Iceland President Olafur Ragnar Grimson on Economic Recovery

I've known you for many years already, Aaron, and I think you're responses to me in this thread have been quite poor by your usual high standards.

It's reasonable, I think, to demand a link for something that would seem unlikely. But Krugman, as you could already see, said explicitly that Latvia was going the way of Argentina because they were operating the exact same policies as Estonia, i.e. refusing to devalue their currency. It follows logic that it was likely that he might also believe the same fate awaited Estonia and any other country following those policies, because he wasn't attacking Latvia for being Latvian, but for operating policies he didn't agree with. In fact it would be more reasonable to demand a link if someone said he didn't think the other Baltic countries would go the same way as Latvia, because that would create a bizarre exception in his argument.

But that's the thing; without being able to read that FT link (which requires registration), the only things I can find that are relevant to the situation are:

1 - Krugman said that Estonia's recovery was being invalidly rhapsodized (I love that word) by austerity hawks, and that it was rather underwhelming, and
2 - Krugman said that Latvia's recovery was being invalidly rhapsodized, that their recovery is far from complete, that implicitly (this is particularly pointed out in the first linked article) their modest partial recovery does not qualify as a refutation of Keynesian economic thought.
3 - Krugman agrees with a guy named Edward Hugh that both devaluation and wage cuts are damaging in terms of being able to pay back debts.

There's other stuff but it's generally in the same vein. Krugman is consistently not making predictions, consistently not saying that Latvia or Estonia are not going to have a recovery / not having a recovery. He's saying that their recoveries are not particularly miraculous, that the fact that they're having a partial recovery does not qualify as a refutation of Keynesian thought, and criticizing the arguments made in the various cases.

The money quote that I think most of this centers around is a prediction, but it's far from a sweeping one:

is pretty much representative of the kinds of things he's written that I can see on the topic. I don't know enough macro to dispute or agree with his position; in fact, I know barely enough macro to understand his position. But it appears to me that his position is being consistently mischaracterized by people - bloggers, journalists, people on the internet, and professional economics writers alike. And I fucking hate that. If you want to argue with his position, feel free. I probably won't be able to meaningfully contribute to that discussion, but I'm sure at least one of the jeffs will be.

But calling, as I've heard people do (not you, banquo, but people in real life) Krugman's position a prediction that Latvia will follow the exact same trend as Argentina is absurd. It's a nuanced prediction which makes a very specific prediction, and testing that prediction would first require a systematic analysis of what "other things equal" even means, followed by at least a moderate effort to control for that, and I have no fucking clue how one goes about that but people haven't been.


edit to add: And it's not like Krugman is pretending that devaluation is a magic wand. He's saying that devaluation should, things being equal, be no more harmful than wage cuts in terms of servicing debts. He makes no claim in that post that it should be more or less harmful in other ways; he's criticizing a very specific defense of of the Latvian policies.

The cost of borrowing (private or public? I assume you mean public) is a financial signal whose meaning differs significantly based on context.

Here's a few:
1. The nature of the goods mix in economy under discussion.
2. Whether the currency is floating, pegged, or independent.
3. How open the economy is to international investment.
4. The level of imports and exports.
5. The perceived stability of the government.
6. The perceived likelihood of government default.
7. The predicted growth rate for the economy
8. Demographic trends.
9. Central bank activity.
10. Business cycle context.

The simplest counter-factual to "borrowing cost = financial viability" is to note that the real interest rate on US government debt has dropped down 3 points since the start of the financial crisis. It's down It's at its lowest level since the 1930s.

Does the US government seem healthier to you than any point since the 1930s?

An aside; is banquo (you know who) from qt3 finally escaping? I can't think of anyone else in this milieu who cares about Estonia so much :)

I think as a general point what is true for small countries is not easily generalisable to larger economies, and definitely not generalisable across an entire economic area. Small countries can survive (at least for a time) in all kinds of economic niches that broader, diversified economies are unable to. Which is why it is so frustrating when modern Estonia, or 80's Germany is held up as a reason to push for global Austerity today.

'Market Confidence' as a respectable idea also winds me up, as it puts me in mind of the same fucktard wisdom that helped cause the financial crisis in the first place and the herd of lemmings style movements of the stockmarkets.

