Why The IMF's Decision To Agree A Lavian Bailout Programme Without Devaluation Is A Mistake

The IMF finally announced it's Latvia "bailout" plan on Friday. The plan involves lending about €1.7 billion ($2.4 billion) to Latvia to stabilise the currency and financial support while the government implements its economic adjustment plan. The loan, which will be in the form of a 27-month stand-by arrangement, is still subject to final approval by the IMF's Executive Board but is likely to be discussed before the end of this year under the Fund's fast-track emergency financing procedures, and it is not anticipated that there will be any last minute hitches (although I do imagine some eyebrow raising over the decision to support the continuation of the Lat peg). The Latvian government admits that some of the IMF economists involved in the negotiations advocated a devaluation of the lat as a way of ammeliorating the intense economic pain involved in the now inevitable economic adjustment. But the government in Riga stuck to its guns (supported by the Nordic banks who evidently had a lot to lose in the event of devaluation), arguing that the peg was a major credibility issue, and the cornerstone of their plan to adopt the euro in 2012.

"It (the programme) is centered on the authorities' objective of maintaining the current exchange rate peg, recognizing that this calls for extraordinarily strong domestic policies, with the support of a broad political and social consensus," said IMF Managing Director Dominique Strauss-Kahn.

In return for the loan the IMF have agreed a "strong package of policy measures" with the Latvian government and these will involve sharp cuts in public sector salaries, and a tight control on Latvian fiscal policy. The IMF have insisted on a substantial tightening of fiscal policy: the government is aiming for a headline fiscal deficit of less that 5 percent of GDP in 2009 (compared with a anticipated deficit of 12 percent of GDP in the absence of new measures) - to be reduced to 3% in 2010 (thus the Latvian economy will face not only tight effective monetary policy in 2010 - via the peg - but also a less accommodating fiscal environment, frankly it is hard to see where the stimulus to economic activity is going to come from here) . Structural reforms and wage reductions will also be implemented, led by the public sector, and VAT will be increased, all with the longer term objective of further strengthening Latvian competitiveness and facilitating the external adjustment. The problem is really how the Latvian population are going to eke it out in the shorter term.

"These strong policies justify the exceptional level of access to Fund resources—equivalent to around 1,200 percent of Latvia's quota in the IMF—and deserve the support of the international community," Strauss-Kahn said.

The loan from the IMF will be supplemented by financing from the European Union, the World Bank and several Nordic countries. The EU will provide a loan of €3.1 billion ($4.3 billion), the World Bank €400 million ($557.6 million), and several bilateral creditors [including Denmark, Estonia, Norway, and Sweden] will contribute as well, for a total package of €7.5 billion ($10.5 billion).

The stabilization program forecasts that the economy will contract 5 percent next year, the Finance Ministry said in a statement yesterday. Revenue is expected to fall by 912 million lati ($1.7 billion) next year and spending will be reduced by 420 million lati.

Strangely the IMF statement was not very explicit the key topic - the currency peg - in the sense that it was a little short on argumentation as to why it considered - despite its well known waryness about such approaches, and having got its fingers very badly burnt in Argentian in 2000 - that it would be best to continue this arrangement in the Latvian case, despite the Fund's strong emphasis on the need to current the large external balances which exist (see Current Account deficit in the chart below).






All we really know about the background to this decision is contained in the statement the IMF posted on its website on December 7:

Mr. Christoph Rosenberg, International Monetary Fund (IMF) Mission Chief, issued the following statement today in Riga :

"Following the IMF's statement on Latvia on November 21, 2008, good progress has been made towards a possible Fund-supported program for the country.In cooperation with the European Commission, some individual European governments, and regional and other multilateral institutions, we are working with the authorities on the design of a program that maintains Latvia's current exchange rate parity and band. This will require agreement on exceptionally strong domestic adjustment policies and sizeable external financing, as well as broad political consensus in Latvia In this context we welcome the commitment made today by the Latvian authorities. All participants are working to bring these program discussions to a rapid conclusion."

