Thursday, February 26, 2015

The Eurozone Counterfactual

Imagine the ECB had not raised its interest rate target in 2008 and 2011, but had lowered it. Also imagine the ECB began its open-ended QE program back in 2009. Would there now be a brighter future for the Eurozone? If the answer is yes, then the Eurozone economic debacle is at its core a monetary policy crisis.

There are compelling reasons to believe the answer to this counterfactual question is, in fact, yes. A comparison of total money spending growth in the United States and Eurozone is one of them. Unlike the ECB, the Fed did cut its target policy interest rate quickly and implement QE programs. Though these programs were flawed, the Fed was able to promote a stable growth path for total money spending because of them. And it did so despite a tightening of U.S. fiscal policy beginning in 2010. This suggest the ECB could have done the same for the Eurozone economy. Instead, it did not and the growth of Eurozone nominal GDP growth faltered.


Another reason to believe this counterfactual is to note that the aggressive monetary easing outlined above would have addressed, at least partially, a key problem in the Eurozone: the real appreciation of the periphery. Monetary easing would have done this by causing inflation to rise more in those parts of the Eurozone that were closer to full employment. These regions happened to be the core, particularly Germany. The price level, in other words, would have increased more in Germany than in the troubled periphery of the Eurozone. Good and services from the periphery would have become relatively cheaper.  This was one way to a real depreciation of the periphery. Instead, monetary easing was not tried and so a real depreciation occurred through painful deflation in the periphery.  

Now some observers will object and say the Eurozone crisis was actually a debt crisis. It was the consequence of irresponsible government spending policies that finally came home to roost. For some countries like Greece this was true, but for others it was not the case. Spain, for example, was actually running primary surpluses for several years leading up to the crisis. It was the crisis that worsened its debt position, not the other way around. This can be seen in the figure below. It shows Spain's debt-to-GDP ratio was falling until the crisis erupted. The flat-lining of nominal GDP growth closely matches the surge in Spain's debt-to-GDP ration. This is no coincidence.


This pattern holds up more generally across the Eurozone as seen in the next figure.


These figures point to the Eurozone crisis not being a debt crisis, but a monetary one.

Others will contend the Eurozone crisis is actually an austerity crisis. They point to a positive relationship between government spending and real GDP growth in the Eurozone. For example, the figure below replicates for 33 European countries a scatterplot produced by Paul Krugman that shows this relationship.1  This figure reveals a strong relationship between these two series with a R-squared of 63%.


However, if one pulls out non-Eurozone countries (red squares) this relationship disappears while it gets even stronger for the Eurozone countries (blue diamonds) with the R-squared going to 72%. This can be seen below:


This scatterplot suggests, then, that some omitted variable unique to the Eurozone over this time was affecting both regional economies and their level of government spending. The obvious candidate is the tight monetary policy of the Eurozone.

One, however, could reasonably object that the above scatterplot does a poor job reflecting austerity since it only looks at government spending. Austerity could also work through tax changes and more generally, through changes in the government budget balance. To account for this possibility and to control for the effects of the business cycle on fiscal policy, I plot below the 'structural budget balance' as a percent of potential GDP against real GDP growth for the 2010-2013. This measure comes from the IMF's Fiscal Monitor and includes all levels of government.



We see in this figure a positive and fairly strong relationship between the stance of fiscal policy and real GDP growth for the IMF group of advanced economies. However, like before, if we pull out non-Eurozone countries (red squares) this relationship largely disappears while it gets even stronger for the Eurozone countries (blue diamonds) as seen in the higher R-squared. 


So once again, the evidence points to something unique to the Eurozone that affects both the economy and  fiscal policy. The Eurozone's tight monetary policy over this period fits the billing. 

It is reasonable to conclude, then, that had the ECB had not raised its interest rate target in 2008 and 2011, but lowered it and had it started its open-ended QE program back in 2009 there would now be a much brighter future for the Eurozone. Instead, we now face the prospect of a Grexit for which the ECB has only itself to blame.

.1There is a slight difference between this figure and Krugman's. Here I use the entire 2010-2013 period as one data point for each of the 33 countries. Krugman actually uses each year from 2010-2013 for 33 countries as data points. I chose the former approach since it creates a cleaner scatterplot and actually improves the fit (i.e. higher R-squared).   

13 comments:

  1. I think 2008 would have been the time to start QE.

    ReplyDelete
    Replies
    1. Okay, I am not set on 2009. I just wasn't sure if it could have happened in 2008.

      Delete
  2. The blue dots in the same region of the red dots are similarly distributed. It's the blue dots at the tails of the distributions that are driving the regressions , and there are no red dots in the tail regions.

    You're really stretching here....

    ReplyDelete
    Replies
    1. Anonymous, on the second graph I agree the outliers like Greece are driving the results. But all I am doing is drawing upon the claims of those who say this is evidence for austerity. In other words, if we remove these outliers the relationship between fiscal policy and real growth disappears and the austerity point to which I am responding is moot. I am trying to be generous here.

      Delete
  3. Very good post David.

    Some of the relationship between fiscal policy and growth will be reverse-causation. But within a fixed exchange rate system, like the Eurozone, we should not be at all surprised to see fiscal policy work, since there cannot be monetary offset at the individual country level.

    ReplyDelete
    Replies
    1. Thanks, that is a good point Nick. Maybe another way of saying it is that the fiscal austerity worsened the existing excess money demand problem.

      Delete
  4. Small snapshot, same conclusion:
    https://thefaintofheart.wordpress.com/2015/02/26/austerity-only-explains-outcomes-when-monetary-policy-is-absent/

    ReplyDelete
    Replies
    1. Marcus, that comparison between Spain and Poland is interesting.

      Delete
  5. Some remarks (last graph): Great work. Note that all the blue dots are to the left of zero... Note that the two largest decreases of GDP are for the two countries with the largest contractions, which is consistent with non-linearities. Considering the extreme differences in size between Euro-members there are reasons to exclude Malta and Luxembourg, using a cut off point of 500.000.

    ReplyDelete
  6. Well, it could be that bad monetary policy is just the weapon of choice in a financial war against the periphery.

    ReplyDelete
  7. All you are doing here is merely interchanging the meanings of debt as money. We all know that European "debt" crisis is really an debt crisis. We all know that Europe has a huge bubble many times larger than US, and we all saw EU's risk premium skyrocketed. Such a phenomenon was not occur for example in the US or Germany or (UK). Clearly, this is also an counterfactual other than monetary policy, and this is a real problem.

    ReplyDelete
  8. If QE was perceived to be temporary how did it affect inflation in the 1-2% corridor as you describe on a previous blog post?

    ReplyDelete