Jump to ContentJump to Main Navigation
Food Price Policy in an Era of Market InstabilityA Political Economy Analysis$

Per Pinstrup-Andersen

Print publication date: 2014

Print ISBN-13: 9780198718574

Published to Oxford Scholarship Online: January 2015

DOI: 10.1093/acprof:oso/9780198718574.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2019. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see www.oxfordscholarship.com/page/privacy-policy).date: 27 May 2019

International to Domestic Price Transmission in Fourteen Developing Countries during the 2007–8 Food Crisis

International to Domestic Price Transmission in Fourteen Developing Countries during the 2007–8 Food Crisis

Chapter:
(p.21) 2 International to Domestic Price Transmission in Fourteen Developing Countries during the 2007–8 Food Crisis
Source:
Food Price Policy in an Era of Market Instability
Author(s):

Kenneth Baltzer

Publisher:
Oxford University Press
DOI:10.1093/acprof:oso/9780198718574.003.0002

Abstract and Keywords

This chapter synthesizes the evidence on price transmission from international maize, rice and wheat markets to domestic markets in fourteen developing countries during the global food crisis in 2007–8. A great variation in the price transmission patterns is observed; from almost no price pass-through in China and India, over close relationship between international and domestic prices in Brazil and South Africa, to substantial domestic price overshooting in Ethiopia and Nigeria. Much of this variation can be explained by price stabilization policies, public policy failure, incomplete market integration, and coinciding domestic shocks.

Keywords:   price transmission, global food crisis, cereal prices, price stabilization, food policy

2.1 Introduction

This chapter synthesizes the evidence of price transmission from international markets to domestic markets during the 2007–8 food crisis experienced by the fourteen countries studied in this volume. It covers the markets for three grain commodities: maize, wheat, and rice.

The degree of price transmission from international to domestic markets during the 2007–8 food crisis varied significantly across countries. The country studies reviewed in this chapter discuss a wide range of factors that may serve to explain this variation, notably active stabilization policies and poor market integration due to high transportation costs and market imperfections. In several countries, domestic prices are largely unrelated to international prices and therefore reflect purely local shocks, such as harvest failures and political turmoil, which are independent from, but coincide with, the global crisis.

The empirical literature on price transmission that has emerged since the crisis generally comes to similar conclusions. Abbott and Borot de Battisti (2011) investigate price transmission patterns for a number of African countries (including a few others, such as China and Brazil, for comparison) by (p.22) plotting international and local commodity price indices and estimating price transmission elasticities. They find great variation in price transmission from almost none in China to virtually complete in Brazil. In Africa, countries like Nigeria and Ethiopia appear to be closely linked to the world markets, whereas most of the other countries show limited and/or lagged responses, suggesting that world market pressures are resisted by domestic market institutions. Abbott and Borot de Battisti also identify certain patterns, such as much greater price transmission for highly traded commodities (for instance rice) compared to non-tradable ones (millet and sorghum), and higher price transmission rates for import-dependent countries, including rice in Senegal, Mali, Burkina Faso, Niger, Malawi, and Uganda; maize in Malawi and Uganda; and wheat in Ethiopia.

Such findings are supported by other studies on countries in Africa (Benson, Mugarura, and Wanda 2008; Cudjoe, Breisinger, and Diao 2010; Minot 2011), Asia (Dawe 2008; Robles 2011), and Latin America (de Janvry and Sadoulet 2010; Robles 2011).

Most African countries experienced commodity price increases which were lower than the international prices, but in a few countries, notably Ethiopia and Malawi, food prices grew more rapidly than on the world market (Minot 2011). This suggests that other shocks than the world market prices were at play. Similarly, Benson et al. (2008) argue that the increasing food prices in Uganda could be better explained by domestic or regional factors (for instance, spill-overs from harvest shortfalls in neighbouring Kenya) rather than by the global food crisis.

Transmission of rice prices in Asia was limited in most cases, notably India, the Philippines, and Vietnam (Dawe 2008). Interestingly, Dawe finds a relatively high degree of pass-through (64 per cent) in China, which is in stark contrast to the almost flat local prices shown by Abbott and Borot de Battisti (2011). However, Dawe’s paper was one of the first analyses on the global food crisis, and the prices of rice were only available up to late 2007, i.e. before the international rice prices accelerated. Based on more recent data covering the international spike in rice prices, Robles (2011) finds that the pass-through of rice prices was actually lower in Bangladesh (34 per cent), a rice importer, than in Pakistan and Vietnam (around 51 per cent), two rice exporters.

De Janvry and Sadoulet (2010) and Robles (2011) report quite low price transmission elasticities in a number of Latin American countries (Costa Rica, Ecuador, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Peru, and the Dominican Republic). Indeed, in Nicaragua (maize), Honduras (maize), and the Dominican Republic (rice), the pass-through may even have been negative. In general, price transmission elasticities are estimated to be below 20 per cent, with notable exception being rice markets in Mexico (p.23) (almost 50 per cent) and wheat-to-bread price transmission in Ecuador (just over 40 per cent).

Through a synthesis of the fourteen country studies and with the support of food price data from Food and Agriculture Organization (FAO)’s Global Information and Early Warning System (GIEWS), this study provides an overview of the price transmission patterns experienced by the countries during the food crisis period. The data is summarized graphically by plotting monthly series of local maize, rice, and wheat prices against international benchmark prices. Additionally, the wealth of information offered by the country-study authors is reviewed in an attempt to gather some broad insights into how food price policies, domestic institutions, and other factors affected price transmission.

2.2 Conceptual Framework

2.2.1 What is Price Transmission?

The fundamental theoretical basis for price transmission is the law of one price (LOP) (Fackler and Goodwin 2001). The LOP can be written as

|pwpd |t
(1)

where pw and pd are the prices of a commodity on the world market and the domestic market respectively and t represents the transaction costs associated with importing or exporting the commodity.1 Equation (1) states that the gap between the international and the domestic prices of a commodity should never be larger than the transaction costs.

The positive transaction costs effectively create a price band between import parity, pw + t, and export parity, pwt. If domestic supply and demand conditions are such that the domestic price lies within the price band, no international trade is profitable. In contrast, if domestic supply is sufficiently large relative to demand to press domestic prices below export parity, incentives for international arbitrage should ensure that domestic prices do not stay too far below export parity for too long. Similarly, relatively tight domestic supplies, which push domestic prices close to or above the import parity, should invite importers to satisfy the excess demand at import parity. In effect, if the domestic price is within the parity bounds and the commodity is non-traded, we would expect the domestic price to be determined by domestic supply and demand condition and be unrelated to international (p.24) prices. Instead, if the commodity is traded, the LOP predicts a close relationship between the international and domestic prices.

In reality, the distinction between isolation from the international markets and world market integration is not as clear-cut as the theoretical discussion above would suggest. For instance, Mozambique and Kenya are both import-dependent in cereals, yet some studies (e.g. Minot 2011) have found that these countries are poorly integrated with the world markets. In contrast, even isolated markets, like the maize market in Ethiopia, appear to display some long-term relationship with the international markets (Loening, Durevall, and Birru 2009). In general, we are likely to find that domestic prices in all countries are determined by a mix of domestic factors as well as transmission from international prices. The relative strength of domestic versus international factors varies greatly from country to country depending on how well countries are integrated with the world market. The rest of this section discusses briefly a number of factors which may influence market integration.

