Transcript trade - Harvard Kennedy School

LECTURE 2: THE TRADE BALANCE
IN PRACTICE
• Question: Is the Marshall-Lerner
condition satisfied in practice?
1) Historical examples
• Italy 1992-93
• Poland 2009
2) Econometric estimation of elasticities
• OLS
• The J-curve
3) Both determinants together: Real exchange rate & income
• Keynesian model of the TB
• Estimation for the case of East Asian countries
Historical examples
(i) Italy devalued in Europe’s 1992 ERM crisis.
The lira’s Real Effective Exchange Rate value & effect on its trade balance.
1992 devaluation
Rise in trade balance
Professor Jeffrey Frankel, Kennedy School,
Harvard University
(ii) Poland’s Exchange Rate Rose 35%
when Global Financial Crisis hit in late 2008.
Zloty/€
4.7
4.5
4.2
4.0
3.7
3.5
3.2
I
III
V
VII
2008
IX
XI
I
III
V
VII
IX
2009
Source: Cezary Wójcik
XI
I
III
V
2010
VII
IX
Poland’s trade balance improved sharply in 2009
while its European trading partners all went into recession.
Trade balance in billions of euros
Source: National Bank of Poland
From FocusEconomics 2014
Contribution of Net X in 2009:
3.1% of GDP > Total GDP growth: 1.7%
=> Poland avoided recession.
A textbook case where depreciation was expansionary:
Poland, the only continental EU member with a floating rate,
was also the only one to escape negative growth
in the global recession of 2009.
% change in GDP
Poland
Lithuania
Latvia
Estonia
Slovakia
Czech Republic
2006
2007
2008
2009
2010
6.2 6.8 5.1 1.7 3.5f 7.8 9.8 2.9 -14.7 -0.6f 12.2 10.0 -4.2 -18.0 -3.5f 10.6 6.9 -5.1 -13.9 0.9f 8.5 10.6 6.2 -4.7 2.7f Source: Cezary Wójcik, 2010
6.8 6.1 2.5 -4.1 1.6f Exchange Rate
Floating
Fixed
Fixed
Fixed
Euro
Flexible
(de facto)
Empirical estimation of export & import elasticities
log of X
demanded
log of EP*/P ≡ Price of foreign goods relative to domestic goods
• Coefficient estimated by OLS regression.
– In logs, so parameters are elasticities.
Common econometric finding
• Estimated trade elasticities with respect to relative prices
often ≈ 1, after a few years have been allowed to pass.
– => Marshall-Lerner condition holds in the medium run.
– e.g., Marquez (2002).
• Some face a higher elasticity of demand for their exports:
– small countries, and
– producers of agricultural & mineral commodities or other
commodities that are close substitutes for competitors’ exports.
Common empirical
observation:
After a devaluation,
trade balance gets worse
before it gets better.
Explanation:
Even if devaluation is
instantly passed through
to higher import prices,
buyers react with a lag.
Also, in practice, it may
take time up front
before the devaluation
is passed through
to import prices.
The trade balance is a function of both the
real exchange rate and income.
• Recall the Keynesian model of the trade balance
from Lecture (iii)
of the pre-semester
Macro Review.
• Micro theory:
The demand for the import or export good,
as for any good, is a function of both price &
income.
Keynesian Model of the Trade Balance
Import demand is a function of the exchange rate & income.
The same for exports: => X = X(E,Y*) M = M(E,Y).
dX
0
dE
dX
 m*  0
dY *
dM
0
dE
dM
m0
dY
If the domestic country is small, Y* is exogenous.
.
Estimated income
elasticities
are mostly
between 1.0 - 2.0.
Estimated price
elasticities (LR)
satisfy the
Marshall-Lerner
Condition.
END OF
LECTURE 2:
THE TRADE
BALANCE IN
PRACTICE
Appendix 1 -- More
historical examples:
EM currency crises
of the 1990s.
After big devaluations
in Mexico
in 1994
and Korea
& Southeast Asia
in 1997,
trade balances “improved”
quickly, but because
of expenditure-reduction,
not expenditure-switching.
Appendix 2–
An application of the marginal
propensity to import
• Why did trade fall
so much more sharply than income
in the 2008-09 global recession?
Professor Jeffrey Frankel, Kennedy School, Harvard University
An application of the marginal propensity to import:
Why did trade fall so much more sharply than income
in the 2008-09 global recession?
2009
Bussière, Callegari, Ghironi, Sestieri, & N.Yamano, 2013,
"Estimating Trade Elasticities: Demand Composition and the Trade Collapse of 2008-2009."
Why did trade fall so sharply in the 2008-09 global recession?
Bussière, Callegari, Ghironi, Sestieri, & Yamano, 2013, "Estimating Trade Elasticities: Demand Composition and the Trade Collapse of 2008-09."
The usual explanations involve trade credit,
inventories, and trade in intermediate inputs.
Bussière et al (2013) argue that Investment, which declined
much more in 2009 than the other components of GDP, has
a higher marginal propensity to import than the other components.
Behavior of real components of GDP
in the 2008-09 recession
GDP
Demand, adjusted
for import-intensity
Imports
& Exports
Investment
Bussière, Callegari, Ghironi, Sestieri, & N.Yamano,
"Estimating Trade Elasticities: Demand Composition and the Trade Collapse of 2008-2009.“