What can the ECB do?

It seems that many economists agree that the ECB has to get more involved in dealing with the current crisis in the Euro area. Politicians (at least some of them) disagree. They cite fear of inflation, take moral stance on past recklessness, argue that sinners have to repent and repay, etc. By now, hundreds of articles have been written on the potential damages from the collapse of the euro area, so I am not going to repeat them here. I want to focus on what the ECB can do to stop the self-fulfilling collapse of several European economies.

The standard view on ECB is that they have to print money in order to buy bonds of governments in trouble like Greece, Italy, Portugal, etc. But there is another thing that they can do. About 18 months ago, the chief economist of Citigroup, Willem Buiter, wrote an article in which he noted that the ECB can use its future profits to stop the acceleration of these negative dynamics on the bonds market. According to his estimates, the ECB has a “non-inflationary loss absorption capacity of … at least €2.4trn and more likely over €3.4trn.” In other words, the ECB can put credibly on the market a firewall of over €2.4 trillion. Certainly this ought to stop the ever-increasing yields on Spanish and Italian debt from rising further.

How does this work? Central banks run very profitable operations – they issue pieces of paper that we call money on which they pay us no interest. With the printed money they buy interest-bearing securities like government bonds. Because of the interest rate differential between the 0% rate on currency and the yield on bonds, central banks generate profits every year. Well, the actual calculation of the profit (seigniorage) is a bit more complicated (see Buiter’s article), but the mechanics are not changed much. This profit is generated even if inflation is 2% (ECB’s mandate).

One way to imagine Buiter’s proposal is the following: Suppose that the ECB issues bonds worth €2.4 trillion (note that they do NOT print money equal to €2.4 trillion). These bonds would most likely have a yield close to the yield on German bonds (2.6% for 30-year bonds) because one can expect that if the euro is alive, the ECB will be mechanically generating this revenue from their activities. After issuing these bonds, the ECB can make an announcement that for countries which are solvent when the yield is 5%, they will not allow these yields to go above 5% (one can adjust the numbers for the state of the cycle, but the idea is that they restrict the yields from increasing due to the lack of confidence). If yields are above 5%, they buy the bonds with the revenue generated through their own issue. If this announcement happens, and if the ECB just starts with some symbolic purchases, yields will go down quite rapidly and yields on Italian debt for example will return to sustainable levels.

Notice that if the ECB has to buy the bonds, then they will be generating yet another profit stream from the difference between their 30-year bond (at 2.6%) and the Italian yield (currently at 6.5%). But even if Italy fails and does not repay the bonds bought by the ECB, the ECB can fully absorb the loss. This is why Buiter calls this quantity the “non-inflationary loss-absorbing capacity” of the ECB.

Is there any magic or a free lunch here? Not really. The seigniorage of central banks has to be transferred by law to the governments of the euro zone. So, effectively what the ECB does is that it takes future government revenues and puts them today to create a firewall. They can commit to having this revenue in the future because of the nature of their operations. Governments – even though they are the eventual recipients of this flow – cannot commit today because the markets do not trust them anymore. The essential role played by the ECB in this case is to put the certain revenue stream on the table and to tell the markets – this future revenues will be used to absorb losses or to repay debt and not for other spending. It is quite likely that they will never have to buy even one government bond out of this facility.

Will the ECB do it? I am somewhat skeptical. It does require a bit of resolve to go beyond the standard thinking.

Ilian Mihov

P.S. On November 16, 2011 Willem Buiter also gave an interview on Bloomberg TV where he talks again about this proposal.

