## _wp08199 - References

## Source details

**Canonical URL:** [_wp08199 - References](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp08199.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp08199.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp08199.pdf.json)

---

### I. Introduction — purpose and approach
- Objective:
  - Develop a model describing Myanmar’s foreign exchange markets, derive the equilibrium exchange rate, quantify distortions on exports and imports from the multiple exchange rate regime, and re-estimate key indicators (GDP, trade openness) using the estimated equilibrium exchange rate.
- Methodology:
  - Extend Rosenberg and Zeeuw (2001) framework to reflect Myanmar-specific exchange rate features.
  - Numerically estimate net efficiency losses and the equilibrium exchange rate that would clear a unified foreign exchange market.
  - Re-estimate the size of the external sector in national accounts using the estimated equilibrium exchange rate.
- Data/time frame:
  - Baseline parameters and estimates reference the period 2004/05–2006/07, with specific point estimates reported for 2006/07.

### II. Structure and mechanics of Myanmar’s multiple exchange rate regime
- Dual/segmented markets:
  - Regime functions as dual foreign exchange markets segmented for public and private sector external transactions.
  - Official exchange rate is exogenously set far below the market-clearing level and used for public-sector accounting and settlement.
  - Private-sector exporters have not faced a foreign exchange surrender requirement since the early 1990s.
- Public sector mechanics:
  - Public-sector exporters must surrender 100 percent of export proceeds to the government and choose export volumes given the official exchange rate.
  - Foreign exchange available for public imports ≈ public-sector export earnings minus reserve accumulation (capital flows assumed negligible).
  - Foreign exchange for public imports is rationed by the Foreign Exchange Budget managed by the Ministry of Finance and Revenue.
  - Foreign exchange revenues include: export receipts of SEEs, export taxes, income tax on inward remittances, utility charges paid by foreign companies for public services, and sales of Foreign Exchange Certificates (FEC).
  - Foreign exchange expenditures mainly: import payments of SEEs and servicing external public-sector debt.
- Parallel/informal markets and instruments:
  - FEC: issued as U.S. dollar equivalent; informal market trading at around K 1,100 per FEC; official FEC center rate at K 450 per dollar; transactions at centers limited.
  - Parallel U.S. dollar rate: around K 1,120 (March 2008) in informal markets.
  - Hundi system used for unrecorded trade and remittances; Hundi rate ≈ parallel rate minus ~1 percent commission.
  - Exchange restrictions limit purchase of foreign exchange for travel and certain transfers; residents with permits can buy US$100–US$200 on presentation of FEC; maximum remittance against FEC = US$10,000 per month or equivalent.
- Implication:
  - Official rate is grossly overvalued → public-sector import demand exceeds available foreign exchange.
  - Most unofficial rates are market-determined; differences among unofficial market rates negligible relative to gap with official rate — analyses assume dual exchange rate for simplicity.

### III. Model framework and key methodological features
- Market segmentation in model: four constant-elasticity markets:
  - Public exports: parameters Uβ and UB.
  - Public imports: parameters Uα and UA.
  - Private exports: parameters Rβ and RB.
  - Private imports: parameters Rα and RA.
- Key exchange rates:
  - OFE = official exchange rate.
  - MCR E = private market clearing exchange rate (parallel market).
  - EQE = equilibrium exchange rate (unified market-clearing rate to be solved numerically).
  - QotE = shadow exchange rate for public import quota.
- Model extensions beyond prior studies:
  - Incorporates private-sector markets under the parallel exchange rate (data available).
  - Introduces explicit foreign-exchange supply constraint on public-sector importers (bounded by public-sector export proceeds net of reserve accumulation).
  - Keeps MCR E and EQE distinct in net efficiency loss expressions.
  - Derives numerical solution methods for EQE and QotE using observed price-quantity combinations and parameter φ (private sector share relative to public sector).
- Efficiency loss/gain expressions:
  - Closed-form expressions for net efficiency losses/gains as functions of implicit tax/subsidy rates, price elasticities (Uβ, Uα, Rβ, Rα), and relative values of EQE, MCR E, OFE, and QotE.
  - Equilibrium condition given by equation (14); QotE by equation (15) (as referenced in source).

### IV. Baseline empirical assumptions and parameter choices
- Elasticity baseline:
  - Unit elasticity assumed for all markets except public exports.
  - Public export elasticity Uβ = 0.2.
- Calibration and observables:
  - OFE and MCR E treated as observable.
  - Normalization assumed: RURU MMXX PPPP ==== 1$$$$ (scaling normalization) to back out other parameters.
  - φ (private sector share in GDP / public sector share in GDP) used to relate private and public import demands under MCR E.
- Sensitivity parameter ranges evaluated:
  - φ in [3/7, 7]
  - Rα/1 in [0.5, 1.5]
  - Rβ/1 in [0.5, 1.5]
  - Uα/1 in [0.5, 1.5]
  - Uβ/1 in [0.3, 1.0]
  - All 32 (= 2^5) combinations evaluated for 2006/07.