So there seems to have been a trade-off here, between the IMF agreeing (reluctantly I think, but this is pure conjecture since there is little real evidence either way) to accept the peg, and the Latvian government agreeing to exceptionally strong adjustment policies. But the question is: was this agreement a good one, and will the bailout work as planned? I think not, and below I will present my argumentation. But before I do, I think it important to point out that the kind of internal deflation process the Latvian government has just accepted is normally very difficult to implement, which is why economists tend to favour the devaluation approach.

Just how large the competitiveness issue is in Latvia's case can be guaged by looking at one common measure of competitiveness, what is known as the country's real effective exchange rate. The REER (or Relative price and cost indicators) aim to assess a country's price or cost competitiveness relative to its principal competitors in international markets. Changes in cost and price competitiveness depend not only on exchange rate movements but also on cost and price trends. The specific REER prepared by Eurostat for its Sustainable Development Indicators is deflated by nominal unit labour costs (total economy) against a panel of 36 countries (= EU27 + 9 other industrial countries: Australia, Canada, United States, Japan, Norway, New Zealand, Mexico, Switzerland, and Turkey). Double export weights are used to calculate the REERs, reflecting not only competition in the home markets of the various competitors, but also competition in export markets elsewhere. A rise in the index means a loss of competitiveness, and as we can see, Latvia has suffered a huge loss of competitiveness since 2005. There is a lot of "correcting" to do here.



The problems of loss of external competitiveness Latvia faces are not new, nor are they unique. Russia may be a lot larger than Latvia, and Russia may also have oil, but Russia's internal industrial core has become uncompetitive, and there is really only one sensible way of attacking this problem, and that is through devaluation, as Standard & Poor's Director of European Sovereign Ratings argues in the extract I cite below. One of the unfortunate side effects of the fact that currency policy has become almost a matter of national strategic importance in Latvia has been that the necessary open-minded discussion of the pros and cons of the situation has not been possible.

Accompanied by generous government spending, the credit boom also fueled inflation, which weighed on the competitiveness of Russia's noncommodity sector. As wage growth averaged nearly 30 percent over the last two years and the ruble-denominated cost of production rose, domestic manufacturers found it very difficult to compete with cheap high-quality imports. As a consequence, entrepreneurs logically avoided manufacturing and, instead, invested in much more profitable and more import-intensive sectors, such as banking, retail and construction.

The resulting structural imbalances were well camouflaged by the extraordinary growth in energy and other commodity prices. For six straight years, the earnings from Russian oil and commodity exports on world markets have increased much faster than the cost of imports, offsetting the less flattering volume effects. From 2003 through this year, the cumulative difference between export and import price inflation in Russia was a fairly remarkable 74 percent. This put upward pressure on the ruble, encouraging borrowers to take loans in dollars or euros at negative real interest rates, under the assumption that the ruble would appreciate indefinitely. But it also provided an important source of financing.
Frank Gill, director of European sovereign ratings at Standard & Poor's in London, writing in the Moscow Times

So the Latvian competitiveness problem has become evident to everyone, and perhaps the best indication of the severity of the problem is the way that people almost laugh at the suggestion that Latvia must now live from exports (exports, what exports?, they say). However it is clear, and especially given the force of the agreed internal adjustment, that domestic demand is now dead as far forward as the eye can see as an effective driver of GDP growth, and, as can be seen in the chart below, exports are going to have a hard time of it, even after growth in other European countries picks up in 2010 (or whenever).

The competitiveness problem can be seen quite clearly in the above chart, as Latvian wage rises became detached from productivity improvements in the second half of 2005 and the rate of increase in exports shrank rapidly, while imports began to enter at a much faster rate. This process eventually itself in the first half of 2007, with import growth at first increasing rapidly, only to subsequently decline, giving in the process some positive increment to GDP from the net trade effect - as exports once more began to accelerate (creative destruction impact) even while imports fell through the floor. However as the external trade environment has darkened, even this expansion in exports has petered out, and inflation adjusted exports are currently hardly growing, and may even turn negative in the coming quarters. 2009 promises in any event to be a very hard year, but without a truly massive correction in relative prices there will be no recovery in 2010 either, and probably not in 2011. Remember, wages are now about to start falling, unemployment is about to start rising, and government expenditure is about to get pruned, so the only possible area for growth is external trade, and any inbound FDI that can be attracted to build productive capacity for exports. On top of which the correction in the current account deficit means that Latvians collectively - government, companies and households - are going to have to start saving, and a rise in net aggregate savings is basically tantamount to a brake on internal demand. So whichever way you look at it, exports are now the name of the game.


