US Banking: Mobile Banking Will Increase Scale Economies
SEE LAST PAGE OF THIS REPORT Howard Mason
FOR IMPORTANT DISCLOSURES 203.901.1635
April 23, 2014
US Banking: Mobile Banking Will Increase Scale Economies
- Despite a strong rally off October 2011 lows, large-bank valuations remain attractive with WFC, JPM, and BAC trading at price-to-book multiples of 60%, 40% and 30% relative to the S&P500 – at or below the low-end of pre-crisis ranges. Valuations have responded to lower normalized returns-on-equity (that we estimate at 13-15% for large banks vs. results of over 20% in some pre-crisis years) but do not reflect either that current returns (of 4%, 9%, and 13% for BAC, JPM, and WFC respectively) are below the new-normal or the re-rating that typically accompanies de-risking.
- The scope for returns to improve arises from a normalizing yield curve (see, for example, our note of March 6th titled “The Dynamics of Net Interest Margins at Selected US Banks”) and lower costs-to-serve as customers shift basic-transaction activity from branches to mobile, online, and ATM channels. This channel migration is at the tipping point; for example, BAC reported that 10% of deposit transactions were executed “remotely” through mobile in 2014Q1 (up from 6% in 2013Q1 and none in 2012Q3) rather than via teller.
- The expense lever is important to bank profitability with the industry generating the same return-on-assets today as coming out of the recession of the early 1990s. This is despite net interest margins being over 100 basis points lower; the reason is that the ratio of expense to average assets is also 100 basis points lower at 2.9%, and we expect this to fall to 2.5% over the next 5-years. Efficiency gains, historically driven by industry consolidation (with large banks operating meaningfully more efficiently than small banks), will increasingly be driven by optimization of the delivery network.
- Our thesis is that the shift to mobile will disproportionately benefit the three largest US depositary institutions because they have the scale to make the necessary marketing, infrastructure, and data-strategy investments and the scope to best optimize across the branch, online, mobile, and ATM channels. More regional institutions will outsource to infrastructure providers such as FIS and FISV (generating opportunity particularly for FIS given its mobile capabilities) but white-labeling makes it more difficult to create the differentiated offerings that large-banks are likely to evolve from proprietary infrastructure such as ChaseNet and even shared infrastructure such as clearXchange (a P2P mobile platform jointly owned by JPM, BAC, and WFC).
- While it may seem counter-intuitive for the benefits of mobile to accrue to banks with the largest branch networks, we believe branches will remain a critical sales channel for meaningful consumer segments and hence important to the scale economies that support investment in mobile infrastructure. WFC CEO John Stumpf reminds us: “Half of the US population and half of businesses in America reside within two miles of our stores. Just think of that. And while people, a lot of times join us online, they pick the location, they pick the provider that they are familiar with or that’s close by them because they still use the store even for the millenials who are very technologically advanced and savvy”.
- Furthermore, as customers continue to migrate to mobile, online, and ATM channels, the largest banks have the greatest opportunity to invest in the associated delivery platforms to reduce cost-to-serve. JPM, for example, reduced branch staffing by 8% in 2013 and an expected 20% over the next two years even though it is not planning a net reduction in branches; it comments that the variable cost of a remote deposit transaction is 3 cents vs. 65 cents for a teller-assisted deposit.
- More typically, banks have reduced branch-counts and the trend is accelerating: industry-wide branches were down 2% in 2013 vs. 0.8% in 2012 and 0.4% in 2011 (see Exhibit). BAC has been particularly aggressive in branch cuts with CEO Brian Moynihan highlighting the expense-save opportunity: “The strategy in our retail segment has been to lower the cost of service, so we could improve our customer experience. We do that by continuing to optimize our delivery network of all types in response to customer behavior changes. Banking centers and basic teller transactions continue to decline as customer move their business to mobile and online transactions … People effectively carry a branch in their pocket. This, coupled with other measures, allowed us to reduce cost in our consumer banking business by 4% from last year’s first quarter.”
