Your SlideShare is downloading. ×
0
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×
Saving this for later? Get the SlideShare app to save on your phone or tablet. Read anywhere, anytime – even offline.
Text the download link to your phone
Standard text messaging rates apply

Firm Heterogeneity and Costly Trad: An Estimation Strategy and Policy Experiments

332

Published on

NES 20th Anniversary Conference, Dec 13-16, 2012 …

NES 20th Anniversary Conference, Dec 13-16, 2012
Firm Heterogeneity and Costly Trade: An Estimation Strategy and Policy Experiments (based on the article presented by Ivan Cherkashin at the NES 20th Anniversary Conference).
Authors: I.Cherkashin (Australian National University, MAE'2003); S.Demidova (McMaster University, MAE'2002); H.L. Kee (World Bank); K. Krishna (Penn State University and NBER)

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total Views
332
On Slideshare
0
From Embeds
0
Number of Embeds
0
Actions
Shares
0
Downloads
3
Comments
0
Likes
0
Embeds 0
No embeds

Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
No notes for slide

Transcript

  • 1. Firm Heterogeneity and Costly Trade:An Estimation Strategy and Policy ExperimentsI.Cherkashin (Australian National University, MAE2003) S.Demidova (McMaster University, MAE2002) H.L. Kee (World Bank) K. Krishna (Penn State University and NBER) December 15, 2012
  • 2. Motivation USA, 2000: African Growth and Opportunity Act (AGOA) * Exporter - Madagascar - Duty free & Quota free - In 2000: Exports to USA - $170 million - In 2004: Exports to USA - $500 million - Exports to ROW: from $750 million to $875 million Europe, 2001: Everything but Arms Initiative (EBA) * Exporter - Bangladesh - Duty free & Quota free - In 2000: Exports to EU - $1.3 billion - In 2004: Exports to EU - $3.0 billion - Exports to the USA increased by $30 million
  • 3. Motivation Widespread trade agreements * Preferential treatment (EBA, GSP (MFN), AGOA) * Intended to help LDCs * Large number of complex eligibility restrictions Limited work evaluating preferences and restrictions * Back of the envelope calculations (No entry) * Little work using models with rm heterogeneity Modeling challenges * Not enough information readily available (on xed costs, market entry costs, documentation costs, parameters of distributions) * Complexities with modeling of rm behavior: multiple choices
  • 4. In this paper... Tractable partial equilibrium model · la Melitz with two dimensions of heterogeneity: * Productivity * Firm/Market specic demand shocks: hierarchy violations Cross-section data based estimation: extends applicability * Cost: ignore dynamics and information therein * Maybe ways to incorporate some such information Estimation procedure to obtain all structural parameters: * Structure of xed costs paid to enter industry or market, to produce, and documentation costs * Parameters of underlying market specic distributions of demand shocks * Parameters of underlying distributions of productivity * Elasticities of substitution
  • 5. Applying estimated model to: Model suitable for policy analysis and applied to Bangladeshi (BD) exports of apparel to US and EU Trade facilitation counterfactuals * Trade preferences, cost of preferences, .. * Fixed cost subsidies * Long-run versus short-run effects Results on * US, EU welfare and BD Exports * SR and LR
  • 6. The Application Apparel producers in Bangladesh 2 main sub-sectors, Woven & Non-Woven goods: only Woven ìMens/Boys Cotton Trousers,î HS 620342) 2 primary markets: EU & US with about 93% of BD exports US has quotas on 65-75% of imports so must get origin, no preferences, loose ROOs EU has preferences, no quotas, and stricter ROOs Size of US and EU potential market is similar
  • 7. Trade Policy in EU and US Preferences with strict ROOs in the EU * Preferences: - Preferential tariff (0% instead of 12-15%) * ROOs: ìYarn Forwardî * Costs of meeting ROOs: - Possibly higher marginal costs (domestic cloth 20% premium) - Documentation costs No Preferences with weaker ROOs in US * ROOs in the US - Assembly only - Costs of meeting ROOs are low
