.052** .167 .505** .783*** -.053** .170 .499** .788*** -.062*** .140 .463** .773*** -.077*** .208 .535*** .810***(.024) (.163) (.205) (.264) (.024) (.163) (.205) (.264) (.024) (.162) (.204) (.264) (.011) (.130) (.181) (.258)

TFB2 -.068 -.193*** -.310*** -.070 -.191*** -.313*** -.063 -.183*** -.309*** -.089** -.210*** -.325***(.053) (.067) (.087) (.053) (.067) (.087) (.053) (.067) (.087) (.042) (.059) (.084)

Literate -.008 .060 .145*** .136** -.006 .063 .148*** .138** -.027* .044 .125*** .109*(.031) (.040) (.051) (.066) (.030) (.040) (.051) (.066) (.014) (.032) (.046) (.066)

Male .220*** .254*** .299*** .131** -.028* .025 .085* .036(.031) (.039) (.047) (.060) (.015) (.031) (.043) (.063)

Stoppage age fixed effects No No No No No No No No No No No No Yes Yes Yes Yes

Number of observations 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798Adjusted R2 .015 .016 .038 .306 .017 .016 .041 .307 .032 .029 .052 .307 .799 .442 .296 .355Joint sign.-level of TFB & TFB2 .031 .196 .002 .000 .026 .184 .003 .000 .010 .130 .002 .000 .000 .002 .000 .000Maximizing TFB 1.224 1.307 1.261 1.223 1.304 1.260 1.113 1.263 1.249 1.163 1.272 1.247Lower limit of 90% CI - .961 .999 - .948 1.000 - .827 .976 -.141 1.012 1.002Upper limit of 90% CI - 1.467 1.398 - 1.466 1.397 - 1.435 1.389 1.393 1.403 1.376

This table presents the results of a series of fixed-effects regressions of the number of descendants in generation t on time to first birth measured in years, i.e. TFB and TFB2. All regressions account for Maternal Founder fixed effects. Birth year and marriageage dummies are included as controls. Furthermore, stoppage age dummies are included in columns 13–16. A dummy indicating unknown literacy is included in the regressions underlying column 5–16. Standard errors clustered at the level of the firstborn arereported in parentheses. * p <0.10, ** p <0.05, *** p <0.01.

TFB and number of descendants for head of lineages born 1660–1685(accounting for Maternal Founder FE)

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)

TFB -.065** .237 .544** .830*** -.066*** .238 .537** .830*** -.075*** .193 .487** .802*** -.079*** .201 .498*** .773***(.025) (.175) (.216) (.280) (.025) (.175) (.217) (.282) (.025) (.174) (.216) (.282) (.012) (.139) (.191) (.277)

TFB2 -.096* -.214*** -.337*** -.097* -.212*** -.337*** -.086 -.199*** -.330*** -.088* -.203*** -.323***(.057) (.071) (.092) (.057) (.071) (.093) (.057) (.071) (.093) (.046) (.063) (.091)

Literate -.028 .027 .124** .118* -.031 .023 .120** .116* -.035** .016 .110** .087(.033) (.043) (.055) (.070) (.033) (.042) (.054) (.070) (.015) (.035) (.049) (.070)

Male .219*** .279*** .315*** .177*** -.035** .056* .116*** .115*(.032) (.040) (.049) (.063) (.015) (.032) (.044) (.066)

Number of observations 3,376 3,376 3,376 3,376 3,376 3,376 3,376 3,376 3,376 3,376 3,376 3,376 3,376 3,376 3,376 3,376Adjusted R2 .019 .020 .042 .335 .020 .020 .045 .335 .036 .036 .058 .337 .804 .451 .306 .379Joint sign.-level of TFB & TFB2 .010 .066 .001 .000 .009 .066 .001 .000 .003 .042 .000 .000 .000 .005 .000 .000Maximizing TFB 1.230 1.270 1.232 1.230 1.268 1.231 1.124 1.219 1.214 1.137 1.224 1.196Lower limit of 90% CI -11.448 .923 .961 -11.038 .910 .959 - .761 .919 -1.02 .883 .886Upper limit of 90% CI 1.505 1.427 1.37 1.504 1.427 1.369 - 1.392 1.357 1.387 1.375 1.343

This table presents the results of a series of fixed-effects OLS regressions of the number of descendants in generation t on time to first birth measured in years, i.e. TFB and TFB2 for heads of lineages born in the period 1660–1685. All regressions account forMaternal Founder fixed effects. Birth year and marriage age dummies are included as controls. Furthermore, stoppage age dummies are included in columns 13–16. A dummy indicating unknown literacy is included in the regressions underlying column 5–16.Standard errors clustered at the level of the firstborn are reported in parentheses. * p <0.10, ** p <0.05, *** p <0.01.

Robustness to alternative cohorts: no restrictions on cohorts (accounting forMaternal Founder FE)

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)

TFB -.064*** .055 .319** .409** -.064*** .050 .310** .400** -.071*** .031 .293** .393** -.090*** .096 .378*** .459**(.017) (.120) (.141) (.199) (.017) (.120) (.141) (.199) (.017) (.119) (.140) (.199) (.008) (.094) (.125) (.198)

TFB2 -.039 -.133*** -.192*** -.038 -.129*** -.189*** -.033 -.126*** -.187*** -.060* -.158*** -.212***(.040) (.046) (.065) (.040) (.046) (.065) (.039) (.046) (.065) (.031) (.041) (.065)

Literate .007 .078** .133*** .133*** .007 .078*** .133*** .134*** -.014 .057** .110*** .121**(.023) (.030) (.037) (.049) (.023) (.030) (.037) (.049) (.011) (.024) (.033) (.048)

Male .222*** .220*** .191*** .088** -.013 .005 .019 .019(.018) (.023) (.027) (.036) (.009) (.019) (.025) (.038)

Number of observations 7,664 7,664 7,664 7,664 7,664 7,664 7,664 7,664 7,664 7,664 7,664 7,664 7,664 7,664 7,664 7,664Adjusted R2 .026 .022 .067 .413 .026 .023 .070 .414 .049 .036 .077 .414 .789 .420 .282 .431Joint sign.-level of TFB & TFB2 .000 .015 .000 .000 .000 .018 .000 .000 .000 .008 .000 .000 .000 .000 .000 .000Maximizing TFB .694 1.200 1.065 .663 1.198 1.060 .457 1.166 1.049 .798 1.196 1.079Lower limit of 90% CI - .741 .466 - .708 .434 - .604 .396 -3.176 .924 .610Upper limit of 90% CI - 1.376 1.269 - 1.378 1.268 - 1.355 1.261 1.167 1.333 1.264

This table presents the results of a series of fixed-effects OLS regressions of the number of descendants in generation t on time to first birth measured in years, i.e. TFB and TFB2 for heads of lineages born in the entire sample period. All regressions accountfor Maternal Founder fixed effects. Birth year and marriage age dummies are included as controls. Furthermore, stoppage age dummies are included in columns 13–16. A dummy indicating unknown literacy is included in the regressions underlying column5–16. Standard errors clustered at the level of the firstborn are reported in parentheses. * p <0.10, ** p <0.05, *** p <0.01.

Summary statistics for females

(1) (2) (3) (4)Mean Median S.D. Count

Children 9.42 10 3.66 2,058Grandchildren 45.99 43 27.40 2,058Great-grandchildren 187.65 159 142.74 2,058Great-great-grandchildren 341.04 206.5 408.07 2,058Years from marriage to first birth (TFB) 1.23 1.04 0.49 2,058Literate 0.68 1 0.47 1,192Fraction of literate children 0.72 1 0.36 1,872Fraction of surviving childrenb 0.59 0.60 0.20 2,058Fraction of surviving children with known literacyb 0.62 0.60 0.40 2,044Age at first marriage 19.34 18.7 3.79 2,058Age at last delivery 38.27 40.3 6.46 2,058a The moderate increase in the mean and median number of descendants from the third to the fourthgeneration (i.e. from great-grandchildren to great-great-grandchildren) reflects the fact that these cohortsare less fully observed. Furthermore, since men produce children at lager ages than women, this effect ismore pronounced among men.b Survival is recorded at the average marriage age, i.e. 23 years.

Summary statistics for males

(1) (2) (3) (4)Mean Median S.D. Count

Children 10.03 10 4.32 1,740Grandchildren 48.94 45 28.77 1,740Great-grandchildren 187.53 159 137.10 1,740Great-great-grandchildren 238.38 136.5 293.17 1,740Years from marriage to first birth (TFB) 1.16 0.99 0.44 1,740Literate 0.64 1 0.48 1,030Fraction of literate children 0.76 1 0.34 1,576Fraction of surviving childrenb 0.60 0.60 0.20 1,740Fraction of surviving children with known literacyb 0.57 0.50 0.39 1,728Age at first marriage 26.62 25.9 4.41 1,740Age at last delivery 46.31 46.9 8.81 1,740a The moderate increase in the mean and median number of descendants from the third to the fourthgeneration (i.e. from great-grandchildren to great-great-grandchildren) reflects the fact that these cohortsare less fully observed. Furthermore, since men produce children at lager ages than women, this effect ismore pronounced among men.b Survival is recorded at the average marriage age, i.e. 23 years.

TFB and number of descendants for head of lineages born –1685 (usingGLM w/ neg. binom.)

Number of descendants in:Gen. 1 Gen. 2 Gen. 3 Gen. 4 Gen. 1 Gen. 2 Gen. 3 Gen. 4 Gen. 1 Gen. 2 Gen. 3 Gen. 4 Gen. 1 Gen. 2 Gen. 3 Gen. 4

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)

TFB -.063*** .120 .371** .616*** -.063*** .090 .324* .573** -.081*** .028 .260 .525** -.090*** .049 .346** .573**(.018) (.148) (.185) (.230) (.018) (.148) (.186) (.231) (.017) (.145) (.184) (.231) (.010) (.121) (.167) (.227)

TFB2 -.047 -.133** -.232*** -.039 -.119* -.219*** -.026 -.106* -.209*** -.035 -.137** -.224***(.048) (.060) (.074) (.049) (.061) (.074) (.048) (.060) (.074) (.039) (.054) (.073)

Literate -.002 .143*** .222*** .227*** -.009 .138*** .219*** .225*** -.056*** .098*** .180*** .195***(.020) (.030) (.038) (.046) (.019) (.030) (.038) (.046) (.011) (.025) (.035) (.047)

Male .295*** .379*** .402*** .306*** -.004 .094*** .150*** .194***(.019) (.027) (.035) (.041) (.011) (.023) (.032) (.042)

Number of observations 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798Joint sign.-level of TFB & TFB2 .000 .446 .059 .000 .000 .445 .086 .001 .000 .139 .041 .000 .000 .016 .001 .000Maximizing TFB 1.273 1.399 1.328 1.159 1.363 1.306 .546 1.229 1.255 .702 1.262 1.277

This table presents the results of a series of GLM regressions, with a negative binomial distribution and a logarithmic link function, of the number of descendants in generation t on time to first birth measured in years, i.e. TFB and TFB2. Birth year andmarriage age dummies are included as controls. Furthermore, stoppage age dummies are included in columns 13–16. A dummy indicating unknown literacy is included in the regressions underlying column 5–16. Standard errors clustered at the level of thefirstborn are reported in parentheses. * p <0.10, ** p <0.05, *** p <0.01.

TFB and number of descendants for head of lineages born –1685 (usingGLM w/ neg. binom. – accounting for Maternal Founder FE)

Number of descendants in:Gen. 1 Gen. 2 Gen. 3 Gen. 4 Gen. 1 Gen. 2 Gen. 3 Gen. 4 Gen. 1 Gen. 2 Gen. 3 Gen. 4 Gen. 1 Gen. 2 Gen. 3 Gen. 4

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)

TFB -.041** .126 .412** .547** -.042** .131 .414** .557*** -.052*** .101 .387** .547** -.073*** .195* .468*** .595***(.020) (.146) (.179) (.215) (.020) (.146) (.178) (.215) (.020) (.145) (.177) (.215) (.010) (.118) (.160) (.216)

TFB2 -.054 -.162*** -.234*** -.056 -.163*** -.238*** -.050 -.158*** -.236*** -.087** -.191*** -.256***(.047) (.058) (.068) (.047) (.058) (.069) (.047) (.058) (.068) (.038) (.052) (.069)

Literate -.006 .059* .121*** .119** -.006 .061* .124*** .121** -.020 .055* .118*** .104*(.026) (.036) (.045) (.056) (.025) (.036) (.045) (.056) (.013) (.029) (.040) (.055)

Male .248*** .280*** .282*** .127** -.015 .044 .083** .072(.027) (.036) (.044) (.054) (.014) (.029) (.040) (.055)

Number of observations 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798Joint sign.-level of TFB & TFB2 .045 .207 .001 .000 .037 .189 .001 .000 .009 .097 .000 .000 .000 .000 .000 .000Maximizing TFB 1.172 1.271 1.169 1.174 1.27 1.169 1.01 1.226 1.155 1.116 1.225 1.159

This table presents the results of a series of GLM regressions, with a negative binomial distribution and a logarithmic link function, of the number of descendants in generation t on time to first birth measured in years, i.e. TFB and TFB2. All regressionsinclude dummies for Maternal Founder fixed effects (the results without the Maternal Founder Fixed Effects is presented in Table A.7). Birth year and marriage age dummies are included as controls. Furthermore, stoppage age dummies are included incolumns 13–16. A dummy indicating unknown literacy is included in the regressions underlying column 5–16. Standard errors clustered at the level of the firstborn are reported in parentheses. * p <0.10, ** p <0.05, *** p <0.01.

Robustness to including extinct lineages – accounting for Maternal Founderfixed effects

Log number of descendants in: Log 1+number of descendants in:

Gen. 1 Gen. 2 Gen. 3 Gen. 4 Gen. 1 Gen. 2 Gen. 3 Gen. 4

(1) (2) (3) (4) (5) (6) (7) (8)

TFB -.046* .151 .462** .773*** -.041* .414* .822*** 1.169***(.026) (.172) (.208) (.264) (.021) (.215) (.294) (.336)

TFB2 -.071 -.180*** -.309*** -.149** -.293*** -.427***(.057) (.068) (.087) (.068) (.093) (.108)

Literate -.022 .060 .153*** .138** -.014 .058 .144* .180**(.033) (.042) (.053) (.066) (.027) (.056) (.076) (.087)

Male .321*** .321*** .349*** .131** .276*** .456*** .597*** .478***(.033) (.040) (.050) (.060) (.027) (.051) (.069) (.076)

Number of observations 4,240 4,002 3,933 3,798 4,240 4,240 4,240 4,240Adjusted R2 .052 .044 .068 .307 .054 .065 .084 .241Joint sign.-level of TFB & TFB2 .073 .083 .003 .000 .053 .046 .002 .000Maximizing TFB 1.067 1.283 1.249 1.387 1.403 1.368This table presents the results of a series of fixed-effects regressions of the number of descendants in generation t on time to first birth measured inyears, i.e. TFB and TFB2. All regressions account for Maternal Founder fixed effects. Birth year and marriage age dummies are included as controls. Adummy indicating unknown literacy is also included in the regressions. In columns 1–4, the restriction of at least one observed great-great-grandchildis relaxed and extinct lineages drop out of the sample in the relevant generations. In columns 5–8, the same extended sample is used, but the outcomeis ln(1 + Di,t), where the added number 1 ensures that the logarithmic transformation is defined and all lineages remains included. Standard errorsclustered at the level of the firstborn are reported in parentheses. * p <0.10, ** p <0.05, *** p <0.01.