### V. Baseline quantitative results (selected, period 2004/05–2006/07)
- Equilibrium and shadow exchange rates:
  - EQE estimated to lie between K 350–450 per U.S. dollar (Box 1 baseline) and elsewhere reported around K 400–500 per U.S. dollar.
  - Public-sector market-clearing exchange rate estimated between K 75–95 per U.S. dollar.
  - Shadow exchange rate for public imports QotE estimated at about K 130–200 kyat per U.S. dollar.
- Macroeconomic aggregates using EQE as accounting rate:
  - Using EQE increases nominal GDP by about 10–12 percent compared to official GDP figures (Box 1 baseline).
  - Trade openness (Exports + Imports)/GDP:
    - Official-statistics measure: less than 1 percent.
    - Re-estimated using EQE: about 20–23 percent.
  - Public-sector share in GDP increases to 12–13 percent from 7–8 percent (official).
  - Comparative note: Using MCR E as accounting rate would increase nominal GDP by about 30 percent, trade openness to about 50 percent, and public-sector share to about 20 percent.
- Net efficiency effects by market (baseline):
  - Public-sector exports: net efficiency effect positive ≈ 20–35 percent of total public export value (net efficiency gain for public sector).
  - Public-sector imports: net efficiency effect negative ≈ 70–73 percent of total public import value (large subsidy cost).
  - Private-sector exports: efficiency gain ≈ 42–45 percent of total private export value.
  - Private-sector imports: efficiency loss ≈ 92–110 percent of total private import value.
  - Overall: private sector incurs net efficiency losses larger than public sector; aggregate regime yields net efficiency losses for public sector as well.
- Net efficiency losses relative to GDP:
  - Using GDP based on EQE: net efficiency losses about 16–18 percent of nominal GDP during 2004/05–2006/07.
  - Using GDP based on MCR E: net efficiency loss estimates around 14–15 percent of nominal GDP.
  - Box 1 summary: total efficiency loss caused by the multiple exchange rate regime in 2006/07 ≈ 14–17 percent of GDP.

### VI. Detailed table highlights (selected figures preserved exactly as presented)
- Table 3 (selected entries):
  - 2006/07: Private Sector Market Clearing Exchange Rate = 1279.9, Official Exchange Rate = 5.7.
  - Trade values (in billions of kyat): Total = 13,674.
  - Public Sector Exports (value 4,089): net distribution reported as -4,071 (Union Government), 2,655 (State Economic Enterprise), 0 (Private Traders), -1,416 (Total? as presented).
  - Public Sector Imports (value 3,277): net distribution reported as 3,263, -873, 0, 2,390.
  - Private Sector Exports (value 3,081): net distribution 0, 0, -1,371, -1,371.
  - Private Sector Imports (value 3,226): net distribution 0, 0, 3,563, 3,563.
  - Additional summary figures shown: -808, 1,782, 2,191, 3,165, 424, 86, 195 (as presented).
- Table 4 (selected entries for 2006/07):
  - Equilibrium Exchange Rate reported: 18,490 (kyat/US$) in one row; Market Rate (Private Sector Clearing Rate) = 1280; Official Exchange Rate = 5.7.
  - Equilibrium Rate (total) row: -4.4, 9.6, 11.9, 17.1; value 424; Relative to Trade Values 21.2, 13.4.
  - Market Rate (total) row: -3.7, 8.1, 9.9, 14.3; Market Rate 1280; value 22,125; Relative to Trade Values 53.4, 21.8.
- Table 5 (sensitivity summaries, 2006/07):
  - Baseline Case: Net Efficiency Losses (Relative to Trade Values) by flow: Public Exports -0.35, Public Imports 0.73, Private Exports -0.45, Private Imports 1.10; Net Efficiency Losses (Percent in Estimated Nominal GDP) = 17.1; Nominal GDP = 424.
  - Robustness averages/medians across 32 parameter combinations:
    - Average: Net Efficiency Losses (Relative to Trade Values) -6.4, 12.5, -7.2, 17.1; Percent in Estimated Nominal GDP = 16.0; Nominal GDP = 475; Openness measure 163; Share of Public Sector in GDP 354; other metrics 1.12, 23.1, 13.9.
    - Median: Net Efficiency Losses -4.7, 11.9, -6.6, 16.9; Percent in Estimated Nominal GDP = 12.8; Nominal GDP = 463; Openness 122; Share of Public Sector 317; metrics 1.12, 22.9, 13.9.
  - Full range across 32 combinations: net efficiency loss ranged from 6 percent to 33 percent of GDP in 2006/2007 (as reported in sensitivity analysis).

### VII. Sensitivity analysis — directional findings
- Key directional effects reported:
  - Higher private import elasticity Rα → more depreciated EQE and higher total net efficiency losses.
  - Higher public import elasticity Uα → more depreciated EQE but overall slightly lower total net efficiency losses.
  - Higher private export elasticity Rβ → increases total net efficiency losses.
  - Higher public export elasticity Uβ → leads to a more appreciated EQE but increases total net efficiency losses due to larger private import losses.
  - Larger φ (greater public import opportunities relative to private) → reduces total net efficiency losses.
- Notable extreme cases in Table 5:
  - Maximum NWL case parameters: 1/α(R)=1.5, 1/α(U)=0.5, 1/β(R)=1.5, 1/β(U)=0.3, φ=6.5 → Percent in Estimated Nominal GDP = 32.6; Nominal GDP = 450.
  - Minimum NWL case parameters: 1/α(R)=0.5, 1/α(U)=1.5, 1/β(R)=0.5, 1/β(U)=0.1, φ=6.5 → Percent in Estimated Nominal GDP = 5.9; Nominal GDP = 405.

