much fiscal stimulus. The problem was not the effort to expand
4.3 percent of its GDP in 1986 to 1.2 percent in 1990. America's
global current account deficit correspondingly fell from 3.7
percent of our GDP in 1987 to under 1 percent in 1991 (even
excluding our receipts from allies to help finance the Persian
Gulf war) .
Japan's global surplus has now soared because these two key
variables were again permitted to get out of line. Japan burst
the "bubble economy" with tight monetary policy and has thus
suffered from slow growth in domestic demand since 1991 — turning
it back to reliance on export-led growth. That was in turn
possible because the yen weakened by 25-30 percent in 1989-90
despite continued steady improvement in Japan's underlying
competitive position . In textbook fashion — and as my colleagues
at the Institute for International Economics and I predicted at
the time — Japan's surplus began to soar two years later. Most of
the blame for the inappropriate weakening of the yen,
incidentally, lies not with the Japanese but with (1) the markets
for ignoring the underlying fundamentals of the Japanese economy
elseif (getClientWidth() > 430)
and (2) the US Treasury and the G-7, which went to sleep at the
switch and let it happen.
It is clear how Japan can correct its global surplus:
restore domestic demand growth to at least 4 percent annually
through expansionary fiscal policy, taking advantage of the fact
that its budget surplus remains at 3 percent of its GDP, and work
with its G-7 partners to make sure that the yen remains at the
much stronger level to which it has risen over the past six
months.^ If sustained around current levels, the recent yen
a ppreciation guarantees a sharp fall in Japan's surplus by 1995-
96 — bringing it within the Administration's goal of 1-2 percent
of Japan's GDP. Hence the Administration's major initiative in
this area should be an effort to achieve Japanese for, better. G-
7) agreement to a system of target zones (or its weaker variant
of reference ranges as adopted by the G-7 in the Louvre Accord in
1987 and maintained for a time thereafter) that would prevent
renewed weakening of the yen .
Trade Policy Initiatives
Macroeconomic measures are not enough, however. Our new
book analyzes in some depth a dozen sectors that range across the
entire Japanese economy: in agriculture, high-tech
manufacturing, medium tech manufacturing, financial services, and
some other services. The picture that emerges is of a very
exclusionary, restrictive economy that pervasively combines
exclusionary corporate practices (including but ranging well
beyond the famous keiretsu) with governmental intervention and
regulation (some of which has now disappeared but leaves a legacy
that makes penetration by outsiders — outsider Japanese as well as
foreigners — extremely difficult.
Some of these problems are structural in the sense that they
cut across a number of sectors. Japan is clearly a "different"
domestic demand but the policy mix which was pursued, notably the
stubborn opposition of Japan's Ministry of Finance to expansionary
fiscal policy even though the country is running a large budget
It is comforting from an American standpoint to note that
the yen (and, to a minor degree, the Swiss franc) are the only
currencies against which the dollar has declined over the past six
or twelve months. The trade-weighted average of the dollar has
risen over both periods.
economy in many respects. However, Japan is also changing very
rapidly and convergence between our two economies — some in the
direction of Japanese practices, some in the direction of our
practices, some toward intermediate outcomes — is already
occurring. Policy changes can accelerate this process and we
recommend a new series of Structural Convergence Talks (SCTl that
would be directed to the objective.
In addition, however, it will be necessary to address a
number of the specific barriers on a sector-by-sector basis. ^
The key here is for the Administration to make a careful
assessment of the potential payoff from among the wide range of
potential candidates; our illustrative analysis (Table 5.5)
suggests that a meaningful deal on mainframe computers, for
example, would be worth 35 times as much as a deal on
supercomputers. Past negotiating agendas have been determined too
frequently instead via the "squeaky wheel" principle, i.e., the
industries that complained the most got the most attention
whether or not the likely benefits to the economy were
Having selected which sectors to pursue, the second key
issue is how to pursue them. The preferred course is always to
seek elimination of a specific government policy that limits
imports, such as public procurement in the cases of construction
contracting or supercomputers. If the problem is exclusionary
practices among Japanese firms, the proper remedy is antitrust
When neither of the preferred options is available and it
can be demonstrated persuasively that access to a Japanese market
is artificially hindered, it may be necessary to seek Japanese
agreement to temporary use of quantitative measures of increased
penetration — to manage the trade in a way that will unmanage the
market. I have no philosophical problem with such efforts: they
expand trade, expand the number of participants in the targeted
market, increase competition, reduce prices and reduce cartel
activity. They bear no resemblance to traditional protection —
which closes markets, reduces the availability of the product,
raises prices and creates or shores up cartels. There would
certainly be nothing new in "managed trade" (of either variety) :
the Reagan and Bush Administrations forced Japan to accept
"voluntary export restraints" (VERs) on its sales to the United
States of automobiles, steel and machine tools, and quantitative
import targets on semiconductors, autos and auto parts.