Competitiveness inside and outside of the Euro area

The story of how the creation of the Euro has led to a large discrepancy in unit labor costs among members of the Euro area is often used to explain the large current account imbalances in these countries, which are partly responsible for the current crisis. These imbalances cannot be easily addressed because of the absence of national currencies. It all starts with a chart that looks like this one:

Data is from the OECD and it shows how Greece, Ireland, Italy, Portugal and Spain saw their labor costs increase relative to Germany since the creation of the Euro. This eroded their competitiveness and it is one of the reasons why we find ourselves in a crisis. I wrote last week about how including France in this picture would challenge the conclusion that it is all a problem of the club-med countries (plus Ireland). As I argued there, France looks a lot like all of those countries, Germany is the outlier.

I was curious to see what other countries looked like during those years. So I added the Netherlands and this what I now had:

There are some differences and the increase in unit labor cost was lower in the Netherlands and France than in the other two countries but the real message is the same as before: Germany is the outlier in this story.

So I was now curious to include other countries, those that are not part of the Euro area. Here is what happens if we now include Sweden, the UK and the US.

It is becoming very difficult to see which country is which with the exception of Germany! This picture makes it even more evident that the Euro effect on unit labor costs is very much driven by one country. And that comparing Germany to countries outside of the Euro area leads to the same conclusion.

The picture above is missing something: it is not correcting for movements in exchange rates, it is just looking at unit labor costs from a domestic perspective (measured in local currency). What happens if we convert the numbers for the US and the UK into Euros?

Now the story is very different for the UK and the US. In both cases their exchange rate has depreciated significantly. In the case of the US it is a trend that started in 2000 when he Euro reached its minimum (about 83 cents of a dollar). In the case of the UK, the pound retained its value relative to the Euro until the recent crisis when it collapsed. By 2009, and measured in Euros, the unit labor costs of the UK look similar to that of Germany. In the case of the US they look even lower because of the large depreciation of the dollar relative to the Euro since 2000. Let's not forget that despite the fall in the US dollar, the current account of the US still had significant deficits during all those years.

What I learn from this analysis is that from a domestic point of view, ignoring exchange rates, what happened in Germany in terms of labor costs was very unique. No other country looks like Germany and whether we look at Southern Europe or Northern Europe or we go to the UK or the US we get a similar evolution of labor costs. When you add the exchange rate we find that in the case of the UK and the US, during these years the depreciation of their currencies have made the labor costs fall when measured in foreign currency. How sustainable is this trend is unclear. We have already seen the Euro falling relative to its peak but about 20% which has undone some of what we see in the picture above (the picture above finishes in 2009 when the Euro was higher in value than today).

Antonio Fatás

S&P warning or warning about S&P analysis?

The fact that S&P is now issuing a warning to all Euro countries regarding their sovereign ratings is not a surprise. All these governments are facing difficulty managing their fiscal policy so there is a potential risk in government bonds, a risk that maybe we have been ignoring for too long. And the behavior of S&P is consistent with their earlier practices: Japan or the US are not AAA in their ratings and they seem to like to make announcements around specific events so that people listen to what they have to say.

What is more surprising is that their announcement made it to the front page in all newspapers and seemed to be moving markets. I did not learn much from their announcement except that I confirmed my earlier impression that rating agencies have very little to say when it comes to sovereign debt. Their analysis is shallow and in many cases misleading or even wrong (e.g. in their recent calculations regarding the US fiscal outlook). They are also the same agency that was mispricing risk in the years previous to the crisis. I did a random search of their assessment of the mortgage risk market in 2006 and I found the following document regarding GMAC mortgage corporation:

The rankings for GMACM as a Residential Special Servicer and a Residential Subprime Servicer are affirmed at ABOVE AVERAGE. The rankings reflect the company's experienced management team and staff, comprehensive and effective policies and procedures, pervasive internal control environment, sophisticated technology platform, and comprehensive standards for monitoring key performance metrics. GMACM continues to be a highly efficient loan servicer for a wide variety of residential mortgage loan products and investors. GMACM's management team effectively minimizes portfolio risk through strong internal controls, well-defined risk management methodologies, proactive default management policies and practices, substantial technology enhancements, and effective business strategies. Management continues to execute its business paradigm of strategically aligning various servicing functions with its human capital resources to maximize servicing performance, develop platform-wide best practices, enhance technology, and minimize employee turnover through enhanced career-pathing opportunities. (January 2006)

Sounds good. Too bad that they had to be bailed out by the US government a couple of years later.

