Back Matter

Author(s):
Stefania Fabrizio
Published Date:
July 2012
    Share
    • ShareShare
    Show Summary Details
    Appendix 1: List of Low-Income Countries

    The group of LICs analyzed in this work is formed by the 70 countries eligible for support from the Poverty Reduction and Growth Trust for which data were available,1 which are, by region:

    Sub-Saharan Africa

    Benin, Burkina Faso, Burundi, Cameroon, Cape Verde, Central African Republic, Chad, Comoros, Democratic Republic of Congo, Republic of Congo, Côte d’Ivoire, Eritrea, Ethiopia, The Gambia, Ghana, Guinea, Guinea-Bissau, Kenya, Lesotho, Liberia, Madagascar, Malawi, Mali, Mozambique, Niger, Nigeria, Rwanda, São Tomé and Príncipe, Senegal, Sierra Leone, Tanzania, Togo, Uganda, and Zambia.

    Middle East and Europe

    Afghanistan, Armenia, Djibouti, Georgia, Kyrgyz Republic, Mauritania, Moldova, Sudan, Tajikistan, Uzbekistan, and Republic of Yemen.2

    Asia

    Bangladesh, Bhutan, Cambodia, Kiribati, Lao People’s Democratic Republic, Maldives, Mongolia, Myanmar, Nepal, Papua New Guinea, Samoa, Solomon Islands, Timor-Leste, Tonga, Vanuatu, and Vietnam.

    Latin America and Caribbean

    Bolivia, Dominica, Grenada, Guyana, Haiti, Honduras, Nicaragua, St. Lucia, and St. Vincent and the Grenadines.

