Published on 4th March 2021 in German at MakronomMagazin
https://makronom.de/wie-sich-die-corona-schulden-entsorgen-lassen-38585
Like other pandemics before it, the current
Corona pandemic will eventually end - either "medically" or "socially". The medical
end occurs when the number of people who fall ill drops sharply. The social
end takes place mainly in people's minds. It occurs when the fear of the
disease decreases, people no longer want to accept the restrictions - and learn
to live with the disease.
The same applies to the economic policy
course, which will have to deal with the aftermath of the crisis. In Germany,
the public dispute about the Corona pandemic has been very poisoned for quite
some time, and unfortunately this increasingly applies to economic policy
debates as well. Thus, in the meantime, it seems that even invective with ad hominem attacks in the ostensibly serious daily
press regarding the financing of the pandemic measures is acceptable. This
article is intended to help keep a cool head in the debate by outlining the
various options for dealing with the national debt.
Historical lessons
The social and economic consequences of the
virus have acted as an exogenous shock on public debt. The pandemic will leave
behind high public debt/GDP ratios. There is ample historical evidence to learn
from: The last 150 years have provided us with enough illustrative material on
how advanced nations have dealt with debt. The Great Depression of 1929-33, the
world wars of the 20th century and the world financial crisis of 2007-09 were
also exogenous shocks that resulted in high government debt ratios in advanced
countries. The Federal Reserve's Volcker monetary shock inflated emerging
market sovereign debt in the 1980s via the dollar, commodity prices and
interest rates. High public debt was also endogenous in these countries,
for example as a result of balance sheet currency mismatches or loss-making
state-owned enterprises.
Three notable periods of public debt relief
can be identified, in the decades before World War I and in the period from the
end of World War II to the beginning of the 1970s. The mean value of government
debt often reached around 150% of GDP before being reduced to a value of around
40%. What were the main drivers of
this erosion?
- A recent study by Eichengreen et al. describes three successful episodes
of debt consolidation before World War I: Great Britain after the
Napoleonic Wars, the United States in the last third of the 19th century
and France in the decades before 1913. The Napoleonic Wars, the
Franco-Prussian War and the US Civil War were the three most expensive
military conflicts of the 19th century, resulting in debt-financed war
spending. Debt relief after these wars was carried out in all three states
primarily through primary surpluses in view of positive interest
rate-growth differentials (interest rates were higher than growth).
- After the Second World War, debt ratios in the developed
economies fell rapidly. They reached the pre-World War I level in the
1960s. However, this was primarily caused not by budget surpluses but by the rapid expansion of nominal GDP. The post-war boom
had various labels: Erhard's "economic miracle years" in
Germany, "Les Trente Glorieuses" in France, "Il Boom
Economico" in Italy. In the case of Germany, the debt cancellations
of 1948 and 1953 by the Allied victorious powers also depressed the debt
ratio.
- Towards the end of the 20th century, public debt ratios in the
poorest countries fell rapidly and initially remained at a fairly low
level. The main reason for this was the HIPC initiative of the Bretton
Woods sisters in 1996 in favour of mostly African low-income countries.
These had not benefited from the Brady Plan of 1989, which had mainly provided debt relief to Latin American
emerging countries.
Today, after a year of the Covid pandemic,
the Maastricht criterion of a 60% ceiling for the debt-to-GDP ratio is being
blatantly missed, especially in Latin Europe. How can the highly indebted EU
states get off their Covid debt? And: does that even make sense with the
current interest "costs", which are close to or occasionally below
zero?
Debt dynamics
Public debt as a percentage of GDP only
declines if nominal GDP growth exceeds interest and is not compensated by the
interest-adjusted budget deficit (primary deficit). In the (first) Covid year
2020, the public debt ratio skyrocketed despite massive EU support, especially
in Latin Europe, which suffered additionally from the loss of tourism.
The purchase of government bonds by
the ECB and the national central banks is thus broadly equivalent to debt
relief for the state
In 2020, the ratio of growth to interest
rates was naturally unfavourable due to the covid recession. But if the euro
countries recover economically in the current year, this relationship will be
reversed. On the one hand, market interest rates are currently still negative
or very low, on the other hand, governments de facto no longer have to pay
interest on their outstanding bonds held by the Eurosystem: The government pays
interest to the central bank, which now holds the bonds, but the central bank
usually returns this interest income to the government. So the purchase of government
bonds by the ECB and the national central banks is broadly equivalent to debt relief for the government.
Debt relief
Led by Thomas
Piketty, over 100 economists have proposed debt relief. They argue that this would allow
governments to issue new debt unencumbered by old debt to finance major
projects. In doing so, the signatories also refer to the London Debt Agreement
of 1953, a radical plan to cancel half of Germany's foreign debt and create
generous repayment terms for the remaining debt. The agreement boosted Germany's
economic growth by creating fiscal space for public investment, lowering
borrowing costs and stabilising inflation.
