Transcript presented at - Harvard University

Currency Wars
Jeffrey Frankel
Harpel Professor of Capital Formation & Growth,
Harvard University
American Enterprise Institute, March 18, 2013
1
Is “currency wars” just a catchy phrase?
It is another way of saying
“competitive devaluation,”
• a kind of beggar-thy-neighbor policy


to use language of the 1930s,
one motive at Bretton Woods
for fixed exchange rates;
• or “manipulating exchange rates…
to gain an unfair competitive advantage
over other members…”

to quote from IMF Article IV(1)iii.
2


The phrase has
been applied
to the Fed’s QE.
But it does not in truth fit well.
• A key point often missed:
even the direction of the effect of one
country’s monetary expansion on the
trade balance is ambiguous:
• the expenditure-increasing effect
when the expanding country expands
counteracts the expenditure-switching
effect when the exchange rate responds.
3
National authorities will & should pursue economic
policies in their own countries’ interests.
At times cooperation is fruitful,
• whether by norms,
• multilateral meetings like the G20,
• or formal institutions: WTO & IMF.
• But there is little point in trying to establish
international cooperation if the nature
of the spillover effects is not relatively clear
and agreed upon.


Pollution & tariffs, for example,
are negative externalities, not positive.
It is not clear, with monetary policy.
4




If US unemployment is high & inflation low,
the Fed will naturally choose
an easy monetary policy (low i).
If the macroeconomic situation
is the reverse in Brazil, its central bank
will naturally choose a tight monetary policy
(high i ).
Also naturally, capital will flow from the US
to Brazil and will in turn appreciate the Real.
But that is the beauty of floating rates.
5
Brazil’s inflation has been high
so a strong currency should help.
6



The exchange rate movement will shift
demand toward exporters in the US and away
from those in Brazil, other things equal.
But such “casualties of war” are not even
unexpected collateral damage;
they are the point of the two chosen
monetary policies:


If the goal is to stimulate demand for US goods
& cool off demand for Brazilian goods,
why shouldn’t the exporters in both countries
share in that process, via net foreign demand
alongside construction & the other sectors
that are sensitive to i via domestic demand?
7
More of a problem arises if one of the countries
had been targeting or fixing the exchange rate.
• It won’t necessarily want to abandon
a proven exchange rate regime
at the first sign of trouble.
• Capital controls & sterilization of reserve
flows might help delay the adjustment,
• but a persistent capital flow will eventually
force the country either to allow its exchange
rate to adjust or its money supply.
8
In the case of China, 2004-2011,this meant a
choice between allowing some appreciation of
the RMB and an increase in the money supply.

They did some of both


but more of the latter than the former:
the monetary inflow eventually turned
inflationary, as expected: in 2008 & 2011.
9

True, an impressively wide array
of countries prefer weaker currencies
to stronger ones, as a means to improve
their trade balances.
• True, too, not all can depreciate at once,

nor improve their TB at the same time.
• This does not mean that they are guilty
of violating agreements or norms,

especially if they have not devalued but merely
stuck with a pre-existing exchange rate regime:
• float, fixed, or band.
10
Uncoordinated monetary expansion does not even
necessarily leave the world in a worse equilibrium.
It might just give the world what it needs.

Eichengreen & Sachs

(1985, 86)
for the 1930s
argue the opposite of the conventional wisdom
• re beggar-thy-neighbor competitive devaluations.


To the extent that each country devalued against
gold, even though they could not all succeed in
improving their trade balance,
they raised the price of gold & thereby increased
the value of the global money supply,
which is what a world in Depression needed.
• The same may apply today

Eichengreen
(2013).
11
The accusation is especially misplaced
against a country like the US
• which was the target of the phrase’s
original coiner, Guido Mantega.
• The US authorities have not intervened in
the fx market nor talked down the $.
• $-depreciation is probably not at the top
of the list of the effects that the FOMC
has in mind when deciding QE.
E12
Yes, the usual channel of transmission, via
short-term interest rates, is all-but-ended
by the Zero Lower Bound.

