Revised
Economic advice and regime change in Portugal
Jorge Braga de Macedo[1]
FEUNL & IICT
In
the mid XV century, a philosopher king wrote The Loyal Counsellor - a classic of Portuguese literature[2]. This may have started a
tradition of treasuring advice, with the useful caveat that advisers might not
be loyal. This paper provides three different examples of loyal economic advice
about regime change. The emphasis is on the exchange rate regimes following the
April 25th revolution and leading to membership in the eurozone. All
three examples reflect the legacy of economic advice from the MIT Group in the
mid 1970s, one of the motivations for the 30th anniversary conference
at which this paper was first presented.
Particular
attention is given to the exit from the policy of stability of the Portuguese escudo with respect to the pound sterling
and the dollar which had lasted almost fifty years[3]. After an unannounced
policy of gradual depreciation during 1976 and a discrete devaluation in
February 1977, a crawling peg was announced in August and remained in place
until after the beginning of the regime change which led to the adoption of the
euro. Economic advice from the MIT Group provided a policy framework for
macroeconomic stabilization and accompanying structural measures. Instead, the
other examples focused on devaluation and currency reform respectively.
Moreover, unlike the crawling peg and entry into the exchange rate mechanism,
the first Gulbenkian conference on the Portuguese economy enhanced public
debate about advice[4].
The
importance of communication of economic advice is put into perspective in the
concluding section, where earlier currency reforms are brought to bear on the
argument. In effect, the hyperinflation of the 1920s, the devaluations of the
1970s and the realignments of the 1990s notwithstanding, Portuguese currency
experience reveals a striking resilience of the commitment to a fixed nominal
price of foreign currency.
The
paper is divided into four sections. Section 1 describes the exit foretold.
This process of adjustment to dramatic external and internal shocks, the oil
crisis and the revolution, featured a succession of stand-by agreements with
the IMF lasting almost until membership in the Exchange Rate Mechanism (ERM) of
the European Monetary System in 1992. This “IMF standard” was associated with a
crawling peg seeking to stabilize the real effective exchange rate escudo described in Section 2[5].
The
search for real stability places the currency regime in the perspective of the
research carried out by the MIT group and my interaction with six graduate
students stationed at the Bank of Portugal during the Summer of 1976[6].
As a recently decommissioned junior lieutenant I interviewed them for Nation and Defense, a military journal,
and one of the questions raised the issue of advice on economic reforms. As
executive secretary of Economics, a
newly established economics journal, I also arranged for the publication in
English of a short description of their model of the Portuguese economy. The
late Rudi Dornbusch, one of the supervisors of the MIT group, advocated the
crawling peg instead of a repetition of discrete devaluations as part of the
IMF stabilization packages[7].
Section
3 describes the third example, a gradual regime change towards currency
stability and convertibility, which entailed the combination of what I called a
multi-annual fiscal adjustment strategy (MAFAS) and a pre-pegging exchange rate
regime (PPERR) in Bliss and Macedo (1990). I have also claimed that a focal
point in the process which allowed the escudo to be part of the euro was
the decision to enter the ERM shortly after the signature of the
The
closing section 4 attempts to differentiate between advice on entry and exit
when it comes to the six currency regimes listed in Table 1. In two of the last
three decisions, advice involved the central bank, either alone or in
connection with the IMF and academic consultants from MIT. Advice on the last
change, on the other hand, relied on embedded economic advisers. The conclusion
comes back to the adjustment process, emphasizing that macroeconomic and
structural reforms are complementary.
Table 1 Advice on regime change for
the real and the escudo
|
Regime |
Entry date |
Advise |
Exit date |
Reason |
Reference |
|
new currency |
1243 |
|
1797 |
fin crisis |
Macedo et al. (2001) |
|
gold standard |
1854 |
Parliament |
1891 |
fin crisis |
Reis (1996) |
|
gold standard |
1931 Jul |
civil society |
1931 Jul |
£ crisis |
|
|
Stability w/ respect £, $ |
1931 Sep |
Bank of |
1977 Feb |
Devaluatn |
Macedo et al. (2001) |
|
Crawling peg |
1977 Aug |
IMF, MIT |
1990 May |
PPERR |
Dornbusch (1981) |
|
Exchange Rate Mechanism |
1992 Apr |
DG ECFIN |
1998 May |
Creation
€ |
Macedo
(2007) |
Source:
Abbreviations explained in references and text. .
1.
An
exit foretold
1.1.
From revolution to devaluation
In
early 1975, Serge Christophe Kolm, a well known, if somewhat heterodox French
economist, came to the Centre for Military Sociology, where all the
revolutionary captains met, and gave a talk on the economics of the socialist
transition, initiated by the widespread nationalization of March 11th.
He also spoke about regime change in terms of the exchange rate. As shown in
Table 1, the last devaluation against sterling had taken place in September,
1931 (precisely 82 days after restoring gold convertibility!). When the pound
devalued against the dollar in 1949 and in 1967, the escudo revalued against its former numeraire currency. Such a long
period of exchange rate stability or appreciation had elapsed that when Kolm
said: ‘you guys have to devalue - now’, the captains replied ‘in this country
we don’t do that!’[8]
Kolm (1977), titled Socialist
Transition – The Political Economy of the Left, is summarized in his
website, where it is claimed that “in all cases the economic policies could have been better on two
grounds. One is a better economic analysis concerning notably the dynamics of
inflation, exchange rate policy, and the effects of various measures. The other
issue is that the instruments of distribution can be chosen to be much less
disruptive of market efficiency, notably in basing taxes or subsidies on much
less elastic items”. In comparing macroeconomic
adjustment in
He then uses the open economy income identity to illustrate the simultaneous
determination of the level and composition of aggregate demand but not with the
simplicity of Dornbusch (1980, p. 29) who lists the identities between the twin
deficits and the private savings gap and the twin deficits and the excess of
broad money creation and domestic credit to non banks before the reminding us
that identities do not “tell us what determines what”. Kolm (1977) wants to
bring out the consequences of an increase in wages (W) and a decline in profits
(P) with the
S - I = CAB+GDEF
CAB =X-M= exports minus imports
GDEF = G-T= expenditure minus revenue.