However, the idea that a selective default
on the debt held by the ECB would be without consequences is unrealistic. Such a decision would at least have the effect
of closing the umbrella offered by the ECB today and increasing the cost of
re-leveraging or refinancing the remaining debt. Debt relief limited to
individual countries is also ruled out because of the immediate contagion
effects. Finally, from a purely accounting point of view, the benefit of
repudiation would be zero in the short term: after all, the debt repurchased by
the ECB was acquired by the central banks of the Eurosystem, which on-lend the
proceeds accruing to it to their finance ministers.
Inflation
As long as the ECB sticks to the inflation
target of 2% (and does not allow medium-term overshoots), it must reduce base
money M0 if it misses the target. The ECB sells either government bonds or its
own interest-bearing bonds, thus taking back the seigniorage it granted the
government when it bought the bonds.
It would be different if the ECB allowed more inflation in the future, i.e. did
not counteract it when inflation was above 2%. Then it would not have to sell
the bonds (or issue its own bonds). In this case, higher inflation would reduce
the real value of government debt that is not on the central bank's balance
sheet and that has been issued at very low interest rates in recent years.
Governments would win first, while private bondholders would have to
"pay" for the higher inflation. Nominal interest rates would rise,
lowering the price of the long-term government bonds that these investors
bought at negative or zero interest rates. The inflationary surprise loss would
subsequently make government bond financing more difficult and more expensive.
In addition, tax revenues could fall in real terms when inflation is high as a
result of the so-called Tanzi effect.
Financial repression
Carmen Reinhart and Bélen Sbrancia have
shown in a widely acclaimed study how governments could get rid of
their debt burdens by means of financial
repression. Financial repression, like a tax on bondholders and savers
through negative or below-market real interest rates, reduces government debt.
It is most successful in liquidating debt when accompanied by moderate
inflation.
After World War II, capital controls and
regulatory restrictions created compulsory buyers of government debt and
limited the erosion of the tax base. From 1945 to 1980, interest rates in
advanced countries were negative about half the time - so savers paid on top
when they lent money to the government. Britain and the United States
liquidated debts averaging 3 to 4% of GDP annually as a result. In Australia
and Italy, where inflation was particularly high, liquidation rates averaged
more than 5%. Average annual savings in interest expenditure for a sample of
twelve countries range from about 1% to 5% of GDP for the entire period
1945-80.
However, the fiscal yield of financial
repression will be lower today in the Eurozone than it was then. Even Greece
was able to sell an eight times oversubscribed ten-year government bond
to investors at 0.8% just a few weeks ago. In addition, 80% of Greek government
debt is held by public creditors such as the Euro Stability Fund (ESM). As part
of the Pandemic Emergency Purchase Programme (PEPP), the ECB is also buying
Greek government bonds, despite the securities being classified as
non-investment grade.
Austerity
The key determinant of future debt stability
is the ratio of interest rates to growth, more specifically the average cost of
debt minus nominal GDP growth. This interest-growth differential is the black
box of debt dynamics: countries where nominal GDP grows at a rate identical
to the average cost of debt can stabilise the debt-to-GDP ratio by maintaining
a balanced primary budget. States whose GDP in nominal local currency grows
faster than the average cost of debt can stabilise debt by running a primary
budget deficit (the size of this debt-stabilising primary balance or DSPB is
determined by the level of debt relative to GDP in the previous year and the
difference between interest and growth). In contrast, countries that pay more
on their debt than nominal GDP grows must run primary surpluses to stabilise
the debt-to-GDP ratio.
The interest rate-growth differential
is the ´black box´ of debt dynamics
Using a historical database of average
effective government borrowing costs for 55 countries over a period of up to
200 years, Paolo Mauro and Jing Zhou documented in an IMF study that negative interest
rate-growth differentials occurred more frequently in both advanced and
emerging market economies and often persist over long historical periods. In
such periods, the debt ratio erodes with comparatively little fiscal
discipline.
At the IMF, such a result cannot simply be
left standing. Therefore, Mauro and Zhou point to the low information content
of average interest costs: the default history of sovereigns shows that after
longer periods of low differentials based on average effective interest rates,
marginal borrowing costs can suddenly and sharply increase and countries can be
excluded from the financial markets in the short term. And even if interest rates are quite low,
this does not mean that there is a fiscal free lunch in dynamically
efficient countries, as Ricardo Reis,
currently much respected among "fiscal hawks", argues. The label "dynamic efficiency" does not apply to Germany, as the Federal Republic saves too much at the expense of
current generations and produces high current account surpluses.
The Maastricht criterion has lost all
credibility and thus binding after the Covid pandemic
The following table gives an impression of
the scope of the task: The "debt ratio stable primary budget
position" (DSPB) reflects the calculations of the rating agency S&P for the necessary
primary budget of the countries to avoid a further increase in the respective
debt ratios until 2023.* In Belgium, France, Italy and Spain, the primary
balance would have to increase by more than 10 percentage points of GDP
compared to the negative balance of 2020 in order to consolidate the budget in
this way. The interest-adjusted primary budget will have to come down from the
negative balance of 2020 at the latest after the end of the pandemic. In
principle, this can only be done by increasing tax revenues or cutting
spending, whereby an increase in tax rates can be just as counterproductive in
fiscal terms as spending cuts are harmful in terms of growth policy, even if
the EU's reconstruction fund, financed for the first time by common European
bonds, will cushion some hardships.
