[R-G] Contradiction of American Hegemony

[Yoshie Furuhashi]

Underlying the conditions that led to the current liquidity problem
are two trends: American workers', and America's, increasing inability
to make ends meet, requiring bigger and bigger debts; and global
financial liberalization that has led the states worldwide to
accumulate larger and larger foreign exchange reserves,* the trend
particularly aggravated by the ways in which Japan, China, and
petrodollar exporters are financially integrated into US economy.

These two trends, which stem from the contradiction of American
hegemony -- American economy has become increasingly hollow, but the
Washington consensus has become global, which helped create the
so-called "global savings glut," helped lower interest rates, and
helped inflate asset bubbles here and elsewhere -- demands a solution
that squarely addresses the contradiction of American hegemony.

* <http://www.aeaweb.org/annual_mtg_papers/2006/0108_1015_1101.pdf>
The Social Cost of Foreign Exchange Reserves
Dani Rodrik
Harvard University
December 2005

. . . . . . . . . . . . . . . . . . . .

II. The Rapid Rise in Reserves

Figure 1 shows the massive increase in developing countries' foreign
exchange reserves in recent years. Reserves have risen from a range of
6-8 percent of GDP during the 1970s and 1980s to almost 30 percent of
GDP by 2004. Reserves begin to trend sharply up just around 1990, the
year that is commonly identified with the onset of the era of
financial globalization. Note that there is no similar jump in the
reserves held by industrial countries, which have remained roughly
steady at below 5 percent of GDP since the 1950s. As the figure shows,
the trend for developing countries looks identical regardless of
whether China is included in the sample or not. In other words, the
increase in recent years cannot be attributed to China's efforts to
prevent the appreciation of the yuan.

Prior to the era of financial globalization, countries held reserves
mainly to manage foreign exchange demand and supply arising from
current account transactions. The traditional rule of thumb for
Central Banks was that they should hold a quantity of foreign exchange
reserves equivalent to three months of imports. Therefore at least
part of the increase in reserves may be due to the increased
commercial openness of developing countries. But as Figure 2 shows,
the increase in reserves is equally evident when looked at in relation
to imports. Prior to 1990, developing country reserves fluctuated
between 3 and 4 months of imports. They now stand at a record high of
8 months of imports. Once again, there has been no corresponding
increase in the industrial countries' reserves-imports ratio, which
still stands at less than 3 months.

It is pretty clear that the increase in developing country's reserves
is related to changes not in real quantities (such as imports or
output) but in financial magnitudes. Financial liberalization has led
to an explosion in financial assets and liabilities since the 1980s,
with which reserves have barely kept pace. For example, Figure 3 shows
the ratio of reserves to M2 in a sample of emerging market economies.
The figure reveals that the increase in Central Bank reserves starting
around 1990 has served simply to restore the reserves-M2 ratio to the
levels that prevailed in the pre-liberalization period. Moreover, this
ratio has remained more or less flat since the early 1990s, indicating
that reserves are barely keeping pace with the expansion of bank
liabilities in these countries. It seems clear therefore that
developing countries began to accumulate reserves as a consequence of
financial liberalization and globalization, and that they actually
embarked on this path before it became part of the conventional policy
wisdom.

The policy guidance that the IMF provides to emerging nations on
reserves was summarized by Stanley Fischer in 2001 in the following
manner:

     An IMF staff study discussed by our Executive Board
     last year agreed that holding reserves equal to short-term
     debt was an appropriate starting point for a country with
     significant but uncertain access to capital markets. But it is
     only a starting point. Countries may need to hold reserves
     well in excess of this level, depending on a variety of factors:
     macro-economic fundamentals; the exchange rate regime;
     the quality of private risk management and financial sector
     supervision; and the size and currency composition of the
     external debt.
          This analysis is now reflected in the way we treat reserve
     adequacy in our lending and surveillance activities. (Fischer
     2001)

The rule that countries should hold liquid reserves equal to their
foreign liabilities coming due within a year is also known as the
Guidotti-Greenspan rule, after a principle enunciated by Pablo
Guidotti (then deputy finance minister of Argentina) and subsequently
endorsed by Fed Chairman Alan Greenspan (see Greenspan 1999). As
Figure 4 shows, most emerging market economies had short-term
debt/reserves ratios that were significantly above unity in the early
1990s. Since then, practically all of them have built up enough
reserves to abide by the Guidotti-Greenspan-IMF rule, most with some
room to spare. The only exception in 2004 was Argentina, a country
that was just coming out of a severe financial crash.