### VIII. Policy-relevant implications and main conclusions
- Unification benefits:
  - Unifying multiple rates would allow Myanmar to benefit from more efficient resource allocation.
- Distortions and measurement effects:
  - The multiple-rate, nontransparent system creates distortions, imposes high costs on participants, and likely leads to underestimation of GDP and external sector size in official statistics.
  - Using EQE for accounting increases nominal GDP by about 10–12 percent and trade openness to about 20–23 percent (from less than 1 percent).
- Aggregate cost:
  - Total efficiency loss caused by implicit tax and subsidy operations under the regime in 2006/07 reported at about 14–17 percent of GDP (baseline), with sensitivity range 6 percent–33 percent of GDP across parameter combinations.
- Caveats:
  - Estimates contingent on parameter assumptions and partial-equilibrium, static framework.
  - Actual efficiency loss likely larger because distortions from unrecorded trade (e.g., Hundi) and dynamic general-equilibrium effects are omitted.

*Source — content unit _wp08199 - References (extracted pages and summary from the supplied PDF content).*

### References..............................................................................................................

### _wp08199 - References

### I. Introduction — purpose and approach
- Objective: develop a model describing Myanmar’s foreign exchange markets, derive the equilibrium exchange rate, quantify distortions on exports and imports from the multiple exchange rate regime, and re-estimate key indicators (GDP, trade openness) using the estimated equilibrium exchange rate.
- Methodology:
  - Extend the model by Rosenberg and Zeeuw (2001) to reflect features specific to Myanmar’s exchange rate regime.
  - Numerically estimate net efficiency losses and the equilibrium exchange rate that would clear a unified foreign exchange market.
  - Re-estimate the size of the external sector in national accounts using the estimated equilibrium exchange rate.
- Comparative context: earlier studies on multiple exchange rates cited include Tarr (1990), Rosenberg, Ruocco, and Wiegard (1999), and Rosenberg and Zeeuw (2001), but their methodologies are not directly applicable to Myanmar.

### II. Multiple exchange rate regime in Myanmar — structure and mechanisms
- Characteristic features:
  - The regime takes the form of dual foreign exchange markets that are effectively segmented for public and private sector external transactions.
  - Official exchange rate is exogenously set far below the market-clearing level and is used for accounting purposes and settlement of external transactions in the public sector.
  - Private sector exporters have not faced a foreign exchange surrender requirement since the early 1990s.
- Public sector mechanics:
  - Public sector exporters are obligated to surrender 100 percent of their export proceeds to the government and choose export volumes given the official exchange rate.
  - Foreign exchange available for public imports is effectively the difference between public sector export earnings and the accumulation of international reserves, assuming capital flows are negligible.
  - Foreign exchange for public imports is rationed by the Foreign Exchange Budget managed by the Ministry of Finance and Revenue.
  - Foreign exchange revenues include export receipts of State Economic Enterprises (SEEs), export taxes, income tax on inward remittances, utility charges paid by foreign companies for public sector services, and sales of Foreign Exchange Certificates (FEC).
  - Foreign exchange expenditures consist mainly of import payments of SEEs and servicing external public sector debt.
- Market segmentation implications:
  - Because the official rate is grossly overvalued, public sector import demand exceeds available foreign exchange.
  - Most unofficial rates are market-determined; differences among those unofficial market-determined rates are negligible relative to the gap between the official rate and market-determined rates. For simplicity, analyses assume a dual exchange rate system.

### III. Key quantitative findings reported
- Estimated equilibrium exchange rate that would clear a unified foreign exchange market: around K 400–500 per U.S. dollar.
- Effect on nominal GDP when using the derived equilibrium exchange rate as the accounting rate: increases nominal GDP by about 10–12 percent compared to official statistics.
- Trade openness (exports + imports divided by nominal GDP):
  - Official-statistics measure: less than 1 percent.
  - Re-estimated using derived equilibrium rate: increases to about 20–23 percent.
- Total efficiency loss caused by the multiple exchange rate regime in 2006/07: estimated at about 14–17 percent of GDP.
- Time frames and data references:
  - Baseline parameters and estimates reference the period 2004/05-2006/07 and specific point estimates for 2006/07 are reported.
  - The paper also includes figures and tables mapped to pages and topics (e.g., Box 1, Figures 1–4, Tables 1–5) that document segmented foreign exchange markets, parallel market premia, and net efficiency loss calculations.

### IV. Policy-relevant implications (as framed by the paper)
- Unification of multiple rates would allow Myanmar to benefit from more efficient allocation of resources.
- The current multiple-rate, nontransparent system creates distortions, imposes high costs on participants, and likely leads to underestimation of GDP and external sector size in official statistics.

