Default: its not (necessarily) the end of the world

One of the things about sovereign debt defaults, whether they be hard or soft, is that there is a meme around them. A meme is a sociological equivalent of the gene. Coined by Richard Dawkins it is defined as “an idea style or behavior that spreads in a population”

One popular meme around sovereign debt is that a default is an enormously damaging event.  We are told that countries that defaults will be cut off from international capital flows, will be locked out of the market, that it would lead to vast falls in national income, we are all but told that the dead shall rise and that  a plague of locusts will swarm o’er the land.   As the song says however,  it ain’t necessarily so. When we hear dire warnings from bond traders and international financial markets, its well to remember the musings of James Carvill, the Democratic strategist.  “I used to think if there was reincarnation, I wanted to come back as the president or the pope or a .400 baseball hitter. But now I want to come back as the bond market. You can intimidate everybody.”

Default should, like any other economic tool, never be ruled out ex ante. Like any other tool its a banace between costs and benefits. The meme is that the costs are, by definition, so large, as to make the concept itself ludicrious. We have a dearth of evidence based thinking in ireland. But, in ay case,  let’s have a look at some of the recent evidence. In looking at this evidence we need to be conscious of two things. First the majority of this research pertains to emerging markets. Right now neither Ireland not Greece is an emerging markets…   it is not clear exactly what we can be classified as, perhaps a demerging market,  but the researches are typically around countries which would generally have a lower level of development and national income than either Ireland or Greece. Secondly the majority of the instances that researchers examined have been for countries which have had a greater degree of freedom over their exchange rates than the case is with either Ireland or Greece.  Thus the findings below cannot necessarily be taken as being likely to map to the Irish, or Greek, case should a sovereign default emerge. Nonetheless they do serve to give some counter to these mimetic  arguments .

A 2004 IMF working paper  took a look at  instances of default, restructuring, and assorted changes to sovereign bond payoffs, for 60 countries over 20 years, and finds “the probability of market access is not strongly influenced by default in the previous year“, and they say this is robust.  In other words, default once done does not appear to automatically lock a country out of the markets even in the next year. This is consistent with the argument that the markets look forward, not back, and that they look back only in so far as it gives a guide to likely future performance. They do not punish per se.

Gelos, Gaston, Sahay, Ratna and Sandleris, Guido, Sovereign Borrowing by Developing Countries: What Determines Market Access? (November 2004). IMF Working Paper No. 04/221 

 A 2006 Bank of England paper suggest that there is some increase in the cost of borrowing after a default, but also suggest that the total losses in terms of reductions in national income and the total time it takes to get out of the crisis are lower for countries that actually do restructure. This suggests that countries that reschedule their debts and start afresh with creditors face  lower cost of finance and quicker renewed access to this finance.  Like so often, Shakespear (Macbeth, Act 1 Scene 7) got it right “If it were done when ’tis done, then ’twere well it were done quickly”

De Paoli, Bianca Glenn Hoggarth and Victoria Saporta, Financial stability Paper Number One, July 2006, Costs of Sovereign Default

A 2008 IMF working paper subsequently published looked at 200 years of sovereign defaults, and suggests  that there are four kinds of cost. Reputational costs, trade exclusion costs, costs to the domestic economy in terms of reduced national income, and political costs. The find these reputational costs, basically being increase in interest rates/changes in the ratings of the country are “significant but short lived”, that the other costs are also significant but shorter lived, but that the political costs are “dire for incumbent government and finance ministries” they further state “the most robust and striking finding is that the effect of default is short lived as we almost never can detect defects beyond one or two years” .  They do find some effect on bilateral trade, but again this is concentrated in the years around the default. This is consistent with the arguments by Andrew Rose.

Borensztein, Eduardo and Panizza, Ugo, The Costs of Sovereign Default. IMF Staff Papers, Vol. 56, No. 4, pp. 683-741, 2009. 

Rose, Andrew (2005), “One Reason Countries Pay their Debts: Renegotiation and International Trade”, Journal of Development Economics, 77:189-206.

 A paper published in 2010 in  Journal of development economics by researchers from the Central Bank of Chile finds  “we do not find evidence that countries. shut their doors to  defaulters investments abroad” . In other words, a country that defaults does not find that it encounters difficulties in itself making investments abroad. There is no evidence that investment by residents of countries which default find these investments barred or seized by those defaulted upon. They do find some effect on foreign direct investment, with declines (but concentrated on the year immediately around the default) in FDI by the countries defaulted upon.

Miguel Fuentes, Diego Saravia, Sovereign defaulters: Do international capital markets punish them?, Journal of Development Economics, Volume 91, Issue 2, March 2010, Pages 336-347,

A 2010 IMF authored paper by the same team as the  2004 paper and in effect updating and refining it finds “the probability of market access is not influenced by countries frequency of default and that defaulters resolved quickly does not reduce significantly the probability of tapping the markets” they  do find that there is a period of lockout, but this is on average less than two years. This finding of relatively short lockout is also found in a paper by other researchers  who note that 50% of defaulters regain at least partial access within a year. Indeed, the time to regain access seems to be dropping, with some finding suggesting that access can take place within a few months and others suggesting that it takes longer.

Gelos, Gaston, Sahay, Ratna and Sandleris, Guido  Sovereign borrowing by developing countries: what determines market access  Journal of International economics, 83, page 243-254 

Dias, Daniel  and Richmond, Christine Regaining Market Access: What Determines Duration of Exclusion?”  (Working paper, UCLA) 

Papers such as the Borensztein  paper noted above typically find that there are declines in output in the year of the default episodes. More recently again a 2011 Journal of development economics paper by researchers from UNCTAD finds that far from defaults leading to declines in output defaults actually seem to be associated with output rising.  They use quarterly data, not the annual data of other researchers. The 2010 “the negative effects of the default  on output are likely to be driven by the  anticipation of default, independent of whether or not the country ultimately decides to validate it”. In other words if you’re been punished anyway you may as well go ahead and do it. A similar finding for unemployment is also evident in their paper.

Levy-Yeyati, Eduardo and Ugo Panizza  the elusive costs of sovereign defaults, Journal of Development Economics, volume 94, page 95-105


So why then do we find this meme? As ever, Rabo, the “straight talking bank” may provide an answer. In a recent economic note they state ” More specifically, the economic costs of sovereign default, as estimated by scholars, are found to be less drastic than most believe possible. The political costs of default, on the other hand, are non-negligible. The expected time of remaining in office is sharply reduced after a government throws in the towel.” This finding is from the Borensztein paper noted above. Perhaps therefore when we discuss default we should remember that we are dealing with political economy, and the two elements of that are equally as important.


One thought on “Default: its not (necessarily) the end of the world

  1. Pingback: 'I wanna return as a bondmarket - then I can intimidate everyone': Debunking Default

Comments are closed.