1. Increases in Global Commodity Prices: The Macroeconomics and Policy Responses of Developing Countries Prof. Jeffrey Frankel, Harvard University V Jornada Monetaria, Banco Central de Bolivia Crisis Alimentaria, Inflación y Respuestos de Política 18 July, 2011, La Paz

2. 2 Prices of food & other commodities are again raising concerns.  French President N. Sarkozy placed food price volatility on the G-20 agenda for 2011.  Many fingers of blame have been pointed:  at global climate change  At “evil speculators,”  at the US Fed,  at growth in China…

3. 3 The questions I. Why are prices of food & other commodities so high? II. What is the record with microeconomic policies to try to reduce the volatility of the prices of imported commodities? III. How should a food importer set monetary policy? IV. How does the answer change for a country that also exports commodities (e.g., minerals), with world prices that are equally volatile?

4. 4 2008 & 2011 commodity price spikes were as high as the 1970s Nominal prices 2010=100 Real prices * = nominal in 2000 A.Saiki, Dutch Nat.Bk.

5. 5 I. Why are prices for food & other commodities so high?  Two competing views of how prices are determined  “Commodities are goods.”  => Prices are determined by the flow of supply & demand and their current economic fundamentals,  such as disruptions from weather or politics. Vs.  “ Commodities have become like assets,”  especially those commodities that are storable.  => They are determined by calculations regarding expected future fundamentals and alternative returns;  – in other words, by speculators.

6. 6 The asset model:  If commodities are storable & markets work efficiently  the expected future change in the price of the commodity (relative to the interest rate), should help determine today’s price.  Via 3 channels:  (i) the decision whether to harvest/ cull/ log/ extract today  vs leaving the crop/deposits in the fields/ground until tomorrow;  (ii) the decision whether to hold inventories,  or to sell them today;  (iii) the decision whether to go long in futures markets,  or to go short.

7. 7  In support of the flow/goods view:  In support of the flow/goods view:  2008 observation: Paul Krugman argued that inventories (stockpiles) of oil, etc., had not risen  as they should if the high prices were caused by an increase in the expected rate of return to holding commodities.  2011 observation: If flows did not matter, the release of oil from US SPR & other countries’ stockpiles (30 m barrels each) on June 23 would not have caused prices to fall 4% that day. Both views are partly right (and partly wrong)

8. 8  In support of the asset view:  In support of the asset view:  Although individual commodities are impacted by “flow” fundamentals,  such as Russian drought, US ethanol subsidies, instability in oil producing countries, etc.,  it cannot be a coincidence that the prices of virtually all fuels, minerals, & farm products rose sharply in 2008 and again in 2011.  True, fuels & grains are partial substitutes in production; but this cannot be the full answer. Both views are partly right (and partly wrong)

9. 9 As the asset view predicts, real commodity prices are inversely related to the real interest rate Real US interest rate Real $ food price index, Moody’s excl. oil 1951-2007 monthly A.Saiki, Dutch Nat.Bk.

10. 10 Establishing that the asset model is important does not imply that speculators are destabilizing.  Speculation often reflects fundamentals.  It can be stabilizing –  dampening volatility when high prices are recognized as temporary;  and carrying the message of likely future changes in fundamentals.  But occasionally speculators are destabilizing, carried away by speculative bubbles.  E.g., Credit Default Swaps, 2006

11. 11 High commodity prices (in $) over the last 5 years can be attributed broadly to two factors: 1. Low real interest rates 2. Strong growth in China & other Emerging Markets

12. 12 Macroeconomic determinants of real commodity prices De trended G-4 GDP US real interest rate ConstantR 2 / / nobs All commodities price index 1992Q1-2010Q4.033 (.004) t = 7.6 -.047 (.010) t = - 4.7.962 (.037) t = 26.3.58/ / / 76 Real price of oil 1987Q1- 2010Q4.039 (.006) t = 7.0 -.072 (.011) t = - 6.7.879 (.046) t = 19.2.56/ / / 95 Real price of wheat 1990 Q1 – 2010 Q4.025 (.006) t = 4.0 -.028 (.013) t = - 2.1 1.247 (.052) t = 26.3.23/ / / 83 A.Saiki, Dutch Nat.Bk.

