1. The large recent fluctuations of international agricultural commodity prices and policies to deal with them in light of future market developments Alexandros Sarris Professor of Economics, University of Athens, Presentation at the Bank of Greece 18 March 2011

2. Overview of presentation Recent global agricultural markets in perspective Causes of agricultural commodity market price increases during 2007-8 and recently The medium term outlook for global agricultural markets Agricultural market volatility and its determinants Policies to deal with agricultural market volatility

3. Recent commodity price movements

4. Recent commodity price movements have not been uniform (source World Bank)

5. Recent crude oil price movements

6. Recent world wheat price movements

7. Recent world maize price movements

8. Recent world rice price movements

9. Very recent agricultural market developments Agricultural commodity prices rose 5.6 percent in February, up for a ninth straight month. Most of the gains were in raw materials and beverages. Food prices rose 2.9 percent, just 0.7 percentage points more than the dollar depreciation. Cotton prices surged 19 percent due to a number of supply shortfalls, strong demand and and low stocks, while rubber prices jumped 10 percent as wet weather affected production in south-east Asia. Coconut oil and palmkernel oil prices (close substitutes) rose 10.9 percent and 8.3 percent, respectively, due to a steep decline in Malaysian palmkernel oil production, causing stocks to fall to a multi-year lows. Cocoa prices increased 10 percent because of political conflict in Côte d’Ivoire, while coffee increased 8-9 percent on low stocks and tight supplies. Wheat prices gained 6.6 percent due to concerns about the winter wheat crop in the U.S. and China, and strong international buying. Maize prices rose 11 due to tight U.S. inventories and concerns that the U.S. might not be able to rebuild its current low level of stocks.

10. Latest cereal market assessments FAO’s latest (March 2011) forecast confirms a tightening of the global cereal supply and demand balance in 2010/11. A decline in world production in 2010 in the face of growing demand is expected to result in a sharp drawdown of world stocks. Reflecting this prospect, international cereal prices have increased sharply with export prices of major grains up at least 70 percent from this time last year. » The latest estimate for the world cereal production in 2010 is 8 million tonnes more than was anticipated in December but still slightly below 2009. This month’s upward revision reflects mostly higher estimates for production in Argentina, China and Ethiopia. » The forecast for world cereal utilization in 2010/11 has been revised up by 18 million tonnes since December. The bulk of the revision reflects adjustments to the feed and industrial utilization of coarse grains. Larger use of maize for ethanol production in the United States and statistical adjustments to China’s historical (since 2006/07) supply and demand balance for maize are the main reasons for the revision. » World cereal stocks for crop seasons ending in 2011 are forecast to fall sharply because of a decline in inventories of wheat and coarse grains. A plunge in stocks of coarse grains at the global level as well as for major exporters is expected to push down stock to use ratios of coarse grains to the lowest in three decades.

11. Recent cereal market developments

15. World food commodity price index 1990- 2011

16. World cereal commodity price index 1990-2011

17. World oils commodity price index 1990- 2011

18. World dairy commodity price index 1990- 2011

19. World meat commodity price index 1990- 2011

20. World sugar price index 1990- 2011

21. Commodity prices in long term perspective (current prices)

22. Commodity prices in long term perspective (real prices)

23. Cereal commodity prices in long term perspective (current prices)

24. Cereal commodity prices in long term perspective (real prices)

25. Is there an end to cheap food? Real international prices of grains have tended to decrease but since mid 1980s tendency seems to have stopped and may have reversed in 2008-9

26. Real prices of vegetable oils have tended to decrease but since mid 1980s tendency seems to have stopped

27. Real prices of livestock commodities have tended to decrease albeit at slowing pace since mid 1980s

28. Real prices of sugar and beverages have tended to decrease but since mid 1980s tendency seems to have stopped

29. What determines long term commodity prices? Supply of agricultural commodities highly elastic at low wages Demand for agricultural commodities quite inelastic Opposite case for non-agriculture Implication: Differential productivity gains can alter terms of trade between agriculture and non-agriculture

30. How do productivity gains affect agriculture and non-agriculture? Productivity affects agriculture differently than non-agriculture P Q p c p’ d Q P S S’ a b Panel A. Agricultural Commodity SectorPanel B. Non-agricultural sector p p’ a S S’ b D D

31. Declining terms of trade for agricultural commodities has been due to faster rates of total factor productivity growth for agricultural than non-agricultural products Rate of growth of TFP has been faster in agriculture than in non-agriculture The rate of growth of TFP in agriculture seems to be higher than that of manufacturing. “Globalization” of agricultural research, has contributed to faster TFP growth in agriculture, Incidence of productivity advances largely on consumers (through lower prices) and little to producers. Annual TFP growth in agriculture does not appear to have slowed down for the world. Hence most likely reason for real price leveling must be lower inputs and faster demand growth

