As
the representatives from the European Union, the IMF
and Greek government are trying to flesh out the deal
how Greece can use EU's and IMF's funding to remedy
its fiscal position, the main question hovering above
these negotiations is whether Greece can and should
follow Estonia's example in massively cutting public
spending. Estonia's public spending was in 2009 a
drastic 12% less than in 2008 and this kept public
deficit below 2% of GDP in 2009. Both the European
Commission and the IMF have hurried to praise Estonia's
efforts; in credit markets Estonia's credit default
swaps (bets on how likely it is that Estonia defaults
on its public debt) valuation has been halved since
the beginning of the year (that is, the likelihood
of default has halved). While it is odd that Estonia's
debt (one of the lowest in Europe, just over 6% of
GDP at the end of 2009; in fact, Estonia has government
reserves amounting to almost 9% of GDP) is being valued
at all as it is not publicly traded to begin with,
it is a sign of a general tendency to see in Estonia
an example par excellence of handling the crisis.
Indeed, the entire issue of public debt seems to have
taken the center stage in the ensuing financial crisis
on the both sides of the Atlantic, see for instance
IMF's warning.
It is true that Estonia's numbers (public debt and
expenditure, budget cuts) are sheer magic in the European
context. As is the fact that these numbers have been
accompanied by no street riots; moreover, the government
is now probably even more popular than before the
crisis. There's justifiable cause for envy and this
begs the question: should Greece, with 13% of public
deficit and 113% of public debt (both as percentage
of GDP), follow Estonia, bite the bullet and get down
to slashing budget and wages? Similarly to Greece,
Estonia has no exchange rate policy option as the
Kroon is pegged to the euro. (Of course, pegs can
always be un-pegged but there seems to be no political
will in Estonia or in Europe to do this and by now
it is also probably too late as devaluation would
now make all the efforts of 2009 futile.) In other
words, the question actually is whether the so-called
internal devaluation works and if so then precisely
how.
The euphoria around Estonia should die rather quickly
when one looks at the GDP performance in 2009. It
fell nearly 15%; Greece's GDP contracted just 2%.
The figure below shows real GDP growth rates in Estonia,
the Baltics taken together (for comparison) and Central
Europe (Czech Republic, Hungary, Slovak Republic and
Slovenia; simple averages; all data from Eurostat).
The GDP data show one intriguing figure that really
begs the question: how come Poland is still growing?
Well, not only is Poland growing, its exports haven't
really slowed down at all during the crisis, as is
evident from the next figure below.
A look at the real effective exchange rates clarifies
the story with a rather amazing picture: Poland's
floating currency seems to have done what it's supposed
to do in the crisis like that – devalue. This has
considerably softened the crisis and contrary to the
Baltic and other Central European economies, Poland
has very quickly regained its export competitiveness.
It is clear that at least partially this was deliberate
policy as is explained by the Polish Central Bank
president in a Financial Times piece published posthumously
last week. It remains to be seen whether Poland remains
on this course of sanity.
That devaluation works in situations like we are
currently experiencing shouldn't really come as that
much of a surprise. There's an interesting parallel
from the Great Depression and its aftermath, depicted
in the figure below (the figure comes from Barry Eichengreen
and Jeffrey Sachs, “Exchange Rates and Economic Recovery
in the 1930s ”, The Journal of Economic History, Vol.
45, No. 4 (Dec., 1985), pp. 925-946, p. 936).
So, what is the price paid by Estonia and the other
Baltic states? Again, one would expect that with the
heavily overvalued currencies, the main suffering
would take place in the labor market and this is indeed
what is happening in the Baltics. While the Baltic
media is fanfaring that the unemployment figures didn't
not raise last week for the first week since the crisis
started, it seems clear that under conditions of weak
domestic demand (engendered through anemic borrowing
and uncertainty amid wage cuts) and lagging productivity
growth (especially if compared to Germany and Scandinavia
– Baltic and Central European export industries are
largely part of German or Scandinavian production
networks) there's no quick end to the crisis in sight.
While the respective national statistical offices
report some price deflation and falling real wages
in the Baltic economies, the former are merely cosmetic
and the latter in fact contribute to collapsing domestic
demand. It has led to positive external balance over
the last quarters, but this is not due to exports,
but rather because of the breakdown of the capacity
to import. Without actual devaluation, in other words,
such internal devaluation is bound to last for years
in the Baltics and to a lesser extent in the Central
European economies generally (other than in Poland,
of course). Growth is bound to stay anemic and, because
of lagging productivity, exports simply cannot pick
up fast and strongly enough to offset the weak domestic
demand. Polish companies at the same time could take
advantage of their relatively higher export competiveness.
However, without European fiscal transfers, most Baltic
and Central European economies would have similar
levels in public deficits as Greece, and, accordingly,
raising level of debts. See figure below. (Fiscal
transfers from the European Union are annual transfers
through the so-called structural funds. Here, the
EU fiscal transfers include funds from three main
sources: Cohesion, Rural Development and Fisheries
Fund; calculations by the author.) Continued high
unemployment will lead, especially in the Baltics,
to increasing public debt – or to even more unemployment
until the situation becomes untenable for some who
then emigrate.
Thus, one can draw the following conclusions: first,
in essence, Estonia and the Baltic economies are simply
mirror images to Greece and other PIIGS. In the former,
the private sector carries the cost of the crisis,
in the latter mostly the public sector (social safety
nets); because of the in-built inflexibilities, both
are cases of free riding: the former export public
deficits, the latter unemployment to the rest of Europe.
Second, Central European economies, except Poland,
seem to follow Germany (weak domestic demand, high
levels of exports) through high level of integration
into latter's production networks and tailwind Germany's
export machine. This assumes recovery in Europe and
the world. The danger is that these economies should
match Germany's productivity growth or else their
effective exchange rates start to appreciate again.
This is precisely the trap the PIIGS fell into. And
third, Poland, with floating currency and relatively
large domestic market, seems to be faring the best
so far among Eastern European economies. Such divergence
within Central and Eastern Europe will probably become
only more pronounced in the coming years.
It is important to understand that these woes have
a common cause: unifying rather unequal countries
into an economic union. That this is bound create
problems for PIIGS was brilliantly predicted by Jan
Kregel already in 1999. Simply put, the eurozone assumed
and still tacitly assumes either growing German wages
or growing productivity in the rest of Europe. Neither
has been the case.
The Baltic economies with pegs, and with insisting
on keeping the pegs, have simply tied the noose around
their own necks, and trading monetary stability for,
first, high financial fragility, and second, now for
very probably long-term high unemployment and debt
deflation in the private sector. This will probably
result in waves of emigration, growing social problems
and the like. In other words, the costs have been
shifted to the future and they are more than likely
to equal Greece troubles in fiscal terms. Estonia
is Greece in disguise. It remains to be hoped that
the EU and the IMF recognize that and refrain from
simplistic fiscal retrenchment that make problems
only worse as Greek domestic demand and, accordingly,
government fiscal position will only weaken further.
This results, as we have seen in Estonia, in real
economic depression, which is GDP contraction in double
digits.
* This is a write-up of the author's
presentation at last week's Hyman Minsky conference
at the Ford Foundation in New York. For a look at
the US debate on financial regulation, see Paul Krugman's
talk from the same event.
The article was originally published in the faculty
blog of the Department of Public Administration, Tallinn
University of Technology, Estonia.
April
21 , 2010.
|