john bongaarts population council, new york süssmilch

Africa’s slow fertility transition

John Bongaarts Population Council, New York

Süssmilch Lecture

Max Planck Institute, Rostock 3 Sep 2015

0

1

2

3

4

1900 1950 2000 2050 2100 2150

Billi

ons

Population projections for sub-Saharan Africa

2015 projection 2002/4 projection

2015 population

Source, United Nations

0

10

20

30

40

2000 2020 2040 2060

Crude birth and death rates sub-Saharan Africa

Crude birth rate Crude death rate

2015 2002/4 2002/4 2015

Source: United Nations

0

1

2

3

4

5

6

7

8

1980 1990 2000 2010 2020

Birt

hs p

er w

oman

TFR trends in sub-Saharan Africa

BeninBotswanaBurkina FasoBurundiCameroonCape VerdeCentral African RepublicChadComorosCongo (Brazzaville)Congo Democratic RepublicCote d'IvoireEthiopiaGabonGambiaGhanaGuineaKenyaLesothoLiberiaMadagascarMalawiMaliMauritaniaMozambiqueNamibiaNigerNigeriaRwandaSao Tome and PrincipeSenegalSierra LeoneSouth AfricaSudanSwazilandTanzaniaTogoUganda

Source: DHS

Socioeconomic development

Mortality decline

Fertility

Family planning Program

Diffusion processes

Determinants of fertility

Hypotheses

1. Africa’s development is slow 2. Africa is exceptional 3. Family planning programs are lacking

Outline 1. Fertility and development trends

• Levels • Pace

2. African exceptionalism 3. Impact of family planning programs 4. Conclusions

0

1

2

3

4

5

6

7

8

1960 1980 2000 2020

Birt

hs p

er w

oman

Total fertility rate

Africa Other LDCs

Development indicators 1970-2010 - GDP per capita (at PPP) from the PWT - Education, % with primary + (Wittgenstein) - Life expectancy at birth (UN 2013) - Percent urban (UN 2014).

500

1000

2000

4000

8000

1960 1980 2000 2020

$ pe

r cap

ita

GDP per capita(PPP) Other LDCs Africa

0

20

40

60

80

100

1960 1980 2000 2020

%. w

ith P

rimar

y+

Education Other LDCs Africa

30

35

40

45

50

55

60

65

70

75

1960 1980 2000 2020

Year

s

Life expectancy Other LDCs Africa

0

10

20

30

40

50

60

70

1960 1980 2000 2020

%

Percent urban Other LDCs Africa

Average at the time of transition onset

Sub-Saharan Africa

Other LDCs

TFR decline % 10 10

Year of transition onset 1994 1975

GDP/cap(log) 6.9 7.7

Education (% primary+) 29 42

Life expectancy 51 59

Percent urban 29 40

Conclusions 1) African transitions later in time

Consistent with conventional theory 2) But early relative to level of development Consistent with diffusion theories

Outline 1. Fertility and development trends

• Levels • Pace

2. African exceptionalism 3. Impact of family planning programs 4. Conclusions

-0.05

0

0.05

0.1

0.15

1960 1980 2000 2020Birt

hs p

er w

oman

/yea

r Total fertility rate, pace

Other LDCs Africa

-2-10123456

1960 1980 2000 2020

% y

ear

GDP per capita Pace

Other LDCs Africa

0

1

2

1960 1980 2000 2020

%/Y

ear

Education pace

Africa Other LDCs

-0.2

0

0.2

0.4

0.6

0.8

1960 1980 2000 2020

Year

s

Life expectancy Pace

Africa Other LDCs

00.10.20.30.40.50.60.7

1960 1980 2000 2020

%

Percent urban Pace Other

LDCs Africa

Average pace at the time of transition onset

Sub-Saharan Africa

Other LDCs

TFR 0.09 0.15 Year 1994 1975 GDP/cap(log) 0.008 0.034 Education (% primary+) 1.2 2.0 Life expectancy 0.12 0.47 Percent urban 0.29 0.57

Conclusions 3) African transitions are slow because the pace of development is slow