2.2.2 Imperfect Market Integration

Transaction costs cover costs of transportation due to poor infrastructure (particularly if countries are landlocked), imperfectly competitive markets, regulatory costs, and tariffs as well as taxes. The higher the transaction costs are, the more likely it is that domestic prices fall within the parity bounds and commodities are non-traded. This is often the case with some of the basic staples in sub-Saharan Africa (SSA), such as millet, sorghum, cassava, and teff, but it also sometimes happens with internationally traded commodities. In such cases, the LOP no longer applies and we should not expect significant price transmission across borders. Yet, sometimes it still takes place. Demand substitution may link non-traded with traded commodities. As the price of the traded commodity increases following an international price shock, demand may shift towards non-traded commodities resulting in higher prices on these as well. Such cross-commodity price transmission may be weak, as Resnick (Chapter 14) indicates in the case of Senegal; or relatively strong, as appeared to have been the case in Ethiopia (Rashid 2011).

The trade status of an individual country tends to persist over time. Countries are often consistently net importers, net exporters, or separated from international markets. There are, however, exceptions. For instance, due to improvements in agricultural productivity, Malawi has shifted from import dependence to becoming largely self-sufficient in maize (Chirwa and Chinsinga, Chapter 7). South Africa, which is a regional trade hub for many commodities, varies between being net importer and net exporter of maize. Such trade regime shifts complicate price transmission analysis. Moving (p.25) from trading to non-trading status may disrupt the price transmission mechanism. Shifting from exporter to importer (or vice versa) may cause sharp changes in domestic prices independently from international price movements as the domestic price changes from export parity to import parity. Needless to say, considering the trade status and particularly changes in trade status is important for evaluating price transmission.

Many of the country studies in the sample, notably for Egypt, South Africa, Bangladesh, Senegal, and Zambia (Chapoto, Chapter 8; Ghoneim, Chapter 12; Kirsten, Chapter 19; Raihan, Chapter 11; Resnick, Chapter 14), suggest that the domestic food supply chains are characterized by high concentration and non-competitive behaviour. Imperfect competition in the supply chain adds additional margins to the transaction costs and thus influences the degree of price transmission. A large literature on supply chain price transmission argues that imperfect competition may be an important explanation for asymmetric price transmission, i.e. the observation that increasing prices are transmitted relatively strongly down the value chain, whereas lower prices are transmitted incompletely and/or with significant lag. Such asymmetries also complicate price transmission analyses.

Public policies may heavily influence the degree of price transmission. All countries in the sample use fiscal regulatory instruments such as tariffs, subsidies, and value added tax (VAT), which directly add to or subtract from the transaction costs. Non-fiscal government market interventions, such as non-tariff trade barriers, parastatal grain traders (China, India, Vietnam, Ethiopia, Malawi, Zambia, and Egypt), and price controls (notably Senegal), disrupt the price transmission mechanisms in ways that are harder to generalize. Most countries responded to the global food crisis by expanding existing policies or introducing new regulations. The next section investigates in more detail these policy responses’ likely impact on price transmission.

2.3 Political Intervention

During and after the global food crisis period, most governments in developing countries pursued a range of policies in attempts to reduce the transmission of the higher international prices to the domestic markets or to limit their adverse consequences (e.g., see Demeke, Pangrazio, and Maetz 2011 for a review). The fourteen country studies in the present sample describe similar policy responses (see Bryan, Chapter 3, for a comprehensive synthesis). The discussion of the possible impacts of policy interventions on price transmission is organized under three types of policies: trade policies, domestic policies, and macroeconomic policies. I end this section by briefly discussing issues surrounding design and implementation of policies.

(p.26) 2.3.1 Trade Policies

Border policies were pursued by many of the countries in the form of export restrictions and reductions of import tariffs (see Bryan, Chapter 3). Export restrictions are discussed first and import tariff waivers second.

By restricting exports, governments sought to reduce or completely disrupt the link between the international and the domestic prices. Obviously, the policy is likely to have the greatest effect when the export restriction is binding, i.e., if the country is already exporting the commodity or would have done so in the absence of the restriction. This is not to say that export restrictions are necessarily ineffective for an importer. If a cereal importer’s price stabilization policies are effective or the country experiences a bumper harvest, domestic prices may drop below export parity, thus creating incentives for exporting to neighbouring countries that are stabilizing local prices less aggressively. Still, a consistent food importer may face additional obstacles to changing trade status in the short term as prospective exporters need to establish new export channels, demonstrate compliance with quality standards, etc. (Dawe 2010).

Export restrictions were implemented by China, Egypt, Ethiopia, India, Kenya, Malawi, Vietnam, and Zambia (Bryan, Chapter 3). They were, however, not likely to be equally binding everywhere. China, India, Vietnam, and Egypt appear to have had both the capacity (indicated by exports in the recent years) and the incentive (the international prices being higher than the domestic equivalents) to export rice, and in India’s case, wheat. In contrast, it is doubtful that the export bans in Ethiopia and Kenya had much of an impact on price transmission. Although these countries have exported small quantities of maize in the recent past, the local prices have been much higher than the international prices. Malawi and Zambia are borderline cases: both have been exporting a small maize surplus to neighbouring countries recently. The domestic prices were roughly at par with international prices during the summer of 2008, whereas the prices in neighbouring countries, such as Mozambique and Kenya, were considerably higher. It is therefore quite possible that exports could have continued had the ban not been in effect (Chirwa and Chinsinga, Chapter 7, and Chapoto, Chapter 8, both report that some informal trade did take place in spite of the bans).

Suspending import tariffs may briefly halt or reverse the increase on domestic prices, but it will not sever the link to international prices. On the contrary, eliminating tariffs reduces transaction costs which may strengthen rather than weaken the market integration. Several countries suspended import tariffs on grains to partially compensate for the increasing international prices (Demeke et al. 2011). However, in many cases the tariffs were very low to begin with and hence the tariff waivers might only have had a marginal effect on price transmission.

(p.27) Bangladesh, Egypt, and Senegal temporarily removed tariffs on rice and wheat, but they were already relatively low, between 2 and 10 per cent, so the impact must have been modest (United Nations Conference on Trade and Development (UNCTAD) TRAINS (Trade Analysis and Information System) database). Tariffs were much higher in Kenya and Nigeria and the tariff waivers have likely softened the impact of the higher international prices. It is noteworthy that these two countries experienced significant domestic price shocks in spite of the policy responses (see the discussion on price transmission in Kenya and Nigeria in sections 2.4.3 and 2.4.4 below).

2.3.2 Domestic Policies

Domestic policies cover interventions that are designed to adjust domestic prices directly (e.g., food subsidies, suspension of VAT, or direct price controls), policies implemented to increase domestic food supply (e.g., release of grains from strategic grain reserves and long-term improvements in agricultural production), and social protection policies. Investments in agriculture and other efforts to improve agricultural productivity are long-term measures which are unlikely to have a major effect on price transmission patterns within the food crisis period. Also, social protection policies (such as income transfers and food for work programmes) should not affect price transmission directly, although they may contribute to maintaining food demand (Demeke et al. 2011). These policies are therefore not the focus of this synthesis.

Food subsidies and elimination of domestic duties work largely the same way as tariff waivers, except that they affect domestically produced commodities and not just imported food. The impact of higher international prices on domestic retail prices is reduced, but market integration should, if anything, be strengthened by the lower transaction costs. Such price support policies were attempted in Bangladesh, China, Egypt, Ethiopia, Kenya, Mozambique, Senegal, Vietnam, and Zambia (Bryan, Chapter 3), but it is not possible to get a complete overview of their likely impact. We have little information on the size of the subsidies and VATs, and we know generally little about the extent to which subsidies were applied (universally or specifically targeted).

