adaptation to land constraints: is africa different?

1

Adaptation to land constraints: Is Africa different?

Derek HeadeyInternational Food Policy Research Institute (IFPRI)

Thom JayneMichigan State University (MSU)

Outline1. About the project2. Introduction (background on existing theory &

evidence)3. Expanding land use (extensification)4. Intensifying agriculture5. Reducing fertility rates6. Diversifying out of agriculture7. Conclusions

This paper is part of a Bill & Melinda Gates Foundation project on emerging land issues in African agriculture

The motivation for the project was the observation of various puzzles of Africa agriculture: apparent land abundance in Africa, but much of Africa has major land constraints, and very, very small farms

In addition to five African case studies (Ethiopia included), we decided to look at the cross-country evidence on agricultural intensification

That is what I am presenting today

About the project

Some 215 years ago, Malthus argued that pop. growth cyclically outstrips agricultural productivity

Strong assumptions: high exogenous fertility rates, land constraints, zero ag. productivity growth

In much of the world, economic history has not been kind to Malthus, because of “induced innovations”

Whilst “induced innovation” is associated with Hayami and Ruttan, plenty of prior research looked at particular elements of induced innovation

More generally, “responding to incentives” is at the heart of economic theories

1. Introduction

Land expansionMalthus’ theory depends on land constraints, but

people have been adept at expanding the land frontier through colonialization, tech. and infrastructure

e.g. recent surge in global food prices has prompted “land grabs” in Africa & land expansion more generally

Agricultural research in Brazil led to massive land expansion in 1990s and 2000s (opening the cerrado)

Of course, for specific countries, land expansion may not be an option

1. Introduction – Land constraints

Agricultural intensificationBoserup (1964): as land to labor ratios shrink, people

intensify agricultural production – use more inputs per hectare to get more output per hectare

Boserup described transition from land-abundant technologies (slash-and- burn, long fallow) to land-scarce technologies (short fallow, adoption of plow, increased fertilizer use, irrigation)

She also emphasized increased labor inputs, and transition from communal to private property rights

Binswanger et al generalized the theory in 1980s

1. Introduction - intensification

1980s saw substantial empirical literatureBroadly supports Boserup’s theory, but lots of

complexity Binswanger emphasizes that land constraints interact

with access to markets, and agroecological factorsFor example, irrigation and high rainfall allow multiple

cropping – not possible in all agroecologies, howeverMarket access can be an driver of intensification, but

might also interact with land constraintsAnd institutions matter – e.g. literature in 1990s

unfavorably compared Ethiopia to Kenya


Policy-induced intensificationOne weakness of Boserup’s theory is that endogenous

intensification takes place over the long runBut Africa’s population has doubled in last 40 yearsHence, much of the ag-economics literature focuses on

policy-induced intensification - e.g. Green RevolutionOf course, many scientific successes in agricultureBut Binswanger emphasized that adoption of

technologies is typically a function of land-labor ratios, agroecology and market access (“Boserup matters!”)


Reducing fertility ratesMassive economic & demographic lit. on fertilityEconomics sees fertility as a choice variable If land is becoming constraint (and labor is not), then

farmers will have less children . . . all else equalBut children serve other purposes (consumption

goods, old age security), so fertility response to land constraints may be low

Moreover, demographic literature emphasizes “supply” constraints: family planning, female education, etc

Not obvious there is a strong endogenous mechanism


Diversifying out of agricultureMajor omission from 1980s literature was discussion

of nonfarm economy, which is large in many countries If land is a constraint, why not migrate?Of course, farmers do migrate, but viability of

migration in domestic economy is a general equilibrium issue: are there nonfarm jobs?

Rural nonfarm economy (RNFE) often felt to be driven by agric. productivity, infrastructure, education

Policies matter: RNFE does not spontaneously emerge


The African contextWhat about international migration?Has boomed in last 20 years: remittances to LDCs

grown by 1600% from 1990 to 2010.Moreover, not just small islands: Philippines, Pakistan

and Bangladesh hugely dependent on remittances, and they are all much larger than most African countries

Are land constraints driving rural people to explore international migration as a way out of farming?