Well shit, that would be one of the few people to whom I do extend enough social capital to do things like register on random websites. This is why you should keep your online identities consistent, or at least linked! >.<

How can people not have noticed that it's Tim!? Estonian wife, relatively recent father and lives in Frankfurt? How many people could it be?

In addition, and quite importantly, he also claims that internal devaluation is method where the burden of readjustment is put to a greater degree on earners of income rather than holders of capital, and that this has real and detrimental effects on human welfare.

That seems right (as in, now that you mention it it seems correct that he made that claim), but I am so not capable of evaluating it.

I'm no economist, but the argument makes at least some sense. An internal devaluation means that your stacks of cash will be worth just as much afterwards as they are today, but your income will decrease. With a currency devaluation, whatever decrease in real value happens to a person's income will also happen to whatever wealth exists in that currency. Of course, I can imagine that the time leading up to a currency devaluation or a country abandoning a currency peg will be a whole lot of people trying to transform their assets to a foreign currency creating its own problems.

Krugman said, and this is a clarification of his "Latvia" blog post of 2008:

"the Baltic nations, in particular, seem well positioned to follow in Argentina’s footsteps"

Well that's not a prediction per se, but it is a belief that they are "well positioned" to go the way of Argentina, i.e. bank runs, massive defaults, pariah status. But their futures didn't even remotely pan out that way.

Put it this way, if an economist said the stock market was "well positioned" to crash in 2013 and it instead rose significantly, I think most people would agree that economist was "wrong".

Financial viability is relative to other countries, not to other time periods. Nobody can lend money to the US of 1930 :).

And by financial viability I mean the likelihood that a country will repay its debts, and the US is a very safe bet, relatively, even if its economy is faltering.

Why thanks!

Having my real name associated with all the insults thrown at me over the years makes me worry about my google profile. Imagine my son searching for me in a few years time and google auto-completing "asshole" or "low social capital" after my name ;).

This is good news. It really does matter if someone has a track record you can reference on subjects that rely greatly on personal credibility.

Could you be more specific?

Briefly: it doesn't really address Krugman's point, it takes for granted that austerity "worked," and it then goes on to make some apples to oranges comparisons. Basically it reads like its written by a Very Serious Person who jus knows austerity is the way to go.

1. You completely changed the definition of your terms in your response.
2. If "financial viability" for a single country can't be compared across time periods, how can you evaluate whether Estonia's situation got better or worse during the financial crisis? What is the term even useful for, if it's just a proxy for international government debt repayment risk at a specific time? If you stay the same and a meteor wipes out Germany your "viability" gets better.
3. Estonia government debt has apparently been tiny for years and years, and Estonia at near-zero risk of not paying it as a result. How is this relevant again?
4. How does a number proxy for "likelihood government will default on its debt" result in changes to the regular economy?

Looking back through the thread:

Nope. Estonian capital flows are nothing of note. The bleeding of fleeing capital at the peak of the crisis has stopped; in 2010 things got bad. It's not remotely true to say "capital flooding back into the countries", however; it's stopped fleeing, but not much is coming in either. Not sure what your liquidity term there means.

I don't think I did. Maybe you misunderstood what I meant originally. I probably wasn't very clear.

But you weren't trying to show how a country has improved it's economy, you were trying to show how financial viability, or the potential for a country to repay its debts, isn't related to its cost of borrowing. Of course if you were thinking I meant something different (see above) we might be on completely cross paths here (and wasting each other's time).

Near zero? Not quite. Fitch's credit rating for Estonia fell to BBB+ in 2009. I imagine if Estonia went spending crazy it would have gone even lower. But as they kept their finances in check it's now risen to A, and then A+ after joining the Euro.

Because the cost of borrowing for a government is often reflected in private interest rates within the country, and also results in higher taxes to maintain payments on the rising interest on loans, itself caused by the risk of default.

2. Ability to repay public sectors debts is one component of many that contributes to government debt interest rates. That's because they're almost always the safest financial asset in the country and used as a proxy investment variable for inflation, in addition to the rate the government pays on its debts. So saying "government debt bond rates = default risk" is very misleading the vast bulk of the time, as that's a tiny contributor even when the economy is bad. Defaults are very rare even in financial crises.