Price controls may completely disrupt the price transmission mechanism if they are successfully enforced. Among the fourteen study-countries, only Ethiopia, Malawi, and Senegal attempted to control prices directly. The experiences of Malawi and Senegal demonstrate the difficulties associated with fixing prices when enforcement mechanisms are lacking. In Malawi, a price band was supposed to be maintained through open market grain procurement and sales by ADMARC, a parastatal trader, but the agency lacked the necessary funds for its operation (Chirwa and Chinsinga, Chapter 7). In Senegal, the government promised to subsidize rice distributors in return for (p.28) observing the price ceiling, but the policy backfired when the government was unable to disburse the funds on time (Resnick, Chapter 14).

Releasing grain from public stocks should not by itself directly affect price transmission if domestic markets are perfectly integrated with the world market. Any excess supply at prevailing world market prices would simply be exported. However, if markets are imperfectly integrated, possibly as a result of export restrictions, expansion of domestic supply should help depress domestic prices.

Among the fourteen study-countries, Bangladesh, China, Ethiopia, India, and Nigeria attempted to expand domestic supply by releasing stocks (Bryan, Chapter 3). It is difficult to evaluate the likely impact of such policies. Most of the country studies do not specify exactly how much grain was released, and the ones on Bangladesh, Egypt, Malawi, Senegal, and Zambia report that private traders responded to their governments’ attempts at stabilizing prices by hoarding grain. If such behaviour is widespread, private stockpiling of grain could render public efforts at stabilizing prices largely ineffective.

2.3.3 Macroeconomic Policies

Macroeconomic policies, such as fiscal and monetary policies, tend to be blunt instruments for achieving food price stabilizing objectives. However, the food crisis and the food price policies discussed above have macroeconomic repercussions, so it is instructive to briefly review the role played by macroeconomic policies during the food crisis period.

Consider first fiscal policies. Attempting to stabilize food prices or alleviating the impacts of higher food prices is very expensive. Most of the food price policies pursued by the fourteen countries either reduce government revenue (suspension of import tariffs and VAT) or increase outlays (food subsidies, social transfers, and agricultural investments). They are not likely to be sustainable if needed for an extended period of time (Abbott and Borot de Battisti 2011). As a result, food price policies may temporarily reduce price transmission, but once fiscal constraints force governments to roll back the policies, the link between the international and the domestic prices is re-established, possibly with a lag. Several countries in the sample approached the fiscal limits during the crisis, partly due to aggressive grain market regulation. For instance, Egypt managed to expand relatively generous food subsidies by cutting other government expenses (such as fuel subsidies) (Ghoneim, Chapter 12), while price policies in Malawi and Senegal were rendered ineffective by lack of funds (Chirwa and Chinsinga, Chapter 7; Resnick, Chapter 14).

Several of the country studies view the food crisis from a monetary perspective. Admassie (Chapter 6) suggests that relatively accommodating monetary policies in the years prior to the food crisis were responsible for a (p.29) significant portion of the observed Ethiopian food price inflation. In addition, the food crisis also coincided with rising general inflation in India and Bangladesh (Raihan, Chapter 11; Ganguly and Gulati, Chapter 16). One of the key responses to the crisis in these countries was tightening monetary policy. In contrast, Brazil found room for loosening monetary policies slightly by extending public credit, despite a strong focus on anti-inflationary policies (Mueller and Mueller, Chapter 18).

In a theoretical fully flexible economy, monetary policies should not influence the price transmission mechanism. Demand-driven inflation would increase domestic cereal prices denominated in local currency, but the exchange rate would depreciate by the same rate thus cancelling the effect of inflation on prices measured in foreign currency (US$). Economies are, however, not fully flexible and it is unlikely that exchange rate movements completely negate inflationary pressures, particularly in the short run. For instance, Ethiopia pursued a strongly inflationary monetary policy up to and during the food crisis period while placing strict controls on foreign exchange. As a result, the currency depreciated only marginally and the general inflationary pressures drove US$-denominated cereal prices higher (Minot 2011).

Exchange rate movements also greatly affect how US$-denominated price increases are transmitted to local-currency prices (Dawe 2008; Abbott and Borot de Battisti 2011). In this study, all price transmission evidence presented is measured in US$ prices in order to fully concentrate on the price transmission mechanism. Still, it is worthwhile briefly considering how exchange rate variations impacted on how domestic consumers experienced the food crisis. Consider for instance Brazil, where the price of maize rose by 122 per cent in US$ terms between June 2006 and June 2008. However, as the Brazilian real appreciated relative to the US$ over the same period, local currency denominated prices increased only by around 60 per cent. Most countries in the sample saw currency appreciation relative to the US$ (although not to the extent that Brazil did), but there are also exceptions. Due to the inflationary monetary policies in Ethiopia, the birr depreciated during the food crisis period and the price of maize rose by 189 per cent in US$ terms and 220 per cent when measured in local currency units.2

2.3.4 Policy Implementation

Policy interventions only work as intended if they are designed and implemented properly. Many of the country studies provide examples of (p.30) interventions that have gone awry due to lack of information, poor government capacity, and political concerns, or corrupt practices.

Policy makers do not operate in an environment of perfect information, particularly when dramatic events, such as the global food crisis, call for swift action. Many of the studies indicate that government interventions based on poor information have been less effective or have even exacerbated the situation. For instance, incorrect harvest forecasts led the government of Vietnam to reduce the number of export licenses in a situation where Vietnam could have benefited from improved terms of trade (Nguyen and Talbot, Chapter 15). In contrast, the government of Malawi apparently overestimated the maize harvest and entered an export contract with Zimbabwe that could not be filled (Chirwa and Chinsinga, Chapter 7). One of the key lessons learned by the South African government from an earlier local food crisis in 2002–3, was the importance of accurate and timely information which led the government to establish a network for monitoring food prices more closely (Kirsten, Chapter 19).

Whether or not appropriate policies are designed, many governments, particularly in SSA, have limited capacity to implement the policies properly. Several of the country studies report that policies could not be enforced. For instance, substantial informal cross-border trade took place in Kenya, Malawi, and Zambia in violation of export bans (Chirwa and Chinsinga, Chapter 7; Chapoto, Chapter 8; Nzuma, Chapter 9). Malawi and Senegal imposed price controls, but they were largely ignored (Chirwa and Chinsinga, Chapter 7; Resnick, Chapter 14). Also, many of the policies pursued by the governments, such as food and input subsidies, import and VAT waivers, and social safety nets, put severe strain on the countries’ fiscal resources.

An important constraint on policy implementation is the political economy of food price policy generating rent-seeking behaviour. The political environment shapes the choice, specific design, and implementation of food price policies, and policy implementation is also often influenced by politics and corruption. For a more in-depth analysis of the political economy of food price policy, see Watson (Chapter 5).