However, there are major practical problems with
quantitative market target industries ("involuntary import
expansion" arrangements, or IIEs) . First, any numerical goal
(like the 20 percent market share for semiconductors) is bound to
be arbitrary. Moreover, our negotiators may well shoot too low —
the analysis of several sectors in our book shows that the actual
impact of market liberalization was far higher than anyone
anticipated at that time. Second, no government can control the
outcome — and hence the "success" or "failure" of the agreement,
including the implications for retaliation, are also arbitrary.
Third, a policy that relies on such devices is inherently subject
to political capture, i.e., the "squeaky wheel" syndrome, in both
the exporting and importing countries. Fourth, precedents are
set and the initiator of the idea could at some future point
easily become its target.
'' Unlike the situation in the early and middle 1980s, when
the United States persuaded Japan to restrict its exports of
automobiles and steel to the United States, there are now no
significant US barriers to Japanese exports.
Having said all that, we endorse "IIEs" if no other policy
tool is available and the sector is sufficiently important. But
they should be used sparingly. Their deployment should always be
temporary. They should be used to expand trade (IIEs) rather
than to reduce trade (VERs) . They should apply to imports from
all countries rather than just the United States. Every effort
should be made to find superior alternatives in each individual
In its current negotiations with Japan, the Administration
has clearly stated that the "quantitative indicators" that it has
in mind are not limited to IIEs (or any other type of benchmark) .
Our study of the various sectors suggests that a case-by-case
approach will be necessary, tailoring the criteria against which
to judge "success" to the nature of each individual sector. I
suspect this will in fact be the most difficult and most critical
aspect of the talks when they finally begin in earnest by
addressing specific sectors, but that solutions can ultimately be
found that both countries can accept.
An Assessment of the Administration's Effort
Our new study thus largely supports the basic thrust of the
Administration's Japan policy to date : pressing the Japanese to
boost domestic demand through expansionary fiscal policy, a
stronger yen, support for global trade liberalization through the
Uruguay Round in the GATT, and sector-specific liberalization
including through structural changes that will open the Japanese
economy increasingly to external trade and investment. In the
course of implementing this policy, I would suggest four
modifications in the present strategy.
First, the Administration has been quite explicit about the
nature of the macroeconomic commitment it has sought from Japan:
a sharp cut in that country's global current account surplus . In
its daily rhetoric, however. Administration officials have often
spoken of seeking significant cuts in — or even elimination of —
Japan's bilateral trade surplus with the United States . These
are two very different policy targets and the Administration
should carefully focus its attention on the former.
Targeting the bilateral trade imbalance makes no sense
either conceptually or pragmatically. In a world of multilateral
trade, it would be ridiculous to target bilateral balance with
any country. We would be decidedly unhappy if the European
Community or Mexico adopted a policy of eliminating their
bilateral deficits with us.
Moreover, Japan will still run a sizable bilateral trade
surplus with the United States even when it fully meets the
Administration's target of cutting its global current account
surplus to 1-2 percent of its GDP. There would still be such a
bilateral imbalance if the United States fully eliminated its
global current account deficit. The reason is the geographical
pattern of trade: Japan imports huge amounts of primary products
from Saudi Arabia, Australia and Brazil and must export similarly
huge amounts of manufactured products to high-income industrial
countries like the United States.
