The gold standard, the euro, and the origins of the Greek sovereign debt crisis.
Dellas, Harris ; Tavlas, George S.
The planners of a European monetary union would be well advised to
study the reasons the pre-World War I gold standard was a successful
monetary regime.
--Anna J. Schwartz (1993)
The entry of Greece into the eurozone in 2001 was widely expected
to mark a transformation in the country's economic destiny. During
the decade of the 1980s, and for much of the 1990s, the economy had been
saddled with double-digit inflation rates, double-digit fiscal deficits
(as a percentage of GDP), large current-account imbalances, very low
growth rates, and a series of exchange rate crises. Adoption of the
euro--the value of which was underpinned by the monetary policy of the
European Central Bank (ECB)--was expected to produce a low-inflation
environment, contributing to lower nominal interest rates and longer
economic horizons, thereby encouraging private investment and economic
growth. The elimination of nominal exchange-rate fluctuations among the
former currencies of members of the eurozone was expected to reduce
exchange rate uncertainty and risk premia, lowering the costs of
servicing the public sector debt, facilitating fiscal adjustment, and
freeing resources for other uses.
And that is precisely what happened--at least for a while. In the
years immediately prior to and immediately after Greece's entry
into the eurozone, nominal and real interest rates came down sharply,
contributing to high real growth rates. From 2001 through 2008, real GDP
rose by an average rate of 3.9 percent per year--the second-highest
growth rate (after that of Ireland) in the eurozone. Inflation, which
averaged almost 10 percent in the decade prior to eurozone entry,
averaged only 3.4 percent over the period 2001-08. Then, beginning in
2009, everything changed as Greece became the center of a major
financial crisis. Interest rates on long-term government debt soared
from the low single digits prior to the crisis to a peak of 42 percent
in early 2012; the country had to resort to two successive adjustment
programs (in May 2010 and March 2012) with official international
lenders; and the Greek government restructured its debt. Between the end
of 2008 and mid-2012, the economy contracted by a cumulative 20 percent
(and it continues to contract), and the unemployment rate jumped from
less than 8 percent to about 25 percent. Like Odysseus's return
trip home from the Trojan War, the road to Ithaca led to a Tartarean
hell.
What happened? And why did it happen? To answer these questions, we
begin by describing the origins of the Greek financial crisis,
highlighting the crucial role of growing fiscal and external imbalances.
Next, we identify what we believe was a key factor that abetted those
imbalances--namely, the absence of an automatic eurozone adjustment
mechanism to reduce members' external imbalances. To illustrate our
argument, we compare the adjustment mechanism in the eurozone with the
adjustment mechanism for the participants of the classical gold-standard
regime of the late 19th and early 20th centuries. Are there major
differences between the working of the gold-standard adjustment
mechanism and that of the eurozone? What are the lessons that can be
drawn from a comparison between the gold standard and the eurozone? We
address these questions in what follows.
[FIGURE 1 OMITTED]
The Years of Living Dangerously
As mentioned, Greek interest rates came down sharply in the years
immediately prior to, and immediately after, the country's entry
into the eurozone. Figure 1 shows the monthly interest-rate spread
between 10-year Greek and German government bonds for the period
1998-2012. (1) The spread fell steadily, from over 600 basis points in
early 1998 to about 100 basis points one year prior to Greece's
eurozone entry. By the time Greece entered the eurozone in 2001, the
spread had fallen to around 50 basis points; it continued to narrow
subsequently, declining to between 10 and 30 basis points from late
"2002 until the end of 2007. During the latter period, the absolute
levels of nominal interest rates on the 10-year Greek instrument
fluctuated in a range of 3.5-4.5 percent, compared with a range of
5.0-6.5 percent in the year prior to eurozone entry.
Although entry to the eurozone contributed to a period of low
interest rates and rapid real growth, deep-seated problems in the Greek
economy remained unaddressed, reflecting a procyclical fiscal policy; as
a result, the country continued to run large fiscal and external
deficits. Figures 2 and 3 show data on fiscal deficits and government
debt as a percentage of GDP. Several features stand out with regard to
the period 2001-09. First, fiscal deficits consistently exceeded the
Stability and Growth Pact's limit of 3 percent of GDP during the
entire period, rising to 9.8 percent of GDP in 2008 and 15.6 percent of
GDP in 2009. (2) Second, the widening of the deficits was mainly
expenditure-driven; between 2005 and 2009 the share of government
spending in GDP rose by 9 percentage points (to 54 percent), with the
bulk of the rise occurring between 2006 and 2009, a period that featured
a government run by a conservative party. Third, beginning in 2007, the
deficits underpinned an unsustainable increase in the government
debt-to-GDP ratio, culminating in the crisis that erupted in late 2009.
The large and widening fiscal deficits contributed to growing
current account deficits. There are two main series on the Greek current
account. One is compiled by the Bank of Greece, based on information on
international transactions reported by commercial banks. The other, used
in the national accounts and by the European Union, is derived from
customs information, (3) Both series are plotted in Figure 4. Both show
that the deficit was large (in relation to GDP) upon Greece's entry
in the eurozone, and grew even larger in the following years. The Bank
of Greece data show that the current account deficit rose from about 7
percent of GDP in 2001 to almost 15 percent in 2008, before declining to
about 14 percent in 2009. The national account series shows the current
account deficit rising from 11.5 percent of GDP in 2001 to almost 18
percent in 2008, before declining to about 15 percent in 2009.
[FIGURE 4 OMITTED]
Figure 5 compares the current account positions of Greece, Germany
(the center country of the eurozone), and the eurozone as a whole, based
on national accounts data to ensure consistency. The reason that we
compare Greece with Germany will become clear later when we discuss the
adjustment mechanism in the eurozone. Two points are important to
mention. First, during the period 2001-09 the current account of the
eurozone as a whole was roughly in balance. Second, the current account
of Germany went from essentially a balanced position in 2001 to a
surplus of around 6 percent of GDP in 2008, a swing of some 6 percentage
points, almost the same percentage as the increase in Greece's
current account deficit during the same period. (4)
[FIGURE 5 OMITTED]
Figures 6 and 7 show the relative contributions of the public and
private sectors, respectively, to the evolution of the current account
balances of Greece, Germany, and the eurozone as a whole. Again, several
points stand out. In the case of Greece, the widening of the current
account deficit was caused entirely by the behavior of the public
sector; (5) net public saving (relative to GDP) fell from around minus
four percent in 2001, to minus 15 percent of GDP in 2009. During the
same period, net private saving (relative to GDP) in Greece rose, from
about minus 7 percent to around minus 1 percent of GDP. For Germany and
the eurozone as a whole net public saving increased from 2001 through
2007, before declining in 2008; net private saving rose in both Germany
and the eurozone as a whole.
