Databases of financial crises

On the economic history of financial crises

In my dissertation defense, my examiners and I spent a great deal of time talking about what kind of economic history emerges from the variable-based approach that is employed by economists such as Carmen Reinhart and Ken Rogoff. I’m not trained as an economist, but economic history is an important part of my discipline as well. The variable-based approach refers to long-term databases of dummies that indicate the presence of absence of a variety of financial crises. For example, Reinhart and Rogoff have a book called This Time is Different ( where they describe a variety of crises (banking, stock market, inflation, sovereign debt default, currency) for the last 800 years, in a large sample of countries.

I wrote a chapter on my dissertation on their dataset, and I used their sample of crises on a dataset of employment in nine economic sectors. In brief, however, such work contains serious problems with defining and determining thresholds for financial crises. One problem is that, in Reinhart and Rogoff’s book, a banking crises involves state intervention into the banking sector. Obviously, this makes a banking crisis as political as they are financial. Collecting the data is another problem, as it demands reviewing quite a bit of history.

The Wall Street Journal last week had a post on its MoneyBeat blog ( about a new database documenting financial distress on an index for a sample of advanced countries from 1967 and 2007, by Christina Romer and David Romer. Christina Romer previously served on Pres. Obama’s Council of Economic Advisers, and she recommended the President support and execute a spending program, which she helped to put together, in early 2009. That alone should suggest that the Romers take very different stances—in wider political circles but also within the discipline of economists—than do Reinhart and Rogoff, who, by the way, advocated for austerity measures and produced research (now determined to suffer from debilitating flaws in their data collection and analysis) to support their reasoning. I am very glad that we have this new dataset, because it is based on a very different method than the financial crisis dataset of Reinhart and Rogoff.

I’ll be writing up analysis in the coming days, as I fiddle with the Romer dataset. I need to read the paper that accompanies the dataset and also construct some databases.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s