13. 13 Regardless what determines global commodity prices, small & developing countries must take them as given. Locally, in a small open economy,  price is exogenous in terms of $ =>  The exchange rate is mostly passed through to local-currency prices of the commodity.

14. 14 The effects of high agricultural prices  Consumers are hurting worldwide,  especially the poor,  for whom food takes a major bite out of household budgets.  for whom food takes a major bite out of household budgets.  Popular discontent over food prices has fueled political instability in some countries,  most notably in Egypt & Tunisia,  contributing to the Arab Spring uprisings.

15. 15 Macroeconomics  If an increase in the price of food is due to a boom in Aggregate Demand:  then it can be offset by tighter monetary policy,  aided by currency appreciation  if the exchange rate is flexible. GDP P AS AD ' AD

16. 16 Macroeconomics  An increase in the price of food imports is an adverse Aggregate Supply shock:  => higher inflation for any given level of GDP,  or lower GDP for a given level of inflation.  Especially if inflation is measured by CPI. GDP CPI AS AS ' AD

17. 17 Monetary policy can’t offset such a supply shock  If monetary policy is tightened to prevent the CPI from rising,  the result may be a severe recession.  The central bank cannot fight a shift in the terms of trade. GDP CPI AS AS ' AD AD'

18. 18 II. Microeconomics The record with policies to reduce the volatility of the prices of commodities  In theory, government stockpiles might be able to smooth price fluctuations, releasing stocks in times of shortage and buying when prices are low.  In theory, government stockpiles might be able to smooth price fluctuations, releasing stocks in times of shortage and buying when prices are low.  But the record in practice is not encouraging.

19. 19 The record with microeconomic policies to reduce the volatility of the prices of commodities, continued  In rich countries, the farm sector is usually powerful politically;  => stockpiles of food products are used to keep prices high rather than low.  The EU’s Common Agricultural Policy is a classic example –  disastrous for EU budgets, economic efficiency, and consumer pocketbooks.

20. 20 Microeconomic policies, continued  In developing countries, farmers often lack power.  Example: African countries adopted commodity boards for coffee & cocoa  at the time of independence.  The original rationale: to buy the crop in years of excess supply and sell in years of excess demand, thereby stabilizing prices.  In practice the price paid to cocoa and coffee farmers was always below the world price.  As a result, production fell.

21. 21 Microeconomic policies, continued  Politicians seek to shield consumers via price controls on staple foods & fuel.  But the artificially suppressed price usually  discourages domestic supply, and  requires rationing to domestic households.  Shortages and long lines can fuel political rage as well as higher prices can.  Otherwise, the policy can require imports in order to satisfy excess demand,  Thus price controls raise the world price even more.

22. 22 Microeconomic policies, continued  Some food-producing countries use export controls to insulate domestic consumers from a world price rise.  In 2008, India capped rice exports,  and Argentina did the same for wheat exports,  as did Russia in 2010.  Results:  Domestic supply is discouraged.  World prices go even higher.

23. 23 An initiative at the G20 meeting of agriculture ministers in Paris in June deserved to succeed:  Producing and consuming countries in grain markets should cooperatively agree to refrain from export controls and price controls.  The result would be lower world price volatility.  One hopes for steps in this direction, working through the World Trade Organization.

24. 24 Another initiative at the G20 meeting of agriculture ministers in Paris in June deserved to succeed:  Bio-fuel subsidies should be eliminated.  Ethanol subsidies, such as those paid to American corn farmers, do not accomplish avowed environmental goals, but do divert grain and so help drive up world food prices.  So far, the initiative has failed.  The US is the biggest obstacle.

25. 25 The overall lesson for microeconomic policy  Attempts to prevent food prices from fluctuating or to insulate consumers generally fail.  Even though enacted in the name of reducing volatility and income inequality, their effect is often different.  Better to accept volatility and cope with it,  e.g., well-designed transfers to the poor,  along the lines of Oportunidades or Bolsa Familia.

26. 26 III. Back to macroeconomics: Monetary policy for a food-importing country  The choice of monetary regime:  Fixed exchange rate or floating?  Inflation Targeting ?  See Frankel (2011) for references.