  • 8. Overview of Results Exports * Trade facilitation by EU raises BD exports to EU by a lot * Cross-market effects: Also raises BD exports to US by a lot, and welfare * Large export effects of trade facilitation * Fixed cost subsidies and exports: - Up to 81-1 leverage - Applying subsidy at later stage produces greater results Welfare * Welfare results: ìwin-win-winî scenarios possible * Fixed cost subsidies differ in their welfare effects * Broader policy relevance: trade as aid, role of US quotas
  • 9. Related Literature Demidova, Svetlana, Hiau Looi Kee, and Kala Krishna (2012). ìDo Trade Policy Differences Induce Sorting? Theory and Evidence From Bangladeshi Apparel Exporters,î Journal of International Economics 87(2), 247ñ261. Eaton, Jonathan, Samuel Kortum, and Francis Kramarz (2011). ìAn Anatomy of International Trade: Evidence from French Firms,î Econometrica 79(5), 1453-1498. Das, Sanghamitra, Mark Roberts, and James Tybout (2007). ìMarket Entry Costs, Producer Heterogeneity, and Export Dynamics,î Econometrica 75(3), 837-873. Krishna, Kala (2006). ìUnderstanding Rules of Origin,î Estevadeordal et al. (eds.), Rules of Origins, Oxford University Press. Melitz, Marc J. (2003). ìThe Impact of Trade on Intra-Industry Reallocations and Aggregate Industry Productivity,î Econometrica 71(6), 1695-1725.
  • 10. Demand Side Utility function in country j : s j 1 ! s s1 j j Uj = Âi2Wj [Xij ] sj Demand in country j for apparel from country i is:   s s1 j R 1 s j 1 j Xij = v w sj q (w )] w 2Wij [ ij ( )] [ ij sj dw Demand for a particular variety qij (w ) :    sj pij (w ) qij (w ) = vij (w ) Xij Pij
  • 11. Marginal Costs, Pricing, and Revenue Prot of rm:   1 1 wt ij p ij (f, vij , tij , t ij ) = (1  tij ) pij (f)  (1tij ) rj af qij (f) 1 sj t ij pij (f) = (1  tij ) sj  1 af tij is market specic tariff, t ij are market specic transportation costs, f is rm specic productivity, a 6 1 is cost disadvantage ROOs: * If meet ROOs, a < 1, and tij = 0. * If do not meet ROOs, a = 1, and tij > 0.
  • 12. Model Timing Stage 1 Stage 2 Stage 3 Pay f and export Low without Productivity Exiters meeting Entrants ROOs Survivors `s are learnt with high Potential by firms demand Entrants shock High Productivity Market Pay f and d, Entrants Entrants and invoke ROOs Pay entry costs Pay fixed costs to access a desirable `s are learnt and market or markets (EU and/or US) by firms for Exiters randomly draw and learn a market and firm specific all markets productivity demand shock tried
  • 13. Stage 3: If ROOs are an Option, then Trade-off Firms enter the market and pay to meet ROOs Firms enter the market, Firms choose but do not pay not to enter to meet ROOs and do not pay Firms draw demand shock, but stay out Two bounds for EU Only lower bound for US
  • 14. Data Bangladeshi customs data (ìuniverseî) for 2004 nancial year. In this version * HS 620342 sub-code only (about 800 rms). * Distribution of prices and quantities for AUS, OUS, OEU rms. * Shares of AUS, OEU, and OUS rms. * Share of rms invoking ROO in EU market. * Do NOT use panel dimension of the data. UN Comtrade database * Total US and EU imports of woven apparel from Bangladesh * Total US and EU imports of woven apparel
  • 15. Estimation Outline Latent: Ex-Ante Parameterization Exogenous MODEL - Solve for demand shocks cutoffs Tariffs, TFP distribution TF Weibull: 2 parameters - Solve for productivity cutoffs Quota price Guess - Calculate price indices In EU: Weibull, guess Transportation Demand shocks 2 parameters, Distributions EU, US In US: Weibull, guess Market Size: 2 parameters EU, US Market entry costs: Market entry costs: fm,EU in EU, fm,US in fm,EU in EU and US Truncated Truncated Share of Truncated fm,US in US Price AUS, OEU, OUS demand shocks Quantity Distributions Distributions and ROO firms distributions Fixed costs of f production: f Data: Data: Data: Data: Documentation Price Quantity Share AUS, OEU, Demand Shocks d costs: d Distributions Distributions OUS, ROO Calculated firm by firm using price, quantity Elasticities of Elasticities: data, price substitution EU and US indices and OBJECTIVE elasticities of Industry entry costs: New guess FUNCTION fe substitution fe
  • 16. Some Identication Intuition Matching shares of AUS, OUS, OEU rms helps match variance of demand shock distributions - more variance tends to raise OUS share. Matching shares of rms that meet ROOs helps identify d . f Matching the position of the quantity distributions helps pin down f . Matching distributions sheds light on remaining parameters.
