## onlineannex21 - 0.6 in economies that do not invoice in US dollars.

## Source details

**Canonical URL:** [onlineannex21 - 0.6 in economies that do not invoice in US dollars.](https://www.imf.org/-/media/files/publications/esr/2019/english/onlineannex21.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/esr/2019/english/onlineannex21.pdf.md)
- [Structured JSON version](/-/media/files/publications/esr/2019/english/onlineannex21.pdf.json)

---

### Dominant findings on invoicing and short-run external adjustment
- Short-run trade volume elasticities vary with degree of US dollar invoicing:
  - In the economy with high US dollar invoicing: export volume elasticity is negligible; import volume elasticity is -0.3.
  - In the economy with no US dollar invoicing: export volume elasticity is 0.1; import volume elasticity is -0.2.
- Short-run trade-balance response:
  - A 10 percent depreciation is estimated to increase the trade-balance by about 0.25 percentage points of GDP.
- Medium run:
  - Exchange rate pass-through estimates, quantity elasticities, and trade-balance effects are quantitatively similar across countries with low and high US dollar invoicing.
- Interpretation:
  - Dominant currency pricing (DCP) affects the composition of external adjustment mostly in the short-run.

### Unweighted regressions and broader sample implications
- Unweighted regressions (giving greater weight to small economies where USD invoicing is more prevalent) produce:
  - Higher exchange rate pass-through estimates than weighted regressions, driven by larger coefficients on the US dollar exchange rate at both short- and medium-term horizons.
  - Short-term trade volume responses remain asymmetric: negligible for exports and negative for imports.
  - Over the medium term, asymmetry persists: the export quantities response is only one fourth of the import volume elasticity (in absolute value).
  - The trade balance still improves over the medium run, more so than in the short run.

### Identification and IV evidence
- Endogeneity concern addressed with an instrumental variable (IV) strategy.
- Instrument construction:
  - High-frequency US monetary policy and risk-aversion shocks identified via a sign-restricted VAR on daily US 10-yr bond yield, US stock market index, and US NEER.
  - Shocks accumulated to annual frequency and interacted with annual change in reserves as a share of GDP (FX intervention) of importer and exporter to obtain cross-sectional variation.
  - Instruments: shocks interacted with FX intervention of exporter and importer.
- IV results (AE-to-EM country-pair focus):
  - IV and OLS estimates are qualitatively similar and support DCP predictions.
  - The coefficient on the US dollar exchange rate dominates that of the bilateral rate in the short-run for both price and volume equations.
  - Relative to OLS, US dollar exchange rate pass-through and trade volume elasticity estimates are somewhat larger with IV, but differences are not statistically significant.

### Global Value Chains (GVC): conceptual effects on prices and quantities
- Backward integration (BWD—sourcing imported intermediates from upstream suppliers):
  - Makes exporters’ marginal costs sensitive to exchange rates vis-à-vis upstream suppliers.
  - Expected effects of a depreciation vis-à-vis all other currencies (Text Table 2):
    - Exports: prices + (stand-alone); prices + (BWD); quantities + (stand-alone) and - (BWD).
    - Imports: prices + (stand-alone); prices + (FWD); quantities - (stand-alone) and + (FWD).
- Forward integration (FWD—exports of intermediates re-exported downstream):
  - Makes exporters’ bilateral trade affected by downstream buyers’ exchange rates through demand channels.
  - Demand shifters depend on the share of exports used as intermediates and downstream partners’ exchange rate movements.