27. 27 Fixed vs. floating exchange rates  Fixed exchange rates have many advantages,  especially providing an anti-inflationary anchor for monetary policy.  But floating exchange rates have many advantages too,  especially accommodating trade shocks:  appreciating when the terms of trade improve,  depreciating when they worsen;  thus automatically stabilizing the balance of payments

28. 28 Floating  Bolivia might consider allowing more exchange rate flexibility  to the extent that the economy is in danger of overheating.  Some appreciation – a natural consequence of rapid growth and high real interest rates – would help hold down inflation.  But then some alternative nominal anchor would be needed,  to help meet central banks’ mandate of preventing inflation in the long run.

29. 29 Inflation Targeting (IT) became the favorite choice of economists  after the failures of currency pegs in the Emerging Market crises of the 1990s.  Three South American countries officially adopted IT in 1999, in place of exchange rate targets:  Brazil,  Chile,  Colombia.  Mexico had done so earlier, after the peso crisis of 1994.  Peru followed in 2002, switching from official money targeting.  Guatemala officially entered a period of transition to IT,  under a law passed in 2002.

30. 30  It apparently anchored expectations and avoided a return to inflation in Brazil, for example, despite two severe challenges:  the 50% depreciation of early 1999, (exit from the real plan), and  the similarly large depreciation of 2002, (Lula shock). In some ways, Inflation Targeting has worked well

31. 31 But the 2008-10 global financial crisis revealed some drawbacks of Inflation Targeting, much as the 1994-2001 EM crises revealed some drawbacks of exchange rate targeting.  One vulnerability is asset bubbles.  Another is terms of trade changes.  Another is terms of trade changes.

32. 32 The terms of trade (1)  A worsening of the terms of trade can take the form of either:  (1) A rise in global prices of imports such as food  (2) A fall in the global prices of export commodities.  (1) A CPI target, interpreted literally, forces the central bank to respond to an increase in $ import prices with a monetary tightening so severe that the currency appreciates in proportion  in order to hold local-currency prices of imports flat.  It is the opposite of accommodating the terms of trade,  causing likely problems for the balance of payments.

33. 33 The 4 inflation-targeters in Latin America show correlation ( currency value in $, import prices in $ )  > 0 ;  > correlation before they adopted IT;  > correlation shown by non-IT Latin American countries.

34. 34 Table 1: LACA Countries’ Current Regimes and Monthly Correlations of Exchange Rate Changes ($/local currency) with Dollar Import Prices Import price changes are changes in the dollar price of oil. Exchange Rate Regime Monetary Policy 1970-19992000-20081970-2008 ARG Managed floating Monetary aggregate target -0.0212-0.0591-0.0266 BOL Other conventional fixed peg Against a single currency -0.01390.0156-0.0057 BRA Independently floating Inflation targeting framework (1999) 0.03660.09610.0551 CHL Independently floating Inflation targeting framework (1990)* -0.06950.0524-0.0484 CRI Crawling pegs Exchange rate anchor 0.0123-0.03270.0076 GTM Managed floating Inflation targeting framework -0.00290.24280.0149 GUY Other conventional fixed peg Monetary aggregate target -0.03350.0119-0.0274 HND Other conventional fixed peg Against a single currency -0.0203-0.0734-0.0176 JAM Managed floating Monetary aggregate target 0.02570.26720.0417 NIC Crawling pegs Exchange rate anchor -0.06440.0324-0.0412 PER Managed floating Inflation targeting framework (2002) -0.31380.1895-0.2015 PRY Managed floating IMF-supported or other monetary program IMF-supported or other monetary program-0.0230.34240.0543 SLVDollar Exchange rate anchor 0.10400.05300.0862 URY Managed floating Monetary aggregate target 0.04380.11680.0564 Oil Exporters COL Managed floating Inflation targeting framework (1999) -0.02970.04890.0046 MEX Independently floating Inflation targeting framework (1995) 0.10700.16190.1086 TTO Other conventional fixed peg Against a single currency 0.06980.20250.0698 VEN Other conventional fixed peg Against a single currency -0.05210.0064-0.0382 * Chile declared an inflation target as early as 1990; but it also had an exchange rate target, under an explicit band-basket-crawl regime, until 1999. * Chile declared an inflation target as early as 1990; but it also had an exchange rate target, under an explicit band-basket-crawl regime, until 1999. LAC Countries’ Current Regimes and Monthly Correlations of Exchange Rate Changes ($/local currency) with $ Import Price Changes Table 1 IT coun- tries show correl- ations > 0.