3. The financial crisis and US downgrading fairly clearly showed that the credit rating's evaluation systems for government bonds are not useful information. Even if it was, the risk of Estonia defaulting was very close to zero before, during, and after the financial crisis, just because they don't have jack shit for debt. Their rate on government bonds was overwhelmingly determined by all the other contributing factors.

There was a risk of inflation or devaluation reducing the net present value of the earnings stream, but that's not "default". You can see this by looking at their time series for ratings - the local currency rating was A+ -> A- -> A+. The foreign currency rating dropped a bit more due to the currency risk, down to BBB. Still, neither was remotely a default risk.

4. These are not exogenous variables.

In summary:

1. Contrary to the bullshit you read in the papers, government bond rates are very, very rarely about risk of government default.
2. Contrary to the bullshit you read in the papers, austerity = internal devaluation = forced economy-wide wage cuts to "reduce government debt and thereby fix the economy" is completely pointless; no one ever forces through economy-wide wage cuts to reduce government debt when times are good.

To grossly simplify, it's really a variant on the Keynesism framework of economic management. During a boom, unit labor costs go up to match the investment level. When a recession hits, investment collapses, but sticky prices means unit labor costs don't go down. Therefore, the labor market doesn't clear, unemployment stays high, economic growth drops, and so on.

Forcing economy-wide wage cuts - partially by government spending cuts and government wage cuts - is an alternative solution to the sticky prices problem - wages go down, the labor market clears, the economy recovers.

It's only worked a few times when it's been tried; 1990s Barbados is the other example I've heard. My immediate guess is that 1) have to actually be able to cut internal wages when you try; it's damn hard, 2) you need to be able to export your way out, and 3) a low debt load, so the increased debt ratio doesn't kill you.

There's also the political economy angle; imposing economy-wide wage cuts is likely to be really unpopular; generally you'd not expect a democracy to pull that sort of thing off.

So why has Greece, Italy and Spain all had to pay much higher interest rates than Germany, if interest rates aren't influenced by potential default?

"one component of many" and "very rare even in financial crises" does not mean "not a factor" or "never happens".

Yeah, those European governments are at a very real risk of default, or at the very least leaving the Euro and devaluing like crazy to escape. Greece basically already has, and rest of the problematic Euro countries are really in a shit situation - high debt levels, high interest rates, apparently no help coming from the ECB.

So you say they almost never happen, but they are a very real risk? Enough that they can make the rate for Germany so low they are almost paying Germany to take their money, and so high for Greece that it goes off the scale (38% for a two year bond is one figure I read). But yet you don't think it's a significant factor. I'm a bit confused about what you are saying.

Back in 2008, the Baltic countries were seen as one of the "problematic" countries. That's why their credit rating got cut, and why respected economists were saying how they would "go the way of Argentina". If they didn't deal with that global fear of their imminent collapse, that would have led to high interest rates, like Italy, Hungary, Spain and Greece have suffered. Instead they took another route and regained the market's confidence.

Just repeating things endlessly doesn't make it so.

edit: also, Scott Sumner has some very good advice when trying to understand economics: never reason from a price change. Interest rates are a price, most of your reasoning is based on them changing. E.g., Estonia did austerity and interest rates went down; Estonia has different interest rates than Greece, Spain, etc. This is bad economic reasoning, and it's why I seem so very skeptical!

You should go review Jason's posts again, specifically this graph:

Something I can probably clear up here - "go the way of Argentina" for Krugman meant:

1) Don't abandon the currency peg,
2) Suffer years of terrible economic problems and high unemployment.
3) Currency peg comes under increasing levels of self-fulfilling speculative attack.
4) Estonia runs out of reserves to defend the peg.
5) Followed by finally abandoning the currency peg.

He's an international trade specialist which is why he's so focused on stuff like this.

By contrast, I don't think anyone was predicting debt issues for Estonia, because Estonian debt was incredibly low.



Estonia spends about 40% of GDP through the government. They'd have to collect no taxes at all for a year to even get to Sweden's level of public debt.

