Why is it that Italy causes such a stir in financial markets when
proposing a budget? Is it politics or is the stability of the financial
system at stake? In our assessment, the best way to avert a crisis is to
allow market forces to play out. Let me explain.
We all “know” Italy is in trouble. Well, before we jump to conclusions,
let’s look at a few charts. Above is the Italian unemployment rate; it’s come
off a high level, but still elevated. When policy makers call for structural
reform, it is a codeword for increasing flexibility in the labor market, i.e.
making firing easier. If firing workers is difficult, companies won’t hire
workers. It’s also in this context that providing a so-called basic income is
criticized by some as providing a disincentive to join the labor force (aside
from cementing higher deficits for years to come). Basic income means you get
paid, whether you work or not. In practice, the devil is in the details, as
European workers have long enjoyed unemployment benefits; streamlining such
benefits might actually save the government money. That said, Germany’s low
unemployment, to a significant extent, may be due to the fact that welfare
benefits were curtailed in 2002 (with a social democrat as chancellor),
providing an incentive for workers to join the workforce.
Moving on, consumer confidence in Italy is actually doing just fine:
However, economic sentiment shows not everything is great – a problem as
it suggests investments may be held back:
The big elephant in the room is Italy’s deficit. Italy’s total debt
is larger than Germany’s,
although the German economy is almost twice as large as Italy’s. As a
percentage of Gross Domestic Product, Italy’s debt at 131.8% is rather high.
Still, as can be seen from the chart below, Italy’s annual budget deficit has
been shrinking for several years, most recently at 2.3%
Looking at the above chart, I can’t help but muse that weak or
technocratic governments can’t get much spending done, but strong governments
can. Italy currently has a populist government that intends to provide an
economic stimulus. Keep in mind Italy is in its 66th government since 1946
(also note EU elections are coming up next year, with the coalition partners
in Italy competing against one another on the EU stage).
The chart below depicts the spreads of French, Italian and Spanish 10-year
bonds over that of Germany. Germany has the lowest bond yield in the
Eurozone; as such, the spread is the difference in yield, reflecting the
higher borrowing costs of these countries for 10-year debt:
As can be seen, Italian debt had rallied up to a few days ago when the
market expectations of a deficit below 2% for next year dwindled as
concessions to the coalition partners lead to an announcement that next
year’s deficit will be 2.4% (note: this is still a draft budget; and it’s
also a target deficit, the actual deficit may well be different).
The reason why folks outside of Italy care about any of this is because
Italy’s finances matter to the rest of the Eurozone. Notably consider the
long-term and short-term charts of the euro versus Italian bond yields
(y-axis for yields is inverted):
To be more specific, Italy has started to matter since the financial
crisis.
All that said, Italy has made progress in some areas that are worth
mentioning, notably non-performing loans:
As a point of reference, the U.S. federal government estimates its budget
deficit to be 4.2% of GDP in fiscal 2018. The possibly most significant
difference between the U.S. and Italy is that the U.S. has its own currency.
As such, the risk of U.S. government defaulting on its debt is minimal as it
could always print its own currency; this creates a risk that the dollar’s
purchasing power may erode, but minimizes default risk. In contrast, Italy is
part of the euro and, as a result, has waived its right to print its own
money.
In relinquishing authority to print money, Eurozone countries like Italy
can be thought of more like states in the U.S., such as California. The key
difference, however, is, that the U.S. government has broad fiscal powers,
whereas there is not even an EU Treasury Secretary. The 2018 EU budget is
€160 billion, a rounding error compared to the U.S. federal budget of over
US$ 4 trillion.
All of this is by design as countries, such as Germany, don’t want to cede
fiscal control to Brussels, afraid deficits would balloon. Ironically, a lot
of the populist movement sweeping developed countries is about taking control
back to the national level. As such, Italy should conceptually embrace that
the EU doesn’t have its own Treasury. What Italy has a problem with is that
Brussels is reviewing Italy’s budget in an attempt to tell it what to do.
That’s because the Eurozone has agreements on deficits, notably to keep
annual deficits below 3%; and, for highly indebted countries, to actively
pursue measures to reduce their debt to GDP levels. And while 2.4% deficits
may not sound like much, when the economy grows slowly, total debt-to-GDP
will increase even with a deficit of “only” 2.4%.
This makes the Italian budget a political battle: firebrand politicians
versus bureaucrats.
Some Italian politicians have also indicated deficits wouldn’t matter if
the European Central Bank (ECB) explicitly guaranteed Italian debt, purchased
it in unlimited amounts.
In my humble opinion, policy makers should be realistic and admit:
- Italian politicians cannot print their own money; and
- EU bureaucrats won’t be able to stop Italian politicians
from pursuing their agenda.