_Italic: Source — content unit _wp08199 - References (extracted pages and summary from the supplied PDF content)._

### Box 1: Myanmar’s Multiple Exchange Rate System

### Box 1: Myanmar’s Multiple Exchange Rate System

### Structure of the exchange rate and foreign-exchange arrangements
- Multiple exchange rate regime: an official exchange rate coexists with informal market exchange rates.
- Official exchange rate: pegged at K 8.50847 per SDR since 1977, valuation allowed to differ within a margin of +/- 2 percent.
- Foreign Exchange Certificates (FEC):
  - Issued as U.S. dollar equivalent; traded in informal markets at around 1,100 kyat per FEC.
  - Official FEC exchange centers set center rate at K 450 per dollar (transactions at centers limited).
  - FEC used by residents as substitute for U.S. dollar in domestic transactions; use of foreign currencies by residents not legal.
  - FEC used for large value items (e.g., air tickets); local employees of international/foreign organizations usually paid in FEC.
  - Holders of FEC and residents with foreign currency earnings may open foreign currency accounts.
  - Foreign firms often required to pay for utilities in U.S. dollars or FEC.
- Parallel/informal markets:
  - U.S. dollars exchanged with kyats at a premium over the FEC rate in informal parallel markets (at around K 1,120 in March 2008).
  - Hundi system widely used for unrecorded trade, repatriation of profits, and U.S. dollar cash remittances; Hundi rate close to parallel U.S. dollar rate but at a deduction of about 1 percent for commissions.
  - Market-determined U.S. dollar exchange rate and Hundi rate constitute main exchange rates for private-sector international transactions.
- Foreign-exchange controls supporting multiple rates:
  - Private imports require a license and are restricted by availability of export earnings.
  - Exchange restrictions limit purchase of foreign exchange by residents for foreign travel, transfers of remittable portion of wages by nonresidents, and payments relating to invisible and other current international transactions.
  - Residents with official permits to travel abroad allowed to buy equivalent of US$100-US$200 on presentation of FEC from own foreign currency accounts; maximum remittance against FEC is US$10,000 per month or its equivalent.

### Model framework and methodology
- Objective: develop a model reflecting Myanmar’s multiple exchange rate regime to calculate equilibrium exchange rate and net efficiency losses from implicit taxes/subsidies.
- Market segmentation in model: four constant-elasticity markets considered:
  - Public exports (supply function with parameters Uβ and UB).
  - Public imports (demand function with parameters Uα and UA).
  - Private exports (supply with parameters Rβ and RB).
  - Private imports (demand with parameters Rα and RA).
- Key exchange rates in model:
  - OFE = official exchange rate.
  - MC R E = private market clearing exchange rate (parallel market).
  - EQE = equilibrium exchange rate (unified market-clearing rate to be solved numerically).
  - QotE = shadow exchange rate for public import quota.
- Model extensions beyond prior studies:
  - Incorporates private-sector markets under the parallel exchange rate (data available for Myanmar).
  - Introduces explicit foreign-exchange supply constraint on public sector importers (bounded by public-sector export proceeds net of reserve accumulation).
  - Keeps private market clearing exchange rate and equilibrium exchange rate distinct in net efficiency loss formulae.
  - Derives numerical solution methods for EQE and QotE using observed price-quantity combinations and a parameter φ (private sector share relative to public sector).

### Efficiency loss/gain expressions (conceptual)
- Net efficiency losses for each market derived in closed-form expressions as functions of:
  - Implicit tax rate (tax = MCROF E E/1− by definition) or implicit subsidy rate (sub).
  - Price elasticity parameters (Uβ, Uα, Rβ, Rα).
  - Relative values of EQE, MCR E, OFE, and QotE.
- Public export net efficiency loss/gain (relative to public export value) expressed in equation form (see equations (5)–(9) in source).
- Public import net efficiency loss/gain (relative to public import value) expressed in equation form (see equations (10)–(11) in source), explicitly accounting for foreign-exchange quotas QotU M ,.
- Private export/import net efficiency losses (relative to private trade values) expressed in equations (12) and (13), functions of respective elasticities and EQE relative to MCR E.
- Equilibrium exchange rate condition given by equation (14) (market-clearing equality across four markets); QotE defined by equation (15).

### Baseline empirical assumptions and parameter choices
- Elasticity assumptions:
  - Baseline: unit elasticity assumed for all markets except public export market.
  - Public export elasticity Uβ assumed equal to 0.2 (smaller responsiveness due to SEEs' obligations).
- Observables:
  - OFE and MCR E treated as observable.
- Calibration approach:
  - Assume without loss of generality RURU MMXX PPPP ==== 1$$$$ (scaling normalization) to back out other parameters from observed price-quantity combinations.
  - Use φ (private sector share in GDP / public sector share in GDP) to relate private and public import demands under MCR E.