Kolm (1977) also brings out the consequences on money creation, which depend on the composition of money supply, as they must under fixed exchange rates, with an increase in domestic assets matching the decline in foreign assets if S=I. Other identities in Dornbusch (1980) provide a simpler link between national accounting and “financial programming”: net foreign assets are indexed by the consolidated banking system (b, which includes the central bank), or by the government (g), whereas domestic credit is to non-banks (nb) or the government. In other words:
CAB = DNFAb + DNFAg
GDEF = DDCg - DNFAg:
In
an annex, Kolm (1977) describes the absence of sterilization by the Bank of
Portugal of the precipitous decline in foreign exchange reserves and the fall
in credit to the private sector by the nationalized banking system as “the
Portuguese monetary paradox”. It is this paradox that motivates the
recommendation of devaluation before reserves are exhausted and borrowing
abroad becomes necessary, as it did in
The
reaction of the revolutionary captains to this loyal advice gives a good idea
of what the public opinion was amongst the intelligentsia about the exchange
rate regime or their ability to communicate an unpopular decision effectively.
The advice the revolutionary government received to devalue the currency by
about 20% as soon as possible was not followed. Quoting again from Kolm’s
website summary of this book: “Since the publication of Socialist Transition, all governments or political parties
considering a fast reduction of economic inequalities in the framework of a
market economy have asked the author or other members of his teams for advice
(the first instances were all the governments of Southern Europe in the early
1980’s). This permitted some advances, and to begin with the avoidance of the
most dramatic failures of the previous cases (although advice was sometimes not
followed or was misused, and politics has other dimensions)”
1. 2. The interaction with the MIT Group in
the run-up to the Gulbenkian conference
After
the revolutionary fervor subsided, the Centre for Military Sociology was
renamed the Institute for National Defense, and started Nation and Defense, a journal which still exists, attempting to
disseminate more reformist ideas and thus bridge the gap between civilian and
military opinion[10].
The last entry of the first issue was an interview to the MIT Group on the
Portuguese economic situation. My introduction explained that the six questions
had been answered in writing; in the spirit authorized by the Governor of the
Bank of Portugal[11].
The interviewees had been called students in some op-ed pieces so I mention
that they were third year graduate students, preparing their Ph.D. The subtext
was an attempt to dispel the idea that the members of the group were “too”
young.
I
conclude by introducing them in alphabetical order: “Andrew Abel, economics
major at Princeton University, had published work on econometrics, Miguel
Beleza, one of the best students of the last few years in the Higher Institute
for Economic and Financial Sciences (ISCEF), with two years of teaching
experience there and also at Nova and Catolica Universities, Jefferson[12] Frankel, economics major at
Swarthmore College, Raymond Hill, economics major at Princeton University who
presented at the same conference a paper on money and credit written with
Teodora Cardoso, economist at the Bank of Portugal and, at last, Paul Krugman,
economics major at Yale University, who published an article on deflationary
effects of devaluation in the Journal of International Economics made
important interventions at the same conference”[13].
Once
again, the idea behind the sole interview to the MIT students was to point out
to the military intelligentsia that these experts had been working all summer
for the Bank of Portugal: they could not be considered instant advisors. On the
other hand, the importance of their being outsiders also stood out: they had
come at the invitation of the Bank of Portugal, it was arm’s length from the
government, and so they were able to see what everybody was doing in the
different ministries. Indeed, one of the outcomes of the exercise had been to
uncover a lot more data on the Portuguese economy than the advisers thought
existed. Hence, if different ministries would talk more with each other, they
could share this data more effectively and make sense of it through a common
analytical framework: “when assembling data from a variety of published and
unpublished sources we found that more information is available about
Portugal’s economy than one might have expected. Of equal importance was the
discovery that the process of producing a quantitative model - even a model
based on limited data - had a beneficial effect on the whole of our research
effort. This effect occurred in two ways. First, the modeling effort helped
define the areas in which research was most needed: if what was known about,
say, the import sector was found to be insufficient to allow us to model the determination
of imports, imports became a subject of further investigation”[14]
Second
message from the interview was a statement about the importance of
macroeconomics. Rather than the usual tussle on macro versus micro, the
statement reflects the simplicity of the model of the Portuguese economy
published in the first issue of Economia,
the Journal of the Catholic University of Portugal. This model was also
described the first chapter in the two volume of conference proceedings, and it
seemed to me the simplest, most compelling and, according to my interview, the
most important of the conference [15]. Yet, if you look at Expresso,
which was then (and perhaps continues to be) the staple weekly news magazine,
what Paul called prime journalistic real estate, this was not the hit of the first
Gulbenkian conference[16].
If you give attention to institutional detail, use the information that exists on it and then analyze both through a simple model in a way that allows you to derive policy implications, then you really are making a difference as adviser, you become a loyal economic counselor. The MIT advisers came and quickly solved an economic problem, (like paratroopers keeping with military jargon). Other schools of thought might love to start the complicator first (if it can be made complicated, why keep it simple?)[17]:
A lasting
beneficial effect of the loyal economic advice of MIT was that the construction
of a model required that the analyses of different aspects of the economy be
consistent with each other. The example illustrates the simultaneous
determination of the level and composition of aggregate demand, through the identities
in the celebrated Dornbusch (1980): “a balance of payments projection and an
investment projection together imply a projection of savings”.
The model in Abel et al (1976)
jointly determines output, consumption, and imports given a level of autonomous
spending. But it allows relative prices (e.g. the price of imported goods
relative to wages) to magnify both the favorable and the unfavorable effects of
increased investment. A simulation showing that the multiplier effects are
fairly small and that the final effect of policy changes on the trade balance
is often very different from the direct effect leads to a clear warning against
single adjustment measures such as devaluation: “policies which aim at
improving the balance of payments without accompanying policies to regulate
domestic demand are likely to have disappointing results”.
