Learn how the World Bank Group is helping countries with COVID-19 (coronavirus). Find Out

Speeches & Transcripts

Fiscal Consequences of Food and Agricultural Commodities Inflation

October 11, 2011

Otaviano Canuto, World Bank Vice President and Head of Network, Poverty Reduction and Economic Management (PREM) Remarks for the GAIM/GMA, Geopolitical Risk, Macroeconomics and Alternative Investment Conference New York

As Prepared for Delivery

While food and agricultural commodities inflation may have harmful fiscal impacts -- especially during crises -- fiscal policy remains one of the main tools to promote crisis recovery, economic growth and equity. Acknowledging this dual nature in the relationship of fiscal policy and food/commodities inflation is critical when stakes are so high. On April 16, 2011, President Zoellick stated that: "[F]ood prices [are] the biggest threat today to the world's poor… Already 44 million people have fallen into poverty as a result of rising food prices over the last year… Further 10% rise in the food price index could push 10 million more people into poverty". On August 15, President Zoellick insisted again in reference to global food prices -- and financial markets volatility—that "Persistently high food prices and low food stocks indicate that we are still in the danger zone".1

But fiscal consequences of food and agricultural commodities inflation are not uniform -- in fact, some countries may benefit from such inflations. The impact that global food inflation has on fiscal balances -- but also external balances, macro stability, growth and welfare -- will critically depend on each country's position as net importer or net exporter of such commodities. This is the same case as the toll of losers -- importing countries -- we have grown familiar with as international oil prices spike. In the case of food, global shocks may have dramatic consequences. Countries in the Horn of Africa such as Somalia, Kenya or Ethiopia currently face famine and humanitarian emergency situations, respectively, while Tanzania and Uganda are net exporters of grains (maize on the main). In the Andean Region, Venezuela is the only net importer of food; in contrast, Bolivia, Colombia, Ecuador and Peru are all net food exporters. But Bolivia is the only net exporter of cereals and vegetable oils, that is, of those foodstuffs whose price increases dramatically spiked vis-à-vis coffee and bananas -- which drive Ecuador, Peru and Colombia net food exporting position.

Because of these differences, averages can be deceptive.2 For a sample of 139 developing countries, we estimate that net imports of food averaged about 0.78% of each country's GDP for the period 2002-2009. When we separate net importers --70% of the sample—from net exporters, net food imports for the former average 2.2% of their GDP, while the latter average 1.9% of their GDP in net food exports. Interestingly, each group averages also conceal large variations. In fact, a substantive proportion of net importing countries have net food imports totaling 2% to 5% of their GDP. Those net import shares clearly put such countries in a highly vulnerable position to global food price inflation.

Vulnerability also has a domestic angle, closely related to the macroeconomic strength of the economy, including -- of course -- fiscal balances. Price shocks on net food importers3 may expectedly widen current account deficits; put additional pressure on exchange rates; likely cause a shortage of foreign reserves; substantively build domestic inflationary pressure; and exacerbate popular pressures for increased expenditures (not always on most efficient uses). For example, in the course of the Arab Spring, the Jordan government has recently overturned its food subsidy slash in 2008 and has introduced tax exemptions on thirteen foodstuffs. In Peru, the food price crisis in 2007/8 put pressure on the government, who implemented a rushed and short-lived food aid program distributed by the military. The fiscal space and public debt that countries have will determine how much countries are able to increase public expenditures in order to limit the welfare consequences of domestic price increases. Alternatively, they may also allow pass-through of international prices, which may not be politically feasible in the presence of high domestic inflation. Countries might also reduce import tariffs of food/fuel, a decision that is also limited in practice by fiscal balance and public debt. World Bank estimates for a sample of 58 countries4 that 40% of countries reduced taxes on food; 30% controlled prices and/or resorted to consumer subsidies; and 20% introduced export restrictions as responses to the previous food price crisis. And more than 15% of the sampled countries did nothing! When fiscal balances and debt restrict the use of effective macro policies and other factors such as weak governance and civil unrest all coalesce, global food inflation may contribute to food insecurity catastrophes as the on-going famine in Somalia.

