Download Malthusian Trap in East Africa: Analysis of Indicators and more Summaries Statistics in PDF only on Docsity!
ARXIV.ORG
EAST AFRICA IN THE MALTHUSIAN TRAP?
A statistical analysis of financial, economic, and
demographic indicators
Andrey Korotayev & Julia Zinkina
Abstract. A statistical analysis of financial, economic, and demographic indicators performed by
the authors demonstrates (1) that the main countries of East Africa (Uganda, Kenya, and
Tanzania) have not escaped the Malthusian Trap yet; (2) that this countries are not likely to
follow the “North African path” and to achieve this escape before they achieve serious successes
in their fertility transition; (3) that East Africa is unlikely to achieve this escape if it does not
follow the “Bangladeshi path” and does not achieve really substantial fertility declines in the
foreseeable future, which would imply the introduction of compulsory universal secondary
education, serious family planning programs of the Rwandan type, and the rise of legal age of
marriage with parental consent. Such measures should of course be accompanied by the
substantial increases in the agricultural labor productivity and the decline of the percentage of
population employed in agriculture.
The Malthusian trap
1
is a rather typical for pre-industrial societies situation when the growth of
output (as it is accompanied by a faster demographic growth) does not lead in the long-range
perspective to the increase in per capita output and the improvement of living conditions of the
majority of population that remains close to the bare survival level (see, e.g. , Malthus, 1978
[1798]; Artzrouni & Komlos, 1985; Steinmann & Komlos, 1988; Komlos & Artzrouni, 1990;
Steinmann, Prskawetz, & Feichtinger 1998; Wood, 1998; Kögel & Prskawetz, 2001; Korotayev,
Malkov, & Khaltourina, 2006b; Korotayev & Khaltourina, 2006; Clark, 2007; Conley, McCord,
& Sachs 2007; Korotayev et al. , 2011 ).
In complex pre-industrial societies the Malthusian trap was one of the main generators of state
breakdowns and sociopolitical collapses frequently resulted in millions of deaths (see, e.g. ,
Korotayev & Khaltourina, 2006; Korotayev, Malkov, & Khaltourina, 2006 b ; Chu & Lee, 1994;
(^1) Using the terminology of non-linear dynamics one can also denote it as the low-level equilibrium attractor (ср.
Nelson, 1956).
Nefedov, 2004; Turchin, 2003, 2005 a , 2005 b ; Turchin & Korotayev, 2006; Turchin & Nefedov,
2009; Usher, 1989; van Kessel-Hagesteijn, 2009; Korotayev et al. , 2011; for a recent case in East
Africa see André & Platteau 1998)^2.
The global modernization, and first of all the increase in per capita productivity due to major
technological advances in the recent centuries in conjunction with demographic transition
secured the systematic excess of output growth rates over population growth rates, which
allowed most social systems to escape the Malthusian trap (see, e.g. , Artzrouni & Komlos, 1985;
Lucas, 1998; Galor & Weil, 1999; Kögel & Prskawetz 2001; Korotayev, Malkov, & Khaltourina,
2006 a ; Pereira, 2006).
However, these modernization processes started later in Sub-Saharan Africa than in the rest of
the world; and even in the recent decades the Malthusian trap tended to produce state
breakdowns in this region (see, e.g., André & Platteau 1998). What is more, some parts of
Tropical Africa (and, first of all, East Africa) appear to be trapped in the Malthusian Trap till
now (though contrary views continue to be expressed [see first of all Kenny, 2010]). This looks
especially salient if we compare the long term per capita food consumption dynamics in North
Africa, on the one hand, and the three main countries of East Africa (Kenya, Tanzania, and
Uganda), on the other (see Fig. 1):
(^2) Note that Malthus himself considered warfare (including, naturally, internal warfare) as one of the most important
results of overpopulation (in addition to epidemics and famines). What is more, he regarded wars, epidemics, and famines as so-called “positive checks” that checked overpopulation in pre-industrial systems (Malthus 1978 [1798]). Thus, in pre-industrial societies bloody political upheavals frequently turned out to be a result of the respective social systems being caught in the Malthusian trap.
This escape was accomplished first of all due to the fact that in this period all the countries of
this region managed to achieve very high GDP growth rates that exceeded significantly the
population growth rates, which resulted in very substantial increases in per capita GDP^6 (see
Fig. 2):
Fig. 2. Relative Dynamics of Population, GDP, and GDP per capita
in North African Countries during Their Escape from the Malthusian Trap
Algeria, 1971-1985 (100 = 1971 level) Egypt, 1971-1985 (100 = 1971 level)
Morocco, 1971-1985 (100 = 1971 level) Tunisia, 1971-1985 (100 = 1971 level)
Libya, 1960-1970 (100 = 1960 level)
Data source : Maddison 2010.