Robustness to additional sample restrictions: more than one child andmarriage occurring after turning 15 years – accounting for MaternalFounder fixed effects

Log number of descendants in:Gen. 1 Gen. 2 Gen. 3 Gen. 4 Gen. 1 Gen. 2 Gen. 3 Gen. 4

(1) (2) (3) (4) (5) (6) (7) (8)

TFB -.053** .204 .560*** .891*** -.052** .129 .461** .779***(.021) (.157) (.201) (.263) (.025) (.175) (.218) (.279)

TFB2 -.082 -.211*** -.344*** -.057 -.180** -.310***(.051) (.066) (.086) (.058) (.073) (.093)

Literate .025 .090** .175*** .162** -.016 .052 .139*** .123*(.027) (.038) (.050) (.065) (.031) (.041) (.052) (.068)

Male .251*** .272*** .314*** .135** .225*** .269*** .318*** .153**(.029) (.038) (.047) (.061) (.031) (.039) (.048) (.061)

Number of observations 3,738 3,738 3,738 3,738 3,604 3,604 3,604 3,604Adjusted R2 .044 .037 .060 .318 .035 .031 .055 .314Joint sign.-level of TFB & TFB2 .011 .104 .001 .000 .038 .294 .009 .000Maximizing TFB 1.248 1.323 1.296 1.140 1.282 1.256Lower limit of 90% CI - 1.059 1.095 - .798 .973Upper limit of 90% CI - 1.464 1.417 - 1.458 1.396This table presents the results of a series of fixed-effects regressions of the number of descendants in generation t on time to first birth measured inyears, i.e. TFB and TFB2. All regressions account for Maternal Founder fixed effects. Birth year and marriage age dummies are included as controls.A dummy indicating unknown literacy is also included in the regressions. In columns 1–4, the sample is restricted to heads of lineages who produced atleast two children. In columns 5–8, the sample is restricted to heads of lineages who were married after their 15th birthday. Standard errors clusteredat the level of the firstborn are reported in parentheses. * p <0.10, ** p <0.05, *** p <0.01.

Robustness to additional control variables: number of marriages and spousalmigration (accounting for Maternal Founder FE)

(1) (2) (3) (4) (5) (6) (7) (8)

TFB -.059*** .282* .590*** .828*** -.059** .117 .423** .720***(.022) (.157) (.200) (.263) (.024) (.161) (.204) (.264)

TFB2 -.109** -.225*** -.328*** -.054 -.168** -.289***(.051) (.066) (.087) (.052) (.067) (.087)

Literate -.013 .059 .143*** .135** -.007 .067* .148*** .140**(.028) (.039) (.050) (.066) (.030) (.040) (.051) (.066)

Male .168*** .204*** .255*** .116* .198*** .206*** .232*** .037(.028) (.037) (.046) (.060) (.032) (.039) (.047) (.061)

Total number of marriages fixed effects Yes Yes Yes Yes No No No NoTotal number of marriages of spouse fixed effects Yes Yes Yes Yes No No No NoImmigration status of spouse fixed effects No No No No Yes Yes Yes YesEmigration status of spouse fixed effects No No No No Yes Yes Yes Yes

Number of observations 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798Adjusted R2 .188 .116 .097 .312 .035 .035 .062 .315Joint sign.-level of TFB & TFB2 .009 .027 .000 .000 .013 .169 .005 .000Maximizing TFB 1.289 1.311 1.263 1.073 1.260 1.244Lower limit of 90% CI .420 1.063 1.024 - .716 .935Upper limit of 90% CI 1.509 1.446 1.393 - 1.445 1.392This table presents the results of a series of fixed-effects regressions of the number of descendants in generation t on time to first birth measured in years, i.e. TFB andTFB2. All regressions account for Maternal Founder fixed effects. Birth year and marriage age dummies are included as controls. A dummy indicating unknown literacy isalso included in the regressions. In columns 1–4, dummies for the total number of marriages experienced during the lifetime of the heads of lineages, as well as dummies forthe total number of marriages experienced by the first spouses of the heads of lineages, are included. In columns 5–8, dummies indicating the immigration and emigrationstatuses of the head of the first spouses of the heads of lineages are included. Standard errors clustered at the level of the firstborn are reported in parentheses. * p <0.10,** p <0.05, *** p <0.01.

Robustness to gender distinction – sample restricted to females (accountingfor Maternal Founder FE)

(1) (2) (3) (4) (5) (6) (7) (8)

TFB -.079*** .374* .985*** 1.568*** -.080*** .380* .993*** 1.581***(.030) (.214) (.278) (.349) (.030) (.215) (.278) (.349)

TFB2 -.140** -.347*** -.560*** -.142** -.349*** -.564***(.069) (.088) (.109) (.069) (.088) (.110)

Literate -.040 .072 .120 .145(.043) (.067) (.084) (.107)

Number of observations 2,058 2,058 2,058 2,058 2,058 2,058 2,058 2,058Adjusted R2 .089 .064 .091 .267 .089 .064 .092 .267Joint sign.-level of TFB & TFB2 .008 .057 .000 .000 .007 .056 .000 .000Maximizing TFB 1.334 1.418 1.401 1.339 1.422 1.402Lower limit of 90% CI .366 1.24 1.263 .428 1.247 1.267Upper limit of 90% CI 1.58 1.538 1.500 1.583 1.542 1.501This table presents the results of a series of fixed-effects regressions of the number of descendants in generation t on time to first birth measuredin years, i.e. TFB and TFB2 for female heads of lineages. All regressions account for Maternal Founder fixed effects. Birth year and marriage agedummies are included as controls. A dummy indicating unknown literacy is also included in the regressions. Standard errors clustered at the level ofthe firstborn are reported in parentheses. * p <0.10, ** p <0.05, *** p <0.01.

Robustness to additional control variables: location of birth and death –accounting for Maternal Founder fixed effects

(1) (2) (3) (4) (5) (6) (7) (8)

TFB -0.062∗∗∗ 0.106 0.416∗∗ 0.695∗∗∗ -0.062∗∗∗ 0.032 0.391∗ 0.653∗∗

(0.024) (0.163) (0.205) (0.267) (0.023) (0.161) (0.202) (0.255)TFB2 -0.052 -0.168∗∗ -0.284∗∗∗ -0.023 -0.154∗∗ -0.262∗∗∗

(0.053) (0.067) (0.088) (0.053) (0.066) (0.084)Literate -0.008 0.062 0.149∗∗∗ 0.144∗∗ -0.015 0.037 0.097∗ 0.074

(0.031) (0.040) (0.051) (0.066) (0.031) (0.040) (0.052) (0.067)Male 0.214∗∗∗ 0.253∗∗∗ 0.298∗∗∗ 0.134∗∗ 0.203∗∗∗ 0.208∗∗∗ 0.226∗∗∗ 0.033

(0.031) (0.039) (0.048) (0.061) (0.031) (0.038) (0.046) (0.059)Constant 3.861∗∗∗ 5.949∗∗∗ 6.288∗∗∗ 6.228∗∗∗ 2.256∗∗∗ 4.130∗∗∗ 5.229∗∗∗ 4.676∗∗∗

(0.366) (0.649) (0.667) (0.935) (0.428) (0.681) (0.703) (0.921)Birth parish fixed effects Yes Yes Yes Yes No No No NoDeath parish fixed effects No No No No Yes Yes Yes Yes

Number of observations 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798Adjusted R2 0.034 0.030 0.050 0.308 0.066 0.098 0.130 0.370Joint sign.-level of TFB & TFB2 .009 .155 .002 .000 .009 .425 .011 .000Maximizing TFB 1.024 1.234 1.223 .700 1.264 1.243Lower limit of 90% CI - .646 .875 - .607 .892Upper limit of 90% CI - 1.424 1.379 - 1.461 1.402This table presents the results of a series of fixed-effects regressions of the number of descendants in generation t on time to first birth measured inyears, i.e. TFB and TFB2. All regressions account for Maternal Founder fixed effects. Birth year and marriage age dummies are included as controls. Adummy indicating unknown literacy is also included in the regressions. In columns 1–4, dummies indicating the birth (or baptism) parish of the headsof lineages are included. In columns 5–8 dummies indicating the death (or burial) parish of the heads of lineages are included. Standard errors clusteredat the level of the firstborn are reported in parentheses. * p <0.10, ** p <0.05, *** p <0.01.

Robustness to additional control variables: month of marriage and first birth– accounting for Maternal Founder fixed effects

(1) (2) (3) (4) (5) (6) (7) (8)

TFB -0.061∗∗ 0.151 0.482∗∗ 0.798∗∗∗ -0.050∗∗ 0.181 0.500∗∗ 0.787∗∗∗

(0.024) (0.163) (0.205) (0.265) (0.024) (0.166) (0.208) (0.274)TFB2 -0.067 -0.190∗∗∗ -0.318∗∗∗ -0.073 -0.194∗∗∗ -0.314∗∗∗

(0.053) (0.067) (0.087) (0.054) (0.068) (0.089)Literate -0.008 0.062 0.146∗∗∗ 0.137∗∗ -0.004 0.065 0.153∗∗∗ 0.140∗∗

(0.030) (0.040) (0.051) (0.067) (0.030) (0.040) (0.051) (0.067)Male 0.224∗∗∗ 0.255∗∗∗ 0.296∗∗∗ 0.123∗∗ 0.220∗∗∗ 0.257∗∗∗ 0.298∗∗∗ 0.129∗∗

(0.031) (0.039) (0.048) (0.060) (0.031) (0.039) (0.047) (0.060)Month of marriage fixed effects Yes Yes Yes Yes No No No NoMonth of birth of firstborn fixed effects No No No No Yes Yes Yes Yes

Number of observations 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798Adjusted R2 0.032 0.028 0.051 0.307 0.032 0.032 0.056 0.308Joint sign.-level of TFB & TFB2 .010 .109 .001 .000 .037 .169 .001 .000Maximizing TFB 1.126 1.267 1.253 1.242 1.286 1.254Lower limit of 90% CI - .865 .994 - .904 .968Upper limit of 90% CI - 1.433 1.389 - 1.452 1.398This table presents the results of a series of fixed-effects regressions of the number of descendants in generation t on time to first birth measured in years, i.e.TFB and TFB2. All regressions account for Maternal Founder fixed effects. Birth year and marriage age dummies are included as controls. A dummy indicatingunknown literacy is also included in the regressions. In columns 1–4, dummies indicating the months of marriage of the heads of lineages are included. Incolumns 5–8, dummies indicating the month of birth of the the firstborn of the heads of lineages are included. Standard errors clustered at the level of thefirstborn are reported in parentheses. * p <0.10, ** p <0.05, *** p <0.01.

Robustness to additional control variable: birth order (accounting forMaternal Founder FE)

(1) (2) (3) (4) (5) (6) (7) (8)

TFB -.062*** .139 .462** .773*** -.062*** .144 .456** .752***(.024) (.162) (.204) (.264) (.024) (.162) (.204) (.265)

TFB2 -.062 -.183*** -.309*** -.063 -.181*** -.303***(.053) (.067) (.087) (.053) (.067) (.087)

Literate -.006 .064 .149*** .137** -.008 .064 .149*** .141**(.030) (.040) (.051) (.066) (.030) (.040) (.051) (.067)

Male .219*** .250*** .296*** .132** .220*** .248*** .293*** .131**(.031) (.039) (.047) (.060) (.031) (.039) (.047) (.061)

Firstborn .016 .064** .042 -.019(.023) (.031) (.038) (.047)

Birth order fixed effects No No No No Yes Yes Yes Yes

Number of observations 3,798 3,798 3,798 3,798 3,798 3,798 3,798 3,798Adjusted R2 .032 .030 .052 .307 .031 .030 .052 .308Joint sign.-level of TFB & TFB2 .010 .135 .002 .000 .009 .139 .002 .000Maximizing TFB 1.117 1.264 1.249 1.133 1.261 1.240Lower limit of 90% CI - .828 .976 - .807 .949Upper limit of 90% CI - 1.436 1.388 - 1.434 1.384This table presents the results of a series of fixed-effects regressions of the number of descendants in generation t on time to first birth measured inyears, i.e. TFB and TFB2. All regressions account for Maternal Founder fixed effects. Birth year and marriage age dummies are included as controls.A dummy indicating unknown literacy is also included in the regressions. In columns 1–4, a dummy for the firstborn status of the heads of lineagesis included. In columns 5–8, dummies the birth order of the heads of lineages are included. Standard errors clustered at the level of the firstborn arereported in parentheses. * p <0.10, ** p <0.05, *** p <0.01.

