Capital One Financial Stock: Mature Bank With One Special Property (NYSE:COF) – Seeking Alpha

Capital One Financial Stock: Mature Bank With One Special Property (NYSE:COF) – Seeking Alpha

Capital One Bank

ProArtWork/iStock Unreleased via Getty Images

ProArtWork/iStock Unreleased via Getty Images
Recently, I published an article on Synchrony Financial (SYF). Since I liked its business model, but I found it at the time trading close to its fair value, I extended my search to its peers in search of potentially undervalued efficient US banks. Capital One Financial Corporation (NYSE:COF), or Capital One, seemed like the most likely candidate since, as Synchrony Financial, it has a relatively simple portfolio structure. Analysis has shown that Capital One is in a mature phase of its development, as its portfolio grows at the speed of the overall markets in which it operates. Its portfolio is of high quality, and it has become obvious that high loan loss reserves, formed in the aftermath of March 2020, can be safely released now. This is an ongoing process, that is resulting in record high profitability result being reported, even in low interest rate environment. In turn, release of loan loss allowances will allow Capital One to finance its portfolio growth using lower proportion of generated earnings. With this in mind, Capital One seems to be undervalued at current price level, and investing now could yield above average market return in the long run. Hence, I am bullish on Capital One.
Since its original founding in 1988, and subsequent spin off and public listing in 1994, Capital One has grown to be the 11th largest bank in the USA by asset size according to the Q3 2021 data.
Banks by asset size
(Source: Federal Reserve)
Originally, Capital One was the equivalent of today’s FinTechs. It came to the saturated banking market offering only credit cards, but in a way different to their competitors. Using customer level data and introducing new technologies in portfolio management and their marketing activities, they quickly managed to expand their portfolio. As of 1998, Capital One also started expanding into auto financing, and today they are one of the largest auto finance companies in the USA. Up until 2012, rapid portfolio and assets under management growth was driven by organic and growth by acquisition. Since then, Capital One has entered a more mature phase in its corporate life.
COF assets (Source: Author’s Calculations)
Today, its portfolio is split between credit card exposures, consumer financing and commercial banking. Up until the middle of 2018, they had a minor portfolio of home loans, however, they have made a decision to divest from this business due to low profit margins that could be achieved from it. In light of today’s low interest rate, this was a good strategic move.
Loan portfolio structure over time
(Source: Author’s Calculations)
As I mentioned, Capital One is in a mature phase of its corporate lifecycle. The chart below tries to quantify this by comparing the cumulative segment growth rates with the same segment overall growth for the whole of the USA banking sector.
Portfolio segments cumulative growth vs total USA segment values
As it can be seen, for all three major segments of its portfolio, Capital One has followed the development of the overall lending market in the USA.
This is even more visible when the development of Capital One loan portfolio is compared to its respective segment’s development in the total USA banking sector. In addition, comparison is made with hypothetical total market portfolio, whereby in each snapshot, relative weights of Capital One segments were applied to total USA banking sector segment development (the USA replica portfolio).
USA replica vs COF portfolio
(Source: Author’s Calculations)
As the figure shows, Capital One strategic decisions on product mix had little influence on the portfolio growth in recent times. In most cases, Capital One follows the overall loan market development.
Measured on the last four quarters, five-year compounded growth rate for the USA replica portfolio was between 3.40% and 4.24%. For the same period, Capital One compounded growth rates were between 3.36% and 4.91%. In my forecasts, I have assumed that similar growth rates will be maintained in next 5 years, with gradual decrease after that to the levels of long-term inflation rate.
One of the more noticeable recent developments in the Capital One results is that Q3 2021 brought record high reported return on equity (ROE), even though low interest rates are prevailing for more than 12 months being reported in trailing numbers. As the graph shows, the latest reported ROE was above 20%, which is high above its historical levels.

COF return on equity
Data by YCharts

This high ROE value was mainly caused by the release of provisions booked in the wake of March 2020 and expectation of increased levels of customer defaults. As this failed to materialize, Capital One initiated loss allowance releases. Compared to the historical levels of loss allowances to total gross loans, it could be expected that additional USD 1.5 billion could be released before historical levels are reached again.

