This is the third in a series of articles on the importance - or the influence - of asset quality on valuations. The first article was on energy MLPs (Master Limited Partnerships) and detailed how investors can use bonds ratings and bond yields along with historical DCF (Distributable Cash Flow) projection accuracy to assess risk that results in earnings volatility. The safer investment grade midstream MLPs sell at lower yields and high Price/DCF ratios. The safer MLPs were also the ones that continued with their distribution growth during the credit crisis.
The second article was on Health Care REITs (Real Estate Investment Trusts) and detailed their portfolio by asset type. Some asset types generate higher percentages of private pay incomes, while others generate more income from Medicare and Medicaid. The market prefers assets with high private pay earnings. REITs with portfolios that are heavy in private pay sources sell at significantly higher price to FFO (Funds from Operations) ratios.
Both prior articles stressed that what a company owns strongly influences its price-to-earnings valuation and the yield at which the stock sells. That is a message started in prior articles on BDCs (Business Development Companies), a sector where BDCs with portfolios with senior secured loans sell at higher P/E (or price-to-earnings) ratios and lower yields than BDCs with portfolios of higher yielding subordinated debt. The "quality influences valuation" message will continue in this article on regional banks. This article will end with a short section - and two spreadsheets - showing why dividend growth investors should have an interest in this sector.
Regional banks are highly leveraged companies that - with a few exceptions - primarily own loans. Loans vary as a percent of earning assets. The lower the loans are as a percent of earning assets, the higher the ownership of securities. Banks are legally limited on the type of securities they can own. They strongly tend to own low risk and short duration securities that currently come with tiny yields. Banks with the better net interest margins strongly tend to have a higher percentage of earning assets in loans.
There are many metrics that can be used to assess the quality of a loan portfolio. Banks provide a wealth of data in their earning releases - and the quality of their disclosure in those releases varies widely. I would like to have loan to value data - but few banks provide it. For corporate loans, I would like to have debt-to-EBITDA (Earnings before Interest, Taxes, Depreciation and Amortization) data and interest coverage ratios - but none provide that. For consumer loans, I would like to have data on FICO scores. We can only use data for loan portfolio comparisons that all banks provide. And that data is on non-performing assets.
The data I will use adds non-performing loans; non-performing securities; foreclosed loans or "other real estate owned;" performing loans over 90 days past due; and restructured loans to calculate a non-performing asset (NPA) number. I do not use the individual bank calculations of NPAs. Several banks omit some of those above listed components in their own calculation of NPAs. I take my NPA sum and divide it by total assets to get an NPA percent. The article will provide data that gives evidence to the fact that banks' valuations are strongly correlated to that NPA/assets metric. The NPA percentage strongly influence price-to-book ratios. NPA data also influences P/E ratios.
Let's dive into the data. First I will show year-to-date performance and yields. Then I will show a short EPS history along with P/E data and current year EPS projection spreads. Then I will show a spreadsheet on NPA trends over the last eight quarters. Then I will show data on the correlation of metric attributes and valuations.
South-East, South-West & Pacific Regional Banks 09-13-13
The Q3-13 dividend is used for yield calculations. The Q2-13 book value is used in the price/book ratios - and the change in book is from Q2-12 to Q2-13. Some banks accelerated their Q1-13 dividends into payments in calendar 2012. Special dividends are not included in these stats. Accelerated dividends are treated as if they were paid in calendar Q1-13. CBSH at times pays dividends in new shares or stock splits of 105 for 100 shares. This results in a shrinking book per share metric. PB has had NPA growth due to acquisitions. LTM = last twelve months.
|12-30-11||Current||Dividend||Current||Div/EPS||Pr/Bk||Percent Change||LTM % Change|
|Bank of OK||BOKF||54.46||64.11||0.380||2.37||32.14||1.49||17.72||19.81||1.28||10.94||0.00||1.61||-6.42|
|The KBE Bank ETF (dominated by the mega-cap banks) is up 27.11% year-to-date - with divs it is up 28.05%.|
|With the 10-yr Treasury at 2.88% and the sector average yield (on Q3 Divs) at 2.17% - the spread is -71 basis points.|
|Weeding out banks paying sub ten cent dividends, the average yield would be 2.66% and average price/book would be 1.65.|
|* The change in the dividend and book is the Last Twelve Months. They are the ratios of the Q3-13 div to the Q3-12 div and the Q2-13 book to the Q2-12 book.|
Let's pause and look at the banks that have significantly underperformed the sector average gain year-to-date - First Horizon National, F.N.B. Corporation, Hancock Holding Company, Trustmark Corporation and Westamerica Bancorporation. Four of those five have had EPS projection downgrades during the year. Four of those five have lacked dividend growth. Sector average LTM book growth has been 2.39%. Only one of the banks has outperformed the sector average on the book growth metric. Four out of five have underperformed in their NPA reduction. The underperformance in those key metrics explain their year-to-date price changes.