### Key quantitative baseline results (period 2004/2005 to 2006/07)
- Estimated conceptual exchange-rate levels:
  - Equilibrium exchange rate EQE estimated to lie between K 350–450 per U.S. dollar.
  - Public sector market clearing exchange rate estimated between K 75–95 per U.S. dollar.
  - Shadow exchange rate for public imports QotE estimated at about K 130–200 kyat per U.S. dollar.
- Effects on nominal GDP and aggregates (using EQE as accounting rate):
  - Using EQE increases nominal GDP by about 11–12 percent compared to official GDP figures.
  - Trade openness (Exports + Imports)/GDP increases to about 21–23 percent from less than 1 percent measured using official statistics.
  - Public sector share in GDP increases to 12–13 percent from 7–8 percent indicated by official statistics.
  - (Comparison note: Using MCR E as accounting rate would increase nominal GDP by about 30 percent, trade openness to about 50 percent, and public sector share in GDP to 20 percent—reported for comparison in the source.)
- Net efficiency effects by market (baseline):
  - Public sector exports: net efficiency effect positive at around 20–35 percent of total public export value (net efficiency gain for public sector due to tax revenue exceeding exporters’ surplus loss).
  - Public sector imports: net efficiency effect negative at around 70–73 percent of total public import value (subsidy expenditure larger than efficiency gain to importing SEEs).
  - Private sector exports: efficiency gain about 42–45 percent of total private export value.
  - Private sector imports: efficiency loss about 92–110 percent of total private import value.
  - Overall: private sector incurs net efficiency losses larger than the public sector; multiple exchange rate regime designed to benefit public sector in practice yields net efficiency losses for public sector as well.
- Net efficiency losses relative to GDP:
  - Net efficiency losses appear relatively constant during 2004/05 to 2006/07 at around 16–18 percent of nominal GDP using GDP estimates based on EQE.
  - Using GDP estimates based on MCR E yields net efficiency loss estimates around 14–15 percent of nominal GDP.
  - Aggregate figures are an intuitive ballpark; caveat that simple aggregation of partial-equilibrium results may overlap and is not exact.

### Sensitivity analysis and robustness
- Parameter variation:
  - Explored sensitivity to five basic parameters: φ, Uβ, Uα, Rβ, Rα across intervals:
    - φ in [3/7, 7]
    - Rα/1 in [0.5, 1.5]
    - Rβ/1 in [0.5, 1.5]
    - Uα/1 in [0.5, 1.5]
    - Uβ/1 in [0.3, 1.0]
  - All 32 (= 2^5) combinations evaluated for 2006/2007.
- Robustness outcomes:
  - Median/average across combinations not very different from baseline:
    - EQE estimated around K 470 per U.S. dollar.
    - Nominal GDP about 12 percent higher than official statistics.
    - Trade openness roughly 23 percent of GDP.
    - Public sector share about 14 percent of GDP.
    - Total net efficiency losses about 13–16 percent of GDP.
  - Full range across 32 parameter combinations: estimated net efficiency loss ranged from 6 percent to 33 percent of GDP in 2006/2007.
- Directional impacts summarized:
  - Higher private sector import elasticity Rα leads to more depreciated EQE and higher total net efficiency losses.
  - Higher public sector import elasticity Uα leads to more depreciated EQE but overall slightly lower total net efficiency losses.
  - Higher private export elasticity Rβ increases total net efficiency losses.
  - Higher public export elasticity Uβ leads to a more appreciated EQE but increases total net efficiency losses due to larger private import losses.
  - Larger φ (greater public import opportunities relative to private) reduces total net efficiency losses.

### Main conclusions from the analysis (summary of findings)
- Myanmar’s multiple exchange rate regime creates dual foreign-exchange markets effectively segmented between public and private sector external transactions.
- Baseline analysis indicates:
  - EQE to clear unified market could be about K 450 per U.S. dollar.
  - Using EQE for accounting increases nominal GDP by about 10–12 percent versus official statistics; trade openness rises to about 20–23 percent from less than 1 percent.
  - Total efficiency loss caused by implicit tax and subsidy operations under the regime is about 14–17 percent of GDP in 2006/07, with private importers mainly hurt.
- Caveats:
  - Estimates are contingent on parameter assumptions; sensitivity analysis yields a wide margin (net efficiency loss between 6 percent and 33 percent of GDP across parameter combinations).
  - Actual efficiency loss likely larger because distortions from unrecorded trade and dynamic effects are omitted in the static partial-equilibrium analysis.