Why Keep The Peg?

Given all the problems that having the peg are likely to create, what then are the arguments for maintaining it? Well frankly, such arguments are hard to find at this point, in the sense that there are relatively few people, at least in the English language, who are willing to stick their neck out and try to justify what, in my humble opinion, is virtually the unjustifiable, and the implicit consensus among thinking economists would seem to be that this is a bad idea. The decision does, however, have its advocates, and Anders Aslund of the Peterson Institute has been bold enough to have a try, so, in the interests of balance and try and get some purchase on what the arguments might be, I am reproducing his argument in its entirety.

Why Latvia Should Not Devalue
by Anders Aslund December 9th, 2008

Latvia has a severe financial crisis, the preconditions for which have long been evident. A fixed exchange rate to the euro led to an excessive speculative influx of capital, boosting Latvia’s private foreign debt to 100 percent of GDP. Inflation soared to 16 percent, and the current account this year to 15 percent of GDP. Latvia’s budget has traditionally been almost in balance.

For most countries, devaluation would appear inevitable, and some argue that Latvia has to devalue its currency, the lat. But Latvia’s circumstances are peculiar, making the standard cure not only inappropriate but harmful. A severe wage and social expenditure freeze would be a better prescription, along the lines of a preliminary agreement on macroeconomic stabilization reached on December 8 among the Latvian government, the European Commission, the International Monetary Fund (IMF), and the Swedish government.

Now the questions are how much financing Latvia needs, who will give it, and on what conditions? The key outstanding issue has been whether Latvia should devalue or not. But given that Latvia—and Estonia—are experiencing high inflation with close to balanced budgets, devaluation is neither necessary nor desirable. A freeze of wages and social transfers would be preferable for both economic and political reasons.

First of all, thanks to Latvia’s limited GDP, $27 billion in 2007, sufficient international financing can be mobilized. The combination of IMF, EU, and Nordic funding should be sufficient.

Second, devaluation is likely to aggravate inflation and it could start a snowball effect of higher inflation and repeated devaluations. A devaluation would not be less than 20 percent and it would cause greater social and economic disruption.

Third, the great number of mortgages held in euros would force a massive blow-up of bad debt and mortgage defaults, which in turn would seriously harm the population, the housing sector, and the banking sector and thus the economy as a whole. Such a banking crisis is not necessary. One of the three big banks, Parex Bank, has already gone under, but the other two, the Swedish banks Swedbank and SEB, are strong enough to hold, if no devaluation occurs.

Fourth, Latvia’s main macroeconomic problem is inflation. Devaluation would initially aggravate inflation, while a wage and social expenditure freeze would sharply reduce inflation. High inflation has led to the excessive current account deficit. Latvia does not suffer from any structural terms of trade shock

Fifth, a freeze on wages and public expenditures would strengthen the budget, while devaluation is likely to lead to severe budget strains.

Sixth, the Latvian population seems politically committed to the fixed exchange rate, and it seems prepared to take a freeze of incomes and public expenditures, and if necessary even cuts. Therefore, devaluation could lead to undesirable and unwarranted political convulsions.

Finally, devaluation in Latvia would inevitably drag down Estonia as well, and all the effects would be doubled, while Estonia might hold its own without Latvian devaluation. Lithuania, which does not really have any serious financial problems, could also be harmed. I would have recommended that the Baltics abandon their fixed exchange rates a few years ago, but this is the wrong time to do so.