2.4 Price Transmission Patterns

The local grain markets in each of the fourteen countries in the country-study sample responded to the rising international grain prices in a variety of ways. Although no two stories are exactly the same, I attempt to identify some common patterns by classifying the fourteen countries into four categories: ‘free traders’, ‘exporting stabilizers’, ‘importers’, and ‘the isolated’. This classification is based on an assessment of the local markets’ exposure (p.31) to international price shocks, largely determined by the countries’ trade status and the strength and effectiveness of the food price policies pursued by the countries as detailed in the country studies. I chose the classification scheme to provide a rough guide to what kind of price transmission patterns we should expect to find. The classification is as follows:

  1. 1. Free traders: Brazil and South Africa are well-integrated into the global cereal markets as net importers, net exporters, or both (depending on harvests). They responded to the food crisis with few and relatively weak food price policies and we would therefore expect them to display a high degree of price transmission. As we will see, this is also largely what we find, although the picture is distorted a bit by trade regime shifts.

  2. 2. Exporting stabilizers: China, India, and Vietnam are all net exporters of rice, and they reacted forcefully to the rising international prices with strong food price policies. We would expect to see relatively weak price transmission in these countries as they have a powerful instrument, export restrictions, at their disposal. We find that in China and India these policies effectively stabilized local prices during the period, but the Vietnamese policies appeared to have had a limited impact.

  3. 3. Importers: Bangladesh, Egypt, Kenya, Mozambique, and Senegal are consistently dependent on imports of their main staples. They are exposed to international price volatility and in contrast to exporters they have very few means available for stabilizing local prices. Thus, we would expect price transmission to be relatively high. As we will see, the picture is rather mixed due to local factors and distorting policies.

  4. 4. The isolated: Ethiopia, Malawi, Nigeria, and Zambia are poorly integrated into global cereal markets and are largely self-sufficient in their main staple. In these cases, we would expect that domestic prices are primarily determined by local supply and demand factors and largely unrelated to global prices. Interestingly, we find that in spite of the countries’ relatively isolated status, they all experienced sharply increasing food prices during the crisis period.

In the following, I summarize the price transmission patterns observed in the countries in these four groups, and discuss some of the main factors influencing the patterns. Ideally, this summary would present estimates of price transmission elasticities. There are, however, technical and conceptual problems involved with such methods, when applied to the food crisis period, primarily that the standard methods may not be strictly valid.3 (p.32) Although these issues may be solvable, such attempts are beyond the scope of this synthesis. Instead, to give as clear and transparent picture of the price transmission patterns as possible, I provide a graphical representation of the international and domestic prices, and discuss in a bit more detail the main factors influencing the patterns.

2.4.1 Free Traders

The first group of countries consists of Brazil and South Africa. I call the group the free traders as both countries appear to be closely integrated into the world grain markets. In addition, they were pursuing fairly liberal food price policies during the crisis (as detailed in Mueller and Mueller, Chapter 18; and Kirsten, Chapter 19). This is also reflected in the relatively close co-movement of local maize, rice, and wheat prices with their international equivalents as illustrated in Figures 2.1, 2.2, and 2.3.

International to Domestic Price Transmission in Fourteen Developing Countries during the 2007–8 Food Crisis

Figure 2.1 Maize prices in Brazil and South Africa

Note: International: US Gulf, no. 2’, yellow maize. Brazil: national average, yellow maize, wholesale. South Africa: Randfontain, white maize, wholesale. US Gulf, no. 2 refers to the specific maize variety used to designate the international maize price. This is the most commonly used benchmark for an international maize price.

Source: FAO GIEWS food price database.

International to Domestic Price Transmission in Fourteen Developing Countries during the 2007–8 Food Crisis

Figure 2.2 Rice prices in Brazil

Note: International: Bangkok, ‘Thai 100% B’. Brazil: national average, wholesale. Thai 100% B refers to the specific rice variety used to designate the international rice price. This is the most commonly used benchmark for an international rice price.

Source: FAO GIEWS food price database.

International to Domestic Price Transmission in Fourteen Developing Countries during the 2007–8 Food Crisis

Figure 2.3 Wheat prices in Brazil and South Africa

Note: International: US Gulf, no. 2, hard winter red. Brazil: national average, wholesale. South Africa: Randfontain, wholesale. US Gulf, no. 2, hard winter red refers to the specific wheat variety used to designate the international wheat price. This is the most commonly used benchmark for an international wheat price.

Source: FAO GIEWS food price database.

In most markets, the prices move close together. There are, however, a few exceptions. In South Africa, the price of maize seems to be a lot more volatile (p.33) than the international prices. There are relatively large gaps between the South African and the international maize price in 2002, 2004, 2006, and the second half of 2007, but the gaps seem to disappear briefly in 2003, 2005, and 2008. The South African maize market is characterized by substantial variation in production around what is needed to satisfy demand (FAOSTAT 2013a). As a result, South Africa shifts continuously between being a net importer and a net exporter of maize. When South Africa is a net importer, as in 2002, 2004, 2006, and 2007 (according to data from FAOSTAT 2013b), the domestic price approaches import parity, whereas periods of net exports (2003, 2005, and 2008) drive prices towards export parity (NAMC 2007, 2009). Kirsten (Chapter 19) finds that when such trade regime shifts are accounted for, the evidence suggests that South African grain markets are highly integrated into the world markets.

The only other market in this group which experienced regime shift during the crisis period is the Brazilian market for rice. Brazil is traditionally a rice net importer, but at the peak of the crisis in 2008, the country exported around 81,000 tons more than it imported, compared to a net import of 500,000 tons in 2007 (FAOSTAT 2013b). A shift in trade status could account (p.34) for the relatively modest transmission of the international price peak as the domestic price declines (relative to international prices) from import parity towards export parity.

Both Brazil and South Africa are consistent net importers of wheat (FAOSTAT 2013b), so the domestic wheat prices tend to move closely with international prices (NAMC 2009). In Figure 2.3, discrepancies are evident during and after the global food crisis period, but the overall impression of substantial market integration remains.

2.4.2 Exporting Stabilizers

China and India successfully managed to stabilize local rice prices, primarily by banning exports. Vietnam also introduced export restrictions, but the stabilization policies were less effective. Common features of these three countries are that they are all net exporters of rice (India and Vietnam being two of the largest rice exporters in the world), and that grain trade tends to be heavily regulated. It therefore appears to have been relatively easy for these (p.35) countries to control domestic grain supply, and stabilize prices. Evidence of this is illustrated in Figures 2.4 and 2.5.

International to Domestic Price Transmission in Fourteen Developing Countries during the 2007–8 Food Crisis

Figure 2.4 Rice prices in China, India, and Vietnam

Note: International: Bangkok, Thai 100% B. China: Hunan province, first quality, wholesale. India: Delhi, wholesale. Vietnam: Hanoi, retail.

Source: FAO GIEWS food price database.

International to Domestic Price Transmission in Fourteen Developing Countries during the 2007–8 Food Crisis

Figure 2.5 Wheat prices in India

Source: FAO GIEWS food price database. International: US Gulf, no. 2, hard winter red. India: Delhi, wholesale.

Whereas the rice prices in China and India as well as wheat prices in India were virtually flat during the food crisis period, the price of rice in Vietnam did show a partially muted response to the surging international prices. One explanation may be that the Vietnamese prices are retail prices (wholesale prices were not available) which include additional marketing and profit margins. Another possibility is provided by Nguyen and Talbot (Chapter 15): the Vietnamese government sets rice export limits (by advice of the Ministry of Agriculture and Rural Development (MARD)) based on projected rice surplus and not rice prices per se. In the spring of 2008, MARD forecast a lean rice crop which prompted the government to reduce the maximum export quota by one million tons to 3.5–4 million tons for 2008, and impose a temporary three-month moratorium on signing new export contracts. In any event, the projection turned out to be faulty and rice farmers produced a bumper crop. The Vietnamese policies took the tip off the price spike but were not sufficient in isolating the domestic market completely.