Effect of number of children on long-run reproductive success – accountingfor Maternal Founder fixed effects

Log number of descendants in:Gen. 2 Gen. 3 Gen. 4 Gen. 2 Gen. 3 Gen. 4 Gen. 2 Gen. 3 Gen. 4

(1) (2) (3) (4) (5) (6) (7) (8) (9)

Log Number of Children 2.334*** 3.446*** 4.491*** 2.339*** 3.459*** 4.545*** 2.384*** 3.515*** 4.619***(.524) (.842) (1.078) (.525) (.851) (1.091) (.520) (.837) (1.075)

(Log Number of Children)2 -.371 -.728** -1.280*** -.376 -.738** -1.306*** -.388* -.749** -1.324***(.229) (.367) (.474) (.230) (.373) (.481) (.228) (.366) (.474)

Literate .068 .151** .171* .067 .150** .170*(.047) (.070) (.092) (.047) (.070) (.092)

Male .129*** .266*** .307***(.046) (.071) (.092)

Number of observations 4,240 4,240 4,240 4,240 4,240 4,240 4,240 4,240 4,240First stage F (Kleibergen-Paap) 30.701 30.701 30.701 33.094 33.094 33.094 31.891 31.891 31.891Joint sign. of linear and squared terms .000 .000 .000 .000 .000 .000 .000 .000 .000Maximizing number of log(1+D) 3.142 2.367 1.754 3.113 2.343 1.74 3.069 2.345 1.744Exp(Maximizing log(1+D)) 22.15 9.665 4.777 21.488 9.412 4.697 20.52 9.433 4.72Lower limit of 90% CI 1.618 1.770 1.473 1.473 1.756 1.465 2.081 1.767 1.472Upper limit of 90% CI 6.012 8.556 2.805 2.805 8.445 2.757 57.776 7.475 2.702This table presents the results of a series of fixed-effects 2SLS regressions of the log number of descendants in generation t, i.e. ln(1 +Di,t), where Di,t is the number of descendantsthat the head of household i has in generations t, t = 2, 3, 4, on the number of children and the number of children squared, i.e., ln(1 + Di,1) and (ln(1 + Di,1))2, instrumented byvariation in TFB. The added number 1 ensures that the logarithmic transformation is defined for extinct lineages and all that lineages remain in the sample. All regressions accountfor Maternal Founder fixed effects. Birth year, marriage age, and stoppage age dummies are included as controls. Since the second stage of these 2SLS regressions is quadratic in theendogenous regressor, it is necessary to instrument for both the linear and the squared terms in order to identify the parameters. Thus, following Wooldridge (2010), pp. 267–268, azeroth stage is introduced to the analysis, where Di,1 is first regressed on TFB and all the second-stage controls to obtain predicted values of the number of children. The predictednumber of children from the zeroth stage, D1,i is transformed by ln(1 + D1,i ) as well as (ln(1 + D1,i ))2, and these transformed terms are then used as excluded instruments in thesecond stage. A dummy indicating unknown literacy is included in the regressions underlying columns 4–9. Standard errors clustered at the level of the firstborn are reported inparentheses. * p <0.10, ** p <0.05, *** p <0.01.

0 41 60

38 142

1 5 10 15 20Number of children

1 5 10 15 20 25 30Number of children

1 50 100 150 200Number of grandchildren

1 200 400 600 800 1000Number of great-grandchildren

1 1000 2000 3000 4000Number of great-great-grandchildren

Age at first marriage (All)

10 15 20 25 30 35 40 45 50 55 60 65 70Age at marriage (years)

Age at first marriage (Females)

Age at first marriage (Males)

0 41 60

r of c

Return to presentation

38 60 80 100Weeks from marriage to first birth

f g.-g

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23Number of children

The Agricultural Origins of Time Preference

Oded Galor and Omer Ozak

“Patience is bitter, but its fruit is sweet.”

– Aristotle

Oded Galor and Omer Ozak The Agricultural Origins of Time Preference 1 / 115

Introduction Human Evolution and Economic Development

Main Objectives

Geographical origins of human traits

Coevolution of human traits and the growth process in the

course of human history

The effect of the economic environment on the evolutionary

processes that shape the composition of human traits

The impact of the evolution in the composition of human traits

on the growth process

Main Objectives

Main Hypothesis

The geographical origins of human traits and their coevolution

with the growth process is critical for the understanding of:

Main Hypothesis

During the Malthusian epoch

Malthusian forces affected the size and composition of the population

Intergenerationally transmitted traits that generated higher income

Increased reproductive success

Became more prevalent in the population

Evolutionary forces

Increased the representation of traits that were complementary

to the growth process, reinforcing the growth process

Contributed to regional variation in economic performance

across the globe

Main Hypothesis

Evolutionary forces

across the globe

Main Hypothesis

Evolutionary forces

across the globe

Main Hypothesis

Evolutionary forces

across the globe

Main Hypothesis

Evolutionary forces

across the globe

Main Hypothesis

Evolutionary forces

across the globe

Main Hypothesis

Evolutionary forces

across the globe

Main Hypothesis

Evolutionary forces

across the globe

Introduction Delayed Gratification

Long-Term Orientation and Development

Ability to delay gratification (Time Preference, Long-Term Orientation, Patience, Future

Orientation) is a fundamental determinant of:

Academic achievements

Human and physical capital formation

Technological advancement

Comparative Development

Cross Country Variation in Long-Term Orientation

Introduction Hypothesis

Main Hypothesis

The distribution of time preference across the globe reflects

historical variations in the natural return to agricultural

investment across regions

Pre-industrial agro-climatic characteristics that were conducive

to higher natural return to agricultural investment

Triggered selection, adaptation and learning processes

Persistent positive effect on

• Prevalence of long-term orientation

• Economic behavior: education, saving, non-smoking, and

technological adoption

Main Hypothesis

Formation of Time Preference: Mechanism

Occupational Choice

Learning

Reproductive success

Intergenerational transmission of time preference

Occupational Choice

Learning

Occupational Choice

Learning

Occupational Choice

Learning

Individuals with higher long-term orientation:

Chose more lucrative investment opportunities (associated with

delayed return)

Gained socio-economic status, which reinforced their outlook on

long-term orientation

Generated higher income & reproductive success

Transmitted their enhanced long-term orientation to their

offspring

delayed return)

offspring

delayed return)

offspring

delayed return)

offspring

delayed return)

offspring

delayed return)

offspring

Introduction Structure

Structure of the presentation

1 Introduction

2 Model

3 Data

4 Country-Level Analysis

5 Second Generation Migrants

6 Individual-Level WVS

7 Conclusions

Model Basic Structure

The Model

Overlapping generations economy

Agricultural stage of development

no financial markets

no storage technology

Continuum of three-period lived individuals

Childhood - passive economic agents

1st working period

2nd working period

The Model

1st working period

2nd working period

The Model

1st working period

2nd working period

The Model

1st working period

2nd working period

The Model

1st working period

2nd working period

The Model

1st working period

2nd working period

The Model

1st working period

2nd working period

The Model

1st working period

2nd working period

Model Production

Feasible Modes of Production for Individual i

Two feasible modes of production:

Endowment ModeInvestment Mode

Income generated by member i of generation t

(yi ,t , yi ,t+1) =

(R0,R0) under endowment mode

(1,R1) under investment mode

where R1 > R0 > 1 Malthusian

Model Production

Endowment Mode

Investment Mode

(yi ,t , yi ,t+1) =

Model Production

(yi ,t , yi ,t+1) =

Model Production

(yi ,t , yi ,t+1) =

Model Production

(yi ,t , yi ,t+1) =

Model Production

(yi ,t , yi ,t+1) =

Model Individuals

Member i of generation t

Preferences

ui ,t = ln ci ,t + β it [γ ln ni ,t+1 + (1− γ) ln ci ,t+1]

Discount factor ≡ βit = 1/(1 + ρit) ∈ (0, 1)

Rate of time preference ≡ ρit ≥ 0

Model Individuals

Preferences

Model Individuals

Preferences

Model Individuals

Budget Constraints:

First working period

ci ,t ≤ yi ,t =

R0 under endowment mode

1 under investment mode

Second working period

τni ,t+1 + ci ,t+1 ≤ yi ,t+1 =

R1 under investment mode

Cost of raising a child ≡ τ > 0

Model Individuals

Budget Constraints:

ci ,t ≤ yi ,t =

τni ,t+1 + ci ,t+1 ≤ yi ,t+1 =

Model Individuals

Budget Constraints:

ci ,t ≤ yi ,t =

τni ,t+1 + ci ,t+1 ≤ yi ,t+1 =

Model Individuals

Budget Constraints:

ci ,t ≤ yi ,t =

τni ,t+1 + ci ,t+1 ≤ yi ,t+1 =

Model Individuals

Budget Constraints:

ci ,t ≤ yi ,t =

τni ,t+1 + ci ,t+1 ≤ yi ,t+1 =

Model Individuals

Budget Constraints:

ci ,t ≤ yi ,t =

τni ,t+1 + ci ,t+1 ≤ yi ,t+1 =

Model Optimization

Optimization

Optimal consumption and number of children

ci ,t =yi ,t ;

ci ,t+1 =(1− γ)yi ,t+1;

ni ,t+1 =γ

τyi ,t+1

Model Optimization

Optimization

ci ,t =yi ,t ;

ci ,t+1 =(1− γ)yi ,t+1;

ni ,t+1 =γ

τyi ,t+1

Model Optimization

Optimization

ci ,t =yi ,t ;

ci ,t+1 =(1− γ)yi ,t+1;

ni ,t+1 =γ

τyi ,t+1

Model Optimization

Optimization

ci ,t =yi ,t ;

ci ,t+1 =(1− γ)yi ,t+1;

ni ,t+1 =γ

τyi ,t+1

Model Optimization

Time Preference and Mode of Production

Indirect utility

v i ,t =

lnR0 + β i

t [lnR0 + ξ] under endowment mode

ln 1 + β it [lnR

1 + ξ] under investment mode.

ξ ≡ γ ln(γ/τ) + (1− γ) ln(1− γ)]

⇒ Threshold level of time preference

β =lnR0

lnR1 − lnR0∈ (0, 1)

Model Optimization

Indirect utility

v i ,t =

lnR0 + β i

ln 1 + β it [lnR

ξ ≡ γ ln(γ/τ) + (1− γ) ln(1− γ)]

β =lnR0

lnR1 − lnR0∈ (0, 1)

Model Optimization

Indirect utility

v i ,t =

lnR0 + β i

ln 1 + β it [lnR

ξ ≡ γ ln(γ/τ) + (1− γ) ln(1− γ)]

β =lnR0

lnR1 − lnR0∈ (0, 1)

Model Optimization

Indirect utility

v i ,t =

lnR0 + β it [lnR

0 + ξ] under endowment mode

ln 1 + β it [lnR

ξ ≡ γ ln(γ/τ) + (1− γ) ln(1− γ)]

β =lnR0

lnR1 − lnR0∈ (0, 1)

Model Optimization

Indirect utility

v i ,t =

lnR0 + β i

ln 1 + β it [lnR

ξ ≡ γ ln(γ/τ) + (1− γ) ln(1− γ)]

β =lnR0

lnR1 − lnR0∈ (0, 1)

Model Optimization

Indirect utility

v i ,t =

lnR0 + β i

ln 1 + β it [lnR

ξ ≡ γ ln(γ/τ) + (1− γ) ln(1− γ)]

β =lnR0

lnR1 − lnR0∈ (0, 1)

Model Optimization

Indirect utility

v i ,t =

lnR0 + β i

ln 1 + β it [lnR

ξ ≡ γ ln(γ/τ) + (1− γ) ln(1− γ)]

β =lnR0

lnR1 − lnR0∈ (0, 1)

Model Optimization

Indirect utility

v i ,t =

lnR0 + β i

ln 1 + β it [lnR

ξ ≡ γ ln(γ/τ) + (1− γ) ln(1− γ)]

β =lnR0

lnR1 − lnR0∈ (0, 1)

Model Optimization

Indirect utility

v i ,t =

lnR0 + β i

ln 1 + β it [lnR

ξ ≡ γ ln(γ/τ) + (1− γ) ln(1− γ)]

β =lnR0

lnR1 − lnR0∈ (0, 1)

Model Optimization

Mode of Production

endowment mode if β it ≤ β

investment mode if β it ≥ β

Threshold level is decreasing in the rate of return to investment

∂R1< 0

Model Optimization

Mode of Productionendowment mode if β it ≤ β

∂R1< 0

Model Optimization

∂R1< 0

Model Optimization

∂R1< 0

Model Optimization

Time Preference, Income and Fertility

Income

(yi ,t , yi ,t+1) =

(R0,R0) if β i

t ≤ β

(1,R1) if β it > β

Fertility

ni ,t+1 =

γτR0 ≡ nE if β i

t ≤ β

γτR1 ≡ nI if β i

t > β

⇒nI > nE

Model Optimization

Income

(yi ,t , yi ,t+1) =

(R0,R0) if β i

t ≤ β

(1,R1) if β it > β

Fertility

ni ,t+1 =

t ≤ β

t > β

⇒nI > nE

Model Optimization

Income

(yi ,t , yi ,t+1) =

(R0,R0) if β i

t ≤ β

(1,R1) if β it > β

Fertility

ni ,t+1 =

t ≤ β

t > β

⇒nI > nE

Model Optimization

Income

(yi ,t , yi ,t+1) =

(R0,R0) if β i

t ≤ β

(1,R1) if β it > β

Fertility

ni ,t+1 =

t ≤ β

t > β

⇒nI > nE

Model Optimization

Income

(yi ,t , yi ,t+1) =

(R0,R0) if β i

t ≤ β

(1,R1) if β it > β

Fertility

ni ,t+1 =

t ≤ β

t > β

⇒nI > nE

Model Evolution of Time Preference

Evolution of Time Preference

Parents transmit their time preference to their children

β it+1 =

β it if β i

t ≤ β

φ(β it ;R

1) if β it ≥ β

Engagement in the investment mode enhances long-term

orientation

β it+1 =

β it if β i

t ≤ β

φ(β it ;R

1) if β it ≥ β

orientation

β it+1 =

β it if β i

t ≤ β

φ(β it ;R

1) if β it ≥ β

orientation

The Evolution of Time Preference within a Dynasty

itE)( 1RE

);( 1RitEI

it EE �1

it 1�E

i0EE 1

Evolution of the Composition of Each Generation

Evolution of population of each type in generation t

LEt = (nE )tLE0

LIt = (nI)tLI0

Fraction of endowment type in generation t

θEt ≡LEt

LEt + LIt= θEt

Vanishes asymptotically

limt→∞

θEt = 0

LEt = (nE )tLE0

LIt = (nI)tLI0

θEt ≡LEt

LEt + LIt= θEt

limt→∞

θEt = 0

LEt = (nE )tLE0

LIt = (nI)tLI0

θEt ≡LEt

LEt + LIt= θEt

limt→∞

θEt = 0

LEt = (nE )tLE0

LIt = (nI)tLI0

θEt ≡LEt

LEt + LIt= θEt

limt→∞

θEt = 0

Average time preference

βt = θEt βEt + (1− θEt )βIt

βEt ≡ average time preference of endowment type

βIt ≡ average time preference of investment type

Converges to the steady state of the investment type

limt→∞

θEt = 0⇒ limt→∞

βt = limt→∞

βIt = β(R1)

Increases in return to investment

∂β(R1)

∂R1> 0

limt→∞

βt = limt→∞

βIt = β(R1)

∂β(R1)

∂R1> 0

limt→∞

βt = limt→∞

βIt = β(R1)

∂β(R1)

∂R1> 0

limt→∞

βt = limt→∞

βIt = β(R1)

