Brazil: Analyzing the Crisis
And Prospects for Recovery
André Lara Resende
The following text is an edited transcript of remarks made by André Lara Resende to
the 1999 annual meeting of the Trilateral Commission in Washington, D.C. André Lara
Resende was Special Advisor to the President of Brazil from August 1997 to April 1998, and
then President of the National Bank of Development (May-November 1998).
I will talk briefly on two schematic
interpretations of what went wrong in Brazil. Of course, you could simply say nothing was
wrong domestically and that we were simply the victims of the contagion from the crisis of
emerging markets. I dont think this view should be taken seriously. The important
question anyway is why Brazil was vulnerable to such a crisis.
Why Was Brazil Vulnerable? Exchange Rate Policy, Inadequate Fiscal Adjustment
The first view is that the root of the problem in Brazil was the exchange rate policy.
When the real was introduced in July 1994, it managed very successfully to stabilize
prices after three decades of chronic inflation. The real was left free to float from July
1994 until approximately April l995 and became overvalued in relation to the dollar. The
starting parity was one real per dollar, but the parity went up to 83 cents of the real
per dollar, and then came back slightly. The reason that the real revalued was the fact
that extremely high domestic interest rates had to be maintained to consolidate the
stabilization of prices. The resulting differential between domestic and external interest
rates attracted a lot of short-term capital.
For many analysts, this was a huge mistake and there was no reason whatsoever to allow
the real to revalue immediately after the stabilization program began. The revaluation of
the real was, however, very important to reverse some of the currency
substitutionthe dollarizationin the Brazilian economy; and the importance of
this revaluation in bringing inflation down quickly cannot be underestimated. However, in
April 1995, the exchange rate policy was changed, and though it was presented as an
exchange rate band, it was, in fact, a crawling peg. The only difference between this
crawling peg and the crawling peg of the inflationary period was that now it had no fixed
relation with domestic inflation; Brazil simply devalued by approximately a rate of six or
seven percent per year (on a monthly basis, 0.5 or 0.6 percent) with respect to the U.S.
According to this view, these actions were not enough to reverse the original
revaluation of the real; inflation was under control, but the exchange rate was
dangerously overvalued. This put domestic industry in a difficult situation, squeezed
between an extremely high cost of capital due to high interest rates, on the one side, and
foreign competition due to an overvalued exchange rate, on the other. As a result, the
large trade surplus that Brazil had maintained for many years was reversed. The current
account deficit started to increase rapidly. The government insisted on playing the card
of high interest rates to finance the increasing current account deficit. Therefore, when
we were hit by the crisis in Asia, and then in Russia, confidence went down and to float
the real was imperative to stop losing reserves. According to this view, the use of an
overvalued exchange rate for so long, combined with high interest rates to attract
short-term speculative capital, now risked threatening domestic inflation stabilization.
According to a second view, the problem in Brazil was not really related to the
exchange rate; it was simply a question of not being able to promote the necessary fiscal
adjustment. When the Real Plan was introduced, it included a very specific constitutional
amendment to guarantee some kind of fiscal relief. This was, however, only a transitory
measure that would maintain fiscal balance for a few years. Everybody knew that fiscal
equilibrium in the long run would depend on institutional reforms. Resistance from
pressure groups with interests in maintaining a heavy state sector delayed these reforms.
The government had to go on financing a rapidly increasing public debt with higher and
higher interest rates. The fiscal deterioration caused a loss of confidence. Low
confidence ended up provoking the crisis when other emerging markets collapsed. According
to this view, the exchange rate was not in question, and if the real had been devalued,
the only winners would have been those forces against the structural reform and the quest
for fiscal equilibrium It would have been the victory of those accustomed to living under
an apparatus of protection from outside competition and dependent on public subsidies.
Of course, it is possible to recognize in these two views the two classical cases of
wrong relative prices and the necessity of expenditure-switching policies, on the one
hand, and, on the other, of too much expenditure, low public saving, and the necessity of
expenditure-reducing policies. I think that to accept one of these extreme views and to
reject the other is incorrect; virtue most of the time is in the middle, and both have
some merit. There is no doubt that if we had managed to make more progress on the fiscal
side we would have been less vulnerable to the external crisis. At the same time, it is
clear that a more devalued domestic currency would have helped significantly, since the
current account deficit to be financed would have been smaller. Interest rates could,
therefore, have been lower. As a result, there would have been more growth, less
unemployment and a better fiscal situation. These two questionsexchange rate policy
and fiscal equilibriumshould be handled together.
After the Asian crisis in 1997, Brazil started to lose reserves. Interest rates, down from
levels in excess of 50 percent at the beginning of the real stabilization program to
levels around 20 percent, had to be sharply increased again to face the external shock.
Higher interest rates led to a further deterioration of the fiscal situation and to a
rapid increase of domestic public debt. A large and increasing current account deficit
associated with successive failures to reduce the public sector deficit eroded confidence,
and Brazil started to lose reserves. Late last year, it became clear that the country
should try to have a preventive arrangement with the IMF. An agreement was reached, with
the understanding that Brazil would not devalue. In fact, this was a point that Brazilian
negotiators stressed as a condition for the accord. In January, with foreign reserves
continuing to fall, in spite of the agreement with the Fund, Brazil tried to change the
exchange rate policy. The crawling peg was abandoned and an exchange rate band, with
explicit rules of intervention, was announced. After two days it had failed miserably.