And S&P was also the same rating agency that was very pleased with fiscal policy in advanced economies during the years when it was really badly run (during the good years when we should have seen s healthy surplus).

And do not panic if they downgrade all the Euro government bonds, we still have Liechtenstein rated AAA. And if you get tired of government bonds you can invest in the gold market. Visit the S&P web site today and right there were they have their announcement on the warnings to Euro governments you can also find a great video explaining how Gold has dropped to a strong support area, picture below for your entertainment - I wondered if they used a similar graphical analysis to calculate the probability that the German government will default on its debt obligations.

Antonio Fatás

Internal Devaluation German Style

When countries are members of a single currency are, such as the Euro, they cannot depreciate their currencies to boost their exports. The only way to produce the equivalent of a depreciation is to keep costs growing at a lower rate than other countries. This can only be achieved through wage moderation - not in absolute terms but relative to productivity growth. This is sometimes called an internal devaluation and it is normally thought as being more difficult to achieve than a straight devaluation or depreciation of the currency because it involves changes in wages.  Many see this today as a challenge for Souther European countries as they might have been losing competitiveness relative to the other Euro countries and now they cannot just use their exchange rate to gain it back.

The recent OECD economic outlook talked about all this and had a chart that I am reprinting below.

The chart shows unit labor costs of selected Euro countries. The striking pattern of the chart comes from Germany. Germany managed to keep unit labor costs constant while all other countries in the chart saw increasing labor costs since the creation of the Euro. It is important to notice that this is not just about the usual suspects, France looks very similar to all the "club med" Euro countries. What is really remarkable is the behavior of Germany!

What I find interesting in this chart is that with the creation of the Euro, Germany managed to "engineer" such an increase in competitiveness while it did not manage to do it when it had its own currency. The German Mark, as any of the other large currencies in the world (the US dollar) fluctuated in directions that might not always have been in the interest of the country. In that sense, one could argue that Germany had a stronger control of its real exchange rate post-1999 than before. Of course we are talking about the intra-Euro exchange rate. Relative to the US, there is still an exchange rate that Germany cannot control: the Euro/Dollar. And this runs contrary to the way we normally think about exchange rates. It is when you control your own currency that we assume that you can better to influence your real exchange rate.

Just to complement the chart above, I produced two additional charts with the behavior of exports and GDP (both in real terms) during the same period. There are, of course, many other factors that affect these two variables but it is interesting to check what happened to both during the same years. Did Germany benefit from the behavior in unit labor costs?

When we look at exports we see that Germany did better than any of the other countries, although the behavior is not as extreme as one would expect from the behavior of unit labor costs. When we look at GDP then Germany is not an outlier at all and in fact it is one of the countries with the lowest performance during these years (only Portugal and Italy performed worse). There are reasons why we do not expect Germany to grow faster than the other countries (Germany has a higher GDP per capita than all of them) so the absolute comparison might be misleading but it still provides a sense on how the behavior of unit labor costs was reflected in economic activity and exports.

Antonio Fatás

No bailout, just monetary policy: ECB versus US Fed

We enter yet another interesting week in Europe with the same discussions on how high interest rates will be in Italy or Spain, rumors on a possible IMF program for Italy (doubtful) and pressures on the ECB to do more. So far ECB officials argue that "bailing out" Euro governments will violate their legal framework and it is a bad idea. Without going again into the arguments of whether the ECB can and should buy Euro government debt, here is a quick comparison between the US Fed and the ECB in terms of holdings of government debt.

The chart below measures the holding of US government debt at the US Fed and Euro government debt at the ECB. They are measured in billions of local currency (USD for the US, EUR for the ECB).