    Appendix 2: Selected Economic Indicators
    GDP growth

    in percent
    Inflation

    in percent
    International reserves

    in months of imports1
    Fiscal balance

    in percent of GDP
    Current account

    balance incl. FDI

    in percent of GDP
    Gross public debt

    in percent of GDP
    200920102011201220092010201120122009201020112012200920102011201220092010201120122009201020112012
    Afghanistan, I.R. of20.98.27.17.2−12.27.78.43.2−1.60.90.0−1.8−0.24.81.5−1.3
    Armenia−14.12.14.64.33.57.38.83.35.74.74.54.2−7.7−4.9−3.8−3.1−7.4−7.8−5.5−4.640.239.241.541.4
    Bangladesh5.96.46.36.15.48.110.17.44.13.63.02.5−3.5−2.9−3.2−4.44.33.00.80.1
    Benin2.72.63.84.32.22.12.83.07.77.06.45.9−3.3−0.4−1.7−1.6−7.4−5.1−6.0−5.228.331.130.830.5
    Bhutan6.78.38.18.58.67.06.55.013.312.712.212.22.12.4−4.3−6.4−8.0−3.3−9.5−14.864.166.076.089.1
    Bolivia3.44.15.04.53.32.59.84.814.914.915.116.10.62.01.81.27.16.85.96.140.536.632.431.4
    Burkina Faso3.27.94.95.62.6−0.61.92.06.45.96.56.8−5.3−5.8−4.3−3.1−3.1−3.1−1.5−5.126.127.129.129.9
    Burundi3.53.94.24.810.76.48.712.55.04.94.85.1−8.6−3.9−3.1−3.1−15.3−12.5−15.5−16.248.250.049.044.2
    Cambodia−2.06.06.76.5−0.74.06.45.64.33.94.04.5−4.1−2.9−2.2−2.4−0.42.5−1.70.628.729.929.228.8
    Cameroon2.03.23.84.53.01.32.62.56.25.55.34.8−0.1−1.1−1.4−0.4−1.8−0.9−1.9−1.510.612.114.514.7
    Cape Verde3.75.45.66.41.02.15.04.94.03.53.13.4−6.3−10.6−10.2−9.0−7.7−4.5−7.1−5.868.673.273.978.3
    Central African Rep.1.73.34.15.03.51.52.82.65.13.93.63.8−0.1−0.8−2.00.2−6.0−7.0−5.3−5.036.841.937.032.8
    Chad−1.213.02.56.910.1−2.12.05.01.11.32.13.1−9.9−5.22.32.55.3−8.60.7−0.530.532.629.829.7
    Comoros1.82.12.23.54.82.75.83.35.35.44.84.62.72.1−1.6−0.9−6.4−8.3−12.0−10.855.351.847.142.9
    Congo, Dem. Rep. of2.87.26.56.046.223.514.812.50.51.21.31.5−5.21.2−7.7−6.2−1.24.65.19.6124.333.846.650.5
    Congo, Republic of7.58.85.07.04.35.05.95.27.06.512.618.54.816.019.520.312.023.526.227.957.223.822.022.2
    Côte d’Ivoire3.82.4−5.88.51.01.43.02.54.65.14.94.7−1.6−2.3−6.4−3.89.06.62.51.167.066.870.054.8
    Djibouti5.03.54.85.11.74.07.11.95.84.64.34.2−4.6−0.50.40.00.4−2.4−3.51.859.856.153.753.6
    Dominica−0.70.30.91.50.03.34.21.93.53.43.23.1−0.2−2.6−1.8−1.4−12.7−15.2−15.6−14.253.455.155.955.8
    Eritrea3.92.28.26.333.012.713.312.31.72.02.12.9−14.7−16.1−16.2−13.5−2.7−1.32.24.8145.7144.8134.4127.6
    Ethiopia10.08.07.55.536.42.818.131.22.22.32.82.4−0.9−1.3−2.1−4.0−2.3−1.1−2.9−6.032.236.739.434.1
    Gambia, The6.76.15.55.54.65.05.95.56.55.25.55.6−2.4−4.9−3.4−2.7−5.5−7.4−11.5−7.857.057.857.756.9
    Georgia−3.86.45.55.21.77.19.65.04.23.74.44.1−6.5−4.8−2.2−2.3−5.1−5.4−5.2−3.837.339.136.838.0
    Ghana4.07.713.57.319.310.78.78.72.63.13.43.8−5.8−7.4−4.2−2.32.50.2−2.2−0.536.237.438.137.5
    Grenada−7.6−1.40.01.0−0.33.44.23.24.03.32.32.3−5.3−3.1−5.3−4.8−11.1−12.2−13.6−12.498.298.6101.9104.3
    Guinea−0.31.94.04.24.715.520.613.82.91.52.02.0−7.2−14.2−12.21.0−15.2−18.0−17.1−6.077.088.685.582.2
    Guinea–Bissau3.03.54.84.7−1.61.14.62.08.26.06.06.12.9−0.2−1.9−1.6−4.3−4.5−5.2−6.7163.849.045.143.5
    Guyana3.34.45.36.03.03.75.85.84.34.54.44.5−3.5−2.7−2.4−2.5−1.10.7−3.2−6.761.260.260.459.3
    Haiti2.9−5.46.17.53.44.17.38.02.35.85.14.3−4.42.10.2−4.6−2.9−0.1−1.2−4.727.717.112.619.0
    Honduras−2.12.83.53.58.74.77.87.92.62.93.03.1−4.7−2.9−3.1−2.50.0−1.0−1.6−1.424.126.327.627.8
    Kenya2.65.65.36.110.64.112.17.43.53.03.03.3−5.2−6.0−5.4−4.8−4.4−5.7−7.1−6.847.650.451.250.2
    Kiribati−0.71.83.03.58.8−2.87.75.0−12.6−9.1−16.3−16.1−29.6−22.4−30.9−26.7
    Kyrgyz Republic2.9−1.47.06.06.87.819.19.44.63.84.03.8−1.3−6.1−8.0−7.74.72.3−3.0−2.758.062.655.254.6
    Lao People’s Dem. Rep7.67.98.38.40.06.08.76.72.22.01.92.0−6.5−4.2−2.3−1.6−6.9−7.4−5.8−4.361.760.155.352.2
    Lesotho3.13.65.15.15.93.46.55.11.72.33.43.1−3.9−4.6−14.90.60.6−12.5−17.2−2.338.434.138.241.9
    Liberia4.65.66.99.47.47.38.81.62.22.42.12.2−12.0−6.5−3.6−3.3−20.8−3.31.5−0.1194.013.413.314.5
    Madagascar−3.70.61.04.79.09.210.38.54.03.22.93.0−3.1−0.4−1.3−2.3−11.9−4.1−5.3−5.733.734.036.436.0
    Malawi9.06.54.64.28.47.48.611.50.81.91.20.9−5.01.5−4.2−3.8−4.4−0.2−4.3−2.340.135.138.641.2
    Maldives−7.57.16.54.64.04.712.18.42.12.42.01.1−20.8−16.0−15.0−13.6−14.7−20.7−19.7−18.152.259.362.970.5
    Mali4.55.85.35.52.21.32.82.36.14.64.24.3−3.3−1.5−2.3−1.7−1.2−5.6−5.3−4.424.229.630.027.2
    Mauritania−1.25.25.15.72.26.36.26.31.01.01.41.6−5.1−1.9−2.8−3.8−10.8−5.2−2.5−3.8101.586.262.064.2
    Moldova−6.06.97.04.50.07.47.97.83.93.63.83.9−6.3−2.5−1.9−1.2−6.3−5.0−6.2−6.429.126.623.621.7
    Mongolia−1.36.411.511.86.310.210.214.33.94.46.78.0−5.01.20.9−1.91.910.30.6−4.1
    Mozambique6.36.87.27.53.312.710.87.25.24.64.44.6−5.5−3.9−6.1−6.8−3.3−2.2−4.7−4.241.537.839.042.7
    Myanmar5.15.55.55.58.28.26.73.74.74.55.37.1−3.6−3.9−3.2−2.81.41.92.84.844.642.845.247.7
    Nepal4.44.63.53.812.69.69.58.05.95.25.15.0−3.0−1.4−2.1−3.14.4−2.2−0.6−0.139.535.933.933.5
    Nicaragua−1.54.54.03.33.75.58.38.23.43.33.23.0−1.9−0.5−0.2−0.5−5.2−6.7−5.5−8.280.980.377.476.0
    Niger−0.98.05.512.51.10.94.02.02.83.03.33.9−5.5−2.5−2.2−0.9−11.1−2.8−10.6−7.415.716.217.717.8
    Nigeria7.08.76.96.612.513.710.69.08.76.67.88.2−10.2−8.50.42.216.48.716.514.215.217.315.716.3
    Papua New Guinea5.57.09.05.56.96.08.48.74.44.34.86.0−9.6−0.31.00.4−2.8−4.7−4.3−3.0
    Rwanda4.17.57.06.810.32.33.96.55.45.25.85.20.30.4−1.5−3.8−5.1−5.2−3.8−8.023.023.224.426.1
    Samoa−5.1−0.22.02.114.4−0.22.93.06.4−4.1−11.3−8.2−5.1−3.1−8.1−12.7−13.3
    São Tomé & Príncipe4.04.55.06.017.013.311.47.44.83.65.65.2−16.9−11.0−17.41.5−18.8−25.3−18.6−24.131.071.482.484.1
    Senegal2.24.24.04.5−1.71.23.62.55.04.24.95.4−5.0−5.2−6.2−5.4−4.7−3.9−5.4−5.232.038.040.041.5
    Sierra Leone3.25.05.151.49.217.818.011.04.92.92.82.7−3.2−6.9−5.1−2.3−4.4−2.0−7.6−3.761.864.761.133.2
    Solomon Islands−1.26.55.66.17.11.06.05.03.35.76.47.01.65.91.71.5−2.03.9−4.0−6.928.925.723.120.6
    St. Lucia−1.34.42.02.6−0.23.32.52.52.72.92.82.8−4.0−5.9−8.2−5.80.5−2.4−7.6−8.264.566.172.078.8
    St. Vincent & Grens.−2.3−1.8−0.42.00.40.62.51.42.74.03.43.3−3.2−5.7−3.2−3.7−14.9−16.1−14.4−12.564.966.869.571.2
    Sudan4.66.5−0.2−0.411.313.020.017.51.10.91.11.3−4.8−3.2−2.8−3.0−8.5−2.3−3.1−3.983.671.678.287.3
    Tajikistan3.96.56.06.06.56.513.610.01.41.61.71.8−5.2−3.0−4.9−4.2−5.62.4−2.0−4.136.636.737.038.6
    Tanzania6.76.46.16.111.810.57.09.45.04.84.24.2−4.8−7.0−8.5−6.5−8.2−6.9−6.7−8.037.140.145.048.9
    Timor Leste, Dem. Rep. of12.96.07.38.60.14.910.56.02.9239.3238.6210.2179.3245.4227.1196.9167.6
    Togo3.23.73.84.41.93.24.02.84.94.03.33.1−2.8−1.6−3.9−4.1−6.3−6.5−6.8−6.767.832.327.627.3
    Tonga−0.30.31.41.73.44.05.94.8−2.6−3.6−3.2−3.5−11.1−9.4−11.3−11.2
    Uganda7.25.26.45.514.29.46.516.96.35.94.64.3−2.4−5.0−7.6−6.5−3.1−4.21.4−2.222.223.623.025.3
    Uzbekistan8.18.57.17.014.19.413.111.811.19.712.915.83.12.73.34.64.710.811.210.311.010.012.614.0
    Vanuatu3.52.23.84.24.32.82.22.94.84.74.44.1−0.7−2.7−1.3−0.9−4.0−3.0−3.6−4.4
    Vietnam5.36.85.86.36.79.218.812.12.31.41.52.2−9.0−5.7−4.0−3.80.82.11.62.451.252.850.348.1
    Yemen, Republic of3.98.0−2.5−0.53.711.219.018.07.75.53.31.9−10.2−4.0−7.1−6.1−10.9−7.1−8.2−7.149.940.642.944.4
    Zambia6.47.66.76.713.48.59.17.54.03.33.53.9−2.6−3.1−3.1−6.07.58.49.97.425.624.625.128.7
    Medians
    All LICs3.35.45.25.54.65.07.85.94.33.94.03.9−4.1−2.9−3.1−2.9−4.1−3.3−4.3−4.441.539.140.041.5
    Sub–Saharan
    Africa3.65.55.15.86.64.56.56.04.93.93.94.1−4.3−3.9−3.8−2.9−4.4−4.2−5.2−5.137.835.938.436.7
    Asia4.46.46.36.16.36.08.45.64.24.34.64.7−3.6−2.9−2.3−2.8−2.0−2.2−3.6−4.147.947.847.847.9
    Middle East and Europe3.46.45.35.23.67.311.38.64.43.83.93.9−5.2−3.1−2.8−3.1−6.0−3.7−3.3−3.845.039.942.242.9
    Latin American and Caribbean−1.32.83.53.33.03.45.84.83.43.43.23.1−3.5−2.7−2.4−2.5−2.9−2.4−5.5−8.261.260.260.459.3
    Net oil exporters4.67.93.86.64.36.08.46.74.65.14.84.7−6.5−3.2−1.4−0.4−1.8−2.3−1.9−1.553.536.636.337.1
    Net oil importers3.25.25.35.54.74.97.75.84.13.83.93.9−4.0−2.9−3.2−3.1−4.4−3.3−5.2−5.040.539.140.041.5
    Sources: IMF, World Economic Outlook database; and IMF staff calculations.

    Next year’s imports of goods and services.

    Appendix 3: How Do Global Commodity Prices Behave? A View from the Literature

    Research suggests that the impact of shocks on commodity prices is long-lasting though the persistence varies. Cashin, Liang, and McDermott (1999) find that the typical half-life of shocks to commodity prices is five to eight years, though depending on the commodity, it could vary from permanent to as short as one year. According to Cashin, McDermott, and Scott (2002), the likelihood of an end to a slump/boost in prices is independent of the time spent in the slump/boost, though price increases have become more persistent than in the past.

