Bank size is an important consideration in designing a scorecard. First, small banks with limited and unstable funding base and heavily dependent on NIM income for bottom-line profit, hence, are more likely to fail than big banks when economic adversity strikes. The larger diversified banks, with their more varied business, are generally in a stronger position to recover from external shocks. Data from the US FDIC (Federal Deposit Insurance Corporation)
provide supporting evidence of the high failure rate, as shown in Table
. Banking is pro-cyclical and the period covers four recessions or periods of depressed economic activity to observe the behaviour of banks.
US bank failures and assistance 1984–2017
The data support the following tendencies:
Small banks are more likely to fail than big banks. Of the 2835 banks that
failed, 95.5% were small, local, or community banks with assets of less than $10 billion. The figure includes the 1000+ S&Ls that failed between 1986 and 1995. Throughout the period, only one big bank, Washington Mutual Bank, with total assets of $307 billion, failed on 25 September 2008 in the aftermath of the financial crisis and the Great Recession.
The high rate of small bank failures was evident in the aftermath of the Great Recession. The number of total bank failures jumped from 25 in 2008 to 147 in 2009 and continued to climb to 157 in 2010. Still, few large banks failed. The largest bank failure in 2008 was Washington Mutual Bank (mentioned before); in 2009, Colonial Bank, Montgomery, Alabama, with approximately $25 billion in assets; in 2010, Westernbank Puerto Rico, with approximately $12 billion in assets.
The bigger the bank the more likely it would seek and obtain government assistance rather than fail.
We observe that of the 3222 small banks with assets ≤$10 billion, 14.2% of them (400 banks) received assistance, whereas for the larger banks, assistance rather than failing was more prevalent. Six of the 18 banks were in the $10–$25MM category, and three of the bigger four banks were in the assistance group. 55
In regard to balance sheet size, the classification of bank size is arbitrary. The US Federal Reserve defines “large banks” as those with consolidated assets of $300 million or more.
In contrast, a prudential definition of a big bank size was based on the $50 billion threshold, which came about in the aftermath of the 2008–2009 financial crisis and the 2010 Dodd-Frank Act, also known as the Systemic Risk Designation Improvement Act. The law required bank holding companies with more than $50 billion in assets to be subject to enhanced prudential regulations. In May 2018, Congress approved a bill to dismantle key parts of the 2010 Dodd-Frank act that decided which banks were designated “too big to fail”. Under the new rules, a bank is 56 systematically important if it has $250 billion or more in assets. That covers 13 G-SIBs (global systemically important banks), compared to 44 in 2008 in the aftermath of the financial crisis. These banks face the strictest banking regulations. As we said before, we ignore the question of bank size, though it is an important rationale for having separate scorecards.
The second consideration is that the Basel capital ratios are just the
international requirements, regardless of bank size. At the national level, regulators use their own discretions, and usually impose tougher rules than Basel. Data for the G-7 countries and China, shown in Fig.
, show that average capital ratios exceeded the Basel minimum requirement of 10.5%. One explanation for this behaviour of banks worldwide is that the cost of higher capital requirements is offset by the lower risk premium for funding. A bank with a bigger capital buffer is stronger and better able to attract cheaper capital, thus enabling it to maintain its level of lending. Inasmuch as the market drives capital buffering, we would expect that the average bank’s capital ratios would exceed the regulatory minimum and would be closer to the mid-point of the rating scale. In other words, a bank must do even better than the minimum just to be in the upper 50 percentile. We incorporated this second consideration in creating the ranges for capitalisation in Table
Regulatory capital to risk weighted assets. (Source: Federal Reserve Economic Data)