Content uploaded by Ernest L. Molua
Author content
All content in this area was uploaded by Ernest L. Molua on Dec 17, 2018
Content may be subject to copyright.
Content uploaded by Ernest L. Molua
Author content
All content in this area was uploaded by Ernest L. Molua on Sep 29, 2014
Content may be subject to copyright.
Policy Options
Climate variability, vulnerability and
effectiveness of farm-level adaptation options:
the challenges and implications for food
security in Southwestern Cameroon
ERNEST L. MOLUA
Department of Agricultural Economics, Georg-August University
Goettingen, Platz der Goettinger Sieben 5, 37073 Goettingen
Germany. Tel: 49 551 394811. Fax: 49 551 399866.
Email: emolua@uni-uaao.gwdg.de
ABSTRACT. The risks associated with increasing climate variability pose technological
and economic challenges to societies which are dependent on agriculture for their liveli-
hood. In Southwestern Cameroon the natural variability of rainfall and temperatures
contribute to variability in agricultural production and food insecurity. This paper
explores the impact of climate variability in Southwestern Cameroon on food avail-
ability. It examines farm household’s vulnerability to food availability relating to
climate, and reviews the interplay of climate, agriculture, and prospects for food security
in the region. An econometric function directly relates farm income and precipitation, in
order to statistically estimate the significance of farm-level adaptation methods. The
results reveal that precipitation during growing and adaptation methods through
changes in soil tillage and crop rotation practices have significant effects on farm returns.
An essential precondition for food security and overall agricultural development in
Southwestern Cameroon is a dynamic agricultural sector brought about both by steady
increase in agricultural production and by greater efforts in farmer support, to enable
farm households to take advantage of the opportunities and to minimize the negative
impacts of climate variation on agriculture.
Key words: Cameroon, climate variability, farm households, vulnerability, food security
1. Introduction
Climate variation, climate change and their potential impacts have given
rise to heated debates within and outside the farming world. According
to the United Nations’ Intergovernmental Panel on Climate Change,
Environment and Development Economics 7: 529–545 © 2002 Cambridge University Press
DOI:10.1017/S1355770X02000311 Printed in the United Kingdom
The author greatly appreciates the comments of Stephan von Cramon-Taubadel,
Manfred Zeller, the Editor of this journal and the three anonymous reviewers for
their detailed comments that helped improve the earlier draft.
anthropogenic greenhouse gas emissions are significantly altering the
earth’s climate. Climate models predict that mean annual global surface
temperature will increase by 1–3.5C by 2100 and the global mean sea level
will rise by 15–95 cm with changes occurring in the spatial and temporal
patterns of precipitation (IPCC, 1990, 1997). Some studies have asserted
that the potential impacts could be severe on developing countries (Parry,
1990; IPCC, 1997; Winters et al., 1999). One of the basic concerns to devel-
opment experts is food security, an issue with local, national, and
international dimensions. Climate variation can enhance or diminish a
local area’s comparative advantage in agriculture. Heightened year-to-
year variation of climate and changing local factors can markedly affect
income from agriculture production, costs to consumers, and food scarcity.
Cameroon’s economy is predominantly agrarian. The exploitation of
agricultural and other natural resources remains the driving force for the
country’s economic development, contributing some 60 per cent of the
total value of exports (Tchoungi et al., 1996; EIU, 1999). Cameroon’s food
and agricultural policy is aimed at providing guidance to farmers, pro-
moting the use of improved inputs, adequate funding of research
institutes, consolidating the achievements of food self-sufficiency, and
achieving food security. However, the lofty policy objectives have been
tested by recent developments following the enactment of the Enhanced
Structural Adjustment Facility (ESAF) in 1997 and the prevailing environ-
mental constraints. Climate-induced production variability has had
immediate and important macroeconomic impacts.
Agricultural production depends on climate variables, such as tempera-
ture, precipitation, and light. Farm households’ ability to grow enough
food to feed themselves and their animals is determined to a large extent
by the weather. Examining the effects of altered weather on agriculture-
dependent households, Downing (1992) asserted that change in global
climate variables may present risks to future livelihoods. Shifts in tem-
perature and precipitation are, therefore, important parameters for
agrarian societies in regions such as Southwestern Cameroon which are
dependent on agriculture for their livelihood. Therefore, what types of
adaptations and policies will be necessary to take advantage of the benefits
and to minimize the negative impacts of increased climate variability on
agricultural production and food security in Southwestern Cameroon? To
address this question this paper sets out to (i) examine farm household
vulnerability and associated risks pertaining to climate variability; (ii)
identify and assess the economic significance of farm-level adaptation
strategies in response to climate variability; and (iii) highlight the chal-
lenges to future food security in the region associated with farmers’
vulnerability to changing climate variables.
This paper contributes to the already existing body of literature in two
ways. Firstly, it adds to the understanding of how climatic conditions
affect rural households. Secondly, the paper constructs an econometric
model that identifies climate-related factors that influence farm-level
income. An improved understanding of the ways in which climate vari-
ability affects society and the environment is a prerequisite for attempting
to improve agricultural productivity and increase farm incomes. An
530 Ernest L. Molua
assessment of the impact of weather variables on food production and
farm-level income will contribute to efforts aimed at ensuring both
increased food availability through sustainable domestic production,
storage and/or trade, as well as farmers’ adaptation to a long-term climate
change anomaly.
The remainder of this paper is organized as follows. The second section
briefly discusses the issue of climate variability and climate change. The
third section describes the study area, the nature and source of data used.
It further reviews the vulnerability of farms in the region, the methods of
farm-level adaptation employed, the challenges for food security and then
estimates the effect of the adaptation methods on net farm return. The
results of the analyses are succinctly presented and discussed. In section 4,
the implications of the findings and prospects for food security are exam-
ined. Some policy recommendations based on the empirical findings are
presented in section 5.
