Introduction
If you study political sciences, there is a slight chance you will be offended by the current blog post. Lets just make it clear that you have been paying money to receive education on how to lie to people, so suck it :-)
I was never into politics. Coming from a logic-oriented background I found the fuzzyness and strategic manipulation of politics to be something not deemed worth of exploring; e.g. what person in their right mind does not like yes or no questions? In that regard, on every occasion that avoiding such conversations is not possible, I try to isolate the logic behind things and see through all the bs that political speech is about. You see politics for me is the art of saying that different truths exist over the exact same situation. That is a large nope for any type of logician.
Nevertheless, with me being Greek it would be hard to avoid or stay unaffected with the current debt crisis that came into everyone's lives during the last five years. With the crisis being among many and a politic problem, I've spent quite some time arguing with people whether the crisis was exclusively the Greek's fault or were they wronged by the Germans or any type of other theory that could explain why we are in the situation we currently are.
When arguing with people there is one specific thing I found to be generally true; when the tensions are high the human mind tends to tackle the problem by creating solutions of a very specific type; binary solutions. More specifically in politics people tend to argue that their solutions/arguments are better than any other type of solution. With me being a mild nerd, but nerd enough to find that the infinity introduced by the operating keyword any does not fit well with me, I had to come up with a specific approach that could show what the problem was as objectively as possible.
And here comes the kick from the statistician inside me. Why not use objective methods like Bayesian estimators[1] to see how the problem evolved from before the crisis to where we are right now? Maybe that would shed some objective light in the very very shady situation we have at hand. The research question I would like to answer in the end of this post the following: Was there any way of knowing what that the crisis would happen eventually? If yes, why did it happen?If no, what happened? Whose to blame for all that? End of intro.
Data gathering
If you have done any type of data analysis you would know that gathering data, cleaning them up and generally bringing them to a point that they can be analysed can be a pain. I had to go with data that are more or less clean or need minimal cleaning, so that the work could be done in a single or maybe two weekends. For that reason I chose Eurostat[2] which is the European statistics service. They provide nice tables, and their data are clean-ish, downloadable and they have a huge collection of tables. So analysing those table wouldn't be too much of a hassle. Neeto.
For my initial analysis I chose two metrics that seem to be receiving most of attention the by the media: GDP of a country and unemployment rate of a country. With that in mind I went to the Eurostat database and searched for those two metrics. I found some tables consisting of entries regarding the GDP and unemployment rate of the countries in the Eurozone from 2006 to 2015. Many countries did not have entries in the years 2014 and 2015 so I chose to analyse year 2013 and also one year pro-crisis so that I had I good comparison base. Since the data start from year 2006 I went by what Eurostat gave me as time t=0.
Now I had a question that needed answer, and I also had the data. What is left is to make sense of their connection. This leads us to:
Results
Now that the data are cleaned and nice, I had to choose what to do with them. I firstly chose to cluster them using a Bayesian method known as Gibbs sampling[3]. What is nice about this method is that it is mathematically proven to converge to the true underlying distribution of the data as the number of iterations approaches infinity. What this means in simpler terms is that if you need to group your data and don't know how, Gibbs sampler will tell you what the statistically correct way of grouping your data is. Simple, right? Lets see what I found.
 | |
| Figure 1. 2006 GDP percentage |
 | |
| Figure 2. 2013 GDP Percentage |
The top picture shows how the sampler fits the GDP percentages of countries in the Eurozone for the year 2006. The actual data passed to the sampler are the ploted as circles lying by the x-axis. I intentionally removed all the country labels so that I could be as objective as possible. What the sampler does is it finds the number of distributions that best fit the data according to science. The number of distributions it outputs is arbitrary; it could be one it could be ten but in our case it is two.
Personally, it came as no surprise that I had two clusters fitting my data. This means that in 2006 we had two major groups of economies in the Eurozone. A group with higher GDP and a group with lower GDP. If also pay attention to the detail you'll see that the distribution with the higher variance is the one further away from point 0. In 2006 we had more rich than poor countries. Well, wasn't it obvious? For some people yes, for some people no. For most people, thankfully, yes. A question I had now was how was this imbalance between economies dealt with? Also how does the 2006 graph compare to the 2013 graph?