But, aside from depreciation,
that still leaves effects via:
• longer term and private bonds,
• the credit channel,
• expectations regarding inflation,
• equity prices,
• And real estate prices.
13
Japan comes a little closer,

in that:
• Japan went back to buying $ in Sept. 2010;

and some members of the new Abe
government in 2013 were initially foolish
enough to mention ¥ depreciation explicitly.
14
China of course intervened heavily.
The RMB was undervalued
by most measures
from 2004 to 2009
(less so, by 2012).
• But countries have a right to opt for a fixed
exchange rate regime.
• China was continuing a regime
that had previously been in place,

which does not sound like “manipulation.”
15


Appreciation was probably
in China’s interest.
And it was reasonable
for others to propose that China
allow appreciation as part of a deal
voluntarily agreed among sovereign states
• in exchange, for example, for the US fiscal reform.

But this is different from charging Beijing
with having violated international norms
or rules and from threatening retaliation.
16
Also, China has adjusted much of the way, since 2009.
Appreciation versus the US $, 2005-12
1.5
1.4
CNY/USD,
2005M06=1
1.3
real
1.2
nominal
1.1
1.0
0.9
2005
2006
2007
2008
2009
2010
2011
Indeed, very few of the countries accused of
participating in the currency wars have undertaken
discrete devaluations in recent years or acted to
weaken their currencies by switching exchange rate
regimes. These are the sort of deliberate policy
changes connoted by a phrase like “manipulation”


Switzerland perhaps comes the closest.
But the Swiss franc was so “overvalued”,
even at the new rate set Sept. 2011, that the
SNB cannot really be accused
of unfair undervaluation.
18
Conclusion

The accusations fly in both directions.
• The US accuses China of unfairly
manipulating its currency.
• EM countries retaliate by accusing
the advanced countries of pursuing
a currency war.

My view: In both directions, the charges
of wrong-doing are not merited.
19
20
http://ksghome.harvard.edu/~jfrankel/index.htm
Appendix I: Origin of
“Currency Wars,” in 2010

Warning from Brazil’s Finance
Minister Guido Mantega (9/27/2010):
“We’re in the midst of
an international currency war,
a general weakening of currency. This threatens us
because it takes away our competitiveness.”

I.e., countries everywhere are trying
to push down the value of their currencies,
to gain exports & employment,
• a goal that is not globally consistent.
21
Singapore took the renewed inflows
heavily in the form of reserves,
while India & Malaysia in 2010 took them
in the form of currency appreciation.
↑ moremanaged
floating
High EMP
less-managed floating
Low EMP
(“more appreciation-friendly”)
→
22
GS Global ECS Research
In Latin America, renewed inflows
were reflected mostly as reserve accumulation in Peru,
but as appreciation in Chile & Colombia.
more-managed floating
less-managed floating
(“more appreciation-friendly”)
23
GS Global ECS Research

Renewed flows to EMs in 2010-2013
were met with $ purchases in FX intervention
•
•
•
•
•
Brazil,
Korea,
Thailand,
Peru …
By 2011, even Chile, the cleanest of the floaters,
was intervening to dampen appreciation.
24
Currency Wars,

China had long been intervening
to prevent the RMB from appreciating.


continued
The U.S. tried to enlist help from countries like Brazil
in pressuring China to abandon its undervaluation.
The Brazilian Minister’s response:
The $ is as much a part of the problem as the RMB.
• He was referring to QE2 that fall, which like any other
monetary easing could be expected to depreciate the $.

More attacks on Fed action -• China & Germany: $ depreciation is a deliberate salvo in currency wars
• Sarah Palin, Rick Perry & John Taylor: QE2 “debauches the currency.”
25
Appendix II: China
26
1
2005
-1
-.5
0
.5
The Balassa-Samuelson Relationship
-3
-2
-1
0
1
Log of Real Per capita GDP (PPP)
2
coef = .23367193, (robust) se = .01978263, t = 11.81
Source: Arvind Subramanian, April 2010,
“New PPP-Based Estimates of Renminbi Undervaluationand Policy Implications,” PB10-08, Peterson Institute for International Economics
Undervaluation of RMB in the regression estimated above = 26%.
Estimated undervaluation averaging across four such estimates = 31%.
Compare to estimate for 2000
= 36%.
As recently as 2009
25% .
(Frankel 2005)
(Chang 2012) :
27
China Adjusts, 2009-12



Various key measures suggest that China
has achieved a substantial share of the
needed trade adjustment since 2009:
Its trade surplus peaked at $300 billion in
2008, and has been declining since then.
Substantial real appreciation of the RMB has
brought it closer to equilibrium.
• Some nominal appreciation +
• Some inflation &, especially, wage increases
China’s external balance has adjusted.