∂β(R1)

∂R1> 0

limt→∞

βt = limt→∞

βIt = β(R1)

∂β(R1)

∂R1> 0

limt→∞

θEt = 0

⇒ limt→∞

βt = limt→∞

βIt = β(R1)

∂β(R1)

∂R1> 0

limt→∞

βt = limt→∞

= β(R1)

∂β(R1)

∂R1> 0

limt→∞

βt = limt→∞

βIt = β(R1)

∂β(R1)

∂R1> 0

limt→∞

βt = limt→∞

βIt = β(R1)

∂β(R1)

∂R1> 0

Model Comparative Statics

Cross-Country Differences in Return to Investment

);( Ait REI

it EE �1

it 1�E

);( Bit REI

BE AE )( ARE )( BRE

Model Comparative Statics

Effect of an Increase in Return to Investment

itE)( 1RE

);( 1RitEI

it EE �1

it 1�E

1)( 1 RR '�E'E E

);( 1 RRit '�EI

Model Testable Predictions

Testable Predictions

Descendants of inhabitants of regions characterized by higher yield

(conditional on growth cycle) have higher long-term orientation

The expansion in the spectrum of potential crops in the post-1500

period increased long-term orientation in society

An increase in the crop growth cycle generates conflicting effects on

Holding the crop yield constant, longer growth cycle reduces the

return to investment

Longer duration of investment mitigates the aversion from

delayed consumption

Data Data

FAO/GAEZ project

Major cropsOver global 5′ × 5′ grid (10km × 10km)Potential yield

Unaffected by time preference

Low level of inputs and rain-fed agriculture

Reflecting early stages of development

Agro-climatic conditions

Unaffected by human intervention

USDA Nutrient Database for Standard Reference

Caloric content per gram for each crop

Data Data

FAO/GAEZ project

Major crops

Over global 5′ × 5′ grid (10km × 10km)Potential yield

Data Data

FAO/GAEZ project

Major cropsOver global 5′ × 5′ grid (10km × 10km)

Potential yield

Data Data

FAO/GAEZ project

Data Data

FAO/GAEZ project

Data Data

FAO/GAEZ project

Data Data

FAO/GAEZ project

Data Data

FAO/GAEZ project

Data Data

FAO/GAEZ project

Data Data

FAO/GAEZ project

Data Data

FAO/GAEZ project

Data Data

Caloric Suitability Index (CSI)

Potential Crop Yield

Cell: Calories per hectare per year of the most productive crop

Potential Crop Growth Cycles

Cell: average of days elapsed from planting to harvesting for themost productive crop

Data Data

Data Crop Yield

Potential Crop Yield pre-1500CE Post-1500CE

Data Crop Growth Cycle

Potential Crop Growth Cycle pre-1500CE Post-1500CE

Data Crop Return

Potential Crop Return pre-1500CE Post-1500CE

Data Potential vs Actual Yield

Potential vs Actual Yield

Data Most Productive Crops pre-1500CE

Most Productive Crops pre-1500CE

Other Rule

Most Productive Crops post-1500CE

LTO, Crop Yield, Growth Cycle and Return - Old World

Top Crop All Crops LTO

Region Crop Yield Cycle Return Yield Cycle Return

Europe Barley 8371 125 68 6117 112 52 66

Asia Rice 8709 139 63 5973 127 46 64

SSA Pea 4495 190 23 4180 189 22 20

Country-Level Analysis Long-Term Orientation Measure

Country-Level Analysis

Long-Term Orientation

Country-level measure (Hofstede, 1991)

“the fostering of virtues oriented toward future

rewards, in particular, perseverance and thrift”

0 (Short-Term) to 100 (Long-Term)

Country-Level Analysis Correlations

Correlations: Long-Term Orientation and Development

(a) GDP per capitaOded Galor and Omer Ozak The Agricultural Origins of Time Preference 40 / 115

(a) SchoolingOded Galor and Omer Ozak The Agricultural Origins of Time Preference 41 / 115

(a) Growth rateOded Galor and Omer Ozak The Agricultural Origins of Time Preference 42 / 115

Crop Yield and Long-Term Orientation

Whole World Old World

(1) (2) (3) (4) (5) (6) (7) (8)

Crop Yield 7.43*** 9.84*** 9.06*** 9.46*** 13.26*** 15.23***(2.48) (2.88) (2.62) (3.41) (2.55) (3.58)

Crop Growth Cycle -0.70 -3.18(3.96) (4.03)

Crop Yield (Ancestors) 11.58*** 13.31***(2.15) (2.94)

Crop Growth Cycle (Ancestors) -3.15(3.52)

Absolute latitude 2.85 1.88 1.68 4.72 3.99 4.76 3.87(4.05) (3.85) (4.33) (3.29) (3.63) (4.15) (4.71)

Mean elevation 4.98* 5.97** 6.09** 5.56** 5.96** 4.58 4.87(2.87) (2.96) (3.03) (2.48) (2.46) (2.99) (3.03)

Terrain Roughness -6.24** -5.72** -5.72** -6.74*** -6.72*** -6.40** -6.29**(2.51) (2.75) (2.75) (2.53) (2.49) (2.83) (2.82)

Neolithic Transition Timing -6.46** -6.31** -4.75* -4.08(2.87) (3.06) (2.60) (2.66)

Neolithic Transition Timing (Ancestors) -4.77** -4.31*(2.24) (2.30)

Continent FE Yes Yes Yes Yes Yes Yes Yes YesAdditional Geographical Controls No Yes Yes Yes Yes Yes Yes YesOld World Sample No No No No No No Yes Yes

Adjusted-R2 0.54 0.60 0.62 0.61 0.66 0.66 0.61 0.61Observations 87 87 87 87 87 87 72 72

Country-Level Analysis Identification

Identification Strategy

Potential Concerns:

Reverse causality:

Time preference ⇒ actual return to agricultural investment

Choice of cropsChoice of technology

Remedy:

Exploit variation in potential (rather than actual) return toagricultural investment

Potential Concerns:

Reverse causality:

Remedy:

Potential Concerns:

Reverse causality:

Remedy:

Potential Concerns:

Reverse causality:

Choice of crops

Choice of technology

Remedy:

Potential Concerns:

Reverse causality:

Remedy:

Potential Concerns:

Reverse causality:

Remedy:

Potential Concerns:

Reverse causality:

Remedy:

Potential Concerns:

Omitted Variable:

Geography

& Growth Cycle

Institutions

Culture

& Growth Cycle

Potential Concerns:

Omitted Variable:

Geography

& Growth Cycle

Institutions

Culture

& Growth Cycle

Potential Concerns:

Omitted Variable:

Geography

& Growth Cycle

Institutions

Culture

& Growth Cycle

Potential Concerns:

Omitted Variable:

Geography

& Growth Cycle

Institutions

Culture

& Growth Cycle

Remedy:

Account for the confounding effects of:

Geographical characteristics(e.g., absolute latitude, elevation, roughness, distance to waterways, etc.)

Continental FEsAdjusting for the roots of contemporary populations

Analysis of 2nd generation migrants:

Accounting for host country FEsFocusing on portable component of the effect of crop yieldIndividual characteristics (e.g., gender, age, religion, etc.)

Regional analysis:

Accounting for country FEsIndividual characteristics (e.g., gender, age, religion, etc.)

Remedy:

Regional analysis:

Remedy:

Regional analysis:

Remedy:

Continental FEs

Adjusting for the roots of contemporary populations

Regional analysis:

Remedy:

Regional analysis:

Remedy:

Regional analysis:

Remedy:

Accounting for host country FEs

Focusing on portable component of the effect of crop yieldIndividual characteristics (e.g., gender, age, religion, etc.)

Regional analysis:

Remedy:

Accounting for host country FEsFocusing on portable component of the effect of crop yield

Individual characteristics (e.g., gender, age, religion, etc.)

Regional analysis:

Remedy:

Regional analysis:

Remedy:

Regional analysis:

Remedy:

Regional analysis:

Accounting for country FEs

Individual characteristics (e.g., gender, age, religion, etc.)

Remedy:

Regional analysis:

Empirical Specification

LTOi =β0 + β1crop yieldi + β2crop growth cyclei

γ0jXij + γ1YSTi + δc∆i + εi ,

LTOi ≡ Long-Term Orientation measure

Xij ≡ Geographical controls

YSTi ≡ Years since transition to agriculture

∆i ≡ Continental FEs

Crop Yield and Long-Term Orientation

(1) (2) (3) (4) (5) (6) (7) (8)

Crop Yield 7.43*** 9.84*** 9.06*** 9.46*** 13.26*** 15.23***(2.48) (2.88) (2.62) (3.41) (2.55) (3.58)

Crop Growth Cycle -0.70 -3.18(3.96) (4.03)

Crop Growth Cycle (Ancestors) -3.15(3.52)

Mean elevation 4.98* 5.97** 6.09** 5.56** 5.96** 4.58 4.87(2.87) (2.96) (3.03) (2.48) (2.46) (2.99) (3.03)

Terrain Roughness -6.24** -5.72** -5.72** -6.74*** -6.72*** -6.40** -6.29**(2.51) (2.75) (2.75) (2.53) (2.49) (2.83) (2.82)

Neolithic Transition Timing -6.46** -6.31** -4.75* -4.08(2.87) (3.06) (2.60) (2.66)

Neolithic Transition Timing (Ancestors) -4.77** -4.31*(2.24) (2.30)

Continent FE Yes Yes Yes Yes Yes Yes Yes YesAdditional Geographical Controls No Yes Yes Yes Yes Yes Yes YesOld World Sample No No No No No No Yes Yes

Partial Correlation: Crop Yield and LTO

(a) Ancestry Adjusted (b) Old World

Potential Concerns: Selection

High long-term orientation individuals settled in regions whichreward LTO

Note: This selection process will result in the same geographicalorigins in LTO

Remedy:

Exploit natural experiment associated with Columbian Exchange

Changes in the spectrum of potential crops in the post-1500period Crops

→ Assignment of potentially superseding crops to existingindividuals across regions Random

Remedy:

Potential Concerns: Historical vs Contemporary Effect

Geography

HistoricalPotential Yield

ContemporaryPotential Yield

Long-TermOrientation

Geography

Pre-1500CE Crop Yield and LTO ∆ > 0 Natives

(1) (2) (3) (4) (5) (6) (7) (8)

Crop Yield (pre-1500) 5.67** 5.98*** 7.28*** 8.82*** 12.23*** 15.21***(2.40) (2.09) (2.29) (3.13) (2.84) (3.51)

Crop Yield Change (post-1500) 7.88** 8.77*** 9.83*** 7.95*** 10.53***(3.08) (2.69) (3.11) (2.56) (3.30)

Crop Growth Cycle (pre-1500) -3.77 -7.65(4.17) (4.80)

Crop Growth Cycle Change (post-1500) 0.16 0.31(1.90) (1.73)

Crop Yield (Ancestors, pre-1500) 8.62*** 10.56***(2.01) (2.35)

Crop Yield Change (Anc., post-1500) 8.03*** 9.86***(2.03) (2.28)

Crop Growth Cycle (Ancestors, pre-1500) -7.31**(3.59)

Crop Growth Cycle Change (Anc., post-1500) 0.77(1.60)

Continent FE Yes Yes Yes Yes Yes Yes Yes YesGeographical & Neolithic No No Yes Yes Yes Yes Yes YesOld World Sample No No No No No No Yes Yes

Excluding the Persistence of Development Channel

Agricultural productivity (crop yield)

→ Population density

→ Urbanization

Persistence of pre-industrial development

→ Income, education, etc.

→ Long-term orientation

→ Urbanization

Excluding the Pre-Industrial Development Channel

Population Density Urbanization GDP per capita

1500CE 1500CE 1800CE 1870CE 1913CE

(1) (2) (3) (4) (5) (6) (7) (8)

Crop Yield (Anc., pre-1500) 11.05*** 11.52*** 10.01*** 11.08*** 11.54*** 11.54*** 14.19*** 12.66**(2.53) (2.33) (3.68) (3.68) (3.18) (3.22) (5.08) (5.02)

Crop Yield Change (post-1500) 10.76*** 10.40*** 8.77** 9.96*** 10.05*** 10.22*** 15.55*** 14.92***(2.89) (2.78) (3.35) (3.35) (3.23) (3.37) (3.22) (3.29)

Crop Growth Cycle (Anc., pre-1500) -8.06* -10.43*** -5.06 -7.30 -8.60* -8.75* -12.58* -10.28(4.06) (3.63) (5.28) (5.37) (4.68) (4.84) (6.44) (6.46)

Crop Growth Cycle Ch. (post-1500) -0.46 -1.06 1.06 0.55 0.07 0.03 2.14 3.31(1.72) (1.84) (2.91) (2.95) (2.37) (2.41) (3.38) (3.35)

Population density in 1500 CE 3.76**(1.86)

Urbanization rate in 1500 CE 1.90(2.24)

Urbanization rate in 1800 CE -0.57(1.22)

GDP per capita 1870 10.57***(3.65)

GDP per capita 1913 10.99***(3.53)

Semi-Partial R2

Crop Yield (Anc., pre-1500) 0.08*** 0.09*** 0.04*** 0.04*** 0.07*** 0.07*** 0.09*** 0.07**Crop Yield Change (post-1500) 0.05*** 0.05*** 0.03** 0.03*** 0.04*** 0.04*** 0.10*** 0.09***Crop Growth Cycle (Anc., pre-1500) 0.02* 0.03*** 0.00 0.01 0.02* 0.02* 0.04* 0.03Crop Growth Cycle Ch. (post-1500) 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01Population density in 1500 CE 0.01**Urbanization rate in 1500 CE 0.00Urbanization rate in 1800 CE 0.00GDPpc 1870 0.05***GDPpc 1913 0.05***

Continental FE Yes Yes Yes Yes Yes Yes Yes YesGeography & Neolithic Yes Yes Yes Yes Yes Yes Yes Yes

Excluding Other Channels

Average Suitability PCA

Future Time Reference (FTR)

→ Long-Term Orientation

Agricultural Suitability Plow Future Time Reference

(1) (2) (3) (4) (5) (6) (7) (8) (9)

Crop Yield (Ancestors, pre-1500) 12.02*** 11.46*** 10.36*** 12.85*** 12.80*** 12.72*** 13.05*** 14.10*** 13.95***(2.69) (2.91) (3.32) (2.65) (2.67) (2.70) (2.75) (2.77) (2.80)

Crop Yield Change (post-1500) 10.70*** 10.50*** 10.03*** 10.93*** 10.93*** 11.17*** 10.30*** 9.89*** 10.13***(2.71) (2.70) (2.73) (2.77) (2.78) (2.76) (3.16) (2.88) (3.02)

Crop Growth Cycle (Ancestors, pre-1500) -7.63* -7.71* -8.04* -10.02** -10.13** -10.50*** -10.87** -10.05** -10.21**(3.85) (3.94) (4.09) (3.94) (3.92) (3.94) (4.14) (3.80) (3.97)

Crop Growth Cycle Change (post-1500) -0.90 -0.96 -1.16 -1.30 -1.40 -1.63 -1.09 -0.86 -0.97(1.62) (1.68) (1.76) (1.69) (1.66) (1.61) (1.62) (1.72) (1.70)

Land Suitability 0.83(2.07)

Land Suitability (Ancestors) 2.34(3.20)

Plow 1.62(3.17)

Plow (Ancestors) 3.35(3.92)

Strong FTR -3.68**(1.68)

Strong FTR (Ancestors) -2.59(1.76)

Semi-Partial R2

Crop Yield (Ancestors, pre-1500) 0.07*** 0.05*** 0.03*** 0.08*** 0.08*** 0.08*** 0.08*** 0.09*** 0.09***Crop Yield Change (post-1500) 0.05*** 0.05*** 0.04*** 0.05*** 0.05*** 0.05*** 0.04*** 0.03*** 0.04***Crop Growth Cycle (Ancestors, pre-1500) 0.01* 0.01* 0.02* 0.03** 0.03** 0.03*** 0.03** 0.02** 0.02**Crop Growth Cycle Change (post-1500) 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00Land Suitability 0.00Land Suitability (Ancestors) 0.00Plow 0.00Plow (Ancestors) 0.00Strong FTR 0.02**Strong FTR (Ancestors) 0.01

Continental FE Yes Yes Yes Yes Yes Yes Yes Yes YesGeography & Neolithic Yes Yes Yes Yes Yes Yes Yes Yes YesAdjusted-R2 0.68 0.67 0.68 0.67 0.66 0.67 0.70 0.72 0.70Observations 85 85 85 87 87 87 71 71 71

Excluding Other Cultural Channels

Long-Term Orientation is correlated with other cultural traits.