35. 35 Why is the correlation between the import price and the currency value revealing?  Some central banks claim to target core CPI, i.e., excluding volatile food & fuel components.  But then the currency of a commodity importer should not respond to an increase in the world price by appreciating.  If anything, floating currencies should depreciate in response to such an adverse terms of trade shock.  When these IT currencies respond by appreciating instead, it suggests that the central bank is tightening monetary policy to reduce upward pressure on the CPI.

36. 36 Wanted !  New candidate variable for nominal target.  The economic variable should be:  simpler for the public to understand ex ante than core CPI, and yet and yet  robust with respect to supply shocks.  robust with respect to supply shocks.  “Robust with respect to supply shocks” means that the central bank should not have to choose ex post between 2 unpalatable alternatives:  an unnecessary economy-damaging recession or  an embarrassing credibility-damaging violation of the declared target.

37. 37 Trade shocks  If the supply shocks are terms of trade shocks, then the choice of CPI to be the price index on which IT focuses is particularly inappropriate.  Alternative: an output-based price index  such as an export price index, the GDP deflator, or PPI;  My preference: a price index for final sales.  The important difference is that  import goods show up in the CPI, but not in the output-based price indices,  and vice versa for export goods: they show up in the output-based prices but much less in the CPI.

38. 38 My proposal – Product Price Targeting (PPT): Call it IT, but target an output price index instead of CPI  I.e., target a broad index of all domestically produced goods.  I.e., target a broad index of all domestically produced goods.  The central bank in practice would not hit the target exactly,  in contrast to the way it could hit exactly a target for the exchange rate, the price of gold,  or even the price of a basket of 4 or 5 mineral or ag. commodities.  There would instead be a declared band for the PPI target, which could be wide if desired, just as with the targeting of the CPI, M1, or other nominal variables.  Open market operations to keep the price index inside the band  could be conducted in terms of either foreign exchange  or domestic securities. PPT

39. 39 One advantage of PPT (Product Price Targeting) if food import prices are volatile.  It does not let the currency become overvalued when import prices rise,  as a strict CPI target would. PPT

40. 40 IV. How do the answers change for a country that also exports commodities (minerals), with world prices that are equally volatile? If the $ price of mineral exports is highly correlated with the $ price of food imports, then it is not a terms of trade issue. Monetary policy is free to pursue domestic goals of growth and price stability which includes tightening/appreciation to prevent overheating.

41. 41 Table 2: Major Commodity Exports in LAC countries and Standard Deviation of Prices on World Markets Source: Global Financial Data Bolivia exports a range of minerals and other commodities. The leading export, natural gas, has the most variable price of all major commodities.

42. 42 RANK BY VOLATILITY Standard deviationof log ofprice indices Country / Region Terms of Trade (as reported by EIU) Calculated Terms/ Trade Export Price Index in US$ Import Price Index in US$ Latin America0.407 0.2100.373 Middle East & N.Afr.0.291 0.3600.246 Main CIS0.285 0.4370.256 Sub-Saharan Africa0.136 0.3170.221 Greater China0.046 0.2280.209 United States0.042 0.1120.149 North America0.021 0.1250.139 Eastern Europe0.017 0.2470.235 Appendix 1: Volatilities of terms of trade, export prices & import prices Lat.Am. has the highest terms of trade volatility, - <= import prices are also volatile even though oil exporting regions have more volatile export prices. and not as highly correlated with export prices (as Africa).

43. 43 The terms of trade (2)  A worsening of the terms of trade can take the form of either:  (1) A rise in global prices of imports such as food  (2) A fall in the global prices of exports.  (2) Accommodating the terms of trade means allowing the currency to rise or fall in value along with world prices of the export commodity.