Abel et
al. (1976) end
with a plea for more work on the role of monetary policy and the supply side of
the economy. The lack of an adequate analysis of the monetary mechanism in
Like the interview, the paper had to avoid the word
“devaluation” and call instead for “measures to improve the balance of
payments”. The label did not matter at some level, what was important is that the
explicit aggregate model helped drive the point home in a way that Kolm’s call
to save the revolution did not. Indeed, after the Gulbenkian conference, there
was still a lot of hesitation, but public opinion caught on, and devaluation
became inevitable.
This story of an exit foretold remained after February 1977, because
the underlying causes for devaluation remained in place. This led to another policy package six
months later, where interest rates were increased further and the exchange
regime became a crawling peg based on inflation differentials with forward
cover by the Central Bank. The MIT Group hinted that a crawling peg could be
used to blunt the impact of inflation and retain trade competitiveness[18]. The major feature of the
IMF agreement of May 1978 was to recognize explicitly the connection between
the rate of crawl and the rate of monetary expansion, and thus to set tight
ceilings on domestic credit creation and the public sector borrowing
requirement. The current account turned into a surplus in the third quarter and
so did the balance on non-monetary transactions, while real wages continued to
fall to their warranted level. The rate of crawl accommodated the domestic
inflation rate in 1978 but declined to 17 per cent in 1979, in the light of the
improvement in the external position.
2. Searching for real stability
2.
1. Follow-up on budgetary procedures and monetary myths
The main
follow-up points to the Gulbenkian conferences do not appear to be macro, but
have relevance for policymaking institutions along Tommasi (2001) lines. This
is the fact that in
Jurgen von Hagen and others have been comparing the various stages
of the budgetary process through very detailed institutional analysis. They do
this because the formal and informal rules governing the drafting of the budget
law (formulation), its passage through the legislature (adoption) and its
implementation distribute strategic influence among the participants in the
budget process and regulate the flow of information. Around 2000 they carried
out a survey in the European Union, according to which the Minister of
Finance’s power over the budget has the lowest score in
That
simple fact is based on detailed institutional knowledge, a point that needs to
be hammered through because it complements the macroeconomic approach favored
by Abel et al. (1976). I will go as far as saying that this detailed
institutional knowledge is also needed because the low power of the Minister of
Finance over the budget is often not seen as a problem. In fact, the opposite
is true: a strong Minister of Finance only fits in a dictatorship[21].
Monetary
myths can be harsher and more rigid than institutions. Especially when they
appear to be rooted in history: “during the gold standard we had financial
stability and democracy. But there was a financial crisis then a revolution.
Then, with the republic, we had hyper inflation and democracy. The late 1920s
brought back financial stability, with a political dictatorship. So when
democracy returns, there has got to be inflation!”
This
was indeed one of the problems the MIT Group looked at in the mid 1970s. When
stable governments came back ten years later and exchange rate stability followed
in the early 1990s, the reaction was that… there’s got to be dictatorship
somewhere! More subtle, for sure, than under Salazar, but keep looking, you
will find some form of dictatorship. This asserts that political freedom and
financial freedom are inimical, instead of being complementary.
This
is a serious issue, not unique to
2.2. The adjustment process from devaluation to
crawling peg
Graph
1, from the AMECO data base, shows the nominal (left scale) and real (right
scale) effective exchange rate with respect to the ECU/euro from 1976 to 2001.
The former shows that the number of escudos per ecu increased by a factor of
six over a twenty five year period, an average depreciation of close to 8% per
annum. The latter, an index based in 2000 using unit labor costs relative to
the partners in the eurozone, has a range from 80 to 110, and the pattern of
depreciation only lasts until the late 1980s. This section discusses the
interaction with the MIT Group in the run-up to the first Gulbenkian conference,
the lack of progress on budgetary procedures, due in part to monetary myths,
and then the introduction of the crawling peg.
This is a period of sustained
nominal and real depreciation, which quickly restored external balance and
showed that the structural rigidities were not such as to require
administrative controls instead of the conventional combination of
expenditure-switching and expenditure-reducing macroeconomic policies. Krugman
and Macedo (1979), based on a variant of the simple dependent-economy model, identify
three phases in the macroeconomic adjustment process. These phases are defined
by the gaps between actual and potential output on the one hand and the actual
and warranted real wage on the other, under the assumption that an increase in
output leads to a current account deficit if the real wage does not fall
correspondingly.
The first phase, l974-5, is
characterized by the opening of the two gaps relative to the situation of full
employment and surplus prevailing in early 1973. While the increase in the
labor force increased potential output and terms of trade deterioration implied
a decline in real income, the revolution brought a decline in actual output and
an increase in real wages. Stabilization policy succeeded in moderating the
decline in output via increased government expenditure. On the other hand,
price controls and a fixed exchange rate ensured that the substantial increases
in nominal wages were not offset by increases in the price of goods. During
this phase the reported current account deficits remained manageable and in
late 1975 the MIT group recommended a disregard for external financing
constraints[23].
By that time, the decline in remittances and widespread capital flight had
virtually wiped out foreign exchange reserves, but there remained the abundant
and undervalued gold reserves of the earlier regime.
Thus the expansionary policy stance
continued in the second phase, 1976-7. In particular, aggregate real
expenditure (including public administration) increased by 7 per cent 1976,
narrowing the output gap. Despite a sliding devaluation in the first half of
the year, imports continued to grow while exports declined. In the face of the
worsening current account deficit, the MIT economic advice in the summer of
1976 was a 30 per cent once and for all effective, devaluation, which would
reduce real wages by 6 per cent and restore cost competitiveness to the 1973
level.
Such unpopular measures could not be
taken before the December local elections, and were thus delayed until the
pressure to borrow foreign exchange made a policy package inevitable;
relaxation of price controls, ceilings on wage increases, modest increases in
interest rates and an effective devaluation of less than 15 per cent were
announced at the end of February. While these measures may have succeeded in
narrowing the wage gaps they had a stronger effect in narrowing the output gap,
for the payments situation continued to deteriorate. The government fell before
the year end but it was still a socialist-led coalition which signed the letter
of intent to the IMF on
In conclusion, Kolm’s advice of a 20% devaluation in early 1975 was
between the 30% suggested in mid-1976 and the realization of 15% in early 1977.