2. The issue of climate variability and change
Humanity cannot accurately predict what the next season will bring.
Farmers, input suppliers, marketers, and governments would all like to
know because it is critical to decision making. The uncertainty and risks
(chance of incurring a loss, environmental degradation, etc.) associated
with climate has fuelled the urge to study the climate phenomenon and to
generate predictions in agriculture and the economic system. A growing
body of literature has examined and reviewed the nature and potential
impact of climate variation and climate change (Tobey, Reilly, and Kane,
1992; Easterling et al., 1993 and Rosenzweig et al., 1993). Nordhaus (1991)
and Cline (1991) have examined the monetary impact of climate change.
Studies on the impact of change in climate variables on agriculture include
Adams et al. (1990), Parry (1990), Reilly, Hohmann, and Kane (1994) and
Rosenzweig and Parry (1994). While uncertainties accompany predictive
climate modelling, increasing agreement on global climate trends arising
from varying methodologies in different climate centres is cause for
concern among agricultural and food policy planners. Climate variation
could have a variety of effects on both human and natural systems.
Together with changes in soil water availability, the increased occurrence
of climate fluctuation, climatic extremes, and crop diseases could lead to
an overall reduction in agricultural yields and serious food shortages.
The World Health Organisation considers long-term changes in global
climate variables as a serious threat to sustainable public health (WHO,
1990). Health experts expect a rise in climate-related diseases, such as heat
strokes and an increased incidence of vector borne diseases, such as
malaria, to accompany extremely wet and dry years. Other studies have
warned about the consequences of increased water shortages on food and
fibre production. A large unknown is the effect of climate change on
extreme weather events, such as droughts, floods, and storms.
By definition, climate variation refers to change in one or more climatic
variables (rainfall, temperature, wind, etc.) over a specified time (CDIAC,
1990). Long-term fluctuation in temperature, precipitation, wind, and
other aspects of the earth’s climate relates to climate change (ibid.). Farmers
Environment and Development Economics 531
and civil society are both vulnerable to impacts of change in climate vari-
ables on agricultural production. Agricultural decisions have complex
interactions with climate, but these decisions must be made months before
the impacts of weather are realized. Climate forecasts could make it poss-
ible to adjust farm decisions to reduce unwanted impacts and take
advantage of favourable conditions. Climate information can be grouped
into three categories: (i) Historical climate information (for example
average rainfall and its variability, (ii) information from real-time moni-
toring, and (iii) forecast information on future climate (seasonal,
year-to-year and long-term forecasts). The discussion in this paper is
limited to the impact of average rainfall variability on food security. Long-
term annual rainfall is projected by some General Circulation Model
(GCM) experiments to decrease for Sahel and Southern Africa (IPCC, 1997;
Hulme and Sheared, 1999) and expected to increase in the highlands of
east Africa and equatorial central Africa (IPCC, 1997).
3. The interplay of climate and food security in Southwestern
Cameroon
3.1. Agroecological setting
Southwestern Cameroon, situated between 5.15N and 9.15E, is largely
comprised of sub-humid agro-ecology. Climatologically, it is characterized
by moderate to high year-round temperatures and the weather is con-
trolled by equatorial and tropical air masses. There are two rainy seasons,
April–July and September–November with average daily temperatures of
25C. The Global Information and Early Warning System of the FAO
reports an increasing fluctuation in rainfall and a noted overall decline in
the last decade with precipitation reported to be below the mean monthly
average for 1961–1990 (FAO, 2000a). The rainy season is increasingly char-
acterized by shorter periods with torrential rainfall (ibid.). Abundant
precipitation was reported in August–September of 1996 as reaching
5,000 mm, above the expected maximum of 2,700 mm, and a high year
mean of 1,700 mm. In a region where much of the precipitation is associ-
ated with wave disturbances that give rise to strong convective activities
and short intense storms with peak intensities exceeding 100 mm/h,
intense rains result in soil compaction, soil erosion and high run-off. This
leads to physical and chemical deterioration of the soil in the region.
Fluctuation and seasonal variability in rainfall are constraints to crop pro-
duction. Weather problems have been exacerbated by soil physical
characteristics. In years with more than average rainfall, the low water
holding capacity of the soil results in water deficiency at critical phases in
crop development.
Thus, farming practices have developed as a response to environmental
and agroclimatic dictates. The agricultural sector employs about 80 per
cent of the population. Farming and food production systems are based on
the rudimentary rainfed semi-subsistence agriculture. Characterized by
shifting cultivation and fallow rotation involving multiple cropping and
integrated crop-livestock production systems, farming is the major source
of food and income for the households. Throughout the region, including
the marshy coastal creeks in the south and the rocky volcanic soils in the
532 Ernest L. Molua
center, land in the production systems is allocated principally to the pro-
duction of tubers (Dioscorea spp., Colocasia spp., Manihot esculenta), cereals
(Zea mays), legumes (Phaseolus vulgaris, Vigna spp.) and fruit trees (Musa
spp., Mangifera indica, Citrus spp., Artocarpus spp.). As in other regions of
Cameroon, the cereals provide 43 per cent of calories and 46 per cent of
protein of the average energy intake of 1980 kcal/day/person (FAO,
2000b). However, the highest market potential for the region are perennial
tree-based production systems. The rural economy largely depends on
plantation crops such as banana, rubber, oil palm, sugar cane, and medic-
inal products. Livestock such as goats and sheep as well as poultry are
common, while cattle are less so.