So we have 4 adaptations to land constraints In this paper we focus on international evidence, and

on whether and how Africa adapts to land constraintsWhy be especially concerned about Malthus in SSA?Many reasons:1. Very poor, and poverty still heavily rural: history of

famine & drought; progress might be deceptive2. Rural poverty closely associated with small farms;

most Africa farms have a few hectares or less3. Low inherent agric. potential (incl. low irrigation)

1. Introduction

5. Rapid population growth (double by 2050); suggests that farm sizes will only get smaller

6. Climate change: secular changes in climate, but also likelihood of more shocks

7. Very limited success with industrialization; urban jobs mostly in low-wage informal services sector

1. Introduction

So we are going to explore how countries have adapted to farm constraints

Framework based on decomposing growth in farm income:

+Growth in rural population is the sum of fertility & net migration:

1. Introduction

h𝑆 𝑟𝑖𝑛𝑘𝑖𝑛𝑔 𝑓𝑎𝑟𝑚𝑠𝑖𝑧𝑒𝑠

Our overarching objective is to assess international experience in these 4 adaptations to land pressures

There is a large literature exploring Boserup’s hypothesis, as well as policy-induced intensification

There is much smaller literature on land expansionThere is essentially no literature on farm sizes &

fertility ratesAnd there is some indirect literature on farms sizes,

rural nonfarm activity and migrationFor each of these adaptations, we also ask whether

Africa is different, and why?

1. Introduction

In terms of data and methods, we make use of:1. FAOSTAT ag production and land data;2. Census (FAO) and survey data on farm size

distributions3. DHS data on rural fertility rates & occupations4. Some WB data on remittances We combine these data in an unusually rich data set

on agricultural and rural development (though we also acknowledge that some of the

numbers are fairly speculative)

1. Introduction

On methods, our approach is necessarily exploratoryEstablishing causation is an under-recognized

problem with Boserup’s theoryProblems of simultaneity, omitted variables, selection

biases, parameter heterogeneity. Some examples:1. Agroecological (AE) factors & market access jointly determine

settlement patterns and intensification2. Boserupian intensification depends on AE potential3. Unsuccessful intensification encourages out-migration4. Policies promote intensification, discourages out-migration IV rarely plausible in cross-country setting, but we do

make an effort to add as many controls as possible

If farm sizes are shrinking, why not expand land use? Africa is typically thought of as land abundant, but

this neglects the heterogeneity within Africa

2. Land expansion

Region Period

Hectares per agric. worker

(FAO)

Hectares per holding

(censuses)

Used land as % of potentially

cultivable land

Africa - high densityb

(n=5)1970s 0.84 1.99 32.72000s 0.58 1.23 43.8

Africa - low densityb

(n=11)1970s 1.65 2.65 17.22000s 1.37 2.82 24.7

South Asia 1970s 0.78 2.01 129.5(n=5) 2000s 0.55 1.19 135.9China & S.E. Asia 1970s 0.80 2.08 71.2(n=4) 2000s 0.68 1.58 83.0

Several important facts & mysteries emerge from census, FAO and FAO-IIASA data:

1. Farm sizes are shrinking in high-density Africa. 2. Some high-density countries still have unused land,

but smallholders face major constraints to using that land (e.g. Ethiopia, Madagascar).

3. Even in countries with unused land (e.g. Ethiopia), there are major constraints to using new lands: different agronomics, disease burdens, infrastructure

4. Farm sizes are unchanged (on average) in low density Africa, but still very small on average

2. Land expansion

3. Agricultural intensification In the framework above, the most welfare-relevant

indicator of intensification is just output per hectare Boserup focused more on cropping intensity, and the

ag-econ profession & CGIAR looks a lot at yields But switching to high value crops is obviously also a

potentially important adaptation, especially in SSA. So I’m going to show you a series of graphs, and then

some more formal econometric tests. Note that I also decompose agricultural output per

