## EXECUTIVE SUMMARY ## Source details **Canonical URL:** [EXECUTIVE SUMMARY](https://www.imf.org/-/media/files/publications/sdn/2017/sdn1702.pdf) ## Other formats - [Markdown version](/-/media/files/publications/sdn/2017/sdn1702.pdf.md) - [Structured JSON version](/-/media/files/publications/sdn/2017/sdn1702.pdf.json) --- ### Executive overview - Purpose: Assess the magnitude and direction of the bias introduced by single deflation relative to the recommended double deflation (2008 SNA, paragraph 15.136). - Approach: Apply single deflation and single extrapolation to official GDP data for countries that use double deflation and compare results with official double-deflation estimates. - Case study sample: Eight countries—Belgium, Brazil, Canada, France, Japan, Korea, the Netherlands, and the United States. - Main headline findings: - Errors from single deflation can be significant. - The direction of the bias cannot be predicted accurately; it varies across countries and over time. - When output prices move faster than input prices, single deflation tends to overstate value added growth; when output prices move more slowly than input prices, single deflation tends to understate value added growth. - Large commodity price changes can exacerbate the error: single deflation can produce significantly overstated GDP in oil-importing countries during sharp oil price declines and understated GDP in oil-exporting countries. ### I. Introduction - Volume measurement importance: Separates price and volume to produce GDP at constant prices (real GDP) and value added in volume terms. - Three techniques to derive volume measures: - Double deflation: relies on independent deflators for input and output prices and tends to yield the best results. - Single extrapolation: uses a volume indicator to extrapolate value added; considered the second best. - Single deflation (direct deflation): applies one deflator across the board and typically reduces accuracy. - Research contribution: Quantifies the order of magnitude of bias by applying single deflation to countries that currently use double deflation and comparing with official series. ### II. Overview of methods - Double deflation: - Outputs and inputs are deflated separately using relevant output and input price indices. - Intermediate inputs are measured at purchasers’ prices whereas output is measured at basic prices or producers’ prices (2008 SNA, paragraph 18.27). - Data intensive; in practice often focused on major industries (e.g., oil mining and refining, major crops). - Single deflation: - Output and input prices are deflated with the same index (often the output price). - Produces value added in volume terms by deflating current value added directly by one price indicator. - Underlying assumption: prices of outputs and inputs change at the same rate. - Some countries use the consumer price index (CPI) as the single deflator, with limitations noted. - Single extrapolation: - Output is deflated using the output price index and value added is extrapolated from a base period using the growth of output in constant prices. - Considered acceptable second best, provided an up-to-date base year (five years ago or less) is used. - Assumes input-output technical coefficients do not change in the short term; requires regular benchmark updates. ### III. Simulation design and scope - Simulation method: - For each activity, take gross value added at current prices (official difference between output and intermediate consumption at current prices). - Derive the output implicit price index as output at current prices divided by output in volume terms. - Deflate gross value added at current prices by the output price index to obtain single-deflation volume estimates by activity. - Aggregate single-deflation estimates to the total economy level. - Measure impact as the difference between single-deflation estimate and the official double-deflation estimate of total value added. - Data and collaboration: - Selected countries have sound deflation methods and data availability by economic activity. - Worked with national statistics agencies; some provided detailed unpublished data. - Exercises conducted at a minimum detail of two-digit ISIC Rev. 4. ### Key empirical and practical findings - Empirical evidence: - Single-deflation methods can result in significant overall errors in volume estimates. - Differences between single deflation and official double-deflation series can be both positive and negative, reflecting relative movements of output and input prices. - Single extrapolation reduces the error in some but not all countries when applied to the same data. - Practical implications during commodity price shocks: - In oil-importing countries during sharp declines in oil prices, single deflation can significantly overstate GDP and GDP growth. - In oil-exporting countries during such declines, single deflation can understate GDP. ### IV. Moving toward double deflation (guidance) - Practical stepwise approach: - Prioritize compilation of output and input price indices for industries with large potential bias. - Use double deflation for major industries (for example, oil mining and refining, major crops) where input prices are readily available. - Apply single extrapolation as a second-best where appropriate, ensuring base years are updated (five years ago or less). ### V. Conclusions - Double deflation is the recommended method (2008 SNA) and is applied by virtually all advanced economies. - Only a few G20 countries still predominantly use single deflation. - The simulation for the eight selected countries demonstrates that single deflation can introduce significant and variable biases; direction depends on whether output prices change faster or slower than input prices. - Single extrapolation can be a useful second-best in some contexts but does not uniformly reduce bias across countries. --- ### 12. Our results show that single deflation can introduce a significant error in the gross ### Key findings on single deflation bias - Single deflation can deviate from official gross value added growth estimates by up to 3–4 percentage points in some years. - Differences over time appear to be autocorrelated for the majority of countries. - Some countries show large differences during 2008–09 (in particular, Japan, Korea, United States). - Single-deflation estimates lead to a systematic underestimation of GDP growth in commodity importers during the global commodity price boom. - For commodity exporters, particularly Brazil and Canada, single deflation tends to overstate GDP growth during the commodity price boom from early 2000 until the global financial crisis; these trends reverse when commodity prices begin to fall. ### Empirical magnitudes and country-level summary statistics (differences between single deflation and official estimate of GDP, percentage points) - On average, single deflation produces an understatement for five countries: Belgium, France, Japan, Netherlands, United States. - On average, single deflation produces an overstatement for three countries: Brazil, Canada, Korea. - In absolute terms, the largest impact of single deflation is noted for Japan, Korea, and Brazil. - The effect on European Union countries is relatively small; France and Canada show the smallest differences with respect to the official estimates: 0.36 percent and 0.41 percent, respectively. - Selected notable event magnitudes: - In Brazil and Japan, the large increase of output prices relative to input prices in 2009 would lead to a growth bias of 3.4 percentage points when using single deflation. - In Korea in 2008, single deflation would have produced an understatement of GDP growth by 4.8 percentage points. - For Japan, mean absolute difference using single extrapolation shows a dramatic reduction compared with single deflation: 0.49 vs. 1.21 percentage points. - Table 2. Summary statistics on the difference between single deflation and official estimate of GDP (Percentage points) - Belgium (Sample Period 2000–13): Mean Difference –0.50; Mean Absolute Difference 0.75; Standard Deviation 0.77; Minimum Difference –1.92; Maximum Difference 0.70 - Brazil (Sample Period 2001–13): Mean Difference 0.04; Mean Absolute Difference 1.14; Standard Deviation 1.58; Minimum Difference –2.79; Maximum Difference 3.36 - Canada (Sample Period 2000–12): Mean Difference 0.05; Mean Absolute Difference 0.41; Standard Deviation 0.55; Minimum Difference –0.84; Maximum Difference 1.22 - France (Sample Period 2000–13): Mean Difference –0.20; Mean Absolute Difference 0.36; Standard Deviation 0.37; Minimum Difference –0.85; Maximum Difference 0.36 - Japan (Sample Period 2000–14): Mean Difference –0.74; Mean Absolute Difference 1.21; Standard Deviation 1.30; Minimum Difference –2.51; Maximum Difference 3.36 - Korea (Sample Period 2000–14): Mean Difference 0.18; Mean Absolute Difference 1.21; Standard Deviation 1.69; Minimum Difference –4.75; Maximum Difference 2.02 - Netherlands (Sample Period 2000–14): Mean Difference –0.25; Mean Absolute Difference 0.61; Standard Deviation 0.67; Minimum Difference –1.33; Maximum Difference 0.86 - United States (Sample Period 2000–15): Mean Difference –0.33; Mean Absolute Difference 0.86; Standard Deviation 0.99; Minimum Difference –1.68; Maximum Difference 2.09 ### Drivers of the single-deflation error - The direction and size of errors reflect different movements of output and input prices experienced in these countries. - The single-deflation error is highly correlated with changes in the commodity terms of trade and exchange rates: - An increase in the commodity import price results in an understatement of GDP growth using single deflation. - An increase in the commodity export price produces an overstatement of GDP growth using single deflation. - A currency appreciation leads to an overstatement of GDP growth using single deflation because appreciation improves the terms of trade and increases output prices relative to input prices (the opposite happens in cases of depreciation). - Country-specific comments: - Belgium: GDP bias depends on the weight of the industry where single deflation is applied and the size of its output-input ratio. - Japan: Large positive bias in 2009 because single deflation does not capture the sharp fall in crude oil prices. - Korea: Sharp depreciation of the Korean won-US dollar exchange rate in 2008 led to a sudden increase in input prices, producing a large understatement of growth. - Netherlands: Volatile prices lead to large differences; applying single deflation to services such as trade, government, and education is particularly misleading. - United States: Single deflation can produce misleading results when economic growth is changing