McWilliams hits the nail on the head on Irish Credit Crunch Solution

Banknotes from all around the World donated by...Image via Wikipedia

David McWilliams, echoing Martin Feldstein in the FT earlier this week, suggests a Central Bank refinancing of banks to introduce liquidity into the system using the fact we are in an economic and monetary union. McWilliams' sense of social justice ensures the developers, whom he blames for the present mess (and he's not too far wrong there) will get short shrift from this deal. This is an excellent idea from McWilliams.

The key insight of the post is that EMU is not just a set of constraints on monetary policy---it is also an opportunity to provide free liquidity, at least on the relatively small scale Ireland would require.

McWilliams tells us, in a lovely turn of phrase:

Monetary union is a two-way street. While we can’t affect our interest rates, we can engineer liquidity.

A longer post on this when I start teaching Economics of EU Integration in a few weeks.

The Dossing times accuses McWilliams of Abuse of Statistics

You and All Your Misperceptions

While I wouldn't go as far as Simon from The Dossing Times in accusing David McWilliams of coming to "plain silly" conclusions which represent an "abuse of statistics", I do think Simon has a much simpler (and less time and data intensive) method of discovering why the increase looked so big than mine: McWilliams just compared January 2008 with August 2008, and saw a huge increase. But there are seasonal effects here, as Simon correctly points out when you look year to year, which make this increase look more than regular.

So it's more of a slight gaffe than a deliberate abuse of statistics, and McWilliams should know better, but he has shown us (or at least to me) data source when looking back at Irish emigration patterns. A more sophisticated analysis might be in order here. Anyone interested in helping?

McWilliams incorrect on the Return to 1980's levels of Emigration

Irish economists are reaching for any evidence of an economic meltdown with which to terrorize the public. Commentator David McWilliams, echoing Prof. Kelly (see my previous post), gets medieval on the return of 1980's style emigration. He references GAA transfer figures as representing

a huge increase in young men moving from Irish clubs all around the country to clubs in London and New York.

The idea behind the piece is very good, using a Freakonomics-style capturing of economic choices (in this case, whether to stay in Ireland or go somewhere else looking for work) through an unusual data set.

(If anyone wants to work with me on it, there is an important story to be told using this data, and it goes back many years, so get in touch if you like working with this type of stuff. I've obtained the data back to 2000.)

David is to be congratulated for looking under this particular rock for signs of our imminent economic disaster, because even though I don't think he found much, the fact that this data exists at all is interesting enough to warrant a head nod of gratitude.

I do have some problems with his interpretation of the data. I didn't believe his story, so I went and got the transfer data myself. They come from two GAA pages, here and here.

The data is monthly, and shows player movement from club to club. Some of the clubs are in London, and the UK more generally, the US, Australasia, and 'Europe'. When you take the numbers leaving for foreign soil and divide them by the numbers of people actually moving in a given month, and compare the years 2007 and 2008, you get this picture below. Click on the charts to get the data I used to construct the figures.


Each bar represents a relative group in group ratio, so in January 2007, a figure of 0.09 means only 9% of the people moving moved abroad. We see a big increase in movement over the summer months, with July 2008 being a big (but not huge) increase.

Then when you look at the absolute numbers moving, the story changes again, as the figure below shows:


Now the way to read the figure is to just count up the numbers moving anywhere, not just abroad. It's a straight summation. And do the numbers moving go up? They do not. They go down. Way down. So not an increase, but a decrease in movement over time from 2007 to 2008.

Here is the same figure as the one above, but this time just for people traveling abroad.


Again, the way to read the figure is just to add up the columns, so 101 people transfered to a club outside Ireland in April 2007. The only huge increase is in May 2007, well before the subprime crisis hit in August 2007, and not repeated through the economic doldrums we are in now.

Even looking at the raw numbers---total foreign departures in 2007 to August (to make the data ranges the same in both years) are 723 out of 2062 people transferring, so a ratio of 35%. In 2008, total foreign departures to August are 588 out of 1346, a ratio of 0.44, but still a decrease in the absolute amounts of people moving abroad to date.

So based on this simple exercise, I'd have to say David's conclusions are incorrect about rising levels of Emigration. The data just don't tell that story so clearly.