Greek debt sustainability and official crisis lending.
Schumacher, Julian ; di Mauro, Beatrice Weder
ABSTRACT The International Monetary Fund and the European Stability
Mechanism softened their crisis lending policies repeatedly to deal with
the Greek debt crisis, but the analysis of debt sustainability still
acts as the gatekeeper for access to official financing. We explore the
underlying mechanics of debt sustainability analysis and show that the
standard model is inappropriate for Greece since it ignores the highly
concessional terms of Greek debt. Greek debt has been restructured
repeatedly, and now two-thirds of the stock contains grant elements of
about 54 percent. The present value of outstanding Greek debt is
currently about 100 percent of GDP and will rise to about 120 percent
under the new program. Greek debt sustainability therefore is less a
problem of the debt stock. By simulating different paths of the gross
financing needs, we show that there may be liquidity problems over the
medium to long terms (in particular, in 2035 and beyond). However, our
estimation of the financing need is subject to high uncertainty and
mainly depends on whether Greece will be able to regain access to
markets at reasonable terms.
**********
Many people hold strong views on Greek debt. Just taking a casual
look at the level of Greece's debt, which the International
Monetary Fund (IMF) has recently projected to rise to 200 percent of
GDP, it seems obvious that Greek public finances cannot possibly be
sustainable (IMF 2015a, 2015b). What is not obvious is how this can be
subject to dispute among the main creditor institutions. Greek official
sector debt sustainability assessments have been quite volatile, but by
the beginning of 2015 the verdict of the main official creditor
institutions--hereafter referred to as the troika for short--was that
Greek debt was sustainable. (1) Eight months later, however, at the time
of this writing, the troika has split over the question of debt
sustainability; while the Europeans are pushing ahead with a new program
for Greece, the IMF is holding out. It seems that the diagnosis of debt
sustainability is not so obvious after all.
One reason that a diagnosis of debt sustainability is complicated
could be that politics plays a role, in particular the political
feasibility of adjustment. Indeed, theory has long emphasized that
sovereign debt is different from corporate debt, precisely because
politics and institutions are crucial in determining a country's
capacity and willingness to repay. (2) From this perspective, debt
sustainability would depend, among other things, on the particular
political coalition, the strength of political institutions, and even on
the egos of decisionmakers and their negotiating power, both at home and
abroad. Thus, debt sustainability would not only be unobservable and
country-specific but also time-varying and highly volatile. Political
positions can change very rapidly, as showcased by the turmoil caused by
the Greek government's turnover in January 2015.
However, this is a perspective that neither the IMF nor the
eurozone can adopt. As a matter of principle, they must ensure equal
treatment across members and cannot constantly change the goalposts in
accordance with shifting political circumstances. Thus, they need to
deploy a framework to assess debt sustainability that can be applied to
the entire membership. (3) In addition, this framework should be
designed with the goals of both protecting the debtor country from
overborrowing and protecting the resources of the creditor institution.
An "unsustainable" verdict should preclude the official sector
from lending into cases of insolvency and should require some form of
debt restructuring first (IMF 2014a, 2014c). Understanding the basis of
official debt sustainability analysis is therefore crucial. The first
contribution of this paper (section I) is to present the models for
sustainability assessment employed by the IMF and the European Stability
Mechanism (ESM), and then to review the impact of the Greek debt crisis
on the overall framework for international crisis lending.
The paper's second contribution (section II) is to evaluate
present Greek debt sustainability in light of these models. We show that
neither of the current frameworks sufficiently takes into account the
extent of Greece's dependence on official sector funding. We
analyze Greek debt using a present-value approach to account for the
concessionality element. Our main finding is that the overall debt
contains a grant element of 37 percent and the European loans of up to
60 percent. We also show an analysis of projected repayment flows, or
gross financing needs. While the main projection shows a critical level
of more than 15 percent over the long term, this result is highly
sensitive to assumptions about market conditions in the coming decades.
We are not the first to argue that the face value of (gross)
sovereign debt may be a misleading measure. For rich but highly indebted
countries like Japan, it has long been suggested that government debt
should be measured in net rather than gross terms by deducting the value
of government assets from the debt stock (IMF 2013b). A more recent
suggestion is that sovereign debt should be expressed according to
international accounting standards, including measuring the debt stock
at fair values (Serafeim 2015; Kazarian 2015). We are more in line with
Daniel Dias, Christine Richmond, and Mark Wright (2014), who show that
measuring debt in present value, rather than face value, enables
cross-country comparisons and discourages the hiding of true
indebtedness behind convenient debt profiles. But our main point applies
to Greece and countries borrowing from the ESM. For them, looking at
gross debt will lead to a misdiagnosis, because it does not appreciate
the concessional nature of European crisis lending.
The remainder of the paper is organized as follows. In section I,
we discuss the role of debt sustainability analysis within the European
and IMF crisis-lending frameworks. In section II, we highlight the
uncertainty of Greek debt sustainability. And in section III, we provide
our policy conclusions.
I. The Official Sector's Lending Framework and the Role of
Debt Sustainability Analysis
The lending frameworks of the IMF and the ESM set the parameters
under which countries in financial distress are considered illiquid,
rather than insolvent, and can therefore receive emergency financing
without first restructuring existing debt. The principle of committing
to not lend into unsustainable debt dynamics is justified by several
reasons, most importantly by the need to protect taxpayer resources and
prevent the debtor from accumulating excess debt, which would make
future adjustment and eventual restructuring more costly. Nevertheless,
this commitment is regularly tested, since the official sector will be
tempted to lend even in highly doubtful cases, hoping that conditions
will improve, and thus avoid the immediate costs of a restructuring, a
policy known as "kicking the can down the road." The
temptation to lend into insolvency is even larger if there are fears of
global or regional contagion resulting from a restructuring.
Consequently, debt restructurings tend to do too little, come too
late, and are too costly (Levy Yeyati and Panizza 2011; IMF 2013a).
Moreover, such time-inconsistent policies of official lenders may result
in overborrowing. The presence of an international lender of last resort
creates incentives for private creditors to lend without regard for risk
in the expectation of an official bailout. The costs of overborrowing
and delayed restructuring are then mostly borne by local taxpayers,
since official lenders tend to be repaid (Buchheit and others 2013). By
governing the decision to provide emergency funds or insist on debt
restructuring and relief first, the lending frameworks of the IMF and
ESM in practice act as sovereign debt restructuring regimes. The
analysis of debt sustainability is their main gatekeeper.
I.A. The ESM and IMF Crisis Lending Frameworks
The European crisis lending framework, as laid out in the 2012
treaty establishing the ESM, provides for rule-based decisionmaking for
the granting of emergency loans. Article 13 requests that an application
by a member state will be considered based on an assessment of three
criteria through the European Commission (EC) in conjunction with the
European Central Bank (ECB): (i) The risks to the financial stability of
the euro area as a whole; (ii) the sustainability of public debt (if
appropriate, in conjunction with the IMF); and (iii) the actual or
potential financing needs of the applicant member state.
In principle, ESM loans will only be extended if the member
state's public debt is sustainable. However, the treaty does not
give clear guidance on how to proceed if the results of the ESM-EC-ECB
debt sustainability analysis indicate an unsustainable situation. (4)
Specifically, there are no provisions that would require a debt
restructuring to unlock ESM access in a case where the sustainability
analysis suggests an unsustainable debt.