If you look carefully at the last months you see a small increase in the ECB holdings that reflect their recent  attempts to bring some stability to financial markets and keep bond yields under control. But how does it compare to the US Fed actions over the last two years?

Let's first compare the end point. Today the US Fed holds about 1.7 trillion of US government debt. This represents about 11% of the 15 trillion of outstanding US government debt. For the ECB, the holdings of government debt are about 550 billion (EUR), which is less than 7% of the total outstanding debt of Euro governments (around 8.3 Trillion).

If we look at the evolution over the last years we can see that from mid-2009 the US Fed has doubled the holdings of government debt. If we use 2008 as the starting point then we are looking at an increase of more than 300%. The decrease in holdings of government debt in the years 2008-09 corresponds to the period when the Fed was exchanging treasury bills against other assets such as MBS.

The ECB has also been increasing its holdings of government debt. From 2008 to today it has increased its holdings by about 66%. But this increase is not that different from the previous trend and there seems to be very little change during the crisis except in the last months.

The actions of the Federal Reserve are part of its monetary policy strategy and not an attempt to provide a hidden bailout to the US government (not everyone agrees on this but let's leave that discussion for the future). The comparison between the two central banks and the fact that the ECB is so reluctant to consider a similar action in the current environment where financial and macroeconomic stability are at risk makes the ECB decision even more difficult to understand.

Antonio Fatás

Ratings deflation: a world without AAA bonds?

First it was Greece, then Portugal, Ireland, Spain, Italy and France and Germany are maybe coming next. It seems that soon there will be no Euro countries left with AAA rating. Japan lost it a while ago and the US could follow when the next super-committee does not agree on what to cut or which taxes to increase. What happens if all countries lose their AAA-rating?

We know that in some cases we know that the impact of going down in the ratings is minimal - as it has been in the case of Japan where the government keeps funding very large deficits at low interest rates.

But what is more interesting, given current circumstances, is whether the rating can be seen as an absolute measure of the probability of default or a relative one. If you check what agencies say, they will not give you a definite answer to this question.

To the question "Are Credit Ratings absolute measures of default probability?", Standard and Poor's answers:

"Since there are future events and developments that cannot be foreseen, the assignment of credit ratings is not an exact science. For this reason, Standard & Poor’s ratings opinions are not intended as guarantees of credit quality or as exact measures of the probability that a particular issuer or particular debt issue will default."

Which is not too informative. And then they add:

"For example, a corporate bond that is rated ‘AA’ is viewed by Standard & Poor’s as having a higher credit quality than a corporate bond with a ‘BBB’ rating. But the ‘AA’ rating isn’t a guarantee that it will not default, only that, in our opinion, it is less likely to default than the ‘BBB’ bond." 

This is more informative but not much of a surprise: One would hope that at least the relative ranking of ratings says something about the relative default risk.

So if ratings are relative and they are all going down (deflation has also reached rating agencies), it might not matter much for investors. This might be a signal that the world is more volatile than what we thought, and this is not good news, but as long as nothing else look safer than these bonds, it might not change much the portfolio strategy of investors.

There is however a risk: that investors seek safe returns somewhere else. This happened to some extent in the period 2003-2007 where investors looked for returns in assets other than government bonds because of their low yields. And financial markets were very good at creating assets that offered higher returns and that apparently where as safe as government bonds. But we know that this story did not end very well.

We can imagine going forward a world where investors move away from government bonds not so much because of low yields but because of the perception of risk, and theysearch for yield in assets that appear less volatile or where the expected return more than compensates for the potential risk. We have seen this behavior in the gold market, the exchange rate market (with the Swiss Franc until the central bank said enough). But if deflation in ratings continues it is very likely that we will see similar phenomena in many other markets.

