Tuesday, November 30, 2010

The European crisis = the Euro crisis?

"The Euro is to blame for the current crisis in Europe". I am sure this sentence sounds familiar to many. The argument is simple: as Euro members cannot devalue their currencies anymore, they do not have an option to improve their economic conditions (by favoring exports), growth suffers and their high levels of debt become unmanageable. The Euro area is not an optimum currency area (it lacks labor mobility, fiscal transfers, etc.) so this was a crisis waiting to happen. Paul Krugman says it here, Simon Johnson says it here and you can find many more articles in the business press repeating these arguments.

Let me (partially) disagree with that statement and just bring an alternative view to this issue. It is not a view that denies the importance of exchange rates or the fact that the constraints of the Euro area (one monetary policy might not fit all) might be hurting Euro economies but I think that it is healthy to question our priors on the importance of the exchange rate in explaining some of the empirical phenomena we observe these days in Europe.

A couple of disclaimers before I present my arguments: these are difficult questions to answer. When looking at historical episodes, one needs to control for all variables and their behavior during that time. I will not do that here. Partly because those arguing about the costs of the Euro do not do it either, partly because it will involve heavy statistical analysis which might not be suitable for a blog.

A second disclaimer: I will use Spain as an example because it tends to be used as an example of the problems faced by Euro members and because it is seen as the next in line to ask for funding (it is not related to my nationality...).

Spanish unemployment has reached 20% during the current crisis and there is a sense that unemployment will remain very high for the foreseeable future. The logic says that if Spain could devalue, unemployment would fall very fast. How fast? Let's go back in time to when Spain could and did devalue. The previous recession in Spain (if we do not count the slowdown in 2002 as a recession) was in 1992/1993. At that point Spain had a fixed exchange rate relative to other European currencies. Prior to the crisis it had gone through a period of real exchange rate appreciation (because of high inflation) that had eroded its competitiveness. Very similar to what we saw before the current crisis. The difference is that in 1993 Spain decided to devalue its currency (three times between June 1992 and September 1993). What happened to unemployment during the years that followed?

Below is the picture of the unemployment rate before and after the devaluation. We cannot compare with the current crisis yet, we will need to wait a few years before we see how fast or slow unemployment comes back to a "normal" level, but as we can see in the chart below, unemployment remained extremely high in the years that followed the devaluation of the peseta. Growth picked up but not fast enough to generate employment at a fast enough rate.


Second observation: if staying outside of the Euro area produces significant benefits, a quick comparison between the UK and Spain should be very revealing. The UK has seen a large devaluation of the Pound during the crisis while Spain had no option to devalue its currency.

Below are two charts that summarize the fiscal tensions that each of these two countries are witnessing in terms of fiscal policy as well as the overall economic growth before and after the crisis. There are many ways to read these charts and I am aware that I am abstracting from many other variables but keep in mind that they could make the argument stronger or weaker depending on how you read those other variables. For example, one could argue that the real estate bubble in Spain was more significant than the one in the UK (which is true if one looks at activity in the construction sector as a measure of the bubble) so it is likely that the crisis had a much larger effect on Spain than the UK. One could argue, on the contrary, that the crisis of the financial sector had a larger impact on the UK than in Spain so it should be the UK economy that had a sharper downturn. I ignore all these arguments and simply compare a couple of variables during these years (by the way, data coming from the World Economic Outlook Database, IMF).

My reading of these two charts is that differences so far are small. The best argument one could use to highlight the costs of the Euro for Spain would be the forecasted growth for 2010, below that of the UK (but this ignores a better performance during the previous two years as well as possibly the positive effects of the Euro in the years that preceded the crisis). Apart from this, the rest looks very similar. In fact it looks as similar as if I were to compare a pair of two random European countries (within or outside of the Euro area).



A more direct comparison of the effects of the Euro could potentially be seen below in a chart of the relative performance of Spanish and UK exports since 1995. Despite the deterioration of competitiveness of Spain because of the Euro (and higher inflation), exports grew at a faster rate than in the UK (data is from WDI, World Bank, exports measured in current USD, last available year 2008).


So if the Euro is not to blame what do we make out of all the efforts of countries like Germany or France to try to save the offenders (Greece, Ireland, Portugal and Spain) from default? Isn't this linked to the Euro project and a potential failure of this project? Maybe, but there is something which is at least as important: in an integrated area (the European Union not to be confused with the Euro area), there are so many economic linkages, both in trade and capital flows, so that the performance of each individual economy is strongly linked to the performance of others. This is not because they share a currency but because they are highly integrated. Sweden has announced that they will provide funding to Ireland - this is not because they share a currency (they do not), but because there are strong links (e.g. financial links) that makes it in its best interest to do so. The Mexican bailout after the 1994/95 crisis had nothing to do with the sharing of a currency but with the economic interests of the US (and the IMF) in not seeing Mexico go into an even deeper crisis. And in that case, as it would happen with Spain or Ireland today, the existence of a currency in Mexico made matters worse as the devaluation sent all the USD liabilities of the Mexican economy to an unsustainable level.