[FIGURE 6 OMITTED]
[FIGURE 7 OMITTED]
Eruption
The global financial crisis that erupted in August 2007, following
the collapse of the U.S. subprime mortgage market, initially had little
impact on Greek financial markets. Spreads on the 10-year instrument,
which were in a range of 20-30 basis points during January-July of 2007,
remained in the vicinity of 30 basis points for the remainder of 2007
and the first few months of 2008 (Figure 1). With the collapse (and
sale) of Bear Stearns in March 2008, spreads widened to about 60 basis
points, where they remained until the collapse of Lehman Brothers in
September of 2008. The latter event brought spreads up to around 250
basis points during the first few months of 2009, but they gradually
came back down to about 120 basis points in August and September of
2009.
Then came a double shock in the autumn of 2009. Two developments
combined to disrupt the relative tranquility of Greek financial markets.
First, in October the newly elected Greek government announced that the
2009 fiscal deficit would be 12.7 percent of GDP, more than double the
previous government's projection of 6.0 percent. In turn, the 12.7
percent figure would undergo further upward revisions, so that the
outcome was a deficit of 15.6 percent of GDP. Second, in November 2009
DubaiWorld, the conglomerate owned by the government of the Gulf
emirate, asked creditors for a six-month debt moratorium. That news
rattled financial markets around the world and led to a sharp increase
in risk aversion. In light of the rapid worsening of the fiscal
situation in Greece, financial markets mad rating agencies turned their
attention to the sustainability of Greece's fiscal and external
imbalances. The previously held notion that membership of the eurozone
would provide an impenetrable barrier against risk was destroyed. It
became clear that, while such membership provides protection against
exchange-rate risk, it cannot provide protection against credit risk.
The two shocks set off a sharp and prolonged rise in spreads, which
continued into early 2012. As shown in Figure 1, the spread on the
10-year sovereign widened from about 130 basis points in October 2009,
to around 900 basis points one year later. (6) The widening took place
despite a May 2010 agreement between the Greek government and the
International Monetary Fund, the European Central Bank, and the European
Commission for a 3-year 110 billion [euro] adjustment program. By early
2012, the Greek-German spread had widened to about 4,000 basis points
and it became clear that Greece's rising debt burden was no longer
sustainable. In March 2012, the Greek government agreed to a new 130
billion [euro] adjustment program with the official lenders, and the
country restructured its sovereign debt.
To summarize the above discussion, Greece entered the eurozone with
a current account deficit of about 10 percent of GDP (depending on the
current account series used); the current account deficit widened
sharply in the following years, leading to the crisis that erupted in
late 2009. The major factor underpinning Greece's large and growing
current-account deficits was the decline in net public saving. Germany,
in contrast, went from a current account deficit of about 1 percent of
GDP in the early 2000s to a surplus of about 6 percent by the time of
the outbreak of the Greek crisis. As we will see, changes in the
relative competitive positions of Greece and Germany help explain the
movements of the external positions of the two countries. A question
that arises is how a country like Greece, which entered Europe's
monetary union with an external deficit of about 10 percent of GDP, was
able to function for many years without adjusting its external position.
To address that question, we turn to a comparison of the adjustment
mechanisms under the classical gold standard and the euro.
The Gold Standard and the Euro
A prominent feature of recent discussions about the eurozone has
been a comparison of the functioning of that area's fixed exchange
rate regime with that of the gold standard of the late-19th and
early-20th centuries (see Buttonwood 2010, Eichengreen and Temin 2010,
Boone and Johnson 2012, James 2012). In this connection, James (2012: 1)
stated: "The European Monetary Union, as many of its critics
maintain, looks a lot like the pre-1913 gold standard, which imposed
fixed exchange rates on extremely diverse economies." How relevant
is the gold standard metaphor? (7) In this section, we briefly describe
key characteristics of the gold standard of the late 19th and early 20th
centuries, and compare adjustment to external imbalances between the two
regimes during noncrises periods. The latter comparison, in particular,
will help shed light on the underlying origins of the present crisis in
the eurozone.
What Was the Classical Gold Standard?
Essentially, the gold standard was a regime under which the
exchange rates of the participating countries moved within narrow limits
approximating their respective gold points without the support of
exchange restrictions, import quotas, or related controls. (8) The
authorities of the participating countries maintained these fixed prices
by being willing to buy or sell gold on demand at that fixed price. The
classical gold standard, under which the circulation of domestic
currencies was tied (to varying degrees) to gold, and international
settlements were made primarily in gold (and, to a lesser extent, in
pound sterling), prevailed in its most pure form during the period from
1880 to 1913 (see Bloomfield 1959: 9, Bordo 1981: 2, Eichengreen 1996:
42). (9)
Bordo (1981, 1993) compared the performance of the classical gold
standard with its successor regimes, including the gold exchange
standard that operated from the mid-1920s until the mid-1930s, the
Bretton Woods regime that operated from the mid-1940s until the early
1970s, and the managed float that began in the early 1970s. Overall,
Bordo (1993: 182) found that the classical gold standard performed less
well relative to other regimes in terms of the stability of real
variables but achieved the lowest rate of inflation and the highest
degrees of inflation and interest-rate convergence, raising the
following question: "Why was ... the classical gold standard so
unstable [in terms of real variables] yet so durable?"
Three (interrelated) characteristics of the classical gold standard
appear to have contributed to its durability. First, it
"facilitated adjustments to balance of payments
disequilibrium" (Bloomfield 1959: 22; see, also, Eichengreen 1992:
29-66). Second, it operated with "virtually no instances of major
or sustained 'runs' on any of [the] currencies [of the leading
participating countries]" (Bloomfield 1959: 21); indeed,
devaluations of currencies on the gold standard "were highly
exceptional" (Bloomfield 1959: 2). Third, it provided "an
effective defense against [inflationary policies] of a kind that time
and again [had previously] led to the debasement mad depreciation of
once-proud currencies" (Friedman 1953: 179).