rapidly. ### Assessment of alternative methods - Single extrapolation: - Reduces the difference with the official estimate for five out of eight countries. - Japan: mean absolute difference single extrapolation 0.49 vs. single deflation 1.21 percentage points. - France shows a substantial increase of the error using single extrapolation, although it remains well below the size of the error from single deflation identified for other countries. - Single extrapolation generally provides better short-term results because the assumption of a constant input-output ratio in volume terms tends to hold in the short term, but it does not always improve results. - Double deflation: - 2008 SNA recommends the use of double deflation rather than single deflation. - Double deflation should be adopted in particular for activities with large fluctuations of input-output prices and volumes over time. ### Steps, resource implications, and implementation guidance for adopting double deflation - Short-run assessment: - Authorities should assess the possible direction and size of the bias introduced by single deflation to identify economic activities most likely affected. - Moving from single deflation to single extrapolation may be a useful intermediate step and does not require new data. - Medium- and long-term requirements: - Update benchmark estimates to reflect the relative size of the various economic sectors. - Develop price indices for outputs (for example, producer price index) and inputs (for example, import price indices). - Generate appropriate independent deflators for output and intermediate consumptions by economic activity. - Resource implications: - Establishing new price indices requires additional human resources and new surveys; starting up new price indices may take one or two years. - Adoption of double deflation requires additional resources for compilation, validation, and analytical tables. - Ideally, historical series should be revised backward to avoid a break in the GDP series. - Communication: - Revisions to GDP should be communicated to users before new estimates are made public. - Provide illustrative examples to help users understand why double deflation better mirrors economic reality. ### Conclusions and recommendations - Annual national accounts estimates using the single deflation method produce significant errors; this note is the first to demonstrate the materiality of the bias introduced when single deflation is used. - The bias can be exacerbated during large commodity price or exchange rate changes: - During a sharp decline of the oil price, single deflation can produce significantly overstated GDP data in oil-importing countries. - Conversely, in oil-exporting countries, single deflation could produce an understatement of GDP. - Regular updates of the base year are required; regular updates at intervals of (at least) five years are required. - Main recommendation: - Countries should take steps to adopt double deflation. - When double deflation is not realistic in the short term, countries can explore whether single extrapolation may constitute a useful interim improvement. - Countries should investigate changes in the input-output structure in the recent past and its likely evolution in the future. - Double deflation should be adopted in particular for activities with large fluctuations of input-output prices and volumes over time. --- ### Annex Table 1.2. A Panel Regression Model to Correlate the Single Deflation Errors with Key Drivers of Terms of Trade ### Objective and data - Purpose: Quantify relationship between GDP estimation error due to single deflation and relative changes in input and output prices, reflecting changes in global commodity prices and exchange rates. - Indicators used: - (1) country-specific commodity import price index (based on Gruss (2014)) - (2) country-specific commodity export price index (based on Gruss (2014)) - (3) nominal effective exchange rate (NEER) (from IMF’s International Financial Statistics database) - Sample: Balanced panel for eight countries, 2001–2012 - Method: Panel Least Squares with country fixed-effects - Periods included: 12 - Cross-sections included: 8 - Total panel (balanced) observations: 96 - Dependent variable: GDP Estimation Error due to Single Deflation (Growth Differential in Percentage Points) ### Estimated coefficients and significance - Constant (C) - Coefficient: 0.000171 - Std. Error: 0.000720 - t-Statistic: 0.237254 - Prob.: 0.8130 - DLOG(Commodity Import Price) - Coefficient: -0.065010 - Std. Error: 0.010837 - t-Statistic: -5.998611 - Prob.: 0.0000 - DLOG(Commodity Export Price) - Coefficient: 0.023978 - Std. Error: 0.005725 - t-Statistic: 4.188237 - Prob.: 0.0001 - DLOG(Nominal Eff. Exch. Rate) - Coefficient: 0.079016 - Std. Error: 0.019383 - t-Statistic: 4.076628 - Prob.: 0.0001 ### Model fit and specification statistics - Effects: Cross-section fixed (dummy variables) - R-squared: 0.520817 - Adjusted R-squared: 0.464442 - Mean dependent var: -0.002645 - S.D. dependent var: 0.011301 - S.E. of regression: 0.008270 - Akaike info criterion: -6.644898 - Log likelihood: 329.9551 - Hannan-Quinn criter.: -6.526126 - F-statistic: 9.238510 - Prob(F-statistic): 0.000000 - Durbin-Watson stat: 2.342848 ### Key empirical findings from the regression - The three estimated coefficients (commodity import price, commodity