Sharing a currency has its costs and benefits. The costs are mostly about losing the ability to manage monetary policy as a stabilizing tool. These costs tend to be more visible during times of crisis when we are asking monetary policy to act strongly. So no surprise that this is a good time to ask ourselves whether the Euro is working and whether it was a good idea in the first place. But to reach a conclusion we need to look a broader set of issues that go beyond sharing a currency and a central bank. If we simply stick to the textbook conclusion of why fixed exchange rates are bad, we might miss the bigger picture.

Antonio Fatás

Tuesday, November 23, 2010

Two percent or a bit below

In recent speeches, Ben Bernanke has referred to a(n) (implicit) target of around 2% for inflation in the US. The U.S. Federal Reserve, together with the Bank of Japan, is one of the few central banks in advanced economies without an explicit inflation target.

Bernanke’s reference to this number is in the context of concerns that QE2 (second round of quantitative easing) will increase inflation in the US and with this message he wants to reassure the public that inflation will remain low. The exact words that he is using are

“FOMC participants generally judge the mandate-consistent inflation rate to be about 2 percent or a bit below.”

It is interesting that the expression “2 percent or a bit below” happens to be almost identical to the way the ECB currently refers to its mandate for price stability.

Some history about the ECB mandate: Originally, the ECB was given the mandate of maintaining “price stability”. This mandate was later (October 1998) made precise by the governing council of the ECB as “a year-on-year increase in the Harmonised Index of Consumer Prices (HICP) for the euro area of below 2%”. At that point there was also a reference to this target, or to price stability more generally, a a mandate “to be maintained over the medium term”.

This initial definition of price stability by the ECB received some criticism because it was leaving too much room for interpretation: Is any inflation rate below 2% consistent with price stability? Is deflation consistent with price stability or even desirable? This led to a redefinition of “price stability" by the governing council of the ECB in May 2003. At that point, the mandate of price stability was defined as “maintaining inflation rates below but close to 2% over the medium term”. So 2% becomes a ceiling (for the medium term) and the goal is to be close enough to the ceiling. Clearly, in this definition there is a sense of asymmetry: 2.1% inflation is worse than 1.9%.

The words chosen by Ben Bernanke are almost identical to the words chosen by the governing council of the ECB where “below but close” has been replaced by “or a bit below”.

What is the record of the ECB? Clearly the ECB has managed to keep inflation very close to 2% and as such we could call it a success. However, given the definition of price stability, it seems that inflation has remained many more months above than below 2%. Below is a picture of Euro inflation from the launch of the ECB until the recession of 2007/2008. Inflation is close to 2% but it would be more accurate to describe it as “above but close to 2% than “below but close to 2%”. Or using Bernanke’s words “2% or a bit above”.

If we add the last two and a half years of data (picture below), towards the end of 2007 inflation increased substantially and then it fell during the recession, below the 2% target. Today, inflation is once again approaching the 2% target.

Antonio Fatás

Thursday, November 18, 2010

Not-so-balanced risks when it comes to inflation

One more posts on the inflation outlook in the US related to our earlier post. With the release of the last CPI data from BLS, we continue to see a downward trend in inflation. In particular, we have seen the smallest annual change in core inflation since 1957. This is a good opportunity to reproduce the chart below (via Mark Thoma, original source San Francisco Fed).

It compares the evolution of core inflation in the US until September 2010 with the evolution of core inflation in Japan in the months that preceded deflation.

Quite similar so far. The US is today where Japan was in mid 1995. Let's hope that the months ahead show a divergent path between the two lines.

Antonio Fatás

Tuesday, November 16, 2010

How negative should real interest rates be?

Standard monetary policy is about setting short-term nominal interest rates. Most macroeconomic models assume that inflation is sticky (constant) in the short run and by moving nominal interest rate the central bank is actually setting the real interest rate and by doing so influencing spending (consumption and investment) decisions. Of course, these spending decisions might depend on long-term interest rates and therefore we also need to understand how short-term interest rates affect both nominal long-term rates and inflation over a longer horizon (where we cannot assume that inflation is constant).

We can use this logic to think about the most recent quantitative easing policies announced by the Fed. That's what Mark Thoma does very well today in his blog. One issue that I am missing in his analysis is how we think about real interest rates (not just nominal) in the current context. This is very much related to the defense that some Fed officials have done over the last hours of their policies. For example, in his interview with the WSJ, Janet Yellen argues that QEII (the next round of quantitative easing) is not intended to raise inflation. That the Fed is happy with an inflation rate below but close to 2%.