Potential concern:

→ Potential yield determines other cultural traits

→ Other cultural traits determine LTO

Potential concern:

Excluding Other Cultural Channels Corr

Cultural Indices

RestraintvsIndulgence

Trust Indivi-dualism

PowerDistance

Coope-ration

UncertaintyAvoidance

(1) (2) (3) (4) (5) (6) (7)

Crop Yield (Ancestors, pre-1500) 10.03*** 6.58 -7.11* -10.88 6.69 -7.60 3.03(3.05) (3.99) (3.72) (6.59) (5.92) (5.98) (5.55)

Crop Yield Change (Anc., post-1500) 9.03*** 7.91** -0.53 -3.05 2.50 -1.51 -0.39(2.16) (3.10) (3.48) (2.62) (2.18) (2.23) (2.21)

Crop Growth Cycle (Ancestors, pre-1500) -5.98** -4.59 0.35 2.20 -2.50 3.50 4.06(2.75) (3.57) (3.47) (3.82) (4.11) (4.15) (4.33)

Crop Growth Cycle Change (Anc., post-1500) -0.77 2.02 1.96 -3.72 -0.89 3.00 -0.05(1.60) (2.42) (2.09) (3.18) (2.90) (2.51) (3.24)

Land Suitability (Ancestors) 2.33 0.91 -6.17 6.94 7.75* 12.54*** 6.08(3.15) (4.86) (5.10) (4.99) (4.22) (3.91) (3.98)

Neolithic Transition Timing (Ancestors) -7.58** -0.19 0.56 -0.60 -2.13 1.22 -8.88**(3.04) (4.62) (4.09) (3.32) (4.40) (5.85) (3.77)

Continental FE Yes Yes Yes Yes Yes Yes YesAll Geographical Controls Yes Yes Yes Yes Yes Yes YesAdjusted-R2 0.68 0.41 0.46 0.68 0.39 0.46 0.60Observations 85 83 83 60 60 60 60

(1) (2) (3) (4) (5) (6) (7)

Crop Yield (Ancestors, pre-1500) 10.03*** 9.38*** 10.30*** 13.54** 11.47* 12.76* 11.17*(3.05) (3.21) (3.41) (6.49) (6.78) (6.78) (6.53)

Crop Yield Change (Anc., post-1500) 9.03*** 8.55*** 8.97*** 7.45*** 6.88** 7.11*** 6.84***(2.16) (2.53) (2.23) (2.47) (2.63) (2.53) (2.50)

Crop Growth Cycle (Ancestors, pre-1500) -5.98** -5.71* -6.05** -5.53 -5.14 -5.75 -5.29(2.75) (3.08) (2.76) (4.88) (5.32) (5.14) (4.89)

Crop Growth Cycle Change (Anc., post-1500) -0.77 -0.88 -0.71 0.17 -0.61 -1.16 -0.59(1.60) (1.71) (1.84) (3.11) (3.11) (3.20) (3.03)

Restraint vs. Indulgence 2.18(2.22)

Trust 0.63(3.10)

Individualism 4.80(3.96)

Power Distance -0.45(3.90)

Cooperation 3.95(4.20)

Uncertainty Avoidance 1.18(6.06)

Land Suitability (Ancestors) 2.33 2.30 2.35 -2.71 -1.13 -3.67 -1.61(3.15) (3.30) (3.51) (4.93) (4.76) (5.54) (5.32)

Neolithic Transition Timing (Ancestors) -7.58** -7.49** -7.51** -7.86 -8.03 -8.22 -7.53(3.04) (3.05) (3.14) (5.32) (5.34) (5.07) (5.91)

Continental FE Yes Yes Yes Yes Yes Yes YesAll Geographical Controls Yes Yes Yes Yes Yes Yes YesAdjusted-R2 0.68 0.68 0.67 0.59 0.58 0.59 0.58Observations 85 83 83 60 60 60 60

Robustness

Including Cells with Zero Caloric Output Table Zeroes

Daily Return Table Daily

Trade Table Trade

Population Age Structure

Life-Expectancy Table Age

Current Income

Climatic Variability Table Climatic

Spatial Autocorrelation (Cliff and Ord, 1973; Conley, 1999)

Omitted Variable Bias (Altonji, Elder, and Taber, 2005; Bellows and Miguel,

2009; Oster, 2014) Table AET Table AET Changes

Robustness

Trade Table Trade

Current Income

Robustness

Trade Table Trade

Current Income

Robustness

Trade Table Trade

Current Income

Robustness

Trade Table Trade

Current Income

Robustness

Trade Table Trade

Current Income

Robustness

Trade Table Trade

Current Income

Robustness

Trade Table Trade

Current Income

Robustness

Trade Table Trade

Current Income

Crop Yield, LTO and Technological Adoption

Major Technological Changes (Probit)

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

Crop Yield (pre-1500) 0.10** 0.13** 0.15*** 0.17** 0.30*** 0.29***(0.05) (0.05) (0.05) (0.06) (0.05) (0.06)

Crop Yield Ch. (post-1500) 0.06 0.09* 0.16*** 0.21***(0.05) (0.05) (0.04) (0.06)

Crop Cycle (pre-1500) -0.13 -0.22*** -0.21**(0.08) (0.08) (0.09)

Crop Growth Cycle Ch. (post-1500) -0.12* -0.23*** -0.19***(0.06) (0.06) (0.07)

Geographical Controls No Yes Yes Yes Yes YesLanguage Family FE No No No No Yes YesContinental FE No No No No No Yes

Pseudo-R2 0.04 0.13 0.15 0.18 0.43 0.45Observations 86 86 86 86 86 86

Crop Yield, LTO and Education

Years of Schooling in 2005

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

Crop Yield (Ancestors, pre-1500) 0.93*** 0.90*** 0.90*** 0.90*** 0.84*** 0.88***(0.24) (0.30) (0.24) (0.29) (0.23) (0.28)

Crop Growth Cycle (Ancestors, pre-1500) -0.08 -0.05 -0.04 -0.04 0.03 0.03(0.20) (0.23) (0.19) (0.23) (0.24) (0.32)

Crop Yield Change (post-1500) -0.05 0.02 0.09(0.27) (0.26) (0.34)

Crop Growth Cycle Change (post-1500) 0.00 0.02 0.08(0.16) (0.16) (0.17)

Geographical Controls Yes Yes Yes Yes Yes YesTiming of Neolithic No No Yes Yes Yes YesContinental FE No No No No Yes Yes

Adjusted-R2 0.52 0.51 0.53 0.52 0.59 0.58Observations 129 129 129 129 129 129

Second Generation Migrants Motivation

Second Generation Migrants Analysis Data

Accounts for host country FEs(geography, institutions, culture)

Accounts for individual characteristics(e.g., age, gender, education, etc.)

Focus on portable component of the effect of crop yield

Correlations: Long-Term Orientation and Education

Years of Schooling

Second Generation Migrants All Individuals

(1) (2) (3) (4) (5) (6) (7) (8)

Long-Term Orientation 0.35*** 0.37*** 0.36** 0.32** 0.79*** 0.88*** 0.70*** 0.63***(0.13) (0.14) (0.14) (0.13) (0.05) (0.05) (0.05) (0.04)

Country FE No Yes Yes Yes No Yes Yes YesSex & Age No No Yes Yes No No Yes YesPray & Health No No No Yes No No No Yes

Adjusted-R2 0.01 0.10 0.10 0.11 0.04 0.15 0.19 0.21

R2 0.01 0.13 0.13 0.16 0.04 0.15 0.20 0.21Observations 705 705 705 705 42016 42016 42016 42016

Income

Second Generation Migrants Empirical Analysis

LTOic =β0 + β1crop yieldip + β2crop growth cycleip

γ0jXipj + γ1YSTip +∑j

γ2jYij + δc∆i + εi ,

LTOic ≡ Long-Term Orientation of individual i in country c

Xipj ≡ Geographical controls in parent’s country of origin

YSTip ≡ Years since transition to agriculture in parent’s country of origin

Yij ≡ Individual controls (age, sex, education, marital status, health status, religiosity)

∆i ≡ Host country fixed effects

Crop Yield and Long-Term Orientation in Second Generation Migrants

Either Parent Mother Father Both

(1) (2) (3) (4) (5) (6) (7) (8)

Crop Yield (Ancestors, pre-1500) 2.29*** 2.61*** 2.99*** 3.44*** 2.70** 3.34*** 5.63** 6.11**(0.80) (0.97) (1.10) (1.30) (1.04) (1.13) (2.43) (2.54)

Crop Yield Change (post-1500) 0.52 0.65 0.32 0.87 0.57 0.52 1.83 2.15(0.65) (0.61) (0.71) (0.77) (0.85) (0.89) (1.29) (1.76)

Crop Growth Cycle (Ancestors, pre-1500) -0.82 -1.17 -1.84 -2.07(1.00) (1.56) (1.32) (2.54)

Crop Growth Cycle Change (post-1500) -0.10 -0.92 0.48 -0.07(0.63) (0.68) (0.78) (1.33)

Country FE Yes Yes Yes Yes Yes Yes Yes YesIndividual Characteristics Yes Yes Yes Yes Yes Yes Yes YesAll Geographical Controls & Neolithic Yes Yes Yes Yes Yes Yes Yes Yes

Crop Yield and Saving in Second Generation Migrants

Saving

Either Parent Mother Father Both

(1) (2) (3) (4) (5) (6) (7) (8)

Crop Yield (Ancestors, pre-1500) 0.04** 0.06** 0.04* 0.06** 0.05** 0.07** 0.02 0.03(0.02) (0.03) (0.02) (0.03) (0.02) (0.03) (0.03) (0.03)

Crop Yield Change (post-1500) 0.03* 0.04** 0.04*** 0.04** 0.02 0.04** 0.08*** 0.07**(0.01) (0.02) (0.01) (0.02) (0.02) (0.02) (0.02) (0.03)

Crop Growth Cycle (Ancestors, pre-1500) -0.04 -0.03 -0.05 -0.03(0.03) (0.04) (0.04) (0.04)

Crop Growth Cycle Change (post-1500) -0.01 0.00 -0.02 0.02(0.02) (0.01) (0.02) (0.02)

Country FE Yes Yes Yes Yes Yes Yes Yes YesIndividual Characteristics Yes Yes Yes Yes Yes Yes Yes YesGeography & Neolithic Yes Yes Yes Yes Yes Yes Yes Yes

Crop Yield and Smoking in Second Generation Migrants

Smoking

Either Parent Both

Habit Ever Habit Ever

(1) (2) (3) (4) (5) (6) (7) (8)

Crop Yield (Ancestors, pre-1500) -0.02** -0.02*** -0.02** -0.03** -0.04*** -0.08*** -0.05*** -0.13***(0.01) (0.01) (0.01) (0.01) (0.02) (0.02) (0.02) (0.03)

Crop Yield Change (post-1500) -0.02** -0.00 -0.00 0.06 -0.01 -0.02(0.01) (0.01) (0.02) (0.04) (0.03) (0.03)

Crop Growth Cycle (Ancestors, pre-1500) 0.02 0.04** 0.02 0.10***(0.01) (0.02) (0.02) (0.03)

Crop Growth Cycle Change (post-1500) -0.00 0.00 -0.00 0.04*(0.02) (0.04) (0.03) (0.03)

Individual Controls Yes Yes Yes Yes Yes Yes Yes YesRegion FE No Yes Yes Yes Yes Yes Yes YesYear FE No Yes Yes Yes Yes Yes Yes YesGeographical Controls & Neolithic No No No Yes Yes Yes Yes Yes

Individual-Level WVS

Individual-Level Analysis (WVS) Data

Individual-level analysis:

Country FE(geography, institutions, culture)

Individual-Level WVS Empirical Analysis

LTOircw =β0 + β1crop yieldrc + β2crop growth cyclerc

γ0jXrc + γ1YSTrc +∑j

γ2jYircwj + δcw∆cw + εircw

LTOircw ≡ Long-Term Orientation of individual i in region r of country c in wave w

Xrc ≡ Geographical controls in region r of country c

YSTrc ≡ Years since transition to agriculture in region r of country c

Yircwj ≡ Individual controls (age, sex, education, income)

∆cw ≡ Continent/Country and Wave fixed effects

Crop Yield and Long-Term Orientation (WVS)

Long-Term Orientation (OLS)

(1) (2) (3) (4) (5) (6) (7) (8)

Crop Yield (pre-1500) 0.025*** 0.040*** 0.036*** 0.032*** 0.032*** 0.031*** 0.066***(0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.003)