The IMF's lending "of last resort" to countries in
financial trouble is based on multiple variables. First and foremost, a
country's maximum loan volume is determined by its
"quota"--a blended measure of a nation's GDP, financial
openness and volatility, and official reserves (IMF 2008). In normal
circumstances, countries are allowed to borrow up to 200 percent of
their quota during a 12-month period, and not more than a cumulative 600
percent of their quota. The Greek quota, for instance, currently stands
at SDR 1.1 billion, or about $1.5 billion, which would have limited the
maximum Greek borrowing from the IMF to about $9 billion. (5) However,
in exceptional circumstances, member countries are allowed to borrow
more than the normal limits under the "exceptional access"
policy. (6) To obtain exceptional access under the rules prevalent at
the time of the first Greek program, four criteria had to be met (IMF
2004, p. 4):
i. The country is under exceptional balance-of-payments pressure
exceeding the normal limits.
ii. A debt sustainability analysis indicates a high probability
that the debt will remain sustainable. If the debt sustainability
analysis cannot conclude this with high probability, exceptional access
may be granted on grounds of systemic concerns (a "systemic
exemption").
iii. The country has good chances of regaining access to private
markets before the bailout ends.
iv. The country has a policy program convincingly promising
success, as well as the institutional quality to implement the program.
As with the ESM framework, the IMF's framework requires an
in-depth debt sustainability analysis of the country's debt stock.
Before the introduction of the systemic exemption, the outcome of this
analysis determined whether debt restructuring was required before a
loan could be granted (IMF 2014c). Only if the debt level was deemed
sustainable with high probability could exceptional access be granted
without recourse to debt restructuring. The introduction of an exemption
in cases of systemic concerns was therefore a major softening of the
lending framework.
I.B. The Softening of the Lending Framework in Response to the
Greek Crisis
European and international institutions of crisis lending were
profoundly affected by the Greek crisis. In the case of the eurozone,
the crisis led to the very creation of a multilateral institution for
emergency financing. The previous regime only foresaw offering financial
assistance for balance-of-payments crises in European Union (EU) members
outside the eurozone. Inside the eurozone, fiscal crises were to be
avoided by the threat of "no bailouts," as enshrined in
Article 125 of the Treaty on the Functioning of the European Union.
The Greek crisis exposed the time inconsistency of the "no
bailout" promise. Faced with the threat of an imminent Greek
sovereign default and high uncertainties about the direct and indirect
costs of a default to the monetary union, eurozone member states found a
quick fix to circumvent the "no bailout" clause: They granted
a credit line of up to 80 billion [euro] in bilateral loans through a
special vehicle, the Greek Loan Facility (GLF). (7) This exceptional
vehicle was replaced first with the creation of a multilateral structure
(the European Financial Stability Facility, EFSF) and then through a
treaty establishing the permanent ESM.
Introducing a permanent facility for emergency financing amounted
to a significant reform of the eurozone architecture. It added a
supranational fiscal buffer for large crises and established a new
regime of conditional bailouts. As noted above, three criteria for
access to ESM funding are the main governors of this new regime, and
should ensure that loans are extended only in cases of sustainable debt
dynamics. However, the first and the third criteria are bound to be
fulfilled in any crisis. If any default or restructuring is considered
to raise doubt about the "integrity of the euro area as a
whole," this test becomes meaningless as a commitment device. (8)
Thus, the only real test is the analysis of debt sustainability.
The Greek debt crisis further affected the European financial
architecture as loans from European partner countries were being
restructured. As presented in detail below, Europeans made a series of
concessions and restructured their original loans multiple times. This
official restructuring was silent but had a permanent impact on the
Greek debt profile and on the institutions for crisis lending in Europe.
The conditions of Greek loans were passed on to other crisis countries
and became a de facto new ESM lending policy. As a consequence, European
crisis lending conditions are now highly concessionary, with average
maturities of up to 32.5 years, decade-long grace periods, and a
pass-through of ESM funding costs to program countries. These lending
terms are closer to those of the World Bank for long-term lending to
low-income countries than to IMF-type, short-term balance-of-payments
assistance. (9)
For the IMF, the first Greek program also brought about an
important change in lending policies. The May 2010 stand-by program
granted Greece exceptional access to draw 30 billion [euro], more than
3,000 percent of its quota. The yardstick for granting such a high level
of access was the debt sustainability criterion. Under the baseline
scenario, the IMF projected Greece's public debt as a share of GDP
to peak in 2013 at 149 percent and to gradually decline by 2020 to 120
percent, although it flagged many risks to this baseline scenario (IMF
2010a). On balance, the IMF considered debt to be sustainable over the
medium term; however, it noted that the significant uncertainties
"make it difficult to state categorically that this is the case
with a high probability" (IMF 2010a, p. 20). Under the
then-existing "exceptional access" policy, this statement
would have precluded the IMF from approving the program without first
requiring debt restructuring. The quick-fix solution was to introduce a
"systemic exemption" from the rule due to the high risk of
international spillovers.
This solution implies that the IMF could lend to insolvent
countries, provided that spillovers are seen to be large. (10) The
systemic exemption eventually became a permanent feature of IMF
exceptional access policies (IMF 2014c). (11) A former director of the
IMF, Susan Schadler, put it like this: "The framework constraining
the discretion of the IMF in severe debt crises broke down in its first
serious test" (Schadler 2013, p. 14). (12) IMF staff members have
proposed eliminating the systemic exemption on the grounds that it is
inequitable and excessively open ended. (13) Instead of keeping the
vague option of extending loans on grounds of systemic risk concerns, a
recent staff proposal suggested that a one-time debt reprofiling
(prolongation of maturities without reduction in principal or interest)
should always be required in cases of doubtful debt dynamics (IMF
2014a). (14) By January 2016, the main shareholders of the IMF had
accepted this argument, and the IMF announced a new lending policy
abolishing the systemic exemption (IMF 2016).
However, the verdict on debt sustainability still constitutes an
important condition for access to IMF and ESM lending. We next turn to
the mechanics of these analyses.
I.C. The Mechanics of IMF and ESM Debt Sustainability Analysis
Both the IMF's and ESM's methodologies for analyzing debt
sustainability require an analysis of the debt stock in a static
framework using observed data about the current situation and in a
dynamic framework using forecast data (IMF 2013b, 2013d; European
Commission 2014). Forecasting requires a comprehensive macroeconomic
model that at a minimum includes growth, inflation, interest, and
exchange rates, as well as fiscal policies, and is therefore subject to
uncertainty. Besides the benchmark assumptions, the data are also
exposed to a series of robustness checks and stochastic analyses in
which alternative data trajectories are considered. (15)
The results of these exercises, along with the static indicators,
are then compared with a set of thresholds that designate an increased
risk of debt distress. These thresholds are derived by running early
warning systems, in the spirit of the "signaling approach"
suggested by Graciela Kaminsky, Saul Lizondo, and Carmen Reinhart
(1998); Kaminsky and Reinhart (1999); and Emanuele Baldacci and others
(2011). A related alternative is the regression-based approach suggested
by Aart Kraay and Vikram Nehru (2004). In the signaling approach, a
signal of an impending crisis is triggered if the realized value of a
set of macroeconomic and financial variables exceeds a critical value of
the variable's distribution. If a signal is triggered and a crisis
erupts in the following predefined projection period (such as the
24-month period in Kaminsky and Reinhart [1999]), the signal is recorded
as a "good" positive; if no crisis occurs, it is counted as a
"false" positive. Likewise, if there is a crisis but no signal
has been recorded in the preceding projection period, every observation
without a signal is recorded as a "false" negative. If no
crisis erupts, and no signal was triggered, the observation counts as a
"good" negative signal. The critical value of the distribution
is chosen so as to minimize the equally weighted sum of false positive
and false negative signals.