hectare into cereal yields, cereal cropping intensity and high value non-cereals

3. Agricultural intensification

AFG

ALB

DZAAGO

ARG

ARM

AZE

BGD

BLR

BEN

BTN

BOLBIHBWA

BRA

BGR

BFA

BDIKHMCMR

CAFTCD

CHL

CHNCOL

COMZARCOG

CRI

CIV

DOM

ECU

EGY

SLV

ERI ETH

FJI

GAB

GMB

GEO

GHA

GTM

GINGNB

GUY HTI

HND INDIDNIRN

IRQ

JAM

JOR

KAZ

KENPRKKGZ LAO

LVA

LBN

LSO

LBRLBY

LTU

MKD

MDG MWI

MYS

MLI

MRTMEX

MDAMNGMNE

MAR

MOZ

MMR

NAM

NPL

NIC

NER

NGA

PAK

PANPRY

PER

PHL

ROM

RUS

RWA

SEN

SRB

SLE

SOM

ZAF

LKA

SDN

SWZ

SYR

TJK

TZA

THA

TMPTGO

TUN

TUR

TKM

UGAUKR

URY

UZBVEN

VNM

ZMBZWE

02

00

04

00

06

00

0A

gricu

ltura

l ou

tput p

er

he

catr

e (

20

05

int. d

olla

rs)

0 200 400 600 800Agricultural population density (person per sq km)


AFG

ALB

AGO

ARG

ARMAZE

BGD

BLR

BEN

BTN

BOL

BIH

BRA

BGR

BFA

BDI

KHM

CMR

CAF

CHL

CHN

COL

COM

ZARCOG

CRI

CIV

DOM

ECU

EGY

SLV

ETH

FJI

GAB

GMB

GEO

GHA

GTMGINGNB

GUY

HTI

HND

IND

IDN

IRN

IRQJAM

JOR

KEN

PRK

KGZ

LAO

LVA

LBN

LSO

LBRLTUMKD

MDG

MWI

MYS

MEX

MDA

MARMOZ

MMR

NPL

NIC

NGA

PAKPAN

PRY

PER PHL

ROMRUS

RWASEN

SRB

SLE

ZAF

LKA

SWZ

SYR

TJK

TZA

THA

TMP

TGO

TUR

TKM

UGA

UKR

URY

UZB

VEN

VNM

ZMB

ZWE

05

00

100

01

50

0C

ere

al o

utp

ut pe

r h

ect

are

($

/ha)



AFG

ALB

DZA

AGO

ARGARMAZE

BGD

BLR

BEN

BTN

BOLBIH

BWA

BRA

BGR

BFA

BDI

KHM

CMR

CAF

TCD

CHL

CHN

COL

COM

ZAR

COG

CRICIVDOM

ECU

EGY

SLV

ERI

ETH

FJIGAB

GMB

GEO

GHA

GTM

GIN

GNBGUY

HTI

HND

IND

IDN

IRNIRQ

JAM

JOR

KAZ

KEN

PRKKGZ

LAO

LVA

LBN

LSO

LBR

LBY

LTU

MKD

MDG MWI

MYS

MLIMRT

MEXMDA

MNG

MNE

MAR

MOZ

MMR

NAM

NPL

NIC

NER

NGA

PAK

PAN

PRY

PER

PHL

ROM

RUS

RWA

SEN

SRB

SLESOM

ZAF

LKASDN

SWZ

SYR

TJKTZA

THA

TMP

TGO

TUN

TURTKM

UGA

UKR

URY

UZBVEN

VNM

ZMB

ZWE

05

01

00

150

Cere

als

cro

pp

ing in

tensi

ty (

%)


Cropping intensity in non-Africa sample is heavily explained by irrigation:R-sq = 0.56


AFG

ALB

AGO

ARG

ARM

AZEBGD

BLR

BEN

BTNBOLBIH

BRA

BGR

BFA

BDI

KHM

CMRCAF

CHL

CHN

COL

COMZARCOG

CRI

CIV

DOM

ECU

EGY

SLV

ETH

FJI

GAB

GMB

GEO

GHA

GTM

GINGNBGUY

HTI

HND

INDIDN

IRN

IRQ

JAM

JOR

KEN PRK

KGZ

LAOLVA

LBN

LSO

LBR

LTU

MKD

MDG MWI

MYS

MEX

MDA

MAR

MOZ

MMR

NPL

NICNGA

PAKPAN

PRY

PER PHL

ROM

RUS

RWA

SEN

SRB

SLE

ZAFLKA

SWZ

SYR

TJK

TZA

THA

TMPTGO

TUR

TKM

UGAUKR

URY

UZBVEN

VNM

ZMBZWE

01

00

02

00

03

00

04

00

05

00

0N

on

-sta

ple

s o

utp

ut (%

tota

l cro

p o

utp

ut)