According to the second figure we now have economies with negative GDP(yes, we all know where the finger points) and generally a shift towards GDP of ~0 in some of the countries in the Eurozone. You can also see that the distribution with the higher variance is now the one on the left in comparison to 2006. Countries are poorer in the 2013 screen-shot. Well, to be fair, that was more or less expected; a worldwide crisis does not increase GDP. But what happened to the unemployment rates over the same period and how do these two correlate?
According to the second figure we now have economies with negative GDP(yes, we all know where the finger points) and generally a shift towards GDP of ~0 in some of the countries in the Eurozone. You can also see that the distribution with the higher variance is now the one on the left in comparison to 2006. Countries are poorer in the 2013 screen-shot. Well, to be fair, that was more or less expected; a worldwide crisis does not increase GDP. But what happened to the unemployment rates over the same period and how do these two correlate?
 |
| Unemployment rates 2006 |
 |
| Unemployment rates 2013 |
The results of the unemployment rate clustering were very surprising, for me at least. Even though there was a difference between the GDP's of the countries, unemployment rates in 2006 were so close that could be actually grouped together. Since more jobs means more spendings, governments with less money in their budget had to find those money from somewhere to reduce their unemployment rates. After doing some research I found out how this whole thing worked. The answer was deficit spending[4] programmes. But what does that mean?
Normally, a country is allowed to spend as much money as it collects from taxes. If the country wants to spend more, it has to borrow that money. This type of spending is called deficit spending. With Greece being in the Eurozone since 2000, it was able to create a
lot of deficit spending programmes due to lower interest rates that the package of being in the euro came with. (Corrupt)Politicians promised jobs, and after getting elected they had
to come through their promises so that they would continue to stay in power. Consequently, they borrowed money to create new jobs, and when the time to repaying those debts was nigh, they just applied for new loans. Their thinking was: as long as we can borrow new money, accumulating debt is ok. Ostrich-type thinking mode on, what unique prerogatives politicians have. Of course when there is no money to borrow, the whole thing collapses. With that and Murphy's law[6] in mind, we arrive in 2008 and the American housing crisis.
The number one economy in the world would receive a huge blow from the housing bubble, Lehman brothers went bankrupt while holding an staggering 600 billion dollars in assets and hence came the worst crisis since the great depression[7]. They say that in the economy everything is intertwined and this was definitely the case for the relation of the European-America system. What this meant for the small country of Greece was that borrowing money to repay debts was no longer an option. Oh boy oh-boy here comes Johnny. And the collapse came, but not just for Greece. Other countries where also affected and in the 2013 unemployment figure we have clearly two clusters in the data. Countries that had stable economies and where affected less by the crisis, and countries that were affected a lot and their unemployment rate sky-rocketed with Greece of course being classified in the latter group.
Contrary to what I would expect and while being in the midst of this worldwide crisis the countries of the Eurozone decided that instead of letting Greece go bankrupt, it would be wiser to issue the highest loan in the history of a country at peace. Quoting wikipedia "On 2 May 2010, the European Commission, European Central Bank (ECB) and International Monetary Fund (IMF), later nicknamed the Troika, responded by launching a €110 billion bailout loan". To make things worse, even €110 billion were not enough and a year later another €130 billion were issued again as bailout money. The thing that makes you wonder, is why did the Europeans not let Greece default? Well, the simple answer is that they would lose money too. A lot of them actually. You see, Greek banks had also borrowed money from German and French banks too. Letting Greek banks default would also create a legal precedent that other high risk countries like Spain and Portugal could exploit so that they did not pay back their debts. See? That's a bad version of being intertwined. Especially if you're the loanee. Also, not a very nice period to be a minister of economics I would say. So loans where issued and banks where saved. What happened eventually was the shift from a private sector debt to a public sector debt. The banks no longer owned money. It was the Greek government that owned money now. Not the road Ireland chose btw, but that's a story for another post.
The number one economy in the world would receive a huge blow from the housing bubble, Lehman brothers went bankrupt while holding an staggering 600 billion dollars in assets and hence came the worst crisis since the great depression[7]. They say that in the economy everything is intertwined and this was definitely the case for the relation of the European-America system. What this meant for the small country of Greece was that borrowing money to repay debts was no longer an option. Oh boy oh-boy here comes Johnny. And the collapse came, but not just for Greece. Other countries where also affected and in the 2013 unemployment figure we have clearly two clusters in the data. Countries that had stable economies and where affected less by the crisis, and countries that were affected a lot and their unemployment rate sky-rocketed with Greece of course being classified in the latter group.