COF loans
Data by YCharts

Viewed on the level of individual segments, loan loss allowance dynamics are the following:
Allowance coverage ratios by segment
(Source: Q3 2021 Earnings presentation)
In order to understand the earnings power of Capital One, it is important to take into consideration properties of its portfolio development, impact loss allowance release had so far on its past and future profitability and, finally, how the capital funding needs will be impacted by the previous two. This is covered in the next sections.
To calculate the fair value of COF and convert its narrative into numbers, I used a three-stage discounted free cash flow model. Stages I assumed were the following:
The company’s free cash flows to equity (FCFE) were valued as a going concern and discounted back at an estimated cost of equity. The estimated value of preferred shares and outstanding options was subtracted from the calculated present value of FCFE to arrive at a total value of equity for common shareholders. Total equity value was then divided by the current total number of shares outstanding to obtain a fair value per share.
The following are the assumptions I used in the intrinsic value calculation for them:
I assumed that the risk-adjusted assets will grow at a slightly smaller rate than the historical growth rate of volume of aggregated auto, credit card and commercial loans in the USA. For the first 5 years in my forecast, I used 3.5%. This growth I am attributing only to the overall inertia of the lending market, and it is completely independent of COF’s ability to attract additional extraordinary business. After year 5, I assume that over the next 5 years, the growth rate will slowly decrease to 2%, which is close to the last 10-year inflation rate in the USA. The last assumption will also be subject to simulations, in order to measure the sensitivity of my forecasts.
COF’s latest reported return on equity was 21.9%. However, this result, when compared to its historical levels, was achieved solely based on the loss allowance provision release and has not been generated from the increased volume of loans or extraordinary interest income. Using the post-tax amounts, a total of USD 3,444 million was generated thought loan loss allowance release in 12 months leading to Q3 2021. In effect, this reduces net income after tax from reported USD 12,531 million to USD 9,087 million, and thus the ROE to 15.77%. One should also note that in Q3 2021, interest income and loan book size are on the level of Q4 2017, when ROE was 4.20%. I performed similar analysis in my Synchrony Financial valuation. However, COF does show some better developments. Contrary to SYF, average share of loss allowances in last 5 years, compared to total gross loans, are still below current levels by 0.6%. This in turn would allow COF to release additional USD 1,556 million in forthcoming period. This is a non-cash event, however it will increase COF regulatory capital and thus lending capacity in the future. In my valuation, I assumed that the release would occur over the next three years. Effectively, increasing their after tax earnings by USD 389 million for each of next three years. This would allow COF to report ROE above their historical average values even in low interest rate environment for few years more. However, as I mentioned, I recalculated COF’s ROE to exclude this effect, and I assumed that normalized ROE will gradually revert to its historical mean value of around 10% in next five years.
Reinvestment needs are measured through regulatory capital requirements, or the amount of Tier 1 capital banks needs to hold against their risk-weighted assets in order to be able to withstand financial distress (Tier 1 ratio).
Portfolio quality and capital requirements
(Source: Author’s Calculations)
As the figure above shows, COF risk-weighted assets (RWA) develop in a similar proportion to total loans, indicating a stable risk profile of its portfolio over time. Assuming that, in the future, RWA will develop according to the forecasted loan portfolio growth, one can calculate the amount of cash needed to fund the capital needs through the expected Tier 1 Capital ratio. Since banking is a mature industry, I base my expectations on having a look at the current state of the USA banking sector using the data on ROE and Tier 1 Capital distributions from FFIEC (joint banking supervisory body in the USA).US banks average as of JUN 2021
(Source: Author’s Calculations)
Based on its normalized ROE, COF is better than 75% of the US banks while based on its long-run average risk coverage, it is closer to the median bank. Also, looking at the data from FRED, this relationship seems to be holding over time.
COF vs USA banking sector Tier 1 capital ratio development
(Source: Author’s Calculations and FRED data)
Assuming that this property will hold even in the future, in my forecasts, I assumed that over the next 10 years, the COF Tier 1 ratio will slowly approach 15.17% while it will still maintain the profitability of 75th percentile of the current banking distribution. I am basing my forecasts of the future on current values, as banking as an industry is a mature and saturated industry where surprises are rare and few. In the long term, COF will have to come closer to the average profitability results, or in its case, the 75th percentile.