### Empirical construction of GVC shifters
- Backward GVC (supply-side shifter) constructed as:
  - Sectoral marginal-cost sensitivity: ∆lnMGC_a,B = GGC_a,B · ∆ln e_a,B
  - GGC matrix built from WIOD input-output (including direct and indirect imported intermediate content).
  - Country-sector marginal cost changes aggregated to bilateral level using sectoral composition of exports.
- Backward USD sensitivity:
  - Analogous measure using exporter’s exchange rate vis-à-vis the US dollar (∆ln e_a$,B).
- Forward GVC (demand-side shifter) constructed as:
  - GGF_a b,B = X_a b,t^I · (I − Ã_b,t)^−1 · R_b,t, capturing share of exports from a to b that are reexported.
  - Demand shock: ∆lnD_a b,B = GGF_a b,B · ∆ln e_b,B
  - USD-based forward shifter: ∆lnD_a→b,B_USD = GGF_a b,B × ∆ln e_b$,B

### Econometric specification and aggregation
- Baseline bilateral regressions augmented with:
  - Bilateral and USD exchange rates (contemporaneous + 3 lags).
  - Backward and forward GVC shifters (bilateral- and USD-based), each with 3 lags.
  - Country-pair fixed effects and year fixed effects in main specifications.
- Trade-balance effects computed by combining estimated price and quantity elasticities and GVC-induced shocks:
  - Trade-balance impact formula uses sums of price and quantity coefficients and the GVC shock magnitudes as functions of exchange rate movements.

### Main empirical results with GVC shifters (weighted baseline)
- GVC shifters have expected signs and are often economically and statistically significant:
  - Backward GVC shifters (marginal-cost shocks) increase export prices and reduce export quantities.
  - Forward GVC shifters (demand shocks) increase import prices and quantities.
- Net implications for trade elasticities and trade balance:
  - For countries more integrated into GVCs (both backward and forward), trade volume elasticities fall while pass-through estimates increase.
  - Quantitatively, with an average degree of GVC integration:
    - Standalone import elasticity falls by about half; export elasticity declines by about one third.
    - Trade balance response (percent of GDP) is dampened: it drops by about half in the short term and by about one third in the medium term, relative to a country with no GVC integration.
- Including GVC shifters alters “stand-alone” estimates relative to the baseline DCP model:
  - Stand-alone pass-through estimates increase; trade volume elasticities are somewhat higher.
  - Coefficients on the US dollar exchange rate fall somewhat in absolute terms but remain broadly consistent in sign and significance.

### Robustness checks (overview)
- Results hold qualitatively and quantitatively across robustness exercises:
  - Unweighted regressions (Table 8).
  - Alternative importer demand controls:
    - Importer real domestic demand growth instead of real GDP growth (Table 9).
    - Fitted importer-time fixed effects as an import demand proxy (Table 10).
  - Using only direct components of GVC integration (Table 11).
  - Excluding intra-Euro area trade (Table 12).
  - Modeling global factors more granularly by replacing time fixed effects with world GDP growth, inflation, real export shocks, VIX, and real oil prices (Table 13).

*Source: IMF staff estimates and analysis in onlineannex21 - 0.6 in economies that do not invoice in US dollars.*

### 0.6 in economies that do not invoice in US dollars.

### onlineannex21 - 0.6 in economies that do not invoice in US dollars.

### Dominant findings on invoicing and short-run external adjustment
- Short-run trade volume elasticities vary with degree of US dollar invoicing:
  - In the economy with high US dollar invoicing: export volume elasticity is negligible; import volume elasticity is -0.3.
  - In the economy with no US dollar invoicing: export volume elasticity is 0.1; import volume elasticity is -0.2.
- Despite compositional differences, the short-run trade-balance response is similar across degrees of USD invoicing:
  - A 10 percent depreciation is estimated to increase the trade-balance by about 0.25 percentage points of GDP.
- Over the medium run, exchange rate pass-through estimates, quantity elasticities, and trade-balance effects are quantitatively similar across countries with low and high US dollar invoicing.
- Interpretation: dominant currency pricing (DCP) affects the composition of external adjustment mostly in the short-run.