(p.36) The Indian wheat and rice prices show an upward trend in the second half of 2009 (Figures 2.4 and 2.5). Ganguly and Gulati (Chapter 16) refer to this as a form of delayed price transmission. However, the increases in prices have less to do with (belated) transmission of price changes in the international market, but may be better explained by specific factors within India. India experienced relatively poor harvests in 2009, and following the food crisis of 2007–8, the government started to buy larger quantities of grain in an effort to expand the strategic grain reserves. Also, the MSPs for grains were adjusted upwards in the wake of the crisis, and the resulting higher farm gate prices were transmitted down through the value chain.

2.4.3 Importers

Five countries in the sample, Bangladesh, Egypt, Kenya, Senegal, and Mozambique consistently depend on imports for the supply of their main staple, rice in Bangladesh and Senegal, wheat in Egypt, and maize in Kenya and Mozambique.

Figure 2.6 shows the rice prices in Bangladesh and Senegal plotted against the international prices. I have also included the Indian rice price as India is the main supplier of Bangladeshi rice imports (Hossain and (p.37) Deb 2010). In fact, judging from Figure 2.6, the Bangladeshi rice price appears to be more closely related to the Indian price than the international (Thai export) price. Although both Senegal and Bangladesh were exposed to the international price volatility, they experienced the global crisis rather differently. In Bangladesh, the price shock is visible but quite modest. In Senegal, the rice price rose sharply and has stayed at a relatively high level ever since.

International to Domestic Price Transmission in Fourteen Developing Countries during the 2007–8 Food Crisis

Figure 2.6 Rice prices in Bangladesh and Senegal

Note: International: Bangkok, Thai 100% B. India: Delhi, wholesale. Bangladesh: Dhaka, wholesale. Senegal: Dakar, imported, retail.

Source: FAO GIEWS food price database.

Rather than explaining why the rice prices increased in Bangladesh in the early 2008, a more interesting question is why they did not rise further. Raihan (Chapter 11) reports that the government in Bangladesh did seek to stabilize the rice prices but they had limited tools at their disposal. The most effective short-term policy response was the suspension of a 5 per cent import tariff on grains, which had at best a marginal effect. More importantly, Bangladesh managed to secure a supply of around half a million tons of rice from India in February 2008, just before the international price skyrocketed. However, it took a long time for the two governments to agree on a price and the first delivery arrived in April, just as the domestic rice harvest was about to hit the market (Hossain and Deb 2010). The 2008 season turned out to produce a bumper (p.38) harvest, partly due to a massive supply response. In fact, for the first time in almost a decade, Bangladesh produced more than it used (FAOSTAT 2013a).

However, the question remains why private traders did not utilize the arbitrage opportunities and export rice from Bangladesh. The government did impose a ban on exports, but Dawe (2010) suggests that it had little effect in practice. As a traditional rice importer, Bangladesh lacks the capacity (in terms of quality assurance mechanisms and reputation) to export large quantities of rice in the short term.

Like Bangladesh, Senegal had few options available for reducing the impact of higher international rice prices. Import tariffs were already very low, and their suspension must have had little impact. To make matters worse, Senegal suffered two consecutive seasons of poor harvests in 2006–7 and 2007–8 due to shortages in agricultural inputs and irregular rainfall patterns (Resnick, Chapter 14). According to data from FAOSTAT (2013a), production in 2006 and 2007 was down almost 19 per cent compared to the average of the preceding three years. In response, the Senegalese government attempted to regulate rice prices directly. Senegal has had a long tradition of fixing bread prices, but as such price ceilings were extended to rice and other commodities, they could not be enforced and were therefore largely ignored. In response, the government struck an agreement with rice distributors by offering subsidies in return for lower retail prices. However, the subsidies turned out to be unsustainable and might actually have exacerbated the crisis. As the government was unable to pay out the subsidies on time, rice distributors started to stockpile rice in anticipation of future subsidy payments. As a result, rice prices continued to rise (Resnick, Chapter 14).

In the case of Egypt, it is difficult to assess the extent to which international prices were transmitted to domestic markets due to the lack of data: wholesale prices and retail prices are only available from January 2008 onwards. The evidence, shown in Figure 2.7, indicates a relatively stable retail price, but the apparent price stability was broken by short-lived spikes in late 2008, 2010, and 2011. The 2008 spike may represent a delayed response to the global food crisis; international wheat prices were also climbing in 2010 and 2011. Even so, any close relationship between international and Egyptian prices is not obvious. This rather mixed picture is consistent with the literature. Despite substantial wheat imports, Baffes and Gardner (2003) find that Egyptian wheat markets were very poorly integrated with global markets. However, Conforti (2004) and Rapsomanikis, Hallam, and Conforti (2006), show a long-term relationship between the Egyptian wholesale wheat prices and the international wheat prices after (but not before) 1989, despite strong regulation of Egyptian wheat markets.

International to Domestic Price Transmission in Fourteen Developing Countries during the 2007–8 Food Crisis

Figure 2.7 Wheat prices in Egypt

Note: International: US Gulf, no. 2, hard winter red. Egypt: Lower Egypt, retail.

Source: FAO GIEWS food price database.

An elaborate social protection infrastructure already existed in Egypt prior to the crisis. Large quantities of locally produced and imported wheat are (p.39) procured by the General Agency for the Supply of Commodities (GASC, a government agency), milled in public mills, and processed in public bakeries into baladi bread which is sold in public shops at subsidized prices. In addition, a range of subsidized basic commodities (such as sugar, oil, and rice) are distributed to a large part of the population on the basis of ration cards. Although the social safety net is hugely expensive, inefficient, and mired by corruption and massive leakage, it does appear to have shielded the poor beneficiaries from the worst impact of the global food crisis (Ghoneim, Chapter 12).

Figure 2.8 shows the wholesale price of maize in Kenya and Mozambique, together with the international price and the maize wholesale price in South Africa, the main supplier of the two countries. The maize prices in Kenya and Mozambique rose substantially during the food crisis period but as the international and South African prices collapsed in the second half of 2008, domestic prices stayed at a high level. In the case of Mozambique, maize prices peaked half a year after the global crisis.

International to Domestic Price Transmission in Fourteen Developing Countries during the 2007–8 Food Crisis

Figure 2.8 Maize prices in Kenya and Mozambique

Source: FAO GIEWS food price database. International: US Gulf, no. 2, yellow maize. Kenya: Nairobi, wholesale. South Africa: Randfontain, white maize, wholesale.

Closer inspection of the two countries suggests that domestic factors played a major role in the domestic maize price formation. Although Kenya and Mozambique are both consistent maize importers, Minot (2011) cannot find significant long-term relationship between the international (or South (p.40) African) and domestic prices (however, he does find evidence of rice market integration in Mozambique).

