Why the IMF Supports the Latvian Currency Peg


My co-RGE blogger Edward Hughsome days ago wrote a compelling piece on why the IMF’s decision to agree a Latvian bailout program without devaluation was a mistake.

ByChristoph Rosenberg

He doesn’t mince words: Sticking to the present currency peg, Edward maintains,  constitutes “virtually the unjustifiable” according to “the implicit consensus among thinking economists.”

I happen to be the IMF’s mission chief who over the six weeks before Christmas led the discussions on the Latvian program. (And I also consider myself a thinking economist.) Let me therefore explain what considerations went into our decision to support for the Latvian authorities’ strong preference for a program built around the present currency peg. Along the way, I will also try to dispel Paul Krugman’s and Edward Harrisons’ claim that Latvia is the new Argentina..

Before I get started, I would agree with much of Hugh’s analysis of what has gone wrong in Latvia. The post-EU accession boom, fueled by rapid credit and wage growth, brought Chinese-style growth rates but led to an overheated economy (a point made by the IMF as early as 2005). The bubble burst in mid-2007, when foreign banks slowed their lending, and output has been decelerating ever since. The global financial crisis, by triggering the liquidity problems at Parex (the country’s second largest bank) dealt the final blow. As Edward rightly points out, the underlying problems go deeper, notably a loss of competitiveness (wage growth exceeded productivity and most resources went into non-tradables) and large stock vulnerabilities (the private sector built up high external debt, much of it short term).

The IMF-supported program addresses both the short-term liquidity problem–by providing official financing at low interest rates–and the long-term issues associated with an overvalued real exchange rate and the private sector’s debt overhang. Where I disagree with Edward Hugh, Paul Krugman and others is that the latter can only be addressed by nominal devaluation. The alternative route to external balance, adjustment via factor prices, may be drawn-out and painful. But at this point in time it corresponds more to the circumstances in Latvia than meets the eye from blogsphere. Here are nine reasons why.

First and foremost, this is the Latvians’ program. So even if we at the IMF were to favor a devaluation (which we aren’t), a program without national ownership would be doomed to fail. Edward believes that the Latvian’s very strong consensus in favor of the peg is misguided because it is rooted in misinformation. Based on my experience in Latvia, I don’t agree. The policymakers there are keenly aware of the choice they face. Why else would the Latvian parliament, with votes from the opposition, approve a supplementary budget that entails 7 ppt fiscal adjustment, including a 25 percent wage cut for all public servants? When he presented the program to parliament, Prime Minister Godmanis could not have  been more explicit about the alternatives and the sacrifices ahead. I myself spoke to the national tripartite commission (government, employer and employee associations) and I can only say that these people know what sticking to the currency peg will entail.

Secondly, a devaluation in Latvia would have severe regional contagion effects, especially given the fragile global funding environment. The spill-overs could well go beyond pressures on countries with fixed exchange rate in the Baltics and South-East Europe. For example, market confidence in foreign banks invested in the Baltics and similar countries would likely be affected, with implications for their ability to access wholesale financing

This is why, thirdly, Latvia’s preference for the peg is strongly supported by all foreign stakeholders, including the EU and its Nordic neighbors. They have put their money where their mouth is, providing three quarters of the total financial package of EUR 7.5 billion that backs the peg. In a remarkable show of solidarity, three new EU member states—the Czech Republic, Poland and Estonia—also contributed. The Swedish, Danish and Norwegian banks operating in Latvia are doing their part by publicly committing to support the liquidity and capital needs of their Baltic subsidiaries. Given the long-term bricks and mortar investments made in the region, it seems unlikely that they will cut their losses and pull out, as Japanese banks did during the Asian crisis.

Fourth, a devaluation would not significantly reduce Latvia’s external financing needs. While it would shrink the current account deficit further, private sector roll-over rates might not improve because the higher external debt to GDP ratio would likely result in credit agency downgrades to junk status and trigger the immediate repayment of most syndicated loans. Once unhinged, the peg may come under speculative pressures again and even larger external financing may be needed to credibly defend it at its new level.

Fifth, and now I turn to the internal adjustment discussed by several bloggers, there are advantages to a U-shaped  adjustment via factor price compression over the V-shaped recovery that is often  associated with a devaluation. (We estimated that a 15 percent devaluation, through the associated balance sheet effects, would have suppressed growth in 2009 by another 2 ppt, but led to a faster recovery in 2010)  As Mary Stokes points out, devaluation—depending on its size—would lead to a wave of defaults in a concentrated period of time. Latvia’s banks (both domestic and foreign) and its legal system are at this point not prepared for such a shock. By drawing out the likely rise in bankruptcies and nonperforming loans, the authorities have bought time to improve the country’s insolvency  regime, strengthen banks’ capital base and allow private debt restructuring. The program contains a host of measures to address these deficiencies as a matter of urgency.

Sixth, as again nicely argued by Mary, it is questionable whether a devaluation would quickly boost exports, given the global environment and the structure of its exports. Moreover, Latvia is a very small open economy and many of its retailers operate across in all three Baltic states and set uniform prices (for all three countries) in euros. The pass-through would be high and and the effect on the real exchange rate small. Re-orienting the economy towards tradables will require structural reforms which are envisaged in the program.

Seventh, Latvia has a very flexible economy, especially a quite nimble labor market. Wage cuts up to 25 percent may seem large, but let’s not forget that this comes after the tripling of nominal wages during 2001-07 (doubling in real terms) and that most of it will be accomplished by cutting bonuses of up to five monthly salaries. Latvia once before went through such factor price adjustments while maintaining its peg, during the Russian crisis in 1998. While circumstances may differ now, this experience suggests that it could do so again, especially in an environment where reduced employment options in the EU make emigration less attractive than a few years ago.

Eight, the large-scale fiscal adjustment under the program is in line with experience of successful such episodes elsewhere and therefore provides some assurances that it will not undermine the currency peg. Despite the pro-cyclical withdrawal of stimulus, budget restraint is needed to support real devaluation, reduce financing needs and make room for the additional fiscal costs of bank restructuring. Roughly one third of the adjustment comes from revenue measures (increases in indirect rather than direct taxes to support wage deflation) and two thirds from cutting expenditures (wages and spending on goods and services). It was important to us at the IMF and the authorities that the 2009 budget fully protects two essential expenditure categories: cofinancing of EU-supported capital projects and social spending, which is set to increase as a share of GDP compared to 2008. The program also contains institutional reforms to put the expenditure reduction on a permanent footing.

Finally, Latvia has a clear exit strategy from its currency predicament:  euro adoption. The authorities are determined to meet the Maastricht criteria in 2012. As I argued in a blog last year (“Avoiding the Portuguese trap”), entering the euro zone will not do away with the hard lifting necessary to address the competitiveness problems and high external debt. But it would once and for all remove the threat of a devaluation and thus bring much-needed investor confidence.

Latvia is at a crossroad. It needs to engineer a fundamental reorientation of its economy at a time when the global economy is in crisis. I concede Edward’s “biggest condemnation” (as he puts it) that, at least in the short term, the program does not contain policies to stimulate the economy. But, as I argued above, a devaluation would in the present circumstances unlikely deliver such a stimulus and may in fact make matters even worse. With fiscal policy constrained, sticking to the peg and toughing it out seems to be the only option for now, as long as the time is used to implement structural reforms the improve the financial system and reorient the economy towards exports. This is not to deny that this course of entails risks, not least a deflationary spiral or a fading of the present political determination to achieve the adjustment by fiscal, structural and income policies. In my view, these are risks worth taking, for Latvia’s and the region’s sake.

Christoph Rosenberg