For instance, in the analysis by Kaminsky and Reinhart (1999), the
threshold of the deficit-to-GDP ratio that minimizes the sum of false
positives and false negatives is the 86th percentile of the historical
deficit distribution. If a country's realized deficit in any given
year exceeds the 86th percentile of that country's distribution of
deficits, a crisis signal is triggered. While the original contributions
by Kaminsky, Lizondo, and Reinhart (1998) and Kaminsky and Reinhart
(1999) suggest country-specific distributions, the approach used today
by the IMF (Baldacci and others 2011; IMF 2013b) and the ESM (Berti,
Salto, and Lequien 2012; European Commission 2014) chooses thresholds
based on the pooled distributions of all countries in the sample.
Similarly, the IMF thresholds for low-income countries are derived
from a regression model in which a crisis indicator is regressed on the
threshold variables; the maximum thresholds are then set so that the
predicted crisis probability remains below predefined values (Kraay and
Nehru 2004).
Table 1 shows the thresholds for the various frameworks. While the
ESM does not distinguish between different countries, the IMF framework
has different variables for market-access and low-income countries. In
addition, the values for advanced economies and emerging markets are
different, and within low-income countries values are further
differentiated according to the institutional quality.
A country is only considered at low risk of debt distress if its
debt stock and predicted future development do not exceed these
thresholds, both under the benchmark scenario and under the robustness
scenario with more negative assumptions. If the indicators exceed the
thresholds in the baseline scenarios, the probability of debt distress
is considered high. The middle ground is more ambiguous; a moderate risk
rating is assigned if the thresholds are breached in the robustness
scenarios but remain below the critical values under the baseline
assumptions.
The general framework is therefore similar for all countries, in
both the IMF framework and the ESM framework. However, the IMF analysis
significantly differentiates between low-income countries and
market-access countries along at least two dimensions that are not
contained in the ESM analysis, as follows.
First, as the name suggests, market-access countries are assumed to
borrow predominantly at market terms from market sources. This requires
specific assumptions about the type and cost of market financing,
including modeling the coupon, maturity, and currency structure of the
debt. For low-income countries, borrowing from capital markets is
considered an option, but many low-income countries rely mostly on
official financing.
Second, for market-access countries, the stock of debt is
considered at nominal values. The liabilities therefore only consist of
the principal repayments, without taking into account coupon payments or
the life of a debt instrument. The face-value measure of the debt stock,
FV, is thus given by
(1) FV = [T.summation over (t=0)] [A.sub.t],
where [A.sub.t] represents principal repayments in year t.
This is different for low-income countries, which receive most of
their financing from official sources and whose debts are computed and
analyzed in discounted present values. The present value, PV, is
computed to include all discounted cash flows of the principal,
[A.sub.t], and coupons, [C.sub.t]:
(2) PV = [T.summation over (t=0)] [C.sub.t]/[(1 + d).sup.t] +
[A.sub.t]/[(1 + d).sup.t].
When computing meaningful present values, the key decision is
choosing an appropriate discount rate, d. In the current IMF framework,
d is set to a constant rate of 5 percent (IMF 2013c). This choice is
justified by the fact that a more elaborate discounting model would
increase the degrees of freedom in the analysis, thereby making
cross-country comparisons more difficult. Conceivable alternatives that
have been discussed in the literature range from the London Interbank
Offered Rate (commonly known as LIBOR) (Easterly 2001), to higher
(constant) rates of between 7 and 10 percent (Chauvin and Kraay 2005;
Andritzky 2006; Dikhanov 2006), to discounting based on the
country's (or a reference country's) sovereign yield curve
(Cruces and Trebesch 2013). In this paper, we do not take a stance on
which of these approaches should be preferred; instead, to maximize
comparability, we apply the IMF's discount rate of 5 percent. (16)
Both FV and PV only measure debt stocks, without taking into
account funding pressures in any given year. To complement the analysis
for this dimension, the gross financing needs, GFN, measure the
difference between the debt service obligations and the
government's available income for debt payments, the primary
balance, PB:
[GFN.sub.t] = [A.sup.*.sub.t] + [C.sup.*.sub.t] + [A.sup.N.sub.t] +
[I.sup.N.sub.t] [FV.sup.N.sub.t-1] - P[B.sub.t],
and
[FV.sup.N.sub.t] = [FV.sup.N.sub.t-n] + [GFN.sub.t] -
[A.sup.N.sub.t] = [t.summation over (i=0)] [GFN.sub.i] [A.sup.N.sub.i],
where [A.sup.*.sub.t] and [C.sup.*.sub.t] denote the debt
repayments and interest that are already scheduled as of the date of the
analysis, [A.sup.N.sub.t] denotes the repayment of newly issued debt to
cover previous periods' GFN, [I.sup.N.sub.t] represents the
interest burden on newly issued debt, and [FV.sup.N.sub.t] is the stock
of newly issued debt. The future refinancing terms of the projected
funding shortfalls--that is, the interest rate and maturity of newly
issued debt--must be assumed. For instance, under an assumed maturity of
5 years, [A.sup.N.sub.t] will be equal to [GFN.sub.t-5].
II. Debt Sustainability Analysis in Greece: Past and Present
Shortfalls
Ideally, the mechanical application of these tools should provide a
clear result if a government's public finances are sustainable or
not--independently of whether such a result is taken seriously in
official lending frameworks. Yet in practice, significant uncertainties
open a wide range of possible outcomes of the analysis, not least
because it requires assumptions about the feasibility of future
budgetary assumptions (House and Tesar 2015).
[FIGURE 1 OMITTED]
II.A. Serial Misdiagnoses and Restructuring
Indeed, over the course of the Greek crisis, debt dynamics have
been repeatedly underestimated. We divide the recent history into four
stages, with the first stage starting in the fall of 2009 (figure 1). At
that time, Greece was already in an "Excess Deficit Procedure"
for breaching the limits of the Maastricht Treaty, and it had committed
to bringing its deficit back to 3 percent of GDP during the coming year.
In the summer of 2009, the IMF estimated the current deficit at 6.2
percent and warned that debt dynamics would become unsustainable unless
policies were radically changed (IMF 2009). The debt-to-GDP ratio was
expected to rise above 100 percent in 2009 and to increase further, to
more than 120 percent, within 2 years. The IMF analysis concluded that
fiscal consolidation was immediately required to achieve sustainability
but that fairly modest adjustments of 1.5 percent of GDP would be
sufficient. In October 2009, Greece revealed new deficit estimates of up
to 12.5 percent, at the same time acknowledging that it had misreported
previous numbers, which also turned out to be substantially higher. (17)
Concerns about fiscal sustainability deepened and triggered a confidence
crisis.
The second stage began in May 2010 with the first joint bailout by
eurozone governments and the IMF. While the dynamic was seen as
considerably more negative than before and the debt stock was seen to
peak at close to 150 percent of GDP, the verdict was still that the debt
was sustainable, if not with high probability. The official loans were
justified by invoking the systemic exemption for the first time and
trusting that debt restructuring would not be necessary.
