by Marc Chandler
27 December 2014
from
EconomyWatch Website
At
best, it is the holiday season when hopes and
wishes abound.
It may be
as it was for Ebenezer, "an undigested bit of
beef, a blot of mustard, a crumb of cheese, a
fragment of undone potato."
At worst,
it is spiked eggnog and the peppermint schnapps.
Implement What's Already There
How else could reasonable people, like former ECB President
Jean-Claude Trichet, say that there is a "window of
opportunity" for a new agreement on foreign exchange among the major
high-income countries?
Other economists go further and
fancifully include China too in a new global accord.
Trichet talks of a conceptual convergence among the countries whose
currencies make up the IMF's
Special Drawing Right, the,
He notes that these central banks have
all adopted inflation targets around 2%, adopted similar rules aimed
at limiting financial risk, conduct surveys of member banks, and use
similar communication tools, like press conferences.
Do not confuse the adoption of best practices, an evolving process,
with coordination. It cannot conceal that, when shocked (The
Great Financial Crisis), there were different policy
responses. Those policy responses have produced different economic
outcomes.
The dramatic moves in the foreign
exchange market among the major currencies reflect this economic
divergence.
Even where Trichet sees a common understanding, like the near 2%
inflation target, there are significant differences. So much that on
the eve of the financial crisis, during Trichet's presidency, the
ECB hiked rates in July 2008, due largely to the inflation
implications of $150 per barrel oil.
Trichet also led the ECB to hike rates
twice in H1 2011. These were quickly unwound when Mario Draghi
replaced Trichet later that year. During both periods,
the Federal Reserve was easing
policy.
Note too that the ECB defines its mandate of price stability by
targeting near-but-lower-than 2% CPI.
The headline rate is noisy, subject to
perturbations that create volatility, but are of little durable
economic consequence. The Bank of Japan targets core inflation. Its
definition of core excludes fresh food, but not energy.
The Federal Reserve defines its mandate
of price stability as 2% increase in the core personal consumption
expenditure deflator. Its definition of core excludes food and
energy.
Starting in 2015, the ECB will adopt a rotating voting system for
meetings every six weeks, and for the release of a report on the
deliberations. This is strikingly similar to the Federal Reserve.
However, it is not the basis of coordination.
The problems faced by the high-income countries do not lend
themselves to enhanced coordination. The international arena does
not have the remedy for the economic downturn sparked by the hike in
Japan's retail sales tax in April. Europe's problems are also
largely self-induced. Their rules and ideology (ordo-liberalism)
pose high hurdles to stronger aggregate demand.
The U.S. has just strung together its
strongest six-month economic performance in over a decade.
Not only is there no basis for new coordination, but it is not clear
that there is a material need. The thinking at the highest levels
has evolved over the nearly thirty years since
the Plaza Agreement (September
1985).
Because currency markets are so
vulnerable to political forces that can lead to ruinous
beggar-thy-neighbor competitive currency devaluations, policymakers
at the highest levels have agreed to forswear such use.
The "strong dollar policy", first articulated by Robert Rubin
as Treasury Secretary nearly twenty years ago, needs understanding
in this context.
It is the economic equivalent of arms
control. There are certain weapons that most nations have agreed not
to use first. Similarly, there are certain economic practices, such
as targeting foreign exchange prices, to which the major
industrialized countries have agreed to refrain.
There have been a few exceptions over the years. There was
coordination in 2000 to support the euro. In 2011, there was
coordinated attempt to push the yen lower. These are the exceptions.
Sometimes countries flirt with the threshold.
During Shinzō Abe's campaign as
Prime Minister of Japan, some senior advisers suggested targets for
dollar-yen (and the Nikkei). The push back resulted in a fresh
commitment to the financial arms control agreement in early 2012.
The ECB's Draghi confirmed taking this
threshold seriously, and recently asked whether the ECB would
consider buying foreign bonds. I have suggested that the ECB could
take a page from the Swiss National Bank's playbook and buy foreign
bonds, namely U.S. Treasuries that would be one form of QE that
would remain true to the letter of its mandate and still expand its
balance sheet.
Draghi dismissed the possibility by
saying that it was too close to intervention.
This is why we need to be exceptionally careful when some claim a
country is engaged in a currency war. When the Brazilian finance
minister claimed the U.S. was engaged in a currency war by pursuing
unorthodox monetary policy, the U.S. defended its right to pursue an
appropriate policy given its domestic economy.
Some claim that the Eurozone and Japan are engaging in a currency
war.
If they are, it begs a certain policy
response. However, the ECB and BOJ are pursuing unorthodox monetary
policies because their economies are stagnant or worse, and they are
fighting deflationary forces.
Even with the rapid expansion of the
BOJ's balance sheet, the pace of core CPI has been slipping for
several months and now, when adjusted for the sales tax hike, stands
at 0.7%.
Given the relative cyclical performance, a decline in the euro and
yen is part of the adjustment process. Even though those currencies
are already undervalued, according to the OECD's measure of
purchasing power parity, further decline is both expected and
necessary.
We note that even assuming an unbiased
investor, the premium the U.S. pays over Germany for borrowing (both
2-year and 10-year money) is the most in seven years.
Rather than be seduced by the siren calls of currency wars, U.S.
Treasury Secretary Jack Lew noted that a stronger dollar was
a symptom of the strengthening of the U.S. economy. Yes, some U.S.
corporations will attribute disappointing earnings to the strength
of the dollar, but this is often difficult to verify, and two
companies in the same industry frequently report different currency
impacts.
The real issue is that analysts and the
media often shy away from pressing companies on their hedging
program.
In any event, the U.S. economy has gotten stronger, exports are at
record highs, U.S. employment has accelerated even as the dollar has
risen. Yes, someday, it is conceivable that an appreciating dollar
bites harder, but that day is not today or tomorrow.
Moreover, it is important to keep the
dollar's strength in perspective.
On a broad, trade-weighted measure
adjusted for inflation, it is only now reaching the average over the
last ten years. It remains more than 5% below its post-crisis high.
The case for a new global accord, some economists say, is to
safeguard the U.S. expansion.
If that is truly the goal, it is not
clear how a global accord makes it more likely.
-
What is the U.S. to do?
-
Is it practical for the U.S. to
lean against a strong dollar by providing more monetary and
fiscal stimulus?
-
Is that really the basis for a
quid pro quo with Germany to reduce its current account
surplus?
-
Is a new global accord necessary
to prompt France or Italy into implementing the kind of
structural reforms that they have long promised?
-
Does that really provide a basis
for China to reduce its household savings, boost
consumption, or open up its capital account?
The G7 and G20 have called for
international imbalance reductions and to move toward
market-determined exchange rates where they do not exist. This is
the framework of the existing global accord. There is no need for a
new one.
Fuller implementation of the existing
one is preferable.
|