With earnings season right around the corner, I thought I'd share a little preview of an analysis I will be updating once the big banks file their 10Ks (typically occurs a few weeks after earnings are announced). The figures below are based on last year's results, but the adjustments are worth considering now while earnings are being announced. There are plenty of valuation models out there, but I consider the analysis below as a basic starting point because it's simple and intuitive, but more importantly, it helps better assess how much capital a bank has and what they are earning with it.
Sifting Through The Noise
Bank earnings have been hard to decipher due to so many recurring 'one-time' items. Investors want to see a clear trend line, and adjusted earnings have helped some with that, but in reality the raw or true economic performance of banks like Bank of America (NYSE:BAC) and JPMorgan (NYSE:JPM) have been hidden by the massive reset that happened during the financial crisis. And, while the earnings damage is apparent, that reset also affected comparative tools like book value and tangible book value. For some banks, this means that the capital available to generate future earnings may truly be less than reported, but that returns are better than they appear, and for others the opposite is true.
In this article I've identified and adjusted the earnings and the balance sheet of Bank of America for the major one-time items and accounts that I think are clouding the company's results and profile. There are several other adjustments that could be made (appears some won't have to be made anymore), but the ones chosen are the most material (80/20 principle) and I think the end results are fair.
Noting that the adjustments were similar to ones that would be made for an EVA based analysis, the numbers below are essentially a modified (or at the least 'my') version that I think is more consistent with how someone owning a bank should look at earnings and returns - which is after required returns have been charged. This, however, is a theoretical concept so I don't expect everyone to adopt it, but I think it's helpful for comparisons between banks (and one's past performance) due to the fact that it charges each company for the amount of capital they manage for shareholders. And, as I will explain, I don't think this measure is inconsistent with internal benchmarks BAC uses to evaluate its own performance. With that said, let's get into the calculations:
From above, net income is based on numbers I pulled from 10Ks. The adjustments are as follows:
1. Provision expenses are nice and all but they are non-cash and supposed to reflect the quality of future income streams. Substituting this projection with actual net charge-offs switches the focus from future to current results, and because of this these numbers are more in-line with the bank's actual and ongoing performance. As you can see, provision expenses have declined and are lower than net charge-offs in all periods that I've highlighted. The bank's future outlook is better than the current one, and BAC is still paying for the past with the charge-offs. I don't expect for this trend to continue indefinitely, and when it does end the adjusted income calculation will be positively impacted.
2. The adjustment to taxes is self explanatory. We are looking for what the business actually paid. And, as you can see, even with a large tax asset (covered below) the bank paid more cash than they expensed on the income statement in 3 of the last 5 years.
3. The adjustments for intangibles (amortization and goodwill impairments) are helping us sort out non-cash charges. These two adjustments increased adjusted net income in each of the last 5 years.
This again is fairly straight forward, we are trying to place a value on the amount of shareholder capital that the bank is managing. The adjustments are as follows:
1. The allowance account is a theoretical box of future losses, it isn't an actual subtraction from assets. Bank of America has these assets and they are trying to make money with them. The total balance is added back for this adjustment, and this is consistent with the adjustment we made to earnings for provision expenses.
2+3. Again, we are looking for tangible assets the bank can use to grow earnings. These two accounts have value, but not for the purpose of this analysis, and so they are subtracted here - also consistent with the adjustments we made to earnings.
4. This is probably the most controversial adjustment I've made up to this point, but it's important to note that a Net Deferred Tax Asset is an intangible receivable that is the result of a loss - the worst type of asset (as far as its origin and ability to generate income is concerned). I have wrote about this account a few times in the past (here, here, and here). I may not be able to convince you of this now, but this is an important adjustment for this analysis because it truly evens the playing field for when we expand to a look at the bank in comparison to its peers. And, while it is a negative adjustments, remember this move increases the rate of returns that we will calculate for BAC. The bank's results truly are handicapped without this adjustment.
In addition, the Net Deferred Tax Asset is a catch-all account - a filler if you will. For instance, the bank has tax liabilities, this is just the net balance. And, as you will see when we get to Wells Fargo (NYSE:WFC)(after 10Ks are filed), a net deferred Tax Liability is actually an unpaid-hidden asset. Because of this, WFC's net deferred tax liability will be added to its capital account, which in-turn will increase the amount shareholders should expect the bank to earn. Plus, even though this account is listed as a liability on WFC's balance sheet, the bank still benefits from cash-tax payments that are less than the bank's accounting-tax expense. Think of this and the adjustment to earnings for actual taxes paid as the result of the bank's ability to profitability manage the cards they have to grow earnings - we are trying to calculate/judge/analyze the results.
I mentioned earlier that there was a correlation between this analysis and the bank's internal performance measures, and these capital adjustments are what I was referring to. As you can see below (from Q4 2014 earnings announcement), BAC's allocated capital adds up to just $129.5 billion. This is not far off from my calculation of $136.4 billion - but the bank's calculation is also based on average balances and not the period-end balances that I used.