    Higher persistence of oil price increases in the 2000s could be due to increasing oil scarcity. As discussed in the fall issue of the World Economic Outlook (IMF, 2011d), the strong momentum in oil demand growth, in particular in emerging economies, combined with the recent downshift in oil supply trends, has likely resulted in oil scarcity. The analysis suggests that the period of higher-than-average prices could be rather long, as the effect from resource scarcity on price is strengthened by low elasticity of oil demand and supply to the price.

    Movements in prices of commodities are usually synchronized, possibly driven by several factors. The recent episodes of surges in commodity prices have been widespread across the whole range of commodities, raising questions over co-movements and their determinants in the prices. Several factors have been considered as determinants of fluctuations in commodity prices. Svensson (2008) and Wolf (2008) consider global demand as a determinant of commodity prices. Frankel (2008) believes that real interest rates on bonds are a driver for commodity prices. Vansteenkiste (2009) finds that oil prices, the U.S. dollar effective exchange rate, the real interest rate, and increasingly global demand are important in explaining commodity price movements. On the other hand, Lombardi, Osbat, and Schnatz (2010) find evidence in support of only world industrial production and the U.S. dollar real effective exchange rate’s significance as determinants in commodity price fluctuations.

    Appendix 4: Methodology for Vulnerability Indicator Underlying the Growth Decline Model

    This appendix describes the methodology developed by the IMF staff (IMF, 2011a) to capture LICs’ vulnerabilities to a growth recession when hit by exogenous shocks. The approach taken is to identify observations (country-years) in which a country is hit by a large external shock. Policies and structural variables that predict whether the country also experiences negative growth are then identified. Such countries are considered “vulnerable.”

    Dependent Variable: Identifying Shock Episodes and Real Output Drops

    Large negative shocks events in LICs are identified if the annual percentage change of the relevant variable falls below the 10th percentile in the left tail of the country-specific distribution.3 In particular, shock episodes include one or more of the following six shocks occurring over the period 1990–2009: (i) external demand; (ii) terms of trade; (iii) FDI; (iv) aid; (v) remittances; and (vi) climatic shocks (large natural disasters).4 Within the sample of identified shock episodes, a crisis is defined as a large real output drop when the following two conditions hold: (i) the post-shock two-year average (t and t+1) level of real output per capita falls below the pre-shock three-year trend; and (ii) output per capita growth is negative at time t.

    Selection of Vulnerability Indicators

    Several indicators were considered, based on empirical studies of growth declines and protracted growth slowdowns in the event of exogenous shocks. Those that were retained can be constructed for a majority of LICs and capture the flow and stock vulnerabilities in the external and public sectors as well as institutional weaknesses identified in past studies and IMF surveillance. These include lagged values of:

    • Overall economy and institutions: real GDP growth; the World Bank’s Country Policy and Institutional Assessment index; and the Gini coefficient;

    • External sector: reserve coverage (gross international reserves in months of imports) and real growth of exports of goods and services; and

    • Fiscal sector: overall fiscal balance in percent of GDP; public debt in percent of GDP; and real government revenue growth.

    Methodology

    The approach examines a range of indicators one by one to identify variables and thresholds that separate crisis and non-crisis cases in a given data set. For each of the individual indicators, the approach involves searching for a split that minimizes the combined percentages of missed crises (Type I error) and false alarms (Type II error). Thresholds that yield the best split map indicator values into zero-one scores. These indicators are then aggregated into sectoral indices using weights that depend on the individual indicator’s ability to discriminate between crisis and non-crisis cases.

    The overall vulnerability index, which ranges from zero (low vulnerability) to one (high vulnerability), is a summary measure of underlying vulnerability. Indicator-based ratings (“low,” “medium,” or “high”) are derived from the vulnerability index.

    Appendix 5: Methodology of the Tail-Risk Scenario of Higher Global Commodity Prices

    The scenario of higher global commodity prices uses market expectations in commodity futures options. Risk-neutral probability density functions are derived for expected future spot prices over different time horizons. The values at the upper two standard deviation level implied by the risk-neutral density functions are used to construct commodity price levels in the adverse scenario. Given that futures options are traded only for a small set of commodities, we construct probability density functions for the following major commodities: crude oil, copper, corn, and soybeans. Estimated price deviations of crude oil from the baseline, 20.7 percent in 2011 and 48.3 percent in 2012, correspond to $124.5 and $148.3 a barrel in 2011 and 2012, respectively, compared to the baseline prices of $103.2 and $100. We then apply the estimated price deviations of copper from the baseline (21.3 percent in 2011 and 36.3 percent in 2012) to all other base metals, and the average deviations for corn and soybeans (25.1 percent in 2011 and 30.8 percent in 2012) to the other food commodities. Commodity price levels under this adverse scenario are somewhat higher than those in April 2011 WEO, reflecting increased volatility in commodity markets since late April, although not far out of the ordinary relative to historical trends.

    Impact of Higher Global Prices for Commodities on LICs’ Growth

    The methodology for analyzing the impact of higher global prices for commodities on LICs’ growth is similar to that used to analyze the impact of the slower global growth scenario on LICs’ growth (Appendix 6), except that only the impact from the terms of trade is considered (changes in terms of trade are found to impact growth only for the most open economies (top quartile of the distribution of LICs).

    Impact of Higher Global Prices for Commodities on Inflation

    The country-specific impact of higher global food and oil prices on domestic inflation is derived using regression analysis. For the sample period January 1996 to March 2011, monthly year-over-year domestic headline inflation is regressed on contemporaneous and up to 12-month lags of international food and oil price inflation (in U.S. dollars), contemporaneous and one lag of changes in the nominal exchange rate, and time dummies. For each country, the optimal lag length for international food and oil price inflation is chosen via the Akaike information criterion. A symmetric lag structure for food and oil price inflation is assumed for simplicity. For the optimal lag structure, the cumulative multipliers for international food and oil prices are calculated as the sum of the contemporaneous and lagged coefficients on international food and oil price inflation, respectively. If the estimated value of the cumulative multiplier is not different from zero with statistical significance, or negative, or data are not available, the average cumulative multiplier for the peer group of LIC commodity exporters or non–commodity exporters is used.

    Impact of Higher Global Prices for Commodities on the External Sector

    A first-round effect on a country’s trade balance stemming from a shock to global commodity prices is estimated by multiplying the change in the price (index) of individual commodities grouped in five categories (crude oil, food, metals, agricultural raw materials, and others) and the country’s commodity export and import weights derived from trade data for the period 2005–08.

    Two types of price changes are considered. First, price changes are taken from the IMF’s October 2011 baseline projections relative to the October 2010 WEO baseline commodity exports. Second, the impact of an adverse scenario of further increases in global prices for commodities is analyzed (see the discussion earlier in the paper).

    The analysis of the effect on the overall balance of payments is analogous to the approach followed in the double dip growth scenario discussed above and uses (i) financing gap, (ii) reduction in reserves, and (iii) potential for import compression.

    • The external financing gap in percent of GDP is defined as the difference in the overall balance of payments between the scenario and the baseline external balances, in percent of GDP.

    • The level of reserves after the shock, in months of imports, is calculated under the assumption that shortfalls in financing on account of the shock are entirely absorbed by a commensurate reduction in reserves. Countries are assumed to make use of available reserve buffers up to a reserve floor of three months of imports.

    • Import compression in percent of GDP is the change in imports that would compensate for the financing shortfall in order to maintain reserves at three months of imports. Import compression is zero if, after the shock, the estimated level of reserves (from the second metric) is above three months of imports.