Regression No. R1 R2 R3 R4

Dep. var.Agric. output

per haCereal output

per haCereal crop

intensityNon-cereal

output per haPopulation density 0.33*** 0.18*** 0.20*** 0.28***Density*Africa -0.11** -0.23*** -0.01 -0.01Road density 0.14*** 0.09** -0.03 0.19***Number of ports 0.14*** 0.21*** 0.03 0.15***Urban agglom (%) 0.29*** -0.09 0.31*** 0.31***Regional fixed effects? Yes Yes Yes Yes Sign of SSA dummies? + in E.Africa Zero Neg. + in E.AfricaAE controls Yes Yes Yes Yes No. Obs 243 243 243 243R-square 0.8 0.74 0.67 0.79

Table 4. Log-log estimates of agricultural value per hectare and its three components

Regression No. R1 R2 R3 R4

Dep. var.Fertilizers per hectare

Cattle/oxen per hectare

Irrigation per hectare

Capital per hectare

Population density 0.76*** 0.42*** 0.59*** 0.24***

Density*Africa -0.32** 0.15* -0.47*** -0.10***

Road density -0.08 0.31*** 0.04 0.07**

Number of ports 0.50*** 0.07 0.24*** 0.12***

Urban agglom (%) 0.38 0.03 0.24** -0.03

Regional fixed effects Yes Yes Yes Yes Sign of SSA dummies? Zero Neg. Zero ZeroAE controls Yes Yes Yes Yes No. Obs 0.73 0.77 0.92 0.77R-square 0.69 0.74 0.91 0.73

Table 5. Log-log estimates of specific agricultural inputs

Stylized facts Potential explanationsLow productivity of cereals sector

Low fertilizer application

Agronomic constraints (e.g. low soil fertility) Poor management practices, low human capital High transport costs (see regression 1 in Table 4); Low rates of subsidization (structural adjustment)

Low adoption of improved varieties

More varied agroecological conditions and crop mixLower returns because of limited use of other inputs (e.g. irrigation); Lower investment in R&D

Low use of plough/ tractors

Tsetse fly in humid tropics Feed/land constraints in some densely populated areas

Low rates of irrigation

Hydrological constraints; High costs of implementation and maintenance; Poor access to markets limits benefits

Noncereals

High non-cereal output per hectare

Agroecological suitability; Colonial introduction of cash crops; Non-perishable cash crops (cotton, coffee, cocoa, tea, tobacco) not limited by poor infrastructure and isolation

Table 7. Potential explanations of Africa’s agricultural intensification trajectory

02

46

8R

ura

l fe

rtili

ty r

ates

(#

child

ren

)

0 500 1000 1500Rural population density (person per sq km)

Non-Africa gradient

African gradient

Figure 3. Rural fertility rates and rural population density

3. Reducing rural fertility rates

ALBARMARMARMAZE BGDBGDBGD BGD

BGD

BEN

BEN

BEN

BOLBOLBOLBOLBOL

BWA

BRA

BRA

BFABFABFA

BDI

BDI

KHM

KHMKHM

CMR

CMRCMR

CAF

TCD

TCD

COLCOL

COLCOL

COL

COL

COM

ZAR

COGCIV

CIV

DOM

DOM

DOMDOMDOMDOMECU

ECU

SLV

SLV

ERIERI

ETH

ETH

ETH

GAB

GHA

GHA

GHAGHAGHA

GTMGTMGTMGTM

GINGIN

GUY

HTIHTIHTIHND

INDIND

IND

IDNIDNIDNIDN

IDNIDN

KAZKAZ

KENKENKEN

KEN

KENKGZ

LSO

LBR

LBR

MDG

MDG

MDG

MDGMWI

MWIMWIMWI

MLIMLI

MLIMLI

MRT

MEX

MOZ

MOZ

NAMNAM

NAM

NPL

NPLNPL

NPL

NICNIC

NERNER

NER

NGA

NGA

NGANGA

PAK PAKPRY

PRY

PERPERPERPERPERPER

PHLPHLPHL

PHL

RWA

RWA

RWARWA

RWA

SEN

SENSEN

SEN

SEN

SLE

LKA

SDN

SWZ

TZA

TZA

TZATZATZA

THA

TMPTGO

TGO

TURTUR

TKM

UKR

UZB

VNM VNM

ZMBZMB

ZMB

ZMB

ZWE

ZWE

ZWE

ZWE

ZWE

EGYEGYEGYEGYEGY EGY

JOR

JOR

JORJOR

JOR

MAR

MAR

MARTUN

02

46

81

0D

esir

ed fe

rtili

ty (

# c

hild

ren)