What countries were members of the classical gold standard club?
(10) Historians distinguish between core members of the club and
peripheral members. Core countries included France, Germany, and the
United Kingdom, with Belgium and the Netherlands also sometimes
considered part of the core category. Among the common features of these
countries are that each had relatively well-developed financial markets
and a central bank. The periphery included Canada, South Africa, the
United States, and parts of Latin America (Argentina, Brazil, Mexico),
Asia (including Australia, New Zealand), and Europe (Austria-Hungary,
Greece, Italy, Spain, Portugal, Russia, Switzerland, and the
Scandinavian countries). (11) Their economies were typically financially
less developed than those of the core countries and most of them did not
have a central bank during at least part of the gold standard period.
Some of the peripheral countries participated in the gold standard
only during part of the 1880-1913 period. Additionally, some countries
that were not formally on the gold standard nevertheless followed
policies that were consistent with a fixed price of their currencies
against gold in an effort to "shadow" the gold standard. With
regard to countries that are sometimes considered to have been members
of the European periphery, the following particular circumstances merit
comment.
* Greece joined the gold standard in January 1885 but dropped out
in September of 1885, because, as Lazaretou (2004: 14) noted, the
government failed to control the fiscal deficits and thus to support the
credibility of the system. It rejoined the gold standard in 1910. Given
the very limited duration of its participation in the gold standard, and
the inconsistency of its policies with such participation, in what
follows we do not consider Greece to have been a member of the
periphery.
* Italy joined the gold standard in March 1883 but dropped out in
February 1894 (Fratianni and Spinelli 1984). Following a period of
floating exchange rates from 1894 until 1902, the monetary authorities
shadowed the gold standard. As Bordo and Kydland (1996: 78) argue,
"The monetary authorities acted as if they were on the gold
standard.... Money growth was low and the budget was often in
surplus." Thus, Italy's economic indicators were in line with
those of full-time periphery members. (12)
* Portugal was a member of the gold standard from 1854 until 1891;
its departure from the gold standard in 1891 was, in part, related to
domestic political instability, following a failed attempt to establish
a Republic (Duarte and Andrade 2004). After its departure from the gold
standard, Portugal shadowed the gold standard, without committing to it
(Soto 1999: 468).
* Spain was a member of the gold standard from 1874 until 1883; it
suspended gold convertibility in 1883, in the aftermath of a sovereign
crisis (Martin-Acena 1994: 13647). Spain then pegged its currency to the
pound sterling in an attempt to shadow the gold standard (see Soto 1999:
468). (13)
The Adjustment Mechanism
The gold standard proved to be a remarkably durable regime, with
periods of tranquility tar outlasting crisis periods, while the
tranquility of the eurozone has proved to be short-lived. What accounted
for the gold standard's tranquility? In answering this question, we
must distinguish between the operation of the gold standard at its core
and at its periphery.
The Core--For its core participants, the gold standard possessed an
adjustment mechanism that served to reduce external imbalances (Scammell
1965, Eichengreen 1996). Consider first the operation of the gold
standard in the absence of capital flows. To simplify the discussion,
let us assume a two-country world comprised of Greece and Germany, in
which Greece runs a trade deficit and Germany runs a trade surplus. Let
us also assume that only gold coins circulate and prices and wages are
flexible in both countries. In such a situation, the gold standard
adjustment process--called the price-specie-flow mechanism--worked as
follows:
* Greece experiences a gold outflow, decreasing the money supply
and reducing credit growth (perhaps reducing the quantity of credit) in
that country, causing prices and wages to fall.
* Germany experiences a gold inflow, increasing the money supply
and raising credit growth in that country, causing prices and wages to
rise.
* As a result of the change in relative prices, Greece's
exports rise and its imports fall, eliminating its trade deficit. The
opposite occurs in Germany.
Capital flows reinforced the overall self-equilibrating character
of the system as it operated in the late 19th and early 20th centuries.
Typically, the central bank of a country experiencing a trade deficit
would increase its discount rate, reducing its holdings of domestic
interest-bearing assets and drawing cash from the market (Eichengreen
1996: 28). This action produced two main effects. First, the money
supply and credit growth in the country raising rates declined, reducing
(or eliminating) the need of gold outflows from that country. In fact,
capital could flow into the country as a result of the rise in the
discount rate, smoothing the required adjustment. Second, the rise in
interest rates would reduce economic activity in the country concerned,
decreasing prices and, thereby, contributing to the elimination of the
country's external imbalance, through both relative-price
adjustment and the decrease in aggregate demand.
The pre-World War I gold standard operated in the above manner
among the core countries (Scammell 1965: 35). These countries possessed
the institutional capacity to make their commitment to the gold standard
credible. Thus, they were able to issue debt denominated in their own
currencies, each of which represented a certain amount of gold.
What did this institutional capacity comprise? Effectively, it
included the following elements. First, the requirements of fixed
exchange rates and free convertibility dominated the requirement of
domestic economic stability--that is, external balance took precedence
over internal balance (Friedman 1953: 166-67). Because there was no
well-articulated theory connecting changes in monetary policy with
domestic economic conditions, there was little or no pressure on central
banks to adjust interest rates in response to changes in those
conditions (Eichengreen 1992: 30). Consequently, economic agents were
assured that the authorities would take any necessary actions to restore
external equilibrium without the need to adjust the nominal exchange
rate or to restrict convertibility. Second, the monetary authorities
operated within an environment in which "the public sector was in
general only a small one, where fiscal policy and debt management policy
in their modern sense were virtually unknown, and where government
budgets were for the most part in balance" (Bloomfield 1959: 20).
(14) Third, should the budget fall into a deficit there was no question
of the authorities' commitment to restore balanced budgets, if need
be, by raising tax rates. This commitment was rendered credible by their
institutional capacity to raise taxes--that is, the core countries
possessed strong legal frameworks, well-developed public
administrations, and efficient bureaucracies. As a result of their
credible commitment to sound finance, the core economies could run small
budget deficits, if needed, to respond to extraordinary shocks without
raising concerns about their ability to service their debts.