China’s trade surplus is much smaller now.
Its overall balance of payments, too,
was negative some months in 2012
Adjustment of relative prices

The famous “China price”:
• Ever since China rejoined the world economy
3 decades ago, its trading partners have
snapped up exports of manufacturing goods,
• because low Chinese wages made them
super-competitive on world markets.

But in recent years, relative prices have
partly adjusted
• following the laws of market economics.
Adjustment of relative prices, continued

The change in relative prices is reflected
as real exchange rate appreciation.
• This comprises, in part, nominal appreciation
• and, in part, Chinese inflation.
• The government would have been better advised
to let more of the real appreciation take the form
of nominal appreciation ($ per RMB).
• But it didn’t, so it showed up as inflation instead.
• See appreciation in chart below.
The natural adjustment process was delayed.


1st, because the authorities intervened to
keep the exchange virtually fixed against
the dollar, in the years 1995-2005 and
2008-2010.
2nd, workers in China’s increasingly
productive coastal factories were not paid
their full value.
• The economy has not completed its transition
from Mao to market, after all.

As a result of these two delaying
mechanisms, Chinese continued to
undersell the world.
But then two things happened.

1st, the yuan was finally allowed to appreciate
against the $ during 2005-08 & 2010-11,
by 25% cumulatively


=17% + 8%.
2nd, and more importantly,
labor shortages began to appear =>
China’s workers at last began to win rapid wage hikes.
• Major cities raised their minimum wages
sharply over each of the last 3 years:


22% on average in 2010 & 2011;
somewhat less in 2012, in response to slowing demand:
• 8.6 % in Beijing, 13% in Shenzhen & Shanghai.

Meanwhile another cost of business,
land prices, rose even more rapidly.
The RMB rose against the $ during 2006-08,
returned to peg in mid-2008,
and then appreciated again in 2010-2011.
34
Overheating also shows up in rapid rise of land prices
Real Beijing land prices
35
Real appreciation


The RMB’s real appreciation against the $
from 2009 to 2012 has amounted to 12%,
reducing the degree of undervaluation by roughly half,


depending on whether one measures it against the $
or against all currencies.
In any case, China’s real exchange rate is already
closer to this measure of equilibrium than are most
countries’ exchange rates (Cheung, Chinn & Fuji, 2010).
5 types of adjustment are gradually taking place
in response to the new high level of costs
in the factories of China’s coastal provinces:

1st, some manufacturing is migrating inland,
• where wages & land prices are still relatively low.



2nd, export operations are shifting to Vietnam or Bangla Desh
• where wages are lower still.
3rd, Chinese companies are beginning to automate,
• substituting capital for labor.
4th, they are moving into more sophisticated products,
• following the path blazed by Japan, Korea, & other Asian tigers


in the “flying geese” formation.
5th, multinational companies that had in the past
moved some stages of their production process to
China, out of the US, or out of other high-wage
countries, are now moving back.


All five of these ways of reallocating
resources represent the economic
process operating as it should.
None of this comes as news
to most observers of China.

But many Western politicians are unable
to let go of the syllogism that seemed
so unassailable just a decade ago:
•
•
•
•

(1) The Chinese have joined the world economy;
(2) their wages are $0.50 an hour;
(3) there are a billion of them, and so
(4) their exports will rise without limit:
“Chinese wages will never be bid up in line
with the usual textbook laws of economics
because the supply labor is infinitely elastic.”
But it turns out that the laws of economics do
eventually apply after all -- even in China.
Expansion of the services sector.
This 6th dimension of adjustment still lags behind,

despite the consensus in favor of it.

China has had great success in manufacturing
• especially via exports.


Now it needs to help the other side of the economy
catch up: services, via domestic demand
• Housing, retail, education, environmental quality,
• health care, pensions, social safety net.
Some of this could be done via government spending
• as China did in 2009; but that was mostly heavy investment.