I mention in my introduction that markets function best when market forces
are allowed to play out. In the context of the above statement there’s a
simple solution. Mr. Draghi should rescind his July 2012 statement that he
would do “whatever it takes.” You cannot have your cake and eat it.
Long before the financial crisis, when Wim Duisenberg was still the head
of the ECB (he’s the predecessor to Trichet, who in turn was Draghi’s
predecessor), I argued that the way to make the Euro sustainable in the
long-term would be for it to embrace, rather than fight, differences in
fiscal discipline in the Eurozone. Let the policy makers do what they want to
do and be held accountable, but not by Brussels bureaucrats: let policy
makers be held accountable by the markets.
If we look back at the financial crisis, arguably the most powerful
incentive for reform has been the market. Since the financial crisis, several
rescue programs have been created, although it had also become clear that
governments prefer to pursue extreme measures before asking for help when the
cost of the help is to give up sovereign fiscal control. In this model,
countries pursue their own fiscal policy; if they need help, they give up
sovereign control. In that context, let me mention that I don’t endorse some
of the chaos in the markets at the peak of the financial crisis. But be aware
that crises will always happen; they are, however, less stressful when sound
institutional processes are in place.
I have long argued that much of the work of European regulators should be
focused on making the system more robust to failure. In the U.S., we have
bank failures all the time, yet there’s no crisis because there is a robust
process in place (the FDIC steps in). Similarly, especially in recent years, many
steps have been undertaken to make the Eurozone more robust. In my
assessment, that work needs to continue with the goal for the Eurozone to be
able to stomach a sovereign default within the Eurozone.
There are those who argue Italy is too big to fail. That shouldn’t stop
anyone from pursuing sound regulatory and monetary policy. Incidentally, just
as regulatory policy is working to make the system more robust, monetary
policy is moving towards holding countries more accountable. Consider:
- The ECB is expected to phase out its quantitative easing
(“QE”).
- A first interest rate hike in quite some time may well
come before Draghi retires.
- When Draghi retires, my best guess is a less dogmatic
ECB chief will take charge. Consider that a lawyer became head of the
Fed. The pendulum is swinging away from activist central bank chiefs.
- Inflationary pressures are gradually increasing in the
Eurozone. Note the ECB’s inflation target is on headline inflation, not
core inflation (Duisenberg and Trichet respected that; Draghi has all
but ignored this despite his claim he pursues his QE policies to achieve
the ECB’s legislated mandate). Draghi has indicated US fiscal stimulus
may cause an inflationary spillover to the Eurozone; and that EU fiscal
policy is also expansionary (more so if Italy implements its proposed
budget!).
I mention the above because, in my assessment, they point to higher rates
in the eurozone, as well as higher spreads. Higher spreads because, in our
analysis, the main achievement of quantitative easing throughout the world
has been a reduction of risk premia, i.e. a tightening of spreads, making
risky assets appear less risky. As QE goes into reverse, risk premia ought to
rise, causing spreads to increase.
In plain English, Italy should get used to paying more for its debt.
Policy makers and market participants alike should stop the illusion that
Italy will some day pursue German fiscal policy. It will not be the end of
the world, nor the end of the Eurozone because risk friendly capital will
gladly hold Italian debt – at the right price. The problem with a risk-free
illusion is that you coerce risk-averse capital to hold risky securities.
That’s a problem because the moment a reality check kicks in, risk-averse
capital jumps ship. Aligning expectations with reality fosters healthy
markets; artificial barriers create fragile markets bound to break under
pressure.
As I indicated, we are moving in this direction as risk premia are bound
to rise and regulators will continue to work to strengthen financial institutions.
In this context, I argue the flareups we see in the markets will, with
hindsight, be considered noise. I’m not suggesting there won’t be turmoil,
but those who seek the relative safety of German bunds will pursue those;
others interested in the higher yield may purchase Italian debt. What is
unhealthy is to chase yield and live under the illusion that the additional
yield is risk free.
The argument against this tough-love argument is that it could break
countries, give rise to extreme parties. I would counter that in arguing that
QE has been a key reason why populist governments have gained influence as
one of its many unintended consequences. Printing money does not solve
problems, only shifts the burden. At the peak of the financial crisis, I argued
the best short-term policy may be a good long-term policy. I stand by that:
stability is a result of a predictable, sound institutional framework. When
the rules are clear, it ultimately benefits all stakeholders. And when
mistakes have been made, someone has to pay for it. That includes governments
that have spent too much money. The Eurozone isn’t structured to offload
one’s debt problems to a central bank magic wand.
Please sign up for our free Merk Insights, explore our
chartbooks and follow me on
LinkedIn, Twitter, Facebook or Patreon.
Axel Merk
President & CIO, Merk Investments
|