3.2. Review of vulnerability and food insecurity
During the last decade, managing for climate variability has become a very
significant feature in developed countries, in order to reduce the negative
consequences on the national economy. Skilful seasonal forecasts with
agricultural systems analytical framework have provided an opportunity
for systems managers to better tailor enterprise management decisions in
the seasons ahead. However, this has not been the case for most devel-
oping countries, especially Cameroon, where farmers rely on past
experience and where rainfall variability is one of the major sources of
income fluctuation among farmers and other primary producers.
The vulnerability or susceptibility of farm households to adverse conse-
quences of climate variables on food availability and accessibility in
Cameroon is examined with cross-sectional evidence obtained from 120
households in surveys carried out in 1996 and 1999, respectively in
Southwestern Cameroon. Three key components impeding food security
within the subsistence production structure are identified as (a) inade-
quate food production by farm households, (b) distribution and marketing
constraints, and (c) low household incomes and food procurement.
Inaccessibility to food arises from physical and economic constraints in the
region. Physical inaccessibility is due both to the inadequacy of production
and to the inefficiency of distribution systems including storage, trans-
portation, and marketing. Economic inaccessibility is due to the inability of
groups of farm households to establish entitlement over a requisite
amount of food. An indication of vulnerability is the inability of farm pro-
duction and storage to meet food consumption requirements (Molua,
1999). The challenge of improving food security in the region is heightened
by the increasing unpredictability of the nature of rainfall. Due to lack of
weather information and inability to accurately predict weather con-
ditions, the attendant variation in weather has resulted in yield and price
risk for farmers in the regions. Farm objectives increasingly aim to spread
such risk through diversification of resource use.
Irrigation is almost unknown in the region and a high incidence of pests
(for example stem borer, army worms, and the variegated grasshopper-
Zonocerus variegatus) and crop diseases further constitutes a real hazard to
expanding food production. This is further compounded by soil acidity
and associated aluminium toxicity. Furthermore, land tenure constraint is
gradually converting the traditional extensive fallow system to continuous
Environment and Development Economics 533
cropping. These factors have resulted in declining crop yields and soil
fertility. The vulnerability of farmers in the region may be exacerbated if
changes in the global climate occur as predicted by the world’s leading
atmospheric and oceanic scientists (IPCC, 1990). However, while there
appears to be a consensus on the nature of global climate change, the trend
and nature of this change at regional levels such as in Central Africa or
Southwestern Cameroon are not yet clear. Nonetheless, observed short-
term climate variations are already affecting farm-level food production
and development programs pursued by both the government and NGOs.
Farm-level adaptation is pivotal in translating climatic challenges and
agricultural response into changes in production, prices, food supply, and
welfare. Sampled farmers were asked about methods used to protect crops
from climatic constraints such as the stormy and torrential rain during the
rainy season and in the reported instances of long dry seasons. An exam-
ination of responses revealed that about 60 per cent adjusted or modified
their farming practices to suit the prevailing climatic constraints. Of this
group, about 90 per cent employed indigenous techniques and 10 per cent
employed improved soil and crop management technologies rec-
ommended by research teams at the Institute for Agronomic Research and
Development (IRAD) and the extension service of the provincial depart-
ment of agriculture. Of those farmers who reported adaptations to climate
constraints, about 40 per cent employed two or more methods including
the following: change in planting and harvesting dates, tillage and rotation
practices, substitution of crop varieties, increased fertiliser/pesticide appli-
cations, irrigation during dry spells, and the construction of drainage
systems to control water run-off (see table 1).
An increasing number of farmers rely on crops with shorter growing
seasons and integrate indigenous trees and plants such as Annona muricata,
Prunus africana and Pygeum africana in their food crop farms. The farmers
report that traditional technologies for conserving soil fertility (as depicted
in table 2), mixed cropping, and multiple cropping are aimed at buffering
the farming system against climate variability and increasing farm yield
and income.
The dearth of financial resources and inadequate funding of institutions
to promote farmer adaptations, the lack of credit schemes, as well as the
implementation of the ESAF and the resulting cut-back in government
534 Ernest L. Molua
Table 1. Types of climate-induced adaptation practices in the farming system of
Southwestern Cameroon
Methods of adaptation Percentage of farmers*
Change in planting and harvesting dates 40
Change in tillage and rotation practices 60
Substituting crop varieties 10
Increased fertiliser and pesticide applications 5
Employing irrigation techniques 2
Construction of drainage systems 20
Note: * Some farmers employed more than one adaptation method.
Source: Survey data 1999.
Environment and Development Economics 535
Table 2. Observed indigenous soil and water conservation practices in Southwestern Cameroon
Item Adopted practices Traditional/indigenous practices
Soil conservation Constructing broad bed and furrows. Establishing field bunds with waste weirs.
Constructing continuous contour bunds. Planting Crotalaria on field bunds.
Planting elephant grass strips on field boundaries.
Moisture conservation Contour farming. Normal tillage and interculture operations.
Soil mulching.
Water run-off control Constructing broad bed and furrows. Using farm debris (waste) to establish weirs.
Constructing field drains and boundary water-ways.
Gully control Using stone checks at regular intervals to Establish tree checks on boundaries to reclaim
stabilise gullies. gullies.
Water harvesting Percolating tanks. Diverting run-off from gullies to perennial crops.
Growing Eucalyptus spp. at farm boundaries.
Source: Survey data 1996.
support services (withdrawal of farm input subsidies, the elimination of
qualitative commercial bank lending aimed at farmers and reduction of
funding for research centres), have impeded research and extension
efforts, strained input intensive agriculture, and resulted in a dearth of
ancillary infrastructure to support food production. Hence only 20 per cent
of the farmers reported to have been visited by extension officers within a
two-year-period and none of the farm households purchased farm insur-
ance against crop failure. In light of attendant food insecurity, Colocasia
spp. and Manihot esculenta is grown by most households for the dual
purpose of cash and food, with cassava increasingly viewed as a security
crop as it is more resistant to adverse weather and tolerant to a certain
degree of soil acidity.