Consistent with conventional theory

Outline 1. Fertility and development trends

• Levels • Pace

2. African exceptionalism 3. Impact of family planning programs 4. Conclusions

012345678

100 1000 10000 100000

Birt

hs p

er w

oman

$ per year

TFR by GDP/capita, 2010

Africa Other LDCs

012345678

100 1000 10000 100000

Birt

hs p

er w

oman

$ per year

TFR by GDP/capita, 2010

Africa Other LDCs

012345678

0 50 100

Birt

hs p

er w

oman

% primary+

TFR by education, 2010

Africa Other LDCs

012345678

40 60 80 100

Birt

hs p

er w

oman

Years

TFR by life expectancy, 2010

Africa Other LDCs

012345678

0 50 100

Birt

hs p

er w

oman

Percent urban

TFR by percent urban, 2010

Africa Other LDCs

TFR 2010

Africa effect 1.18** GDP/cap -0.36* Education -0.019*** Life expectancy -0.02 Percent urban 0.00 R2 0.84 N 71 Year 2010

Socioeconomic dev. Mortality decline

Cost and benefits of children

Fertility preferences

Demand for contraception

Use of contraception

Fertility

TFR Contraceptive prevalence

Desired Family size

Africa effect 1.18** -10.9* 1.24* GDP/cap (log) -0.36* 0.83 0.3 Education -0.019*** 0.51*** -2.7*** Life expectancy -0.02 0.65 -0.04 Percent urban 0.00 -0.15 0.01 R2 0.84 0.86 0.72 N 71 71 39 Year 2010 2010 Latest DHS

Conclusions: African fertility is high relative to development Consistent with theories about African exceptionalism (e.g. Caldwell)

Caldwell (1992) : ”(1) African traditional society stressed the importance of ancestry and descent. …younger generations assisted the older generations .. for males at least, high fertility ultimately brought substantial economic returns… (2) Polygyny led in West and Middle Africa to separate spousal budgets. The father was spared much of the cost of rearing children. (3) There was strength and safety in numbers. Communal land tenure meant that large families could demand a greater share of the land… (4) Family planning programs were nonexistent or weak ..regarded as foreign or as incompatible with African culture

Outline 1. Fertility and development trends

• Levels • Pace

2. African exceptionalism 3. Impact of family planning programs 4. Conclusions

Socioeconomic dev. Mortality decline

Cost and benefits of children

Fertility preferences

Demand for contraception

Use of contraception

Fertility

Family planning program

Cost of contraception

Unmet need for contraception

Contraceptive access /information

0

20

40

60

80

1960 1980 2000

Perc

ent o

f cou

ples

Met and unmet need for contraception,developing world

Potential demand forcontraception

Unmetneed

Currentuse

0

10

20

30

40

50

60

70

# pr

egna

ncie

s Planned and unplanned pregnancies

Africa

Planning status Unintended Intended

Pregnancy outcome Abortion Unintended birth Intended birth

Source: Guttmacher

0

10

20

30

40

50

1974 1976 1978 1980 1982 1984 1986

Con

trace

ptiv

e P

reva

lenc

e (%

) Successful FP experiment in Matlab,

Bangladesh

Experimental area

Control area

Source: Cleland et al. 1994

0 2 4 6

IndonesiaPhilippines

IranJordan

BangladeshPakistan

KenyaUganda

RwandaBurundi

Births per woman

Fertility impact of weak vs strong FP programs

Weak Strong

0

1

2

3

4

5

6

7

0

10

20

30

40

50

60

1990 1995 2000 2005 2010 2015

Birt

hs p

er w

oman

% m

arrie

d w

omen

Rwanda reproductive trends

TFR Contraceptive use

Conclusion: 1) High unmet need for contraception and

large numbers of unplanned pregnancies 2) Family planning programs can reduce

fertility by about 1.5 births per woman

0

1

2

3

4

5

6

2000 2020 2040 2060 2080 2100 2120

Billi

ons

Population projection variants sub-Saharan Africa Variant

High Medium Low

Source: United Nations

Causes of slow fertility decline in Africa 1) Slow pace of development 2) African pro-natalism 3) Weak or non-existent FP programs

Sources • Bongaarts John, “Africa’s unique fertility transition”

Paper prepared for the US National Academies of Science Workshop on Recent Trends in Fertility in Sub-Saharan Africa, June 14-15, 2015, Washington DC.

• J. Bongaarts, J. Cleland, J. Townsend, J. Bertrand, and M. Das Gupta, Family Planning Programs for the 21st Century: Rationale and Design, New York Population Council (2012).

© 2015 The Population Council. All rights reserved.

NOT FOR CITATION OR CIRCULATION. The content of these slides are provided for the purpose of scientific discussion only. Content may be changed based on finalized analyses. Prior to finalization, this content may not be reproduced, stored, or transmitted in any form or by any means without prior permission.