Crop Yield Change (post-1500) 0.053*** 0.054*** 0.055***(0.002) (0.002) (0.003)

Crop Growth Cycle (pre-1500) -0.007** -0.018***(0.003) (0.003)

Crop Growth Cycle Change (post-1500) 0.025*** 0.026***(0.002) (0.002)

Crop Yield (Ancestors, pre-1500) 0.043***(0.002)

Crop Yield Change (Anc., post-1500) 0.041***(0.002)

Crop Growth Cycle (Ancestors, pre-1500) -0.005*(0.003)

Crop Growth Cycle Change (Anc., post-1500) 0.018***(0.002)

Wave & Continent FE Yes Yes Yes Yes Yes Yes Yes YesGeographical Controls & Neolithic No Yes Yes Yes Yes Yes Yes YesIndividual Characteristics No No No Yes Yes Yes Yes Yes

Robustness

Results are robust to:

Estimation method Probit

Cells that experienced change in crop post-1500 Table

Weighted Observations Table

Country Fixed Effects Table

Robustness

Regional Analysis

Share of Individuals in WVS Region with Long-Term Orientation

Unweighted Weighted: Area Weighted: Area Share Area Share

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)

Crop Yield 0.049*** 0.046*** 0.053*** 0.097*** 0.032** 0.031** 0.039*** 0.032**(0.012) (0.013) (0.017) (0.033) (0.012) (0.013) (0.015) (0.013)

Crop Growth Cycle -0.010 -0.047** -0.024** -0.036*** -0.027*** -0.036***(0.012) (0.021) (0.010) (0.009) (0.009) (0.008)

Crop Yield (Ancestors) 0.077*** 0.133*** 0.043** 0.041**(0.020) (0.032) (0.017) (0.017)

Crop Growth Cycle (Anc.) -0.012 -0.050*** -0.027*** -0.037***(0.013) (0.018) (0.009) (0.009)

Continental FE Yes Yes Yes Yes Yes Yes No No No No No NoCountry FE No No No No No No Yes Yes Yes Yes Yes YesGeographical Controls No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes YesOld World Sample No No No No No No No No No No Yes YesWeighted by Region Area No No No No Yes Yes Yes Yes No No Yes NoWeighted by Region’s Share No No No No No No No No Yes Yes No Yes

Adjusted-R2 0.22 0.25 0.25 0.28 0.28 0.37 0.72 0.72 0.86 0.86 0.72 0.86Observations 1356 1356 1356 1356 1356 1356 1356 1356 1356 1356 1143 1143

Conclusions

Crop Yield and the Adoption of Lengthy Production Processes:

Aceto Balsamico and Parmiggiano Reggiano

Conclusions

Concluding Remarks

Co-evolution of human traits and the economic environment iscentral for the understudying of comparative development

Variations in natural pre-industrial agricultural productivityacross regions

⇒ persistent effect on the distribution of time preference across

the globe

⇒ Education

⇒ Saving

⇒ Smoking

⇒ Technological Adoption

Conclusions

Concluding Remarks

the globe

⇒ Education

⇒ Saving

⇒ Smoking

Conclusions

Concluding Remarks

the globe

⇒ Education

⇒ Saving

⇒ Smoking

Conclusions

Concluding Remarks

the globe

⇒ Education

⇒ Saving

⇒ Smoking

Conclusions

Concluding Remarks

the globe

⇒ Education

⇒ Saving

⇒ Smoking

Conclusions

Concluding Remarks

the globe

⇒ Education

⇒ Saving

⇒ Smoking

Conclusions

Concluding Remarks

the globe

⇒ Education

⇒ Saving

⇒ Smoking

Conclusions

The Agricultural Origins of Time Preference

Oded Galor and Omer Ozak

“Patience is bitter, but its fruit is sweet.”

– Aristotle

Additional Results Model

Malthusian Framework: Endowment Sector

Production function

Y Et = At(L

Et )(1−α)Xα, α ∈ (0, 1)

Output endowment sector ≡ Y Et

Technological level ≡ At

Labor in investment mode ≡ LEtFixed amount of land ≡ X = 1

Boserupian technological progress

At ≡ A(LEt ) = R0(LEt )α

Per capita output

R0(LEt )α(LEt )(1−α)Xα

LEt= R0 Back

Production function

Y Et = At(L

Et )(1−α)Xα, α ∈ (0, 1)

Per capita output

LEt= R0 Back

Production function

Y Et = At(L

Et )(1−α)Xα, α ∈ (0, 1)

Per capita output

LEt= R0 Back

Production function

Y Et = At(L

Et )(1−α)Xα, α ∈ (0, 1)

Per capita output

LEt= R0 Back

Production function

Y Et = At(L

Et )(1−α)Xα, α ∈ (0, 1)

Labor in investment mode ≡ LEt

Fixed amount of land ≡ X = 1

Per capita output

LEt= R0 Back

Production function

Y Et = At(L

Et )(1−α)Xα, α ∈ (0, 1)

Per capita output

LEt= R0 Back

Production function

Y Et = At(L

Et )(1−α)Xα, α ∈ (0, 1)

Per capita output

LEt= R0 Back

Production function

Y Et = At(L

Et )(1−α)Xα, α ∈ (0, 1)

Per capita output

LEt= R0 Back

Malthusian Framework: Investment Sector

Production function

Y It = At(LIt )(1−α)Xα, α ∈ (0, 1)

Output investment sector ≡ Y ItTechnology level ≡ At

Labor in investment mode ≡ LItFixed amount of land ≡ X = 1

At ≡ A(LIt ) = R1(LIt )α

Per capita output

Y ItLIt

=R1(LIt )α(LIt )(1−α)Xα

LIt= R1 Back

Production function

Y It = At(LIt )(1−α)Xα, α ∈ (0, 1)

Per capita output

Y ItLIt

LIt= R1 Back

Production function

Y It = At(LIt )(1−α)Xα, α ∈ (0, 1)

Output investment sector ≡ Y It

Technology level ≡ At

Per capita output

Y ItLIt

LIt= R1 Back

Production function

Y It = At(LIt )(1−α)Xα, α ∈ (0, 1)

Per capita output

Y ItLIt

LIt= R1 Back

Production function

Y It = At(LIt )(1−α)Xα, α ∈ (0, 1)

Labor in investment mode ≡ LIt

Fixed amount of land ≡ X = 1

Per capita output

Y ItLIt

LIt= R1 Back

Production function

Y It = At(LIt )(1−α)Xα, α ∈ (0, 1)

Per capita output

Y ItLIt

LIt= R1 Back

Production function

Y It = At(LIt )(1−α)Xα, α ∈ (0, 1)

Per capita output

Y ItLIt

LIt= R1 Back

Production function

Y It = At(LIt )(1−α)Xα, α ∈ (0, 1)

Per capita output

Y ItLIt

LIt= R1 Back

Additional Results Post-1500CE Data

Potential Crop Yield post-1500CE Back

Potential Crop Growth Cycle post-1500CE Pre-1500CE

Potential Crop Return post-1500CE Pre-1500CE

Additional Results Crop Choice Robustness

Total vs. Daily Yield

Additional Results Ancestry Adjusted Maps

Potential Crop Yield (Ancestry Adjusted) and LTO

(a) Potential Crop Yield (b) Long-Term Orientation

Additional Results Ancestry Adjusted Maps

Potential Crop Yield (Ancestry Adjusted) and LTOUnadjusted

(a) Potential Crop Yield (b) Long-Term Orientation

Continental Distribution of crops (and their variants) pre-1500CE Back

Crop Continent Crop Continent

Alfalfa Asia, Europe Palm Heart North Africa, SubsaharaBanana Asia, Oceania, North Africa Pearl Millet Asia, North Africa, SubsaharaBarley Asia, Europe, North Africa Phaseolus Bean AmericaBuckwheat Asia Pigeon Pea Asia, SubsaharaCabbage Europe Rye EuropeCacao America Sorghum North Africa, SubsaharaCarrot Asia, Europe Soybean AsiaCassava America Sunflower AmericaChick Pea Europe Sweet Potato AmericaCitrus Asia, Europe Tea AsiaCoconut America, Oceania Tomato AmericaCoffee North Africa Wetland Rice Asia, SubsaharaCotton America, Asia, Europe, North Africa, Subsahara Wheat Asia, Europe, North AfricaCowpea Asia, North Africa, Subsahara Wheat Hard Red Spring Asia, Europe, North AfricaDry Pea Europe, North Africa Wheat Hard Red Winter Asia, Europe, North AfricaFlax Asia, Europe, North Africa Wheat Hard White Asia, Europe, North AfricaFoxtail Millet Asia, Europe, North Africa Wheat Soft Red Winter Asia, Europe, North AfricaGreengram Asia, Subsahara Wheat Soft White Asia, Europe, North AfricaGroundnuts America White Potato AmericaIndigo Rice Asia, Subsahara Yams Asia, SubsaharaMaize America Giant Yams Asia, SubsaharaOat Europe, North Africa Sorghum (Subtropical) North Africa, SubsaharaOilpalm North Africa, Subsahara Sorghum (Tropical Highland) North Africa, SubsaharaOlive Europe, North Africa Sorghum (Tropical Lowland) North Africa, SubsaharaOnion America, Asia, Europe, North Africa, Subsahara,

OceaniaWhite Yams North Africa, Subsahara

Changes in Crop Yield and Growth Cycle and their Correlates (Anc.) Back

Change Yield (Anc.) Change Growth Cycle (Anc.)

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

Crop Yield (Ancestors, pre-1500) -0.05 -0.29** -0.17 -0.26* -0.28* 0.16 0.26* 0.36* 0.35 0.31(0.13) (0.14) (0.14) (0.15) (0.15) (0.12) (0.16) (0.18) (0.25) (0.28)

Crop Growth Cycle (Ancestors, pre-1500) 0.77*** 0.56*** 0.59*** 0.69*** -0.32 -0.36 -0.54** -0.40(0.19) (0.19) (0.20) (0.24) (0.24) (0.26) (0.26) (0.36)

Absolute Latitude -0.74* -0.74* -1.00* -1.24**(0.40) (0.42) (0.52) (0.62)

Mean Elevation 0.03 -0.22 0.43 0.36(0.24) (0.25) (0.30) (0.30)

Terrain Roughness -0.15 -0.02 -0.19 -0.20(0.15) (0.16) (0.15) (0.14)

Distance to Coast or River 0.04 -0.05 -0.01 -0.04(0.12) (0.10) (0.14) (0.15)

Landlocked 0.08 -0.01 -0.24* -0.21(0.10) (0.08) (0.14) (0.16)

Island -0.11 -0.12 0.17 0.14(0.13) (0.15) (0.11) (0.14)

Pct. Land in Tropics -0.80* -0.59* -0.77 -0.69(0.46) (0.35) (0.52) (0.48)

Pct. Land in Temperate Zone 0.19 0.12 0.38 0.48(0.17) (0.16) (0.31) (0.31)

Pct. Land in Tropics and Subtropics 0.78 0.48 0.25 0.27(0.52) (0.47) (0.49) (0.51)

Precipitation -0.04 -0.07 0.17 0.05(0.17) (0.20) (0.17) (0.25)

Temperature -0.29 -0.67 -0.31 -0.59(0.36) (0.41) (0.47) (0.61)

Continental FE No No Yes No Yes No No Yes No Yes

Adjusted-R2 -0.01 0.24 0.52 0.40 0.53 0.01 0.05 0.03 0.21 0.18Observations 87 87 87 87 87 87 87 87 87 87

Potential Crop Yield, Growth Cycle, Agricultural Suitability and LTO Back

Principal Components

Component 1 Component 2 Unexplained

Crop Yield (Ancestors, pre-1500) 0.71 0.71 0.00Crop Growth Cycle (Ancestors, pre-1500) 0.71 -0.71 0.00

Eigenvalues 1.40 0.60Proportion Variance 0.70 0.30

Observations 87

Principal Components

Component 1 Component 2 Unexplained

Crop Yield Change (post-1500) 0.71 0.71 0.00Crop Growth Cycle Change (post-1500) 0.71 -0.71 0.00

Eigenvalues 1.12 0.88Proportion Variance 0.56 0.44

Observations 87

Additional Results Principal Component Analysis

Potential Crop Yield, Growth Cycle, Agricultural Suitability and LTO Back

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

PC2 Pre-1500 Crop 17.38*** 17.75*** 18.53*** 12.52*** 13.37*** 11.79*** 10.90*** 10.71***(2.69) (2.70) (3.10) (2.35) (3.27) (3.22) (3.21) (3.34)

PC2 Crop Change 0.55 0.77 8.82*** 8.74*** 8.22*** 7.93*** 6.39**(2.66) (2.88) (2.20) (2.46) (2.34) (2.35) (2.75)

PC1 Pre-1500 Crop 1.25 1.10 0.74 0.75 3.08* 4.02** 2.72 3.11(2.05) (2.05) (1.57) (1.57) (1.69) (1.89) (2.80) (2.85)

PC1 Crop Change 1.30 3.28 8.04*** 7.22*** 6.95*** 6.29*** 4.86(3.04) (2.49) (2.24) (2.40) (2.12) (2.26) (3.01)

Neolithic Transition Timing (Anc.) -6.46** -7.05** -9.88**(3.02) (3.17) (4.06)

Land Suitability (Anc.) 2.34 4.28(3.20) (3.50)

Continent FE No No No No No Yes Yes Yes Yes YesGeographical Controls No No No No No No Yes Yes Yes YesOld World Sample No No No No No No No No No YesAdjusted-R2 0.33 -0.01 0.32 -0.02 0.33 0.62 0.66 0.68 0.68 0.63Observations 85 85 85 85 85 85 85 85 85 70

Pre-1500CE Crop Yield and LTO

Grids that Experienced Change in Crop post-1500 Back

(1) (2) (3) (4) (5) (6) (7) (8)

Crop Yield (pre-1500) 4.97** 8.52*** 7.40*** 6.65** 7.75*** 7.97**(2.28) (2.46) (2.58) (2.98) (2.81) (3.66)

Crop Yield Change (post-1500) 4.36* 5.81** 5.58* 7.59**(2.46) (2.55) (2.83) (2.93)

Crop Growth Cycle (pre-1500) 0.06 -1.55(2.58) (3.97)

Crop Growth Cycle Change (post-1500) -4.50** -4.87**(2.18) (2.36)

Crop Yield (Ancestors, pre-1500) 8.21*** 7.85**(2.34) (3.26)

Crop Yield Change (Ancestors, post-1500) 6.09*** 7.31***(2.13) (2.25)

Crop Growth Cycle (Ancestors, pre-1500) -0.95(3.16)