To take a stab at it:

Estonia was under basically no risk of defaulting, ever. This is in line with the vast majority of cases.

Greece is an outlier because of how thoroughly fucked they are, and they were at risk of default (in fact, they have already effectively defaulted), and for them it's a significant problem and significant factor.

But making a comparison between Estonia and Greece is a non-starter. Their circumstances are so wildly different that you can't meaningfully draw conclusions from a comparison.

fwiw I'm of the opinion that basically all economic comparisons to Greece are inoperative; Greece is just too much of a basketcase to derive any generalizations.

Assuming that your version of Krugman's claim is true, he was still wrong! :)

But by your own definition, Estonia would get into debt problems that would risk it quickly not being able to repay these debts, or at least have great difficulty. If (your definition of) Krugman's prophecy were to come true, they would have to run up a huge debt to defend the Euro peg. They couldn't spend reserves, because they didn't have any. Their balance was just a little in the black before 2008 - certainly not enough to fight off a Soros level speculative attack.

And bonds can be for as long as ten years - plenty of time for a small country to get into big trouble.

"Prophecy" is an interestingly loaded word to use, and more than slightly inappropriate.

I wasn't comparing Greece with Estonia, but Greece with Germany. The reason being that they have the same currency, so they both have their bonds denominated in the same currency and are so both have the exact same devaluation risk. The argument Jason put forward was that the risk of default had almost no bearing on the cost of borrowing, and yet that seems to be exactly the difference between Greece and Germany's interest rates.

Let me clear this up for you:

Jason said:

"So saying "government debt bond rates = default risk" is very misleading the vast bulk of the time, as that's a tiny contributor even when the economy is bad. Defaults are very rare even in financial crises."

Then when I asked:

"So why has Greece, Italy and Spain all had to pay much higher interest rates than Germany, if interest rates aren't influenced by potential default?"

(note: zero reference to Estonia in that question)

And Jason answered:

"Yeah, those European governments are at a very real risk of default"

Which seems to be a complete contradiction to me.

My original argument was that Estonia's low cost of borrowing is highly influenced by the country's financial viability, i.e. it's ability to repay (now and 10 years into the future). Jason said that was not true, and despite Krugman saying that Estonia risked getting into financial trouble by going the route they did, he thinks they had absolutely no chance of getting into financial trouble. I think that's easy to say now, in 2013, when Estonia has navigated its way safely through very troubled waters, but back in 2008 I think people were very worried about many countries, especially small, former Eastern European ones, getting into trouble. And when people worry, they demand more interest to allay those fears.

Actually banquo this was your original argument, near as I can tell: "What austerity has achieved is a healthy bank balance. This, it has been argued, has led to the financial markets feeling confident in the Baltic countries. This has led to capital and liquidity flooding back into the countries. This has led to business regaining its lifeblood. This has led to the countries experiencing record growth."

Your phrasing is a bit wishy washy ("it has been argued,") but whatevs. Since that post I'm not really sure what point it is you're trying to make. Obviously risk is one input into whatever function determines the interest rates on sovereign debt, but (as Jason's pointed out) Estonia's debt load has never been particularly high so there's not a lot of reason to believe that it's ever been a significant factor. Linking this back to the broader issue of austerity vis a vis Estonian economic performance the point remains that a) Estonia's performance since 2008 isn't all that impressive and b) you haven't even come close to making any kind of evidence-based case for austerity being a major driver (much less the prime driver) of what recovery Estonia has enjoyed. Insofar as what you're trying to argue represents the above quote, Jason's already shown that the idea that capital flooded back into the country is hogwash.

Again: "the vast bulk of the time" does not mean "never", and "very rare" does not mean "nonexistent".

Hehe if we're going for pedantry, it's worth noting that none of the PIIGS have defaulted (with the arguable exception of Greece, and that depends on how you define "is").

Again, there is no plausible scenario where Estonia has debt repayment troubles, because they didn't have any debt, and there was no plausible scenario where they'd acquire enough in the crisis to do so.

That's not how speculative attacks work; defending a currency peg is not done by a government acquiring debt. Hence my insistence on this point.