The third stage was the time of reckoning, reached in mid-2011,
when official sector creditors acknowledged that debt restructuring was
unavoidable (IMF 2011). The well-publicized and well-documented part of
this stage was the restructuring of private debt, which took place in
March 2012. The process involved retroactively changing bond contracts
by legislative action and a good measure of coercion by governments on
financial institutions; but the result was a high participation rate and
a severe haircut, with present value reduction of over 60 percent
(Zettelmeyer, Trebesch, and Gulati 2013).
The official sector restructuring was more silent. Over time,
European public sector loans were restructured, deeply and repeatedly.
Table 2 shows the timeline for Greek debt restructurings through the two
main public loan vehicles for Greece, the GLF and the EFSF. Interest
rates on bilateral loans in the GLF were lowered in three steps between
2010 and 2013, reducing the interest margin over the floating 3-month
Euro Interbank Offered Rate (commonly known as EURIBOR) from 300-400
basis points to 50 basis points. Even more pronounced were the
extensions of the grace period, from 3 to 10 years, and of the maturity,
from 5 to 30 years. EFSF loan conditions were restructured in a similar
way, most importantly by almost doubling the average maturity of the
loans to more than 30 years. (18) ESM lending policies were later
aligned.
After the combined private and official debt relief, the troika
concluded that Greek debt was finally sustainable. Moreover, the
assessment became gradually more optimistic, and the IMF released a new
debt projection in June 2014. For the first time, no further increase in
the debt stock was projected--it seemed that the peak had been left
behind. By the beginning of 2015, the troika viewed Greece as being on a
good path. In its request to the German parliament for an extension of
the Greek program, the German Ministry of Finance justified the
extension with a "confirmation of debt sustainability" by the
European Commission, and explained that the "debt sustainability
has improved since the last program review of April 2014" (German
Ministry of Finance 2014, p. 4).
The fourth and ongoing stage is characterized by conflict between
official creditors about debt sustainability, which at the time of this
writing remains unresolved. The IMF has made debt relief a condition for
participation in a third Greek program. In July 2015, it published two
new debt projections within a short time. It then argued that the
systemic exemption can no longer be invoked for Greece and that it will
not participate in funding a new program unless there is further debt
restructuring on the European side. The Europeans have decided to go
ahead with financing without restructuring the existing loans again,
leaving burden sharing within the official sector an unresolved
conflict.
II.B. Uncertainties in the Analysis of Greece's Debt Stock
While Greece had been exclusively relying on private financing
between the introduction of the euro and the start of the European
sovereign debt crisis in 2009, Greek debt today is dominated by official
loans. As of the end of July 2015, before the third program had been
negotiated, more than 80 percent of its current outstanding debt was
owed to official creditors. The average maturity was 15.7 years, with an
average interest rate of 2.7 percent. (19)
Figure 2 shows the debt repayment profile of Greece by creditor.
(20) Only the relatively large amount of short-term debt (Treasury
bills) and the remaining holdout bonds that were not restructured in
2012 require repayments to private investors within the next 8 years.
Afterward, the remaining private sector involvement (PSI) bonds amortize
over a period of 20 years, stretched out through the maturity extensions
of the 2012 debt restructuring. The bulk of the debt is owed to the
EFSF; to other eurozone governments through the GLF; to the IMF; to the
ECB; and to other members of the European System of Central Banks.
Notably, the official European loans through the GLF and the EFSF only
start becoming due in 2020 and 2023, respectively, and repayments are
stretched out until 2054.
[FIGURE 2 OMITTED]
As explained above, the official loans are extended at highly
favorable terms (table 3). This generates a significant element of
concessionality. Using the discount rate of 5 percent to compare the
face value of the EFSF and GLF loans to their present value reveals
considerable "grant elements" of up to 61 percent. On average,
the Greek debt stock contains a grant element of 37 percent. (21)
The recently negotiated third program over 86 billion [euro] is
likely to increase this concessionality. The new ESM program will have
the same 32.5 year average maturity as the EFSF loans, with
amortizations beginning in 2034, and similarly favorable interest rates.
Assuming that the IMF will contribute circa 10 percent of the total
volume, in line with the currently outstanding share of IMF and European
commitments, and request a maturity of 5 years, the average grant
element will rise to more than 40 percent.
II.C. Uncertainties in the Analysis of Flows
In its most recent debt sustainability analysis, the IMF has
expressed a similar judgment. Furthermore, the IMF has acknowledged that
the nominal gross debt-to-GDP ratio is no longer a meaningful metric to
evaluate sustainability (IMF 2015a, p. 11). The recent analysis suggests
that gross financing need (GFN) should be looked at instead, as these
needs capture the funding pressure in any given year. Using values
similar to those indicated in the recent IMF debt sustainability
analysis (annual real GDP growth of 1.75 percent, average interest rate
of 6 percent, average maturity of 5 years, and primary balance of 3.5
percent), the GFN is projected to increase relative to GDP above
critical thresholds of 15 to 20 percent from the mid-2030s onward
(figure 3). In the short to medium terms, however, the GFN remains below
this critical value.
[FIGURE 3 OMITTED]
Two factors explain the increase in GFN from the 2030s onward.
First, the European loan repayment schedules fall together with
redemptions of private sector bonds after the grace periods on the
official loans have ended in the mid-2020s. However, even under modest
growth assumptions, the annual obligations do not exceed 15 billion
[euro], so in isolation they remain well below 15 percent of GDP. The
second, more crucial factor in this analysis is therefore the refinanced
debt from the relatively low GFN in the coming 10 to 15 years. Every
annual funding gap is refinanced at the assumed market interest rate of
6 percent with a maturity of 5 years. Over the very-long-term horizon
considered in this analysis, the compounding of these relatively
expensive terms (as compared with today's low-interest environment)
leads to the very high GFN displayed in figure 3.
For instance, the total projected debt payments in 2050 amount to
62.1 billion [euro] under the chosen parameters. Of these, 42.6 billion
[euro] consists of the repayments of debt projected to be issued in 2045
to cover the financing gap in that year. Another 14.1 billion [euro]
consists of the projected interest due on the outstanding debt stock,
and only 5.3 billion [euro] are the interest and principal payments
scheduled to the EFSF as of 2015. Together with a projected primary
surplus of 11.5 billion [euro], the 2050 GFN comes down to 50.6 billion
[euro]. In other words, only about 10 percent of the GFN for that year
is based on terms known today, and the remainder rests on the accuracy
of the assumptions about market conditions over the coming 35 years.
This long projection horizon implies considerable uncertainty. The
gray area in figure 3 displays the results of a simple Monte Carlo
simulation of the projected GFN, showing the 25th and 75th percentile of
realizations. (22) For 2050, the interquartile range reaches from 6 to
27 percent. The policy implications of these two outcomes for the
requirement of debt restructuring today would of course be fundamentally
different. While a low GFN would imply healthy expected finances, a
value of 27 percent would even breach the higher bound of the GFN range
mentioned in table 1 by a wide margin.