export price, NEER) are highly significant and the model explains more than half of the variation in the single deflation error (R-squared 0.520817). - Signs and magnitudes: - A 1 percentage point increase in the commodity import price causes, on average, an understatement of GDP growth by 0.07 percentage points. - A 1 percentage point increase in the commodity export price leads, on average, to an overstatement of GDP growth by 0.02 percentage points. - A 1 percentage point increase in the NEER (currency appreciation) leads, on average, to an overstatement of GDP growth by 0.08 percentage points. - The single-deflation error associated with a given change in commodity import prices is about four times as large as the error associated with an equally-sized change in commodity export prices, reflecting that the bulk of the sample are net commodity importers. ### Interpretation and caveats - The estimated coefficients capture average effects for the eight-country sample and cannot be generalized to yield precise predictions for any single country. *Source: sdn1702 - EXECUTIVE SUMMARY; sdn1702 - Annex Table 1.2. A Panel Regression Model to Correlate the Single Deflation Errors* ### EXECUTIVE SUMMARY ______________________________________________________________________________ 4 ### EXECUTIVE SUMMARY ### Executive overview - Purpose: Assess the magnitude and direction of the bias introduced by single deflation relative to the recommended double deflation (2008 SNA, paragraph 15.136). - Approach: Apply single deflation and single extrapolation to official GDP data for countries that use double deflation and compare results with official double-deflation estimates. - Case study sample: Eight countries—Belgium, Brazil, Canada, France, Japan, Korea, the Netherlands, and the United States. - Main headline findings: - Errors from single deflation can be significant. - The direction of the bias cannot be predicted accurately; it varies across countries and over time. - When output prices move faster than input prices, single deflation tends to overstate value added growth; when output prices move more slowly than input prices, single deflation tends to understate value added growth. - Large commodity price changes can exacerbate the error: single deflation can produce significantly overstated GDP in oil-importing countries during sharp oil price declines and understated GDP in oil-exporting countries. ### I. Introduction - Volume measurement importance: Separates price and volume to produce GDP at constant prices (real GDP) and value added in volume terms. - Three techniques to derive volume measures: - Double deflation: relies on independent deflators for input and output prices and tends to yield the best results. - Single extrapolation: uses a volume indicator to extrapolate value added; considered the second best. - Single deflation (direct deflation): applies one deflator across the board and typically reduces accuracy. - Research contribution: Quantifies the order of magnitude of bias by applying single deflation to countries that currently use double deflation and comparing with official series. ### II. Overview of methods - Double deflation (description): - Outputs and inputs are deflated separately using relevant output and input price indices. - Data intensive: requires a broad range of price measures for both outputs and inputs. - Intermediate inputs are measured at purchasers’ prices whereas output is measured at basic prices or producers’ prices (2008 SNA, paragraph 18.27). - In practice, many compilers focus double deflation on major industries (e.g., oil mining and refining, major crops). - Single deflation (description): - Output and input prices are deflated with the same index (often the output price because it is readily observable). - Produces value added in volume terms by deflating current value added directly by one price indicator. - Underlying assumption: prices of outputs and inputs change at the same rate. - Some countries use the consumer price index (CPI) as the single deflator, with limitations noted (CPI measures prices of goods and services purchased by domestic households and is measured at purchasers’ prices). - Single extrapolation (description): - Output is deflated using the output price index and value added is extrapolated from a base period using the growth of output in constant prices. - Considered acceptable second best, provided an up-to-date base year (five years ago or less) is used. - Assumes input-output technical coefficients do not change in the short term; requires regular benchmark updates. ### III. Double vs. single deflation: simulation design and scope - Simulation method: - For each economic activity, take gross value added at current prices (official difference between output and intermediate consumption at current prices). - Derive the output implicit price index as output at current prices divided by output in volume terms. - Deflate gross value added at current prices by the output price index to obtain single-deflation volume estimates by activity. - Aggregate single-deflation estimates to the total economy level. - Measure impact as the difference between single-deflation estimate and the official double-deflation estimate of total value added. - Data and collaboration: - Selected countries have sound deflation methods and data availability by