This escape was further supported by the decrease of the fertility rates (see Fig. 3):
(^6) Note that Libya made the respective breakthrough a bit earlier, in the 1960s.
population
GDP
GDP per capita
100
population
GDP
GDP per capita
population
GDP
GDP per capita 100
population
GDP
GDP per capita
population
GDP
GDP per capita
Fig. 3. Dynamics of the Total Fertility Rate (TFR, births per woman)
in North and East Africa (1960–2011)
Data source : World Bank 2013: SP.DYN.TFRT.IN.
As we see, in this respect, North Africa also stands in sharp contrast with East Africa, where the
total fertility rate remains very high till now.
Of course, in recent decades the TFR somehow declined in East Africa too, but the extent
of this decrease was rather small, and it was mostly compensated by the decrease of the death
rate (which was interrupted somehow by the HIV/AIDS epidemic, but resumed with its
decline
7
). As a result, in a contrast with North Africa which observed a very significant decline
of the population growth rates in the 1980s and 1990s, East African population growth rates
remained mostly at such levels that were typical for North African countries before the start of
the fertility decline in this part of the world (see Fig. 4):
(^7) See, e.g., World Bank 2013: SP.DYN.LE00.IN, SP.DYN.IMRT.IN, SH.DYN.MORT, SP.DYN.CDRT.IN;
Ngallaba et al. 1993: 71–78; Bureau of Statistics [Tanzania] and Macro International Inc. 1997: 97–104; Opiyo, Omolo, Imbwaga 2010; National Bureau of Statistics [Tanzania] 2000: 85–92; 2005: 123–130; 2011: 117–126; Kaijuka et al. 1989: 53–66; Statistics Department [Uganda] and Macro International Inc. 1996: 97–104; Uganda Bureau of Statistics 2001: 97 – 108; 2007: 109 – 118; 2012: 98 – 104; National Council for Population and Development et al. 1989: 55–72; 1994: 83–92; Kichamu 1999; Muindi, Bicego 1999; Bicego, Kichamu 1999; Otieno, Omolo 2004; Kenya National Bureau of Statistics 2010: 103–112.
Algeria Egypt Libya Morocco Tunisia Uganda Kenya Tanzania
Fig. 6. Relative Dynamics of Population, GDP, and GDP per capita
in Kenya, 1980–2009 (100 = 1980 level)
Data source : World Bank 2013: SP.POP.TOTL, NY.GDP.MKTP.PP.KD.
As we see, between 1980 and 2009 the Kenyan economy grew rather substantially, by about
150%. But the Kenyan population in those years grew as substantially – and, as a result, per
capita GDP in Kenya in 2009 was almost the same as in 1980.
However, the situation in Uganda and Tanzania does not appear as simple. In both countries
in recent years one could observe a rather substantial growth of per capita GDP (see Fig. 7):
Fig. 7. GDP per capita (PPP, constant 2005 international $) dynamics in
Tanzania and Uganda, 1994– 2011
Data source : World Bank 2013: NY.GDP.PCAP.PP.KD
Population GDP GDP per capita
Tanzania Uganda
Why was not this converted in the growth of the per capita food consumption? It appears very
important to answer this question, as the positive values of per capita GDP growth rates observed
recently in most countries of Tropical Africa have made some experts believe that “(almost) the
whole World – including most IF NOT ALL of Africa – shares features of Malthus’ vision for
escaping the trap he outlined” (Kenny 2010: 192, our emphasis).
To answer this question it appears necessary to consider the recent economic growth in East
Africa by sector. Let us start with the dynamics of production of services in Uganda (see Fig. 8):
Fig. 8. Relative Dynamics of Population, Services Production, and Services
Production per capita in Uganda, 1994–2011 (100 = 1994 level)
Data source : World Bank 2013: NV.SRV.TETC.KD, SP.POP.TOTL; the services production index was calculated on the basis of the World Development Indicator NV.SRV.TETC.KD: “Services, value added (constant 2000 US$)”.
As we see, in these years Ugandan successes in this sector of its economy were really
spectacular. Between 1994 and 2011 Ugandan population almost doubled. But growth rates in
population
services production
services production per capita
Fig. 10. Relative Dynamics of Population, Agricultural Production, and Agricultural
Production per capita in Uganda, 1994–2011 (100 = 1994 level)
Data source : World Bank 2013: NV.AGR.TOTL.KD, SP.POP.TOTL; the agricultural production index was calculated on the basis of the World Development Indicator NV.AGR.TOTL.KD: “Agriculture, value added (constant 2000 US$)”.
As we see, between 1994 and 2011 the growth of agricultural production in Uganda was still
quite substantial (60%). But the population growth in Uganda in those years was even higher.
And as a result the agricultural production per capita in Uganda declined. Note that this
correlates rather well with the data on almost perfect stagnation of the agricultural labor
productivity in Uganda in the recent years^9 (see above Fig. 5).