The IMF has made debt operation a precondition for continued
involvement in Greece and has proposed that restructuring could take the
form of doubling the maturities on the European loans. However, due to
the significant grant element of the EFSF and GLF loans, a pure
reprofiling of this part of Greek debt will achieve a relatively smaller
reduction in the present-value debt stock than an extension of
maturities of earlier liabilities. Average maturities on the GLF and
EFSF loans are already more than 30 years. For every 1 [euro] due in 30
years, a doubling of maturities reduces the present value of this
liability by only 19 cents. Conversely, extending the term of 1 [euro]
coming due next year by only 10 years reduces the present value of that
obligation by 38 cents.
A debt operation that only extended the European loans would
therefore be relatively less efficient in achieving present-value debt
stock reductions than would a restructuring that included shorter-term
liabilities. Furthermore, it would not ease financing needs until the
mid-2020s, when those loans start coming due.
An extension of GLF or EFSF maturities would indeed bring down the
projected GFN by reducing the amount that has to be refinanced at the
assumed unfavorable market terms. But the effectiveness of such an
operation rests strongly on the accuracy of the macroeconomic,
financial, and fiscal projections. If market interest rates remain
elevated for Greece, GDP growth remains sluggish, or a primary balance
of 3.5 percent proves elusive over the next four decades, any return to
private sector funding will be difficult. Conversely, a more positive
outcome would make a maturity extension of the official loans obsolete.
III. Policy Implications
The analysis of this paper has policy implications on three levels.
First, there are implications for the ongoing debate about a
restructuring of Greek debt. We show that the nominal debt stock
projections paint far too bleak a picture of the actual burden.
Evaluated in present-value terms, Greek debt stands at about 100 percent
and will rise to about 120 percent under the new program, which is not
exceptionally high for advanced countries. Nevertheless, even in
present-value terms, Greece still breaches the thresholds of the
standard debt sustainability analysis for both market-access countries
and lower-income countries. The projection of GFN over the short to
medium runs, however, does not provide significant reasons for debt
relief. The long-term projections, while sending signals of critically
high funding pressure, are marked by very large uncertainties and thus
are not a reliable basis for deciding the restructuring need.
Consequently, the debt stock and projected payment flows show that,
despite the extraordinary amount of private and public debt relief
Greece has already received, further debt restructuring may be
advisable, although that conclusion is far less certain than commonly
argued. (23) If required, a debt operation should focus on the horizon
over which the payment flow projections are relatively reliable and a
restructuring would be relatively more efficient in reducing the debt
stock. But in the short to medium runs, the repayments are mostly owed
to the IMF and the ECB, and only a further extension of grace periods by
the European partners would reduce the risk of default on their loans.
Thus, an efficient debt restructuring will have to answer the question
of burden sharing and seniority within the official sector first.
Furthermore, a mere extension of maturities without a reduction in
the nominal value of long-run obligations will only extend the
interdependencies of Greece and its European creditors. In their paper
in this volume, Reinhart and Trebesch (2015) show that such long-term
financial dependencies create significant political tensions. Political
decisionmakers in the current debate should be well aware of such
frictions when engaging in debt operations, which can prolong such
potentially bruising negotiations for decades to come. This political
economy argument speaks for outright debt relief rather than further
prolongations. But a second historical lesson from Reinhart and
Christoph Trebesch (2015) is that Greece has been prone to quickly
overborrow again as soon as the previous debt crisis has been overcome.
This suggests that debt relief should only be granted after Greece has
demonstrated that it is able and willing to break away from the
historical pattern, and lends support for a process like that for the
Highly Indebted Poor Countries, whereby multilateral debt forgiveness is
granted after an extended track record of good policy has been
established.
The second set of implications concerns the mechanics of assessing
debt sustainability in the official sector. Both the ESM and the IMF
still apply the market-access framework to Greece, not taking into
account the effective present-value debt relief that has already been
granted. This is especially paradoxical in the case of the ESM, since
the grant element of up to 60 percent is only contained in European
loans. Nevertheless, the headline number of roughly 200 percent of
nominal debt-to-GDP stock is still used in public and in negotiations.
While the low-income country framework of the IMF and World Bank does
account for grant elements in official lending, it does not seem
appropriate for a case like Greece. In particular, the low-income
country framework sets different levels for debt sustainability
depending on the quality of institutions and policies. This may be sound
in principle, but not feasible in practice inside the eurozone.
Finally, there are broader implications for the European monetary
union as well as the international monetary system. The Greek debt
crisis has profoundly changed the architecture of the eurozone. In
addition to spurn ng reforms in EU fiscal governance, it has led to the
creation of the ESM as a permanent crisis-lending mechanism. Moreover,
Greece contributed to softening the ESM's lending framework and to
transforming it from an institution like the IMF to one like the World
Bank. The importance of this last step has not been sufficiently
recognized. On one hand, it has implications for the ongoing debate on
fiscal union, since the large grant element in European crisis lending
has added a fiscal buffer. On the other hand, the repeated softening of
lending conditions signals that the ESM has commitment problems and
strengthens the case for establishing an effective regime for sovereign
debt restructuring in the eurozone.
In contrast, the IMF's recent reforms seek to reverse the
softening of its crisis lending framework. By abandoning the systemic
exemption introduced in 2010, future lending decisions by the IMF should
take the results of debt sustainability analyses more seriously again.
Linking emergency loans to an obligatory maturity extension of existing
debt if it cannot be assessed to be sustainable with high probability
may help to overcome commitment problems. However, negative
externalities can always be expected in cases serious enough to require
exceptional access to the IMF. The framework will therefore have to
stand the test of future crises to reveal the true robustness of its
rules.
ACKNOWLEDGMENTS We thank the editors, the seminar participants at
the 2015 Brookings Papers Fall Meeting, and also Reza Baqir, Antonio
Fatas, Nicolas Sauter, Susan Schadler, Christoph Trebesch, and Jeromin
Zettelmeyer for very helpful comments.
References
Aguiar, Mark, and Manuel Amador. 2014. "Sovereign Debt."
In Handbook of International Economics 4, edited by Gita Gopinath,
Elhanan Helpman, and Kenneth Rogoff. Amsterdam: North-Holland.
Andritzky, Jochen. 2006. Sovereign Default Risk Valuation:
Implications of Debt Crises and Bond Restructurings. Berlin: Springer.
Baldacci, Emanuele, Iva Petrova, Nazim Belhocine, Gabriela
Dobrescu, and Samah Mazraani. 2011. "Assessing Fiscal Stress."
Working Paper no. 11/100. Washington: International Monetary Fund.
Berti, Katia, Matteo Salto, and Matthieu Lequien. 2012. "An
Early-Detection Index of Fiscal Stress for EU Countries." Economic
Paper no. 475. Brussels: European Commission.
Buchheit, Lee C., Anna Gelpern, Mitu Gulati, Ugo Panizza, Beatrice
Weder di Mauro, and leromin Zettelmeyer. 2013. "Revisiting
Sovereign Bankruptcy." Washington: Brookings Institution, Committee
on International Economic Policy and Reform.
Chauvin, Nicolas Depetris, and Aart Kraay. 2005. "What Has 100
Billion Dollars Worth of Debt Relief Done for Low Income
Countries?" Working Paper, http://
siteresources.worldbank.org/DEC/Resources/ChauvinKraayWhatHasDebtRelief
AccomplishedSept2005.pdf
Cruces, Juan J., and Christoph Trebesch. 2013. "Sovereign
Defaults: The Price of Haircuts." American Economic Journal:
Macroeconomics 5, no. 3: 85-117.
Dias, Daniel A., Christine Richmond, and Mark L. J. Wright. 2014.