Faced with massive capital outflow, Brazil had to stop defending the band and to adopt the
In my view, the attempt to change the crawling peg rate to a band was technically
correct, but the timing was wrong. It was too late. Immediately after the agreement with
the IMF, the government suffered two important political defeats. First, it failed to
approve a piece of legislation on the Social Security reform which was vital for meeting
the fiscal targets agreed with the Fund. Second, in early January, the new governor of the
state of Minas Gerais, ex-President Itamar Franco, declared a moratorium on that
states domestic and external debt. These two factors, associated with the
substitution of the Central Bank Governor, who had defended the previous exchange policy,
led to a complete loss of confidence in Brazils ability to defend the newly
announced band. The fact that the IMF was surprised by the change of policyand did
not hide the fact that it was not very happywas also very important in aggravating
the confidence crisis. In these circumstances, I think it was the right decision not to
try to defend something that had become clearly indefensible.
What happened next? The real fell from 1.21 per dollar to 1.45. It held there for a few
days. There was some initial optimism that this rate could be sustained. Well, not so.
When the excess reserves of the banks dried up, there was complete illiquidity in the
exchange market, and the real started to devalue more and more, ending up below 2.2 per
U.S. dollar, far beyond any reasonable point of long-term equilibrium. All credit lines to
the country were significantly reduced; even trade lines were called back. Since external
payments still had to be paid, any additional demand on the exchange market only added to
the pressure on the price of foreign reserves.
The Central Bank of Brazil had its hands completely tied by the IMF. The first
tranche of nine billion dollars had already been drawn. Before using the money
to try to stabilize the exchange market, Brazil had to renegotiate with the Fund. So for
all practical purposes, the Central Bankwith about 39 billion dollars of reserves,
counting the IMF moneyhad to act as if it had zero reserves, and could not intervene
in the market for two weeks. After a new agreement was negotiated and the IMF agreed that
the Central Bank of Brazil could intervene in the market, the real recovered to levels
around 1.80 per U.S. dollar.
Prospects for Recovery
Where are we now and what are the perspectives? Well, I think that after a few extremely
rough weeks, things are much better. Those who subscribe to the view that the roots of the
Brazilian problems were the overvalued exchange rate would certainly say that we have a
high probability of a fast recovery. The current exchange rate seems to be, by all means,
undervalued. Brazils exports will recover, even if the world situation does not
improve significantly, and imports have collapsed. The trade surplus will, therefore,
reappear soon. The current account deficit will be reduced; confidence should then come
back and credit lines return. Finance Minister Malan is now with the new governor of the
Central Bank, Arminio Fraga, in Paris to discuss with banks the possibility of
reestablishing the credit lines so that the market can stabilize. A short recession seems
unavoidable but the economy should probably start to recover before the end of the year.
If, however, you take the view that Brazils difficulties are due to the lack of
long-term fiscal adjustment, you would still have to ask if the agreement with the IMF
will finally lead to serious fiscal homework before taking a positive stance on the
economy. Is it sufficient to guarantee that inflation will not come back after the large
nominal devaluation of the currency?
There is one additional restriction that tends to be overlooked in most analyses, which
is the socio-political restriction.
The crucial question is why are interest rates in Brazil still so high? The lack of
confidence is certainly an important part of the answer. High interest rates are due to
the lack of confidence, but high interest rates aggravate the fiscal situation that is the
main cause of the lack of confidence...
A Mercosur Currency Union?
Is there an alternative to this difficult situation? The alternative of adopting a
currency board, of using a fixed exchange rate with a currency fully backed by
foreign reserves, has many supporters. This is, of course, what Domingo Cavallo has
adopted in Argentina and what he defends as the best alternative for countries that come
from a recent experience of hyperinflation, like Argentina and Brazil.
The consensus today on exchange rates seems to be that there is a superiority of the
extremeseither you float or you go to a fixed exchange rate. The float is believed
to be more adequate for large closed countries and fixed exchange rates for small open
economies; so Brazil would clearly qualify as a candidate for the float. This would be
certainly true if Brazil were not coming from a long period of chronic inflation and a
serious crisis of confidence in its currency.
There is strong political resistance to the Currency Board. This resistance seems to
be, at first, of a psychological nature, probably due to the association of currency
boards to the colonial regimes of the past. There are, however, two recent misconceptions
associated with the adoption of the currency board. The first is that the adoption of a
currency board has to be preceded by a forced restructuring of public debt. The second is
the idea that it is not the stock of M0, the monetary base, which has to be fully backed
by external reserves, but indeed M4, the broadest definition of a monetary aggregate. The
reasoning is the following: Since domestic public debt is mostly short-term debt, it
could, in principle, be totally transformed into money; in order to avoid a run it would,
therefore, have to be fully backed by foreign reserves. This is, however, a case of the
classical confusion between stocks and flows. All you need is a guarantee that no
additional currency will be issued without an equivalent increase in the stock of the
reserve currency. The current stock of domestic currency is being held under very adverse
circumstances. Introducing the restriction on the issuance of additional currency will
only increase the quality of the stock of financial assets denominated in the domestic
currency and, therefore, lead to an increase in the demand for money and debt.
There is one additional risk with respect to the idea of a currency board.
There is always the temptation to look at it as a short-term panacea to which you can turn
in a desperate situation. Most of the benefits of the currency board come in the short
run: a higher degree of confidence and lower interest rates. The fiscal constraints and
the restrictions on the ability to react to adverse external shocks, however, are
I believe that to consider the possibility of adopting a fixed exchange rate in the
near future, fully backed by foreign reserves, only makes sense as part of a long-term
program towards a currency union for the Mercosur region. This is not incompatible at all
with a successful way out of the current crisis with a floating exchange rate. On the
contrary, I believe that the monetary union should not be viewed as an easy alternative to
short-term difficulties, but as part of the process of going toward a new world.
Everything seems to indicate that, in the future, there will be fewer currencies. There
seems to be too many currencies and too much instability for an increasingly integrated