  • 17. Some Exogenous Inputs Table 1: Trade Policy Parameters a t tROO t+µ EU 0.85 0.12 0 1.14 US 1 0.2 0.2 1.14+0.07
  • 18. Results: Productivity Distributions AUS rms for both EU and US markets rms t is good OEU, OUS rms distributions of price and quantity t relatively badly Model has OEU and OUS rms being low productivity (high price) compared to the data * High productivity rms need very bad EU or US shock to be OEU or OUS Capacity constraint in real world? * Only demand shock matters if there are capacity constraints * Lets high productivity (low price) rms sell to only one market * Limited quantity
  • 19. Results: Demand Shocks Distribution of demand shocks EU US Estimate Std. Err. Estimate Std. Err. Shape (g) 0.32 0.008 0.17 0.003 Scale (l) 1.39 0.087 0.57 0.020 Implied means and Coefcient of Variation Implied mean shock 10.4 421.8 Coefcient of variation 4.9 30.7
  • 20. Results: Demand Shocks Distributions t well overall US demand shocks mean and variance are higher than in EU Marketing differences: Chain store effect?
  • 21. Other Estimates: Elasticities of Substitution Elasticities of substitution EU US s 1.34 1.45 Std. Error 0.03 0.03
  • 22. Results: Structure of Fixed Costs Fixed costs in $ terms Estimate Std. Error Market Entry Costs EU fm 251,250 19,054 US fm 67,869 5,237 Documentation Costs d 4,240 317 Industry Entry Costs fe 77,348 5,372 Fixed Production costs f 6,404 476
  • 23. Policy Experiments Scenarios * Complete removal of preferences for Bangladeshi rms * Changing costs of meeting ROOs (Marginal & Fixed) * Subsidizing various xed costs * Quota prices: endogenous / exogenous Long-run vs Short-run BD * Mass of industry entrants (ME ) is xed * Firms can only exit markets Track changes in: * Export revenues * Tariff revenues * Welfare * Entry of rms
  • 24. Digression: Price Index and Small Country RBD,j is total Bangladeshi sales to j = EU, US: COMTRADE Rj is total exports of apparel to j = EU, US: COMTRADE (PBD,j )1sj RBD,j = Rj . (PBD,j )1sj + Âi2W(BD) [Pi,j ]1sj (PBD,j )1sj comes from estimation. Solve for Âi2W [Pi,j ]1sj = PBD,j . (BD) ¯ In our simulations we keep this index xed in accordance with our partial equilibrium assumptions.