Actually one of the most important differences between contemporary North and East Africa
is that in North Africa a rather small proportion of population is employed in agriculture,
whereas in East Africa this is the overwhelming majority of the population (see Fig. 11):
(^9) For some peculiarities of Tanzanian case see Appendix 1. On the phenomenon of the agricultural growth rate
lagging behind the overall economic growth rate in Uganda see, e.g., Deininger, Okidi 2001: 124–125.
population
agricultural production
agricultural production per person
Fig. 11. Employment in agriculture, %, in North and East Africa in the 2000s
Data source : World Bank 2013: SL.AGR.EMPL.ZS.
Hence, the majority of the population of the East African countries will remain in the Malthusian
Trap even against the background of an overall growth of GDP per capita if this growth is
achieved through the productivity growth in non-agricultural sectors only with the agricultural
labor productivity of the majority of rural population stagnating or declining.
Hence, a sustainable escape from the Malthusian trap can hardly be achieved in East
Africa without a substantial increase in the productivity of labor of the majority of agricultural
population. However, one of the main obstacles for this is just the rural overpopulation
effectively blocking such increases.
As Roth and Fratkin (2005: 6) note, in East Africa “rapid population growth has affected rural
and urban areas, where farmers increasingly move onto less productive lands to raise their crops
and families”. Population pressure leads to the reduction of the fallow periods^10 , as a result of
which the soil fertility fails to restore to original levels, as a result frequently various innovations
(like fertilizers) just allow to prevent (at least partly) the decline of the current low agricultural
labor productivity instead of increasing it (see, e.g., Bigsten, Kayizzi-Mugarewa 1999: 80).
Hence, the countries of East Africa will hardly be able to achieve a sustainable escape from
the Malthusian Trap before they achieve a serious fertility decline.
However, why could not East Africa follow the “North African” path – that is, achieving the
escape from the Malthusian Trap first, and accomplishing the fertility decline second?
One point is that North Africa escaped from the Malthusian Trap just 1–2 decades after the
population growth rates exceeded 2.5% per year levels (implying the population doubling just
within 30 years), whereas in East Africa such a growth continues for many decades and with any
new decade the escape from the Malthusian Trap before the fertility decline becomes less and
less likely.
(^10) Those analysts who deny the presence of overpopulation problems in Tropical Africa love to say things like “only
12% of all potential arable land is under cultivation [in Sub Saharan Africa]” (Hayes 2012: 114); however, they tend not to take into consideration the point that “new” land being brought under cultivation in Tropical Africa under population pressure in most cases is not virgin, but it is rather fallow land whose straightforward cultivation leads to the reduction of fallow periods, inadequate soil fertility restoration, soil degradation, and decrease of labor productivity, whereas against such a background serious efforts should be taken even in order just to retain the agricultural labor productivity at old level. E.g., “Blackie et al. (1998) observed that because of increasing population density and declining land availability ‘maize is now grown in continuous cultivation rather than as part of a fallow [rotation] which traditionally used to restore soil fertility and reduce the build up of pests and diseases. The soil resource base is now being degraded with a consequent reduction of yield’ ” (Barahona, Cromwell 2005: 159).
Egypt Tunisia Algeria Tanzania Uganda Kenya
Fig. 13. UN Population Division projections of TFR (births per woman)
dynamics in East Africa underlying their medium population forecast for the
main East African countries
Data source : UN Population Division 2013.
Still these projections forecast the following population dynamics for East Africa (see Fig. 14):
Kenya Uganda Tanzania
Fig. 14. UN Population Division medium population forecast for the main
East African countries, millions
Note : black dotted line presents the current population of Russia for comparison. Data source : UN Population Division 2013.
As we see, even with the acceleration of fertility decline forecasted by the UN Population
Division, the population of Kenya and Uganda will exceed the population of Russia in the
second half of the century. Tanzania will reach Russia in terms of population by 2050 and is
projected to have twice the Russian population by 2100. And Tanzania is almost bound to have
population of Russia, as, according to the estimates of the UN Population Division (2013), the
number of children below 5 in Tanzania is already almost the same as in Russia.
Note that without more substantial fertility declines (which would imply the introduction of
compulsory universal secondary education, serious family planning programs of the Rwandan
type [see, e.g. , Kinzer 2007; WHO 2008; Lu et al. 2012]), and the rise of legal age of marriage
with parental consent) East Africa is likely to face two options, if it would like to escape the
Malthusian Trap:
1) to move no less than 50% of its rural populations to dynamically growing urban sectors
(that is, to increase the urban population proportion growth rate substantially in comparison with
what was observed in recent years [see Fig. 12]). Note that, in conjunction with the UN
Population Division projections this would imply the growth of the urban population if East
Africa by about 80 million people by 2050 and by c. 160 million people by the end of this
century. This would be tantamount to the creation in a few forthcoming decades in East Africa of
about 80 new cities of the size of Koeln. Most likely such a scenario would rather produce an
explosive growth of slums, urban overpopulation, social explosions and so on (cf. Korotayev et
al. 2011).