    Impact of Higher Global Prices for Commodities on the Fiscal Sector

    Higher international food and oil prices (compared to the 2011–12 projections in the baseline WEO projections) are assumed to have the following additive impacts on the fiscal sector: (i) unchanged policies effects and (ii) fiscal impact of measures.

    Unchanged policies effects

    These effects on the budget are estimated following the methodology to assess “automatic stabilizers” in the case of a growth shock.

    We calculate the effects by estimating revenue and expenditure country-specific elasticities (separately for fuel price and fuel price changes) based on historical data (typically for the last decade). The elasticities are estimated using fixed-effects panel regressions, conditional on other prices and growth (and controls for time dummies).5 We then apply the elasticity to the assumed increase in international prices and estimate a new set of revenue and expenditure projections for 2011 and 2012 for the 62 LICs in the sample. Results of this simulation are compared with the baseline, the difference being the size of the impact of this first component.

    Policy response

    We also estimate the impact of policies introduced in response to the recent surge in prices. To do so we have collected information on tax and expenditure measures adopted by countries in response to price increases during the last price shock (2007–08). We use the fiscal impact of revenue and expenditure measures separately for food and fuel prices to calculate country-specific semi-elasticities, which measure the cost (in percent of GDP) of measures adopted in response to a 1 percent increase in prices. We multiply these semi-elasticities by the assumed increase in international prices and get the fiscal impact of these measures in each LIC. We use the 2007–08 data in our first round of results. For countries with no information, we use the median semi-elasticity.

    Overall impact

    These are the sum of the unchanged policies impacts and policy responses. Results are provided for each country in the sample.

    Appendix 6: Methodology of the Tail-Risk Scenario of Lower Global Growth6

    The downside scenario, based on the Global Integrated Monetary and Fiscal (GIMF) model developed by the IMF, reflects exclusively the global macroeconomic impact of bank capital being severely eroded by sovereign debt distress in the euro area, which translates into lower growth, and therefore global demand, in different regions of the world, including the United States, the euro area, Japan, and emerging Asia (Table A6.1).

    Table A6.1.GDP Growth Projections under the Downside Scenario
    BaselineDownsideDifference
    201120122011201220112012
    World4.04.02.62.4−1.4−1.6
    United States1.51.80.61.0−0.9−0.8
    Euro area1.61.1−1.8−2.9−3.4−4.0
    Japan−0.52.3−1.31.6−0.8−0.7
    Emerging Asia17.97.67.57.2−0.4−0.4
    Latin America24.23.93.53.2−0.7−0.7
    Rest of the world34.14.12.71.8−1.4−2.3

    Includes China, Hong Kong SAR, India, Indonesia, South Korea, Malaysia, the Philippines, Singapore, Taiwan Province of China, and Thailand.

    Includes Brazil, Chile, Mexico, Colombia, and Peru.

    Includes Argentina, Australia, Bolivia, Bulgaria, Canada, Denmark, Estonia, Israel, New Zealand, Norway, Russia, South Africa, Sweden, Switzerland, Turkey, United Kingdom, and Venezuela.

    The macro growth impact is calculated in a two-step process. In the first step, a closed economy dynamic stochastic general equilibrium (DSGE) model with a banking sector is used to estimate the own-country GDP effects of the banking sector responding to restore its regulated capital adequacy ratio. In the second step, the higher costs of borrowing, for both firms and households, are then used in the GIMF model to replicate the GDP impacts for each region. Higher lending rates reduce loan volumes and raise returns, shrinking bank balance sheets and raising profitability sufficiently to restore capital adequacy ratios. The resulting sharp fall in investment reduces aggregate demand globally. The larger borrowing costs are imposed in all regions to capture the full global impact.

    Further value-at-risk simulations convert the associated lower global growth into changes in global commodity prices, which in turn are translated into LIC-specific export and import prices.

    Impact of the Slower Global Growth Scenario on LICs’ Growth

    Weak global demand stemming from advanced and emerging market countries and the attendant effect on commodity prices are key transmission channels for LIC growth prospects.

    The analysis is carried out in two stages. First, the elasticity of LICs’ growth to its main trading partners (both advanced and emerging market countries) is estimated using a growth spillover regression for a panel of commodity exporters and non–commodity exporters. Second, alternative projections for the global economy and six relevant regions, along with alternative country-by-country projections for the terms of trade, are used to estimate the potential downside growth impacts for LICs. This calculation makes use of information on trading patterns taken from the IMF’s Direction of Trade Statistics.

    LICs’ elasticities of growth to partner country growth and terms of trade

    Separate regressions are run for commodity exporters and non–commodity exporters, as the former are less affected by partner country growth. Partners’ growth is calculated using weighted averages of trading partners’ GDP growth, with the weights based on the 2008 bilateral trade flows from the Direction of Trade Statistics. For non–commodity exporters, the model finds the elasticity of growth to partners’ growth to be 1.2. In addition, the elasticity of growth to changes in the terms of trade is also found significant (elasticity of 0.1) for the most open countries (i.e., countries in the top quartile in trade openness, measured in terms of the ratio of exports plus imports to GDP).

    For commodity exporters, in the post-1995 period, growth prospects appeared to be aligned with growth in the dynamic emerging market countries (Brazil, Russia, India, and China). The elasticity of GDP growth to these emerging markets’ growth is estimated at 0.35.

    Application to individual LICs

    The IMF Research Department’s Global Projections Model (GPM) provides a baseline and an alternative scenario for growth in six regions of the world (United States, euro area, Japan, emerging Asia, Latin America, and remaining GPM countries). Using trade data for these six regions, average partners’ growth is calculated for each LIC in both the baseline and the alternative scenario.

    Impact of the Slower Global Growth Scenario on the External Sector

    The external sector module for the slower global growth scenario estimates the impact on LICs’ gross financing gaps, the adequacy of reserve buffers, and the potential for import compression. The GPM scenario of lower world growth is assumed to affect the balance of payments of individual LICs through four spillover channels affecting both the current account and the financial account. The main assumptions are:

    • Variation in export prices and import prices (of goods and services) affect the dollar value of exports and imports with an elasticity of 1.

    • A reduction in growth affects the external demand facing each LIC, and the elasticity of export volumes to external demand is equal to 3 (based on Dabla-Norris, Espinoza, and Jahan, forthcoming). The change in external demand is computed using weighted averages of trade partners’ real GDP growth, where the weights are based on the 2008 Direction of Trade Statistics exports flows.

    • A reduction in growth affects remittances from source countries with an elasticity of 1.5 (based on Lueth and Ruiz-Arranz, 2008). Partners’ growth is computed using the weights from bilateral remittances data (2006 World Bank remittances flows and Organization for Economic Cooperation and Development data).

    • A reduction in growth affects FDI from source countries with an elasticity of 21 the year of the shock and an elasticity of 35 the year after the shock (based on Dabla-Norris, Honda, and others, 2010). Partners’ growth is computed using weights derived from bilateral (Organization for Economic Cooperation and Development) FDI data.

    The effect on the overall balance of payments is then analyzed using three metrics: (i) financing gap, (ii) reduction in reserves, and (iii) potential for import compression.

    • The first metric is the external financing gap in percent of GDP, defined as the difference in the overall balance of payments between the scenario and the baseline external balances, in percent of GDP.

    • The second metric computes the level of reserves after the shock, in months of imports, under the assumption that shortfalls in financing on account of the shock are entirely absorbed by a commensurate reduction in reserves. Countries are assumed to make use of available reserve buffers up to a reserve floor of three months of imports.

    • The third metric computes the change in imports divided by GDP (import compression) that would compensate for the financing shortfall in order to maintain reserves at three months of imports. Import compression is zero if, after the shock, the estimated level of reserves (from the second metric) is above three months of imports.

    Impact of the Slower Global Growth Scenario on the Fiscal Sector

    This section presents the methodology used in the fiscal module to assess the public finance implications of an adverse shock to global growth for LICs. It describes the estimation of revenue and expenditure buoyancy ratios for the real GDP growth rate shock and the terms-of-trade shock associated with this adverse scenario. Using these results, fiscal impacts are calculated for each country under the shock scenario for 2011.