44. 44 The high volatility of Latin America’s terms of trade makes it a good candidate for PPT.  Recap of advantages:  Relative to an exchange rate target,  PPT allows accommodation of trade, while yet preserving a nominal anchor.  Relative to an inflation target,  PPT allows appreciation when price of export commodity (minerals) goes up, not when price of import commodity (food) goes up –  whereas a CPI target gets it backwards. PPT

45. 45 The presentation draws on these references by the author  "The Effect of Monetary Policy on Real Commodity Prices" in Asset Prices and Monetary Policy, J.Campbell, ed. (U.Chicago Press, 2008): 291-327.  "The Effect of Monetary Policy on Real Commodity Prices" in Asset Prices and Monetary Policy, J.Campbell, ed. (U.Chicago Press, 2008): 291-327. "The Effect of Monetary Policy on Real Commodity Prices"Asset Prices and Monetary Policy"The Effect of Monetary Policy on Real Commodity Prices"Asset Prices and Monetary Policy  “The Natural Resource Curse: A Survey,” forthcoming in Beyond the Resource Curse, B.Shaffer, ed. (U.Penn.Press, 2011). NBER WP 15836. CID WP195.  “The Natural Resource Curse: A Survey,” forthcoming in Beyond the Resource Curse, B.Shaffer, ed. (U.Penn.Press, 2011). NBER WP 15836. CID WP195. The Natural Resource Curse: A Survey forthcomingShafferCID WP195The Natural Resource Curse: A Survey forthcomingShafferCID WP195  “A Comparison of Product Price Targeting and other Monetary Anchor Options, for Commodity-Exporters in Latin America," Economia, journal of LACEA, 2011. NBER WP 16362.  “A Comparison of Product Price Targeting and other Monetary Anchor Options, for Commodity-Exporters in Latin America," Economia, journal of LACEA, 2011. NBER WP 16362. A Comparison of Product Price Targeting and other Monetary Anchor Options, for Commodity-Exporters in Latin America,16362A Comparison of Product Price Targeting and other Monetary Anchor Options, for Commodity-Exporters in Latin America,16362  “How Can Commodity Exporters Make Fiscal and Monetary Policy Less Procyclical?” forthcoming, Natural Resources, Finance & Development, R.Arezki, T.Gylfason & A.Sy, eds. (IMF, 2011). HKS RWP 11-015.  “How Can Commodity Exporters Make Fiscal and Monetary Policy Less Procyclical?” forthcoming, Natural Resources, Finance & Development, R.Arezki, T.Gylfason & A.Sy, eds. (IMF, 2011). HKS RWP 11-015. How Can Commodity Exporters Make Fiscal and Monetary Policy Less Procyclical?T.Gylfason HKS RWP 11-015How Can Commodity Exporters Make Fiscal and Monetary Policy Less Procyclical?T.Gylfason HKS RWP 11-015  “Combating Agricultural Price Volatility,” Project Syndicate, June 27, 2011. Project SyndicateJune 27, 2011 Project SyndicateJune 27, 2011

46. 46

47. 47 Appendices  Appendix I: Simulations of the implications of PPT, compared to 6 other nominal anchors  Appendix II  Food price graph  Update of commodity-interest rate test

48. 48 Appendix I  Simulations of ability of PPT to stabilize the real prices of traded goods,  relative to CPI targeting  or exchange rate targets  or historical status quo. PPT

49. 49 Jamaica Price of aluminum in Jamaica Simulations under 6 alternative anchors 1) Any nominal target would have eliminated high 1970s inflation. 2) PPI target stabilizes domestic aluminum price better than does CPI. in logs

50. 50 Table 4: Variability of Export Prices under Alternative Currency Regimes (a) Standard Deviation of Nominal Export Prices

51. 51 Historical $ € SDR commodity CPI PPI regime peg peg peg peg peg peg Standard Deviation of Nominal Export Prices

52. 52 Begin with a hypothetical peg to the $.  In the case of Panama, same as historical peg.  In the case of Panama, same as historical peg.  In the other cases, we can simulate precisely what the price of copper, soy, etc. would have been in terms of pesos under the counterfactual,  by using the historical series for the peso/$ exchange rate:  if the peso historically depreciated against the $ by 1% in some given month, we know the price of copper would have been lower by precisely 1% if the peso had instead been pegged to the $.  if the peso historically depreciated against the $ by 1% in some given month, we know the price of copper would have been lower by precisely 1% if the peso had instead been pegged to the $.  In general, the $ pegs would have produced far more stable prices in domestic terms.  This is also true of the other 5 nominal anchors, and just illustrates the tremendous price instability experienced in the 1970s & 80s.  Changes in prices are also more stable under the anchors (Table 4b).