Because there was no question about the authorities'
commitment to (1) do whatever it takes to maintain the price of gold and
convertibility, and (2) balance the budget, for a core country facing an
incipient crisis, "capital flowed in quickly and in significant
quantities," mitigating the crisis (Eichengreen 1996: 31; see,
'also, Friedman 1953: 186). Since production by governments was
mainly composed of nontraded goods and services, the small size of the
public sector typically meant that the nontraded goods sector was a
relatively small one. Consequently, a given size of an internal
devaluation would have a relatively large impact on traded goods,
increasing exports, reducing imports, and facilitating external
adjustment. (15)
The Peripheny--In contrast, the situation among the economies in
the periphery varied. These economies were financially less developed
and, therefore, needed access to international financial markets in
order to finance both private and public investments. At the same time,
their fiscal policy institutions lacked credibility mad international
investors were reluctant to lend to them at low interest rates without
gold or foreign exchange clauses in loan contracts (Eichengreen and
Hausmann 1999). Thus, for the capital-scarce peripheral economies,
participation in a system of hard pegs, such as the gold standard,
addressed the problem of dynamic inconsistency in monetary policy and
acted as a "good housekeeping seal of approval," providing
them access to international capital markets at lower interest rates
than would otherwise have been the case (Bordo mad Rockoff 1996,
Obstfeld and Taylor 2003). However, external shocks (mainly to commodity
prices) and (especially) domestic fiscal shocks sometimes triggered a
sudden stop to capital inflows to those countries, leading to currency
and sometimes debt crises.
The frequency of external debt defaults was much higher for the
Latin American peripheral participants, which had a proclivity for
following expansionary fiscal policies, than it was for the European
peripheral members, which typically ran relatively small fiscal
deficits. (16) In this connection, Beinhart and Bogoff (2011: 91)
reported three cases of external debt default among European peripheral
countries during the period 1880 to 1913: Spain in 1882, Russia in 1885,
and Portugal in 1890. (17) Reinhart and Bogoff also reported 19 cases of
default among the Latin American periphery during the same period.
Table 1 reports data on fiscal balances, government spending, and
current account balances as percentages of GDP for four European core
countries (France, Germany, the Netherlands, and the United Kingdom) and
three European peripheral countries that participated in the gold
standard for most of the period 1880 to 1913 (Denmark, Norway, and
Sweden) along with occasional gold standard participants (Italy, Spain,
and Greece). (18) The data are averages over the period 1880 to 1913.
Several features of Table 1 are important to mention. First, the
fiscal balances were in most cases essentially in balance. Second, in
most cases the share of government spending was 10 percent or less of
GDP. Third, the core countries ran current account surpluses (on
average), while the peripheral countries typically ran current account
deficits (on average). However, the current account deficits of the
peripheral countries typically were small; for each of the peripheral
countries reported in the table, the current account deficit averaged
below 3 percent of GDP. (19) Fourth, the major exception is the case of
Greece, which, in an effort to build its infrastructure following
centuries of occupation, consistently ran large fiscal and external
deficits. It is no coincidence, therefore, that Greece, in contrast to
the members of the gold standard club, suffered a series of sovereign
debt and currency crises under the gold standard.
Figures 8.1 and 8.2 show (long-term) interest-rate spreads against
the United Kingdom for three flirt-time peripheral countries--Denmark,
Norway, and Sweden--along with Greece, Italy, and Spain during the
period 1880-1913. Again, several points are important to highlight.
Typically, spreads under the gold standard were fairly large--in the
range of 100 to 400 basis points--despite the small external and fiscal
imbalances of the participating countries. The case of Greece is the
exception that proves the rule. It was not a member of the gold standard
for most of the gold standard period, and its spreads were far above
those of other European countries during the gold standard years. It
also had by far the largest imbalances mad the largest government
sector.
The upshot of the data on spreads is that investors drew a
distinction between the sovereign risk of the core country--the United
Kingdom--and the sovereign risk attached to the debt of the periphery.
The latter debt carried a risk premium, despite the mostly sound fiscal
and current account fundamentals of the countries concerned. The
knowledge that sovereign credit risk--and, therefore, spreads--would
rise if imbalances rose limited the size and persistence of the
imbalances. In turn, these small imbalances underpinned the durability
of the system.
[FIGURE 8.1 OMITTED]
[FIGURE 8.2 OMITTED]
To summarize, a key feature of the classical gold standard was the
distinction between the operation of the system in the core countries
and its operation in the periphery.
* In the core countries, government budgets were, for the most
part, in balance. The adjustment mechanism operated through the
price-specie-flow mechanism to restore external equilibria. This
mechanism was reinforced, if need be, by the counter-cyclical policies
of national central banks. As a result of the authorities'
commitment to maintain balanced budgets and to restore external
equilibria, the regime proved to be durable, with virtually no sustained
runs on any of the core currencies.
* In the European periphery, attacks on currencies and sovereign
defaults among the countries of the European periphery were rather
infrequent events, reflecting the small external imbalances of these
countries, the size and persistence of which was limited by three
factors. First, fiscal shocks were relatively small in comparison to
such shocks today because of the small size of governments. Second, the
adjustment mechanism for the core countries was also more or less
operational for the European peripheral countries, limiting the size and
persistence of external deficits. Third, as reflected in the spreads on
interest rates, there was no expectation that the core countries would
step in to provide debt relief to a heavily indebted peripheral country.
Therefore, the debt of the European periphery was considered to contain
sovereign risk, limiting the demand for such debt by foreign investors.
Consequently, European peripheral countries found it difficult to
finance large current account imbalances on a sustained basis.
* The experiences of Greece, which, as noted, was not a member of
the gold standard club for most of the gold standard period, and the
Latin American peripheral countries differed from those of the full-time
European members of the club. In particular, Greece and the Latin
American peripheral countries experienced frequent sovereign defaults
and currency crises, typically triggered by domestic fiscal shocks.
A key implication of the experiences of the participants in the
gold standard is the following: Adherence to a hard peg is no panacea
and cannot be sustained without the support of credible fiscal
institutions. (20)
The Euro Standard
As mentioned, the effective functioning of the gold standard
required a high degree of wage-price flexibility so that adjustment to
external imbalances could take place. To bring about the necessary
changes in wages and prices, money and credit flowed from countries
experiencing external deficits to countries running external surpluses.
Ever since the work of Robert Mundell (1961) on the conditions
needed to form monetary unions, the economic literature has placed a
high priority on the necessity of wage trod price flexibility to
facilitate adjustment to external imbalance in the absence of separate
currencies and the capacity to adjust the nominal exchange rate. (21)
However, as our discussion of the gold standard has indicated, wage and
price flexibility is a necessary but not a sufficient condition for the
operation of a fixed exchange rate regime. What is crucial is the
existence of an adjustment mechanism that triggers the necessary changes
in wages and prices.