  • 25. Endogenous Quota Price: Setup Survey: Original quota price in the US market is ~7% * This level is used in estimation Allow quota price to change, keeping quantity exported from BD to US constant (QBD,US ) * Note: Export revenue changes via price index changes Solve for model unknowns & for a new quota price Compare results to exogenous quota price case
  • 26. Long-Run Equilibrium Implications of Policy Changes Baseline No preferences Higher doc. costs No yarn req. Tariff EU: ROO / NO 0% / 12% 12% / 12% 0% / 12% 0% or 12% Tariff in US 20% 20% 20% 20% Cost disadvantage 0.85 1.00 0.85 1.00 Documentation costs d/f 0.66 0.00 1.32 0.66 Endogenous quota price setting Quota license price (change) =0.07 -100% -5.7% +43.4% EU imports from BD 482.3m -31.7% -1.5% +17.1% US imports from BD 233.6m -11.9% -0.1% +1.1% Implied mass of entrants 4,712 -22.3% -0.7% +5.8 Price index in EU 100% +19.1% +0.87% -9.38% Price index in US 100% +1.1 +0.01% -0.1% Share of ROO rms 70.2% 0% 57% 77.7% Tariff Revenue in EU 447k +8,742% +125.9% -34.2% Tariff Revenue in US 46,728k -11.9% -0.1% +1.1% Change in welfare EU ó -480,936k -25,208k +293,418k Change in welfare US ó -68,538k -709k +6,191k Exogenous quota price setting EU imports from BD 482.3m -45.5% -2.24% +22.7% US imports from BD 233.6m -41.6% -1.94% +14.3% Change in welfare EU ó -707,595k -37,343k 391,918k Change in welfare US ó -238,328k -11,193k 82,650k
  • 27. Long-Run Equilibrium Implications of Policy Changes Removal of preferences (lose - lose) Reduces cost of production of ROOs which reduces price charged, but adds tariffs, which raises price charged. Net effect: * Increase in price indices from less entry Welfare Effects * EU: TR increases by 8,742%, CS falls, welfare falls * US: TR falls 11.9%, CS falls, welfare falls * EU policy reduces US welfare: lose - lose US quotas provide insulation: BD quotas made less binding
  • 28. Long-Run Equilibrium Implications of Policy Changes Baseline No preferences Higher doc. costs No yarn req. Tariff EU: ROO / NO 0% / 12% 12% / 12% 0% / 12% 0% or 12% Tariff in US 20% 20% 20% 20% Cost disadvantage 0.85 1.00 0.85 1.00 Documentation costs d/f 0.66 0.00 1.32 0.66 Endogenous quota price setting Quota license price (change) =0.07 -100% -5.7% +43.4% EU imports from BD 482.3m -31.7% -1.5% +17.1% US imports from BD 233.6m -11.9% -0.1% +1.1% Implied mass of entrants 4,712 -22.3% -0.7% +5.8 Price index in EU 100% +19.1% +0.87% -9.38% Price index in US 100% +1.1 +0.01% -0.1% Share of ROO rms 70.2% 0% 57% 77.7% Tariff Revenue in EU 447k +8,742% +125.9% -34.2% Tariff Revenue in US 46,728k -11.9% -0.1% +1.1% Change in welfare EU ó -480,936k -25,208k +293,418k Change in welfare US ó -68,538k -709k +6,191k Exogenous quota price setting EU imports from BD 482.3m -45.5% -2.24% +22.7% US imports from BD 233.6m -41.6% -1.94% +14.3% Change in welfare EU ó -707,595k -37,343k 391,918k Change in welfare US ó -238,328k -11,193k 82,650k
  • 29. Long-Run Equilibrium Implications of Policy Changes Long-run: x2 documentation costs (lose - lose) Fewer rms meet ROOs so lower cost and price, but pay tariffs so higher price. * Small increase in price indices from less entry * Small changes in cutoffs Welfare Effects, Endogenous quota price * EU: TR rises by 125.9%, CS falls, welfare falls by $25.0m * US: TR falls by 0.1%, CS falls, welfare falls by $0.7m EU policy reduces US welfare US quotas would provide insulation: BD quotas made less binding.