It may have neglected Cooper’s law (1971, p. 502), according to which the
likelihood of loss of power doubles in the year of a devaluation and is multiplied by three for the finance minister[24]. In his comment, Diaz
(1971, p. 514) carries the political argument further by looking at the long
run results of large devaluations. “Too often the heroics of once-and-for-all
large devaluations have been followed by negligible long-run results, leaving
in their wake widly fluctuating relative prices which entrepreneurs cannot take
seriously as guides to resource allocation.” In connection with the possible
deflationary effects of devaluation, he points out that “import liberalization
can be interpreted as a way to make industry again part of the trade sector,
and as such a potential gainer from future devaluations” and adds that, in many
developing countries, “industry has become a quasi-home goods sector”
2.3.The crawling
peg
Dornbusch (1981, p. 251) concluded that “the crawl was quite
decidly part of the incomes policy”.[25]. His analysis emphasized
that the announcement of the rate of crawl of the effective exchange rate was
made credible by the provision of forward cover in major currencies at the
implicit discount. While this part of the policy package did not last long, it
allows interesting comparisons between the official and private spot and
forward valuations of an inconvertible currency. Quoting again Dornbusch (1981,
p. 244): “Balance of
payments and exchange rate problems were not at the centre of public policy
discussion in 1974-5. There was no experience with problems of external finance
because there had traditionally been a surplus. In addition, of course, there
were ample exchange reserves. More importantly, the social reorganization was
much more exciting than esoteric questions of trade balance adjustment,
competitiveness or exchange rate management. In fact, there was very little
economic management. Mostly macroeconomics happened as a direct outgrowth of
the revolution and disorganization of the public sector. The Banco de Portugal
paid the bills, albeit reluctantly.
By mid-1976 exchange reserves had
declined very substantially, the Banco de Portugal had started selling gold and
the question of exchange rate adjustments to restore external competitiveness
became a policy issue. By September the government, forced by the Central
Bank's refusal to grant further credit, took steps to control wages,
productivity and imports (…) the policy of gradual depreciation was de facto,
unannounced and was certainly aided by the precipitous depreciation of sterling
during l976. (…)The main disadvantages were linked to foreign exchange
speculation on the part of the nationalized banks (!), emigrants delaying
remittances and firms increasing inventories of importable.
Against the background of
dissatisfaction with the gradual depreciation policy and the continued need to
maintain and, indeed, restore external price competitiveness, the effective
escudo rate was depreciated by 12.5 per cent in February of 1977. The Bank
announced at the same time the intention to maintain henceforth the effective
rate constant. By the end of February 1977 the nominal effective rate had
cumulatively depreciated by nearly 30 per cent since 1973 and by nearly 20 per
cent in the last two months.
The rapidly declining foreign
exchange reserves and the illiquidity of the gold reserves forced
In June 1977 a group of 14 countries
joined to fund a further $750 million loan to
The commitment to support domestic
employment left no serious alternative to balance of payments deficits and
their financing. The questions that had to be addressed, though, were the mix
between domestic restraint and deficit financing, the mix between devaluation
and trade controls, and the leeway for interest rate policy. Discussion in the
press turned increasingly to the prospect that the IMF terms of agreement would
almost certainly involve major devaluation – perhaps as much as 20 per cent.
Accordingly it was not surprising that there was a large bulge in the balance
of payments deficit arising from over invoicing, inventory building and any of
a number of ways of speculating on foreign exchange and import prices (…).
There was, after all, a policy of
raising interest rates in the move to the crawl in Autumn 1977 and, of course,
in the agreement with the IMF in Spring 1978. The general rise in interest
rates conceals the substantial selection and credit rationing, favoring exports
and industries with low import content, in particular construction, that are practiced
by the banking system. Most importantly, the crucial role of remittances was recognized
throughout by special deposit arrangements, including foreign currency deposits
for non-residents.
The question of interest
differentials and a credible depreciation target was difficult not only because
of appalling statistics but also because an inflationary bulge from the
February depreciation, decontrol of consumer prices and increased import
restrictions made it difficult to discern the underlying rate of inflation.
Uncertainty about productivity developments made it difficult to estimate the
trend in costs. Under these circumstances there was a temptation to take the
most recent inflation rumors - 30 per cent and rising was a fashionable number
– and base the depreciation path on that assessment. Very fortunately the Bank
took the view that a lower estimate was a more realistic evaluation of the
underlying rate of cost inflation. A l0-15 per cent rate of trend depreciation
was judged adequate to maintain competitiveness without foregoing the benefits
of inflation stabilization.
The problem of identifying an
appropriate trend depreciation highlights an important aspect of the Portuguese
crawl, namely that it was adopted under conditions of high uncertainty and not
in the circumstances of well-defined, steady state inflation differentials.
Moreover, the crawl was quite decidedly part of an incomes policy. In line with
limits on wage increases, the crawl was to stabilize inflation by lending a
credible government commitment – an announced exchange rate target path and
forward markets at those rates – to the incomes policy package. In this respect
the Portuguese crawl was decidedly an unusual experiment”.
My comment also noted that the
success of the Portuguese stabilization program could be seen in the fact that
interest rate increases were not so high as to attract excessive capital
inflows that could have led to real appreciation via the foreign monetary base.
In this connection, the reluctance to liberalize interest rates entirely was
fortunate, even when another channel for interest arbitrage was taken into
account, because, with parity interest rates, the economy would have been vulnerable
to the ”Southern Cone” problem. Noting that substantially smaller average
covered and uncovered deposit rate differentials were reported in the free
forward market in
I concluded that the determination of
the rate of crawl should attempt to insulate the economy from capital account
disturbances, and thus be based on the performance of the reported current
account rather than on inflation differentials. Furthermore, in order to allow
for delays in current account adjustment and introduce some flexibility for
capital account transactions without destabilizing the real exchange rate it
might be desirable to allow a wider band around the crawling peg. Allowing for
“float in order to fix” would also ease the transition to convertibility in the
perspective European integration[26].
