Is it Banking Crisis Season?

One really has to wonder if having the season move into “fall” is correlated with the fall of the financial sector. While some time in the making, the 2007-08 subprime financial crisis moved into crisis mode during August of 2007 and by early fall central banks had moved to lower discount rates and pump liquidity into the system. From my casual observations, it seems that as we move into autumn, the global financial sector is about to enter another period of instability and uncertainty with the focal points of concern being China and Italy.

This time around, the Bank for International Settlements has noted that China’s banking sector is stressed as the credit-to-GDP ratio has hit 30.1 percent with a value of 10 being a cause for concern. At the same time, as the Chinese banking system is largely controlled by the government, this is apparently means we need not be as concerned as they will provide a bailout. Somehow, knowing that there is a government taking care of this issue is not all that reassuring for me.

As for Italy, the situation there has apparently calmed down after a surge of fear in July that a bailout was needed to avert a banking sector collapse but the situation is still likely to rear its head. As has been noted, non-performing loans in Italy are equivalent to about 20 percent of its GDP– a GDP that has not been growing in particularly robust fashion due to weak domestic demand as outlined in a post by Focus Economics. Of course, weak domestic demand is in turn being undermined by the uncertainty of what is going to happen in the banking sector.

Returning to my point, I have to wonder if a major banking crisis is more likely to occur in the fall and if so why? It turns out, there is some historical research examining the role of seasonality on banking crises. Carlson and Wheelock (2015) examine the impact of the founding of the US Federal Reserve on seasonal pressures and contagion risk in the banking system. They found that deposit flows were quite seasonal before 1914 but seasonal pressures were reduced once the Fed began to provide seasonal liquidity and reserves.

Of course, in this era the importance of agriculture was still a factor as banks were called upon to make loans in the spring (to finance spring planting) and then take repayment in the fall (once the harvest was in). Naturally, I would imagine that a shock such as a bad harvest could lead to some difficulties in the fall come repayment time.

Is there still a seasonal component to modern banking crises and if so what drives it given a world economy driven by a more diversified economy?  Or is it perhaps simply the return of bankers from their summer vacations that sparks extra activity and attention to banking in the fall?

I really don’t know as the area of monetary economic history is generally outside my area of expertise but it may be an interesting question that others have addressed.

Comments

[Stephen Gordon]

Eric Gyhsels had a paper that remarked the same thing - start dates for recessions seemed to be concentrated in the fall. It was the basis for a lot of his work on seasonality. We're accustomed to stripping out the seasonal cycle, assuming that there's no information there about the business cycle; perhaps we shouldn't.

[Frank Restly]

Not sure if it is a global phenomenon, but there is also the election cycle to consider.

I took the data for U. S. real GDP found here:
https://fred.stlouisfed.org/series/A191RL1Q225SBEA#0

I then created a 12 month ( 6 months each direction ) around each presidential inaugural month (Jan 2017, Jan 2013, Jan 2009, etc.)back to 1947.

Average Real GDP year / year growth inside the 12 month window around presidential inauguration: 2.94%
Average Real GDP year / year growth outside the 12 month window around presidential inauguration: 3.22%

[Avon Barksdale]

Maybe. We have to be careful with these kinds of observations. Look elsewhere effects are a big deal. Both CMS and ATLAS at the LHC saw a combined bump that was nearly 4 standard deviations above background. This summer we learned it was a statistical fluke - the bump went away with more data, no new physics. When you measure so many things stuff will just stand out.

Banking crisis season aside, perhaps world wars only start in late summer ;)