    Three indicators of fiscal vulnerability are constructed using these simulation results along with an associated heat map that classifies countries into low, medium, or high vulnerability. A combined overall fiscal vulnerability ranking is also presented.

    Growth Shock

    Revenue buoyancy (BR) is defined as: BR,t = (RtRt−1)/(gtgt−1), where Rt is the revenue/GDP ratio for year t and gt is the real GDP growth rate for year t. The revenue buoyancy for each country is calculated as follows:

    • Median country-specific buoyancy ratios are estimated for revenue and, if available, also tax revenue, on the basis of 2007–10 data.

    • For countries with median revenue buoyancy estimated to be outside the 0.25–1.5 range, we replace the country-specific buoyancy with the estimated country-specific median tax revenue buoyancy. If this is not available (or its value is outside the 0.25–1.5 range), we use the median revenue buoyancy estimated for the sample (excluding countries that have estimated values outside the range).

    The country-specific revenue buoyancy is then multiplied by the corresponding output growth shock (output growth in the shock scenario in 2012 minus the baseline 2012 growth rate) to calculate its impact on the revenue/GDP ratio in 2012 compared to the no-shock baseline. The 2012 expenditure/GDP ratio in the shock scenario is derived by assuming that nominal expenditures remain the same as in the baseline for 2012, while GDP is consistent with the shock scenario assumptions for the same year. This assumes no expenditure adjustment by countries in the face of the shock.

    Terms-of-Trade Shock

    For the terms-of-trade shock, the expenditure and revenue effects are calculated in a similar fashion: we use four price indices (fuel export price, fuel import price, nonfuel export price, nonfuel import price). For each price index we calculate the buoyancy as BX,i,t = (XtXt−1)/pi,t, where Xt is the revenue/ GDP (or expenditure/GDP) ratio in year t and pi,t is the growth rate of the price index i in year t. Given the significant decline in global commodity prices from 2008 to 2009, which is similar to the terms-of-trade shock here, the buoyancy coefficients are calculated using 2009 data. However, if a country has estimated buoyancy greater than four or less than the 25th percentile value for a given price index, we use 2008 data instead. For the countries with buoyancy rates based on 2008–09 data that are outside the range, we use 2010 data.

    To calculate the impact on revenue/GDP (or expenditure/GDP) in 2012 under the terms-of-trade shock, the assumed price shock under the adverse scenario for each price series is multiplied by the respective buoyancy ratio and then weighted and aggregated using the following weights: Wi = (μii)/ [ Σ (μii)] (where μ is the mean and σ is the standard deviation of the index-specific buoyancy ratio, and i indicates the ith price index).

    Combined Shock Impact

    The growth shock and the terms-of-trade shock impacts on the budget components are then combined using the following weights, for all countries: 0.9 for the growth shock and 0.1 for the terms-of-trade shock.7

    Fiscal Indicators

    We use three indicators to measure fiscal vulnerability:

    • Change in Fiscal Balance (CFB): defined as the overall fiscal balance in 2012 under the shock scenario minus the baseline projected (pre-shock) 2012 balance:

      where Rs is the revenue/GDP ratio after the shock impact and Rb is the projected baseline 2012 revenue/GDP ratio (similarly for expenditure/GDP ratios E). This indicates the additional net financing needs to maintain the baseline nominal expenditure plans.

    • Revenue/GDP Growth Rate (or Revenue Growth) (RG): this is the relative change in the revenue/GDP ratio after the shock compared to the 2012 baseline (in percent):

      This indicates how severe the impact of the shock is on a country’s revenue/GDP ratio, adjusting for severity of underlying shock, and the degree of fiscal rigidity in the budget.8

    • Fiscal Space (FS) (see Escolano, 2010): this is defined as the difference between the baseline 2012 primary balance (PB) and the constant PB that is needed to achieve a target debt/GDP ratio of 40 percent in 2031.9 It is an indicator of initial vulnerability intended to capture how much flexibility authorities may have in employing countercyclical fiscal policy when the economy is hit by a negative shock.

    The fiscal space calculation requires the initial (2012) primary balance (PB0) (overall balance plus interest expense), the initial (2011) debt/GDP ratio (D0), the target debt/GDP ratio (DN) (in 2030), and the average projected implied real interest rate (r) and output growth rate (g). The target debt/GDP ratio of 40 percent in 2031 is applied to all countries. Given the fiscal space (FS) definition, we have FS = PB0PB*, where PB* is the constant primary balance needed in each year t = 1, …, N to achieve the target debt/GDP ratio in year N (with N = 20). PB* can be calculated as

    where λ = (rg)/(1 + g). It is assumed that the difference r minus g converges to zero in 30 years (year 2041). In the first 10 years (2012–21) r minus g is kept constant at its 2012–16 average for each country in the calculation. Starting from the average value in 2012, a linear trend in the difference is assumed from 2022 until r minus g converges to zero in 2041.

    A country with an initial positive primary balance will have more fiscal space. Also, a country with more negative λ (i.e., average lower real interest rate and/ or higher output growth) or a lower initial debt/GDP ratio will have a larger fiscal space.

    Appendix 7: Methodology for Assessing the Poverty Impact of Higher Food and Fuel Prices

    The impact of an increase in food prices on poverty is estimated as follows:

    • Estimate the pass-through of exogenous increases in international food prices to domestic prices. This uses information from Ivanic, Martin, and Zaman (2011) on changes in international food prices and retail food prices from June to December 2010 for a sample of 28 developing countries, of which 17 are included in the Vulnerability Exercise for LICs. Country-specific pass-through rates transform changes in international prices into price changes at the retail level, assuming unchanged policies in the short term.

    • Estimate the effects on poverty. The World Bank study estimates elasticities of retail food price changes on poverty incidence for the 28 developing countries,10 based on household survey data. These elasticities are applied to the country-specific projected retail price changes to simulate the impact on poverty incidence.11

    • For LICs for which information on retail food prices and elasticities is not available, the pass-through is assumed to be 50 percent and the elasticities to be the average of similar countries based on region, GDP per capita, and agriculture as a share of GDP.

    The impact of an increase in fuel prices on poverty is estimated as follows:

    • Estimate the pass-through of increases in international fuel prices to retail prices. This is based on two studies by the IMF’s Fiscal Affairs Department, one that provides estimates on the pass-through rates for 20 countries (Arze del Granado, Coady, and Gillingham, 2010), and another one for Middle East, North Africa, and Central Asian countries (Coady and Antonio, 2011). The pass-through estimates for the Middle East and North Africa are based on data for 2008–10, while estimates for other regions are based on data prior to 2008.

    • Estimate the direct effect of higher fuel prices by decile. Multiply the share of household consumption of fuel by the projected percentage price increase.

    • Estimate second-round effects of higher fuel prices on domestic nonfuel prices. Arze del Granado, Coady, and Gillingham (2010) provide estimates of input-output tables (production of other goods and services using fuel as input) for a subset of the countries in their sample. Here other goods and services include the categories: food, non-food, and nonfuel. Elements in the input-output tables measure the increases in the prices of other goods and services when the price of fuel increases by 1 percent. These input-output tables are combined with data on household consumption share of other goods and services by decile to compute the indirect impacts of fuel prices on household consumption by decile.

    • Combine the direct and indirect effects to get the overall impact on household consumption by decile. The overall impact is translated into a new poverty line in nominal prices, which is equivalent to the poverty line of $1.25 per day in real terms.