Antonio Fatás

Solvency or Liquidity? (r-g)

As we continue seeing interest rate spreads increasing in the Euro area, we keep asking the question of whether this is a crisis of solvency or liquidity. The fact that interest rates keep increasing makes it more difficult for governments to meet interest payments, and solvency becomes more likely. If default happens we might never find out what type of crisis we had. Were governments insolvent? Or did the high interest rates and lack of funding pushed them into default? And if we cannot tell ex-post, how can we tell ex-ante (now!)?

Let me look at some historical facts to understand the potential scenarios these countries are facing. After my previous post on Italy, let me look at Spain today, one of the countries that is also seeing spreads rapidly increasing.

The way we normally look at the question of solvency is by asking what type of effort the government needs to do to keep the debt under control. The typical benchmark is to look at the current ratio of government debt to GDP and think of scenarios where this ratio remains constant. In the case of Spain, this ratio is 60-67% measured in gross terms and 48-56% in net terms. The range corresponds to the number for 2010 and the forecast by the end of 2011. Let's just use 65% as the relevant ratio.

To keep this ratio constant, the Spanish government has to deliver a primary balance which is equal to the difference between the interest rate it pays on the debt (r) and the growth rate of GDP (g) multiplied by the debt to GDP ratio [Note: the primary balance is the difference between revenues and spending excluding interest payment on the debt].

With a simple formula: to maintain a stable debt-to-GDP ratio you need a primary surplus of

(r-g)  Debt/GDP

Interest rates and growth rates have to be measured in the same units so we either measure them in nominal or real terms. Let me choose nominal rates. 

Some facts: In the period 2000-2011, average nominal growth in Spain has been equal to 5.23% (out of which 2.17% was real growth). If we exclude the crisis years (2008-2001), average nominal growth was as high as 7.46%. 

I am focusing at the post-Euro years as we want to avoid looking at a different monetary policy framework, but just for the sake of understanding history, in the previous decade (1988-99) Spain grew at a rate of 7.73 in nominal terms and 3.11 in real terms.

What about interest rates? Currently the Spanish government is paying about 4% on average (even if on the margin they are facing rates closer to 6%). If Spain managed to maintain rates at 4%, the required primary balance is

(4% - 5.23%) x 65% = -0.74%

So a deficit of 0.74% will do it. This does not look that different from actual numbers. Over the period 1999-2011, the government primary balance in Spain was -0.57%. If we exclude the years of the crisis and focus on the expansion 1999-2007, the primary balance was +2.24%, a significant surplus. 

In other words, if Spain faced an interest rate of 4%, and even if within the next decade they experienced another massive crisis with four really bad years (as bad as 2008-2011) in terms of low growth and large deficits, the debt to GDP ratio would still remain at 65% ten years from now. "Business as usual" would do it. This is a very conservative scenario where we are asking no change in policies to the Spanish government.

Is 4% a reasonable interest rate? No if you look at markets today. But here is where the self-fulfilling nature of the crisis comes in. The relevant question to me is whether can we build a scenario for Spain that is conservative in terms of growth and fiscal efforts coupled with rates which are not too far to a risk-free rate, and where we feel that we can guarantee with almost 100% confidence that the debt will remain stable. If this scenario is possible, then the interest rate of 4% is justified because we are looking at a world of no default. 

Of course, if we start with an interest rate of 15%, then all the calculations above will send you in the direction of default, which would justify the high interest rate we started with. 

But if both scenarios seem plausible then we are facing multiple equilibria and we need to find a way to coordinate to the good one, the one without default. Looking at last week, it seems that we are going in the opposite direction. So it looks like the only way out is for all of us suddenly become optimistic, or the ECB steps in and helps us coordinate to the good equilibrium.

Antonio Fatás


Update: And Paul Krugman is also puzzled about interest rates in Euro countries when you compare them with fiscal policy outcomes in those countries. He cannot understand how interest rates for Austrian government debt remain so high giving its low debt, low unemployment and a current account surplus.