economic activity. - Worked with national statistics agencies; some provided detailed unpublished data on output, intermediate consumption, and gross value added at current prices and in real terms. - For European Union countries, data were extracted from the Eurostat database of national accounts; for Japan, Korea, and the United States, data were collected from compiling agencies’ websites; Brazil and Canada submitted detailed data. - Exercises conducted at a minimum detail of two-digit ISIC Rev. 4. ### Key empirical and practical findings - Empirical evidence: - Single-deflation methods can result in significant overall errors in volume estimates. - Differences between single deflation and official double-deflation series can be both positive and negative, reflecting relative movements of output and input prices. - Single extrapolation reduces the error in some but not all countries when applied to the same data. - Practical implications during commodity price shocks: - In oil-importing countries during sharp declines in oil prices, single deflation can significantly overstate GDP and GDP growth. - In oil-exporting countries during such declines, single deflation can understate GDP. ### IV. Moving toward double deflation (summary of guidance) - Recognizes double deflation is data intensive and may be focused initially on major industries where input and output prices diverge most (for example, primary producers or homogenous manufacturing activities). - Practical stepwise approach (as outlined in the note): - Prioritize compilation of output and input price indices for industries with large potential bias. - Use double deflation for major industries (for example, oil mining and refining, major crops) where input prices are readily available. - Apply single extrapolation as a second-best where appropriate, ensuring base years are updated (five years ago or less). ### V. Conclusions - Double deflation is the recommended method (2008 SNA) and is applied by virtually all advanced economies. - Only a few G20 countries still predominantly use single deflation (see Table 1 for the predominant methods across G20 countries). - The simulation for the eight selected countries demonstrates that single deflation can introduce significant and variable biases; the direction of bias depends on whether output prices change faster or slower than input prices. - Single extrapolation can be a useful second-best in some contexts but does not uniformly reduce bias across countries. *Source: sdn1702 - EXECUTIVE SUMMARY (PDF chapter/section).* ### 12. Our results show that single deflation can introduce a significant error in the gross ### sdn1702 - 12. Our results show that single deflation can introduce a significant error in the gross ### Key findings on single deflation bias - Single deflation can deviate from official gross value added growth estimates by up to 3–4 percentage points in some years. - Differences over time appear to be autocorrelated for the majority of countries (that is, a positive difference is likely to be followed by another positive difference, and vice versa). - Some countries show large differences during 2008–09, the years of the global financial crisis (in particular, Japan, Korea, United States). - Single-deflation estimates lead to a systematic underestimation of GDP growth in commodity importers during the global commodity price boom. - For commodity exporters, particularly Brazil and Canada, single deflation tends to overstate GDP growth during the period of the commodity price boom from early 2000 until the global financial crisis; these trends are reversed when commodity prices begin to fall. ### Empirical magnitudes and country-level summary statistics (differences between single deflation and official estimate of GDP, percentage points) - On average, single deflation produces an understatement for five countries: Belgium, France, Japan, Netherlands, United States. - On average, single deflation produces an overstatement for three countries: Brazil, Canada, Korea. - In absolute terms, the largest impact of single deflation is noted for Japan, Korea, and Brazil. - The effect on European Union countries is relatively small; in particular France and Canada show the smallest differences with respect to the official estimates (0.36 percent and 0.41 percent, respectively). - Selected notable event magnitudes: - In Brazil and Japan, the large increase of output prices relative to input prices in 2009 would lead to a growth bias of 3.4 percentage points when using single deflation. - In Korea in 2008, single deflation would have produced an understatement of GDP growth by 4.8 percentage points. - For Japan, mean absolute difference using single extrapolation shows a dramatic reduction compared with single deflation: 0.49 vs. 1.21 percentage points. - Table 2. Summary statistics on the difference between single deflation and official estimate of GDP (Percentage points) - Belgium (Sample Period 2000–13): Mean Difference –0.50; Mean Absolute Difference 0.75; Standard Deviation 0.77; Minimum Difference –1.92; Maximum Difference 0.70 - Brazil (Sample Period 2001–13): Mean Difference 0.04; Mean Absolute Difference 1.14; Standard Deviation 1.58; Minimum Difference –2.79; Maximum Difference 3.36 - Canada (Sample Period 2000–12): Mean Difference 0.05; Mean Absolute Difference 0.41; Standard Deviation 