4
Table 1 - Test of the transmission of international maize and millet prices on domestic price,
OLS estimation on monthly data (January 2000- December 2010)
Panel A- Malawi
Dependent variable:
maize price in Malawi (in US $)
Variable
Coefficient
Std. Error
t-Statistic
Prob.
Constant (
)
-0.438
0.196
-2.234
0.0272
maize price in Malawi -1 (in $) ( )
-0.091
0.025
-3.619
0.0004
world (USA) maize price -1 (
0.058
0.035
1.629
0.1056
world (USA) maize price -1 ( )
0.198
0.162
1.221
0.2243
maize price in Malawi -1 (in $) ( )
0.558
0.072
7.665
0.0000
Adjusted R-squared
0.332 Durbin-Watson stat
1.873
F-statistic
17.028 Prob (F-statistic)
0.000
N. obs
130
Notes: the value of is equal to 0.058/0.091 = 0,638; it shows that around 60 percent of the international
price will be transmitted onto the Malawian price in the long run. -1 = one month lag
Panel B- Niger
Dependent variable:
millet price in Niger (in US $)
Variable
Coefficient
Std. Error
t-Statistic
Prob.
Constant (
)
0.517
0.235
2.199
0.0297
millet price in Niger -1 (in $) ( )
- 0.169
0.049
- 3.443
0.0008
world cereal price -1 (
)
- 0.102
0.046
- 2.231
0.0274
world cereal price -1 ( )
0.542
0.155
3.493
0.0007
millet price in Niger -1 (in $) ( )
0.566
0.109
5.193
0.0000
Adjusted R-squared
0.212 Durbin-Watson stat
1.8892
F-statistic
8.406 Prob (F-statistic)
0.0000
N. obs
130
Note: the value of is equal to –0.102/-0.169 = 0.603; it shows that around 60 percent of the international price
will be transmitted onto the Niger price in the long run. -1 = one month lag.
Source: authors’ calculation on FEWSNet and IMF data for the commodity world prices and the MWK/U$
exchange rate; INFOEURO and FXTOP for the CFA/U$ exchange rate; FEWSNet for the maize price in Malawi; and
SIMA for the millet price in Niger.
Thus, while the relation between international and domestic food prices is not as direct as it
is generally assumed, in some countries there is evidence that domestic food prices are
influenced by: (i) the long term trend in domestic production and productivity per capita and
the efficiency of domestic food markets, (ii) seasonal price fluctuations (stemming from intra-
annual climatic variations and market imperfections), and (iii) large food crises and famines.
Often, the silent increase in malnutrition and mortality due to these factors attracts less
attention than the shocks caused on occasion of surges in international food prices. While
increases in domestic food prices linked to food imports reduce the food purchasing power
of the of the urban low and middle classes and may therefore raise poverty rates (and trigger
food riots), their impact on malnutrition and mortality is unlikely to be necessarily more
important than that due to the three types of price changes discussed in this paper. In
5
contrast, limited progress in these three areas is a main factor explaining the slow progress
recorded in achieving the MDG on child malnutrition and child mortality
4
.
In view of the above, this paper focuses on these three sources of food price changes, each of
which requires policy responses which differ among each other and from those needed on
occasion of rises in international food prices. The analysis covers the last decade and
concentrates on Niger and Malawi, two countries which experienced during this period a
marked long run volatility of food prices, persistently large seasonal price fluctuations, and
acute food price spikes on occasion of the food crises of 2005 and 2010 in Niger, and of 2002
and 2008/9 in Malawi. Purpose of this paper is therefore to analyze the importance of each of
these three factors, and their relative contribution to overall domestic price volatility and
changes in child malnutrition
5
over the period considered.
2. Framework to analyze food price volatility and its impact on food security
In this paper we concentrate on household and individual food security, a concept which is
more useful than that of national food security in explaining the levels and variations in
malnutrition. In this approach, food security is reached when the value of the individual’s and
household’s entitlements meet the cost of the household’s energy requirements. One of the
core component of this definition is the concept of ‘entitlements’ introduced by Sen (1981) as
a set of alternative commodity bundles a person can command in a society, using its totality
of rights and opportunities. Sen distinguishes ‘trade entitlements’ (what a person can buy
with the commodities owned), ‘production-based entitlements’ (the right to own what a
person produce with her resources), ‘own-labour entitlements’ (what a person can buy
selling the labor power owned), and ‘inheritance and transfer entitlements’ (the right to own
what it is willingly given by others, including state institutions). In addition, over the short
term, food security depends also on the monetary value of the assets the household decides
to sell to purchase food. Thus, the food entitlements bundle – and food security - of
household group h (h=1,..k) can be written as follows:
Pf
A
Pf
T
Pf
Pw
Qw
Pf
Pj
Qj
Qfsc
Qf
h
h
h
h
h
h
∆
+
+
+
+
=
(2)
where
h
Qf
,
h
Qfsc
,
h
Qj
, and
h
Qw
are the total amount of food available to household group
h, the amount of food produced for self-consumption, the amount of goods (such as animals,
cash crops, firewood, etc.) sold and the amount of casual wage labour performed. In turn,
Pf
,
Pj
and
Pw
are the prices of food, the price of goods sold on the market and the wage rate,
Th
the nominal value of transfers received (including in the form of food aid, cash transfers
on occasion of cash/food-for work, and cash obtained from begging), and
h
A
∆
the income
obtained from the sale of household assets.
Pf
Pj
and
Pf
Pw
are the ‘terms of trade’ between the
prices of goods exchanged and that of food, and between the wage rate and the price of
4
Malnutrition is a key factor in the 3.5 million deaths among children of less than 5 years of age recorded every
years.
5
In this paper we focus on child malnutrition as children are especially vulnerable to changes in food intake, as the
incidence of child malnutrition helps capturing aspects of the distribution of wellbeing within the households not
adequately reflected in other indicators, and as most nutritional data refer to this age group. In a sense, child
malnutrition is the most important indicator of hunger in Africa.