Let's look at the banks that have significantly outperformed the sector average price gain year-to-date - BancorpSouth, City National Corporation, Glacier Bancorp, Prosperity Bancshares, Regions Financial Corporation and Umpqua Holdings. All of those stocks have had upgrades in their EPS projections. Three of the six have had dividend increases. Four of those six have outperformed on the book growth metric. BXS and GBCI have outperformed in their NPA reduction. The superior performance in those key metrics explains their year-to-date price changes.
Banks Price/EPS Ratios 09-13
|EPS / Share||% EPS Growth||Price/EPS||13 EPS Range|
As you will see in the upcoming data, the two of the banks with the highest P/E ratios - WABC, Cullen/Frost Bankers - both have much lower than average NPAs. That is the only metric where they are noteworthy. A history of superior NPA metrics resulted in neither one cutting their dividend during the credit crisis. But WABC has not had recent dividend growth. CFR's dividend growth is nothing special. The book growth for both is below sector average. The forward CAGR projections for both are close to average. The bank with by far the lowest P/E ratio - RF - has by far the worst NPA percentage.
Bank NPAs Ending Q2-13
|Non Performing Assets/Assets Ratio||% Change in NPA Ratio||LTM NPA Chng|
NPAs, year-to-date returns and Price/Book Ratios
The Q2-13 NPAs to total assets ratio goes a long way to explaining the variety of valuations. In some years, this metric also is a tool that explains YTD price changes.
The following companies had Q2-13 NPAs to assets of greater than 1.75%: BXS, Cathay General Bancorp, FHN, RF and ZION. Their mean price gain for the year is 26.90%. Their mean total return for the year is 27.53% - and 3 of the 5 beat the sector mean yearly price gain [26.43%] and they sold at an average price/book ratio of 115.39.
The following companies had NPAs to assets ratios of more than 1.00% but less than 1.75%: BOK Financial (NASDAQ:BOKF), FNB, GBCI, HBHC, TRMK and UMPQ. Their mean price gain for the year is 24.69%. Their mean total return for the year is 27.40% - and 2 of the 6 beat the sector mean yearly price gain and they sold at an average price/book ratio of 134.50.
The following companies had NPAs to assets ratios of less than 1.00%: Bank of Hawaii, Commerce Bancshares, CFR, CYN, East West Bancorp, PB, UMB Financial Corporation and WABC.
Their mean price gain for the year is 27.44%. Their mean total return for the year is 29.35% - and 4 of the 8 beat the sector mean yearly price gain and they sold at an average price/book ratio of 187.94.
2014 has been an atypical year when it comes to the correlation between low NPAs and superior year-to-date returns. There are two reasons for that result. The banks with higher NPAs in Q2-13 had even higher NPAs this time last year. Falling NPAs has resulted in falling - or even negative - loan loss provisions for those banks in 2013. Lower loss provisions have resulted in higher EPS growth. High NPA banks have been the banks that have paid very small or token dividends. Those small payout ratios have resulted in generating some of the best book value growth in the sector. To everything there is a season. And those factors has made this a good season for bad banks. This is a season that began in 2012.
NPAs and P/E Ratios:
The banks with Q2-13 NPAs to assets of greater than 1.75% had a mean 2014 P/E ratio of 13.86.
The banks with Q2-13 NPAs to assets between 1.00% and 1.75% had a mean 2014 P/E ratio of 14.43.
The banks with Q2-13 NPAs to assets of less than 1.00% had a mean 2014 P/E ratio of 15.91.
The data I have just provided on price-to-book ratios and P/E ratios show the valuation impact of asset quality in this sector. Asset quality matters - and it matters in all the sectors where I track data.