53. 53 The middle two columns of Table 4a show what variability of the commodity export prices would have been under an SDR peg or € peg, respectively.  Variability of the domestic price of the commodity export is often lower under the € peg than under the $ peg: for natural gas & oil; iron & steel; copper, aluminum & gold; bananas & sugar; and soy & beef.  A point often missed by observers who read too much into the $-invoicing of international commodity trade:  Although the use of the $ may introduce some dollar-stickiness in the very short run, it does not carry over to the medium run.  When the effective foreign exchange value of the $ rises, $ prices of these commodities tend to fall quickly.  The offset is not fully proportionate; but the point is that the prices are not more stable in terms of $ than in terms of €.  The offset is not fully proportionate; but the point is that the prices are not more stable in terms of $ than in terms of €.  Table 4(a) shows that in some cases (soy, coffee & beef), the SDR basket would stabilize commodity prices better than either the $ or €.

54. 54  Comparison in terms of ability to stabilize domestic price of the principle export commodity (Table 4a) :  The standard deviation of the domestic price of the export commodity is usually lower under PPT than under the CPI target.  In a few cases, it is less than half the size:  Jamaica for aluminum & Uruguay for beef. Th e last two columns in the tables: comparison of effects of a PPI target & a CPI target, which is intended as the contribution of Frankel (2011). Th e last two columns in the tables: comparison of effects of a PPI target & a CPI target, which is intended as the contribution of Frankel (2011).

55. 55  Stabilizing domestic prices with respect to the export commodity is far from the only criterion that should be considered in comparing alternative candidates for nominal anchor.  Stabilizing domestic prices with respect to the export commodity is far from the only criterion that should be considered in comparing alternative candidates for nominal anchor.  Another one is stabilizing domestic prices of other tradable goods.  A valid critique of PEP was that it transfers uncertainty that would otherwise occur in the real price of commodity exports into uncertainty in the real price of non-commodity exportables and importables.  This critique is particularly relevant if diversification of the economy is valued.  In Table 5 we conduct simulations for the domestic prices of import goods, under the same seven alternative regimes.

56. 56  Table 6 shows implications of the alternative regimes, for an objective function that is a weighted average of  the standard deviation of the real price of the commodity export  and the standard deviation of the real price of the import good.

57. 57 Average of leading commodity export standard deviation & oil price standard deviation. Table (6c): Standard Deviation of Real Prices Export Price SD & Import Price SD Averaged Real prices of TGs in general would be more stable under PPI target than CPI target.

58. 58 The export price target wins the competition to reduce relative price variability  by a margin  when we look at the level of nominal prices (Table 6a)  or the level of real prices (Table 6c) ; and  by a smaller margin when we look at nominal price changes (Table 6b).  by a smaller margin when we look at nominal price changes (Table 6b).  The three currency pegs are similar to each other, showing less price variability than the historical regime but more than the commodity peg.  The three currency pegs are similar to each other, showing less price variability than the historical regime but more than the commodity peg.  The PPI target usually gives less relative price variability than the CPI target.  Looking at real price variability in Table 6c,  the only exception is Peru;  the gain is substantial in the case of Jamaica and Uruguay,  smaller for the others. PPT

59. 59 Appendix II: Recent food prices, though high, have not been as high as 1974 in real terms. Nominal prices 2010=100 Real prices * = nominal in 2000

60. 60 If 2008-2010 is added to the regression of real commodity prices against real interest rates (alone),  the relationship weakens a bit.  R 2 falls to.26;  Coefficient is -9.4; t-statistic is still -18.6. 1951-2010 Real US interest rate Real $ commodity price index (Moody’s, excl.oil) A.Saiki, DNB

61. 61