The case of the Horn of Africa is admittedly extreme, but a large number of countries across the world are in a highly vulnerable position to food and fuel inflation. The list includes Rwanda, Tanzania, Mali or Liberia, in Sub-Saharan Africa; Kyrgyz Rep., Bosnia, Latvia, Turkey and Poland, in Eastern Europe; Afghanistan, Mongolia, Bangladesh or India, in Asia; and Jamaica, El Salvador, or Peru in Latin America.

As part of a very simple preliminary exercise based on correlations, Rehman and Van Doorn (2011)5 concentrated on five dimensions of macroeconomic vulnerability of these countries: specifically, their fiscal balances (as % of GDP), external debt (as %GDP), foreign reserves (to imports in months), CPI and current account balance (as % of GDP). These indicators were added up and normalized. Countries like Iraq, Sri Lanka, DRC, India, Sudan, Sierra Leone and Liberia are among the most vulnerable economies in 2010.

But even highly vulnerable countries do not behave similarly over time. Looking at their changes over time -- between 2002 and 2009 -- we identify those who observed the largest deteriorations in their macroeconomic vulnerability vis-à-vis the largest gains. While India, Iraq or Venezuela all pertain to the set of countries seeing their vulnerability deteriorated, others like Liberia, DRC or Cambodia experienced the largest improvements.

From a policy point of view -- both global and domestic -- however, there are three immediate channels through with inflation affect fiscal balances. The first one is that the overall cost of living increases as food prices increase, more so for low income countries where food constitutes a substantive part of the consumption basket. Notorious examples of this transmission abound. Recently in East Asia virulent increases in the domestic price of pork in China are believed to be a key driver of worrisome increases in their CPI throughout 2011.6 This is also the case of very different economies in Central America, whose recent sharp CPI increases are related to substantive increases in the prices of their main staple, beans. In middle income countries with extended safety nets, critical schemes such as pensions, public -- and or minimum -- wages and social transfers may well be indexed to inflation, so the transmission of food and other commodities into general inflation will also have an additional fiscal impact. Examples of this indexation can be found in economic powerhouses in Latin America, such as Mexico, Brazil or Colombia. Also, there are costly consequences of linking consumption subsidies to inflation. In Middle East and Northern African countries, the magnitudes of consumption subsidies -- on key food staples vastly consumed -- is staggering. They are estimated around 10% of GDP in Jordan or about 20% of GDP in Iran.7

But not all are bad news, however: crises may incentivize right decisions on future investments. Permanent (or so perceived) high food inflation makes attractive medium to long-term investments in agricultural research and more productive agricultural techniques. Crises provide the opportunity to take bold decisions. Countries like Brazil, Malaysia or Thailand have made significant progress in agricultural commercialization in recent years and undertaken investments in research and extension; others such as India and Mali have improved publicly accessible market information system, to cite a few examples (World Bank 2010).8

Among the G20 priorities9, the increase in agricultural investments that improve production, research, innovation and dissemination top the list -- alongside investments in infrastructure and more transparent agricultural information. One of the most promising initiatives in this respect is what the World Bank calls "climate-smart agriculture", that is, interventions that simultaneously increase yields (poverty reduction and food security), make yields more resilient in the face of extremes (adaptation), and make the farm a solution to the climate change problem rather than part of the problem (mitigation).10 Research on drought-resistant maize is one of the many examples. China's programs on erosion prevention such as the Loess Plateau or cash payments for carbon soil sequestering pilot projects involving small-holder farmers in Kenya are other examples. In DRC, revenues from carbon payments of a reforestation program are used to put children at school.11

But more resources are simply not enough, even when they are poured in countries, regions or communities most in need. Public agricultural spending in Bolivia provides a good example. In a recent study, Cuesta, Edmeades and Madrigal (2011)12 show that increasing agricultural spending during the last decade had a limited effect on reducing municipal vulnerability to food insecurity. This was partly the result of different categories of agricultural spending having distinctive impacts on vulnerability. Investments in infrastructure and spending on research and extension had the most significant positive impacts on reducing vulnerability; the opposite effect -- that is, increasing vulnerability -- was found on support and development; and administrative spending.