32. Some fundamentals of the world cereal markets Dominance of the US in world exports (maize 60%, wheat 25%). Hence US specific factors important (biofuels, US dollar movements, world dollar reserves, futures market trends, US stocks) Seasonality and inelasticity of supply and demand (supply shocks important) Differences in wheat and maize markets (maize market more sensitive to US events) Peculiarities of rice market (only 6% of global production exported, many more trade barriers)

33. Facts of the 2007-8 world food crisis Long term trends maybe relevant for the understanding of short term market developments (trends in incomes and demand, productivity growth slowdown, decline in inputs to agriculture such as land, labour, irrigation, fertilizer) Food export prices rose very quickly and sharply Prior to the 2007-8 crisis food commodity prices were at an all-time low after declining for most of the previous 30 years Many commodities’ prices rose sharply (petroleum, metals, minerals, fertilizer). Petroleum price to agric price passthrough 0.17 Non-food agricultural commodity prices did not rise as sharply as food commodities (food specific versus common macro factors) US dollar had depreciated against a wide range of currencies (price rises in Euro 25% less) After peaks in mid 2008 food and other commodity prices declined sharply (bubble, overshooting)

34. Timeline of events contributing to the 2007-8 food crisis

36. Explanations of high agrifood commodity prices in 2007-8 (new factors in blue) Strong growth in demand –sustained historically high economic growth world wide (China, India have not become more import dependent) –bio-fuel feedstock demand, particularly for maize and vegetable oils –stronger currencies/ weak USD (weak dollar may have contributed 20 percent to agric price increases) –Low interest rates (enhances demand for speculative stocks) Constrained supply –high energy related input costs... crude oil up since 2000. Crude oil price rises may have contributed 22 percent to agric good price rises –repeated yield shortfalls in key areas – climate change? Low commodity stocks –increased speculation/ demand to rebuild Increased activity on commodity exchange –Increased participation by commodity and other funds Policies and policy changes –tariff liberalization by importers –decoupling of subsidies, reduction in export subsidies, lower public stocks –increased use of export taxes/ bans –biofuel subsidies/tariffs/tax credits etc, changing mandates etc

37. Global ending stocks of wheat and stock to utilization ratios for the whole world and for the world without China

38. Global ending stocks of maize and stock to utilization ratios for the whole world and for the world without China

39. Global ending stocks of rice and stock to utilization ratios for the whole world and for the world without China

40. Speculation. Actual and counterfactual corn futures prices with and without index funds (source Gilbert 2009) The broken line sets innovations to the index investment series to zero.

41. Simulated impact on commodity prices of a 1 % USD depreciation against all currencies

42. Policy actions adopted by a sample of 77 developing countries to deal with high international food commodity prices

43. Effects of export restrictions on rice prices

44. Effects of trade policies on wheat prices

45. The global financial crisis and commodity markets: Demand factors i)Slower rates of GDP growth in OECD economies, but sustained rates of growth in East Asia economies. Likely downward pressure on prices (it happened last year) ii)Reduced oil prices lower agricultural production costs and dampen demands for biofuel feedstocks, but this is reversed this year with high oil prices iii)Portfolio reallocations of international commodity funds and other financial funds, away from commodities, put downward pressure on prices in the short run. Not clear if this has changed recently, but certainly evidence of speculative commodity positions iv)Exchange rate volatility v)Return to market fundamentals? (appears to have happened until recently)

46. The global financial crisis and commodity markets: Supply factors With commodity prices declining production incentives will be dampened and supply response can be smaller. Will not hold this year Freight rates declined and this may have lower import costs after 2008,but this may have changed recently Falling prices presented an opportunity to replenish stocks, but since stockholding is (among others) a function of future price anticipations did this happen? (Yes) Large scale investments in land and production will take some time to augment world supply

47. Real commodity prices 15-40% above 1997-06 levels Production increases in the range of 10-40% Strong expansion of biofuels driven by mandates LAC and Eastern Europe fastest growing production Developing countries driving production, consumption, and trade gains Medium term (10 year) outlook highlights