The case of Greece under both the gold standard and the euro
illustrates the importance of the adjustment mechanism. As noted, Greece
tried to participate in the gold standard but was unsuccessful because
it could not abide by its fiscal requirements. Consequently, it was not
able to benefit from the gold standard adjustment mechanism, which would
have contained the external deficits. In contrast, under the euro,
Greece was able, upon entry, to borrow at near-core interest rates and
at the same time, in the absence of an adjustment mechanism, remain a
member without undertaking fiscal adjustment.
Indeed, the case of Greece between 2001 and 2009, the year in which
the Greek crisis erupted, illustrates the absence of an adjustment
mechanism. Table 2 presents the average growth rates of M3, total
domestic credit, credit to the private sector, and credit to the public
sector for Greece, Germany, and the eurozone between 2001 and 2009. As
shown in the table, M3 growth and credit growth were considerably higher
in Greece, a country with an external deficit during the period in
question, than they were in Germany, a country with an external surplus,
or the eurozone as a whole, which had, essentially, balanced external
accounts during that time. What is particularly striking is the much
faster credit growth to the private sector (16.7 percent) in Greece than
in Germany (2.7 percent). In other words, money and credit flowed in the
opposite direction of that needed to bring about external adjustment.
A consequence of these flows is shown in Figure 9, which presents
real (effective) exchange rates for Greece and Germany during 2001-09;
since Greece and Germany share the same currency, changes in the real
exchange rate mainly reflect changes in the relative domestic (consumer)
prices. As shown in the figure, Greece's real exchange rate
appreciated by about 15 percent against that of Germany, the opposite of
what we would expect on the basis of the countries' external
positions, but entirely consistent with the relative flows of money mad
credit in the two countries.
What caused money and credit to rise at high rates in Greece
between 2001 and 2009? The data on Greek government debt and credit
growth in Table 3 provide answers to that question. Several points are
especially noteworthy. First, while the stock of government debt rose by
147.8 billion [euro] from 9.001 through 2009, domestic holdings of that
debt declined by 222.1 billion [euro]. Foreign holdings of Greek
government debt rose by 169.9 billion [euro], accounting for more than
the overall increase in debt. Consequently, the share of Greek sovereign
debt held by Greek residents fell from 56.6 percent to 21.3 percent
while the share held by nonresidents rose from 43.4 percent to 78.7
percent. Second, Greek banks were large net sellers of Greek government
debt. At the time of Greece's entry in the eurozone in 2001, Greek
banks held very large portfolios of Greek government bonds, a result of
the requirements of the country's highly regulated financial system
of the 1980s and 1990s rather than the banks' free choice of
portfolio composition. This fact is demonstrated by the winding-down of
the banks' holdings of Greek government paper following the
liberalization of the financial sector that was completed in the
mid-1990s. They used the proceeds received from the sale of the Greek
sovereigns, in part, to lend to the private sector. Consequently, credit
to the private sector surged, especially from 2001 until 2008; credit
growth to the private sector accounted for the bulk of the large
expansion of total credit during 2001 to 2009 (Table 2).
[FIGURE 9 OMITTED]
What underpinned the foregoing phenomenon was the perception by
international investors that Greek sovereign debt carried little risk of
default. The decrease in spreads on these financial instruments to about
30 basis points in the mid-2000s (Figure 1) suggests that markets indeed
held such a perception. These perceptions, however, seem to have been
unreasonable in the face of historical evidence linking the probability
of default to the size of public debt relative to GDP. For instance,
during the 1980s Latin American countries typically defaulted at public
debt-to-GDP ratios that were a fraction of that of Greece. Moreover,
Greece's public debt-to-GDP ratio was not only very large but was
increasing during a period of above-average economic growth, a worrying
pattern. The only way the low risk premia can be justified is by
assuming that investors expected that the core of the eurozone would
have no choice but to bail out Greece in the event of a financial/fiscal
crisis. The expectation of a bailout lowered the default risk and
provided an incentive to issue increasing amounts of debt.
It is this circumstance that differentiates the euro regime from
the gold standard and allowed Greece to build up large external
imbalances. In contrast to the euro regime, under the gold standard
there was no expectation that the core members would bail out the
peripheral members. European members of the gold standard periphery, by
and large, maintained low fiscal deficits and current accounts that were
close to balance. If the fiscal policies of the peripheral countries
were not consistent with balanced external accounts, domestic interest
rates would rise and money and credit growth would slow, facilitating
external adjustment. In turn, the small fiscal and external imbalances
under the gold standard underpinned its durability. Countries such as
Greece, the fiscal institutions of which could not conform to the
system's fiscal requirements, were forced off the gold standard.
The price they paid for fiscal profligacy was much higher interest rates
than the peripheral participants of the gold standard.
The above process did not operate in the eurozone because investors
did not draw a distinction between the sovereign debt of the core
countries and the sovereign debt of the peripheral countries, such as
Greece. Consequently, there was no mechanism to adjust money and credit
growth, and Greece was able to run large current account and fiscal
deficits without taking remedial policy measures. This behavior
resembled that of the Latin American countries and of European countries
(such as Greece) that were not members of the gold standard.
Conclusion
The experiences of the core and periphery countires under the
classical gold standard are relevant for the eurozone. The behavior of
the core participants has been quite similar across the two monetary
arrangements (i.e., small budget deficits, or even surpluses, and
sustainable current-account balances). Under the gold standard, European
peripheral countries ran current account deficits, but the size of those
deficits was small relative to those experienced by Greece under the
euro. They were small because fiscal shocks were smaller and, more
importantly, because the adjustment mechanism while imperfect, worked to
mitigate the buildup of external imbalances. Countries with external
deficits would experience higher interest rates, a loss of gold
reserves, and lower money and credit growth. The resulting reduction in
wages and prices would contribute to the restoration of trade
competitiveness.
This mechanism was not operative in the case of Greece under the
euro. In the eurozone, the market's perception that Greek sovereign
debt represented a safe investment--probably founded on the expectation
of a bailout by core countries--suppressed the effect of sovereign
credit risk on Greek interest rates. At the same time, low interest
rates greased the wheels of fiscal expansion by sending the message that
there was no price to be paid for the buildup of sovereign debt. Hence,
while external imbalances were essentially self-correcting under the
gold standard, they were self-perpetuating in the eurozone due to the
perception of an absence of credit risk.