The bilateral imbalance has in fact been relatively stable:
Japan's global surplus has quadrupled over the past three years
(more than $100 billion) but its bilateral surplus with us has
risen by "only" about 25 percent ($10 billion) . The bilateral
imbalance will clearly come down as Japan's global surplus
declines, per the close linkage with the exchange rate as shown
in Figure 2.3, especially as a share of both countries' economies
and of the trade flows between them. It will remain sizable in
absolute dollar terms, however, and we should not regard that as
a major problem. The Administration should recognize this
clearly and thus devote its full attention, rhetorically as well
as in practice, to reducing both countries' global current
account imbalances .
Second, the Administration must recognize that some past
negotiations with Japan have yielded impressive results. As
noted, the Plaza-Louvre period of macroeconomic and currency
coordination in 1985-87 brought an enormous reduction of the
Japanese surplus and the American deficit. Our Table 5.5 shows
that a number of past sectoral negotiations have generated
impressive increases in American exports; the semiconductor deal,
for example, appears to have doubled both US exports to Japan and
the sales there of American-based firms from all of their
production locations around the world. There has been
encouraging progress on a number of the structural issues, such
as Japan's distribution system (where large-scale American
retailers are now expanding rapidly) and its willingness to use
antitrust law (as in the recent egregious case of the glass
Our book demonstrates that enormous problems remain in the
United States-Japan economic relationship. Vigorous efforts
across a wide range of issues are essential to deal with them.
But it would be foolish to ignore the positive results of many
past efforts and to fail to learn the lessons which they provide
for the current effort.
Third, as already noted, the structural issues are so
important, and the prospects for further convergence are
sufficiently promising, that the Administration should pursue
those issues separately from their sector-specific efforts. To
be sure, these two sets of issues intersect; structural issues
are an important component of some sectoral problems and some
sectors cannot be opened without structural reform. But there is
a risk that the structural matters may receive short shrift in a
sector-focused approach, which would be extremely unfortunate
because they offer such a promising prospect for improving the
overall situation over time.
The Bush Administration's Structural Impediments Initiative
(SII) made an important contribution to furthering the process of
policy harmonization and performance convergence. Now that an
American Administration is willing to take seriously the need for
structural reform in this country, the prospects for success in
such talks will be even greater. They should be rechristened the
Structural Convergence Talks (SCT) and pursued on an independent
track in the upcoming negotiations .
Fourth, as also noted above, the Treasury needs to lock in
the stronger yen by getting Japan (and hopefully the entire G-7)
to promptly implement a system of target zones around current
exchange-rate levels.^ Further yen appreciation to about 100:1
against the dollar would be desirable to assure the needed
adjustment of Japan's current account. The Treasury's own target
of reducing the Japanese surplus to 1-2 percent of its GDP in
fact implies a target zone for the yen of about 100:1 to 110:1.
The most important goal now, however, is to avoid any
significant renewed weakening of the yen. Such weakening has
occurred on three previous occasions over the past twenty years —
in 1975-76, 1980-82 and 1989-90 — and triggered huge increases in
Japan's surplus which in turn greatly intensified our trade
problems and the frictions between the two countries. We should
^ The installation of reference ranges, a weaker version of
target zones, by the G-7 in February 1987' was presaged by a
bilateral United States-Japan agreement to set such a range for the
yen-dollar in October 1986.
learn from history and avoid yet another replication of this sad
history. Rapid installation of target zones is the most
effective way to do so.
I would close with one word of advice for the Congress: do
not legislate on this issue . It will take an enormous effort and
an extended period of time for the Administration to achieve its
objectives with Japan, as largely endorsed by our new study.
Legislation of precise negotiating goals, time deadlines or
retaliation requirements could convert a difficult task into an
impossible one. The Administration has full authority under
current law to pursue the talks and back them up with forceful
action as necessary, and I hope that the Congress will restrain
itself as the Administration uses these authorities to the
Figure 2.2 Japan: currant account and raal axchangt rata
Real exchange rale (1985«100)
Current account/export (t«2)
Real axchanga rate (t)
J__l I I I ' ' ■
J I I I l_
1B79 1960 1961 1962 1963 1964 1965 1986 1967 1968 1989 1990 1991 1992
Source: International Monetary Fund. lnt*m»tion»l Finineiil S:$listiea, various iasuas.