Elaborating on the previous point, I
wrote: “it would seem preferable to actually use the expected current account
as an indicator for the rate of crawl, with some weight given to the level of
foreign assets of the banking system. This policy would also make the exchange
rate at which the current account is balanced the long-run equilibrium rate,
but it would allow the capital account to determine the exchange rate in the
short run and thus provide precious information about the type of
convertibility that is best suited to facilitate the catching-up process with
the EEC” [27].
Some years later, I cautioned against
the “soft currency club” option and suggested that the crawling peg was “no
longer altering relative prices” and claimed that “interest rates tend to be
kept low so as to alleviate the burden of public debt” and further predicted
that “to experiment with some form of wider band, as in Italy and Spain, seems
unavoidable. Indeed the pressure on the
Little did I know I would try to do
just that in preparing the 1992 budget, but that the European recession would
undermine the “agreement among the central bank, the ministry of finance and
the spending ministries on a plan of deficit reduction involving both
expenditures and revenues, and perhaps including tax reform” which I claimed
was needed to sustain the next regime change!
3. Float in order to fix: from realignments
to the euro
Entry into the ERM following exit from the crawling peg was as unusual an experiment as the crawl itself. This is especially true because both regimes were seen as intermediate and therefore less credible than a hard peg or a pure float. This view has been challenged (Macedo, Cohen and Reisen 2001, Macedo and Reisen 2004) but may be it also reflects the nature of economic advice.
Table 2 shows the ten steps involved in the regime change, which took from 1989 until 1992, and the thirteen follow-up measures up until the current monetary authorities (eleventh and fourth at the treasury and the bank, respectively). The names of successive Ministers and Governors and a reference to the changes in Prime minister are also presented. Note that, after the completion of the regime change, nine other steps take us to the last realignment of the peseta and the escudo in 2005. Four steps are listed before the euro began circulating in January 2002. Strikingly, there were five Ministers of Finance (and four Prime ministers) in the five years since then.
The regime change began in a way that did not look like it had
anything to do with financial freedom. It was the change in the 1976
constitution, where privatization was forbidden. The constitution was amended
in the summer of 1989, after the first free elections in
The constitutional amendment was postponed several times because it
needed a 2/3 majority in parliament and this required the agreement between the
leaders of the main government and opposition parties. As it turns out this
agreement proved impossible until both were economists who understood the costs
of delaying reform in this area. In the case of Cavaco Silva and Vitor
Constancio, economic advice worked in a top-down manner, but only after two
years of the first majority government since the revolution. This dictated the
speed of the regime change. Right after the amendment, the government started
to work on single financial market legislation and other legal adjustments to currency
stability and convertibility, but it was not approved until 1992 (step 10).
Moreover, currency regime was obscured by the direct effect of the amendment on
the privatization of state owned enterprises, in particular the banks and
insurance companies that had been nationalized on
The rapid
decline in inflation in 1986/87 was accompanied by a multi annual adjustment
program (called PCEDED, or P1) that targeted external balance rather than
budgetary adjustment (as opposed to P2, listed as step 2). It did not imply a
change in regime because there was still no constitutional framework for
private economic activity (step 1) and because the two
steps that followed were not made public[29]. Also the price and wage agreement
of 1988 collapsed because of the rise in inflation. A virtual parity for the
escudo in the ECU basket was decided on
The presentation of the National Framework
for the Transition to Monetary Union (called QUANTUM, or Q1). Similarly, a
series of restrictions on capital inflows implemented by the Banco de Portugal
(whose competencies were reinforced by decree-law during the summer of 1991)
were introduced after Q1 was presented. The refusal to liberalize capital
outflows suggests a reversal in the regime change, especially because it coincided
with the increase in the public service wage grid in 1991. Long term and short
term interest rates rose during the reversal, as did the rate of growth of
wages. The belated change in economic and financial regime might never have
happened if it were not for two economists present at the Gulbenkian conference[30].
Before steps 6, 7 and 8 were completed, entry in to the ERM would not have been
credible[31]..
Table 2: Regime change and its follow up in 33 steps
1. Constitutional amendment, August 1989
2. Revised PCEDED presented to European Commission, November 1989
Miguel Beleza
appointed Minister of Finance January 1990
3. End of crawling peg announced at Monetary Committee, May 1990
4. Public presentation of QUANTUM and beginning of 1st phase of EMU,
July 1990
5. New statutes of Banco de Portugal, November1990
6. Approval of Convergence Program Q2 by ECOFIN,
7. Agreement on price and incomes policy for 1992,
8. Approval of State Budget where financing of Treasury by Banco de
Portugal is prohibited (article nº 58 of law nº 2/92 of 9 de March),
9. Entry of escudo in ERM,
Miguel
Beleza appointed Governor of Banco de Portugal May 1992
10. Structural and financial
policies listed in the Program of XII Government, Convergence Program and State
Budget for 1992: a/ regime of financial management in the public sector
(decree-law nº 155/92, 28 July); b/ rationalization of the use of human
resources in the public sector (decree-law nº 247/92, 7 November); c/ rules
relating to funds for investment, restructuring and internationalization of
enterprises (FRIE, decree-law nº 214/92 de 13 de October, including tax
benefits decree-law nº 289/92, 26 December); d/ general regime of credit
institutions and financial corporations (decree-law nº 298/92, 31 December); e/
implementation by Banco de Portugal of the elimination of exchange controls
announced in August (foregoing a derogation until 1995), 16 December, 1992.
11. Appeal of the Minister of Finance with respect to currency
convertibility, Banco de Portugal, March 1993
12. Realignment of peseta and escudo, May 1993
13. Upgrade of rating of external debt, Standard
& Poor's, May 1993.