0.55; Minimum Difference –0.84; Maximum Difference 1.22 - France (Sample Period 2000–13): Mean Difference –0.20; Mean Absolute Difference 0.36; Standard Deviation 0.37; Minimum Difference –0.85; Maximum Difference 0.36 - Japan (Sample Period 2000–14): Mean Difference –0.74; Mean Absolute Difference 1.21; Standard Deviation 1.30; Minimum Difference –2.51; Maximum Difference 3.36 - Korea (Sample Period 2000–14): Mean Difference 0.18; Mean Absolute Difference 1.21; Standard Deviation 1.69; Minimum Difference –4.75; Maximum Difference 2.02 - Netherlands (Sample Period 2000–14): Mean Difference –0.25; Mean Absolute Difference 0.61; Standard Deviation 0.67; Minimum Difference –1.33; Maximum Difference 0.86 - United States (Sample Period 2000–15): Mean Difference –0.33; Mean Absolute Difference 0.86; Standard Deviation 0.99; Minimum Difference –1.68; Maximum Difference 2.09 ### Drivers of the single-deflation error - The direction and size of errors reflect different movements of output and input prices experienced in these countries. - The single-deflation error is highly correlated with changes in the commodity terms of trade and exchange rates: - An increase in the commodity import price (which increases input prices) results in an understatement of GDP growth using single deflation. - An increase in the commodity export price (which increases output prices) produces an overstatement of GDP growth using single deflation. - A currency appreciation leads to an overstatement of GDP growth using single deflation because appreciation improves the terms of trade and increases output prices relative to input prices (the opposite happens in cases of depreciation). - Country-specific comments from statistical agencies: - Belgium: GDP bias depends on the weight of the industry where single deflation is applied and the size of its output-input ratio. - Japan: Large positive bias in 2009 because single deflation does not capture the sharp fall in crude oil prices; differences between export and import prices are not captured. - Korea: Input prices affected by exchange rate movements; sharp depreciation of the Korean won-US dollar exchange rate in 2008 led to a sudden increase in input prices, producing a large understatement of growth. - Netherlands: Volatile prices lead to large differences between single deflation and double deflation; applying single deflation to services such as trade, government, and education is particularly misleading. - United States: Single deflation can produce misleading results when economic growth is changing rapidly, as substantial changes in prices for intermediate inputs may not be immediately passed through. ### Assessment of alternative methods: single extrapolation and double deflation - Single extrapolation (moving forward GDP at current prices from the base year using the output volume indicator): - Single extrapolation reduces the difference with the official estimate for five out of eight countries. - Improvement is particularly evident in Japan: mean absolute difference single extrapolation 0.49 vs. single deflation 1.21 percentage points. - France shows a substantial increase of the error using single extrapolation, although it remains well below the size of the error from single deflation identified for other countries. - Statistical agencies generally indicate single extrapolation provides better results than single deflation in the short term because the assumption of a constant input-output ratio in volume terms tends to hold in the short term. - However, single extrapolation only partially confirms more accurate results than single deflation; it does not always improve the result. - Double deflation: - System of National Accounts 2008 recommends the use of double deflation rather than single deflation. - Double deflation should be adopted in particular for activities with large fluctuations of input-output prices and volumes over time. ### Steps, resource implications, and implementation guidance for adopting double deflation - Short-run assessment: - Authorities should assess the possible direction and size of the bias introduced by single deflation to identify economic activities most likely affected. - Moving from single deflation to single extrapolation may be a useful intermediate step and does not require new data, only a change in the estimation procedure. - Medium- and long-term requirements: - Update benchmark estimates to reflect the relative size of the various economic sectors. - Develop price indices for outputs (for example, producer price index) and inputs (for example, import price indices). - Generate appropriate independent deflators for output and intermediate consumptions by economic activity. - Resource implications: - Updating national accounts benchmarks is resource intensive but should be part of the regular compilation cycle. - Establishing new price indices requires additional human resources and new surveys; starting up new price indices may take one or two years. - Adoption of double deflation requires additional resources to maintain compilation of appropriate output prices and intermediate consumption prices, adjustments to procedures, supplementary validation, and additional analytical tables. - Ideally, the historical series should be revised backward to avoid a break in the GDP series. - Communication: - Revisions to GDP should be