Finally, Figure 5 shows a geographical breakdown of reserve trends.
The surprise here is that the increase in reserves has not been
restricted to "emerging markets." In fact, the increase in Africa's
reserves is as striking as that of Asia. By 2004, Africa held reserves
worth around 8 months of imports, compared to 6 months in the Western
hemisphere and close to 10 months in Asia. So the reserve buildup is a
phenomenon that affects the world's poorest countries as well.

III. Calculating the Cost of Reserve Holdings

Consider a country that lives by the Guidotti-Greenspan-IMF rule.
Suppose a domestic private firm or bank takes a short-term loan from
abroad of $1 million. The Central Bank now has to increase its
reserves by an equivalent amount. The usual strategy that the Central
Bank will follow is (a) to purchase foreign currency in domestic
financial markets to invest in U.S. government or other foreign
short-term securities and (b) to sterilize the effects of its
intervention on the money supply by selling domestic government bonds
to the private sector. When all these transactions are completed, the
domestic private sector ends up holding $1 million of domestic
government bonds balancing its foreign liability of $1 million, while
the Central Bank has $1 million more in foreign assets and $1 million
less in domestic government bonds.

Three consequences are noteworthy. First, the application of the
Guidotti-Greenspan-IMF rule implies that, even when the process is
initiated by borrowing from abroad, the home economy ends up with no
net resource transfer from abroad. The increase in the private
sector's foreign liability matches the increase in the Central Bank's
foreign assets. Second, short-term borrowing abroad does not enhance
the private sector's overall capacity to invest. This is because the
private sector ends up holding additional government securities equal
in magnitude to its borrowing abroad. And third, aggregating the
domestic private and public balance sheets, the net effect is that the
economy has borrowed short term abroad (at the domestic private
sector's cost of foreign borrowing) and has invested the proceeds in
short-term foreign assets.

. . . . . . . . . . . . . . . . . . . .

Yet the striking fact is that short-term debt exposure has continued
to climb in many countries, even as these same countries were
investing valuable resources in increasing reserve assets. As Figure 7
shows, half of the emerging market economies had higher short-term
debt-GDP ratios in 2004 than they did in 1990. In contrast, none held
lower reserves in relation to GDP. Looking at the group of emerging
market economies in aggregate, the average short-term debt-GDP ratio
has risen from 5.4 (6.5) percent to 6.1 (8.4) percent in weighted
(unweighted) terms between 1990 and 2004, while the reserves-GDP and
reserves-short-term debt ratios have increased by a multiple (Table
1). The minimum that can be said is that there has not been a clear
downward trend in short-term debt exposure, a fact that looks all the
more astonishing when we put it together with the massive boost in
reserves.

. . . . . . . . . . . . . . . . . . . .

Certainly gross fixed capital formation has not been visibly affected
by the vast pool of short-term flows moving into emerging market
economies (Figure 8). In the apt words of Joshua Aizenman (2005, 959),
"the 1990s' experience with financial liberalization suggests that the
gains from external financing are overrated."

. . . . . . . . . . . . . . . . . . . .

V. Concluding Remarks

We are left with the inescapable conclusion that developing countries
on the whole have responded to financial globalization in a highly
unbalanced and far-from optimal manner. They have over-invested in the
costly strategy of reserve accumulation and under-invested in capital
account management policies to reduce their short-term foreign
liabilities.

The reasons are perhaps not hard to fathom. Unlike reserve
accumulation, controls on short-term borrowing hurt powerful financial
interests, both at home and abroad. Somehow "market intervention" in
the form of taxing short-term capital inflows has developed an
unsavory reputation that "market intervention" in the form of buying
reserves does not have. Perhaps it is time to start viewing the
Guidotti-Greenspan-IMF rule as an admonishment that applies to short
term foreign borrowing, and not just reserves.
--
Yoshie