Figure 2.3 United States and Japan: raal dollar-jrtn exchange ratal and bilateral
Uade balances. 197B-92
a. US-Japan bilateral balance u a share of total US-Japan trade.
b. Yan-dollar rate adjusted (or Inflation in both countries.
Source: calculations by William R. Clina from Interrtational IMonetary Fund data in
Internationil Finencial Stitiitics and Direction ol Trade Stitiuic*. various issues.
30 RECONCIUkBLE DIFFERENCES?
Reproduced from K»concilMbl» Diff»rmncmB? Vnifd State* - Jmpmn
Economic Conflict, by C. Fred Bergsten end Marcua Moland.
Haahingtoni Znatitute for Znternatlenal Icononica, 1993. All Kighta Reaerved.
Table 5.5 Apparent and potential US galn» from liberalization In
Japan (millions ot dollars)
App»renl gilnt fiom llber«IlMtlon« f """'If' "'ll* .?*"
Ogt'tnn 1^ Compirt.r.' 1.075
semiconductor.* .416 Wood product. 750
CHru. 100 R.U.I.ng' 150
Fiber optic. » Supercomputer. 30
( Doe. no\ include »n estim.led $1 billion In US firm.' loreisn-wureed produeU.
b Does not include en estim.led $715 million In US firm.' toreisn-wurced produeU.
c. Includes B*ir<* .nribuUble to ToytRU. only.
1*2 RECONCILABLE DIFFERENCES?
Raproducad from Jt*co/icii«l)J* Diff»r»nc»aT Vnifd Bt»fM - Japut
Xconoaic Conflict, by C. Pr*d B«rgat*n and Marcus Noland.
Haahingtont Znatitut* for Zntarnational Bcononiea, 1993. All Kighta Maarvad.
Chairman Gibbons. Let me ask you a couple of questions here.
Let's go to those exhibits that you attached to your testimony,
figures 2.2 and 2.3. I have to admit that I tried to study your book
last night, but there were just too many other things that had to
be accomplished and read. Could you explain that figure 2.2 chart?
Mr. Bergsten. What this chart says is that Japan's global cur-
rent account balance — the solid line — tracks almost precisely, with
a 2-year lag, the trade-weighted exchange rate of the yen — ^tne dot-
ted line. Japan's global trade outcome corresponds almost perfectly
to Japan's global exchange rate 2 years before.
In the first few years of the 1980s, the chart, on the left scale,
shows a period when the yen exchange rate was declining. The way
the index works, a rising line means a declining value of the yen —
a rising dollar. From 1979 to 1982, the yen declined significantlv.
That is the rising dotted line. And, with a 2-year lag, Japan's glob-
al current account surplus rose correspondingly. It rose a little
longer in that time phase, a year or two longer than expected, due
mainly to the enormous lall in world oil prices, but then changed
direction. The exchange rate of the yen began to rise from 1982,
and it rose sharply from 1985 through 1988. That was the period
of the Plaza Agreement, the effort to bring down the dollar and to
increase the value of the yen, the deutscne mark and other cur-
rencies. So from 1982 to 1988 the exchange rate of the yen rose
substantially. That is the declining dotted line. With the 2-year lag,
the black solid line, which is Japan's global current account surplus
2 years later, also declined sharply. That is measured on the right
hand scale. If you want to put it in terms of Japan's GNP, the sur-
plus dropped from 4.3 percent of GNP in 1986 to 1.2 percent in
This chart shows the surplus as a share of Japan's total trade.
It dropped from 40 percent of its exports to a little over 10 percent
of its exports in that 4-year period.
Then from 1988 to 1990 the exchange rate of the yen declined
again, represented here by a rise in the dotted line. That decline
in the value of the yen was not nearly as great as in the earlier
period but it was substantial. It was about 25 percent from its base
in 1988. With the 2-year lag, the solid line then turned up again,
and that is the substantial increase in Japan's global surplus that
we are now experiencing.