*Box 1: Myanmar’s Multiple Exchange Rate System (excerpt).*

### References

### References

### Bibliographic references cited
- Bahmani-Oskooee, Mohsen and Orhan Kara, 2005, “Income and Price Elasticities of Trade: Some New Estimates,” International Trade Journal, Volume XIX, No.2, (Summer), pp 165–178.
- Caporale, Guglielmo Maria and Michael K. F. Chui, 1999, “Estimating Income and Price Elasticities of Trade in a Cointegration Framework,” Review of International Economics, Vol. 7, pp 254–264.
- Huizinga, Harry, 1996, “The Taxation Implicit in Two-Tiered Exchange Rate Systems,” IMF Working Paper No. 96/120 (Washington: International Monetary Fund).
- Larkin, Stuart and U San Thein, 2001, Currency Reform in Myanmar (unpublished, Embassy of Japan, Yangon).
- Reinhart, Carmen, 1995, “Devaluation, Relative Prices, and International Trade-Evidence from Developing Countries,” Staff Papers, International Monetary Fund, Vol. 42, No.2, (June), pp 290–312.
- Romer, Paul, 1994, “New goods, Old Theory, and The Welfare Costs of Trade Restrictions,” Journal of Development Economics, Vol. 43, pp 5–38.
- Rosenberg, Christoph, Anna Ruocco, and Wolfgang Wiegard, 1999, “Explicit and Implicit Taxation in Uzbekistan,” in Herausforderungen an die Wirtschaftspolitik an der Schwelle zum 21. Jahrhundert, Festschrift fur Lutz Hoffmann zum 65. Geburtstag, ed. by Irmgard Nubler and Harald Trabold (Berlin: Duncker und Humbolt), pp. 141–162.
- Rosenberg, Christoph, Anna Ruocco, Wolfgang Wiegard, and Maarten De Zeeuw, 2001, “Welfare Effects of Uzbekistan’s Foreign Exchange Regime,” Staff Papers, International Monetary Fund, Volume 48, No.1, pp 160–178.
- Senhadji, Abdelhak, 1998, “Time-Series Estimation of Structural Import Demand Equations: A Cross-Country Analysis,” Staff Papers, International Monetary Fund, Volume 45, No.2, (June), pp 236–268.
- Senhadji, Abdelhak, and Claudio Montenegro, 1999, “Time Series Analysis of Export Demand Equations: A Cross-Country Analysis,” Staff Papers, International Monetary Fund, Vol. 46, No.3 (September/December), pp 259–273.
- Sinha, Dipendra, 2001, “A Note on Trade Elasticities in Asian Countries,” International Trade Journal, Vol. XV, No.2, (Summer), pp 221–236.
- Tarr, David, 1990, “Second-Best Foreign Exchange Policy in the Presence of Domestic Price Controls and Export Subsidies,” World Bank Economic Review, Vol. 4, No.2, pp 175–193.

### Table 2 — Estimated Price Elasticities of Trade Flows in Earlier Studies (coverage and elasticities)
- Reinhart (1995): 12 (12) countries; Exports elasticities: 0.3 (Average-All Countries), 0.3 (Average-Developing Countries), 0.6 (0.8-quartile in Developing Countries), 0.0 (0.2-quartile in Developing Countries). Imports elasticities: 0.6 (Average-All Countries), 0.6 (Average-Developing Countries), 1.3 (0.8-quartile in Developing Countries), 0.3 (0.2-quartile in Developing Countries).
- Senhadji (1998): 66 (47) countries; Exports elasticities: na for the first group; Imports elasticities: 1.1 (Average-All Countries), 1.1 (Average-Developing Countries), 1.6 (0.8-quartile in Developing Countries), 0.3 (0.2-quartile in Developing Countries).
- Senhadji and Montenegro (1999): 53 (36) countries; Exports elasticities: 1.1 (Average-All Countries), 1.1 (Average-Developing Countries), 1.8 (0.8-quartile in Developing Countries), 0.3 (0.2-quartile in Developing Countries); Imports elasticities: na.
- Caporale and Chui (1999): 21 (2) countries; Exports elasticities: 0.9 (Average-All Countries), 0.7 (Average-Developing Countries); Imports elasticities: 0.5 (Average-All Countries), 0.4 (Average-Developing Countries).
- Sinha (2001): 5 (4) countries; Exports elasticities: 1.2 (Average-All Countries), 1.2 (Average-Developing Countries); Imports elasticities: 0.5 (Average-All Countries), 0.3 (Average-Developing Countries).
- Bahmani-Oskooee and Kara (2005): 28 (7) countries; Exports elasticities: 1.0 (Average-All Countries), 0.4 (Average-Developing Countries); Imports elasticities: 1.2 (Average-All Countries), 0.5 (Average-Developing Countries).
- Tarr (1990): Assumed elasticities.
- Rosenberg and Zeeuw (2001): Assumed elasticities.
- Coverage note: Numbers in parentheses are the number of developing countries.
- Reported ranges in the table header: Exports elasticities 0.3-1.16 and 0.3-1.0; Imports elasticities 0.5-1.5 and 0.5-2.0 (as presented).

### Figure 2 — Exchange rate series (captioned)
- Figure: "Myanmar's Parallel Market Exchange Rate and Its Premium over the Official Exchange Rate from 1960"
- Axes and annotations in the figure: Parallel Market Premium (percent, right scale); Official Exchange Rate (Kyat/US$, left scale); Parallel Market Exchange Rate (Kyat/US$, left scale).
- Time span shown: 1960 through 2006.