    • Estimate the impact on poverty of higher fuel prices. For household expenditure distributed as in Figure A7.1, OA denotes the poverty line of $1.25 per day; the area under curve OB denotes the poverty rate under the baseline; AC denotes the increase in the poverty line in nominal terms as a result of higher fuel prices; and ABDC approximates the poverty impact of higher fuel prices when AC is small. AB can be estimated by (baseline poverty)/ OA* α, where the value of α depends on how the baseline poverty rate is approximated. For example, if we use triangle OAB to approximate baseline poverty, α would be 2. For the distribution shown in Figure A7.1, a would be greater than 2.

    • Pass-through rates, consumption shares, and input-output tables for LICs for which data are not available are imputed based on countries with similar characteristics.12

    Figure A7.1.Household Expenditure Distribution and Poverty Estimation

    Appendix 8: Monetary Policy Responses to Food and Fuel Price Shocks in Low-Income Countries

    Analytical Considerations

    The standard monetary policy advice has been to allow for the direct effects of increases in world food and energy prices on headline inflation, but not for the indirect effects that may be present in the response of wages and, in turn, core prices. This advice, whose theoretical underpinnings can be found in the benchmark New Keynesian model with nominal price rigidities (Blanchard and Gali, 2007), has taken different forms. Some describe it as allowing for first-round effects, but not for second-round effects; while others refer to it as not allowing for pass-through into wages and core inflation. But despite the different forms, the logic behind the policy advice is the same: avoid persistent effects on inflation. The first-round or direct effects—which also include the effects associated with the use of oil as an intermediate production input—capture changes in relative prices in the economy and therefore their impact on headline inflation should be short-lived. In contrast, the second-round effects involve increases in prices that are more persistent, including those that result from pressures to preserve real wage levels.

    A related monetary policy issue is to determine the most appropriate inflation measure that should serve as the target that guides policy decisions by central banks. In general, the literature advocates for excluding from this measure flexible and volatile prices, such as those associated with food and energy, that is, core inflation (Aoki, 2001; Bodeinstein, Erceg, and Guerrieri, 2008; Wynne, 2008; and Mishkin, 2008). By stabilizing core inflation, the monetary authority would stabilize output and implement, to a great extent, the standard policy advice of allowing for direct (first-round) effects while reacting to indirect (second-round) effects.

    The standard policy advice, however, may need to be modified depending on the structural characteristics/distortions of the economy and their interaction with the world food and energy price shocks. Whether monetary policy should heavily react to first-round and second-round effects or, on the contrary, allow for some pass-through may also depend on the central bank’s objectives: concerns about inflation pressures versus the real economic adjustment induced by tight monetary policies that try to offset these pressures.

    The central bank’s concern about the real adjustment, labor market frictions, and the importance of food and oil in consumption baskets may imply the need for some pass-through into wages and core inflation to reduce the painful real adjustment. Of course, the degree of acceptable pass-through should be consistent with keeping medium-run inflation expectations anchored.

    • In the presence of real wage rigidity, the central bank must decide whether to accommodate a higher level of inflation or, instead, keep inflation constant but allow for a larger decline in the welfare-relevant output gap (increase in unemployment). The reason is that when real wages respond sluggishly to labor market conditions, stabilization of inflation and stabilization of the welfare-relevant output gap (and therefore unemployment) present the central bank with a trade-off (Blanchard and Gali, 2007). In fact, when there is substantial workers’ resistance to real wage decline, it may be necessary to have very large increases in interest rates and unemployment, to get wages not to respond (preventing pass-through). So by allowing for some pass-through and therefore some increase in core inflation, monetary policy may lead to a smaller decrease in output.

    • When the share of food in the CPI is high, food price shocks may involve large decreases in real wages to which workers respond demanding higher nominal wages. If the monetary authority allows for pass-through and therefore an increase in wages, the full adjustment in real wages may be delayed. The reason, present in New Keynesian models with sticky prices à la Calvo (nominal price rigidities), but also in the data, is that as wages increase, prices increase more slowly for some time, causing a temporary decline in markups of prices over wages. Thus, for some time, markups are lower and real wages higher than they would be under flexible prices. So higher inflation comes with a smoother real wage adjustment. By permitting a more gradual adjustment of real wages, monetary policies that allow for some pass-through can also decrease the painful adjustment to food shocks.

    When concerns about inflation prevail, other structural characteristics of the economy suggest that countries should perhaps react to first-round effects, as some of the literature advocates targeting headline inflation.

    • Since food is a large part of the consumption basket in developing economies and it has limited substitutability with other goods, food price fluctuations often have a significant impact on overall consumer prices. If food price shocks are large relative to monetary and productivity shocks, monetary policies that react only to core inflation may entail high real exchange rate volatility (Catão and Chang, 2010). As higher real exchange rate volatility may translate into consumption volatility, monetary policy responding to headline inflation (e.g., not allowing for first-round effects) may be superior to policies responding exclusively to core inflation.

    • In the presence of financial frictions that limit consumers’ access to credit in financial markets, a narrow policy focus on core inflation, which excludes food and oil prices, instead of headline inflation, can lead to suboptimal outcomes (Anand and Prasad, 2010). With financial frictions, inflation and output, which is demand determined, may move in opposite directions. In contrast to the benchmark New Keynesian model, stabilizing core inflation is no longer sufficient to stabilize output. With these frictions food and energy prices influence aggregate demand, since this demand is mainly determined by the real wages of credit-constrained consumers. So monetary policy may have to stabilize these prices to ensure that aggregate demand declines when monetary policy is tightened.

    • Central bank credibility has also important implications for monetary policy responses in the context of supply shocks (IMF, 2011d). When monetary policy responds to core inflation instead of headline inflation, tight monetary policies may induce smaller output losses at the expense of higher inflation. But benefiting from this trade-off depends on the credibility of the central bank (Habermeier and others, 2009). If credibility is low, food and oil price shocks can have significant effects on inflation expectations, inducing second-run effects. Lack of credibility may then call for stronger monetary policy actions, including reacting to headline inflation.

    These analytical considerations are relevant for LICs, where food and oil are a large part of the consumption basket, consumers are credit constrained, policy credibility is still limited, and food inflation volatile. Under these conditions, targeting core (i.e., non-food) inflation may result in large headline inflation volatility. For the small number of LIC central banks that are primarily inflation targeters, this may argue for counteracting even first-round effects of food price shocks to enhance policy credibility and anchor inflation expectations. However, this must be weighed against those considerations discussed earlier regarding the implications for the real economy. A strong policy response might be particularly contractionary and painful in real terms at a time when economies are already being hit by a negative supply shock. Therefore, the policy response may have to be more gradual than is optimal from the point of view of inflation objectives alone.

    The manner in which monetary policy is tightened must be adjusted to the policy regime in place. Inflation targeters would need to raise their policy rates. Money targeters, on the other hand, would have to be flexible, since sticking to previously set targets may be an excessively tight policy. More generally, real money balances must fall relative to planned levels so as to contain aggregate demand and reduce inflation pressures.

    The exchange regime may also determine the monetary policy response to commodity price shocks. Since LICs may need a real exchange rate depreciation to ensure external balance, in flexible regimes central banks should consider that this may take place through nominal depreciation, which might be at odds with the need for disinflation. Countries with fixed exchange rate regimes are limited in their policy options. Limited capital mobility could give them some room for monetary tightening.

    In any case, the fact that inflation and monetary aggregates are likely to be above target means that the central bank’s communication strategy is important to help shape expectations and contribute to inflation stabilization. Such communication should explain why targets have been missed, what the policy strategy is for bringing inflation down, and what the central bank will do if inflation does not fall as intended (Habermeier and others, 2009).