48. OECD: Recovery after the crisis

49. Mixed picture for non-OECD economies

50. World Oil price: high levels in nominal terms

51. Medium term: World Agricultural production and trade (Base 1999-2001 =1)

52. Agricultural production and trade Industrial Countries (Base 1999-2001 =1)

53. Agricultural production and trade European Union (Base 1999-2001 =1)

54. Agricultural production and trade - BRIC (Base 1999-2001 =1)

55. Agricultural production and trade LDC Countries (Base 1999-2001 =1)

56. Agricultural production and trade Other Developing (non-LDC, non-BRIC) (Base 1999-2001 =1)

57. Food and fuel drive up demand for wheat + 34 Mt

58. Feed and fuel push coarse grain demand + 50 Mt + 90 Mt

59. US maize production and amount used for ethanol production

60. Biodiesel demand will drive demand increases for vegetable oils + 90 Mt

61. World ethanol outlook

62. World biodiesel outlook

63. Biofuel use of global feedstock production + 90 Mt Share of feedstocks used for biofuel production from global production

64. Most commodity prices at higher average levels in 2019 compare to before the 2007-8 crisis Percentage change in world prices in real terms relative to 1997-06

65. Medium Term: Main points In 2010, despite the wheat shock, a degree of normalcy returned to many markets with production closer to historical levels and demand recovering. But 2011 may be quite different. The macroeconomic environment reflects the start of global economic recovery in late 2009 and a slow transition towards higher growth. Average crop prices over the next 10 years are projected to be above the levels of the decade prior to the 2007/08 peaks, based on a higher cost structure particularly in regions where energy inputs are used intensively. Global agricultural production is anticipated to grow more slowly in the next decade than in the past one. Global sectoral growth will be led by the regions of Latin America and Eastern Europe, and to a lesser extent by certain countries in Asia. Developing countries will provide the main source of growth for world agricultural consumption and trade. Their demand is driven by rising per capita incomes, and urbanisation, reinforced by population growth. Diets are expected to slowly diversify away from staple foods towards increased content of animal protein and processed foods that will favour meat and dairy products. Biofuel markets depend heavily on government incentives and mandates, but prospects remain uncertain.

66. Stylized facts about instability in international agricultural commodities Proper policy response to shocks depends on correct assessment of shocks Real commodity prices have small negative long-term trends, but difficult to detect as signal to noise ratio small. This may have changed in the late 1980s or early 1990s Commodity price distributions have the properties of ‘skewness’ and ‘kurtosis’. Skewness suggests that prices can reach occasional high levels, that are not symmetrically matched by corresponding lows, with prices spending longer in the ‘slump’ than at higher levels. Kurtosis suggest that ‘extreme values’ can occur occasionally. Commodity shocks have “persistence”. Typically it takes about five years for effects of shocks to dissipate. Commodities experience booms and slumps. The average duration of slumps exceeds the duration of booms by about a year. The magnitude of price falls in a slump is slightly larger than the magnitude of price rises in a subsequent boom. The rate of change of prices in a boom is typically faster than the rate of change of prices during a decline. The severity of price boom and slumps is unrelated to their duration. The probability of ending a commodity slump or a boom is independent of the time already spent in the slump or boom. Unrelated commodities do not co-move Commodity prices exhibit varying instability around their long-term trends.

67. Ex-Post (Backward looking) versus Ex-Ante (Forward looking) Volatility Relatively high (or low) ‘volatility’ in the past does not necessarily imply that the series has become more (or less) volatile in the ‘forward looking’ sense. From a policy and welfare points of view, it is the ex-ante volatility (or unpredictability) that matters. Is it just the volatility in the unpredictable components which matter or is it the overall volatility? Probably both but depends on context and issue.

68. How can volatility be measured? Variance but of what? of the series? of the difference or percentage difference of the series? of a series around a long run trend? of the ‘unpredictable’ component of the series? by implied volatility from options data? The appropriate measure can depend on what is the issue at hand.

69. Has volatility increased? Annualized real historic volatility of selected food commodities 1957-2010

70. Annualized real historic volatility of selected food commodities 1957-2010

71. Has volatility increased? Implied price volatilities 1987-2010. Proxies for unpredictability

72. Volatility and unpredictability increase during crises. Average ratio of the price of at the money wheat options in CBOT to the futures price on which the option is written at various advance periods (k’s) (percent)

73. Production does not seem to have become more variable for wheat and maize

74. Production does not seem to have become more variable for rice and soybeans

75. Potential determinants of ex-post commodity market price volatility Past Volatility Trends Stock levels Yields Transmission across commodity price Exchange rate volatility Petroleum price volatility Export concentration Interest rate volatility Other determinants