Full sample gradient

African sample gradient

Figure 4. Desired rural fertility & population density

Figure 5. Unmet contraception needs (%) and rural population density in Africa

BEN

BEN

BEN

BFA

BFA

CMR

CMRCMR

TCD

COM

ZAR

COG

CIVERI

ERI

ETH

ETH

GAB

GHAGHA

GHA

GHA

GIN

GIN

KEN

KEN

KEN

LSO

LBR

MDG

MDG

MWI

MWI

MWIMLI

MLI

MOZ

MOZ

NAMNAM

NER

NERNER

NGA

NGANGA

NGA

RWA

RWARWA

SEN

SEN

SLE

TZA

TZA

TZA

TZA

TGO

ZMB

ZMB

ZMB

15

20

25

30

35

40

Unm

et c

ontr

ace

ptio

n n

eeds

(%

wo

men

)

0 100 200 300 400Rural population density (person per sq km)

Sources

Regression number 1 2 3 4

Dependent variable Actual fertility Actual fertility Desired fertility

Desired fertility

Model Linear Log-log Linear Log-log

b/se b/se b/se b/se

Pop density (per 100 m2)

-0.14*** -0.09*** -0.11*** 0.00

Density*Africa 0.05 0.09*** -0.34*** -0.07***

Female sec. education (%)

-0.02*** -0.05*** -0.01** -0.08***

Ag. output per worker, log -0.58*** -0.13*** 0.01 0.06***

Africa dummy 1.25*** -0.15 2.13*** 0.67***

Number of observations

165 165 164 164

R-square

0.75 0.76 0.77 0.81

Table 8. Elasticities between rural fertility indicators & rural population density

4. Nonfarm diversificationMuch neglected in 1980s literature on BoserupSubsequent literature on both RNFE and migration &

remittances shows that RNF income is bigBut not much specific literature looking at pop densityOn RNF activity, often suggested there is a U-shaped

relationship between farm size and RNFE: landless poor are pushed into RNFE, rich are pulled in

Very difficult to look at rural-urban migration Int. remittances have boomed in last 10 years,

particularly in densely population South Asia – now 22% of rural income in Bangladesh

High density Africa Low density Africa Other LDCs

Country W M Country W M Country W M

Benin 50.4 23.7 Burkina Faso 12.9 8.1 BGD 53.4 44.5

Congo (DRC) 14.0 23.5 Chad 13.7 9.6 Bolivia 71.4 25.9

Ethiopia 34.3 9.7 Cote d'Ivoire 31.7 22.1 Cambodia 36.0

Kenya 47.1 37.3 Ghana 50.1 26.6 Egypt 69.4

Madagascar 17.8 15.3 Mali 44.6 16.0 Guatemala 79.1

Malawi 41.5 36.0 Mozambique 5.2 23.0 Haiti 24.0 19.0

Nigeria 65.5 37.0 Niger 60.2 35.8 India 22.4

Rwanda 7.3 14.2 Senegal 63.7 37.1 Indonesia 59.2 39.5Sierra Leone 25.2 20.1 Tanzania 7.2 10.5 Nepal 90.5 34.2Uganda 15.5 20.3 Zambia 30.1 19.5 Philippines 16.2 42.6

Table 9. Speculative estimates of rural nonfarm employment shares for men and women in the 2000s

