47
food security response was also far better than in 2005. Indeed, in December 2009, the Food
Security Fund stood at nearly 43.000 tons, and the National Security Stock at 22.000 tons, still
below the target fixed earlier on (FewsNet, Niger Food Security Update, January 2010), but
substantially higher than the 5.000 tons available in 2005. (e) The specific measures to
respond to the crises also differed. In 2005, the response basically consisted in carrying out
subsidized sales of millet at about half the market price (100 FCFA per Kg.), a price still out of
reach for the poorest. The 2010 response consisted instead of an immediate free food
distribution of 52000 tons to over one million beneficiaries which were targeted more
precisely than in 2005, subsidized sales of 60.000 tons of cereals to vulnerable households,
cash for work for 23.000 households, cereal banks and the distribution of 18.000 tons of
forage in the most affected areas; (f) Despite this better and more timely response, the
number of children admitted to feeding centres rose as fast, or faster, than in 2005. The
econometric analysis carried out in this paper has shown, however, that the huge increase in
admissions of malnourished children was due to a considerable extent to the trebling of the
number of feeding centres and the adoption of a new nutritional protocol which includes
among the population at risk also children affected by moderate malnutrition, placing the
emphasis on the prevention of malnutrition and not only on its treatment. Clearly, by 2010,
child malnutrition had become – at last - a key policy issue.
Malawi’s 2002 and 2008-9 crises
The two crises share a number of similarities but at the same time differ markedly in several
respects: (a) while in 2002 maize production declined by 34 percent (FAD crisis), in 2008 and
2009 Malawi recorded bumper crops which led the government to declare a large maize
surplus and issue export licenses to private traders; (b) the 2002 crisis can be represented as
an ‘import crisis’ as the drop in domestic output was exacerbated by the decline of maize
imports for the reasons discussed above, while that of 2008-9 was an ‘export crisis’, caused by
the exportation of 400.000 tons of maize to neighboring countries and a subsequent price
escalation which was exacerbated by a tripling of fertilizer prices; (c) in both cases there were
‘informational failures’ (FID crisis) as the government and ADMARC massively over-estimated
the output of cassava and tubers. These “ information failures” explain why the government
was slow to order food imports in 2002 and allowed food exports in 2008-9. In addition, the
2002 market operations of ADMARC were slow and problematic and failed because the prices
offered consistently trailed behind the rise in market prices. ADMARC’s market interventions
in 2008 were equally confusing, led to a sharp rise in maize prices and barred private traders
to import maize from South Africa at lower prices; (d) the response to the 2001-2 crisis was
very slow and inadequate, and only in February 2002 the government declared a national
food crisis, while the donors response was delayed. I
n both 2002 and 2007/8-2008/9, the
nutritional impact of the crisis is not well documented. Yet, there is evidence that the number
of malnourished children admitted to feeding centers doubled in relation to the prior year.
a day. In contrast, two thirds of the emergency appeal launched in April 2010 were delivered as of September
2010.
48
7. Comparative impact of long term, seasonal & famine price changes on child
malnutrition
This section aims at disentangling the relative impact of the three different price components
(long term, seasonal, famine/food crises) discussed in the prior sections on child malnutrition
over the decade 2000-2010, though data limitations on the number of child admissions to
feeding centers force us to limit our econometric test to 2003-9 in the case of Malawi and
2006-10 in that of Niger. To do so, we first decompose the overall monthly millet and maize
prices into three components, i.e. (a) a trend price component
,
(b) a seasonal price
component, and (c) famine price component (Figures 24 and 25). To carry out such
decomposition, we relied on the ‘multiplicative method’
21
. We already extracted the 2000-
2010 price trend from the observed time series of maize/millet prices (see Figures 6 for Niger
and 11 for Malawi). We then decompose with the multiplicative method the difference
between price time series and its trend (see footnote 30 for the method used in the
decomposition) into a seasonal price component, a famine price component, and a residual
term. Figure 23 presents the results of the decomposition into the seasonal price and famine
price component for Niger (the residual term is ignored), and Figure 24 does the same for
Malawi.
Figure 23 - Niger: seasonal and famine price components - monthly data (2000-2010)
0.7
0.8
0.9
1
1.1
1.2
1.3
1.4
J
a
n
-0
0
J
u
l-
0
0
J
a
n
-0
1
J
u
l-
0
1
J
a
n
-0
2
J
u
l-
0
2
J
a
n
-0
3
J
u
l-
0
3
J
a
n
-0
4
J
u
l-
0
4
J
a
n
-0
5
J
u
l-
0
5
J
a
n
-0
6
J
u
l-
0
6
J
a
n
-0
7
J
u
l-
0
7
J
a
n
-0
8
J
u
l-
0
8
J
a
n
-0
9
J
u
l-
0
9
J
a
n
-1
0
J
u
l-
1
0
in
d
e
x
n
.
Seasonal price component
Famine price component
Source: authors’ calculation on SIMA data.
21
The monthly time series (
m
), from 2000 to 2010, of the domestic price (
m
,
c
P
) of the commodities (
c
) taken
into consideration, maize for Malawi and millet for Niger, have been disaggregated into four components: (i) A
long term trend price component (
m
,
c
T
), (ii) a seasonal price component (
m
,
c
S
), (iii) a famine price component
(
m
,
c
F
) and residual random component (
m
,
c
R
) (Harvey, 1990). Examination of the graph for trend and seasonal
components and the F-tests for seasonality, have suggested that these effects interact, to generate the observed
time series, according to a multiplicative model specified as following:
m
,
c
m
,
c
m
,
c
m
,
c
m
,
c
R
*
S
*
F
*
T
P
=
. These
components were estimated with two methodologies. The X-12 Monthly Seasonal Adjustment Method (Findley et
al., 1998) has allowed to distinguish the seasonal and the random effects separately and the long term
components in the aggregate. The famine component is often of irregular length and difficult to estimate due to
the economic assumptions required. In order to overcome these issues, the long term trend component has been
calculated with the Hodrick-Prescot Filter and then subtracted from the above mentioned aggregate for the
estimate of the famine price component.