3.3. Challenges for food security in the region
The Committee on World Food Security defines three specific aims of food
security as: (1) ensuring adequate food production, (2) maximizing the
stability of food supplies, and (3) ensuring access to food particularly on
the part of those in greatest need (ECA, 1989). McCalla (1999) further reit-
erates the importance of access in terms of income, utilization, and
available food supplies. Enhanced productivity and profitability of agri-
cultural holdings would therefore be a major prerequisite to ward-off
temporary food insecurity in agriculture-dependent households.
However, farmers in Southwestern Cameroon are achieving yields of their
principal crops far below levels that have been demonstrated to be achiev-
able in research centres (Molua, 1999). Local farming depends entirely on
the quality of the rainy season. This makes the region vulnerable to inter-
annual climate variability. Three major factors are identified to impede
sustainable food production and increased farm income in the region:
1. Deforestation and soil loss through water run-off. Although detailed infor-
mation is still lacking on the erosivity of the rains and the erodibility of
the soil in the region, the risks of accelerated erosion are high. Land
impoverishment through soil erosion and over cultivation encourages
the loss of valuable top soil and therefore food production. The exten-
sion service department and NGO networks recognize the problem of
soil loss. Different types of scientific agroforestry techniques, coupled
with improved traditional methods of soil conservation are currently
being tested and pushed by extension workers for adoption by farmers.
Increasing deforestation and the associated increase in soil erosion has
prompted the establishment of the Mount Cameroon Project (MCP), a
multilateral initiative funded by the German technical cooperation
agency (GTZ), the UK Department for International Development
(DFID), the Global Environmental Facility (GEF) of the World Bank,
and the Government of Cameroon. The MCP discourages deforestation
and encourages reforestation through forest conservation and agro-
forestry promotion. Results of the impact of such an initiative on
curbing soil loss and improving crop production are still to be exam-
ined.
2. Climatic variability. Climate affects both social and natural systems in
536 Ernest L. Molua
the region through the occurrence of weather extremes and through
inter-annual climate variability. In the farming seasons of 1992/1993
and 1996/1997 (FAO, 2000a), the ability of farmers and rural house-
holds to grow enough to feed themselves was hampered to a large
extent by short rainy seasons (less than average rainfall) with torren-
tial down-pours. These weather patterns had impacts on agricultural
production, slashing crop yields and forcing farmers to adapt agricul-
tural practices in response to altered conditions. In seasons associated
with too much precipitation, farm to market roads are rendered
impassable thus influencing input and output supplies. In line with
Binswanger and Rosenzweig (1993) who examined the impact of risk
associated with the timing of rainfall on input choice, households in
Southwestern Cameroon attempt to meet consumption needs by
intensifying coping strategies. As the environment becomes more
risky, low income and vulnerable households shift production to
more conservative but less profitable modes. Increasingly, farmers
rely on past experience of climate variability to anticipate and respond
to fluctuations in the biophysical systems on which their livelihood
depends.
3. Lack of modern farming inputs. Recent scientific advances aimed at
improving the ability to predict major elements of climate variability
are not available to Cameroon’s farmers. Most of the farmers in the
region, as in most of sub-Saharan Africa (see Jagtap and Chan, 1998)
employ rudimentary non-scientific means of predicting large fluctu-
ations in rainfall. In addition, there is under-capitalization and low
financing of agriculture, with only 4 per cent of the farmers having
access to commercial bank loans. Family sources and local thrift and
loan societies are the major sources for finance. Mean gross margins are
estimated at about 300 US dollars per hectare per year for the family
farms, indicating that 60 per cent of the region’s farmers are unable to
afford modern inputs without government assistance. Given the pro-
duction constraints and attendant inability to produce enough food,
households in the region therefore increasingly tend to diversify their
income base into off-farm and non-agricultural activities in order to
afford imported food from neighbouring provinces. Nonetheless, with
food crop farming being the major occupation, there are obvious mon-
etary constraints in meeting both household food supply and the
purchase of modern farm inputs.
In general, climate variables together with insufficient fertiliser use, poor
control of insect pests and weeds, and non-availability of suitable crop
varieties are identified as major technical reasons for low productivity. The
dearth of weather forecasting services in the region and the lack of dia-
logue between the existing meteorological unit and the agricultural
services further compound the difficulties faced by farmers. Vulnerability
can be expected to increase as the population increases and more marginal
lands are brought into production. With precipitation being a key variable
to alter under long-term climatic change, the impact could be much more
severe on agriculture-dependent households in the region.
Environment and Development Economics 537
3.4. Significance of farm-level adaptation methods
Daily precipitation data are grouped into two critical phases of the growth
cycle of dominant food crops in the region. This includes weekly average
rainfall data, in cubic mm, for March–April (crop sowing) and for
May–August (crop growth) periods. Changes in tillage and rotation prac-
tices, and changes in planting and harvesting dates are identified as the
key adaptation methods employed by 60 per cent and 40 per cent of the
farmers, respectively, to offset the negative impact of climate variables.
Information on atmospheric and soil temperature, though necessary for
such an analysis is not included because of the unavailability of data.
Therefore, to enhance the future prospects for food security in the region,
the significance of adaptation options to climate variables are estimated in
order to guide recommendations for prudent policies and effective exten-
sion efforts.