I didn't talk to and I don't know Seeking Alpha Contributor IP Banking Research, but he/she/they recently tried to tie the difference between tangible shareholder equity ($151.7 billion at end of 2014, difference = $22.2 billion) and allocated capital to the far right column titled 'all other.' IP Banking came to this conclusion after adding up each segment's report, but one titled 'all other' was missing and assigned the difference. However, I found this missing piece and you can see in the presentation below that it is zero. Keep in mind though, IP Banking was using this calculation to determine another aspect of an investment in BAC that we are not approaching in this article. I'm not trying to take away from that article, but I think for our purposes the table below gives some credit to this analysis in that allocated capital is in the neighborhood of what our adjusted capital calculation produced.
Charge for Required Shareholder Returns
For a peer comparison to work, it's important to even the playing field so that accounting adjustments don't negatively affect the outcome. For example, an increase in the allowance account lowers shareholder equity and increases future earnings and returns. Our adjustments will neutralize this, and allow charge-offs to dictate earnings quality, but from a reported earnings perspective a bank (BANK A) with a bloated allowance account will 1) look safer, 2) produce a higher ROE, and 3) have more assets hidden away to continue to produce higher returns than the exact same company (BANK B) with a smaller allowance account. Again, from this perspective BANK A will come away with a favorable assessment, but in reality it was producing the same amount of income with the same amount of assets (though in the past the opposite would be true - lower earnings/returns in allowance build-up period). The above EVA analysis neutralizes the different accounting treatments and values both companies equally by calculating the same figure for adjusted capital and earnings - which brings us to required capital returns.
This is where someone usually comes in and talks about weighted average capital expenses - but I like to think of capital charges as my own personal hurdle rate. An owner of shares can theoretically require whatever rate they want from the capital they lend out to managers, and for the sake of consistency the calculations below display a range of rates so that you can see how your own required return pans out. However, in order to stay within a reasonable range of diehard WACC calculators, I'm going to be using 9.5% for the capital charge in this analysis (and those that follow). This is approximately the risk free rate (3% based on a 30-year note) plus a 6.5% risk premium; I've also highlighted in the table the implied rate (shaded green)- which is the MAX rate you could charge before the bank's EVA turns negative.
Mark Bern, CFA: The fair value for any stock will vary based upon the expected return of each investor. Therefore, my fair value will be what I am willing to pay for expected future returns from the stock which may differ from what other investors are willing to pay. Thus, my fair value may be different from yours.
From the results above it's obvious that Bank of America is still trying to find its way out of the woods. Based on our 9.5% charge to capital, the bank destroyed value in each of the last 5-years. But, this is may not be representative of the bank's core earnings potential.
I wanted to separate this adjustment from the others because it is fairly unique to a thesis backing an investment in Bank of America. The company's litigation charges are large and have yet to disappear completely, but the hope is that they will eventually, and that the following adjustment is more true to the value BAC can create once shareholder income is free from this expensive legal tax.
And as you can see, the view is a lot brighter on the other side of legal fees. In the last two years BAC earned in excess of our 9.5% charge - even greedy investors requiring a 13-14% return would have been happy with the results. But, keep in mind, these returns were helped out by our (negative) adjustments to the capital account (and earnings for litigation expenses). A smaller account = a smaller charge.
It took a long time to get to this point but below is a side-by-side look at both reported and our adjusted figures (2014-2010; from left to right). There's a lot in here for you to look at and think about, but the biggest take away is that the adjusted returns are higher than reported, and that the adjusted capital (or adjusted book value) the bank has to earn with is lower than reported. In addition, a surprise from this analysis was the fairly stable results this model produced. I think more weight should be given to the earnings we calculated after adding back litigation expenses, because it is more consistent and representative of 'core earnings,' but you again can be the judge of that.
Value and Where Do We Go From Here?
Like I said at the beginning, this is a preview of what I will update and report for each of the following banks (in addition to BAC) once they have filed (after announcements) their annual reports: JPM, WFC, Citi (NYSE:C), and U.S. Bancorp (NYSE:USB). As a tool this provides a raw look at how the bank's valuation has changed over the years, but I think having an analysis for each company will be even more revealing.
In addition, there is one more step that ties our EVA calculation to Buffett's principle of market value added. Again this is a theoretical multiple, but the idea is that the capital we calculated for the bank is worth our adjusted value if and only if EVA is positive - and excluding legal fees it was, and amounted to $0.53 per share in 2014.
Using this, I've come up with the following 'MVA' multiple:
Closing Price: $17.89 (year-end 2014)
Less Adjusted Capital: $12.89
=MVA Per Share: $5.00
MVA Multiple = MVA Per Share / EVA per share
$5.00/$0.53 = 9.43X
Note: 9.43X is 22% less than last year's Adjusted MVA multiple. This alone isn't necessarily a signal to buy, but shares are down from the closing price used in the calculation, and if EVA improves the adjusted premium would have fallen again in 2015.
In theory, Market Value Added is the present value of discounted future EVA calculations and forecasts, so another model could be drawn, but for this first round of peer assessments (to avoid more assumptions) I'm going to stick with the trailing multiple.
Since this was the first article in the series I thank you for allowing me to spend more time to explain the thinking behind my math and the benefits of calculating EVA (this article will be used as a reference in the future). Understanding a bank's earning report is sort of like learning a different language, and this tool is meant to help level the playing field.
I hope everyone has a great 2016. This isn't an all-inclusive analysis, but I think the adjustments are basic and material enough to roll with going forward. However, if you think of any others that should be added I'd love to hear about them in the comment section below. Good luck!
Disclosure: I am/we are long USB.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.