The argument I am making applies only to very small economies with basically sound economic policies. Russia and Ukraine are in a very different situation. Both suffer from major structural changes in terms of trade because of slumping commodity prices, and they should let their exchange rates float downward with their terms of trade.

The main arguments in favour of the peg would thus seem to be as follows:

1/ Latvia's situation is exceptional (is that also true of Bulgaria, Estonia and Lithuania?). It is hard to know what to make of this. Certainly the comparison with Ukraine and Russia does not seem appropriate, since these are ultimately competitor countries as far as manufacturing industry goes, and they are devaluing not because of their raw material exports (agriculture and energy) are too high, but because the price of the products from their manufacturing industries are too high due to all the earlier internal inflation, and the attempts to maintain the currency value via the controlled "corridor".

2/ A severe wage and social expenditure freeze would be a better prescription than devaluation. Well they would be a good prescription, but they simply are not possible, since simply freezing things where we are won't work, the imbalances are too large, so we are talking about sharp reductions in wages and public spending (as nominal GDP goes sharply down, then even a 5% fiscal deficit will mean spending has to contract - by 420 million lati according to the budget forecast - although the IMF has agreed to a policy of protecting social expenditure as much as possible).

3/ Then there is the forex mortgage situation. This I agree is a major problem, as devaluation implies default, and an oncost for Sacndinavian banks. But if we are sending the entire Latvian population through all this simply to attempt to avoid defaults on mortgages we are making a mistake, since obviously the sharp rise in unemployment we can expect and the sharp fall in wages can have a similar impact. I mean, one way or another the REER (see above) is going back to the 2005 level, so the mortgages will be just as unaffordable, and in my view the best solution to this would be for the Scandinavian (and Italian - Unicredit) banks to take a haircut, and receive compensation via their domestic bank bailout programmes. This would be a much more equitable sharing of the costs of the forex lending programme having gone wrong. To take another example, Spain is not devaluing from the euro, yet a hefty round of mortgage defaults (and builder bankruptcies) is now expected. So it is really a case of default through one door, or default through the other one. Which way would you like to go, sir?

4/. That devaluation would provoke inflation. Well this is just the point, devaluation would only provoke significant inflation IF Latvia still didn't have an independent monetary policy (to restrain domestic demand), but since part of the reason for devaluation is precisely to recover control over monetary policy again, this argument seems to me not to be completely valid, and it seems to be forgetting the other problem, deflation, which is much more likely to become Latvia's real problem over the next two or three years. Trying to run some form of Quantitative Easing (which is the new "in" term for how best to handle monetary policy in the midst of a liquidity trap, which may well be where Latvia and several other CEE economies are now headed) without independent monetary policy is quite frankly, completely impossible. If we look at the chart for the producer price index I reproduce below, we will see that the PPI (which is normally regarded as an indicator of coming inflation) is no longer climbing, and seems set to start to come down., and this could easily be an early warning signal for forthcoming deflation.

5/. The Latvian population seems politically committed to the fixed exchange rate, and appears prepared to take a freeze of incomes and public expenditures. This may well be true, and is an impression I get when I look at some of the comments on my blog. Many Latvians (and citizens of other Baltic states) have accepted the peg as some indication of "post-independence" indication of national "seriousness", and that any stepping-back from it would be seen as some kind of defeat. I understand this view, but I think it is a mistake, since sometimes it is better to accept defeat in order to live to fight again another day. I think Latvian politicians are to some extent reacting to this kind of pressure, to some extent thinking about their own invested social capital, and to some extent under pressure from Nordic banks. In any event all three of these seem to have more influence than the rational arguments about the advisability of the peg. There is no doubt in my mind that the coming recession will be longer and deeper if the peg is maintained. Indeed I am almost certain that the attempt to sustain it will fail (and that we will see some kind of rerun of Argentina 2000 - in all three Baltic countries and Bulgaria) and really the sooner the population become aware of this the better. Basically what we witnessed in Argentia in 2000 was basically a process of growing battle fatigue and war weariness, as the population were asked to make one sacrifice after another in support of a policy which couldn't work, and only lasted as long as it could. The end product is that when the peg finally breaks the local population will be severely disillusioned, and the politicians will totally lack credibility, which is a sure recipe for chaos, as we saw in Argentina in 2001.