Crop Growth Cycle Ch. (Anc., post-1500) -3.44(2.27)

Continent FE Yes Yes Yes Yes Yes Yes Yes YesAll Geographical Controls & Neolithic No Yes Yes Yes Yes Yes Yes YesOld World Sample No No No No No No Yes Yes

Pre-1500CE Crop Yield and LTO

Natural Experiment in Countries with High Share of Natives Back

Old World

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

Crop Yield (pre-1500) 8.49** 8.58*** 13.78*** 17.55***(3.44) (3.05) (3.47) (3.93)

Crop Growth Cycle (pre-1500) -8.86*(5.01)

Crop Growth Cycle Change (post-1500) 1.03(2.19)

Neolithic Transition Timing -2.84 -1.17(4.47) (4.38)

Continent FE Yes Yes Yes YesGeography No No Yes Yes

Adjusted-R2 0.43 0.52 0.58 0.60Observations 46 46 46 46

Additional Results Robustness Hofstede

Excluding Other Cultural Channels: CorrelationsBack

Correlation Among Cultural Indices

(LTO) (RVI) (Trust) (Ind) (PDI) (Coop) (UAI)

Long-Term Orientation (LTO) 1.00Restraint vs. Indulgence (RIV) 0.53*** 1.00Trust 0.19 -0.07 1.00Individualism (Ind) 0.12 -0.18 0.45*** 1.00Power Distance (PDI) 0.05 0.34** -0.50*** -0.66*** 1.00Cooperation 0.01 -0.09 -0.21 0.05 0.16 1.00Uncertainty Avoidance (UAI) -0.04 0.07 -0.50*** -0.23 0.27* -0.00 1.00

Potential Crop Yield, Growth Cycle, and LTO (Including Grids Not-Suitable for Production) Back

(1) (2) (3) (4) (5) (6) (7) (8)

Crop Yield 5.26** 9.01*** 8.21*** 7.11** 11.59*** 10.79***(2.43) (2.86) (2.61) (3.06) (2.84) (3.51)

Crop Growth Cycle 2.18 1.47(4.00) (4.25)

Crop Growth Cycle (Ancestors) 1.52(4.23)

Absolute Latitude 3.56 2.46 3.01 3.66 4.05 4.98 5.37(4.21) (3.94) (4.35) (3.79) (4.16) (4.62) (5.14)

Mean Elevation 6.20* 7.14** 6.63* 6.73** 6.44* 5.86 5.64(3.26) (3.41) (3.44) (3.35) (3.25) (3.92) (3.84)

Terrain Roughness -6.76** -6.16** -6.09** -7.29** -7.24** -6.55** -6.59**(2.68) (2.95) (2.98) (3.00) (3.00) (3.25) (3.28)

Neolithic Transition Timing -6.81** -7.21** -5.58* -5.84*(3.05) (3.20) (2.84) (2.94)

Neolithic Transition Timing (Ancestors) -5.20** -5.41**(2.53) (2.63)

Continent FE Yes Yes Yes Yes Yes Yes Yes YesAdditional Geographical Controls No Yes Yes Yes Yes Yes Yes YesOld World Sample No No No No No No Yes YesAdjusted-R2 0.50 0.57 0.60 0.59 0.60 0.60 0.56 0.56Observations 87 87 87 87 87 87 72 72

Potential Daily Crop Return, Crop Growth Cycle, and LTO Back

(1) (2) (3) (4) (5) (6) (7) (8)

Daily Crop Return 5.71** 9.40*** 8.39*** 7.00*** 10.83*** 9.28***(2.39) (2.57) (2.44) (2.59) (2.69) (2.82)

Crop Growth Cycle 4.04 4.57(3.58) (3.85)

Daily Crop Return (Ancestors) 9.00*** 7.57***(2.41) (2.63)

Crop Growth Cycle (Ancestors) 4.23(3.79)

Mean elevation 6.44* 7.19** 6.39* 6.78* 6.07* 5.98 5.32(3.38) (3.47) (3.42) (3.42) (3.26) (4.11) (3.84)

Terrain Roughness -6.66** -6.09** -6.10** -7.05** -7.08** -6.15* -6.46*(2.67) (2.94) (2.95) (3.01) (3.01) (3.31) (3.26)

Neolithic Transition Timing -6.13* -6.83** -5.14* -5.78*(3.11) (3.18) (2.93) (2.94)

Neolithic Transition Timing (Ancestors) -4.87* -5.41**(2.62) (2.66)

Continent FE Yes Yes Yes Yes Yes Yes Yes YesAdditional Geographical Controls No Yes Yes Yes Yes Yes Yes YesOld World Sample No No No No No No Yes YesAdjusted-R2 0.51 0.58 0.59 0.60 0.59 0.60 0.55 0.56Observations 87 87 87 87 87 87 72 72

Excluding Trade Channel Back

Suitability Money Transportation Routes

(1) (2) (3) (4) (5) (6) (7) (8) (9)

Crop Yield (Ancestors, pre-1500) 9.00*** 9.84*** 11.48*** 12.03*** 11.27*** 11.61*** 12.37*** 11.17*** 11.73***(2.85) (2.45) (2.73) (3.33) (2.61) (2.67) (3.35) (2.66) (2.76)

Crop Yield Change (post-1500) 10.03*** 10.84*** 11.08*** 11.48*** 11.11*** 10.98*** 11.32*** 11.13*** 11.81***(2.97) (2.72) (3.16) (3.42) (3.09) (3.16) (3.17) (3.14) (3.42)

Crop Growth Cycle (Ancestors, pre-1500) -5.35 -7.71* -8.36* -8.96* -8.79** -8.33* -9.28** -8.56* -9.73**(4.23) (4.29) (4.28) (4.66) (4.38) (4.30) (4.61) (4.42) (4.51)

Crop Growth Cycle Change (post-1500) -0.12 0.27 -0.07 -0.02 -0.10 0.02 0.10 -0.34 0.02(1.70) (1.52) (1.82) (1.79) (1.76) (1.85) (1.77) (1.75) (1.83)

Land Suitability (Gini) -2.11(2.02)

Land Suitability (Range) 2.46(1.65)

Exchange Medium 1000BCE 0.05(2.43)

Exchange Medium 1CE 1.15(3.12)

Exchange Medium 1000CE 4.60(4.32)

Transportation Medium 1000BCE 0.84(3.18)

Transportation Medium 1CE 2.40(4.36)

Transportation Medium 1000CE 1.50(4.39)

Pre-Industrial Distance to Trade Route 0.16(5.98)

Continental FE Yes Yes Yes Yes Yes Yes Yes Yes YesGeography & Neolithic Yes Yes Yes Yes Yes Yes Yes Yes YesAdjusted-R2 0.66 0.67 0.63 0.64 0.63 0.63 0.64 0.62 0.61Observations 84 84 81 81 81 81 81 81 71

Potential Crop Yield and Development Back

(1) (2) (3) (4) (5) (6) (7) (8)

Crop Yield 11.67*** 10.87*** 13.23*** 12.96***(3.80) (3.58) (3.95) (3.90)

Crop Growth Cycle -4.53 -4.73 -4.90 -4.61(4.20) (3.95) (4.00) (4.07)

Crop Yield (Ancestors) 15.52*** 14.42*** 16.39*** 16.31***(2.94) (3.02) (3.04) (3.06)

Crop Growth Cycle (Ancestors) -6.30* -6.27* -6.62* -6.33*(3.54) (3.41) (3.50) (3.49)

Age Dependency Ratio -6.51** -4.37(2.95) (2.84)

Life Expectancy at Birth 7.24* 5.77(4.32) (3.80)

Ln[GPD per capita] 3.67 3.04(3.00) (2.57)

Continental FE Yes Yes Yes Yes Yes Yes Yes YesGeographical Characteristics & Neolithic Yes Yes Yes Yes Yes Yes Yes Yes

Crop Yield, Climatic Risk, and LTO Back

Scale Risk

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

Crop Yield (Ancestors, pre-1500) 10.62*** 9.28*** 10.88*** 11.56*** 10.19*** 9.58*** 11.06*** 11.08*** 10.98*** 11.04***(2.62) (2.49) (2.68) (2.70) (2.97) (2.81) (2.58) (2.62) (2.58) (2.64)

Crop Yield Change (post-1500) 10.23*** 8.85*** 10.75*** 10.72*** 10.23*** 9.85*** 10.77*** 10.84*** 10.74*** 10.74***(2.95) (2.93) (2.92) (2.88) (3.00) (2.93) (2.92) (3.14) (2.92) (3.12)

Crop Growth Cycle (Ancestors, pre-1500) -7.45* -3.79 -8.14* -7.22* -6.31 -4.59 -8.07* -8.16* -8.02* -8.05*(4.30) (4.10) (4.18) (4.32) (4.83) (4.71) (4.09) (4.33) (4.11) (4.33)

Crop Growth Cycle Change (post-1500) -0.60 0.15 -0.47 -0.31 -0.12 0.19 -0.46 -0.48 -0.44 -0.45(1.68) (1.65) (1.73) (1.75) (1.87) (1.82) (1.75) (1.78) (1.74) (1.77)

Total land area 3.04(2.17)

Total land area (Ancestry Adjusted) 7.31***(2.08)

Precipitation Volatility (mean) 0.69(3.05)

Precipitation Volatility (mean) (Ancestry Adjusted) -2.26(3.02)

Temperature Volatility (mean) 4.37(6.44)

Temperature Volatility (mean) (Ancestry Adjusted) 6.70(5.07)

Precipitation Diversification (mean) -0.22(2.95)

Precipitation Diversification (mean) (Ancestry Adjusted) -0.28(2.85)

Temperature Diversification (mean) 0.78(3.05)

Temperature Diversification (mean) (Ancestry Adjusted) 0.05(2.97)

Partial R2

Crop Yield (Ancestors, pre-1500) 0.21*** 0.18*** 0.21*** 0.23*** 0.18*** 0.16*** 0.22*** 0.22*** 0.22*** 0.22***Crop Yield Change (post-1500) 0.15*** 0.13*** 0.16*** 0.16*** 0.15*** 0.14*** 0.16*** 0.16*** 0.16*** 0.16***Crop Growth Cycle (Ancestors, pre-1500) 0.05* 0.01 0.06* 0.05* 0.03 0.02 0.06* 0.06* 0.06* 0.06*Crop Growth Cycle Change (post-1500) 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00Total land area 0.02Total land area (Ancestry Adjusted) 0.14***Precipitation Volatility (mean) 0.00Precipitation Volatility (mean) (Ancestry Adjusted) 0.01Temperature Volatility (mean) 0.01Temperature Volatility (mean) (Ancestry Adjusted) 0.03Precipitation Diversification (mean) 0.00Precipitation Diversification (mean) (Ancestry Adjusted) 0.00Temperature Diversification (mean) 0.00Temperature Diversification (mean) (Ancestry Adjusted) 0.00

Continental FE Yes Yes Yes Yes Yes Yes Yes Yes Yes YesGeography & Neolithic Yes Yes Yes Yes Yes Yes Yes Yes Yes YesAdjusted-R2 0.65 0.70 0.65 0.65 0.65 0.66 0.65 0.65 0.65 0.65Observations 87 87 87 87 87 87 87 87 87 87

Crop Yield and LTO Back

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

Crop Yield 9.67*** 10.14*** 13.58*** 16.57***(2.60) (3.02) (3.01) (3.37)[3.03] [3.38] [3.01] [2.57]{2.46} {2.65} {2.88} {2.95}

Crop Growth Cycle -3.78 -2.92 -5.26** -4.07(2.47) (2.95) (2.61) (2.90)[2.39] [2.67] [2.38] [2.45]{2.34} {2.59} {2.50} {2.54}

Crop Yield (Ancestors) 11.35*** 14.50***(2.56) (2.75)[2.60] [2.46]{2.43} {2.41}

Crop Growth Cycle (Ancestors) -5.05** -4.65*(2.41) (2.59)[2.15] [2.24]{2.28} {2.27}

Continent FE Yes Yes Yes Yes Yes YesAll Geography & Neolithic No Yes No Yes No YesOld World Subsample No No No No Yes YesAET -21.58 -3.00 -5.53δ -4.72 -0.35 -0.66β∗ 11.38 22.02 21.67

R2 0.59 0.70 0.61 0.75 0.56 0.72

Changes in Crop Yield & Growth Cycle and LTO (Selection on Unobserv.) Back

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

Crop Yield Change (post-1500) 11.28*** 9.51***(2.92) (2.92)

Crop Growth Cycle Change (post-1500) -0.67 -1.51(1.84) (1.81)

Crop Yield Change (Anc., post-1500) 10.20*** 8.83*** 11.25*** 8.39***(2.50) (2.36) (2.72) (2.88)

Crop Growth Cycle Change (Anc., post-1500) 0.79 -0.73 0.16 -1.45(1.75) (1.78) (1.87) (1.93)

Crop Yield (Ancestors, pre-1500) 10.03*** 10.74*** 9.90*** 11.31*** 10.46*** 12.18***(2.31) (2.76) (2.30) (2.70) (2.43) (3.05)

Crop Growth Cycle (Ancestors, pre-1500) -11.29*** -6.47 -11.59*** -6.85* -12.27*** -5.69(3.22) (3.90) (3.23) (3.65) (3.38) (4.24)

Change Crop Yield

AET 5.38 6.43 2.93δ 2.13 2.51 1.45β∗ 6.21 6.25 3.32

Change Crop Growth Cycle

AET -1.81 -0.48 -0.90δ -0.94 -0.25 -0.49β∗ -3.06 -3.58 -4.29

Continent FE Yes Yes Yes Yes Yes YesAll Geography & Neolithic No Yes No Yes No YesOld World Subsample No No No No Yes Yes

R2 0.65 0.77 0.67 0.78 0.62 0.76

Additional Results Long-Term Orientation Measure

European Social Survey Back

Third Wave European Social Survey

Academically driven cross-national survey that has beenconducted every two years across Europe since 2001Survey measures the attitudes, beliefs and behavior patterns ofdiverse populations in 25 nations

“Do you generally plan for your future or do you just take eachday as it comes?”