Regression No. R1 R2 R3 R4 R5 R6

Sample Women Women Women Men Men Men

Population density 0.47 0.09 0.15 -0.33 -0.32 -0.31

Density*Africa -0.19** -0.22** -0.15* 0.03 -0.02 -0.02

Africa dummy -0.25 0.1 0.04 -0.43 0.09 0.09

Sec. educ. by gender 0.03 0.11 0.35*** 0.35***

Road density 0.14* 0.15** 0.17* 0.17*

Electricity 0.20** -0.07 0.09 0.09Ag. Output/worker, log 0.46*** 0.01

No. Obs. 162 122 95 74 74 74R-square 0.2 0.53 0.24 0.55 0.55 0.55

Table 11. Elasticities between RNF employment indicators and rural population density for women and men

Figure 6. National remittances earnings (% GDP) and rural population density

DZA

ARG

BGD

BEN

BOL

BRA

BFA

BDI

KHM

CMRCHL

CHN

COL

COG

CRI

CIV

DOM

ECU

EGY

SLV

ETH

GHA

GTM

GIN

HTI

HND

IND

IDN

IRNIRQ

JOR

KEN

LAO

LBN

LBR

LBYMYS

MLI

MEX

MAR

MOZ

NPL

NIC

NER

NGA

PAK

PAN

PRY

PER

PHL

RWA

SEN

SLE

ZAF

LKA

SDN

SYR

TZATHA

TGO

TUNUGA

URYVEN

VNM

ZMB05

10

15

20

25

Rem

ittan

ce e

arn

ing

s (%

GD

P)


Estimator OLS Robust OLS RobustStructure Levels (logs) First difference Levels (logs) First differenceDensity variable Agricultural Agricultural Rural Rural

Population density 0.25*** 0.97** 0.31*** 1.17***Population density*Africa 0.05 -0.94 0.04 -1.22**Total population -0.24*** -1.31** -0.23*** -0.82Lagged remittances -0.21*** -0.24***Lagged population density 0.06 0.06West Africa dummy -0.67* -0.49 Central Africa dummy -1.55*** -1.40*** East Africa dummy -0.90** -0.74* Southern Africa dummy 0.14 0.24 1977-87 dummy 0.15 0.12 1987-97 dummy 0.33* -0.09 0.28* -0.061997-2007 dummy 0.79*** 0.19 0.72*** 0.24*

Number of observations

231 147 231 159

R-square

0.39 147 0.4 0.22

Table 11. Estimating elasticities between national remittance earnings (% GDP) and population density

5. ConclusionsLand pressures are severe in much of Africa, esp. high

density SSA, where small farms are getting smaller, and will continue to get smaller as pop. grows

Yet history shows that rural people are generally adept at adapting to mounting land pressures.

Ag intensification is only part of the adaptationThe question we posed is whether Africa is different In many ways, the answer is yes . . .

Adaptation 1 – Agricultural IntensificationAfrica has intensified agriculture, but largely

through high value non-perishable crops (HVCs)Much less historical success with cereals, and much

less potential given limited potential for irrigationShould we shift emphasis of research and development

strategies from cereals to HVCs?CGIAR, for example, barely looks at cash crops like

coffee, tea, cotton, cocoa, tobacco (even though cash buys food!)

5. Conclusions

Adaptation 2 – Reducing fertility ratesHigher densities (smaller farms) apepar to lead to a

desired reduction in fertility in AfricaBut desired reductions are not met by access to

contraceptive technologiesHigh-density East Africa now shows mixed policiesEthiopia & Rwanda are investing in family planning

(*), but Museveni (Uganda) has resisted family planning (population growth is “a great resource”)

Asian experience suggests FP yields high returns

5. Conclusions

Adaptation 3 – Nonfarm diversificationWeak evidence, but evidence that is there suggests

that nonfarm sector doesn’t just grow without engines like education, infrastructure, agriculture (also true for African cities?)

Boom in overseas migration and remittances is new, and unexpected.

20 years ago, BGD and Pakistan were regarded as too big to benefit from remittances. Not true now.

Why isn’t Africa getting more remittances?

5. Conclusions

Finally, we ask whether the results we find warrant a re-think in the way high density countries pursue rural development

Are SSA countries thinking through the implications of rural pop. growth for farm sizes and rural welfare?

Do SSA countries need rural development strategies that are more integrated with respect to smallholder intensification, commercial farms, family planning, migration and rural nonfarm development?

What are the costs of not doing so?

5. Conclusions

adaptation to land constraints: is africa different?

Business