In Kenya, several years of drought-induced poor harvests generated a need for large imports of maize in 2008 (Benson et al. 2008; Nzuma, Chapter 9). According to FAOSTAT (2013a), production in both 2008 and 2009 was down by 20 per cent compared to the average of the previous three years. As the developing global food crisis prompted neighbours, such as Tanzania and Malawi, to restrict their exports, the Kenyan government struggled to fill import orders throughout 2008. When the import needs were finally satisfied during 2009 (another drought-ridden year), imports of maize jumped from around 114,000 tons in 2007 and 244,000 tons in 2008 to 1.5 million tons in 2009 (FAOSTAT 2013b). Insofar as the massive increase in imports represents a buildup of an acute supply shortage, it is not surprising that the maize prices increased in 2008 and persisted throughout the year. What is, perhaps, surprising is that Kenya was unable to satisfy its import needs for such a long time in the aftermath of the global food crisis. Although the regional export bans stayed in effect throughout 2008, Kenya’s main import supplier of maize was not Malawi or Tanzania but South Africa, which did not restrict maize exports. In fact, according to the data from the UN Commodity Trade Statistics Database (p.41) (COMTRADE), South Africa ended up supplying two-thirds of Kenya’s maize imports in 2009.

The picture is not quite as clear in the case of Mozambique. FEWS NET (2009) reports that the country was hit by floods, rainfall deficits, and wildfires in different areas which affected harvests adversely. It is, however, not clear if these events were particularly severe compared to the earlier years. Maize harvests in 2007 and 2008 were around the same size as the average of the previous three years (FAOSTAT 2013a). Arndt et al. (2008) suggest that part of the domestic price shock was due to the unusually high international prices as the government did little to prevent price transmission (see also Nhate, Massingarela, and Salvucci, Chapter 10). Also, part of the price volatility in Figure 2.8 seems to be seasonal. Prices tend to increase towards the ‘hunger season’ from October to January (FEWS NET 2009), as output from the second harvest is no longer available to cushion the prices (Arndt et al. 2008). The large price spike following the global price peak coincides with this period.

2.4.4 The Isolated

Among the fourteen countries included in the sample, Ethiopia, Malawi, Nigeria, and Zambia are relatively isolated from the international cereal markets. Ethiopia experiences substantial variations in maize production over time (FAOSTAT 2013a), but domestic storage serves to close the gaps between production and use with little reliance on international trade (Tadesse and Guttormsen 2011). Demand for maize in Nigeria tends to follow domestic supply quite closely (FAOSTAT 2013a), although Nigeria is dependent upon imports of rice (Olomola, Chapter 13). Malawi and Zambia have traditionally depended on maize imports from neighbouring countries, but both countries recently managed to become largely self-sufficient in maize and even produced a small surplus for exports (Chirwa and Chinsinga, Chapter 7; Chapoto, Chapter 8; FAOSTAT 2013a, 2013b).

Figure 2.9 shows the maize prices in Ethiopia and Nigeria together with the international reference price. Both countries experience sharply increasing prices during the global food crisis period, but the domestic price spikes are substantially larger than the international price shock. It also takes longer for the domestic prices to come down again—especially in Nigeria.

International to Domestic Price Transmission in Fourteen Developing Countries during the 2007–8 Food Crisis

Figure 2.9 Maize prices in Ethiopia and Nigeria

Source: FAO GIEWS food price database. International: US Gulf, no. 2, yellow maize. Ethiopia: Addis Ababa, wholesale. Nigeria: Kano, wholesale.

There is some debate over the extent to which the Ethiopian food prices were driven by domestic factors or the international crisis. The isolated (and landlocked) nature of Ethiopia together with its very limited international food trade would suggest that the domestic market should be fairly unrelated to international markets. However, the evidence appears to be mixed. In their empirical studies, Ulimwengu, Workneh, and Paulos (2009) and Minot (p.42) (2011) find no long-run relationship between Ethiopian maize, sorghum, or wheat markets. In contrast, Conforti (2004) and Loening et al. (2009) do find long-run co-integration between the Ethiopian and international grain prices.

Admassie (Chapter 6) argues that the Ethiopian crisis was primarily caused by long-running domestic factors, such as increasing grain demand due to economic growth and more well-developed social safety nets, stagnating grain supply, and inflationary monetary policy. The trigger came in 2008, when the surging global energy prices led to a shortage of foreign exchange reserves which prompted the government to ration the foreign exchange available for other purposes such as imports of food.

The inflationary monetary policy explanation for the observed food price inflation has gained a lot of traction, particularly among economists at the World Bank and the IMF (Haji and Gelaw 2012). Foreign exchange controls prevented the currency from depreciating and rapidly rising prices denominated in domestic currency translated directly into high food inflation measured in US$ (Minot 2011).

There is little doubt that Ethiopia experienced high and growing general inflation in 2007 and 2008. It is less clear, however, that food inflation was necessarily driven by demand fueled by expansionary monetary policies. (p.43) Food inflation, particularly in cereal prices, was substantially higher than non-food inflation during the food crisis period (FAO/WFP 2009). While demand-driven inflation does not necessarily entail that all prices grow by the same rate, it is still remarkable that Ethiopia had the largest food inflation rate relative to non-food inflation in the developing world (3.5 per cent per month) in 2008, and negative relative food inflation in both 2009 and 2010 (Headey et al. 2012).

In the light of this discussion, it is unlikely that Ethiopian food prices were determined by solely domestic or international factors. In support of this assertion, Haji and Gelaw (2012) decompose food price inflation into different components and find that a number of factors affected domestic food prices, including domestic price level, world grain prices, domestic fuel prices, and non-food prices. Unfortunately, it is not straightforward to assess the relative weight of the different factors in forming domestic prices.

Admassie (Chapter 6) presents another piece of interesting evidence, reproduced here using data from FAO. Figure 2.10 shows a relatively close relationship between different Ethiopian cereal prices—indeed the maize and wheat prices appear to be much more closely related to each other than their respective international counterparts. This is consistent with the explanation of food prices being driven by inflationary monetary policy. But it could also (p.44) indicate an additional (and sometimes forgotten) channel of price transmission. Price increases in one grain commodity are likely to spill-over to other related commodities as demand shifts from more expensive to cheaper staples.

International to Domestic Price Transmission in Fourteen Developing Countries during the 2007–8 Food Crisis

Figure 2.10 Maize, wheat, and teff prices in Ethiopia

Source: FAO GIEWS food price database.

In Ethiopia, this mechanism appears to be quite strong. Rashid (2011) finds a close relationship between the wheat and maize prices, but the correlation between these and the teff price is limited. Furthermore, his analysis suggests that transmission from maize to wheat is stronger than vice versa. It is possible that whatever influence the global cereal markets have on the Ethiopian prices, it works through a single commodity rather than through multiple commodities simultaneously.

Nigeria has not been studied as closely as Ethiopia and less data is available. However, there appear to be many similarities with the Ethiopian case. Nigeria is self-sufficient in maize, yet maize prices spiked along with (and greatly surpassing) the international price during the 2008 food price crisis (see Figure 2.9). However, unlike Ethiopia, Nigeria also experienced a surge in maize prices in 2005 that was almost as large in nominal terms as the one in 2008. This earlier price spike occurred at a time where international prices were relatively stable which suggests that the Nigerian maize prices are predominantly driven by domestic factors. There was no obvious shortfall in production in 2005, but there was a small surplus in 2006—equivalent to 10 per cent of production (FAOSTAT 2013a). It is possible that a part of the price spikes in 2005 and 2008 reflects unusually low prices in 2006 and 2007, following an increase in production in 2006.

Nigeria imports a substantial part of its rice consumption, but unfortunately no data on the Nigerian rice prices were available on a monthly basis. Yearly averages presented by Olomola (Chapter 13) show that rice prices also rose in 2008, but more detailed price transmission patterns cannot be established for rice.