14. Widening of fluctuation bands to 15%, August 1993
15. Award for best issuer for the Republic and special comendation for
Minister of Finance, August 1993 (Euromoney)
16. Global issue in dollars for the Republic, September 1993
17. Approval of Revised Convergence Program presented on
Eduardo Catroga
appointed Minister of Finance, December
18. Global issue in ECUS for the Republic, February 1994
António de
Sousa appointed Governor, June
19. Realignment of peseta and escudo,
April 1995
New prime Minister
appoints Sousa Franco Minister of Finance October
20. Name of single
currency agreed, December 1995
21. Portugal qualifies
for euro, May 1997
Pina Moura appointed
Minister of Finance
October 1999
22. Irrevocable
fixing of parities, December 1999
Vitor Constâncio
appointed Governor, February
2000
Oliveira Martins
appointed Minister of Finance July 2001
23. Entry in
circulation of euro, January 2002
New prime Minister appoints Ferreira Leite Minister of
Finance, April
New prime Minister
appoints Bagão Félix Minister of
Finance, July 2004
New prime Minister
appoints Campos e Cunha Minister
of Finance, March 2005
Teixeira dos Santos appointed Minister of
Finance, July
After that, for electoral and other reasons, capital controls were reintroduced which nearly reversed the process of financial liberalization and nominal convergence. In the end the regime change materialized just before the turbulence of the ERM, structural measures had been introduced and the rating was upgraded in spite of the downturn (step 13). The widening of the fluctuation bands (step 14) did not prevent a strong return to international borrowing (steps 15, 16, 18).
Decisions had to be taken about the level at which to join, about
the accompanying measures, when to relax capital controls, and technical advice
on those issues came from DG ECFIN (then DG2). The advisers were embedded to
the extent that the Commission is an international organization and
Different types of loyal advice are appropriate for different
occasions and Serge Kolm, Rudi Dornbusch and Hervé Carré were in their
different ways equally effective. There were different people who had to follow
up on a regime change, so in the end we never knew how the fixed exchange rate
regime or the crawling peg ended, whether it was a voluntary or forced exit,
and, as a consequence, perhaps we did not explain fully entry into the ERM. And
this is why it was only accepted by the population several years after
international financial markets had accepted it. This can be seen in the lower
spread relative to
As I described in Torres (1996), between 1985 and 1995, Prime
Minister Cavaco Silva was a former Finance Minister and he ensured the
strategic dominance himself, alternating between Ministers who earned
credibility abroad at the beginning of the mandate and Ministers who sold
stability at home in the run up to new elections. But there was no constituency
for globalization and a residual fear about “choosing” between
The delegation
to the Minister of Finance remained weak under Pina Moura and Oliveira Martins until
Guterres resigned in the wake of defeat in December
2001 local elections. After the opposition won the 2002 general elections, a majority
coalition government reinforced delegation to Ferreira Leite. She acknowledged
the violation of stability pact and introduced emergency restrictive measures
during the downturn. Delegation
was weakened again by a new government of same coalition, who gave conflicting
signals about pursuing budgetary consolidation, precipitating general election in
February 2005. A majority socialist government led by Prime Minister Socrates gave a weak
delegation to the Minister of Finance (former independent central bank
official) – who resigned after 4 months. In 2006,
the reform rhetoric grew louder based on the strategic cooperation between
President of the Republic and Prime-minister. Nonetheless, the strength of the
delegation to the Minister of Finance has not been tested and it is not known
whether Teixeira dos
Markets
believed in the regime change in 1992 but the population did not believe until 1995
because the government that had criticized the stability cum convertibility regime
change followed the same policy. I will never forget, for obvious reasons, what
the leader of the Socialist Party said on election night: ‘I have just been
talking with all the ambassadors of the European Union reassured them that the
stability oriented regime will continue…’[35]. This stated continuity
was a major victory: the political landslide was a financial nonevent. Monetary
myths are still around, but they are less serious because the consequences of
the regime change are better understood, thanks to economic research,
institutional detail, and simplicity.
The
effects of regime change on financial reputation can be assessed via the
volatility of the exchange rate of the escudo against the numeraire, the Dmark.
Using high-frequency time series methods, it is possible to define regimes
endogenously through the daily conditional volatility of the exchange rate[36].
The generalized turbulence of the parity grid began shortly after the escudo
entered with a 6% band, and evidently the most volatile period lasts until
bands were widened to 15%. The period of Dmark shadowing under inconvertibility,
which followed the end of the crawling peg, was however more volatile than the
period of very wide bands – in spite of a realignment of the escudo and the
peseta in April 1995.
Table 3 Conditional escudo volatility
|
Std deviation % chg exchange rate against the DMark |
(% p. day) |
(% p.week) |
|
Shadow DM (21/9/89-3/4/92) |
0,30 |
0,43 |
|
6% band ERM (4/4/92-2/8/93) |
0,63 |
0,47 |
|
15% band ERM (3/8/93-31/12/98) |
0,21 |
0,28 |
Source:
Macedo (2007) and references therein.
Focusing
on the period since the escudo defined a virtual parity in the ERM grid, there
were three regimes, shadowing of the Dmark, 6% and 15% ERM bands. The previous
results are confirmed using intervention data for this period, which is shown
to be effective only under low volatility until the bands were widened to 15%,
i.e. intervention is the more effective the less it is needed. Under high
volatility,
This
is confirmed by a measure of exchange market pressure weighting changes in the
exchange rate, reserves and the interest rate differential (Pereira 2005). The
dominant characteristic of exchange market pressure under shadowing was a 1%
appreciation, during the 6% band it was a 4% depreciation and during the 15%
band a 1% depreciation.
Mean
reversion is also more pronounced during the latter phase, as you need to float
in order to fix. In any event external credibility was reached end 1992 with
full convertibility and confirmed after the widening of the bands in August
1993, even though another realignment was going to take place before the
creation of the euro, bringing to four the totals of realignments of the peseta
involving at least partly the escudo. Table 4 shows that contagion only
occurred from peseta to escudo but not the other way around.