### Unweighted regressions and broader sample implications
- Unweighted regressions (giving greater weight to small economies where USD invoicing is more prevalent) yield:
  - Higher exchange rate pass-through estimates than weighted regressions, driven by larger coefficients on the US dollar exchange rate at both short- and medium-term horizons.
  - Short-term trade volume responses remain asymmetric: negligible for exports and negative for imports.
  - Over the medium term, asymmetry persists in unweighted regressions: the export quantities response is only one fourth of the import volume elasticity (in absolute value).
  - Nonetheless, the trade balance still improves over the medium run, more so than in the short run.

### Identification and IV evidence
- Endogeneity concern: exchange rate is endogenous; instrumental variable (IV) strategy used to strengthen causal interpretation.
- Instrument construction:
  - High-frequency US monetary policy and risk-aversion shocks identified via a sign-restricted VAR on daily US 10-yr bond yield, US stock market index, and US NEER.
  - Shocks accumulated to annual frequency and interacted with annual change in reserves as a share of GDP (FX intervention) of importer and exporter to obtain cross-sectional variation.
  - Instruments: shocks interacted with FX intervention of exporter and importer.
- IV results (AE-to-EM country-pair focus):
  - IV and OLS estimates are qualitatively similar and support DCP predictions.
  - The coefficient on the US dollar exchange rate dominates that of the bilateral rate in the short-run for both price and volume equations.
  - Relative to OLS, US dollar exchange rate pass-through and trade volume elasticity estimates are somewhat larger with IV, but differences are not statistically significant.

### Global Value Chains (GVC): conceptual effects on prices and quantities
- Backward integration (BWD—sourcing imported intermediates from upstream suppliers):
  - Makes exporters’ marginal costs sensitive to exchange rates vis-à-vis upstream suppliers.
  - Expected effects of a depreciation vis-à-vis all other currencies (Text Table 2):
    - Exports: prices + (stand-alone); prices + (BWD); quantities + (stand-alone) and - (BWD).
    - Imports: prices + (stand-alone); prices + (FWD); quantities - (stand-alone) and + (FWD).
- Forward integration (FWD—exports of intermediates re-exported downstream):
  - Makes exporters’ bilateral trade affected by downstream buyers’ exchange rates through demand channels.
  - Demand shifters depend on the share of exports used as intermediates and downstream partners’ exchange rate movements.

### Empirical construction of GVC shifters
- Backward GVC (supply-side shifter) constructed as:
  - Sectoral marginal-cost sensitivity: ∆lnMGC_a,B = GGC_a,B · ∆ln e_a,B
  - GGC matrix built from WIOD input-output (including direct and indirect imported intermediate content).
  - Country-sector marginal cost changes aggregated to bilateral level using sectoral composition of exports.
- Backward USD sensitivity: analogous measure using exporter’s exchange rate vis-à-vis the US dollar (∆ln e_a$,B).
- Forward GVC (demand-side shifter) constructed as:
  - GGF_a b,B = X_a b,t^I · (I − Ã_b,t)^−1 · R_b,t, capturing share of exports from a to b that are reexported.
  - Demand shock: ∆lnD_a b,B = GGF_a b,B · ∆ln e_b,B
  - USD-based forward shifter: ∆lnD_a→b,B_USD = GGF_a b,B × ∆ln e_b$,B

### Econometric specification and aggregation
- Baseline bilateral regressions augmented with:
  - Bilateral and USD exchange rates (contemporaneous + 3 lags).
  - Backward and forward GVC shifters (bilateral- and USD-based), each with 3 lags.
  - Country-pair fixed effects and year fixed effects in main specifications.
- Trade-balance effects computed combining estimated price and quantity elasticities and GVC-induced shocks:
  - Trade-balance impact formula uses sums of price and quantity coefficients and the GVC shock magnitudes as functions of exchange rate movements.