"The Stock of External Sovereign Debt: Can We Take the Data at
'Face Value'?" Journal of International Economics 94, no.
1: 1-17.
Dikhanov, Yuri. 2006. "Historical PV of Debt in Developing
Countries, 1980-2002." Report no. 36153. Washington: World Bank.
Easterly, William. 2001. "Growth Implosions, Debt Explosions,
and My Aunt Marilyn: Do Growth Slowdowns Cause Public Debt Crises?"
Policy Research Working Paper no. 2531. Washington: World Bank.
Eaton, Jonathan, and Raquel Fernandez. 1995. "Sovereign
Debt." In Handbook of International Economics 3, edited by Gene M.
Grossman and Kenneth Rogoff. Amsterdam: North-Holland.
Eaton, Jonathan, and Mark Gersovitz. 1981. "Debt with
Potential Repudiation: Theoretical and Empirical Analysis." Review
of Economic Studies 48, no. 2: 289-309.
EFSF (European Financial Stability Facility). 2013a.
"Amendment Agreement Relating to the Loan Facility Agreement,
Financial Assistance Facility Agreement, and Master Financial Assistance
Facility Agreement between European Financial Stability Facility, the
Portuguese Republic as Beneficiary Member State, and Banco de
Portugal." Lisbon and Luxembourg.
--. 2013b. "Amendment Agreement Relating to the Loan Facility
Agreement, First Financial Assistance Facility Agreement, Second
Financial Assistance Facility Agreement, and Master Financial Assistance
Facility Agreement between European Financial Stability Facility,
Ireland as Beneficiary Member State, and Central Bank of Ireland."
Dublin and Luxembourg.
EU (European Union). 2010. "Loan Facility Agreement between
the Member States Whose Currency Is the Euro and the Hellenic
Republic." Brussels.
--. 2011. "Amendment to the Loan Facility Agreement of June
2011." Brussels.
--. 2012a. "Amendment to the Loan Facility Agreement of
February 2012." Brussels.
--. 2012b. "Amendment to the Loan Facility Agreement of
December 2012." Brussels.
Eurogroup. 2012. "Eurogroup Statement on Greece."
Brussels. http://www.
consilium.europa.eu/press/press-releases/2012/11/pdf/Eurogroup-statement-on-Greece-27-11/
European Commission. 2010. "Report on Greek Government Deficit
and Debt Statistics." Brussels.
--. 2014. "Assessing Public Debt Sustainability in EU Member
States: A Guide." Occasional Paper no. 200. Brussels.
--. 2015. "Greece: Request for Stability Support in the Form
of an ESM Loan." Brussels.
France, Treasury Agency. 2015. Monthly Bulletin, no. 302. Paris.
German Ministry of Finance. 2014. "Request of the Federal
Ministry of Finance: Grants in Favor of Greece: Technical Extension and
Continuation of the Stability Support" [in German]. Berlin: German
Bundestag.
House, Christopher, and Linda Tesar. 2015. "Greek Budget
Realities: No Easy Option." In present volume of Brookings Papers
on Economic Activity.
IMF (International Monetary Fund). 1995. "IMF Approves US$17.8
Billion Stand-By Credit for Mexico." Press Release no. 95/10.
Washington.
--. 2002. "Access Policy in Capital Account Crises."
Washington.
--. 2004. "Review of Exceptional Access Policy."
Washington.
--. 2008. "Reform of Quota and Voice in the International
Monetary Fund: Report of the Executive Board to the Board of
Governors." Washington.
--. 2009. "Greece: 2009 Article IV Consultation--Staff Report;
Staff Supplement; Public Information Notice on the Executive Board
Discussion; and Statement by the Executive Director for Greece."
Country Report no. 09/244. Washington.
--. 2010a. "Greece: Staff Report on Request for Stand-By
Arrangement." Country Report no. 10/110. Washington.
--. 2010b. "Minutes of Executive Board Meeting 10/45-1."
Washington.
--. 2011. "Greece: Fourth Review under the Stand-By
Arrangement and Request for Modification and Waiver of Applicability of
Performance Criteria." Country Report no. 11/175. Washington.
--. 2013a. "Sovereign Debt Restructuring: Recent Developments
and Implications for the Fund's Legal and Policy Framework."
Washington.
--. 2013b. "Staff Guidance Note for Public Debt Sustainability
Analysis in Market-Access Countries." Washington.
--. 2013c. "Unification of Discount Rates Used in External
Debt Analysis for Low-Income Countries." Washington.
--. 2013d. "Staff Guidance Note on the Application of the
Joint Bank-Fund Debt Sustainability Framework for Low-Income
Countries." Washington.
--. 2014a. "The Fund's Lending Framework and Sovereign
Debt: Preliminary Considerations." Washington.
--. 2014b. "IMF Executive Board Discusses the Fund's
Lending Framework and Sovereign Debt." Press Release no. 14/294.
Washington.
--. 2014c. Selected Decisions and Selected Documents of the
International Monetary Fund, no. 37. Washington.
--. 2015a. "Greece: Debt Sustainability Analysis (Preliminary
Draft)." Country Report no. 15/165. Washington.
--. 2015b. "Greece: An Update of IMF Staff's Preliminary
Public Debt Sustainability Analysis." Country Report no. 15/186.
Washington.
--. 2015c. "IMF Lending." Factsheet. Washington,
https://www.imf.org/ external/np/exr/facts/howlend.htm
--. 2016. "IMF Executive Board Approves Exceptional Access
Lending Framework Reforms." Press Release no. 16/31. Washington.
Italy, Department of the Treasury. 2015. "Government Debt:
Breakdown by Instrument as of July 31, 2015." Rome: Ministry of
Economy and Finances.
Kaminsky, Graciela, Saul Lizondo, and Carmen M. Reinhart. 1998.
"Leading Indicators of Currency Crises." IMF Staff Papers 45,
no. 1: 1-48.
Kaminsky, Graciela L., and Carmen M. Reinhart. 1999. "The Twin
Crises: The Causes of Banking and Balance-of-Payments Problems."
American Economic Review 89, no. 3: 473-500.
Kazarian, Paul. 2015. "Greece Sovereign Debt: Asking the Right
Questions." Speech given at Conference on Rethinking Sovereign
Debt. Munich: CESifo.
Kraay, Aart, and Vikram Nehru. 2004. "When Is Debt
Sustainable?" Policy Research Working Paper no. 3200. Washington:
World Bank.
Levy Yeyati, Eduardo, and Ugo Panizza. 2011. "The Elusive
Costs of Sovereign Default." Journal of Development Economics 94,
no. 1: 95-105.
Panizza, Ugo, Federico Sturzenegger, and Jeromin Zettelmeyer. 2009.
"The Economics and Law of Sovereign Debt and Default." Journal
of Economic Literature 47, no. 3: 651-98.
Public Debt Management Agency. 2015. Hellenic Republic Public Debt
Bulletin, no. 77. Athens.
Reinhart, Carmen M., and Christoph Trebesch. 2015. "The
Pitfalls of External Dependence: Greece, 1829-2015." In present
volume of Brookings Papers on Economic Activity.
Sandleris, Guido. 2015. "The Costs of Sovereign Defaults:
Theory and Empirical Evidence." Working Paper no. 02/2015. Buenos
Aires: Torcuato di Tella University Business School, Finance Research
Centre (CIF).