We draw two main conclusions. First, the durability of a monetary
union is crucially dependent on the existence of a well-functioning
adjustment mechanism. Second, adherence to a hard peg is no panacea and
cannot be sustained without the support of credible fiscal institutions.
References
Bloomfield, A. (1959) Monetary Policy under the International Gold
Standard. New York: Federal Reserve Bank of New York.
Boone, P., and Johnson, S. (2012) "The Euro Awaits Its
Verdict." Project Syndicate (22 May).
Bordo, M. (1981) "The Classical Gold Standard: Some Lessons
for Today." Federal Reserve Bank of St. Louis Review 63 (,5): 2-17.
-- (1992) "Gold Standard: Theory." In P. Newman, M.
Milgate, and J. Eatwell (eds.) The New Palgrave Dictionary of Money
and Finance, Vol. 2, 267-71. New York: Stockton Press.
--(1993) "The Gold Standard, Bretton Woods and Other Monetary
Regimes: A Historical Perspective." Federal Reserve Bank of St.
Louis Review 75 (2): 19.3-91.
Bordo, M., and Kydland, F. (1996) "The Gold Standard as a
Commitment Mechanism." In T. Bayoumi, B. Eichengreen, and M. Taylor
(eds.) Modern Perspectives on the Gold Standard, 55-100. Cambridge:
Cambridge University Press.
Bordo, M., and Rockoff, H. (1996) "The Gold Standard as a
'Good Housekeeping Seal of Approval.'" Journal of
Economic History 56 (2): 389-28.
Buttonwood (2010) "The New Gold Standard." The Economist
(3 February).
Cesarano, F.; Cifarelli, G.; and Toniolo, G. (2012) "Exchange
Rate Regimes and Reserve Policy: The Italian Lira, 1883-1911." Open
Economies Review 23 (2): 253-75.
Dellas, H., and Tavlas, G. S. (2009) "An Optimum
Currency-Area-Odyssey." Journal of International Money and Finance
28 (7): 1117-37.
--(2013) "Exchange Rate Regimes and Asset Prices."
Journal of International Money and Finance (forthcoming).
Duarte, A., and Andrade, J. (2004) "How the Gold Standard
Functioned in Portugal: An Analysis of Some Macroeconomic Aspects."
University of Coimbra, GEMF, Working Paper 2004-01.
Eichengreen, B. (1992) Golden Fetters: The Gold Standard and the
Great Depression. New York: Oxford University Press.
--(1994) International Monetary Arrangements for the 21st Century.
Washington: Brookings Institution.
--(1996) Globalizing Capital: A History of the International
Monetary System. Princeton, N.J.: Princeton University Press.
Eichengreen, B., and Hausmann, R. (1999) "Exchange Rates and
Financial Fragility." In Federal Reserve Bank of Kansas City, New
Challenges' for Monetary Policy, 329-68. Kansas City: Federal
Reserve Bank of Kansas City.
Eichengreen, B., and Temin, P. (2010) "Fetters of Gold and
Paper." Oxford Review of Economic Policy 26 (3): 370-84.
Fratianni, M., and Spinelli, F. (1984) "Italy mad the Gold
Standard Period, 1861-1914." In M. Bordo and A. Schwartz (eds.) A
Retrospective on the Classical Gold Standard, 40.5-54. Chicago:
University of Chicago Press.
Friedman, M. (1953) "The Case for Flexible Exchange
Rates." In M. Friedman, Essays Positive Economics', 157-203.
Chicago: University of Chicago Press.
Garganas, N., and Tavlas, G. S. (2001) "Monetary Regimes and
Inflation Performance: The Case of Greece." In R. Bryant, N.
Garganas, and G. S. Tavlas (eds.) Greece's' Economic
Performance and Prospects, 43-96. Athens and Washington: Bank of Greece
and Brookings Institution.
Gibson, H.; Hall, S. G.; and Tavlas, G. S. (2012) "The Greek
Financial Crisis: Growing Imbalances and Sovereign Spreads."
Journal of International Money and Finance 31 (3): 498-516.
Goodhart, C. (1992) "Economic and Monetary Union in Europe: A
U.K. Perspective." In E. Baltensperger and H.-W. Sinn (eds.)
Exchange Rate Regimes and Monetary Unions, 183-99. New York: St. Martins
Press.
Holinski, N.; Kool, C.; and Muysken, J. (2012) "Persistent
Macroeconomic Imbalances in the Euro Area: Causes and
Consequences." Federal Reserve Bank of St. Louis, Review 94 (1):
1-20.
James, H. (2012) "Golden Rules for the Eurozone."
European Voice (15 April).
Lazaretou, S. (2004) "The Drachma, Foreign Creditors, and the
International Monetary System: Tales of a Currency during the 19th and
the Early 20th Century." Bank of Greece Working Paper No. 16.
--(2011) "Financial Crises and Financial Market Regulation:
The Long Record of an 'Emerger.'" Bank of Greece Working
Paper No. 140.
Martin-Acena, P. (1994) "Spain during the Classical Gold
Standard Years, 1880-1914." In M. Bordo and F. Capie (eds.)
Monetary Regimes' in Transition, 135-72. Cambridge: Cambridge
University Press.
Mitchell, B. (2007) International Historical Statistics: Europe,
17, 50-2005, Hampshire, UK: Palgrave Macmillan.
Mundell, R. (1961) "A Theory of Optimum Currency Areas."
American Economic Review 51 (4): 6,57-65.
Obstfeld, M., and Taylor, A. (2003) "Globalization and Capital
Markets." In M. Bordo, A. Taylor, and J. Williamson (eds.)
Globalization in Historical Perspective, 121-88. Chicago: University of
Chicago Press.
Officer, L. (2010) "Gold Standard." Economic History
Services, http://eh.net/encyclopedia/article/officer.gold.standard.
Reinhart, C., and Rogoff, K. (2011) This Time It's Different:
Eight Centuries of Financial Folly. Princeton, N.J.: Princeton
University Press.
Scammell, W. (1965) "The Working of the Gold Standard."
Yorkshire Bulletin of Economic and Social Research 17 (1): 32-45.