  • 30. Long-Run Equilibrium Implications of Policy Changes Baseline No preferences Higher doc. costs No yarn req. Tariff EU: ROO / NO 0% / 12% 12% / 12% 0% / 12% 0% or 12% Tariff in US 20% 20% 20% 20% Cost disadvantage 0.85 1.00 0.85 1.00 Documentation costs d/f 0.66 0.00 1.32 0.66 Endogenous quota price setting Quota license price (change) =0.07 -100% -5.7% +43.4% EU imports from BD 482.3m -31.7% -1.5% +17.1% US imports from BD 233.6m -11.9% -0.1% +1.1% Implied mass of entrants 4,712 -22.3% -0.7% +5.8 Price index in EU 100% +19.1% +0.87% -9.38% Price index in US 100% +1.1 +0.01% -0.1% Share of ROO rms 70.2% 0% 57% 77.7% Tariff Revenue in EU 447k +8,742% +125.9% -34.2% Tariff Revenue in US 46,728k -11.9% -0.1% +1.1% Change in welfare EU ó -480,936k -25,208k +293,418k Change in welfare US ó -68,538k -709k +6,191k Exogenous quota price setting EU imports from BD 482.3m -45.5% -2.24% +22.7% US imports from BD 233.6m -41.6% -1.94% +14.3% Change in welfare EU ó -707,595k -37,343k 391,918k Change in welfare US ó -238,328k -11,193k 82,650k
  • 31. Long-Run Equilibrium Implications of Policy Changes Long-run: no yarn requirement (win - win) Liberalizing preferences raises entrants by around 5.8% in industry 9.4% fall in price index in EU, and 0.1% fall in US from lower cost and price and more entry Large changes in cutoffs Welfare effects: * EU: TR falls by 34.2%, CS rises, welfare rises * US: TR rises by 1.1%, CS rises, welfare rises EU policy raises US welfare: win - win scenario US quotas would insulate: BD quotas made more binding. Also would reduce positive impact on EU as less entry occurs.
  • 32. Long-Run Equilibrium: Extra Experiment Baseline No yarn req. Costless Preferences Tariff EU: ROO / NO 0% / 12% 0% or 12% 0% or 12% Tariff in US 20% 20% 20% Cost disadvantage 0.85 1.00 1.00 Documentation costs d/f 0.66 0.66 0.00 Endogenous quota price setting Quota license price (change) =0.07 +43.4% +49.3% EU imports from BD 482.3m +17.1% +19.0% US imports from BD 233.6m +1.1% +1.2% Implied mass of entrants 4,712 +5.8 +6.6 Price index in EU 100% -9.38% -10.41% Price index in US 100% -0.1% -0.11% Share of ROO rms 70.2% 77.7% 100% Tariff Revenue in EU 447k -34.2% -100% Tariff Revenue in US 46,728k +1.1% +1.2% Change in welfare EU ó +293,418k +327,162k Change in welfare US ó +6,191k +6,964k Exogenous quota price setting EU imports from BD 482.3m +22.7% +25.5% US imports from BD 233.6m +14.3% +16.1% Change in welfare EU ó 391,918k 441,279k Change in welfare US ó 82,650k 92, 933k
  • 33. Short-Run Equilibrium Implications of Policy Changes Baseline No preferences Higher doc. costs No home yarn req. Tariff EU: ROO / NO 0% / 12% 12% / 12% 0% / 12% 0% or 12% Tariff in US 20% 20% 20% 20% Cost disadvantage (a) 0.85 1.00 0.85 1.00 Doc.costs (d/f ) 0.66 0.00 1.32 0.66 Change in mass of rms, % Mass of exporters 485 0.00% 0.00% -0.21% Change in cutoffs % Product.cutoff, EU 0.8508 0.00 0.00% 0.00 Shock cutoff, EU 0.1866 +0.37% 0.00% +0.37% Change in BD revenues before & after tariff RBD,EU 482.3m +0.97% +0.01% +4.64% (1  tBD,EU )RBD,EU 481.8m -11.06% -0.11% +4.68% Approximated change in welfare ($) Price index in EU 100% -1.63% -1.09% -3.67% Tariff revenues in EU 447k +12,964% +130% -43% Change in welfare, EU ó +107,433k +33,712k +111.610k
  • 34. Short Run Effects Turning off entry channel changes effects Removing preferences, increasing documentation costs, and removing Home Yarn requirements raise EU welfare * Removing preferences: TR rises * Increase in documentation costs:TR grows by 130% * No yarn requirement: TR falls as more rms meet ROOs * Removing preferences: Price Index in EU falls * Increase in documentation costs:Price Index in EU falls * No yarn requirement: Price Index in EU falls US is unaffected
  • 35. Long-Run vs Short-Run Effects Lead to opposite welfare conclusions! * Preference removal: LR (-) vs. SR (+) * Higher documentation costs: LR(-) vs. SR(+) * No Home-yarn requirement: LR(+) vs. SR(+) Difference between LR and SR depends on parameter values Fixed entry calculations might be quite misleading!