This
asymmetry would be even stronger if the escudo were not in the ERM, since even
when the two currencies are in, an increase in the interest rate differential
when
The
fears about “geographic fundamentals” often expressed in this context suggest
that for the escudo to always follow the peseta would have prevented any
positive differentiation, whereas ignoring the peseta would not have been
credible. This suggests in turn that having remained outside of the ERM during
the turbulent period would have been likely to exacerbate contagion, or require
a drastic tightening of capital controls instead of their dismantling before
the end of 1992. There is no evidence that the drachma might have a similar
contagion problem, so that ERM entry for the drachma was not as urgent as it
was for the escudo.
Table 4
Contagion between escudo and peseta in a model with variable correlations and
volatility states determined by the interest differential with
|
Effect of
differential on probability of currency going from high to low volatilitye |
Correlation
varies with escudo |
Correlation
varies with peseta |
|
Escudo |
-0,24 |
-0,19 |
|
Peseta |
0,28 |
0,36 |
|
Correlation low
volatilitye |
0,71 |
0,61 |
|
Correlation
high volatilitye |
0,68 |
0,76 |
Source: Macedo
(2007) adapted from Lopes (2006). All coefficients significant at 5%.
4. Advice on entry vs advice on exit
Table
1 above listed three changes in exchange rate regime before the stable escudo
was devalued, a crawling peg implemented and stability and convertibility were
restored in the run-up to the euro. The first regime change involved the
creation of the real in
1435 as a national currency based on gold which replaced the debased
libra. Between 1501 and the Napoleonic wars,
when sterling and the real became inconvertible,
Advice
on the change in the exchange rate regime may not be the one people talk about
these days, but, once again, it goes back to the 1850s. As part of a so-called
regeneration program, a substantial debate took place then about the accession
of the real to the gold standard.
The change in the name of the
national currency after the 1910 Republican revolution (1 escudo=1000 reis) did not involve a change in regime
but only the promise to restore convertibility into gold. Instead a
period of maximum divergence ensued with hyperinflation in the aftermath of the
First World War. In the late 1920s, there was an attempt to
the return to gold parity. And that, again, even though it was done under an
authoritarian regime, involved a lot of consultation with employers’
associations, commerce, industry, and so on. So the stakeholders in society
have been involved in these regime changes. The second entry into the gold
standard in July 1931 was also carefully planned but there was an external
shock and we left with sterling , 82 days later (
A
lot of what we heard today about the importance of product market competition
(Silva Lopes was emphatic on that), the policy design being broad based rather
than a succession of individual sectoral reforms, is something that is well
accepted, and we are also witnessing a flurry of advisors. When Dani Rodrik
comes to town, he places a lot of emphasis on technological changes, and on
getting things to work better or improve the productivity[37].
Instead
of just talking about economic institutions, which is almost as broad as social
capital, Orsetta Causa and Daniel Cohen (2006) distinguish, aside from the
usual factors of production, the role of infrastructure, on the one hand, and
the role of integration with the international trading system. It turns out
that, with respect to those two dimensions,
Talking
about creative destruction at the technological frontier is also very fitting
in this context, and I want to close a little bit on the pedagogic side. In my
development course, I present the Krugman (1991) “history versus expectations”
model: when there are multiple equilibrium development paths, a country can escape
being the prisoner of history, of past policies and achievements by following
an expectations-driven path depending on three observable parameters. One parameter,
which has to do with intertemporal trade, is the interest rate (Olivier
Blanchard 2006 is eloquent about that). It must be lower than the combined
effect of increasing returns to scale and speed of adjustment. Aside from
technology, then, we find flexibility as a guide for institutional change[39].
Flexibility
includes the ability to exploit the complementarity between macroeconomic and structural reforms[40].
Yet
there is relatively little attention to the design of reform packages and its
effects on economic growth, suggesting that, when there
are many distortions, eliminating only few of them may threaten the
sustainability of the reform process. Reforms are more likely to fall prey to
the second-best argument under ready-made policy packages with scant knowledge
about local conditions. A more systemic approach to national economies will
include the concept of complementarity as an input into economic advice. When
starting a reform strategy that deliberately results, in its initial stages, in
a reduction of economic coherence, countries incur a risk. While it may be
rational to bear that risk, it cannot be systematically ignored without facing
long-lasting negative consequences[41].
This
is never easy to change, but if you can convince people that changing is
something that will have to happen, either in a gradual way, as the regime
change of 1989-92, or in a crisis, like the exits of 1891, 1931 or 1977, then
you are coming ahead. And this is one of the roles of advising, it is to
explain exactly what flexibility can do, at no cost, so-called Pareto moves. Joining
the ERM was a good entry into the euro, I hope there will be no exit.
References
Abel, Andrew; Beleza, Miguel; Frankel Jefferson (sic); Hill Raymond; Krugman Paul (1976), A model of the Portuguese economy, Economia I (1), pp.113-120.
Abel, Andrew; Beleza,
Miguel Frankel Jeffrey Hill Raymond Krugman Paul (1977) A Economia Portuguesa: Evolução Recente e
Situação Actual, Conferência,
pp. 31-93.
Abreu,
Alesina, Alberto and
Francesco Giavazzi (2006), The future of
Blanchard, Olivier
(2006), Adjustment within the euro: The difficult case of
Bliss, Christopher and
Macedo, Jorge Braga de, editors
(1990), Unity with Diversity in the European Economy: The Community’s
southern frontier,
Causa Orsetta and Cohen Daniel (2006), The
Ladder of Competitiveness how to climb it,
Cecco, Marcello de (1983), editor, International Economic Adjustment: Small Countries
and the European Monetary System,
Conferência Internacional sobre Economia
Portuguesa (1977),
The German Marshall Fund of the United States Fundação Calouste Gulbenkian,
Lisboa (2 vols).
Conferência Internacional sobre Economia
Portuguesa (1979),
The German Marshall Fund of the United States Fundação Calouste Gulbenkian,
Lisboa (2 vols).
Cooper Richard (1971) Currency Devaluation in Developing Countries in Ranis,
pp. 472-512.