### Main empirical results with GVC shifters (weighted baseline)
- GVC shifters have expected signs and are often economically and statistically significant:
  - Backward GVC shifters (marginal-cost shocks) increase export prices and reduce export quantities.
  - Forward GVC shifters (demand shocks) increase import prices and quantities.
- Net implications for trade elasticities and trade balance:
  - For countries more integrated into GVCs (both backward and forward), trade volume elasticities fall while pass-through estimates increase.
  - Quantitatively: with an average degree of GVC integration:
    - Standalone import elasticity falls by about half; export elasticity declines by about one third.
    - Trade balance response (percent of GDP) is dampened: it drops by about half in the short term and by about one third in the medium term, relative to a country with no GVC integration.
- Including GVC shifters alters “stand-alone” estimates relative to the baseline DCP model:
  - Stand-alone pass-through estimates increase; trade volume elasticities are somewhat higher.
  - Coefficients on the US dollar exchange rate fall somewhat in absolute terms but remain broadly consistent in sign and significance.

### Robustness checks (overview)
- Results hold qualitatively and quantitatively across robustness exercises:
  - Unweighted regressions (Table 8).
  - Alternative importer demand controls:
    - Importer real domestic demand growth instead of real GDP growth (Table 9).
    - Fitted importer-time fixed effects as an import demand proxy (Table 10).
  - Using only direct components of GVC integration (Table 11).
  - Excluding intra-Euro area trade (Table 12).
  - Modeling global factors more granularly by replacing time fixed effects with world GDP growth, inflation, real export shocks, VIX, and real oil prices (Table 13).

*Source: IMF staff estimates and analysis in onlineannex21 - 0.6 in economies that do not invoice in US dollars.*

### References

### onlineannex21 - References

### Exchange rate dynamics and pricing-to-market
- Betts, C. and Devereux, M. 2000. “Exchange Rate Dynamics in a Model of Pricing-to -Market.” Journal of International Economics, 50(1): 215–44.
- Devereux, Michael, and Charles Engel. 2003. "Monetary policy in the Open Economy Revisited: Price Setting and Exchange Rate Flexibility." Review of Economic Studies, 70: 765–84.
- Gopinath, Gita, and Roberto Rigobon. 2008. “Sticky borders.” Quarterly Journal of Economics, 123: 2, 531–75.
- Fitzgerald, Doireann, and Stefanie Haller. 2012. “Exchange Rates and Producer Prices: Evidence from Micro Data. Working Paper, Stanford University.
- Goldberg, Pinelopi Koujianou, and Rebecca Hellerstein. 2008. "A Structural Approach to Explaining Incomplete Exchange-Rate Pass-Through and Pricing-to -Market." American Economic Review 98: 2, 423–29.
- Amiti, Mary, Oleg Itskhoki, and Jozef Konings. 2014. "Importers, Exporters, and Exchange Rate Disconnect." American Economic Review, 104:7, 1942–78.

### Dominant currency paradigm and international price system
- Casas, Camila; Diez, J. Federico, Gopinath, Gita, and Pierre-Olivier Gourinchas. 2017. “Dominant Currency Paradigm: A New Model for Small Open Economies.” IMF Working Paper 17/264, International Monetary Fund, Washington DC.
- Gopinath, Gita. 2015. "The International Price System." NBER Working Paper 21646, National Bureau of Economic Research, Cambridge, MA.
- Gopinath, Gita, Emine Boz, Camila Casas, Federico J. Dıez, Pierre-Olivier Gourinchas, and Mikkel Plagborg-Møller. 2018. "Dominant Currency Paradigm."

### Global trade, dollar, and vehicle currency use
- Boz, Emine and Eugenio Cerutti. 2017. “Dissecting the Global Trade Slowdown: A New Database.” Unpublished Manuscript, International Monetary Fund, Washington DC.
- Boz, Emine, Gita Gopinath, and Mikkel Plagborg-Møller. 2018. “Global Trade and the Dollar."
- Goldberg, Linda, and Cédric Tille. 2008. “Vehicle Currency Use in International Trade.” Journal of International Economics, 76(2):177–192.

_Italic source attribution: onlineannex21 - References_

---


_Source: https://www.imf.org/-/media/files/publications/esr/2019/english/onlineannex21.pdf_