LT Week 2 Flashcards
Trade and productivity
* Learning by exporting
Motivation
Exporting and Firm Performance: Evidence from a Randomised Experiment (Atkin et al., 2017)
To understand the mechanisms driving the profit increases due to learning-by-exporting.
Special settings
Exporting and Firm Performance: Evidence from a Randomised Experiment (Atkin et al., 2017)
“Large differences in productivity across countries. Belief that access to high-income markets help firms in developing countries close this gap is a motivator behind large resources now flowing to market access initiatives, based on principle that exporting improves productivity (learning-by-exporting)
Trade generates efficiency gains which narrow productivity gap and magnify gains from trains relative to models without learning
Ongoing debate whether exporting has causal impact on firm performance, unclear whether if due to learning-by-exporting or outward shifts/movements along PPF”
Theory
Exporting and Firm Performance: Evidence from a Randomised Experiment (Atkin et al., 2017)
“RCT on rug manufacturers in Egypt to examine how exporting affects profits and productivity; helps understand which direction causation is
Use intevention that reduced matching frictions between foreign buyers and a random subset of Egyptian firms, tracking changes in quality and product specifications as well
Uncover if and how much firm productivity responds to exporting”
Empirical design
Exporting and Firm Performance: Evidence from a Randomised Experiment (Atkin et al., 2017)
“Provide subset of firms with opportunity to export handmade carpets to high-income markets
Partner with NGO and Egyptian intermediary to secure export orders
Surveyed sample of several hundred small rug manufacturers in a city; random subsample provided with initial opportunity to fill orders by completing 110m2 of rugs (11 weeks of work); then provided with further orders if met satisfaction of buyer and intermediary
Tracked performance through perioidic surveys of treatment and control firms including quanity and quality
Randomisation process allows for causal effect of exporting to be identified through comparing mean outcomes between treatment and control firms”
Data
Exporting and Firm Performance: Evidence from a Randomised Experiment (Atkin et al., 2017)
“Collected baseline survey before treatment firms were provided with export opportunity, covering firm production, rug quality, household and demographic characteristics
Repeated survey twice more as well as shorter surveys on production and rug quality
Production module records production activity monthly and converts to real values using CPI, covering profits, revenues, expenses, output quantity, specifications, labour hours”
Key findings
Exporting and Firm Performance: Evidence from a Randomised Experiment (Atkin et al., 2017)
“Opportunity to export raises overall performance (profits) of treatment firms by 16-26%, which in part due to demand shock
Despite increases in output prices and labour hours, observe decline in total output among treatment firms –> suggest that high-income buyers demand higher-quality rugs which take longer to produce
Rugs produced by treatment firms were significantly better quality in every dimension
Unadjusted productivity down by 24-28% for treatmenty firms (not taking quality into account)
Increase in quality due to 2 mechanisms: 1) movement along PPF due to selling less rugs but at a higher price (and higher quality); 2) learning-by-exporting causes outward shift of PPF due to raising output per input or producing high-quality rugs, both of which improve efficiency and increase profits due to quality –> implies learning-by-exporting is more important as implies larger gains from trade; quality measured in 11 characterisrtics on scale 1-5”
Interpretation / policy implications
Exporting and Firm Performance: Evidence from a Randomised Experiment (Atkin et al., 2017)
Study represetnative for small firms only rather than large firms manufacturing complex products
Motivation
The Costs of Remoteness: Evidence from German Division and Reunification (Redding and Sturm, 2008)
To provide evidence for the causal importance of market access for economic development
Special settings
The Costs of Remoteness: Evidence from German Division and Reunification (Redding and Sturm, 2008)
Unification of West and East Germany in 1990 serves as a natural experiment for exogenous variation in market access. Border was based on military considerations, sealed and all local economic interactions across the border were ceased
Theory
The Costs of Remoteness: Evidence from German Division and Reunification (Redding and Sturm, 2008)
“Key idea is that West German cities close to border experienced disproportionate amount of market access relative to other cities due to losing nearby trading partners that enabled low transportation costs
Those further away had less reliance on trading partners lost and therefore faced higher transportation costs already
Compare development of West German cities close and far from border, before and after division of Germany”
Empirical design
The Costs of Remoteness: Evidence from German Division and Reunification (Redding and Sturm, 2008)
Data
The Costs of Remoteness: Evidence from German Division and Reunification (Redding and Sturm, 2008)
Panel data across 1919-2002, covering city populations, employment (total and by sector)
Key findings
The Costs of Remoteness: Evidence from German Division and Reunification (Redding and Sturm, 2008)
“Population of WG close to border declined at annualised rate of 0.75% relative to other cities, causing a cumulative reduction in relative size of a third (across 40 year period); emerges in immediate aftermath
Relative decline is more than twice as large for cities with below-median population as for those above-median
Could be due to other explanations: 1) cities close to border specialised in declining industries (coal, mining); 2) cities close to border might’ve faced differential levels of war-related disruption (refugees from EG and destruction); 3) increasing connection between WG and western Europe could’ve elevated population growth in western WG, contributing to relative decline; 4) vulnerable in any new armed conflict in western Europe –> all undermined by new model”