Millions
Tanzania
Kenya
Uganda
Russia-
Fig. 15. TFR Dynamics in Bangladesh (1960–2010)
Data source : World Bank 2013: SP.DYN.TFRT.IN.
Fig. 16. Relative Dynamics of Population, GDP, and GDP per capita
in Bangladesh, 1970–2008 (100 = 1970 level)
Data source : World Bank 2013: SP.POP.TOTL, NY.GDP.MKTP.PP.KD.
Still there was an appreciable (albeit very slow) growth of per capita GDP in Bangladesh in
1980 – 1995. However, this growth failed to be translated into a significant improvement of the
life of the majority of the Bangladeshi population, which is evidenced by the dynamics of
average food consumption that remained during this period at a very low level that is typical for
Total fertility rate, children per
woman
GDP
Population
GDP per capita
Tropical African countries trapped in the Malthusian trap and did not exhibit any tendency to
increase (that is also rather characteristic for this type of countries) (see Fig. 17):
Fig. 17. Average per capita food consumption dynamics (kcal per capita per day)
in Bangladesh, Kenya, Tanzania, and Uganda (1980–1995)
Data source : FAO 2013.
Thus, when fertility in Bangldeshi remained close to Tropical African levels (see Fig. 18 below),
average per capita food consumption levels also remained extremely close to Tropical African
ones (see Fig. 17 above).
Fig. 18. TFR dynamics in Bangladesh, Kenya, Iran, Uganda, and Tazania (1980-1995)
Data source : World Bank 2013: SP.DYN.TFRT.IN.
But why did in 1980–1995 the average per capita food consumption in Bangladesh stall against
the background of quite visible growth of per capita GDP in those years. The point is that this
growth was achieved through the increase in productivity in non-agricultural sectors of
Bangladesh Kenya Uganda Tanzania
Total fertility rate, children per
woman
Bangladesh
Kenya
Uganda
Tanzania
growth of productivity of agricultural labor during this period was blocked by explosive rural
overpopulation was the main cause of the stall in the growth of the average per capita food
consumption (see Fig. 21):
Fig. 21. Labor productivity in agriculture (value added per worker per year, constant 2000 dollars) in
Bangladesh and East Africa, 1980-
Data source : World Bank 2013: EA.PRD.AGRI.KD.
The overall picture of relative dynamics in Bangladesh in the period before it achieved
substantial decline in fertility levels looks as follows (Fig. 22):
Fig. 22. Dynamics of GDP per capita, agricultural labor productivity, and average per capita food
concumption in Bangladesh (1980-1995, 100 = 1980 level)
Data source : World Bank 2013: EA.PRD.AGRI.KD, SP.POP.TOTL, NY.GDP.MKTP.PP.KD; FAO 2013.
Bangladesh Kenya Uganda Tanzania
GDP per capita
Agricultural labor
productivity
Per capita food
consumption
As we see against the background of huge population pressure even quite noticeable per capita
GDP growth failed to get translated into a comparatively noticeable growth of agricultural labor
productivity, which (within the context of the overwhelming majority of population still being
rural) resulted in the absence of any noticeable improvement in the average per capita food
consumption at all.
Note that in some East African countries such a dynamics is observed till now. For example, for
Uganda it looks as follows (Fig. 23):
Fig. 23. Dynamics of GDP per capita, agricultural labor productivity, and average per capita food
concumption in Uganda (1990-2009, 100 = 1990 level)
Data source : World Bank 2013: EA.PRD.AGRI.KD, SP.POP.TOTL, NY.GDP.MKTP.PP.KD; FAO 2013.
As we see the pattern which is observed till now in Uganda is almost identical with the one
which had been observed in Bangladesh before this country managed to achieve a really
substantial fertility decline – quite a noticeable per capita GDP growth fails to get translated into
a comparatively noticeable growth of agricultural labor productivity, which (within the context
of the overwhelming majority of population still being rural) results in the absence of any
noticeable improvement in the average per capita food consumption at all.^12 This demonstrates
again that we should not expect that the comprehensive development will “automatically” bring
about all the necessary fertility declines (this belief seems to be analogous to the famous belief of
(^12) Note that this is not the worst possible scenario for the present-day Tropical Africa (at least in recent decades we
do not observe there any pronounced trends toward decline of those variables. For example, in Kenya where after 1980 population grew as fast as GDP, which resulted in stagnation of per capita GDP and a long-term trend toward decline in agricultural labor productivity and average per capita food consumption.
GDP per capita
Agricultural labor productivity Per capita food consumption