Díaz-Alejandro, Carlos (1971) Comment on Cooper in Ranis,
pp. 513-515.
Dornbush, Rudiger (1981),
Dornbush, Rudiger (1981), Open Economy Macroeconomics,
Eichengreen, Barry and David Leblang, Democracy and Globalization
NBER Working Paper 12450 July 2006
Entrevista ao grupo do MIT, Nação e Defesa, ano 1, Novembro 1976, pp.
185-195.
Esteves Rui, Reis Jaime and Ferramosca
Fabiano (2006), Market Integration
in the Golden periphery The
European Commission
(2006), Public Finances in
2006, European Economy
Griffith-Jones, Stephany; Montes Manuel and Nasution
Anwar, editors (2001), Short-Term Capital Flows and Economic
Crises, Oxford University Press, pp.253-260
Hagen, Jurgen von (2003), Fiscal Discipline and Growth in Euroland: Experience with the Stability and Growth Pact, ZEI Working Paper B 06 2003.
Hall, Robert and Jones Charles (1999), Why do some countries produce
some much more output per worker than others,
Quarterly Journal of Economics
114 (1), pp. 83-116
Hallerberg, Mark Strauch Rolf and Hagen Jürgen von, The design of fiscal rules and forms of governance in European Union countries, European Central Bank Working Paper Series, No. 419 / December 2004
International Monetary
Fund, Euroarea
policies: selected issues Country Report No. 04/235, August
Kolm, Serge C. (1977) La Transition
socialiste la politique economique de gauche, Paris: Editions du Cerf
Kouri, Pentti (1979), Profitability and
Growth in a Small Open Economy, in Lindbeck
pp. 129-142.
Krugman, Paul, History
vs Expectations, Quarterly Journal of Economics, May 1991.
Krugman, Paul and Macedo, Jorge Braga de (1979), The Economic Consequences of the April 25th Revolution, Economia, III (3), pp. 435-483
Landersmann, Michael and Rosati Darius, editors (2004), Shaping the New Europe: Economic Policy Challenges of EU Enlargement, Palgrave, pp. 27-48.
Lindbeck, Assar (1979) editor, Inflation and Unemployment in Open
Economies,
Lopes, José Mário (2006), The Portuguese Escudo in the EMS: Crisis and Contagion, mimeo,
Faculdade de Economia, Universidade Nova de Lisboa.
Macedo, Jorge Braga de (1980), Portuguese
Currency Experience: An Historical Perspective, in Estudos em
Homenagem a J. J. Teixeira Ribeiro, vol. IV, Coimbra: Boletim da Faculdade
de Direito.
Macedo, Jorge Braga de (1981), Comment on Dornbusch
in Williamson pp. 272-278.
Macedo, Jorge Braga de (1982), Portfolio Diversification and
Currency Inconvertibility Three essays in international monetary economics,
Universidade Nova de Lisboa Press.
Macedo, Jorge Braga de (1983), A portfolio
model of an inconvertible currency. The recent
experience of
Macedo, Jorge Braga de (2001), Globalisation and Institutional Change: a development
perspective,
in Malinvaud and Sabourin, pp. 223-268
Macedo, Jorge Braga de (2007), A mudança do
regime cambial português: Um balanço 15 anos depois de Maastricht, Nova Economics Working Paper n.o 502,
February.
Macedo, Jorge Braga de, Cohen, Daniel and Reisen, Helmut (2001), editors, Don´t fix don´t float, OECD Development Centre Study.
Macedo, Jorge Braga de, Eichengreen, Barry e Reis Jaime (1996) editors Currency
Convertibility: The Gold Standard and Beyond,
Macedo, Jorge Braga de and Joaquim Oliveira Martins,
Growth, Reform
indicators and Policy Complementarities (2006),
NBER Working Paper no 12544, September
Macedo, Jorge Braga de
and Reisen, Helmut (2004), Float in order to fix? Lessons from emerging markets for new EU member countries in Szapary, Gyorgy and Jurgen von Hagen, editors, Monetary strategies
for joining the euro, EdgarElgar: National Bank of
Macedo, Jorge Braga de, Álvaro Ferreira da Silva e Rita Martins de Sousa (2001), War, taxes and gold: the inheritance of the real, Transferring Wealth & Power from the Old to the New World, edited by Michael Bordo and Roberto Cortes-Conde, Cambridge: Cambridge University Press, pp. 187-228.
Malinvaud, Edmond and Sabourin, Louis
(2001) editors, Globalisation, Ethical and Institutional Concerns, Pontifical
Academy of the Social Sciences.
Pereira, Luís Brites (2004) The Effectiveness
of Intervention: Evidence from a Markov-Switching Analysis, mimeo, Faculdade de Economia,
Universidade Nova de Lisboa.
Pereira, Luís Brites (2005) Relieving Exchange
Market Pressure: When is Central Bank Intervention Effective? mimeo,
Faculdade de Economia, Universidade Nova de Lisboa.
Ranis, Gustav, editor, Government and Economic Development,
Reis Jaime (1996), First to join the gold standard, 1854, in Macedo, Eichengreen and Reis (1996), 159-181.
Rodrik, Dani (2006), Industrial Development: Stylized facts and policies, mimeo,
Sérgio, Rui (1977), Experiments with Investments Strategies. Preliminary Results of a Numerical Model, in Conferencia, pp. 755-832.
Startakis, Georges and Vaggi, Gianni, editors (2006), Economic Development and Social Change, Routledge.
Tommasi, Mariano (2002), Crisis, Political
Institutions, and Policy Reform. It is not the Policy, it is the Polity, Stupid!,
presented at Annual Bank Conference on Development Economics (ABCDE)
Torres Francisco, editor
(1996), Monetary Reform in Europa, Lisboa: Universidade Católica
Portuguesa.
Williamson John (1981), editor Exchange Rate Rules: The
Theory, Performance and Prospects of the Crawling Peg,
Wynne, W. H. (l95l), State Insolvency
and Foreign Bondholders, Vol. III, Yale University Press.