### Table 3 — Net Efficiency Loss Estimates under the Baseline Parameters (2004/05–2006/07)
- Assumed parameters (baseline): 1/α(R) = 1.0; 1/α(U) = 1.0; 1/β(R) = 1.0; 1/β(U) = 0.2; φ = 6.5.
- Results summary for selected years (values reported in tables):
  - 2004/05 (Private Sector Market Clearing Exchange Rate=921.4, Official Exchange Rate=5.7)
    - Trade values (in billions of kyat): Total = 6,945.
    - Net efficiency losses by agent and flow (values shown in table rows):
      - Public Sector Exports: R contributions -0.994, U contributions 0.742, Con -0.251; value 1,800; net distribution: -1,789 (Union Government), 1,336 (State Economic Enterprise), 0 (Private Traders), -453 (Total? as presented).
      - Public Sector Imports: 0.994, -0.278, 0.000, 0.716; value 1,864; net distribution: 1,852, -519, 0, 1,334.
      - Private Sector Exports: 0.000, 0.000, -0.423, -0.423; value 1,590; net: 0, 0, -673, -673.
      - Private Sector Imports: 0.000, 0.000, 0.938, 0.938; value 1,692; net: 0, 0, 1,587, 1,587.
    - Aggregate indicators shown: Total = 6,945; additional numbers in the table: 64, 818, 914, 1,795, 361, 76, 129 (as presented).
  - 2005/06 (Private Sector Market Clearing Exchange Rate=1104.1, Official Exchange Rate=5.8)
    - Trade values (in billions of kyat): Total = 8,622.
    - Public Sector Exports: -0.995, 0.785, 0.000, -0.210; value 2,170; net: -2,159, 1,703, 0, -456.
    - Public Sector Imports: 0.995, -0.294, 0.000, 0.701; value 2,214; net: 2,202, -650, 0, 1,552.
    - Private Sector Exports: 0.000, 0.000, -0.420, -0.420; value 2,059; net: 0, 0, -866, -866.
    - Private Sector Imports: 0.000, 0.000, 0.919, 0.919; value 2,179; net: 0, 0, 2,003, 2,003.
    - Additional table figures: 43, 1,053, 1,137, 2,233, 440, 93, 168 (as presented).
  - 2006/07 (Private Sector Market Clearing Exchange Rate=1279.9, Official Exchange Rate=5.7)
    - Trade values (in billions of kyat): Total = 13,674.
    - Public Sector Exports: -0.996, 0.649, 0.000, -0.346; value 4,089; net: -4,071, 2,655, 0, -1,416.
    - Public Sector Imports: 0.996, -0.266, 0.000, 0.729; value 3,277; net: 3,263, -873, 0, 2,390.
    - Private Sector Exports: 0.000, 0.000, -0.445, -0.445; value 3,081; net: 0, 0, -1,371, -1,371.
    - Private Sector Imports: 0.000, 0.000, 1.104, 1.104; value 3,226; net: 0, 0, 3,563, 3,563.
    - Additional table figures: -808, 1,782, 2,191, 3,165, 424, 86, 195 (as presented).
- Source for Table 3: IMF staff estimates.

### Table 4 — Net Efficiency Loss Estimates under the Baseline Parameters (Relative to Estimated Nominal GDP)
- Reported variables in the table include: Public Sector Market Clearing Rate (kyat/US$), Shadow Exchange Rate for Public Imports, Value in billions of kyat, Relative to Trade Values, Equilibrium Exchange Rate (kyat/US$), Relative to Estimated GDP.
- Selected reported exchange rates and aggregates for years:
  - 2004/05:
    - Equilibrium Rate (total) row: 0.6, 8.1, 9.0, 17.7; value 361; Equilibrium Exchange Rate 10,135 (kyat/US$); Relative to Trade Values 22.6, 12.3.
    - Official Rate (total) row: 0.7, 9.0, 10.1, 19.8; Official Exchange Rate 5.7; value 9,079; Relative to Trade Values 0.3, 7.0.
    - Market Rate (total)(Private Sector Clearing Rate): 0.5, 6.9, 7.7, 15.2; Market Rate 921; value 11,801; Relative to Trade Values 49.6, 18.7.
  - 2005/06:
    - Equilibrium Rate (total): 0.3, 7.7, 8.4, 16.4; value 440; Equilibrium Exchange Rate 13,599; Relative to Trade Values 21.3, 11.8.
    - Official Rate (total): 0.4, 8.6, 9.3, 18.2; Official Exchange Rate 5.8; value 12,287; Relative to Trade Values 0.3, 7.0.
    - Market Rate (total(Private Sector Clearing Rate)): 0.3, 6.7, 7.3, 14.3; Market Rate 1104; value 15,608; Relative to Trade Values 46.5, 17.6.
  - 2006/07:
    - Equilibrium Rate (total): -4.4, 9.6, 11.9, 17.1; value 424; Equilibrium Exchange Rate 18,490; Relative to Trade Values 21.2, 13.4.
    - Official Rate (total): -4.8, 10.7, 13.1, 18.9; Official Exchange Rate 5.7; value 16,716; Relative to Trade Values 0.3, 8.0.
    - Market Rate (total)(Private Sector Clearing Rate): -3.7, 8.1, 9.9, 14.3; Market Rate 1280; value 22,125; Relative to Trade Values 53.4, 21.8.
- Source for Table 4: IMF staff estimates.