    Selected Country Experiences in 2007–08

    The 2007–08 food and fuel price shocks led to a large and broad-based increase in inflation across LICs, with monetary policy remaining largely passive. The experience differed across countries, depending on the policy mix, policy regimes, and other country-specific factors (e.g., openness, extent of dollarization, share of food items in the CPI basket, and climatic shocks). Pass-through from exchange rate depreciation was an important factor, with pegged exchange rates regimes generally experiencing lower inflation outcomes. Within this group, appreciation of the euro in particular helped alleviate inflationary pressures in the CFA zone. With output in about three quarters of LICs exceeding potential output at the time, aggregate demand pressures likely played a significant role in explaining some of the acceleration in inflation. But more than half of LICs adopted fiscal measures limiting the increase in prices of selected food items and petroleum products, partly suppressing the inflationary pressures. Monetary policy tightening appears to have had a negligible role. The collapse in global food and fuel prices during the onset of the global financial crisis in the second half of 2008 brought inflation down rapidly.

    In SSA LICs, inflation surged from 6.2 percent in March 2007 to 14.8 percent in August 2008 for SSA oil importers. However, the median policy rate was raised by only 1.5 percentage points through late 2008, implying an overall decline in real rates (Figure A8.1). Zambia’s policy response was the most aggressive, with a gradual rate increase of 3 percentage points, although it took two years to bring inflation to its pre-shock level. Reserve requirements were raised significantly in some countries, helping to reduce excess reserves, though the level of excess reserves remained fairly high (Figure A8.2). Exchange rate pressures were not apparent, with fairly muted exchange rate response and continued buildup of international reserves for both floaters and fixers. Ethiopia’s inflation reached a record high of 64 percent among LICs in July 2008, associated with strong money growth and with a large depreciation of the exchange rate. Overall, the sharp reversal in food and fuel prices in the second half of 2008 was primarily driven by the global crisis and collapse in global commodity prices, rather than monetary policy action.

    Figure A8.1.Sub-Saharan Africa: Policy Interest Rates and Broad Money

    Source: IMF, International Financial Statistics.

    Figure A8.2.Reserves of the Banking Sector, 2007–10

    Owing to sharp movements in copper prices, Mongolia was hit hard during the last commodity boom and bust cycle. The stronger copper prices brought faster growth together with a fiscal and current account surplus. When copper prices nosedived in late 2008, exports dropped sharply as did foreign exchange inflows and government revenue. The economy was on the verge of collapse when an IMF-supported Stand-By Arrangement was put in place in April 2009. Mongolia had to allow a significant depreciation in the exchange rate to absorb external shocks and safeguard international reserves (Figure A8.3). During 2007–08, nominal exchange rate rigidities did not prevent real exchange rate adjustments but instead led to runaway inflation. The collapse in copper prices combined with foreign exchange interventions brought a significant decline in reserves. A significant tightening of monetary policy played a key role in facilitating an orderly switch to a flexible exchange rate regime and safeguarding international reserves while keeping inflation on a stable path. The authorities increased policy rates by 425 basis points upfront to signal a regime change and limit capital outflows. Once the exchange rate stabilized and remonetization (after bank runs in end-2008) progressed, the policy rate was decreased gradually.

    Figure A8.3.Mongolia: Gross International Reserves and Exchange Rates, 2005–10

    (January 2005 = 100, increase denotes appreciation)
    References

      Aoki, Kosuke, 2001, “Optimal Monetary Policy Responses to Relative-Price Changes,” Journal of Monetary Economics, Vol. 48, No. 1, pp. 5580.

      Anand, Rahul, and Eswar S.Prasad, 2010, “Optimal Price Indices for Targeting Inflation Under Incomplete Markets,” NBER Working Paper No. 16290 (Cambridge, Massachusetts: National Bureau of Economic Research).

      Arze del Granado, Javier, DavidCoady, and RobertGillingham, 2010, “The Unequal Benefits of Fuel Subsidies: A Review of Evidence for Developing Countries,” IMF Working Paper 10/202 (Washington: International Monetary Fund).

      Blanchard, Olivier, and JordiGali, 2007, “Real Wage Rigidities and the New Keynesian Model,” Journal of Money, Credit, and Banking, Vol. 39, No. 1, pp. 3566.

      Bodeinstein, Martin, Christopher J.Erceg, and LucaGuerrieri, 2008, “Optimal Monetary Policy with Distinct Core and Headling Inflation Rates,” Journal of Monetary Economics, Vol. 55 (Suppl.), pp. S1833.

      Bornhorst, Fabian, SanjeevGupta, and JohnThornton, 2009, “Natural Resource Endowments and the Domestic Revenue Effort,” European Journal of Political Economy, Vol. 25, No. 4 (December), pp. 43946.

      Cashin, Paul, HongLiang, and C. JohnMcDermott, 1999, “How Persistent Are Shocks to World Commodity Prices?IMF Working Paper 99/80 (Washington: International Monetary Fund).

      Cashin, Paul, C. JohnMcDermott, and AlasdairScott, 2002, “Booms and Slumps in World Commodity Prices,” Journal of Development Economics, Vol. 69, No. 1, pp. 27796.

      Catão, Luis A.V., and RobertoChang, 2010, “World Food Prices and Monetary Policy,” IMF Working Paper 10/161 (Washington: International Monetary Fund).

      Coady, David, and MatiasAntonio, 2011, “Recent Developments in Petroleum Product Pricing in Middle East and Central Asia” and “Recent Developments in Petroleum Product Pricing in Africa,” Fiscal Affairs Department (unpublished; Washington: International Monetary Fund).

      Coady, David, RobertGillingham, RolandoOssowski, JohnPiotrowski, ShamsuddinTareq, and JustinTyson, 2010, “Petroleum Product Subsidies: Costly, Inequitable, and Rising,” IMF Staff Position Note 10/05 (Washington: International Monetary Fund).

      Dabla-Norris, Era, RichardAllen, Luis-FelipeZanna, TejPrakash, EteriKvintradze, VictorLledó, IreneYackovlev, and SophiaGollwitzer, 2010, “Budget Institutions and Fiscal Performance in Low-Income Countries,” IMF Working Paper 10/80 (Washington: International Monetary Fund).

      Dabla-Norris, Era, JiroHonda, AminaLahreche, and GenevièveVerdier, 2010, “FDI Flows to Low-Income Countries: Global Drivers and Growth Implications,” IMF Working Paper 10/132 (Washington: International Monetary Fund).

      Dabla-Norris, Era, RafaelEspinoza, and SarwatJahan, forthcoming, “Spillovers to Low-Income Countries: Importance of Systematic Emerging Markets,” IMF Working Paper (Washington: International Monetary Fund).

      Escolano, Julio, 2010, “A Practical Guide to Public Debt Dynamics, Fiscal Sustainability, and Cyclical Adjustment of Budgetary Aggregates,” Technical Notes and Manuals No. 2010/02, Fiscal Affairs Department (Washington: International Monetary Fund).

      Fabrizio, Stefania, 2010, Emerging from the Global Crisis: Macroeconomic Challenges Facing Low-Income Countries (Washington: International Monetary Fund).

      Fosu, Augustin Kwasi, 2010, “Growth, Inequality, and Poverty Reduction in Developing Countries: Recent Global Evidence,” OECD Development Centre, Background Paper for Global Development Outlook 2010, Shifting Wealth: Implications for Development (Paris: Organization for Economic Cooperation and Development).

      Frankel, Jeffrey, 2008, “The Effect of Monetary Policy on Real Commodity Prices,” in Asset Prices and Monetary Policy, ed. byJohn Y.Campbell, National Bureau of Economic Research Conference Report Series (Chicago: University of Chicago Press).

      Gupta, Sanjeev, GerdSchwartz, ShamsuddinTareq, RichardAllen, IsabellAdenauer, KevinFletcher, and DuncanLast, 2008, Fiscal Management of Scaled-Up Aid (Washington: International Monetary Fund).

      Habermeier, Karl, İnciŐtker-Robe, LuisJacome, AlessandroGiustiniani, KotaroIshi, DavidVávra, Turgut Kişinbay, and FranciscoVasquez, 2009, “Inflation Pressures and Monetary Policy Options in Emerging and Developing Countries: A Cross Regional Perspective,” IMF Working Paper 09/1 (Washington: International Monetary Fund).