Exhibit: Branch-Count by Bank (adjusted for completed and, in 2013, announced merger activity and branch transfers)
Large Bank Valuations Remain Attractive
Despite strong rallies from the lows of October 2011 (BAC up 140%, WFC up 75%, and JPM up 60%), the three largest US depositary institutions are trading at the low-end of historical ranges relative to the S&P500 (see Exhibit 1). For example, the 0.78x price-to-book value of BAC is 30% of the 2.6x for the S&P500 index, and we expect this to move into the range of JPM and WFC as the legacy mortgage issues from the financial crisis are resolved. At 40% and 60% respectively, the relative price-to-book values for JPM and WFC are themselves at the low-end of historical ranges and we see the potential for meaningful re-rating to at least the middle of the ranges as uncertainty around regulatory requirements and legal claims declines.
Exhibit 1: Price-to-Book Values Relative to S&P500
We acknowledge that regulatory de-risking of the financial system, including higher capital requirements, has reduced returns-on-equity, and see 13-15% as the “new normal” ranges for BAC, JPM, and WFC. While well below the 20% or more results generated in some years before the financial crisis, this new normal is above current results of 5% and 9% for BAC and JPM respectively, and suggests upside for WFC as well (see Exhibit 2). We do not believe these lower-than-historical returns will permanently suppress relative valuations since regulatory de-risking means they are likely to be more stable with lower risk-of-ruin, and hence command higher multiples.
Exhibit 2: Historical Returns on Equity for BAC, JPM, and WFC
Catalyst for Re-Rating
The catalyst for re-rating will be improved returns as: a normalizing yield curve lifts net interest margins (since bank deposit franchises are dramatically under-earning in today’s low-rate context); the resumption of loan growth as the economy recovers and there is a lesser headwind from the run-off of legacy portfolios; and improving efficiency as the migration of customers to mobile, online, and ATM channels from bank branches reduces the cost-to-serve for basic transactions.
Having discussed the likely lift to net interest margins (NIM) from a normalizing yield curve (see, for example, the note of March 6th titled “The Dynamics of Net Interest Margins at Selected US Banks”), we turn attention in this note to the potential for efficiency gains. As context, we note that despite the difficult rate environment, the average ROA for US commercial banks is at the same level as coming out of the recession of the early-1990s (see Exhibit 3). This is despite the fact that bank net interest margins are over 100 basis points lower and fees as a percentage of average assets are 20 basis points (declining that much since 2000 in large part owing to regulation of overdraft charges and debit interchange).
Exhibit 3: Twenty Years of Financial Results for US Commercial Banks
The net-interest-margin gap has been filled by efficiency gains, with non-interest expense (NIE) now 2.9% of average assets versus 3.9% in 1993 and, to a much lesser extent, reserve release. Efficiency gains are most pronounced at large banks (assets of $10-50bn), where NIE has declined to 2.7% of average assets from 3.3% in 2006, versus small banks (assets of $300-500mm), where NIE has declined to 3.0% of average assets from 3.4% in 2006 (see Appendix A), and efficiency across all banks has improved because of consolidation; the top-11 banks by deposits now control 54% of deposits (see Exhibit 4) up from 28% in 1998.
Exhibit 4: Deposit and Branch Share for Top-11 US Commercial Banks by Deposits
Over the next 5 years, we expect meaningful improvement in bank profitability, particularly at large banks, as the yield curve normalizes; as banks develop new fee streams through leveraging deposit and mortgage relationships to gain share of the investment assets of the affluent and mass-affluent; and as efficiency continues to improve (dropping NIE to 2.5% of average assets from the current 2.9%). The driver of efficiency gains is shifting from industry consolidation (although we expect larger banks to gain share organically) to cost-to-serve decline with customer migrations to online, mobile, and ATM channels. The efficiency impact of the shifting channel-mix is already meaningful, and we are still in the early stages.
BAC, for example, reported that in 2014Q1 more than 10% of all deposit transactions were through mobile applications (vs. 6% in the year-ago quarter and nothing before 2012Q3), and this contributed to year-on-year cost-reductions of 4% in the consumer banking business with net branch reductions of 5% completed in the year. While JPM is not reducing branches (and, in fact, increased its branch count primarily to build out the footprint in California and Florida which together accounted for two-thirds of new builds), the firm is reducing expenses through staff cuts; branch staff fell 8% in 2013 and is expected to fall a further 20% over the next two years. As the firm invests in mobile and branch self-service technologies, the average number of tellers in a new branch has fallen to 2 from the historical 4.
Appendix A: Comparing Efficiency for Large and Small US Banks
Exhibit A1: US Commercial Banks with $10-50bn in Assets
Exhibit A2: US Commercial Banks with $300-500mm in Assets