76. 1.Nearly all agricultural commodities have significant stochastic trends. Pigmeat is the exception. 2.Most of the agricultural commodities have cyclical components with the exception of palm oil. 3.Past volatility is a significant predictor of current volatility for nearly all variables (with and without exchange rate volatility). 4.There is evidence that there is transmission of volatility across agricultural commodities for nearly all commodities (except pigmeat). 5.Oil price volatility a significant predictor of volatility in agricultural commodities in the majority series. 6.As with oil prices, exchange rate volatility impacts on the volatility of commodity prices for 10 out of the 19 series. 7.Stock levels have a significant (downward) impact on the volatility for each of the three series for which there is data on stocks. 8.A number of commodity prices have significant deterministic trends. These trends are positive for some series and negative for others. 9.While some agricultural prices may have recently been relatively volatile, it does not follow that this reflects an overall growth in ex ante volatility. Summary of results of Balcombe (2009) concerning explanations of agricultural market volatility

77. Main factors that will affect medium term agricultural commodity prices and price volatility (new factors in blue) Developments in total incomes and consumption (demand inelasticity) Stocks and stock replenishment rates Shocks to production Petroleum prices Biofuel policies and technology prospects Developments in exchange rates Developments in financial markets and speculative fund positions New investments in agricultural production Country policies vis a vis domestic markets Overall: New factors are likely to dominate. Considerable uncertainty and likely volatility Implications for agri-food trade. International markets may become less reliable sources of food, but may offer new opportunities for growth exports of developing countries

78. New challenges for the international agricultural trading system Recent food crisis created mistrust of the international trading system and moves to promote food self sufficiency Many middle and high income NFIDCs started thinking about investments in food production in other countries with contractual commitments to buy back products. This is likely to change world trade patterns for agricultural products. There will be a growing need for medium and long term supply arrangements with main exporters To promote developments along agricultural comparative advantage, need to create system to assure net food importing countries (both developing and higher income) that their physical import supplies can be guaranteed through imports Will need to create system to manage increased price volatilities. G20 priority Will need system to ensure low income food deficit countries appropriate finance to import in times of high food prices Freer trade has increased concentration. Need to define competition rules for international agrifood trade Lower protection has seen increase in application of standards (especially private ones). Will need system to regulate the proliferation of such standards

79. Cereal import dependence 2007-9 (number of countries with percentage share of imports to total domestic supply in given range)

80. Staple food import risks. Macro issues Transitory versus permanent external shocks Uncertainty about external factors affecting food imports Possible impact of external and internal shocks on domestic economy Price transmission to domestic economy Uncertainty of policy objectives Structure of import trade (public versus private)

81. Staple food import risks. Micro issues Determining import requirements How to fulfil import requirements, namely through imports, or by reductions in publicly or privately held stocks How to minimize overall cost of fulfilling import requirements, given uncertainties in international prices and international freight rates How to manage the risks of unanticipated cost overruns How to finance the transaction Counterparty risk of non-delivery of the agreed supplies Major factor in contract defaults is adverse price movements that have not been hedged adequately by supplier

82. Four ways to manage food import risks avoiding or reducing the risk altogether (by altering domestic production, higher degree of staple food self sufficiency) change the fundamentals of supply and demand, by manipulating directly the markets that create those risks (through for instance buffer stocks for global price stabilization) transfer some of the risk to a third party for a fee. This is the standard approach to insurance do none of the above and just cope Basic problem is market unpredictability

83. Policy options to deal with external high food prices Trade policies (tariff changes, export taxes, restrictions) not very effective Domestic taxation policies: not very effective Stock policies. Not effective and expensive Input and other production subsidies (may work in some cases) Combine small scale market operations with effectively targeted safety nets Regional free trade may enhance regional food security Coordination and information between private and public sectors

84. External insurance systems available in developed countries but not in DCs Government subsidized insurance Futures and options markets OTC risk management products International compensatory finance mechanisms (e.g IMF food facility) ex-post and do not deal with immediate problem In developed countries much more predictability of agricultural prices because of policies (e.g minimum prices)

85. Should physical, public, globally managed grain reserves be developed? Answer: Most likely no Why: Three main challenges in maintaining strategic reserves: determination of optimum stocks is politically loaded – Predicting supply and demand and where the potential shortfalls in the market may be can be extremely difficult – Reserves are dependent on transparent and accountable governance level of costs / losses – Reserves cost money and stocks must be rotated regularly – The countries that most need reserves are generally those least able to afford the costs and oversight necessary for maintaining them – The private sector is better financed, better informed, and politically powerful, putting them in a much better position to compete uncertainties that strategic reserves can bring about in the market place. – Reserves distort markets and mismanagement and corruption can exacerbate hunger rather than alleviate it