From about 1991, the dotted line turns down again. That means
the exchange rate of the yen began to rise and so our prediction
would be that, starting next year, Japan's trade surplus with the
world would again start to decrease. I said before, that crucially de-
pends on keeping the yen at the strong level it has now reached.
I hope that clarifies the relationship. It is not the easiest correla-
tion to understand. We tried to put it in the simplest way possible.
Chairman Gibbons. Well, is the correlation between the yen and
the dollar approximately the same as the yen and the rest of the
Mr. Bergsten. Not always. There have been periods, as in the
early 1980s, when the yen went one direction against the dollar
and the deutsche mark went the other direction. In the last 6
months or so, movements have again been in different directions.
The yen has gone up very substantially against the dollar but the
dollar has gone up very substantially against the deutsche mark
and the European currencies. In fact, I am getting a little worried
that the dollar might now be getting a little too strong against the
European currencies. I have not had that worry for some time.
It is interesting to note that the trade-weighted average of the
dollar is not down over the last 6 months to a year; it is up. The
dollar has declined only against the yen and, by a very modest
amount, against the Swiss frginc. The dollar has gone up consider-
ably against the deutsche mark and most of the other European
continental currencies. The dollar has gone up dramatically against
the pound sterling, the Italian lira, and the other European cur-
rencies that were devalued within the European exchange rate sys-
So the trade- weigh ted dollar is actually up quite substantially
over the last 6 to 12 months. Newspaper reports of declining dollar
are wrong. They mistake the fact that the yen has been rising for
a dollar decline whereas, in fact, the dollar is up against the other
What that means, of course, is that the yen has risen consider-
ably against the dollar but it has risen much more against the Eu-
ropean currencies, so the global average exchange rate of the yen
in this period has gone up by much more than it has just against
the dollar. This means that Japan's global surplus should decline
even more than its bilateral surplus against the United States,
based on the kind of relationships we are talking about.
Chairman Gibbons. Well, as I remember wnen we began this
discussion some years ago on the United States-Japan relationship,
the yen was about 360 to the dollar. Is that about right?
Mr. Bergsten. Before the Smithsonian Agreement of 1971, over
20 years ago, the yen had been at 360 since the time we set that
rate during the occupation.
Chairman Gibbons. And this morning it was 110 to the dollar.
How in the world do the Japanese still maintain this huge trade
surplus with us with that kind of long-term swing in the exchange
Mr. Bergsten. By being very competitive and by the fact that
the 360 rate had become enormously undervalued by the time it fi-
nally was changed.
You may remember that Japan came on the world scene in a
major way in trade in the 1960s and began to run substantial sur-
pluses for the first time in the late 1960s. That was one of the
main triggers for the first big set of exchange rate changes back in
1971 and 1973 but that change, though it seemed big at the time,
only began to catch up with the enormous growth of Japan's pro-
ductivity during the first generation of its postwar recovery. So
subsequent further strengthenings of the yen were required.
I think that toward the end of 1987 we had it about right. At the
end of 1987 the yen hit about 120 to 1 against the dollar, and with
the usual lag Japan's global current account surplus was down to
1.2 percent of its GNP. Japan's bilateral surplus with the United
States was still large in absolute terms, about $40 billion, but it
had come down substantially as well.
If that exchange rate had been maintained, I don't think we
would have this Dig Japanese surplus today or anything like it.
But, as I mentioned, the yen was permitted to weaken again by 25
or 30 percent, returning to 160:1, and that was a big factor in gen-
erating these surpluses.
You raised the question, with the yen at 360 then and at 110
now, and how could it be that Japan still has such a big surplus?
It is simply the difference in productivity growth. Japan s produc-
tivity was growing at 10 to 12 percent a year all through the 1960s
and into the early 1970s when ours was growing — better than to-
days standards — ^by 2 to 3 percent. And then subsequently they
were growing at 4, 5, and 6 percent when we were growing at 1
or 2 percent. Those enormous differences in productivity growth,
compounded over 25 or 30 years, simply require a big change in the
exchange rate in order to equate the competitive positions.
Put another way, Japan has reaped a large part of the benefit
from its enormous productivity expansion through a stronger cur-
rency, meaning it can buy things from foreigners and invest abroad
at much less cost than it ever could in earlier periods.