COF: Partnering with AMZN – “Digital is the New Branch Bank”
SEE LAST PAGE OF THIS REPORT FOR IMPORTANT DISCLOSURES
January 2, 2015
COF: Partnering with AMZN – “Digital is the New Branch Bank”
“We can deploy some of our most critical production workloads on AWS [Amazon Web Services] … and believe we can operate more securely on the public cloud than we can even in our own data centers. This is a game-changer for Capital One … digital is the new bank branch” Rob Alexander, CIO Capital One, November 2015
- We view Norway’s DNB Bank as a template for Capital One’s channel strategy: as a digital-first bank, DNB has achieved an industry-leading efficiency ratio of 40% (versus 55% at COF and 60%+ at European banks – see Chart). Through focusing on digital over branch channels, including eliminating branch-based manual cash-handling services in March 2015, DNK has gone beyond in-branch teller-cuts to a radical redesign of the network itself: the branch count in Norway fell by 40%+ from 244 to 137 over the seven years to 2014 and the bank sees further runway; remaining branches are focused on consultative services for “important events” in customers’ lives.
Chart: Efficiency Ratio (Expenses/Revenue – %) for DNB Bank, Norway
Source: Company Reports
- In support of its digital-first strategy, Capital One has prioritized mobile services, including an intense focus on hiring engineering talent, and its flagship app is the fullest-featured of the large banks and the only bank-branded app in the US to offer mobile payments (through Android’s host card emulation service) along with contextual customization. A differentiated app requires a differentiated IT platform, and COF has worked with AWS to pioneer bank use of the public cloud for API orchestration and on-the-fly access to storage, processing, and network resources.
- CIO Rob Alexander comments that “over the last couple of years, AWS has moved to play more of an important role in our technology strategy at Capital One … the ability to provision infrastructure on the fly is huge for us on productivity and speed to market”.
- CEO Richard Fairbank’s focus on the strategic opportunity, and comments that banks focusing on self-funding digital investment are under-appreciating it, has unsettled investors around near-run earnings implications. We expect this to resolve favorably in 2016 leading to a re-rating of the stock:
- 2016 Expenses: Having used the credit windfall of the last few years to kick-start digital investment, COF is entering a period of secular operating leverage as the efficiency ratio falls from the current 55% to the mid-40s over the next 5 years. CFO Steve Crawford comments that “we continue to see good evidence that those [digital] investments are actually going to turn into positive things on the efficiency ratio” and management guidance is for modest improvement in the efficiency ratio this year.
- 2016 Revenue: We expect positive revisions to 2017 eps from $8.50 to $8.70 given the mix-shift to card (where average balances are growing 10%+ versus 5%- for non-card) and where the revenue-margin at ~17% is more than double that for non-card. Specifically, assuming a flat card margin, consensus seems to call for a sharp decline in non-card margins to 6.6% from 7.1% in 2015 (see Chart). In December, CEO Richard Fairbank indicated that “what we are originating now [in card]… its revenue margin is pretty similar to our existing book” and we expect the mix-shift benefit, along with a lift from home-equity run-off, to overcome firm-wide headwinds (e.g. drag from payment protection and compression in auto and commercial)
Chart: COF – Segment Revenue Margins (Implied Consensus)
Source: Company Reports
- With a return to positive operating leverage in 2016, we expect COF to re-rate to a 1.25x multiple of tangible book value, conservatively consistent with a tangible return on equity of 12-12.5% generating a 2016 price-target of $85, or 18% upside from today’s price of $72, given our forecast for year-end tangible book of $68.
We are buyers of COF with a price-target of $85, or 18% upside from today’s price of $72, representing 1.25x our forecast for year-end tangible book of $68 and conservatively consistent with a tangible ROE of 12-12.5%; we include our model in the Appendix and note the assumption for tangible book value assumes the firm maintains the current CET1 ratio of just over 12% (on a standard basis corresponding to 8% under the advanced approach). The catalyst for re-rating is a return to positive operating leverage as digital investments, kick-started by the credit windfall of the last few years, become increasingly self-funding given COF’s pioneering use of the public cloud in partnership with AWS. More generally, as the customer shift to mobile channels continues for the industry (see Exhibit 1), we expect a digital-first strategy at COF (where mobile internet accounts for more than twice the transactions of fixed internet) to bring the efficiency ratio down over the next 5 years from the current 55% towards the 40% achieved in Norway by DNB Bank.
Exhibit 1: Distribution Shifts in Banking – % of Households Using Each Channel
Source: Capital One Presentation – May 2014
Valuation and Investor Concerns
COF can be a difficult stock for investors to own. The firm does not target growth but rather is opportunistic – originating aggressively when the market offers assets that meet over-the-life hurdle rates, and backing away when competitive action erodes these returns. For example, management has commented that beginning mid-2014 a window of opportunity opened in US credit cards and, after years of industry-lagging growth (accentuated by run-off of the high-balance revolver portfolio which does not meet Capital One’s resiliency requirements because of relatively high-lines and tight-pricing), the firm is now taking share. The improved growth relative to DFS which, since the financial crisis, has built its portfolio more around high-balance revolvers but is now becoming less aggressive in the segment, is particularly striking – Exhibit 2.