Indeed, if anything the position is arguably worse in Latvia at the present time, since the optimum conditions for a free and open debate about the alternatives aren't exactly in place at the moment it seems very hard to know what the population at large would decide if they had complete access to all the arguments.


6/. Finally, devaluation in Latvia would inevitably drag down Estonia as well. This is undoubtedly a consideration in the mind of the IMF (and Lithuania, and Bulgaria) but really all of this will have to be faced by all four countries sooner or later, especially since the only way out of their recession will be, as I am saying, through exports, and most of the other competitor countries (look even what is happening to the Polish zloty and the Czech Koruna as I write) will see the partities of their respective currencies well down on the euro as we enter the recovery.

Where Is Growth Now Going To Come From?

Basically the key argument for devaluation is that it is easier to manage an economy with a low level of inflation (please note I am saying low, very low, certainly below 2%, ask Ben Bernanke or the Japanese is you don't believe me) than it is to manage an economy which is in deflation freefall. The big danger in Latvia is not only that there can be a real (ie price adjusted) contraction in the economy of 5% in 2009 (or more, the economy is down 4.9% year on year in Q3 2008, and things are certainly going to get worse), but that this contraction may be accompanied by price deflation (ie actually falling wages and prices) which means nominal (current price) GDP would decrease by the size of the real contraction plus the fall in prices. Thus we could see a very large drop in nominal GDP in 2009 and 2010. If realised this would be a very difficult situation to handle, and I doubt the people currently taking policy decisions in Latvia are fully aware of the implications (although the IMF economists should know better). In particular the deflationary debt dynamics would be very hard to control, and again, especially without independent monetary policy.

It is important to remember that these loans which have been agreed to are simply that, loans, to guaranteee the external financial stability of the country during the forthcoming correction, but they do not, in and of themselves solve any of the real economy problems. And they will need to be repaid if they are used, and will nominal Latvian GDP heading down, the cost of repaying them effectively goes up in terms of real Lat earnings. This is what debt deflation means.

The International Monetary Fund on Friday said it now expects a net income of
about $11 million in fiscal year 2009, and not a shortfall of $294 million as
previously forecast, as more countries turn to it for rescue loans in a
deepening financial crisis. "The improved income outlook reflects new lending
activity that is estimated to generate additional fund income of about $247
million, assuming all disbursements under the recently approved arrangements are
made as scheduled," the IMF said. Since early November, the IMF has approved
rescue packages for Hungary, Iceland, Ukraine and Latvia as the global crisis
spreads to more emerging economies.

I am citing the above Reuters report, not as a criticism of the IMF - they are simply doing their job as best they can, and under very difficult circumstances - but to remind people that the IMF is effectively a bank, and these are loans, and interest is paid, and there are no "freebees" here, and definitely no "free lunches" - not even in the newly established Latvian soup kitchens.

So we should ask ourselves where growth is going to come from - the growth that will now be needed to repay the capital and interest on these loans. Certainly not from household consumption if we look at the chart below, or from government consumption given the restraint on public spending. The private consumption position can only deteriorate as wages fall and unemployment rises.

Not from manufacturing industry in the short term (until prices correct, and the external recovery starts), and again look at the chart.

And finally don't expect an investment driven recovery (again see chart) until the demand for Latvian exports picks up, and it becomes attractive to start expansing capacity.


Basically I feel the biggest condemnation which can be made of the package which has been announced is that it doesn't seem to contain one single policy for stimulating the economy, and stimulation and a return to growth is what Latvia badly needs by now.

And the worst case scenario outcome of the way all this is being handled (and the issue that actually concerns me the most) is the possibility that young people decide to start migrating out of the country again, in order seek a new future and to start sending money home to help their families confront the difficult circumstances. Since Latvia's population is already declining this would be the cruelest cut of all, and one would have to then ask just what kind of future really awaits this unfortunate country?