Schadler, Susan. 2013. "Unsustainable Debt and the Political
Economy of Lending: Constraining the IMF's Role in Sovereign Debt
Crises." CIGI Paper no. 19. Waterloo, Canada: Centre for
International Governance Innovation.
Schumacher, Julian, and Beatrice Weder di Mauro. 2015. "Debt
Sustainability Puzzles: Implications for Greece." London: Centre
for Economic Policy Research, VoxEU.
http://www.voxeu.org/article/debt-sustainability-puzzles-implications-greece
Serafeim, George. 2015. "Greece's Debt:
Sustainable?" Harvard Business School, Case no. 115-063. Cambridge,
Mass.: Harvard Business Publishing.
Tomz, Michael, and Mark L. J. Wright. 2013. "Empirical
Research on Sovereign Debt and Default." Annual Review of Economics
5: 247-72.
Van Rijckeghem, Caroline, and Beatrice Weder. 2009. "Political
Institutions and Debt Crises." Public Choice 138, nos. 3-4:
378-408.
World Bank. 2014. "IBRD/IDA Management's Discussion &
Analysis and Financial Statements." Annual Report. Washington.
Zettelmeyer, Jeromin, Christoph Trebesch, and Mitu Gulati. 2013.
"The Greek Debt Restructuring: An Autopsy." Economic Policy
28, no. 3: 513-63.
(1.) The troika consists of the European Central Bank, the European
Commission, and the International Monetary Fund.
(2.) A large body of theoretical literature emphasizes this point,
starting with Eaton and Gersovitz (1981) and summarized extensively in
the surveys by Eaton and Fernandez (1995); Panizza, Sturzenegger, and
Zettelmeyer (2009); and Aguiar and Amador (2014). The empirical
literature has identified many political and institutional factors that
have an impact on the probability of sovereign default; for example, see
Van Rijckeghem and Weder (2009), Tomz and Wright (2013), and Sandleris
(2015).
(3.) In its analytical framework, the IMF (2013b, p. 4) does make a
reference to the political feasibility of primary balance adjustments,
but this does not seem to depend on the country-specific political
situation.
(4.) In the following, we refer to the debt sustainability analysis
framework mentioned in the European Stability Mechanism treaty, which is
conducted jointly by the European Commission and European Central Bank,
as the "ESM framework."
(5.) The IMF's country quotas are expressed in terms of SDRs
(Special Drawing Rights), the IMF's reserve asset based on a basket
of four major currencies.
(6.) Access to this exceptional credit facility is determined by
additional variables beyond the country quota. Providing large loans to
countries in financial distress comes at greater risks, and granting
exceptional access has therefore been controversial ever since Mexico
received the first such loan of 688 percent of its quota in 1995 (IMF
1995). With the experience of further emerging markets crises in the
1990s in mind, the IMF formalized this instrument in the early 2000s
(IMF 2002, 2004).
(7.) The volume eventually borrowed was 52.9 billion [euro].
(8.) This interpretation was confirmed by the assessment of the
Greek loan application of July 2015. The EU concluded that although
direct financial risks of a Greek default were small, they would create
"significant doubts on the integrity of the euro area as a whole,
currently and in the future" (European Commission 2015, p. 5).
(9.) The maximum term on World Bank loans, under the International
Bank for Reconstruction and Development, is 35 years, with an average
weighted maturity of 20 years (World Bank 2014). Nonconcessional IMF
loans are due much earlier, with final maturities from 3.25 to 5 years
for Stand-By Arrangements and 4.5 to 10 years for the Extended Fund
Facility; even for concessional loans, the IMF expects repayment within
8 to 10 years (IMF 2015c).
(10.) The fact that this constituted a change in policies was not
obvious to the board and led to a heated discussion once one director
pointed it out. The directors first thought that this exception would
only be applied to Greece, but the Legal Department of the IMF explained
that it would carry over to all member countries due to equal treatment
requirements (IMF 2010b).
(11.) The systemic exemption has already been invoked 34 times for
eurozone programs and reviews.
(12.) Schadler (2013) also discusses whether Greece really
presented circumstances that warranted changing the lending criteria.
She distinguishes between the immediate moment--when the counterfactual
to the IMF's involvement would have entailed large systemic
cost--and the continuing involvement of the IMF over the following
years.
(13.) Moreover, it may increase the risk to the IMF's own
resources and its seniority status, as evidenced by the default of
Greece on payments to the IMF in the summer of 2015.
(14.) During the board discussion of this proposal, some directors
preferred to keep the systemic exemption as a "pragmatic way to
safeguard financial stability in an increasingly integrated world and
avoid the perception of lack of evenhandedness" (IMF 2014b).
(15.) In particular, this includes negative shocks to the primary
balance, real GDP growth, nominal interest rates, the exchange rate, and
contingent liabilities.
(16.) In the online appendix, we report alternative results based
on Greek and German market-discount rates. Online appendixes to all
papers in this volume may be found on the Brookings Papers webpage,
www.brookings.edu/bpea, under "Past Editions."
(17.) Moreover, Eurostat noted that it could not verify the new
figures and flagged the risk of further upward revisions. The ministers
of the Economic and Financial Affairs Council immediately mandated the
European Commission to investigate. The resulting report concludes that
the quality and the governance of Greek fiscal statistics are seen as
insufficient, noting that "revisions of this magnitude ... have
been extremely rare in other EU Member States, but have taken place for
Greece on several occasions" (European Commission 2010, p. 3).
(18.) The conditions of EFSF lending were amended accordingly, and
the other EFSF program countries--Ireland and Portugal--similarly
benefited from debt relief in the form of significantly increased
maturities. The maturity on Ireland's loan was increased from
2016-29 to 2029-42, and on Portugal's loan from 2015-38 to 2025-40,
increasing the average weighted maturity to more than 20 years (EFSF
2013a, 2013b).
(19.) See Public Debt Management Agency (2015). To put this in
perspective, the average maturity terms of Italian and French sovereign
debt are 6.5 and 6.9 years, respectively (Italy, Department of the
Treasury 2015; France, Treasury Agency 2015).
(20.) Not including the new ESM program.
(21.) A previous version of these and the following computations
can be found in Schumacher and Weder di Mauro (2015).
(22.) The analysis rests on 1,000 draws of independent and
identically distributed shocks to the growth rate (g), the interest rate
(r), and the primary balance (pb), each drawn from a normal distribution
with standard deviations matched to the historical data between 2001
(entry into the eurozone) and 2009 (last precrisis year). Specifically,
for a normal distribution N([mu], [[sigma].sup.2]), the shock
distributions are g~N(2.1,3.5), r~N(4.6,0.6), and pb~N(-2.3,3.6), where
[mu] and [[sigma].sup.2] are measured in percent.
(23.) See House and Tesar (2015) for an analysis of the required
budgetary adjustments if no debt reduction is implemented.
JULIAN SCHUMACHER
Johannes Gutenberg University Mainz
BEATRICE WEDER DI MAURO
INSEAD *
* The author is currently on leave from Johannes Gutenberg
University Mainz.