3.4.1. The empirical model
The economic significance of climate variables and adaptation methods
is examined using an econometric approach. Since the data from the
study area include variable non-collinear prices and allow for the calcu-
lation of gross returns, this study uses farm income other than direct
farm output. This is based on the assumption that farmers aim to maxi-
mize farm returns and that their response to market conditions is
constrained by uncontrollable natural exogenous factors. A production
function would definitely estimate the significance of climate and adap-
tation variables on farm productivity. However, the approach is biased
because the production function fails to allow for potential economic
substitutions and for market effects as climatic conditions change.
Fluctuations in climate variables may favour one crop in place of
another.1Employing an income approach allows us to account for the
538 Ernest L. Molua
1Given farmers’ experience, stock of indigenous knowledge and local forecast
methods, when they perceive that temperature of rainfall conditions may change,
adaptive and profit-maximizing farmers may switch from crop A to another crop,
either to climate tolerant crop B or C, to enhance farm returns. In the absence of
price mechanisms, a production function would lead to pessimistic results. The
approach has a potential to prematurely conclude that climate variables such as
inadequate precipitation levels will depress farm output and hence overestimate
damage to farming households. It estimates that yield falls for crop A in poor
weather conditions. In reality, however, adaptive and profit-optimizing farmers
may no longer produce crop A or may reduce input allocation to crop A. Using a
farm income approach thus allows for the incorporation of the complementary
nature of the pricing mechanism (or market mechanism) to compensate for poor
yield, per se. It does not overestimate the damage to households from poor
weather conditions (see Mendelsohn et al., 1994). In addition, farm income is con-
cerned with the maximum value that a given input endowment can generate,
combining the effects of input and output prices. It allows, therefore, for aggrega-
tion and ease of computation in the multiple output production system of arable
farms in Cameroon. If markets are functioning properly, this approach provides
estimates of shadow prices of examined inputs and thus allows for the measure-
ment of the economic significance of the studied variables.
possibility of crop substitutions and market adaptation. By directly mea-
suring farm income, we account for the effect of climate variables on
yields of different crops, the indirect substitution of different adaptation
activities and inputs, as well as the variation in income needed for house-
hold consumption needs.
The farm income function is defined as, R(p,x) max{p.y:y Y(x,p0},
where yrepresents the vector of outputs, xrepresents the vector of inputs,
and pis the vector of output prices. The econometric equation is thus pre-
sented as a Cobb–Douglas (C–D) type function.2A C–D function is used
because it allows for theoretically correct and empirically meaningful
insights as an estimated primal production relation. The empirical model3
is thus specified as
lnR 0 1lnPC 2lnPL 3lnXLB 4lnXFT 5ln(RFG) 6Dmi
7Dmj
Where Ris the gross farm revenue per hectare,4 measured in Cameroon
CFA francs. PCand PLrelate to the price index for crops and livestock,
respectively. XLB and XFT denote the total labour used in man days5and
fertiliser, per hectare, respectively. RFG relates to precipitation during
growth. Evidence in Seleka (1999) reveals that rainfall during the growing
season is an important input in traditional arable agriculture. The RFG
variable allows us to evaluate the rainfall variable (water needed for crops)
Environment and Development Economics 539
2A C–D function is employed not only because of its simplicity and ease of esti-
mation and interpretation but because it also provides a first-order differential
approximation to the estimated dual function regardless of whether the C–D spec-
ification represents the true primal technology. This is exhaustively discussed in
Chambers’ (1994: 261) exposition of the ‘duality’ theory.
3The original Cobb–Douglas function prior to taking natural log is R
OPC
1PL
2XLB
3XFT
4RFG5 exp [(6Dmi) (7Dmj)]. A Box–Cox approach is used
to test for the linearity of the estimated equation and the linear specification is
rejected on the basis of a high 2test statistic.
4The income position of farming households and ability to produce and/or
command access to food is the key to understanding and developing policies to
mitigate food insecurity and vulnerability from changing climate. Farm income
may be determined by prices, weather, and technology. As farms attempt to com-
pensate for unfavourable weather conditions, additional capital items such as
fertiliser, insecticides, local farm operation methods, etc. are adopted. These are
associated with increasing costs of production. Farm income is thus a function of
output prices and some exogenous variables, for example climate. The theoretical
basis for this can be found in time-honoured production economics and farm
investment studies such as Hopkins and Murray (1953), Heady and Dillon (1961),
and Valdés, Scobie, and Dillon (1979).
5After land, labour is the second most important resource in farm production in
Cameroon. A mixture of family and hired labour is used for small-holder agricul-
ture. An estimated 2.7 per cent of farms in Southwestern Cameroon employ only
hired labour, 55.5 per cent use only family labour and 41.8 per cent combine both
sources of labour supply. Labour supply and demand in the region is affected by
other factors including public, traditional and religious holidays, and marketing
days (Molua, 1999).
by its seasonal fluctuations rather than the annual total.6Dmdenotes the
dummy variable to account for the adaptation methods employed by
farmers. is the random error disturbance term and are parameters to be
estimated. ln refers to natural logarithm. The dummy Dmi takes on the
value of 1 when farmers employ tillage and rotation practices as an adap-
tation method and 0 otherwise, while Dmjtakes the value of 1 when
changes in planting and harvesting dates are adopted as an adaptation
method and 0 otherwise. A priori it was expected that the signs of the par-
ameter estimate for both adaptation options would be positive, indicating
that they lead to an increase in farm return.
3.4.2. The empirical results
The coefficients of the parameters and the statistical test results7of inde-
pendent variables obtained from the regression analysis (estimated by OLS
method) are presented in table 3. Most of the signs of the independent vari-
ables fit the tested hypothesis. In line with a priori expectations, farm
revenue increased in terms of output prices of crops (PC) and livestock (PL).