Contrary to what I would expect and while being in the midst of this worldwide crisis the countries of the Eurozone decided that instead of letting Greece go bankrupt, it would be wiser to issue the highest loan in the history of a country at peace. Quoting wikipedia "On 2 May 2010, the European Commission, European Central Bank (ECB) and International Monetary Fund (IMF), later nicknamed the Troika, responded by launching a €110 billion bailout loan". To make things worse, even €110 billion were not enough and a year later another €130 billion were issued again as bailout money. The thing that makes you wonder, is why did the Europeans not let Greece default? Well, the simple answer is that they would lose money too. A lot of them actually. You see, Greek banks had also borrowed money from German and French banks too. Letting Greek banks default would also create a legal precedent that other high risk countries like Spain and Portugal could exploit so that they did not pay back their debts. See? That's a bad version of being intertwined. Especially if you're the loanee. Also, not a very nice period to be a minister of economics I would say. So loans where issued and banks where saved. What happened eventually was the shift from a private sector debt to a public sector debt. The banks no longer owned money. It was the Greek government that owned money now. Not the road Ireland chose btw, but that's a story for another post.
Discussion
So how did we reach this point? It seems to be very unlikely that nobody saw that coming. It took me about a day to come up with the idea, gather all the data and a full day to finish this article. It may be the hindsight effect that makes it easier for people to know what was at fault, and I know that predicting things is not a simple problem, but I find it hard to believe that nobody saw that coming.
All in all, I think that someone must have seen this tsunami coming. Ever since its creation the economic union has included countries with differences in their economy, culture and many other aspects. It is hard to imagine a single centralized monetary policy(what the ECB currently does) that would be able to handle the large variations in every member state. The way I see it, the Eurozone will either have to end up with a central monetary AND fiscal policy(currently every country has its own fiscal policy) that has member-specific rules so that it can cope with differences between country members(some sort of united europe) OR downsize and keep a core of strong economies that are able to keep up with the current general monetary policy(some sort of a country level unit testing[5]).
We currently are in a dire need of radical changes that will change the internal mechanisms of Eurozone. Be it for the better or worse, the only thing I know that change at this point is not optional.
The thing is that the people currently governing are afraid what the political cost of those changes would be; and they should be. A lot of those people would lose their positions. But change is never easy. Contrary to what people believe, change is also not just. Change is what took us so far, and will keep on happening no matter if we like it or not.
The thing is that the people currently governing are afraid what the political cost of those changes would be; and they should be. A lot of those people would lose their positions. But change is never easy. Contrary to what people believe, change is also not just. Change is what took us so far, and will keep on happening no matter if we like it or not.
To sum things up and answer my research questions:
- Was there any way of knowing what that the crisis would happen eventually? - YES
- If yes, why did it happen? - There are inherent flaws at the core of the Eurozone's mechanism that need to be tackled so that such problems can be avoided in the future. If people think that having a central European government is too intrusive, this will lead to the end of the Eurozone as we currently know it; otherwise the europe will end up copying USA and have central monetary and fiscal policies. I'd love to see how this plays out in the average American's ego.
- Who is to blame? - Since blaming is what people like to do, but I'm not very fond of it, I will use a nice formula and leave the weight adjustments to the reader \( blame = w_1*lender + w_2*lendee \).
- What about the people studying political sciences? Are they to blame? -No, but I envy your ability to enter a position of power that affects that many people. Don't get me wrong, you still suck :P
References
[1]https://en.wikipedia.org/wiki/Bayesian_probability
[2]http://ec.europa.eu/eurostat
[3]https://en.wikipedia.org/wiki/Gibbs_sampling
[4]https://en.wikipedia.org/wiki/Deficit_spending
[5]https://en.wikipedia.org/wiki/Unit_testing
[6]https://en.wikipedia.org/wiki/Murphy%27s_law
[7]https://en.wikipedia.org/wiki/Financial_crisis_of_2007%E2%80%9308
[8]https://en.wikipedia.org/wiki/Greek_government-debt_crisis
[6]https://en.wikipedia.org/wiki/Murphy%27s_law
[7]https://en.wikipedia.org/wiki/Financial_crisis_of_2007%E2%80%9308
[8]https://en.wikipedia.org/wiki/Greek_government-debt_crisis
