One of the lasting ideas of the last financial crisis is the concept of banks being "too big to fail" ("TBTF"). These banks were institutions considered so critical to financial markets that their collapse could bring down the entire economy. As the failure of these large institutions was deemed unthinkable, the phrase was used in support of bailouts to avoid catastrophe.
Since most people (i.e. tax payers) find the concept of TBTF rather unpalatable, especially in a capitalistic society, one takeaway regulators had from the near collapse of the financial system as we know it was mitigating TBTF and, hopefully, preventing future bailouts.
Given the regulatory intention and its significance, it seems only natural to investigate the results in the years that have passed since the crisis and assess progress in the industry.
To empirically measure a qualitative concept such as TBTF before and after the crisis, the Herfindahl-Hirschman Index ("HHI") was applied to the Endless Metrics banking data set, with results visualized below.
By this measure, the answer is, yes, TBTF has become less of an issue since banking concentration hit a peak of 849.1 in 2008Q4. Unfortunately, the improvement is fairly slim.
The recent rating of 597.2 in 2018Q2 is the lowest reading since 2004Q1, when the index sat at 573.9. However, it remains well above the lowest reading of 154.1 in 1990Q4.
To consider the data another way, a percentile approach is used with the top 1%, 5%, 10%, and 25% largest banks. The trend for the top 1% and 5% of banks looks fairly similar to the HHI index. For the larger percentiles, the trend is flatter.
While each trend hits their respective minimum and maximum in different quarters, there remains a fairly consistent story with the HHI. Banks were least concentrated around 1990 and have become less concentrated since the 2008-2010 time period.
That being said, in the last 30 years, the top 25% of banks of controlled at minimum 89.0% of assets and currently 94.1% of assets, which is barely off the peak of 96.9%. There are a lot of little banks - they just don't play a big part when it comes to overall numbers for the industry.
When the data is analyzed using the top 1, 5, 25, and 100 largest banks, there is more consistency and support to the story told with the percentiles and HHI analysis.
The new added insight comes from the trend for the largest bank by quarter through time. Back in 1990Q2, the largest bank held only 4.3% of assets in the banking sector. This amount ballooned to 18.4% in 2007Q3. When one bank controls nearly a fifth of all assets in the entire sector, which in 2007Q3 was 2.36 trillion dollars, then too big to fail starts looking pretty frightening no matter how one tries to spin it.
The results can tell two completely different stories depending on how the data is framed. One can say that, "too big to fail is now less than of an issue than it was during the financial crisis" or say that, "the banking sector is more concentrated now than it was just four years before the financial crisis." The data is open to interpretation.
Today, the largest bank holds 2.62 trillion in assets. This is more than any bank has ever held but it now represents only 12.8% of the total banking industry, which sits around 20 trillion in assets.
It's important to keep in mind that measuring assets is a fairly simple way of measuring concentration risk. If the answer to future crises was "hey, big banks, hold less assets" then the regulatory costs post-2008 wouldn't have cost the industry billions of dollars. But, for those of us that don't want to read thousands of pages of the federal register, asset concentration provides an easy framework to consider.
So, what can we conclusively say about banks that are too big to fail? Well...it hasn't gotten worse. Unfortunately, for those hoping to avoid future bailouts, it doesn't look all that much better. Let's just hope all those billions invested to make the industry safer was money well spent.