The positive sign of the parameter estimate for farm-level adaptation
through (i) change in tillage and rotation practices (Dmi) and (ii) through a
change in planting and harvesting dates (Dmj), indicates that farm-level
adaptation methods positively correlate with higher farm returns.
As expected, the sign of the coefficient for precipitation (RFG) is positive,
indicating that an increase in precipitation during the crop growth period
is a positive covariate to income. This further suggests that irrigation in the
growth period, especially during dry spells, would be valuable for stimu-
lating a production increase. In a region in which agriculture is rain
dependent, an increase in precipitation coupled with improved tillage
practices could enhance farm economic return. This finding is in line with
Ozsabuncuoglo (1998) who established a functional relation between
540 Ernest L. Molua
6In Cameroon, up-to-date rainfall data are available at regional weather observa-
tories. This study, at best, needed village-level data. In this analysis, data are
collected by field research teams on daily rainfall during the growing season of
principal crops in each of the sampled villages. A rainfall index is then con-
structed for each village using the Kraus (1977) approach (see Landsea and Gray,
1992). It is estimated as follows: ri
j
rij, where riis the village-level mean
growing season rainfall; rij is daily rainfall data; and Jiis the number of months in
the growing season. The monthly variance is: i
2
j
(r2
ij ri
2). The nor-
malization of rainfall for each village per month (j) is given by Rij (rij i)i,
which essentially indicates the standard deviations from the mean for daily rain-
fall for the growing season. Therefore, the index value of rainfall (RFG) deviation
in each village during the growing season is defined as aj
i
Rij.
7A linear specification for the estimated equation is rejected in a Box–Cox
regression experiment. Therefore, the author concludes that there is a non-linear
relationship between farm income and the examined explanatory variables.
Hence, we can confidently or safely discuss the econometric results from the log-
linear empirical model specified in sub-section 3.4.1.
1
(Ji 1)
1
Ji
wheat production and climate variables in Southeastern Turkey, and
revealed that increments of rainfall during the growing period generates
higher productivity and economic return. There is, however, need for
caution. Given the revealed non-linear relationship between farm income
and rainfall, increasing rainfall may, therefore, to a certain extent be ‘bad’
for farms in the region, especially if it is accompanied by the already
reported erosivity of rains. Notwithstanding, the significance of RFG
explains the reliability of rainfall particularly at the critical growth phase
of crop development. It also further highlights much of the agricultural
potential and success in humid tropical regions.
While farm revenue is directly proportional to fertiliser usage (XFT), over
the examined time horizon, it was observed to decline with increasing use
of labour input (XLB), as revealed by the negative parameter estimate. This
is due to land tenure constraints in the region which allows land to behave
as a quasi-fixed input. An increase in labour would initially lead to an
increase in the production level, as expected. But a continuous per unit
increase in labour employment would lead to a reduction in the produc-
tion level, and farm income by 13 per cent. From the results obtained, it
could therefore be that farmers in the region are over-utilizing the avail-
able land resource, and their farm returns could be optimized by reducing
the amount of labour currently employed.
In general, the coefficient of multiple determination of the model and the
correlation coefficient reveal that the independent variables explain close
to 84 per cent of the variation in the dependent variable. In other words, 84
per cent of the variation in farm income can be attributed not only to
changes in output prices, but also to changes in rainfall during crop
growth, improved tillage and rotation practices and fertiliser usage as
well. The residual variation can be explained by some other variables not
included in the model. The t-statistic indicate that all the independent
Environment and Development Economics 541
Table 3. Regression model explaining farm income
Independent variables Coefficients Estimates t-values
Constant term 05.21 5.23***
Crop price (PC)10.25 2.98***
Livestock price (PL)20.32 2.13**
Labour (XLB)30.13 1.82**
Fertiliser (XFT)40.06 1.66**
Precipitation (RFG)50.38 2.44***
Adaptation methods
Tillage/rotation practices (Dmi)60.23 1.89**
Planting/harvesting dates (Dmj)70.15 2.31**
Adjusted R20.84
F-statistics(model) 26.35
N110.0
Notes: *** Represents significance at the 0.01 level, ** represents significance at
the 0.05, and * represents significance at the 0.10 level of probability testing
for a two tailed t-test. The non-included variable values were insignificant at
the three levels of the probability tests.
variables are statistically significant at the 95 per cent significance level
and, overall, the model is statistically significant at the 95 per cent prob-
ability level, as the F-values indicate.
Second-order tests for judging the goodness of the parameter estimates
provide valuable evidence about the overall validity of the model. The
null-hypothesis in the Goldfeld and Quandt tests for homoscedasticity
(Goldfeld and Quandt, 1965; Koutsoyiannis, 1977) could not be rejected,
implying heteroscedasticity is not a problem among the tested variables.
Furthermore, collinearity diagnostics, estimated correlation, and covari-
ance matrices revealed that multicollinearity is not a very serious problem
among the tested variables. This therefore allows for a discussion on the
implications of the findings obtained from the model.
4. Implications for food security in the region
Access to food through both production and exchange will continue to
depend not only on the productivity and profitability of agriculture, but
also on how well the political climate enables people to respond creatively
to their environment and prospects. In Southwestern Cameroon, an
increase in crop and livestock prices and fertiliser by one percentage point
accounts for about 25 per cent, 32 per cent, and 6 per cent of farm income,
respectively. The employment of farm-level adaptation and improvement
in climatic conditions enhance farm income by 23 per cent, 15 per cent, and
38 per cent, respectively. This would imply that inadequate rainfall, infer-
tile soils, and underdeveloped marketing channels could make for a risky
agricultural environment and low incomes. From the analysis, changes in
rainfall pattern during the growing season alter the economic potential for
agriculture and food security. A 38 per cent response of farm income to
rainfall indicates that the vulnerability of the agricultural sector to varying
lengths and intensities of rainfall fluctuations is clearly high in a region in
which agriculture and farm incomes are currently affected by weather
variables, in which market infrastructure and institutions to facilitate dis-
tribution of deficits are inadequate, and in which per capita income is low
and the ability of farmers to adapt is limited.