Correlations: Long-Term Orientation and Income

Total Household Income

Second Generation Migrants All Individuals

(1) (2) (3) (4) (5) (6) (7) (8)

Long-Term Orientation 0.33** 0.22* 0.22** 0.23** 0.35*** 0.45*** 0.36*** 0.32***(0.14) (0.12) (0.10) (0.11) (0.08) (0.04) (0.04) (0.04)

Country FE No Yes Yes Yes No Yes Yes YesSex & Age No No Yes Yes No No Yes YesPray & Health No No No Yes No No No Yes

Adjusted-R2 0.01 0.40 0.40 0.41 0.01 0.50 0.52 0.53

R2 0.01 0.43 0.43 0.47 0.01 0.50 0.52 0.53Observations 383 383 383 383 29323 29323 29323 29323

Education

Crop Yield, Crop Growth Cycle, and LTO in Second Generation

Migrants Graphs

Long-Term Orientation (Ordered Probit)

Country of OriginMother Parents

(1) (2) (3) (4) (5) (6) (7) (8)

Crop Yield 0.11*** 0.11*** 0.23*** 0.27*** 0.23*** 0.31***(0.04) (0.04) (0.07) (0.07) (0.09) (0.11)

Crop Growth Cycle -0.13* -0.09 -0.10(0.07) (0.07) (0.09)

Crop Growth Cycle (Ancestors) -0.14* -0.10(0.07) (0.08)

Country FE Yes Yes Yes Yes Yes Yes Yes YesSex & Age Yes Yes Yes Yes Yes Yes Yes YesOther Ind. Chars. No Yes Yes Yes Yes Yes Yes YesGeographical & Neolithic No No Yes Yes Yes Yes Yes YesOld World Sample No No No No No No No Yes

Pseudo-R2 0.01 0.02 0.03 0.03 0.03 0.03 0.03 0.03Observations 705 705 705 705 705 566 566 557

Pre-1500 Crop Yield and LTO in Second Generation Migrants Back

(1) (2) (3) (4) (5) (6) (7) (8) (9)

Crop Yield (pre-1500) 2.96** 3.40** 6.45*** 6.50*** 6.65*** 5.08** 7.62**(1.18) (1.32) (2.17) (2.16) (2.15) (2.48) (2.92)

Crop Yield Change (post-1500) 0.44 1.37 1.98 2.29(1.20) (1.40) (1.63) (1.65)

Crop Growth Cycle (pre-1500) -1.60 -2.65 -2.36(2.58) (2.37) (2.53)

Crop Growth Cycle Change (post-1500) -1.27 -0.07 -0.24(0.92) (1.19) (1.29)

Crop Yield Change (Anc., post-1500) 1.00 1.87(1.45) (1.66)

Crop Growth Cycle (Ancestors, pre-1500) -2.42 -3.16(2.53) (2.67)

Crop Growth Cycle Ch. (Anc., post-1500) -1.03 0.13(0.92) (1.17)

Country FE Yes Yes Yes Yes Yes Yes Yes Yes YesSex & Age Yes Yes Yes Yes Yes Yes Yes Yes YesOther Ind. Chars. No Yes Yes Yes Yes Yes Yes Yes YesGeographical Controls & Neolitic No No Yes Yes Yes Yes Yes Yes YesOld World Sample No No No No No No No No Yes

R2 0.06 0.11 0.12 0.12 0.12 0.12 0.15 0.15 0.15Observations 705 705 705 705 705 705 566 566 557

Pre-1500 Crop Yield and LTO in 2nd Gen. Migrants, Grids that Changed Crop post-1500 Back

(1) (2) (3) (4) (5) (6) (7) (8) (9)

Crop Yield (pre-1500) 3.71*** 3.81*** 6.16*** 6.09*** 6.44*** 4.97** 4.85*(1.19) (1.30) (1.59) (1.63) (1.67) (2.42) (2.46)

Crop Yield Change (post-1500) 0.42 -0.25 0.39 0.94(1.58) (1.52) (1.45) (1.47)

Crop Growth Cycle (pre-1500) 0.14 -0.07 0.79(1.88) (2.28) (2.30)

Crop Growth Cycle Change (post-1500) 1.18 2.06 1.01(1.62) (1.63) (1.37)

Crop Yield Change (Ancestors, post-1500) -0.86 0.41(1.49) (1.47)

Crop Growth Cycle (Ancestors, pre-1500) 0.28 0.22(1.86) (2.30)

Crop Growth Cycle Ch. (Anc., post-1500) 1.88 2.24(1.59) (1.62)

Country FE Yes Yes Yes Yes Yes Yes Yes Yes YesSex & Age Yes Yes Yes Yes Yes Yes Yes Yes YesOther Ind. Chars. No Yes Yes Yes Yes Yes Yes Yes YesGeographical Controls & Neolithic No No Yes Yes Yes Yes Yes Yes YesOld World Sample No No No No No No No No Yes

R2 0.06 0.11 0.12 0.12 0.12 0.12 0.15 0.15 0.15Observations 705 705 705 705 705 705 566 566 557

Crop Yield and Long-Term Orientation in Second Generation Migrants Back

Long-Term Orientation (weighted OLS)

All crops All cells Changing cells/crops

(Survey) (Nc ) (N) (Nm) (Survey) (Nc ) (N) (Nm) (Survey) (Nc ) (N) (Nm)

Yield (Ancestors) 7.10*** 15.24***12.16***9.29***(2.48) (3.25) (2.83) (3.42)

Growth Cycle (Anc.) -4.72* 1.46 0.05 4.58(2.43) (3.78) (3.25) (4.43)

Yield (Anc., pre) 7.03*** 15.24***12.29***11.88***(2.39) (2.54) (2.21) (2.86)

Yield Change (post) 0.87 0.50 0.33 -1.75(1.55) (2.61) (2.20) (1.94)

Growth Cycle (Anc., pre) -3.28 2.98 1.61 4.23(2.77) (4.25) (3.90) (4.93)

Growth Cycle Ch. (post) -1.70* 1.11 -0.04 1.34(0.98) (1.69) (1.41) (1.39)

Yield (Anc., pre) 6.38*** 9.39***8.18***8.25***(1.97) (2.68) (2.25) (2.24)

Yield Change (post) -1.46 0.92 0.38 -0.73(1.66) (2.74) (2.43) (2.27)

Growth Cycle (Anc., pre) -0.96 1.26 1.32 -0.45(2.27) (2.49) (2.31) (2.45)

Growth Cycle Ch. (post) 2.49 0.78 -0.70 -2.60(1.59) (1.97) (1.95) (1.95)

Country FE Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes YesAll Controls Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Adjusted-R2 0.05 0.20 0.23 0.27 0.05 0.21 0.24 0.28 0.05 0.17 0.22 0.27

R2 0.13 0.26 0.29 0.32 0.13 0.27 0.30 0.34 0.13 0.24 0.28 0.33Observations 705 705 705 705 705 705 705 705 705 705 705 705

Additional Results WVS

World Values Survey Back

All waves of WVS

cross-national survey conducted every 4-5 years96 countrieswidely used in social research

Long-Term Orientation measure based preference for thrift inchildren

Question Back

“Here is a list of qualities that children can be encouragedto learn at home. Which, if any, do you consider to be

especially important?”

Individual has LTO if mentioned “thrift, saving money and things”

Question Back

“Here is a list of qualities that children can be encouragedto learn at home. Which, if any, do you consider to be

especially important?”

Individual has LTO if mentioned “thrift, saving money and things”

Crop Yield and LTO (WVS) Back

Long-Term Orientation (Probit)

(1) (2) (3) (4) (5) (6) (7) (8)

Crop Yield (pre-1500) 0.010*** 0.016*** 0.014*** 0.013*** 0.013*** 0.012*** 0.026***

(0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001)

Crop Yield Change (post-1500) 0.034*** 0.035*** 0.035***

(0.001) (0.002) (0.002)

Crop Growth Cycle (pre-1500) -0.000*** -0.001***

(0.000) (0.000)

Crop Growth Cycle Change (post-1500) 0.002*** 0.003***

(0.000) (0.000)

Crop Yield (Anc., pre-1500) 0.022***

(0.001)

Crop Yield Change (Anc., post-1500) 0.030***

(0.002)

Crop Growth Cycle (Anc., pre-1500) -0.000*

(0.000)

Crop Growth Cycle Ch. (Anc., post-1500) 0.002***

(0.000)

Wave & Continent FE Yes Yes Yes Yes Yes Yes Yes Yes

Geography & Neolithic No Yes Yes Yes Yes Yes Yes Yes

Individual Chars No No No Yes Yes Yes Yes Yes

Pseudo-R2 0.01 0.02 0.02 0.03 0.03 0.03 0.03 0.04

Observations 217953 217953 217953 217953 217953 217953 217953 176489

Pre-1500 Crop Yield and LTO (WVS) Back

(1) (2) (3) (4) (5) (6) (7) (8)

Crop Yield (pre-1500) 0.025*** 0.040*** 0.036*** 0.032*** 0.032*** 0.031*** 0.066***

(0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.003)

Crop Yield Change (post-1500) 0.053*** 0.054*** 0.055***

(0.002) (0.002) (0.003)

Crop Growth Cycle (pre-1500) -0.007** -0.018***

(0.003) (0.003)

Crop Growth Cycle Change (post-1500) 0.025*** 0.026***

(0.002) (0.002)

Crop Yield (Ancestors, pre-1500) 0.043***

(0.002)

Crop Yield Change (Anc., post-1500) 0.041***

(0.002)

Crop Growth Cycle (Ancestors, pre-1500) -0.005*

(0.003)

Crop Growth Cycle Change (Anc., post-1500) 0.018***

(0.002)

Wave FE Yes Yes Yes Yes Yes Yes Yes Yes

Continent FE Yes Yes Yes Yes Yes Yes Yes Yes

Geographical Controls & Neolithic No Yes Yes Yes Yes Yes Yes Yes

Individual Characteristics No No Yes Yes Yes Yes Yes Yes

Old World Subsample No No No No No No No Yes

Adjusted-R2 0.02 0.02 0.02 0.04 0.04 0.04 0.05 0.05

Observations 217953 217953 217953 217953 217953 217953 217953 176489

Pre-1500 Crop Yield and LTO (WVS), Grids that Experienced Change in Crop post-1500 Back

(1) (2) (3) (4) (5) (6) (7) (8)

Crop Yield (pre-1500) 0.039*** 0.053*** 0.052*** 0.049*** 0.041*** 0.034*** 0.034***(0.001) (0.002) (0.002) (0.001) (0.002) (0.002) (0.002)

Crop Growth Cycle (pre-1500) 0.013*** 0.013***(0.002) (0.003)

Crop Growth Cycle Change (post-1500) -0.008*** -0.011***(0.001) (0.001)

Crop Yield (Ancestors, pre-1500) 0.029***(0.002)

Crop Yield Change (Anc., post-1500) 0.028***(0.002)

Crop Growth Cycle (Ancestors, pre-1500) 0.014***(0.002)

Crop Growth Cycle Change (Anc., post-1500) -0.012***(0.001)

Wave FE Yes Yes Yes Yes Yes Yes Yes YesContinent FE Yes Yes Yes Yes Yes Yes Yes YesGeographical Controls & Neolithic No Yes Yes Yes Yes Yes Yes YesIndividual Characteristics No No Yes Yes Yes Yes Yes YesOld World Subsample No No No No No No No YesAdjusted-R2 0.02 0.03 0.03 0.04 0.04 0.04 0.04 0.05Observations 217953 217953 217953 217953 217953 217953 217953 176489

Crop Yield and Long-Term Orientation (Weighted) Back

Long-Term Orientation (Weighted OLS)

All crops All cells Changing cells/crops

(No) (Survey) (Same N) (Pop) (No) (Survey) (Same N) (Pop) (No) (Survey) (Same N) (Pop)

Crop Yield (Ancestors) 0.048*** 0.047*** 0.056*** 0.015**(0.003) (0.003) (0.003) (0.006)

Crop Growth Cycle (Ancestors) 0.017*** 0.018*** 0.010*** 0.046***(0.003) (0.003) (0.003) (0.006)

Crop Yield (Anc., pre-1500) 0.046*** 0.044*** 0.048*** 0.021***(0.002) (0.002) (0.002) (0.004)

Crop Growth Cycle (Anc., pre-1500) -0.012*** -0.010*** -0.019*** 0.006(0.003) (0.003) (0.003) (0.005)

Crop Yield Ch. (post-1500) 0.052*** 0.051*** 0.062*** 0.038***(0.003) (0.003) (0.003) (0.004)

Crop Growth Cycle Ch. (post-1500) 0.021*** 0.020*** 0.014*** 0.033***(0.002) (0.002) (0.002) (0.003)

Crop Yield (Anc., pre-1500) 0.033*** 0.032*** 0.028*** 0.033***(0.002) (0.002) (0.002) (0.004)

Crop Growth Cycle (Anc., pre-1500) 0.010*** 0.016*** 0.014*** -0.000(0.002) (0.002) (0.002) (0.003)

Crop Yield Ch. (post-1500) 0.032*** 0.031*** 0.041*** 0.026***(0.002) (0.002) (0.002) (0.003)

Crop Growth Cycle Ch. (post-1500) -0.006*** -0.005*** -0.007*** 0.007***(0.001) (0.001) (0.001) (0.003)

Wave FE Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes YesContinent FE Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes YesIndividual Chars Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes YesGeographical Controls & Neolithic Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes YesR2 0.04 0.05 0.05 0.07 0.05 0.05 0.05 0.07 0.04 0.05 0.05 0.07Adjusted-R2 0.04 0.05 0.05 0.07 0.05 0.05 0.05 0.07 0.04 0.05 0.05 0.07Observations 217953 217953 217953 217953 217953 217953 217953 217953 217953 217953 217953 217953

Crop Yield and Long-Term Orientation (WVS Regional Analysis) Back

(1) (2) (3) (4) (5) (6) (7) (8)

Crop Yield (pre-1500) 0.023*** 0.024*** 0.023*** 0.025*** 0.028*** 0.005* 0.055*** 0.005

(0.002) (0.002) (0.002) (0.002) (0.002) (0.003) (0.002) (0.004)

Crop Yield Change (post-1500) 0.043*** 0.046*** 0.006** 0.042*** 0.007**

(0.002) (0.002) (0.003) (0.002) (0.003)

Crop Growth Cycle (pre-1500) -0.011*** -0.009** -0.012*** -0.008

(0.003) (0.004) (0.003) (0.005)

Crop Growth Cycle Change (post-1500) 0.002 -0.007*** 0.002 -0.007***

(0.002) (0.002) (0.002) (0.003)

Wave & Continent FE Yes Yes Yes Yes Yes No Yes No

Individual Chars No No Yes Yes Yes Yes Yes Yes

Country FE No No No No No Yes No Yes

Adjusted-R2 0.02 0.02 0.04 0.04 0.04 0.08 0.05 0.08

Observations 185659 185659 185659 185659 185659 185659 151299 151299

human evolution and economic developmentmedia.virbcdn.com/...zeuthenlectures-lectureiii.pdf ·...

Documents