If the choice of investments is critical, so is the perception of price changes. Right decisions, implying bold sectoral shifts, decisive increases, will not take place if price increases are believed to come and go, that is, to be volatile. The G20 Ministerial Declaration of June 2011 emphasizes that excessive volatility in developed and developing countries affects longer-term agricultural development and global food security. Direct approaches to limiting food price volatility are at best very costly and require a lot of political commitment (say, for example, to develop good global information systems on stocks). The Bank argues that fear of volatility pushes the poor into sub-optimal livelihood strategies such as subsistence food production in dry areas that both prevent escape from poverty and exposes them to an increased risk of loss of livelihood from drought. For poor consumers with little cushion, volatility matters, as unexpected high food prices result in reducing access to other priorities such as education and health, and irreversible damage from early childhood malnutrition. Also, volatility divert efforts and resources to rebuild confidence in global food markets and protect the poor who need help more quickly than the necessary long-term solutions can provide.13

Mitigation strategies have or may have a hefty fiscal cost. Food price increases have led in a number of countries to violent riots and subsequent popular demands for lower prices. Even in the absence of these popular demands, governments are under increasing pressure -- regardless of the fiscal position they find themselves in -- to protect their citizens against those shocks. The extent of the shock, the capacity to confront it, the political economy of the specific country, and the existence of safety nets already operating will all affect the policy choice. In some cases -- as those mentioned above in the Arab world -- the only politically acceptable options may imply a huge bill. But this does not have to be the case if, for example, targeting is properly implemented. Simulations in Russia by Simler and Zaman (2010)14 show how a well targeted (80% of the program reaches the poor) cash transfer scheme that provides a monthly benefit of US$ 13 per beneficiary (or about 12.5% of the poverty line) would completely reverse the increase of 1 per cent point of headcount poverty following the substantive increases of wheat prices in that year (by 136% in the case of buckwheat). At a cost of US$ 1.7 billion or 0.1% of GDP, the scheme is more effective in reducing poverty than a policy that would remove the existing VAT of wheat, which is estimated to cost some US$ 6 billion and would only reduce poverty by 0.5 per cent points.

In practice, however, many poor countries lack safety nets prepared to mitigate food inflation. Grosh et al (2011) show that for the 14 countries most affected by domestic food inflation in the first quarter of 2011, only one country, Georgia, had a strong safety net to effectively respond.15 More comprehensively, the Country Policy and Institutional Assessments, CPIA, on safety nets capacities, show that 60 countries out of 135 LIC and MICs score 3 or less in a 1-6 point scale. These countries either have no safety nets or their scale and funding are inadequate to protect most poor and vulnerable groups and significant leakages exist.

Ultimately mitigation costs should not be seen only as a fiscal burden but an investment in human capital and political stability. The magnitudes of food crises alone, as estimated by the World Bank, prove certainly daunting: 44 million of new poor are estimated by Ivanic, Martin and Zaman (2011)16 as a consequence of the food price increases between June 2010 to July 2011. In the 2007/8 the increase in poverty was estimated as 108 million poor.17 And associated with increasing poverty, there are distributional effects.18 In the on-going Horn of Africa famine the worst affected among the 12 million people at risk of starvation are agro pastoralist communities, migrants and the urban poor, all net food consumers without access to food. In Mexico, however, the worst affected are the urban poor, as it is the case in Mozambique.19 In contrast, it is the middle class in urban areas of Brazil who bear the brunt of food inflation in a predominantly urban country concentrating net consumers in cities.20 Lost production, less productive workers, lower consumption, additional demands from safety nets have all a negative fiscal impact in the short term, as well as an immediate growth effect. But failure to mitigate them may have substantive long term effects. Malnutrition at early age has been long proved to have lifetime consequences in terms of personal and cognitive development and ultimately human capital accumulation and future employment and incomes. For example, Schady and colleagues have shown that economic shocks -- and not only long term declines in GDP -- can have serious effects on infant mortality, more of girls than boys.21 They mention how the effect of infant mortality in the Eighties' Peru was larger than that of late Nineties' Indonesia, which they attribute in part to the protection of social spending on health in Indonesia but not in Peru. Skoufias and Tiwari (2011)22 show that key micro nutrients consumption may change during periods of crisis, and that these changes may not be uniform. In Indonesia, vitamin C is insensitive to income variations of the household, while iron and calcium are sensitive to income shocks.