Not only does short-term climate variability in the study area constrain
production, distribution and consumption of food and household entitle-
ments are threatened as well, thus contributing to temporary food
insecurity. Sen (1981) asserts that a household’s food entitlement consists
of the food that the household can obtain through production, exchange or
extra-legal legitimate conventions. Since households in the region are
highly dependent on agriculture for employment, incomes remain low and
impede their ability to buy food imported from neighbouring provinces.
Though the observed entitlement failure and temporary food insecurity
could be corrected through stockpiling, the dearth of storage and transport
infrastructure impedes the ability to buy and procure food items. A sub-
stantial part of farm produce is lost between the farmer and the consumer,
with losses greater in perishable farm produce such as vegetables, fruits,
and animal products. In the face of inter-annual and intra-seasonal rainfall
variation, the challenges of agricultural production and food insecurity in
542 Ernest L. Molua
Cameroon are clearly enormous and policy makers will be called upon to
design and implement prudent policies if these challenges are to be met.
Overall, the significant response of farm income to rainfall presents a
clear indication that irrigation investments in the region would generate
added value to farm production. This, nevertheless, must be accompanied
by the provision of other agricultural improvement measures that would
ensure the improvement of the procurement system, food marketing,
raising farm incomes, and food endowment. The attainment of sustain-
able food security would therefore have to be anchored around changes
in the present structure of food production and employment so that
more food is produced by farmers employing appropriate agricultural
technologies.
5. Policy recommendations and conclusion
Cameroon is highly dependent on rainfed farming and therefore short-
term fluctuations in weather patterns have significant impacts on farm
income. Farmers adjust to short term climatic anomalies, within given
economic and technological constraints. In order to assist households and
the farming society of the study region to better cope with climate vari-
ability and potential long-term climate changes, government policy must
address and increase farmers’ prospects for better adaptive responses.
Agricultural services and inputs such as fertilisers must be made available
and land tenure constraints relaxed in order to ease the diminishing
returns experienced in farm labour employment. Sustained increases in
farm household income require increased provision of farmer support ser-
vices and enactment of an agricultural insurance scheme to enhance
adaptation to variable climate conditions.
Further effort should be made towards the development and strength-
ening of institutional capacity for mitigating the impacts of erratic weather
conditions. This should be linked with increased farmer access to appro-
priate technologies, climate information, measures to improve the
marketing and distribution networks (marketing channels), and access to
credit facilities. Access to credit and micro-finance has the potential to
smooth income and consumption of poor households (Zeller, 1999).
Improving the procurement and distribution system would entail
boosting local production via the reduction of pre- and post-harvest losses
and the development of adequate infrastructures such as farm-to-market
roads.
In view of the robust and significant influence of the rainfall variable
and the adaptive response, in order to enhance the effectiveness and
resilience of farm-level adaptation, regular information exchange meetings
and consultations would have to be organized between the weather fore-
casting stations, IRAD, the agricultural extension service and farmer
associations. The Ministry of Agriculture, NGOs, and the agronomic
research and climate centres should effectively monitor current climate
and environmental conditions in the region and establish an effective early
warning system. The potential for substantial increases in agricultural pro-
duction and farm incomes exist if farmers are helped to adapt to climate
variability.
Environment and Development Economics 543
References
Adams, R.M., C. Rosenzweig, R.M. Peart, B.A. McCarl, J.D. Glyer, R.B. Curry, J.W.
Jones, K.J. Boote, and L.H. Jr. Allen (1990), ‘Global climate change and US agri-
culture’, Nature 345: 219– 224.
Binswanger, H. and M. Rosenzweig (1993), ‘Wealth, weather risk and the compo-
sition and profitability of agricultural investments’, Economic Journal 103: 56–78.
CDIAC (Carbon Dioxide Information Analysis Center) (1990), ‘Carbon dioxide and
climate glossary, ORNL/CDIAC-39, Oak Ridge National Laboratory, Tennessee.
Chambers, R.G. (1994), Applied Production Analysis: A Dual Approach, Cambridge:
Cambridge University Press, pp. 250–301.
Cline, W.R. (1991), ‘The economics of greenhouse effect’, Economic Journal 101:
320–337.
Downing, T.E. (1992), ‘Climate change and vulnerable places: global food security
and country studies in Zimbabwe, Kenya, Senegal and Chile’, Discussion Paper,
Environmental Change Unit, University of Oxford, Oxford.
Easterling, W.E. III, P.R. Crosson, N.J. Rosenberg, M.S. McKenney, L.A. Katz, and
K.M. Lemon (1993), ‘Agricultural impacts of and response to climate change in
the Missouri–Iowa–Nebraska–Kansas (MINK) Region’, Climatic Change 24: 23–61.
ECA (Economic Commission for Africa) (1989), ‘The challenges of agricultural pro-
duction and food security in Africa’, in The Challenges of Agricultural Production
and Food Security in Africa, Report and Proceedings of an International Conference
by the African Leadership Forum, 27–30 July 1989, Ota, Nigeria, pp. 65–99.
EIU (Economic Intelligence Unit) (1999), ‘Cameroon’, Country Report, Second
Quarter, Economic Intelligence Unit, 15 Regent Street, London.
FAO (Food and Agricultural Organisation) (2000a), ‘Cameroon: climate’, Global
Information and Early Warning System, FAO, Rome.
