45
500-800 children a months, it reached the level of around 1500 in early 2008 and of 2000 in
early 2009.
Regression analysis of child admissions to feeding centres, 2003-2009
In contrast to the case of Niger, data on the admissions of children to feeding centres in
Malawi seem to be heavily influenced by the number of feeding centres in operation and the
availability of food and therapeutic nutrients to be distributed among the children. This
explain for instance why the number of admissions to feeding centres hardly budged on
occasion of the 2001-02 crisis and 2010 price increase. This is why, since no meaningful
regression analysis of the relation between child admissions to feeding centres and maize
prices could be carried out for the famine years of 2001-02, the analysis below focuses only
on the 2003-2009 years. Likewise there are no data on the exchange entitlements as there are
no available time series on
ganyu wages, and the prices of poultry, small animals and other
goods sold by smallholders during the hunger season. With the limited data available, we
were therefore simply tested in a log-log framework (Table 13) the hypotheses about the
impact of maize prices, number of feeding centers, time trend, and hunger season on the
average number of children admitted to feeding centre over 2003-09.
Table 13 - Malawi: Log-log regression analysis of the impact of the maize price on child
admissions to feeding centers (2003-09)
Model 1
Model 2
Constant
3.697***
4.391***
Log maize price
0.670***
0.275**
N feeding centers
0.030***
0.030***
Time trend
-0.020***
-0.012***
Hunger season
0.613***
Adj R2
0.62
0.77
F statistics
45.42***
71.23***
Durbin Watson
1.13
1.70
N.obs
84
84
Source: authors’ calculations on official data
Table 13 confirms that the price of maize affects in a statistically significant way the number
of children admitted to feeding centers though the value of the parameter (elasticity) is
smaller than that of millet in Niger (see Tables 9 and 12). The same is true for the number of
feeding centers, while the opposite holds for the hunger season. In addition, the time trend
has negative and significant parameters suggesting there are some endogenous factors (e.g.
medical improvements) which gradually reduce the admissions of children to feeding
centers.
Lessons from Niger’s 2005 and 2010 and Malawi’s 2002 and 2008-9 food crises
(i) Niger’s 2005 and Malawi’s 2002 crisis. These crises were very similar. In both countries: (a)
the initial trigger was a drop in food output (causing a - food availability decline, FAD - crisis)
which followed however several good harvests, so that food output was not much below the
long term trend; (b) food imports – which the structural adjustment policies of the 1980s-
46
1990s were assigned a central role in ensuring food security – failed because of Niger’s and
Malawi’s landlocked-ness, remoteness, poor transport infrastructure and rise of food prices in
their regions caused by covariant weather
shocks. In the end, the sharp rises in domestic food
prices were largely due to a failure of imports which could not cap domestic prices and cover
the domestic food shortfall; (c) in both cases whatever food surpluses existed in their region
were diverted to larger and richer neighbours, including because of the exchange rate
appreciation of their currencies (MF, regional market failure, crisis). Short term food security
in tiny countries may thus be affected by exchange rate variations driven by export
bonanzas, expansionary policies, fast growth and so on in neighboring large economies; (d)
at least in the case of Niger, inefficiencies in the domestic market raised the differentials
between the producer prices paid to smallholders and consumer prices which benefitted a
minority of large farmers and – especially - the few large oligopolistic traders who dominated
the national market (MF, regional market failure, crisis); (e) in both cases, public policy and
donor response were grossly inadequate (FID, food intervention decline, crisis). Both
governments denied for long the existence of a famine and intervened with considerable
delay and reluctance when deaths by starvation, and a surge in admissions of malnourished
children to feeding centres had risen out of control. The responses of both governments
were however inadequate as their food safety reserves had declined due to the new policy of
relying on food imports to cap domestic prices. Also, the interventions of both governments
were poorly targeted, with the result that in Niger only 72 percent of the poor received food
aid as compared to almost 80 percent of the non poor (World Bank 2009).
Niger’s 2005 versus 2010 crises
A comparison between these two crises shows that many lessons were learned from the
2005 crisis: (a) in both crises the initial trigger was a decline in food production caused by
adverse climatic conditions (FAD crisis). Niger’s fragility to climatic shocks and volatility of
food production is chronic and affects food security and child malnutrition in both famine
and non-famine years. Long term food production policies do not seem to have been able to
reduce output and price volatility and Niger is thus likely to have to face a food emergency
every three-four years. This calls for changes in policies on long term food production and
population growth; (b) while failing in 2005, food imports increased rapidly during 2010 (as
other Sahelian countries were not hit by the drought) (no regional MF). As a result millet
prices rose only modestly during the 2010 lean season, while the prices of exchange
entitlements (e.g. cows and goats) declined less than in 2005. Therefore, the exchange
entitlements of smallholders and herders Qj
h
/Pf fell by only 10-20 percent as opposed to 50-
60 percent in 2005; (c) it is unclear whether also in 2010 the ratio producer/consumer prices
deteriorated; (d) in 2010 the public policy response was substantially better. In 2005 the
failure of the Early Warning System (EWS) - which monitored only food production and prices
- led the Government to develop in 2007 a national contingency plan for food security and
nutrition which included a more comprehensive EWS providing information not only on
cereal output and prices, but also on imports, the nutritional status of the population and
vulnerability of rural households. Thanks to this information, and – especially – to the greater
political will, in 2010 the Government immediately drafted an emergency plan and called for
international aid which was delivered in a more timely fashion than in 2005
20
. The domestic
20
In 2005 the Government requested food aid for 78.000 tons against an estimated deficit of 505.000 tons, while
the donors largely ignored the first appeals by WFP and the government in late 2004. The donor contributions
started to arrive 8 months after the first request, when the cost to save one child’s life had risen from 1 to 80 USD