86. Should commodity exchanges be reformed by: limiting the volume of speculation relative to hedging through regulation; making delivery on contracts or portions of contracts compulsory; imposing additional capital deposit requirements on futures transactions. Answer: probably NO – Speculation is a symptom not a cause of spikes, and has not altered market fundamentals albeit has enhanced spikes. This is true irrespective of exchanges or not

87. Assuring adequate grain supplies for world markets Promote “production reserves” instead of commodity reserves In several OECD countries policies have been instituted to set-aside land. Such policies are largely “decoupled”, namely non-trade distorting, hence acceptable from a WTO perspective. Relevant policies, could include apart from support for land set asides, support for technology and farm human capital skills, incentives to maintain set-aside land in in environmentally sustainable condition, etc. Productive land set-aside could be brought into physical production in high income countries within 6-10 months (the recent supply response is evidence to that)

88. Appropriate policies for assuring grain market access by middle and high income net grain importing countries Investments in food production in other countries with commitments to buy back products Medium and long term arrangements with main exporters Managing import risks through derivative instruments reinsured in international reinsurance market

89. Can staple food commodity imports be hedged with futures and options. Case of wheat in in the Chicago CME Bulk of global wheat imports is obtained from the US, Australia, and Argentina Consider US Gulf price as an indicative price for all wheat imports Gulf and near futures Chicago prices are cointegrated, and adjustment to short term shocks is quite fast Simulations involve buying futures or call options k months in advance of the actual order, and selling them when the actual physical transaction for wheat imports is concluded Assumed that agent buys futures k months in advance of date when need to contract the actual delivery. For call options strike price is parameterized as (1+alpha) times the futures price observed in month t for the contract expiring at or in the nearest month after the period t+k, when the actual transaction will be made

90. Normalized coefficients of variation of wheat import bills k=2 months in advance

91. A proposal to ensure food imports in low income countries net grain importing countries through a dedicated Food Import Financing Facility The major problem faced by LDCs and NFIDCs during periods of food import needs in excess of normal commercial imports, is import financing for both private as well as parastatal entities

92. Basic rationale and concept of a FIFF Purpose: To allow LDCs and NFIDCs to finance commercial food imports in periods of excess import bills Problem to be dealt with: Credit and financing exposure ceilings from developed country financing institutions to LDCs and NFIDCs Concept: Provide additional finance for commercial food imports in excess of normal commercial food imports. In other words increase risk bearing capacity of financial institutions financing food imports How: By inducing increases in credit ceilings and country exposures under specific conditions, via a credible mechanism of intermediation

93. The basic structure of the Food Import Financing Facility (FIFF) Ex-ante (i.e. before onset of marketing year) availability of extra finance, based on estimates of excess food import bills Financing, or guarantees for finance above normal credit line ceilings, availed at normal commercial terms. No subsidies, no conditionalities Excess finance made available to financial institutions of eligible LDCs and NFIDCs (not directly to governments or traders). Domestic financial institutions will deal with local food import traders. FIFF would interpose itself between financial institutions in food exporting countries and financial institutions in eligible food importing countries. FIFF will supplement and augment the existing export financing mechanisms in developed food exporting countries.

94. Trigger conditions High international food prices Domestic production shortfalls Excess food import finance possibility made known and available on basis of estimates of excess food import bills, in advance of marketing year Estimates of excess food import bills will be based on estimates of international prices, domestic production, and imports, by reliable credible institutions.

95. Advantages of FIFF No need for new international institution. Facility can operate as part of existing IFI Ex-ante mechanism, not ex-post No conditionalities for finance Low interest rates, due to lower cost of intermediation Risk pooling of food import risks across many LDCs and NFIDCs Specialized knowledge of food import finance and relevant risk management Low interest rates of excess food import finance Considerable leveraging of funds (with small yearly costs total finance extended can be many times that) Multilateral export credit guarantee mechanism for food exports. Low risks due to sophisticated risk management, hence low cost (a small share of total financing extended) Could be adapted and extended to serve more purposes, such as a special concessionary window

96. Concluding comments Recent food commodity market volatility mostly explained by fundamentals, but price changes seem to be beyond what would be expected. World food commodity market volatility is here to stay, and may increase because of several new factors Speculation has increased and may make commodity markets more unstable Macro factors have become more relevant in commodity markets Commodity market volatility can better be dealt with by risk management rather than by market interference Institutional changes can help expand the range of risk management instruments for commodity agents (more commodity exchanges, FIFF?)

97. THANK YOU