MODELING THE IMPACT OF CLIMATE VARIABILITY ON LOCAL FOOD PRICES - Food Security, Food Prices and Climate Variability - page 137

local grain markets in Sub-Saharan Africa are largely unaffected by world grain prices while
rice markets are typically affected by changes in world rice prices, a result also supported by
Brown et al. (2012). Using a model that incorporates NDVI as one of its drivers was able to
show that 87 out of the 179 locations were influenced by local weather anomalies during the
2003-12 period. It is likely that these areas would benefit the most from interventions that
strengthen the food supply chain and the resulting improvements may help mitigate the
uncertainty arising from weather anomalies (Kshirsagar, 2012).
Price-NDVI model conceptual framework
In a market with functioning information flows, efficient storage and transport infrastructure,
food price changes from one month to the next will mainly reflect differences in the cost of
transport and other transaction costs. As described by Hayek (1948), market prices will adjust
through a dynamic process. Traders and intermediaries would anticipate both a shortfall in the
food deficit location and the resulting increase in local prices. Consequently, they will choose to
buy more from the surplus region and sell more at the deficit location until their profit from the
extra food sold has been maximized, or the demand satisfied. Therefore, a local weather shock
may exert a temporary influence on local price in a chronic food deficit market, but in equilib-
rium price differences will reflect differences in transport costs because traders will move goods
to take advantage of the increase in prices in a market. As a result, local food prices in food deficit
regions will be determined by world prices if the country trades with the others in the world
market, producer prices, tariffs and other market costs, and transport costs (Hayek, 1948).
How might this mechanism fail? There are at several factors that might cause the price
forming mechanisms to diverge from the Hayekian version of “perfect” markets. First, and
perhaps most importantly, ad hoc government policies introduce a great deal of uncertainty
in the expected returns for traders. As Tschirley and Jayne (2010) have argued, this factor was
responsible for Southern Africa's major food crisis in 2008-09. Second, credit constraints, and
credit market imperfections more generally, impede the ability of traders to exhaust arbitrage
opportunities (Kshirsagar, 2012).
Aker et al . (2010) showed that ethnic differences in West Africa are responsible for sub-
national market imperfections and argued that this is partly driven by inter-ethnic trading
frictions. The larger the changes in the quantity and geographic scope of possible trading
opportunities, the more important credit constraints become for traders working in these
markets. Traders cannot access sufficient amount of goods due to lack of credit to make
moving goods from one market to another profitable, particularly when moving between
regions with different ethnic composition.
Weather-related production volumes also make a big difference in the ability of traders to
move goods profitably. Zant (2013) showed that maize markets in Malawi functioned very
differently during periods of average or above average production than during times of below-
average production due to constraints in the ability of traders to move goods. Variations in
production can transform a region from a food surplus to a food deficit region, entirely
changing the trade direction. Because transaction costs represent a dominant component of
market prices, especially in the case of low value, high volume staple food with low produc-
tion costs such as manioc, millet or yams, incomplete or missing transportation or transaction
costs will greatly change a price analysis and lead to inaccurate conclusions (McNew and
Fackler, 1997; Baulch, 1997). In developing countries, lack of complete data on trading or