I understand the importance of having a "low and stable" inflation target but we need to keep in mind that these targets should be interpreted in a medium-term framework, we are not asking the central bank to deliver a constant 2% inflation every month, quarter or year. And given that the Fed has refused to adopt a formal inflation target to keep its flexibility to set inflation on a short-term basis, why do they seem so obsessed with ensuring that inflation always stays at or below 2%? Even the ECB that is some times seen as putting too much emphasis on inflation has let the Euro inflation rate go above 2% during many of the months it has been in existence, so a little flexibility above 2% in the communications of the Fed might not hurt.

We can also think about what all this implies for real interest rates, by asking: what should the level for real interest rates be given current economic conditions? We know that with short-term rates at zero (and they cannot go lower)sending a strong message about inflation being below 2% sets a floor for how low real interest rates can go (the floor is -2%). Estimates of what the appropriate real interest rate is in the current situation (which tend to be made within the context of a Taylor rule) vary but some suggest that real interest rates might need to be even lower than that [By the way, I find this related post by Krugman very useful to understand the logic behind negative real interest rates].

In addition, we have the issue of the dynamics of expectations and actual inflation. It might be that Fed officials by sending a very strong message about not wanting to increase the inflation rate above 2% will keep inflation expectations low and actual inflation remains significantly lower than the 2% "target". My guess is that their conservatism when it comes to inflation is the results of the strong criticisms that they have received (both at home and abroad), which has sent them into a defensive position where they need to reassure everyone that their current policies are not about raising inflation. But this might not be optimal, while anchoring long-term expectations of inflation around a low target is reasonable, there is nothing wrong in admitting that one of the goals of the current policy is to ensure that inflation stops falling and that we go back towards 2% or even higher in the short-term.

Antonio Fatás

Wednesday, November 10, 2010

Maybe economists should only have one hand

We have all heard many jokes about economists and how difficult it is to get them to agree on anything. Most economists always answer a question starting with the words "it depends" and then follow with the expression "on one hand...". I am not here to defend (or criticize) my profession but to point out how difficult it is these days to get a consensus among some basic macroeconomic issues.

As an example, the debate about whether inflation or deflation is more likely, and about whether the aggressive response of central banks is appropriate today is at the heart of some of the most basic issues in macroeconomics. The disagreement could potentially be the outcome of a more uncertain world but seeing how strong are the beliefs of those who express a view on this subject, it seems that we do not have that much uncertainty at the individual level (those who believe that there will be inflation seem quite sure about, same for those who are concerned about deflation) but rather a polarization of views.

One example that I always find interesting is the debate that one finds in the minutes of the monetary policy meetings at the Bank of Japan. When discussing the inflation outlook in Japan in recent yeras, you can always find views on both sides, those who are concerned with deflation and those who are concerned with inflation picking up. Here is a paragraph from the meeting back in April 2010.

"Regarding risks to prices, some members said that attention should continue to be paid to a possible decline in medium- to long-term inflation expectations. One member expressed the view that attention should also be paid to the upside risk that a surge in commodity prices due to an overheating of emerging and commodity-exporting economies could lead to a higher-than-expected rate of change in Japan's CPI."

Of course, given the last 10 years of data in Japan, it seems awkward that some are concerned with the upside risk to inflation. While one cannot completely rule out this possibility maybe erring on the other side, making the mistake of letting inflation be "too high", for a few years would be good for the Japanese economy.

Clearly the US or Europe are not in the same situation as Japan but given some of the recent commentary about inflation I wonder whether we are getting close to a debate with too many hands and too many scenarios that leads to a lack of strong actions in the right direction. One can make mistakes in both directions (too much or too little inflation) and only time will tell in which direction our mistakes go, but given what we know about inflation, inflation expectations and long-term interest rates (all of them are low, stable or falling), it seems that we are worrying too much about the potential mistake of being too aggressive when it comes to monetary policy.

Antonio Fatás

Monday, November 8, 2010

Gold Standard?

Robert Zoellick, president of the World Bank, has an interesting piece on today's Financial Times about the need for a new mechanism for currency co-ordination. I like the idea of stronger co-ordination but I was surprised to read about the potential role of gold:

"The system should also consider employing gold as an international reference point of market expectations about inflation, deflation and future currency values. Although textbooks may view gold as the old money, markets are using gold as an alternative monetary asset today."

In this case I have to agree with the textbook view of gold as the "old money". The best argument in favor of using gold as a reference for currencies or central bank policies is to provide a nominal anchor in order to avoid inflationary policies (i.e. constraining central bank policies). The debate today about currency values/manipulation is about the relative price of different currencies and not about irresponsible monetary policy that might lead to inflation (to be more precise, some are afraid that the current monetary policy stance of many central banks will lead to high inflation but so far there is no evidence of high inflation and some countries are still facing the risk of deflation).

So I am all in favor of debating the need to come back to a system of fixed exchange rates (or even a one-currency world) but I cannot see how gold could play a role in that system.

Antonio Fatás