Table 1. Thresholds for Risk of Debt Distress under
Different Debt Sustainability Analysis Templates
European Commission
EU member states
Nominal debt (c) 90
Gross financing needs (d) 15
Debt profile
Bond spreads (basis points) (e) <231 (f) <276.6 (g)
External financing requirement (h) --
Foreign currency debt (i) 29.82
Nonresident-held debt (j) 49.02
Change in short-term debt (k) 2.76
PPG external debt ratios (l)
GDP --
Exports --
Revenues --
PPG external debt service ratios (m)
Exports --
Revenues --
International Monetary Fund
Market-access countries (a)
Advanced economies
Nominal debt (c) 85
Gross financing needs (d) 20
Debt profile Low Moderate High
Bond spreads (basis points) (e) <400 400-600 >600
External financing requirement (h) <17 17-25 >25
Foreign currency debt (i) --
Nonresident-held debt (j) <30 30-45 >45
Change in short-term debt (k) <1 1-1.5 >1.5
PPG external debt ratios (l)
GDP --
Exports --
Revenues --
PPG external debt service ratios (m)
Exports --
Revenues --
International Monetary Fund
Market-access countries (a)
Emerging markets
Nominal debt (c) 70
Gross financing needs (d) 15
Debt profile Low Moderate High
Bond spreads (basis points) (e) <200 200-600 >600
External financing requirement (h) <5 5-15 >15
Foreign currency debt (i) <20 20-60 >60
Nonresident-held debt (j) <15 15-45 >45
Change in short-term debt (k) <0.5 0.5-1 >1
PPG external debt ratios (l)
GDP --
Exports --
Revenues --
PPG external debt service ratios (m)
Exports --
Revenues --
International Monetary Fund
Low-income countries (b)
Weak Medium Strong
Nominal debt (c) 38 56 74
Gross financing needs (d) --
Debt profile
Bond spreads (basis points) (e) --
External financing requirement (h) --
Foreign currency debt (i) --
Nonresident-held debt (j) --
Change in short-term debt (k) --
PPG external debt ratios (l)
GDP 30 40 50
Exports 100 150 200
Revenues 200 250 300
PPG external debt service ratios (m)
Exports 15 20 25
Revenues 18 20 22
Sources: European Commission (2014), International Monetary Fund
(2013b, 2013d).
(a.) Market-access countries are designated as "advanced economies"
or "emerging markets," based on their World Economic Outlook
classification. Debt profile indicators further disaggregate
market-access countries based on their levels of risk.
(b.) Countries qualify as "low-income" if they are eligible for the
IMF's Poverty Reduction and Growth Trust (PRGT). A country is
eligible for the PRGT if its annual per capita gross national
income is below $1,215 and it does not have the capacity to access
international financial markets on a durable and substantial basis.
Low-income countries are further disaggregated by the quality of a
country's policies and institutions, as measured by the World
Bank's Country Policy and Institutional Assessment (CPIA).
Countries with a stronger CPIA rating are considered to be more
resilient to indebtedness, and therefore higher thresholds are
applied.
(c.) Nominal debt level as a percent of GDP. For EU member states
and market-access countries, nominal debt is expressed at face
value; for low-income countries, at present value.
(d.) Sum of debt payments and individual payments or revenues (such
as privatizations) less the primary balance in any given year, as a
percent of GDP.
(e.) Government bond yield spreads relative to German and U.S.
bonds of similar maturity.
(f.) 10-year benchmark.
(g.) 2-year benchmark.
(h.) Current account balance plus amortization of total short-term
external debt at remaining maturity, as a percent of GDP.
(i.) Public debt held in foreign currency as a percent of total
public debt.
(j.) Public debt held by nonresidents as a percent of total public
debt.
(k.) Annual change in short-term public debt (at original maturity)
as a percent of total public debt.
(l.) Present value of public and publicly guaranteed external debt,
as a percent of various economic indicators.
(m.) Public and publicly guaranteed external debt service, as a
percent of various economic indicators.
Table 2. Greek Loan Conditions over Time, 2010-14
Greek Loan Facility (a) May 9, 2010 June 14, 2011
Margin: 300-400 basis points 200-300 basis points
Grace period: 3 years 4.5 years
Maturity: 5 years 10 years
European Financial
Stability Facility (a)
Margin:
Fees:
Grace period:
Average maturity:
Private
FV reduction:
PV reduction:
Greek Loan Facility (a) Feb 27, 2012 Dec 19, 2012 (b)
Margin: 150 basis points 50 basis points
Grace period: 10 years 10 years
Maturity: 15 years 30 years
European Financial
Stability Facility (a) March 1, 2012 Dec 12, 2012 (b)
Margin: 0 basis points 0-200 basis points
Fees: >15 basis points >5 basis points
Grace period: 0 years 10 years
Average maturity: 17.5 years 32.5 years
Private March 9, 2012 Dec 12, 2012
FV reduction: 53.5 percent 64.6 percent
PV reduction: 64.4 percent 61.4 percent
Sources: For the Greek Loan Facility, EU (2010,2011,2012a, 2012b);
European Financial Stability Facility (2013a, 2013b); Eurogroup
(2012); and Zettelmeyer, Trebesch, and Gulati (2013).
(a.) The first entries for the Greek Loan Facility and the European
Financial Stability Facility denote the original terms of the
loans. The remaining entries denote the terms of subsequent
amendments.
(b.) The November 2012 agreement of the Eurogroup in which the
Greek Loan Facility and European Financial Stability Facility
restructurings of December 2012 were announced contained further
measures to ease the Greek debt burden that were not part of the
implemented agreements. In particular, these include a commitment
to pass on profits from the bond purchases under the European
Central Bank's securities markets program, and that further
adjustments of the loan conditions would be considered conditional
on the successful implementation of the reform program.
Table 3. Greek Debt Composition
Face
value (a) Interest (b)
(billions (billions
Debt of euros) of euros)
Treasury bills 14.8 --
Private sector involvement bonds 30.5 20.7
Holdout bonds 2.8 0.7
Bonds held by the Eurosystem (c) 23.6 4.9
International Monetary Fund 19.5 2.5
European Financial Stability Facility 131.0 27.8
Greek Loan Facility 52.9 13.5
Bank of Greece (f) 4.8 --
Total (without new program) 280.1 70.0
Percent of GDP 156
New program (g)
European Stability Mechanism 77.8 31.7
International Monetary Fund 8 1
Total new program 86.0 32.7
Total 366.1 102.6
Percent of GDP 204
Present Grant
value (c) element
(billions (percent)
Debt of euros) (d)
Treasury bills 14.6 n/a
Private sector involvement bonds 26.8 12
Holdout bonds 2.9 -2
Bonds held by the Eurosystem (c) 24.2 -2
International Monetary Fund 18.6 5
European Financial Stability Facility 51.4 61
Greek Loan Facility 33.2 37
Bank of Greece (f) 3.8 n/a
Total (without new program) 175.6 37
Percent of GDP 98
New program (g)
European Stability Mechanism 31.9 59
International Monetary Fund 8 8
Total new program 39.5 54.1
Total 215.1 41.2
Percent of GDP 120
Sources: IMF (2013d), for the definition of the grant element; for
the data, the various data sources cited in the paper (IMF, GLF,
EFSF).
(a.) Sum of principal payments, as in equation 1.
(b.) Undiscounted sum of interest payments due to each creditor.
(c.) Discounted sum of principal and interest payments, as in
equation 2.
(d.) Defined as (face value - present value) / face value.
(e.) Includes bonds held by the European Central Bank, the European
Investment Bank, and various national central banks.
(f.) Assuming constant amortization.
(g.) Assuming the new program is identical to the previous programs
with respect to the share of IMF and European lending.