In conclusion

  • Watch out carefully! In contrast to the popular contagion fear associated with financial crises, the effects of food crises on countries are determined by their initial net importing position -- a position that may well change across commodities -- and domestic macroeconomic vulnerability.
  • Fiscal costs are important, but so is the capacity of fiscal policy to redress country's vulnerability to food inflation. Short term and immediate consequences mix with longer-term consequences, both in economic and social vulnerability terms.
  • But vulnerability has also a human dimension in millions of new poor or irreversible early childhood malnutrition. Mitigating these effects have a short term cost -- may not be necessarily hefty -- but also longer-term returns in terms of averted human capital destruction, lower growth and increasing instability.
  • Crises provide the right incentives to undertake bold decisions as climate smart agriculture exemplifies.


1 World Bank, press releases, April 16 and August 15, 2011.
2 Author thanks Ralph van Doorn for providing the data on food trade for this analysis.
3 But even net food/fuel exporters may not be immune to price increases. Balance of Payments gains along with rising international prices may well lead to rising domestic prices. Also, for net food exporters but net fuel importers, the price of fuel/fertilizer may increase food production cost. See Rehman and van Doorn (2011), “Macroeconomic Vulnerability to Food and Fuel Prices”, presented 5 May 2011 at World Bank.
4 World Bank (2009) “Rising Food Prices: Policy Options and World Bank Response”. PREM, ARD, DEC.
5 Rehman and van Doorn (2011), “Macroeconomic Vulnerability to Food and Fuel Prices”, presented 5 May 2011 at World Bank.
6 World Bank (2011a) Food Price Watch August 2011, PRMPR. World Bank.
7 P. Verme and H. Sayed (2011), “Hashemite Kingdom of Jordan: An Analysis of Consumption Subsidies” Draft, World Bank.
8 World Bank (2010) “Rising Food Prices: Policy Options and World Bank Response”. PREM, ARD, DEC
9 G20 (2011) “Action Plan on Food Price Volatility and Agriculture” Ministerial Declaration, Meeting of G20 Agricultural Ministers, 22 and 23 of June, 2011, Paris, France.
10 World Bank (2011b) Climate Smart Agriculture website, World Bank.
11 Ibid.
12 J Cuesta, S. Edmeades and L. Madrigal (2011) “Food insecurity and public agricultural spending in Bolivia : putting money where your mouth is?” World Bank Policy Research Working Paper 5604
13 R. Zoellick (2011) “Food Price Volatility Is the Problem” Letter to the Editor, Foreign Affairs.
14 K. Simler and H. Zamman (2010) A Simple Tool to Assess the Poverty Impact of Higher Food Prices: An Application to the Recent Russian Food Price Spike, Presented at World Bank, October 21, 2010.
15 Strong basis for response when one or more programs with high coverage of poor, highly progressive targeting and good administration. See M. Grosh, C. Andrews, R. Quintana and C. Rodriguez-Alas (2011) “Assessing Safety Net Readiness in Response to Food Price Volatility”, Draft, World Bank.
16 M. Ivanic. W. Martin and H. Zamman (2011) “Estimating the Short-Run Poverty Impacts of the 2010-11 Surge in Food Prices” Policy Research Working Paper 5633, World Bank.
17 M. Ivanic and W. Martin (2009) “Implications of Higher Global Food Prices for Poverty in Low-Income Countries”, Agricultural Economics, 39, pp. 405-16.
18 World Bank (2011a) Food Price Watch August, 2011. PRMPR, World Bank.
19 J. Valero-Gil and M. Valero (2008) ‘The Effects of Rising Food Prices on Poverty in Mexico” Agricultural Economics, 39 (Supplement), pp. 485-96.; C. Arndt, R. Benfica, N. Maximiano, A. Nucifora and J. Thurow (2008) “Higher Fuel and Food Prices: Impacts and Responses for Mozambique”, Agricultural Economics, 39 (Supplement), pp. 497-517.
20 F. Ferreira, A. Fruttero, P. Leitte and L. Lucchetti (2011) “Rising Food Prices and Household Welfare: Evidence from Brazil in 2008” Policy Research Working Paper 5652, World Bank.
21 S. Baird, J. Friedman and N. Schady (2010) “Aggregate Income Shocks and Infant Mortality in the Developing World”, IIEP-WP-2010-07, Draft, George Washington University.
22 E. Skoufias, S. Tiwari and H. Zamman (2011) “Can We Rely on Cash Transfers to Protect Dietary Diversity During Food Crises?: Estimates from Indonesia” Policy Research Working Paper 5548, World Bank.