### Table 5 and Sensitivity Analyses — Robustness checks and parameter variation (2006/07)
- Baseline assumed parameters (repeated): 1/α(R) = 1.0; 1/α(U) = 1.0; 1/β(R) = 1.0; 1/β(U) = 0.2; φ = 6.5.
- Sensitivity analysis scenarios presented (selected rows and outputs):
  - Baseline Case (2006/07): Net Efficiency Losses (Relative to Trade Values) by flow: Public Exports -0.35, Public Imports 0.73, Private Exports -0.45, Private Imports 1.10; Total (Relative to Trade Values) -7.7, 12.9, -7.4, 19.3; Nominal GDP (in billion kyat) = 424; Net Efficiency Losses (Percent in Estimated Nominal GDP) = 17.1; Share of Public Sector in GDP = 0.71? (table shows multiple columns; values presented as in the table: Average -0.28, 0.71, -0.43, 0.98; -6.4, 12.5, -7.2, 17.1; 16.0; Nominal GDP 475; Openness measure 163; Share of Public Sector 354; other metrics 1.12, 23.1, 13.9).
  - Variation: Private Import Elasticity (1/α(R)) = 0.5 (others baseline): Public/Private flow impacts: -0.65, 0.81, -0.48, 1.11; Net Efficiency Losses (Relative to Trade Values): -14.9, 14.9, -8.4, 20.3; Nominal GDP = 252; Percent in Estimated Nominal GDP = 11.9; additional reported metrics: 1.06, 13.1, 11.3.
  - Variation: 1/α(R)=1.5 (others baseline): -0.08, 0.69, -0.40, 1.02; Net Efficiency Losses (Relative to Trade Values): -1.8, 11.8, -6.5, 17.1; Nominal GDP = 563; Percent in Estimated Nominal GDP = 20.7; additional metrics: 1.14, 27.3, 15.0.
  - Variation: Public Import Elasticity (1/α(U)) = 0.5: -0.38, 0.85, -0.45, 1.15; Net Efficiency Losses: -8.5, 15.2, -7.5, 20.2; Nominal GDP = 404; Percent in Estimated Nominal GDP = 19.3; metrics: 1.10, 20.3, 13.2.
  - Variation: 1/β(R)=0.5 (Private Export Elasticity): -0.39, 0.74, -0.55, 1.16; Net Efficiency Losses: -8.6, 13.1, -9.2, 20.3; Nominal GDP = 403; Percent in Estimated Nominal GDP = 15.6; metrics: 1.10, 20.2, 13.2.
  - Variation: 1/β(U)=1.5 (Public Export Elasticity): -0.32, 0.72, -0.37, 1.07; Net Efficiency Losses: -7.1, 12.8, -6.2, 18.7; Nominal GDP = 437; Percent in Estimated Nominal GDP = 18.2; metrics: 1.11, 21.8, 13.6.
  - Variation: Public Sector Potential Import (φ) = 0.1 and φ = 0.3 among others, with corresponding changes (examples shown): φ = 0.1 row shows -0.36, 0.69, -0.40, 0.82; Net Efficiency Losses: -7.8, 11.8, -6.5, 13.7; Percent in Estimated Nominal GDP = 11.3; Nominal GDP = 566; metrics: 1.14, 27.4, 15.0. φ = 0.3 row shows -0.36, 0.77, -0.47, 1.39; Net Efficiency Losses: -8.1, 14.0, -8.0, 24.9; Percent in Estimated Nominal GDP = 22.7; Nominal GDP = 319; metrics: 1.08, 16.4, 12.2.
  - Maximum NWL case: parameters 1/α(R)=1.5, 1/α(U)=0.5, 1/β(R)=1.5, 1/β(U)=0.3, φ=6.5; reported flows: 0.00, 0.84, -0.37, 1.37; Net Efficiency Losses (Relative to Trade Values): 0.0, 14.9, -6.1, 23.8; Percent in Estimated Nominal GDP = 32.6; Nominal GDP = 450; metrics: 1.11, 22.3, 13.7.
  - Minimum NWL case: parameters 1/α(R)=0.5, 1/α(U)=1.5, 1/β(R)=0.5, 1/β(U)=0.1, φ=6.5; flows: -0.56, 0.69, -0.55, 0.88; Net Efficiency Losses: -12.4, 12.2, -9.2, 15.3; Percent in Estimated Nominal GDP = 5.9; Nominal GDP = 405; metrics: 1.10, 20.3, 13.2.
- Summary statistics reported in sensitivity table:
  - Average: -0.28, 0.71, -0.43, 0.98; Net Efficiency Losses (Relative to Trade Values) -6.4, 12.5, -7.2, 17.1; Percent in Estimated Nominal GDP = 16.0; Nominal GDP = 475; Openness measure 163; Share of Public Sector in GDP 354; other metrics 1.12, 23.1, 13.9.
  - Median: -0.22, 0.67, -0.40, 0.95; Net Efficiency Losses -4.7, 11.9, -6.6, 16.9; Percent in Estimated Nominal GDP = 12.8; Nominal GDP = 463; Openness 122; Share of Public Sector 317; metrics 1.12, 22.9, 13.9.
- Source for Table 5 and sensitivity analyses: IMF staff estimates.

*Source: _wp08199 - References (IMF staff estimates and cited references).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp08199.pdf_