The maize prices in Malawi and Zambia are depicted in Figure 2.11 along with the international price and the regional reference price in South Africa. Both the domestic price series show an upward trend during the food crisis period, but generally the prices display a relatively high volatility and do not show any obvious relationship with any of the reference prices. Minot (2011) finds weak (insignificant) long-run links between the international prices and a few of the local Malawian markets close to the border, but no evidence of any long-run relationship between the Zambian and international prices. Both countries face very high international trade costs and international trade in maize is strictly under the control of the government (Chirwa and Chinsinga, Chapter 7; Chapoto, Chapter 8). They both have a history of relatively frequent food crises during which local maize prices increase rapidly relative to international prices, latest in 2005–6. All this suggests that prices are mainly (p.45) determined by domestic factors and that international prices should play a very limited role.

International to Domestic Price Transmission in Fourteen Developing Countries during the 2007–8 Food Crisis

Figure 2.11 Maize prices in Malawi and Zambia

Source: FAO GIEWS food price database. International: US Gulf, no. 2, yellow maize. Malawi: Lilongwe, retail. Zambia: national average, white maize, retail. South Africa: Randfontain, white maize, wholesale.

Malawi experienced good weather and a bountiful harvest in 2007 (the surplus was smaller, but still positive in 2008). Zambia faced flooding in parts of the country and output declined slightly in both 2007 and 2008, compared to 2006. However, 2006 was a good year, and harvests in 2007 and 2008 were still in line with a long-term upward trend (FAOSTAT 2013a). Chirwa and Chinsinga (Chapter 7) and Chapoto (Chapter 8) suggest that the local crisis was largely precipitated by government mismanagement combined with private hoarding behaviour. In Malawi, very poor quality of information about the domestic maize supply led the private sector to believe that there was a shortage of maize, inducing speculative hoarding of maize in anticipation of higher prices in the future. In response to the initial price increases, the government banned exports (and cancelled a partially filled export agreement with Zimbabwe) and tried to restrict private domestic trading in an effort to control prices. These initiatives merely reinforced the signals of supply shortages and led to more intensive hoarding of maize. In Zambia, reports of flooding initiated the increase in maize prices, and as the government (p.46) responded by banning exports, speculative forces similar to those in Malawi added to the price pressure.

Although there appears to be little direct relationship between the global food crisis and the local crises experienced in Malawi and Zambia, it is possible that the high global prices acted as catalysts for the local market tensions. Chirwa and Chinsinga (Chapter 7) and Chapoto (Chapter 8) suggest that the local maize prices were largely driven by hoarding, driven by expectations of higher prices. It is possible that these expectations were also influenced by the global food outlook, just as the global food crisis could have inspired greater unease among the government officials and prompted them to misjudge the appropriate policy response.

2.5 Conclusion

Drawing a few general conclusions from fourteen very different stories on how the global food crisis was experienced and managed is not an easy task. Yet, it is possible to identify certain patterns shared by several of the countries. Here are the most important ones: the synthesis categorized the fourteen countries according to their trade status to generate predictions regarding the price transmission patterns. The four categories are:

  1. 1. Free traders: as well-integrated and open agricultural economies, Brazil and South Africa were expected to exhibit a relatively large degree of price transmission. The evidence presented here is consistent with this prediction, although the close relationship between the South African and international maize price is sometimes masked by changes in trade status (between exporter and importer), which cause the domestic price to shift between export and import parity.

  2. 2. Exporting stabilizers: China, India, and Vietnam are exporters of rice, and they all have effective state-control over exports. Although the countries’ strong exporter status should generate a close relationship between domestic and international prices, effective price stabilization policies (primarily in the form of export restrictions) were expected to reduce price transmission substantially in the time of crisis. This prediction is strongly supported in the cases of China and India, but the Vietnamese retail rice prices showed a surprisingly strong pass-through of international prices, despite export restrictions.

  3. 3. Importers: of the fourteen countries in the sample, Bangladesh, Egypt, Kenya, Mozambique, and Senegal are consistently dependent upon imports of their main staple. Unlike exporters, the import-dependent countries have few stabilizing policies available, and price transmission (p.47) was therefore expected to be substantial. However, the evidence painted a rather mixed picture. While Bangladesh managed to contain domestic rice prices at relatively low levels (indeed, price pass-through was smaller than in Vietnam), domestic prices in Kenya, Mozambique, and Senegal rose rapidly during the crisis and stayed high long after the international crisis had subsided. Egypt appears to have been greatly affected by the food crisis as well, but the evidence is more difficult to evaluate due to data limitations.

  4. 4. The isolated: Ethiopia, Nigeria, Malawi, and Zambia are poorly integrated with the international cereal markets and are largely self-sufficient in their main staple. Due to these countries’ relatively isolated status, we should expect domestic prices to be independent from the international prices. It is therefore striking to observe that they all faced rapidly increasing domestic prices during the food crisis period. In Ethiopia and Nigeria, the maize price spike surpassed the international prices by several orders of magnitude, and Malawi and Zambia experienced persistently high prices.

The synthesis discusses two overall reasons for why we observed unexpected price transmission patterns: issues related to the implementation of policies and various domestic factors. Although food price policies aimed at controlling local prices were implemented in most countries, they were not equally effective everywhere. Export bans were not always binding (e.g. Ethiopia and Kenya), and in Vietnam export restrictions were reportedly designed to counter a perceived domestic shortfall rather than the international crisis. Suspended import tariffs were often already low, particularly in import-dependent countries (e.g., Bangladesh, Egypt, Mozambique, and Senegal). Many of the policies, particularly food subsidies and tax exemptions, represented a heavy drain on government finances and were unsustainable in the long term (as acutely felt in Egypt, Malawi, and Senegal). Also, many of the country studies suggest that government mismanagement rendered policies less effective or even exacerbated the crisis (Nigeria, Malawi, Senegal, and Zambia).

It is difficult to evaluate to what extent the local price shocks reflected domestic factors or were driven by the global crisis. On the one hand, it is remarkable that almost all of the countries experienced domestic price shocks that coincided with the global price spike (sometimes with a short time lag), even those (such as Ethiopia, Nigeria, Kenya, and Mozambique) that empirical research suggests are poorly integrated with the world markets. On the other hand, correlation is not the same as causation and certain signs indicate that domestic factors must have played a primary role in many countries. In Ethiopia, Kenya, Nigeria, and Mozambique, local shocks were (p.48) much greater than the global shock, and it is hard to explain how price transmission alone could have generated such a major local price response. In Egypt, Malawi, and Zambia, the domestic prices spiked during the food crisis period, but apart from that, local prices show very little obvious relationship to international prices. Also, several of these countries, including Nigeria, Mozambique, and Zambia have earlier experienced episodes of great price volatility that were clearly not related to international prices. It is also possible to identify local supply constraints that can explain a substantial part of the local price shocks: harvest failures in Kenya and Senegal, flooding in parts of Mozambique and Zambia, generally tight markets combined with inflationary policies in Ethiopia, and poorly designed and implemented food policies in Malawi, Senegal, and Zambia. On a more positive note, a domestic supply response may also explain why the food crisis in Bangladesh was relatively benign: a bumper harvest was beginning to hit the markets just as international rice prices spiked.

References

Bibliography references:

Abbott, P., and A. Borot de Battisti (2011). ‘Recent Global Food Price Shocks: Causes, Consequences and Lessons for African Governments and Donors’. Journal of African Economies, 20 (suppl. 1): i12–i62.