Schularick, T., and Taylor, A. M. (2009) "Credit Booms Gone
Bust: Monetary Policy, Leverage Cycles and Financial Crises,
1870-2008." NBER Working Paper No. 15512.
Schwartz, A. (1993) "Interview with Anna Schwartz.'"
Federal Reserve Bank of Minneapolis, The Region (September).
Smits, J.-P.; Horlings, E.; and van Zanden, J. L. (2000)
"Dutch GNP and Its Components, 1800-1913." Groningen Growth
and Development Center, Monograph Series No. 5.
Soto, M. C. (1999) "The Risks and Returns of Not Being on the
Gold Standard: The Spanish Experience, 1880-1913." Journal of
Economic History 59 (2): 466-439.
Tavlas, G.; Dellas, H.; and Stockman, A. C. (2008) "The
Classification and Performance of Alternative Exchange-Rate
Systems." European Economic Review 52 (6): 941-63.
(1) In 1994, the Greek government set a goal to enter the eurozone
on January 1, 2001. The convergence of Greek economic indicators to
those of other European Union countries contributed after 1994 to the
narrowing of interest-rate spreads prior to eurozone entry. For an
analysis of the Greek economy before eurozone entry, see Garganas and
Tavlas (2001).
(2) The European Union's Stability and Growth Pact requires
that members' fiscal deficits be below 3 percent of GDP and their
debt-to-GDP ratios below 60 percent of GDP. Entry into the eurozone is,
in part, contingent on the satisfaction of these fiscal criteria. In the
case of the debt-to-GDP criterion, countries can be allowed to join if
the debt ratio is seen to he approaching the 60 percent critical value
at a satisfactory, pace. The latter circumstance applied to Greece. In
the year 2000, Greece was allowed entry, into the eurozone with a
debt-to-GDP ratio near 100 percent of GDP (because the ratio was on a
declining path) and a fiscal deficit initially reported at 3.0 percent
of GDP; the latter figure was subsequently revised to 4.5 percent of GDP
after Greece became a member of Europe's monetary union.
(3) Differences between the two are roughly attributed to different
compilation methods and data sources that are difficult to cross-check.
(4) Nonetheless, the nominal magnitudes were very different. In
2008, Germany's current account surplus totaled 154.1 billion
[euro]; Greece's current account deficit was 41.7 billion [euro].
(5) This circumstance differs from those of Ireland, Portugal and
Spain, where the private sector was mainly responsible for the widening
of the current account positions of those countries. See Holinski, Kool,
and Muysken (2012).
(6) Gibson, Hall, and Tavlas (2012) estimate a cointegrating
relationship between Greek spreads and their long-term fundamental
determinants and compare the spreads predicted by this estimated
relationship with actual spreads. The authors find that spreads were
significantly below what would be predicted by fundamentals from the end
of 2004 up to the middle of 2005; by contrast, in 2010, spreads exceeded
predicted spreads by some 400 basis points.
(7) The recent literature focuses on similarities between the gold
standard and eurozone regimes. For example, Eichengreen and Temin (2010)
pointed out that both regimes lack an escape mechanism to facilitate
adjustment, with the escape mechanism being more binding for the euro
regime than it was for the gold standard since, under the gold standard,
countries retained their national currencies so that they could modify
the regime. In our discussion, we focus on a major difference between
the two regimes. For recent analyses of the performances of alternate
exchange-rate regimes, see Tavlas, Dellas, and Stockman (2008) and
Dellas and Tavlas (2013).
(8) The gold points were the points at which it became profitable
to either export or import gold because of deviations between the market
and mint prices of gold (Eichengreen 1996: 196). Effectively, the gold
points functioned as the edges of exchange rate bands under which the
exchange rate could fluctuate without occasioning either corrective gold
flows or central bank intervention (Eichengreen 1994: 42).
(9) As Scammell (1965: 32) put it. "The nineteenth century
gold standard was a gold coinage standard, in which gold coins
circulated domestically and were interchangeable with notes at the
central bank." Although fiat (paper) money was increasingly used
during the classical gold standard to economize on the scarce resources
tied up with commodity money, fiat money became acceptable only because
it was convertible into gold (Bordo and Kydland 1996: 63).
(10) As Bloomfield (1959: 14) observed, "The composition of
the gold standard 'club' changed over the course of the
[1880-1914] period."
(11) There is not a clear consensus among historians as to which
countries comprised the periphery. For example, Eichengreen (1996: 39)
included the United States while Officer (2010) did not. Since the
Federal Reserve System was not established until December 1913, most
historians consider that the United States was part of the gold standard
periphery. This circumstance also applies to Switzerland, since the
Swiss National Bank was not established until 1907. Austria-Hungary and
Russia are sometimes considered to be core countries, but they did not
join the gold standard until the mid-1890s.
(12) Cesarano, Cifarelli, and Toniolo (2012: 253) argued that the
policy of "shadowing the gold standard proved very
successful."
(13) According to Soto (1999: 469), "There was a credible
attempt to adhere to the [gold standard] system, but only the decision
of shadowing was taken."
(14) Goodhavt (1992: 192) made a similar argument:
"Governments generally abided by a balanced budget objective, which
could be managed, in effect, as representing the required fiscal
constraint on national policies."
(15) The high degree of discipline imposed by the gold standard
would make it difficult to implement today. As Bordo (1992: 270) put it,
"[in present circumstances] few countries [would be] willing to
accept the gold standard's discipline."
(16) Eichengreen (1994: 44) described the experiences of the Latin
American periphery as follows: "Latin American countries repeatedly
failed to control their fiscal policies, leading to a monetarization of
budget deficits, the suspension of gold convertibility, and currency
depreciation."
(17) Reinhart and Rogoff (2011) reported that Greece defaulted on
its debt in 1893. As noted, Greece was not on the gold standard in 1893.
(18) The choice of countries was based on the availability of data
for those countries during the period in question.
(19) Although, Table 1 reports average values for the period 1880
to 1913, each of the three series in Table 1 shows low volatility for
each country reported in the table.
(20) The experience of the gold standard provides clear-cut
evidence for the rationale of the EU's Stability and Growth Pact.
(21) Dellas and Tavlas (2009) critically assess the development of
the literature on optimum currency areas.
Harris Dellas is Professor of Economics and Director of the
Institute of Political Economy at the University of Bern, and George S.