Exhibit 2: Year-on-Year Growth in Card Loans (Domestic only for COF)
While economically rational to first-order (i.e. excluding cost-of-capital effects) adherence to positive-NPV investing can be unfriendly to near-run earnings; for example, the marketing budget increased 15% in 2015H2 versus the year-ago period as COF responded to opportunity with more aggressive direct mail and online offers and, because of the seasoning of new loans, the firm is guiding its card loss ratio for 2016 to 4.0% versus below 3.4% likely for FY2015. Furthermore, even excluding marketing expenses, operating expenses have increased 8% year-to-date (versus a 4.5% increase in revenue) as Capital One has invested in digital initiatives; the result is investor concern around expense discipline despite management assurances that the efficiency ratio will improve modestly in 2016. Finally, there are concerns around the taxi-medallion and oil-services portfolios (each about $1bn) and a sense that, given its disproportionate online operations, Capital One will not benefit as meaningfully in an up-rate environment as banks with a relatively larger branch-footprint and presumed lower repricing “betas” on deposits.
The result is that the stock has underperformed and trades cheap relative to its return-on-equity – Exhibit 3. We see this as opportunity and expect the shares to rally 15%+ in 2016 to a price-target of $85 based on a tangible price/book ratio of 1.25x that, in turn, is conservatively consistent with the expected tangible return-on-equity of 12.5%. We include our model in the appendix and note that we have made allowance for write-offs and reserve-builds over the next few quarters for the energy and taxi-medallion portfolios, and incorporated management’s 2016 guidance for a 4% loss ratio in the card business and modest improvement in the firm-wide efficiency ratio from the expected 55% for full-year 2015. More broadly, as the new card vintages continue to season and ambient credit conditions begin to normalize, we model a 17-18% increase in net credit losses in each of 2016 and 2017 compared to an expected 8% increase in 2015.
Exhibit 3: Valuation of COF relative to US Banks
Source: Capital IQ, SSR Analysis
Return to Positive Operating Leverage as a Valuation Catalyst
The catalyst for re-rating of the stock will be a return to positive operating leverage with expense growth held below 5% in 2016 and 2017 (versus 6%+ in 2015) even as revenue growth increases to 6-8% (versus 4.5% in 2015). We note the change in language at Capital One with CFO Steve Crawford commenting in early December that “as credit becomes more of a challenge in the income statement … the way you maintain your return on capital is by having operating leverage” and adding that “we continue to see good evidence that those [digital] investments are actually going to turn into positive things on the efficiency ratio”. While it may have as much to do with roles as resourcing, the financial context is a reassuring complement to the more strategic comments from CEO Richard Fairbank that “every single part of banking … is going to be dramatically changed by digital” and that self-funding digital investment “massively understates the magnitude of what is going on”.
Concrete evidence that the digital strategy is beginning to generate cost-savings, beyond shifting customer activity to the mobile channel where variable costs are pennies/transaction versus nearer a dollar for branch-teller support, was provided in a presentation given by CIO Rob Alexander at the AWS re:Invent conference last November. The punchline was that, as COF moves more operations to the public cloud, it plans to reduce its data center footprint from the current eight in 2014 to five by end-2016 to three by 2018. We cannot quantify the associated saving but note that Norway’s DNB Bank has achieved an industry-leading efficiency ratio of 40% (versus an average of 60%+ at European peers) through focusing on digital channels and reducing branches in Norway by 40%+ from 244 to 137 over the seven years to 2014. Given COF is already the largest digital bank in the US and is committed to a digital-first strategy, we believe DNB represents a template and expect the efficiency ratio to fall from the current 55% to the mid-40s over the next 5 years.
Appendix: COF Model
©2016, SSR LLC, 1055 Washington Blvd, Stamford, CT 06901. All rights reserved. The information contained in this report has been obtained from sources believed to be reliable, and its accuracy and completeness is not guaranteed. No representation or warranty, express or implied, is made as to the fairness, accuracy, completeness or correctness of the information and opinions contained herein. The views and other information provided are subject to change without notice. This report is issued without regard to the specific investment objectives, financial situation or particular needs of any specific recipient and is not construed as a solicitation or an offer to buy or sell any securities or related financial instruments. Past performance is not necessarily a guide to future results. The analyst principally responsible for the preparation of this research or a member of the analyst’s household holds a long equity position in the following stocks: JPM, BAC, WFC, and GS.
In Feb 2014, JPM reported the variable cost for a teller-assisted deposit at 65 cents/transaction assuming a service time of one-minute. ↑