communicated to users before new estimates are made public. - Provide a few illustrative examples to help users understand why double deflation better mirrors economic reality. - Explain that development of new indicators (deflators) adds to the serviceability of economic statistics. ### Conclusions and recommendations - Annual national accounts estimates using the single deflation method produce significant errors; this note is the first to demonstrate the materiality of the bias introduced when single deflation is used. - The bias can be exacerbated during large commodity price or exchange rate changes: - During major commodity price changes such as a sharp decline of the oil price, single deflation can produce significantly overstated GDP data in oil-importing countries. - Conversely, in oil-exporting countries, single deflation could produce an understatement of GDP. - Regular updates of the base year are required to maintain the accuracy of volume estimates of GDP; regular updates at intervals of (at least) five years are required to ensure that weights of the constant price series are representative. - Main recommendation: - Countries should take steps to adopt double deflation. - When double deflation is not realistic in the short term, countries can explore whether single extrapolation may constitute a useful interim improvement with relatively minimal technical requirements. - Countries should investigate changes in the input-output structure in the recent past and its likely evolution in the future. - Double deflation should be adopted in particular for those activities with large fluctuations of input-output prices and volumes over time. *Source: IMF staff estimates (excerpt from "Measure Up: A Better Way to Calculate GDP").* ### Annex Table 1.2. A Panel Regression Model to Correlate the Single Deflation Errors ### Annex Table 1.2. A Panel Regression Model to Correlate the Single Deflation Errors with Key Drivers of Terms of Trade ### Objective and data - Purpose: Quantify relationship between GDP estimation error due to single deflation and relative changes in input and output prices, which reflect changes in global commodity prices and exchange rates. - Indicators used: - (1) country-specific commodity import price index (based on Gruss (2014)) - (2) country-specific commodity export price index (based on Gruss (2014)) - (3) nominal effective exchange rate (NEER) (from IMF’s International Financial Statistics database) - Sample: Balanced panel for eight countries, 2001–2012 - Method: Panel Least Squares with country fixed-effects (dummy variables) - Periods included: 12 - Cross-sections included: 8 - Total panel (balanced) observations: 96 - Dependent variable: GDP Estimation Error due to Single Deflation (Growth Differential in Percentage Points) ### Estimated coefficients and significance - Constant (C) - Coefficient: 0.000171 - Std. Error: 0.000720 - t-Statistic: 0.237254 - Prob.: 0.8130 - DLOG(Commodity Import Price) - Coefficient: -0.065010 - Std. Error: 0.010837 - t-Statistic: -5.998611 - Prob.: 0.0000 - DLOG(Commodity Export Price) - Coefficient: 0.023978 - Std. Error: 0.005725 - t-Statistic: 4.188237 - Prob.: 0.0001 - DLOG(Nominal Eff. Exch. Rate) - Coefficient: 0.079016 - Std. Error: 0.019383 - t-Statistic: 4.076628 - Prob.: 0.0001 ### Model fit and specification statistics - Effects: Cross-section fixed (dummy variables) - R-squared: 0.520817 - Adjusted R-squared: 0.464442 - Mean dependent var: -0.002645 - S.D. dependent var: 0.011301 - S.E. of regression: 0.008270 - Akaike info criterion: -6.644898 - Log likelihood: 329.9551 - Hannan-Quinn criter.: -6.526126 - F-statistic: 9.238510 - Prob(F-statistic): 0.000000 - Durbin-Watson stat: 2.342848 ### Key empirical findings - The three estimated coefficients are highly significant and the model can explain more than half of the variation in the single deflation error (R-squared 0.520817). - Signs of coefficients align with expectations: - A 1 percentage point increase in the commodity import price causes, on average, an understatement of GDP growth by 0.07 percentage points. - A 1 percentage point increase in the commodity export price leads, on average, to an overstatement of GDP growth by 0.02 percentage points. - A 1 percentage point in the NEER (currency appreciation) leads, on average, to an overstatement of GDP growth by 0.08 percentage points. - The regression results imply that the single-deflation error associated with a given change in commodity import prices is about four times as large as the error associated with an equally-sized change in commodity export prices. - The relatively larger impact of commodity import price changes reflects that the bulk of the countries in the sample are net commodity importers, where commodities tend to account for a larger share of inputs than outputs. ### Interpretation and caveats - The estimated coefficients capture the average effect of commodity import and export prices and exchange rate movements in explaining the single deflation bias for the eight countries in the sample. - These model results cannot be generalized to yield precise predictions of the single-deflation error for any single country. *Source: sdn1702 - Annex Table 1.2. A Panel Regression Model to Correlate the Single Deflation Errors* --- _Source: https://www.imf.org/-/media/files/publications/sdn/2017/sdn1702.pdf_