Arndt, C., R. Benfica, N. Maximiano, A. M. D. Nucifora, and J. T. Thurlow (2008). ‘Higher Fuel and Food Prices: Impacts and Responses for Mozambique’. Agricultural Economics, 39: 497–511.

Baffes, J., and B. Gardner (2003). ‘The Transmission of World Commodity Prices to Domestic Markets Under Policy Reforms in Developing Countries’. The Journal of Policy Reform, 6(3): 159–80.

Benson, T., S. Mugarura, and K. Wanda (2008). ‘Impacts in Uganda of Rising Global Food Prices: The Role of Diversified Staples and Limited Price Transmission’. Agricultural Economics, 39: 513–24.

Conforti, P. (2004). ‘Price Transmission in Selected Agricultural Markets’. Commodity and Trade Policy Research Working Paper 7. Rome: Food and Agriculture Organization (FAO).

Cudjoe, G., C. Breisinger, and X. Diao (2010). ‘Local Impacts of a Global Crisis: Food Price Transmission, Consumer Welfare and Poverty in Ghana’. Food Policy, 35(4): 294–302.

Dawe, D. (2008). ‘Have Recent Increases in International Cereal Prices Been Transmitted to Domestic Economies? The Experience in Seven Large Asian Countries’. ESA Working Paper 3. Rome: Food and Agriculture Organization (FAO).

Dawe, D. (2010). ‘Can the Next Rice Crisis Be Prevented?’. In D. Dawe (ed.), The Rice Crisis: Markets, Policies and Food Security. London and Washington, DC: Food and Agriculture Organization (FAO) and Earthscan.

(p.49) de Janvry, A., and E. Sadoulet (2010). ‘The Global Food Crisis and Guatemala: What Crisis and for Whom?’. World Development, 38(9): 1328–39.

Demeke, M., G. Pangrazio, and M. Maetz (2011). ‘Country Responses to Turmoil in Global Food Markets: The Nature and Preliminary Implications of the Policies Pursued in the 2006–08 Episode’. In A. Prakash (ed.), Safeguarding Food Security in Volatile Global Markets. Rome: Food and Agriculture Organization (FAO).

Fackler, P. L., and B. K. Goodwin (2001). ‘Spatial Price Analysis’. In B. L. Gardner and G. C. Rausser (eds), Handbook of Agricultural Economics, vol. 1, Part 2. Amsterdam: Elsevier.

FAO GIEWS (2013). ‘Global Information and Early Warning System: Food Price Data and Analysis Tool’. Rome: Food and Agriculture Organization (FAO).

FAO (Food and Agriculture Organization)/WFP (World Food Programme) (2009). ‘Special Report: Crop and Food Security Assessment Mission to Ethiopia’. Rome: FAO/WFP.

FAOSTAT (Food and Agriculture Organization Statistics Division) (2013a). ‘Food Balance Sheets’. Rome: FAO.

FAOSTAT (Food and Agriculture Organization Statistics Division) (2013b). ‘Trade Database, Crops and Livestock Products’. Rome: FAO.

FEWS NET (Famine Early Warning Systems Network) (2009). ‘Mozambique Food Security Outlook’. FEWS NET and USAID (United States Agency for Interntional Development).

Haji, J., and F. Gelaw (2012). ‘Determinants of the Recent Soaring Food Inflation in Ethiopia’. Universal Journal of Education and General Studies, 1(8): 225–33.

Headey, D., F. B. Nisrane, I. Worku, M. Dereje, and A. S. Taffesse (2012). ‘Urban Wage Behaviour and Food Price Inflation: The Case of Ethiopia’. Ethiopia Strategy Support Programme II Working Paper 41. Addis Ababa: IFPRI (International Food Policy Research Institute) and EDRI (Ethiopian Development Research Institute).

Hossain, M., and U. Deb (2010). ‘Volatility in Rice Prices and Policy Responses in Bangladesh’. In D. Dawe (ed.), The Rice Crisis: Markets, Policies and Food Security. London and Washington, DC: FAO (Food and Agriculture Organization) and Earthscan.

Loening, J. L., D. Durevall, and J. A. Birru (2009). ‘Inflation Dynamics and Food Prices in an Agricultural Economy: The Case of Ethiopia’. Policy Research Working Paper 347. Washington, DC: World Bank.

Minot, N. (2011). ‘Transmission of World Food Price Changes to Markets in Sub-Saharan Africa’. IFPRI (International Food Policy Research Institute) Discussion Paper 1059. Washington, DC: IFPRI.

NAMC (National Agriculural Marketing Council) (2007). ‘The South African Food Cost Review 2007’. Pretoria: NAMC.

NAMC (National Agriculural Marketing Council) (2009). ‘The South African Food Cost Review 2009’. Pretoria: NAMC and the Department of Agriculture, Forestry and Fisheries.

Rapsomanikis, G., D. Hallam, and P. Conforti (2006). ‘Market Integration and Price Transmission in Selected Food and Cash Crop Markets in Sub-Saharan Africa’. In A. Sarris and D. Hallam (eds.), Agricultural Commodity Markets and Trade. (p.50) New Approaches to Analyzing Market Structure and Instability. Cheltenham and Northampton, MA: Edward Elgar and FAO (Food and Agriculture Organization).

Rashid, S. (2011). ‘Intercommodity Price Transmission and Food Price Policies: An Analysis of Ethiopian Cereal Markets’. IFPRI (International Food Policy Research Institute) Discussion Paper 1079. Washington, DC: IFPRI.

Robles, M. (2011). ‘Price Transmission from International Agricultural Commodity Markets to Domestic Food Prices: Case Studies in Asia and Latin America’. Washington, DC: IFPRI (International Food Policy Research Institute).

Tadesse, G., and A. G. Guttormsen (2011). ‘The Behavior of Commodity Prices in Ethiopia’. Agricultural Economics, 42(1): 87–97.

Ulimwengu, J. M., S. Workneh, and Z. Paulos (2009). ‘Impact of Soaring Food Price in Ethiopia’. IFPRI (International Food Policy Research Institute) Discussion Paper 846. Washington, DC: IFPRI.

UN COMTRADE (United Nations Commodity Trade Statistics Database) (2013). ‘United Nations Commodity Trade Statistics Database’. Accessed through World Integrated Trade Solution (WITS). New York: United Nations Statistics Division.

UNCTAD (United Nations Conference on Trade and Development) TRAINS (Trade Analysis and Information System) (2013). ‘Trade Analysis and Information Systems’. Accessed through World Integrated Trade Solution (WITS). Geneva: UNCTAD.

Notes:

(*) The author would like to thank all the authors of the fourteen case studies, upon which this synthesis is based, for their useful comments and quality control of an earlier version of this study. Also, the study has benefitted greatly from constructive input by Shane Bryan, Derrill Watson II, Per Pinstrup-Andersen, and an anonymous reviewer. Finally, a special thanks to Per Pinstrup-Andersen for his encouragements and great patience throughout the process.

(1) Transaction cost is a general concept used to capture a variety of pecuniary and non-pecuniary costs associated with trading, such as transportation costs, import tariffs, waste, spoilage, and opportunity costs due to delays, profit margins, etc.

(2) The numbers are own calculations based on price data obtained from the FAO GIEWS food price database and exchange rates from International Monetary Fund (IMF) financial statistics.

(3) Technically, the price series are likely to be strongly non-stationary during the food crisis period, even in first-difference form, and standard inference methods may therefore not be valid.