Tavlas is a Member of the Monetary Policy Council of the Bank of Greece.
The views are those of the authors and not necessarily those of their
respective institutions. They thank the following individuals for
helpful comments on earlier versions of this article: Elena Argiri,
Hiona Balfoussia, George Chouliarakis, Michele Fratianni, Heather
Gibson, Sofia 'Lazaretou, Patrick Minford, Angeliki Momtsia,
Ifigeneia Skotida, Michael Ulan, Haris Vittas, and Emmanouel
Zervoudakis. They are especially grateful to Elisavet Bosdelekidou,
Dafni Giannikou, and Christina Tsochatzi for excellent research
assistance.
TABLE 1
SELECTED ECONOMIC INDICATORS, GOLD STANDARD COUNTRIES, 1880-1913
(Percentage of GDP, Period Averages)
Fiscal Government Current
Balance Spending Account
Core Countries
France -0.5 11.7 2.7
Germany -2.8 5.4 1.8
Netherlands -2.0 10.0 4.5
UK -0.1 7.6 5.1
Peripheral Countries
Denmark -0.6 6.3 -2.6
Italy -1.0 13.9 0.7
Norway -0.7 7.6 -2.7
Spain 0.2 8.8 1.6
Sweden -0.04 6.6 -2.6
Non-Member
Greece -12.8 23.5 -7.1 *
Inflation Nominal
Rate Debt
Core Countries
France -0.1 95.9
Germany 0.7 39.3
Netherlands 0.1 74.6
UK 0.1 45.9
Peripheral Countries
Denmark 0.4 19.1
Italy 0.2 104.6
Norway 0.9 21.4
Spain 0.4 94.9
Sweden 0.5 16.9
Non-Member
Greece 0.9 143.5
* Trade balance
SOURCES: Martin-Arena (1994); Mitchell (2007); Smits, Horlings, and
van Zanden (2000); Lazaretou (2011).
TABLE 2
MONEY (M3) AND CREDIT GROWTH
(ANNUAL AVERAGES, 2001-2009)
Greece Germany Euro Area
M3 8.8 5.7 7.9
Total Credit 10.0 1.9 6.6
Credit to Public Sector 0.3 -1.8 1.9
Credit to Private Sector 16.7 2.7 7.9
Current Account Balance -13.4 4.4 0.3
(% of GDP)
SOURCES: ECB Statistical Data Warehouse and European Commission
Annual Macroeconomic database.
TABLE 3
GREECE: SELECTED ECONOMIC INDICATORS
2001 2002 2003
Government Debt (% of GDP) 103.7 101.7 97.4
Level of Government Debt 151.9 159.2 168.0
(billions of euros)
Change in Government Debt 10.9 7.3 8.8
(billions of euros)
Domestic 4.4 0.0 -8.2
Foreign 6.5 7.3 17.0
Share of Debt Held by 56.6 54.1 46.3
Residents (% of total)
Domestic banks 26.9 25.1 21.6
Share of Debt Held by 43.4 45.9 53.7
Nonresidents (% of total)
Total Domestic Credit Growth (%) 12.2 8.2 6.9
Credit growth (from domestic banks) -1.0 -2.3 -3.3
to public sector
Credit growth (from domestic banks) 13.2 10.5 10.1
to private sector
2004 2005 2006
Government Debt (% of GDP) 98.9 110.0 107.7
Level of Government Debt 183.2 212.4 224.9
(billions of euros)
Change in Government Debt 15.2 29.2 12.5
(billions of euros)
Domestic -4.6 -5.0 -0.2
Foreign 19.4 34.3 12.7
Share of Debt Held by 40.0 32.1 30.2
Residents (% of total)
Domestic banks 18.3 18.5 19.2
Share of Debt Held by 60.0 67.9 69.8
Nonresidents (% of total)
Total Domestic Credit Growth (%) 9.1 11.6 15.8
Credit growth (from domestic banks) -1.5 1.1 2.6
to public sector
Credit growth (from domestic banks) 10.6 10.5 13.2
to private sector
2007 2008 2009
Government Debt (% of GDP) 107.5 113.0 129.3
Level of Government Debt 239.5 263.3 299.7
(billions of euros)
Change in Government Debt 14.6 23.8 36.4
(billions of euros)
Domestic -9.8 -1.1 6.9
Foreign 24.5 24.9 29.5
Share of Debt Held by 24.2 21.6 21.3
Residents (% of total)
Domestic banks 17.1 15.2 15.5
Share of Debt Held by 75.8 78.4 78.7
Nonresidents (% of total)
Total Domestic Credit Growth (%) 10.6 13.6 2.0
Credit growth (from domestic banks) -2.2 0.7 2.8
to public sector
Credit growth (from domestic banks) 12.8 12.8 -0.8
to private sector
SOURCES: ECB Statistical Data Warehouse and ELSTAT.
FIGURE 2
GREECE: EVOLUTION OF GENERAL GOVERNMENT
DEFICIT, PRIMARY DEFICIT, AND DEBT
Percentage of GDP
Primary Interest General Government Debt
Deficit (EDP) Deficit
2001 -2.0 6.5 4.5 103.7
2002 -0.7 5.5 4.8 101.7
2003 0.7 4.9 5.6 97.4
2001 2.6 5.0 7.5 98.9
2005 1.0 4.5 5.5 110.0
2006 1.3 4.4 5.7 107.7
2007 2.0 4.5 6.5 107.5
2008 4.8 5.0 9.8 113.0
2009 10.4 5.2 15.6 129.3
2010 4.7 5.7 10.3 145.0
2011 2.2 6.9 9.1 165.3
SOURCE: ELSTAT (Hellenic Statistical Authority).
FIGURE 3
GREECE: TOTAL REVENUE, EXPENDITURE, AND DEFICIT
Percentage of GDP
Deficit Total Total
(EDP) Revenue Expenditure
2001 4.5 40.9 45.3
2002 4.8 40.3 45.1
2003 5.6 39.0 44.7
2004 7.5 38.1 45.5
2005 5.5 39.0 44.6
2006 5.7 39.2 45.2
2007 6.5 40.8 47.6
2008 9.8 40.7 50.6
2009 15.6 38.2 53.8
2010 10.3 39.7 50.2
2011 9.1 40.9 50.1
SOURCE: ELSTAT (Hellenic Statistical Authority).