The three banks with the smallest percentage of NPAs - UMBF, WABC and CBSH - have price-to-book ratios of 1.76, 2.33 and 1.86. There are banks with inferior NPA metrics selling at higher price-to-book ratios than some of those. So why would banks with inferior NPA metrics sell at higher price-to-book ratios?
It was said by Vince Lombardi that "winning isn't everything, it is the only thing." I would not say the same thing about NPAs most of the time. (There is an exception to that rule of thumb - when NPA problem are growing, NPAs can be said to be the only thing.) Dividend growth matters. Book growth matters. EPS growth matters. CAGR (compound annual growth rate) projections matter. I believe that I have provided evidence that NPAs are the best metric candidate for being the main thing. But NPAs are not "the only thing." And that evidence can be found in the following example.
GBCI sticks out from the rest, with a price-to-book ratio of 1.91 while still having high NPAs. GBCI sells at an above sector average yield of 2.49%. But the yield is lower than the adjusted sector average that weeds out the token dividend paying banks. The P/E ratio is well above sector average. With 63.70% price appreciation, it looks like a stock that has caught momentum that carried its price much higher than its logical valuation. GBCI may be too richly valued at its current price.
On the other hand, there are other banks with P/E ratios in the 18s. And there are some reasons why GBCI belongs to that group. GBCI is one of the rare banks where loans are less than 50% of its earning assets. As GBCI turns its security portfolio, it will benefit faster than the average bank in this rising interest rate environment. Loans will tend to have longer durations compared to securities - and thus would be a portfolio that turns at a slower pace. GBCI has a superior forward CAGR projection. We are headed back to a world where banks sell a logical "yield plus CAGR" metrics. GBCI may be leading the way into that world. This investor is concerned that GBCI is too much in the lead.
Why dividend growth investors should have an interest in regional banks
This retired investor, who is living off dividends and not touching his capital, mainly invests in consumer staple stocks, MLPs and REITs. I lightly invest in BDCs for additional yield. I have a 5% weighting in regional banks. My 25% allocation to MLPs produces close to 50% of my yearly investment income. That high allocation - or lack of diversification - of "yearly investment income" concerns me. In the here and now, I want to capture the capital gains and income I can get from the high distribution CAGR stocks in the MLP sector. But over time, I want to morph into more evenly diversified sources of yearly investment income. I also want a growing income flow that beats the pace of inflation. I need to own some high dividend CAGR stocks.
Regional bank stocks help meet that need. To show why I have that perception, I have provided two spreadsheets on regional bank dividend growth. The first is from 2007 through 2013. That spreadsheet will scare you. The second is from 2000 through 2006. And there are many examples of attractive dividend growth in that spreadsheet. I have the expectation that the next seven-year period will be more like the examples in the last spreadsheet.
Bank Dividend History [based on Q3 Dividends]
|Distribution/Unit/Quarter||Percentage Distribution Growth||Av Growth|
CBSH pays both cash and share dividends. Dividend growth in 2011 was negative on a cash basis, but positive after the effect of additional shares is taken into account. FHN has paid some dividends in new shares - and those dividends were treated as cash dividends in the stats above.
Bank Dividend History [based on Q4 Dividends]
|Distribution/Unit/Quarter||Percentage Distribution Growth||Av Growth|
Summation This was the third in a series of articles on the importance - or the influence - of asset quality on valuations. Most of the better stocks, that will provide the most secure and growing flow of dividends, will also come with relatively lower yields in the here and now -- and higher price-to-earnings ratios. Many investors will only perceive the lower yield and the higher P/E, and opt for what they perceive to be the better values in higher yielding options. They fail to perceive the attributes that will produce a secure and growing flow of dividends for prolonged periods of time. They most often are not provided the spreadsheets and data that expose those favorable attributes. They are at a loss to explain the spreads one will find between the price-to-earnings ratios between what appear to be similar stocks.
Many investors will choose a stock with 1% (or 100 basis points) more yield over a stock that has 5% (or 500 basis points) higher dividend growth. That is a bad trade-off. Many investors will choose a stock with 1% more yield over a stock that has a significantly more secure dividend based on the assets that is generating the company's income. That is also a bad trade-off. The data I have provided in this and prior articles are my attempt to provide the valuation perceptions that will assist in the correction of those potential errors.