      Heller, Peter, 2005, “Understanding Fiscal Space,” IMF Policy Discussion Paper No. 05/4 (Washington: International Monetary Fund).

      International Monetary Fund, 2010, “Eligibility to Use the Fund’s Facilities for Concessional Financing” (Washington).

      International Monetary Fund, 2011a, “Managing Volatility—A Vulnerability Exercise for Low-Income Countries” (Washington).

      International Monetary Fund, 2011b, “New Growth Drivers for Low-Income Countries: The Role of BRICs” (Washington).

      International Monetary Fund, 2011c, “Revenue Mobilization in Developing Countries” (Washington).

      International Monetary Fund, 2011d, “Target What You Can Hit: Commodity Price Swings and Monetary Policy,Chapter 3 in World Economic Outlook: Slowing Growth, Rising Risks, World Economic and Financial Surveys (Washington, September).

      Ivanic, Maros, WillMartin, and HassanZaman, 2011, “Estimating the Short-Run Poverty Impacts of the 2010–11 Surge in Food Prices,” Policy Research Working Paper No. 5633 (Washington: World Bank).

      Lledó, Victor, and MarcosPoplawski-Ribeiro, 2011, “Fiscal Policy Implementation in Sub-Saharan Africa,” IMF Working Paper 11/172 (Washington: International Monetary Fund).

      Lombardi, Marco, ChiaraOsbat, and BerndSchnatz, 2010, “Global Commodity Cycles and Linkages: A FAVAR Approach,” ECB Working Paper No. 1170 (Frankfurt am Main: European Central Bank).

      Lueth, Erik, and MartaRuiz-Arranz, 2008, “Determinants of Bilateral Remittance Flows,” The B.E. Journal of Macroeconomics, Vol. 8, No. 1(Topics), Article 26.

      Mishkin, Frederic, 2008, “Does Stabilizing Inflation Contribute to Stabilizing Economic Activity?NBER Working Paper No. 13970 (Cambridge, Massachusetts: National Bureau of Economic Research).

      Svensson, L.E.O., 2008, “The Effect of Monetary Policy on Real Commodity Prices: Comments,” in Asset Prices and Monetary Policy, ed. byJohn Y.Campbell, National Bureau of Economic Research Conference Report Series (Chicago: University of Chicago Press).

      Vansteenkiste, Isabel, 2009, “How Important Are Common Factors in Driving Non-fuel Commodity Prices? A Dynamic Factor Analysis,” ECB Working Paper No. 1072 (Frankfurt am Main: European Central Bank).

      Wynne, Mark, 2008, “How Should Central Banks Define Price Stability?Working Paper No. 8, Globalization and Monetary Policy Institute (Dallas: Federal Reserve Bank).

      Wolf, M., 2008, “Life in a Tough World of High Commodity Prices,Financial Times, March4.

    Appendix 2 reports selected economic indicators for individual countries and country groups.

    Global food and fuel prices are expected to have increased by 21 percent and 29 percent, respectively, in 2011; the increase in global food and fuel prices in 2008 was 23 percent and 40 percent, respectively.

    These results are based on a survey of IMF desks covering the group of Poverty Reduction and Growth Trust—eligible countries.

    These included food stamps (Mongolia), transportation subsidies (Central African Republic), school feeding programs (Burundi), fertilizer subsidies (Bolivia), subsidies to partially cover the increase in heating prices (Georgia and Moldova), an increase in social funds (Yemen), and higher transfers to public energy companies to subsidize the energy price paid by consumers (Senegal).

    Fiscal cost is defined as the cumulative budgetary cost of fiscal measures implemented during 2010 and the estimated fiscal impact for 2011.

    Average public debt of LICs has declined in recent years in part because of debt relief under the Heavily Indebted Poor Countries and Multilateral Debt Relief Initiatives.

    This assumes that nominal expenditure plans in 2012 budgets would not be revised after the shock while revenues decline substantially—by around 3 percent of GDP for the median LIC.

    Prior to the global financial crisis, the fiscal deficit for the median LIC was about 1.3 percent of GDP. In 2011, the fiscal deficit is projected in the baseline to have been close to 3 percent of GDP. In 2012, only about one-quarter of countries are projected to have restored the pre-crisis fiscal balance buffer.

    Poverty is defined as consumption below $1.25 per person per day (Ivanic, Martin, and Zaman, 2011). Appendix 7 describes the methodology.

    Elasticities of poverty with respect to growth are taken from Fosu (2010).

    The tail-risk scenario of higher global commodity prices is constructed using fan chart—like analysis based on market expectations embedded in commodity futures options. The commodity price levels in the scenario represent extreme values that could be reached or exceeded with only 7 percent probability based on density functions built from the WEO baseline commodity price assumptions. See Appendix 5 for a description of the methodology.

    The estimates of the policy response assume that governments take revenue and expenditure measures per unit of increase in oil and food prices similar to those in the 2007–08 episode of high global oil and food prices. The measures and fiscal impact of existing policies is calculated using revenue and expenditure elasticities to changes in global oil and food prices (see Appendix 5).

    The debt increase is estimated by maintaining the 2012 increase in the deficit over the long term, and then adding the discounted sum of these deficits at the projected nominal GDP growth rate (calculated as the average of the past 10 years) to the public debt stock.

    Poverty is defined as consumption below US$1.25 per person per day (Ivanic, Martin, and Zaman, 2011). Appendix 7 describes the methodology.

    See IMF (2011c). Regarding taxing natural resources, Bornhorst, Gupta, and Thornton (2009) also find that increases in resource revenue in LICs are offset partially (about a fifth of the increase is wiped out) by a decline in non-resource collection because of weak incentives to collect own revenue (natural resource curse).

    Appendix 8 discusses some of the analytical considerations that may call for adaptations of the standard policy advice in developing countries.

    This group includes all countries eligible for concessional financing from the IMF under the Poverty Reduction and Growth Trust (PRGT), except for Somalia, which has been excluded because of lack of data. The list of PRGT-eligible countries contains essentially those IMF members that (i) have annual per capita gross national income of less than twice the operational IDA cutoff (or three times the operational IDA cutoff for small economies); or (ii) have the capacity for durable and substantial access to international financial markets; and (iii) do not face serious short-term vulnerabilities. Therefore, the set of countries defined in this paper as LICs may differ from classifications used by other institutions. For technical details, see IMF (2010).

    Georgia and Armenia participate also in the emerging market vulnerability exercise and are not subject to the same degree of analysis as the other countries covered by the paper.

    Defining large negative shocks over country-specific distributions better captures cross-country heterogeneity among LICs, particularly with respect to their economic structure and vulnerability to external shocks. It means that each country experiences the same frequency of shocks, so that the focus is on the reaction to the shock.

    FDI, aid, and remittances are measured as ratios to GDP.

    Additional specifications of this basic model have been used to control the robustness of the elasticities, including by assessing possible nonlinear effects of sharp increases in prices and controlling for residuals’ serial correlation and heteroscedasticity.

    The methodology shown here was developed in IMF (2011a).

    These parameters are consistent with results of regression on the relative weight of the growth and terms-of-trade shocks on the budget for the sample.

    The rank correlation between this indicator and the one based on the growth rate of nominal revenue is greater than 0.9.

    The rank correlation is high between this indicator and an alternative indicator using a 65 percent of GDP target in 2030.

    The elasticity measures the change in poverty incidence (in percentage points) from changes in food prices (in percent) and includes the effect of food prices on incomes for households that are food producers. The poverty line is based on the $1.25 a day definition.

    Simplifying assumptions are made with regard to the ability of consumers to substitute into less expensive food items (we simulate an increase in the price index of all food items), the coping strategies of households, and the effects of food price increases on wages.

    The method does not take into account the effects of increases in fuel prices on wages and does not model the effects of behavioral responses, such as the effect on demand and other coping strategies by households.

      Other Resources Citing This Publication