I estimated the cost of equity from the market-implied equity risk premium for the USA. However, contrary to my common approach of using average market beta to derive the cost of equity, in the case of COF, I used a more specific method. Based on five years of data of daily prices for SP500 and COF, I created two separate time series models. These models were then used to forecast 10,000 simulated paths for a broad market index and COF. Using the risk-free rate of 1.65% (10-year T-bond rate as of 21st October) and these simulated price developments, the CAPM model was applied to obtain the distribution of 10,000 simulated beta values.
COF chart
(Source: Author’s Calculations)
Finally, I used the median beta value to calculate the following cost of equity for COF (as of 31/12/2021):
COF cost of equity
For the value of the terminal cash flow, the following assumptions we used:
COF return on equity
(Source: Author’s Calculations)
I allowed for some difference between the cost of equity and return on equity after year 10, assuming that COF will still have a small operational advantage compared to other banks. My assumption for the risk-free rate of 2.5% and portfolio growth of 2% might seem low from a historical perspective. But keep in mind that since I have also included the risk-free rate in my assumption for cost of equity after year 10, by deriving the implied cost of equity for a median bank (9.51%) and adding the difference between current and terminal year risk-free rate to it, that makes this valuation less sensitive to future changes in risk-free rate levels.
Values projected and discounted by these assumptions are as follows:
COF valuation
(Source: Author’s Calculations)
Although it will become relatively less profitable in the future, I expect its portfolio to continue to grow with the overall credit card, auto loan and consumer loan lending market. This growth will slow down over time until it reaches the long-term US inflation rate of 2%. I assumed that the risk profile of COF will not change over time, and that it will be a medium risk bank as it is now. This in turn means that that its Tier 1 capital ratio should be close to that of the median bank in the USA today.
COF valuation(Source: Author’s Calculations)
Over time, as the period of extremely low interest rates passes, the cost of equity will also increase. I did assume that the cost of equity in year 10 will be equal to that of the median US bank, increased for the difference between today’s and year 10’s risk-free rate. Since the terminal value of the risk-free rate will impact the cost of equity value and terminal value, this value will be subjected to simulations as a function of the inflation rate, together with terminal value ROE.
COF valuation
(Source: Author’s Calculations)
From the calculated value of equity, I subtracted the value of preferred equity and outstanding options issued as compensation. Value of preferred equity was calculated using perpetuity formula and by dividing trailing 12-month preferred dividend outflows with current cost of equity. For value of options, fair value was obtained using the Black-Scholes model formula and number of options outstanding as of last available annual report (these numbers are not provided in quarterly statements).
To mitigate the effect of potential calculation error and error in my assumptions, 10,000 simulations were run. This allowed for variability around assumptions of inflation (assuming normal distribution in the range between 1.5% and 2.5%) and ROE after year 10 (using normal distribution around assumed value 10.86% with 1% standard deviation).
Calculated value distribution
(Source: Author’s Calculations)
This results in a median present value per one share of USD 179.34, with the 25th percentile value being USD 170.53, which I would usually use for my estimate of the fair value.
COF measurements
(Source: Author’s Calculations)
Capital One stock price has had a remarkable recovery path since March 2020. Similar to other mature banks, it has not produced any extraordinary results. Just as it had done before the crisis, it has followed the overall loan market in its good and in its bad time, and it will most likely continue to do so. However, in banking, following the overall market in portfolio development, as long as quality of portfolio is high can be a remarkable property. Capital One has done exactly that. Throughout its history, it managed to assess the quality of its portfolio in a way that allows it to operate with low allowances for losses. In the wake of the crisis, these were artificially increased, however their years of prudent lending have repaid and will now enable them to continue servicing their clients as efficient as ever. I believe Capital One to be undervalued. Given its historical price volatility, I am willing to include it in my portfolio if the value of its common equity is below USD 60.3 billion, or if according to the current share count price gets below USD 142 per share.
What is your view on the future development of COF? What do you see as the most likely long-run development of the interest rate environment in the USA? Express your opinion, provide alternative assumptions and different opinions in the comment section below and I will rerun my analysis using your inputs.
This article was written by
Disclosure: I/we have no stock, option or similar derivative position in any of the companies mentioned, and no plans to initiate any such positions within the next 72 hours. 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.


Share This