FAO (Food and Agricultural Organisation) (2000b), ‘Cameroon: agriculture’,
Commodities and Trade Division, FAO, Rome.
Goldfeld, S.M. and R.E. Quandt (1965), ‘Some tests for homoscedasticity’, Journal of
American Statistical Association 60: 539–547.
Heady, E.O. and J.L. Dillon (1961), Agricultural Production Functions, Ames: Iowa
State University Press.
Hopkins, J.A. and W.C. Murray (1953), Elements of Farm Management, 4th edition,
New York: Prentice-Hall Inc., 445 pp.
Hulme, M. and N. Sheared (1999), ‘Climate change scenarios for Zimbabwe’,
Climate Research Unit, University of East Anglia, Norwich, 6 pp.
IPCC (Intergovernmental Panel on Climate Change) (1990), Climate Change: The
IPCC Scientific Assessment, Report from Working Group 1, Cambridge:
Cambridge University Press.
IPCC (Intergovernmental Panel on Climate Change) (1996), Climate Change, 1995:
Impacts, Adaptations and Mitigation of the Climate Change, Cambridge: Cambridge
University Press.
IPCC (Intergovernmental Panel on Climate Change) (1997), ‘The regional impact of
climate change: an assessment of vulnerability’, A Special Report of IPCC Working
Group II: Summary for Policy Makers. R.T. Watson, M. Zinyowera, R.H. Moss, and
D.J. Dokken (eds), Published by the Inter Governmental Panel on Climate
Change, New York.
Jagtap, S.S. and K.A. Chan (1998), ‘Agrometeorological aspects of agriculture in the
sub-humid and humid zones of Africa and Asia’, Agriculture and Forest
Meteorology 10: 59–72
Koutsoyiannis, A. (1977), Theory of Econometrics, 2nd edition, Hampshire:
Macmillan Education, 681 pp.
Kraus, E.B. (1977), ‘Sub-tropical droughts and cross-equatorial transports’, Monthly
Weather Review,105: 1009–1018.
544 Ernest L. Molua
Landsea, C.W. and W.M. Gray (1992), ‘The strong association between western
Sahel Monsoon rainfall and intense Atlantic hurricanes’, Journal of Climate 5:
435–453
McCalla, A.F. (1999), ‘Prospects for food security in the 21st century: with special
emphasis on Africa’, Agricultural Economics 20: 95–103.
Mendelsohn, R., W. Nordhaus, and D. Shaw (1994), ‘The impact of global warming
on agriculture: a Ricardian analysis’, American Economic Review 84: 753–771.
Molua, E.L. (1999), ‘Economic optimisation of smallholder agroforestry systems in
Southwestern Cameroon’, M.Sc. Dissertation, Department of Economics and
Natural Resources, The Royal Veterinary and Agricultural University
Copenhagen, Denmark, 157 pp.
Nordhaus, W.D. (1991), ‘To slow or not to slow: the economics of greenhouse
effect’, Economic Journal 101: 920– 937.
Ozsabuncuoglu, I.H. (1998), ‘Production function estimation for wheat: a case study
of Southeastern Anatolian Project (AP) Region’, Agricultural Economics 18: 75–87.
Parry, M.L. (1990), Climate Change and World Agriculture, London: Earthscan
Publications.
Reilly, J., N. Hohmann, and S. Kane (1994), ‘Climate change and agricultural trade:
who benefits, who loses?’, Global Environmental Change 4: 24–36.
Rosenzweig, C. and M.L. Parry (1994), ‘Potential impact of climate change on world
food supply’, Nature 367: 133–138.
Rosenzweig, C., M.L. Parry, K. Frohberg, and G. Fisher (1993), ‘Climate change and
world food supply’, Environmental Change Unit, University of Oxford, Oxford.
Seleka, T.B. (1999), ‘The performance of Botswana’s traditional arable agriculture:
growth rates and the impact of the accelerated rainfed arable programme
(ARAP)’, Agricultural Economics 20: 121– 133.
Sen, A. (1981), Poverty and Famines: An Essay on Entitlement and Deprivation, Oxford:
Oxford University Press.
Tchoungi, R., S. Gartlan, J.A. Mope Simo, F. Sikod, A. Youmbi, and M. Njatsana
(1996), ‘Case study for Cameroon’, in D. Reed (ed.), Structural Adjustment, the
Environment and Sustainable Development, London: Earthscan Publications:
Chapter 3, pp. 53–81.
Tobey, J., J. Reilly, and S. Kane (1992), ‘Economic implication for global climate
change for world agriculture’, Journal of Agricultural and Resource Economics 17:
195–204.
Valdés, A., G.M. Scobie, and J.L. Dillon (1979), Economics and the Design of Small-
Farmer Technology, Ames, Iowa: Iowa State University Press, 211 pp.
WHO (World Heath Organisation) (1990), Potential Health Effects of Climate Change,
Geneva.
Winters, P., R. Murgai, A. de Janvry, E. Sadoulet, and G. Frisvold (1999), ‘Climate
change and agriculture: effects in developing countries’, in G. Frisvold and B.
Kuhn (eds), Global Environmental Change and Agriculture, Edward Eldgar
Publishing, pp. 239–296.
Wolfe, D.W. (1995), ‘Potential impact of climate change on agriculture and food
supply’, Proceedings of Sustainable Development and Global Climate Change:
Conflicts and Connections, the Centre for Environmental Information, 4–5
December.
Zeller, M. (1999), ‘The role of micro-finance for income and consumption
smoothing’, Paper presented at the Conference on Social Protection and Poverty,
organized by the Sustainable Development Department of the Inter-American
Development Bank, Washington, DC, February 4–5.
Environment and Development Economics 545