This IMF paper gives some details. Basically you can use foreign currency reserve selloffs, raising interest rates, and imposing capital controls to try to defend a peg.

Sometimes in the process of defending a peg if the country has a large amount of foreign-denominated debt the increased debt servicing costs from the interest rate rises can damage the government's debt servicing ability, like Mexico 1994, but it's not mandatory. It certainly doesn't apply to Estonia, as even if their debt interest rates shot through the roof they could still easily service the debt.

Krugman's assumption was that Estonia's currency peg would come under speculative attack due to the fundamentally unsustainable exchange rate, and they'd forced by speculators into the standard currency peg death spiral. For a variety of reasons that didn't happen.

Greece has defaulted in effect, if not in name. I'm a functionalist before I'm a pedant. (And yes, I know that's a misuse of the term functionalist.)

But it's not an idle point. Saying that something is "very rare even in financial crises" does not mean that it never happens. It means that it is the exception, rather than the rule. Pointing out a time when it happened does not qualify as a counterargument; that requires an analysis of financial crises and rates of incidence.

However, I disagree with Jason on a certain point. Default risk is not ever what drives bond rates. What drives bond rates is perceived default risk, which is very different; a negative-speculation "bubble" even in the absence of real evidence can have significant effects.

You're right, actual risk is kind of a meaningless concept. There's just perception, often self-fulfilling.

The Malaysian component of the 1997 financial crisis is a good example of one where debt wasn't a significant deal. They came under attack when all the other pegged regional curriences did; they fought back with capital controls and managed to preserve the peg. Their debt to gdp ratio was virtually unchanged, as was their nominal debt level.

Personally I agree with Krugman that capital controls to stop "rationally stupid" investors from panic selling is not used anywhere near enough.

I'm not, and never will be, in full cognizance of every variable regarding Estonia's economy, so all you have done is proven that capital didn't "flood" back into the country as I thought. That doesn't disprove the point, however, as everything is relative. Estonia's capital flow situation has improved significantly, it's gone from very negative to positive. What they have done is created a stable economy which is now attractive again to investors. If you could show that it somehow got worse, or at least remained bad, then you'd have a case that their policies failed, but that's not true.

When Soros attacked Sterling's EMF mechanism, the UK government lost almost 2% of UK GDP in a single day trying to defend it. Argentina, that country the Baltics were supposed to follow, increased its debt (as a percentage of GDP) from a reasonable 53% to an incredible 165% in a single year. It's perfectly possible for any country to become bankrupt pretty quickly, excepting resource rich countries like Norway and Saudi Arabia.

As I said above, the UK lost 3.5 billion in a single day protecting the pound against speculative attacks in 1997. Did that 3.5 billion somehow not get added to the UK national debt at the end of the year just because it resulted from a currency reserve sell-off?

So he was wrong, then? ;)

So what were the reasons? I'm open to arguments.

I would have thought perceived default risk was default risk, because who decides what the risk is? There is no absolute measure, so surely it's always perceived, i.e. subjective. That's why I said above that Jason's view that Estonia has no risk of default is meaningless if other people believe differently, because that belief alone will drive up their cost of borrowing. And when people have confidence in your ability to repay based on your handling of the economy (maybe the subjectively "correct" way in the eyes of investors) maybe that alone drives the cost of borrowing down.

Trivial counterexample: If a country is running a surplus and servicing old debts, but it's being run by people from Ethnic Group X, a country (we'll call it Racist Country) with systemic distrust of Ethnic Group X might perceive a risk of default. That doesn't mean there is a risk of default. We can look back and say quite confidently that Ethnic Group X was not going to renege on their agreements to pay the debts despite the belief from Racist Country to the contrary; the perceived default risk was totally out of whack with the actual default risk.

Another trivial counterexample: The country of Bobtopia has debt that they're servicing. They're doing so based on a currently successful run of gambling. But as soon as the roulette wheel hits a black instead of a red (that's how roulette works, right? I dunno, bear with me here), they'll default. The perception is that Bobtopia's not in danger of default, but in fact, they're quite likely to do so in the near future.

Depends. There are a number of things you can bet on in roulette, with varying odds.