A handful of regional U.S. banks were downgraded in August by rating agency S&P which included Keycorp, Comerica Inc., Associated Banc Corp., Valley National Bancorp, and UMB Financial Corp.
Each bank was downgraded by one notch as shown in the Rating Monitor app found in LSEG Workspace. To access the app, type “RATMON” in the search bar. Users can customize various filters including time frame, country of issue and sector/industry.
Exhibit 1: Rating Monitor for U.S. Banks
Source: LSEG Workspace
When looking at the StarMine Combined Credit Risk Model (CCR), we found that for every bank mentioned above, CCR provided a strong signal in March 2023 of elevated credit concerns and that a downgrade was likely forthcoming by a rating agency.
As shown in Exhibit 2, using Comerica as an example, CCR (highlighted in blue) aggressively downgraded the StarMine Implied Rating by five notches from ‘BBB’ to ‘B+’ over a 9-day period in March. The StarMine Implied Rating is based on the Probability of Default (technical or bankruptcy) over a 12-month horizon, which can be compared to traditional rating agency ratings.
When compared to the green line (S&P and Moody’s), we see that the credit rating for Comerica was unchanged for the entire year until last week’s downgrade. When the agency and the StarMine Model Implied Rating differ significantly, the agency rating moves toward the Implied Rating 4-5 times more often than it moves away.
While it is no surprise that credit risk rose sharply across the banking sector in March when a handful of banks collapsed, we illustrate the value of CCR beyond this episode. CCR is a multi-factor model that incorporates three standalone credit risk models (Structural Credit Risk, Smart Ratios and Text Mining Risk) and covers approximately 44,000 companies globally.
CCR, like the rest of the StarMine models, are updated daily and are more responsive compared to traditional rating agencies, which may assess a single issuer once or twice a year. In particular, the StarMine Text Mining Model and StarMine Structural Credit Risk Model (both included in CCR) are highly responsive, given the models use of input data such as stock prices along with textual data from news, broker research, transcripts and filings.
Exhibit 2: Combined Credit Risk (Comerica Inc.)
Source: LSEG Workspace
We turn to LSEG Datastream to look at small banks vs. large banks from a balance sheet perspective, which indicates a structural difference amongst the two, as shown in Exhibit 3. Using weekly data from the Federal Reserve (H. 8), we can calculate the loan-to-deposit ratio for the banking system as outlined below (Note: Small commercial banks are defined as those that are not included in the top 25):
- The aggregate banking system has loans and leases of $12.1 trillion compared to deposits of $17.3 trillion, resulting in a loan-to-deposit ratio of 69.9%.
- Within small banks, the loan-to-deposit ratio is much higher at 83.1% compared to 62.4% for large banks.
- The loan-to-deposit ratio is below pre-pandemic levels in part due to higher interest rates (lower demand for loans).
Exhibit 3: Loan to Deposit Ratio for U.S. Banks
While small banks tend to lend out a greater proportion of deposits vs. large banks, this may partially explain the difference in profitability amongst the two. Exhibit 4 highlights the forward 12-month net profit margin estimate for large, mid, and small banks as defined by the S&P 500/400/600 banks index.
Smaller and mid-size banks (which will be part of ‘small’ banks in Exhibit 3) have historically had higher net profit margins vs. large banks. The S&P 400 Banks index has the highest forward profit margin estimate of 31.2%, compared to 28.5% for S&P 600 (small) and 23.7% for S&P 500 (large).
This may be attributable to a range of factors including a higher loan-to-deposit ratio, aggressive loan pricing (in part due to lending to smaller/riskier customers), and holding relatively smaller amounts in short-term securities.
Exhibit 4: U.S. Bank Net Profit Margins
To conclude, banks within the S&P 500 are expected to grow earnings by 12.9% in 2023 according to analyst expectations, while year-over-year growth expectations in 2024 are expected to decline by 4.4%. Over the same period, revenues are forecasted to grow by 10.2% in 2023 and then expected to decline by 1.3% in 2024.
For comparison, banks within the Russell 2000 are expected to see earnings decline by 9.1% in 2023 according to analyst expectations, while year-over-year growth expectations in 2024 are expected to decline by 1.9%. Over the same period, revenues are forecast to grow by 1.5% in 2023 and then expecting to increase by 3.7% in 2024.
LSEG Workspace is a complete solution for research and analytics. It places the most comprehensive market information, news, analytics and trading tools available into a desktop.
LSEG Datastream – Financial time series database which allows you to identify and examine trends, generate and test ideas and develop viewpoints on the market.