  • 36. Fixed Costs Compensation Efciency Baseline Ind. Entry EU entry US entry Docum. Fixed Costs compensated: ó fe EU fm fmUS d f Original (estimated) ó 77,348 251,250 67,869 4,240 6,404 ó Endogenous quota price case Compensation amnt. ó 318 1,826 2,328 3,192 2,117 Market share in EU 482.3m +0.11% +1.68% +1.30% +1.37% +6.54% Market share in US 233.6m +0.04% +0.08% +5.78% +0.06% +3.19% Mass of entrants 4712 +0.22% 0.47% 2.62% 0.39% 12.34% Tariff Revenue in EU 447k +0.12% +2.14% +1.49% -93.1% +86.3% Tariff Revenue in US 46,728k +0.04% +0.08% +5.78% +0.06% +3.19% Change in welfare EU ó 1.9m 28.5m 22.1m 22.7m 111.7m Change in welfare US ó 0.2m 0.5m 33.4m 0.4m 18.4m Policy efciency ó 0.4 5.5 11.4 4.8 24.8 Exogenous quota price case Compensation amnt. 317 1,820 2,001 3,185 1,912 Market share in EU 482.3m +0.28% +2.07% +8.59% +1.76% +14.69% Market share in US 233.6m +0.46% +1.04% +23.6% +0.95% +27.75% Change in welfare EU ó 4.8m 35.2m 146.6m 29.4m 252.6m Change in welfare US ó 2.6m 6.0m 136.8m 5.5m 159.8m Policy efciency ó 1.5 8.3 57.1 7.1 81.2
  • 37. Fixed Costs Compensation Subsidies to xed costs Total 1.5 million dollar (2%) subsidy raises exports by over 0.4-24.8 times * Implemented by giving dollar subsidy per rm and changing till spend 1.5 million * Want to choose which xed costs to subsidize to increase exports Rule of thumb * Compensate late in the entry process * Compensate in the markets with the lowest initial market shares Win-win-win scenario possible: welfare at EU and US rises, BD exports rise
  • 38. Subsidies to xed costs: What is behind large entry impact? Ex-ante prots are very at as mass of entry rises so large entry effects * Low substitution between BD rms means new entrants make room for themselves * Lower BD price index means BD rms steal from ROW rms: small country assumption * This channel does less if substitutability in BD and ROW is reduced Decomposition of policy experiment outcomes into extensive (via margins and via entry) & intensive margins. * Entry part of extensive margin does most of the work. Role of demand shock modelling assumption * Marginal TFP rms with marginal demand shock produces f * Other marginal TFP rms produce more than f on average making marginal TFP rms more important economically Quotas reduce impact in US and in EU: license price will rise, muting rewards to rms => less entry
  • 39. Relation to Krugman (1980) and Chaney (2008) Krugman (1980): homogeneous rms + low s ) tariff wont reduce imports much as goods are poor substitutes Chaney (2008): heterogeneous rms + low s ) tariff reduces imports a lot as marginal rm has little disadvantage from high cost so sells a lot even if its prots are low. Hence, large effect of tariff on trade ows. No free entry in Chaney (2008)! In our paper most of action comes from entry margin.
  • 40. Policy Importance Trade facilitation vs direct aid as aid/development tool Conversely, devastating impact of poor infrastructure, rule of law, corruption,... Such aid may also be in donors narrow interest Approach can be used to evaluate policy interventions

×