The shortcoming of such calculations of
debt dynamics is, among other things, that a higher debt ratio requires a lower
primary surplus; therefore, according to S&P, Greece would manage with less
austerity. The last column therefore shows my calculations of how many years it
would take to squeeze government debt back below the Maastricht criterion of
60% at S&P's projected interest rate differentials to growth. In some
cases, it would take several generations.
Parameters of debt dynamics in some EU
countries (2020)
|
Country |
DEBT RATIO (IN % OF GDP) |
PRIMARY BUDGET (IN % OF GDP) |
DSPB* 2023 |
YEARS TO
MAASTRICHT** |
|
Greece |
205.6 |
-7.1 |
-0.3 |
30 |
|
Italy |
155.8 |
-7.6 |
6.2 |
96 |
|
Portugal |
143.2 |
-4.2 |
0.9 |
32 |
|
Spain |
117.1 |
-9.9 |
6.5 |
22 |
|
France |
113.9 |
-8.5 |
5.6 |
24 |
|
Belgium |
113.8 |
-8.8 |
7.4 |
53 |
|
Median Eurozone |
86.1 |
-6.7 |
4.6 |
13 |
|
Memo: Germany |
66.6 |
-5.6 |
4.0 |
4 |
*Debt ratio stable primary budgetary position (DSPB) until 2023. R=(D/Y)/0.6. **"Years to Maastricht" calculated from J=-(ln R/(g-i,%)). Sources : own calculations; S&P Global (2021), Sizing Sovereign Debt and the Great Fiscal Unwind
Because of many uncertainties, precise
figures are less important here than the message: the Maastricht criterion has,
in my view, lost all credibility and thus commitment after the Covid pandemic.
Perpetual debt
Among others, George Soros or Guy Verhoefstadt have recently called for the issuance of perpetual
bonds, also called consols or perpetuals. These have an
infinite maturity and pay an annual coupon. Their principal is never repaid.
The main argument for issuing perpetual bonds is that the EU should take
advantage of the low interest rate environment to lock in low interest rates
through perpetual bonds. Perpetual bonds are also not subject to refinancing
risk as they never need to be rolled over.
Would perpetual bonds be a low-cost
repository solution? Given today's very low yields, it may be worthwhile for
issuers to take on long-term debt at favourable conditions. But financial
mathematicians are rather unimpressed . With a
positive yield, a perpetual bond is a comparatively expensive source of funding
compared to issuing ten-year bonds, which currently trade at rates below 0%.
The yield is the average interest rate over all coupon terms, weighted by the
present value of each interest coupon. However, it is difficult to make this
calculation for a perpetual bond because we cannot observe interest rates to
infinity. The ECB yield curve stops at 30 years because there are few bonds
with a maturity longer than 30 years.
Yield curve of Eurosystem government
bonds
Source: ECB
The well-rated Republic of Austria (S&P
rating: AA+) is a pioneer in Euroland for government bonds with extremely long
maturities. It issued a 100-year bond with a coupon of 2.1% in 2017 and
increased it to six billion euros in 2019. Currently, the yield is just below
one per cent. The 51-year bond with a coupon of 0.5% of the less well-rated
France (S&P: AA) currently yields similarly high. The question is above all
who will want to invest in such long-dated bonds when
yields rise at the long end.
There is no such thing as a 'jack of
all trades', but any variant will produce winners and losers, whether they are
governments, their taxpayers or bondholders.
Currently, there is a high demand for these
ultra-long maturity bonds, so their monetisation by the Eurosystem is not
necessary. Why? Institutional investors or endowment funds usually use
long-dated bonds to extend the maturity of their bond portfolios. The addition
on their asset side helps them to reflect the maturity structure of their obligations
(liabilities). Austria, like many other developed countries, is considered
low-risk, so investing in the 100-year bond will at least give investors a
positive long-term return - in contrast, many shorter-dated government bonds in
Europe currently have negative returns.
Bonds with ultra-long durations also benefit from "positive
convexity": if investors own long-dated bonds with low coupons (low
payouts), their purchase price increases more when yields fall. Conversely,
when yields rise, investors benefit from an implicit asymmetry of highly convex
bonds. Indeed, the price of a bond with high convexity falls less when yields
rise than it rises when yields fall.
Every variant has winners and losers
And the moral of the story? The multitude
of historical examples of debt consolidation today inform us about different
options: with and without austerity, with and without (de facto) debt repudiation,
with and without financial repression or inflation. Surely, combinations of
different variants are also possible, which allows policymakers to pursue the
strategy of consolidating or disposing of high pandemic debt with multiple
instruments.
It should also have become clear that there
is no such thing as a 'jack-of-all-trades', but that each variant will produce
winners and losers, whether they are governments, their taxpayers or
bondholders. Who wins or loses exactly how much and when must ultimately be
decided in the democratic process. For the selection and weighting of these
options is the responsibility of neither economists nor journalists, but
parliaments, governments and their voters, who should be as well informed as
possible about the consequences of these decisions.