BNY Mellon Isn't the Underdog It Seems

Custody banks have suffered in recent years as low interest rates and clients' risk aversion have weighed on revenue. Banks have responded by cutting costs--though some more than others--but shareholder returns have remained stubbornly subpar. Despite this, we think the wide economic moats of custody banks are firmly intact, and we argue that returns will improve significantly as economic conditions normalize. We expect  Bank of New York Mellon (BK) to benefit disproportionately from improving conditions, and we see it as the most undervalued of the custody banks. Moreover, we think recent criticisms regarding a bloated cost structure and slow growth and calls to break up the bank are overdone and built on misunderstandings of industry dynamics.

Custody banks, also known as trust banks, take custody of and provide safekeeping for assets belonging to clients like mutual funds, pension funds, insurance companies, alternative asset managers, and wealthy individuals. Custody banks typically also administer these assets, providing services such as arranging and recording settlement of any asset sales or purchases, collecting and recording income from the assets, such as dividends or interest payments, and administering corporate actions, such as stock splits and company mergers. These client assets are typically known as assets under custody or assets under administration, depending on the specific arrangement, or a combination.

While traditional custodial activities are the backbone of custody banking, they also serve as a springboard for the banks to provide other value-added services, such as securities lending, foreign exchange, and risk-management middle-office activities. Traditional custody activities are fairly routine, and all custody banks are investing heavily in building out their portfolios of value-added activities.

BNY Mellon,  Northern Trust (NTRS), and  State Street (STT) are often lumped together as the trust banks, but their business models are actually very diverse. Of the custody banks, BNY Mellon is the most focused on investment management, earning about one fourth of its revenue from that segment. It has about $1.6 trillion in assets under management and specializes in actively managed mutual funds. While its current focus is on distribution through institutional channels, it recently announced plans to significantly expand its retail distribution through its Pershing platform and other channels.

We see Northern Trust as the most diversified of the custody banks, since it earns nearly half of its revenue from other business, notably wealth management. Northern Trust is also the smallest of the three custody-focused banks by a wide margin. The bank's wealth management business focuses on serving individuals and families--it serves some 20% of the Forbes 400 richest families. Northern Trust reports its business segments differently than the other two custody banks, breaking out segments by customer type (institutional versus retail) rather than activity (investment servicing versus investment management), which complicates comparisons.

State Street is the closest to being a pure-play custody bank, earning nearly 90% of revenue from investment servicing activities. State Street also has a large passive (index) investment management business, with about $2.3 trillion in assets under management, and is a leading provider of exchange-traded funds.

Custody Banking Is a Naturally Moaty Business
We've long seen custody banking as a wide-moat business, and we've identified three primary sources of moats in the industry: customer switching costs, cost advantages, and efficient scale.

We believe customer switching costs are very high for custody banks. The firms' institutional clients, such as pension funds and asset managers, have technology systems that are deeply integrated with those of their custodians. The work that custodians do, while often banal, is closely tied with the core of their customers' businesses--safety, accuracy, and timeliness are of utmost importance. Errors could cause double payments or misstate account values, for example. Customers are therefore loath to risk the back-office disruption that switching providers could entail. Anecdotal evidence of high switching costs abounds: The California Public Employees' Retirement System retained the services of its custodian, State Street, when the contract came up for renewal in 2011 after suing the firm for "unconscionable fraud" in foreign-exchange pricing in 2009. And rumors persist that a boutique asset manager acquired by BNY Mellon retained Northern Trust as its servicer even after being brought into the corporate fold.

The other two moat sources, cost advantages and efficient scale, are intertwined, in our opinion. Asset custody and administration is a business where scale and scope are inherently important. The business is technology-driven and requires enormous fixed investments in systems--State Street spent $935 million on technology and communications in 2013 alone, for example--but has low variable costs. This makes it nearly impossible for new firms to build the scale and scope necessary to enter the market and offer competitive pricing, and we estimate that the top 10 custody banks control about 80% of global assets under custody. The top four banks alone control about 60% of the market.

As a result, we think the market is a natural oligopoly and benefits from what we call efficient scale: The firms already serving the market are so large that it is very difficult for another firm to enter competitively. We see evidence of this in market shares, which have remained largely stable over the past decade, excluding shifts related to mergers and acquisitions.

Protected by these competitive advantages, custody banks are able to earn very attractive returns--returns on tangible equity have been near 20% in recent years at both BNY Mellon and State Street, well above the banks' 10% costs of equity--and can approach 30% in better interest rate environments. At Northern Trust, which has a more diversified business model, returns on tangible equity have been more modest, averaging just under 10% since 2010, as the low interest rate environment has weighed heavily on its custody banking and wealth management businesses.

Despite Wide Moats, Recent Profits Have Been Subpar
While strong returns on tangible equity continue to provide evidence of wide moats, returns on equity have been more disappointing. Since 2010, return on equity has averaged just 7.2% at BNY Mellon, 9.7% at Northern Trust, and 10.2% at State Street.

Revenue growth, though lumpy, has been sluggish since 2009, and rising regulatory costs have meant that operating margins have fallen. Pretax operating margins have shrunk an average of 14 percentage points since 2009 and now average 26.5%.

In part, these low shareholder returns are structural because BNY Mellon and State Street have built their businesses through acquisition. At the end of 2013, 50% of BNY Mellon's common equity was goodwill, as was 30% of State Street's. Only 7% of common equity was goodwill at Northern Trust, which has relied much more heavily on organic growth.

The biggest drag on revenue at custody banks has been the persistence of very low short-term interest rates. Since December 2008, the Federal Reserve has targeted a fed funds rate of 0%-0.25%. Since 2009, net interest margins at the custody banks have fallen an average of 48 basis points. This is due in large part to the low interest rate environment, though it should be noted that structural changes at the businesses of BNY Mellon and State Street have contributed to the drop. The impact of low short-term rates is cumulative: As assets mature and roll off banks' balance sheets, managers are forced to replace them with lower-yielding assets. Initially, managers may be able to offset this by decreasing deposit rates, but they rarely fall below zero.

In the case of the trust banks, ultralow interest rates have an additional avenue to negatively affect revenue: money market fee waivers. The trust banks are important providers of money market funds, particularly BNY Mellon and Northern Trust. These funds typically charge fees of 0.15%-0.60% and invest client funds in short-duration assets like time deposits and U.S. Treasuries. In recent years, as yields on these assets have fallen to levels below the fees charged, trust banks have been forced to waive hundreds of millions of dollars in associated fees.

All Will Benefit From Rising Rates, but BNY Mellon Should Outperform
We anticipate that BNY Mellon will benefit disproportionately from a 100-basis-point increase in short-term interest rates. While the timing of such an increase is uncertain, we see near-zero rates as unsustainable in the long term; we expect rates to rise as demand for credit increases from cyclical lows and economic optimism increases. At the same time, we think net interest margins are likely to remain below peak levels, as high debt/GDP ratios across the developed world means that the ability of borrowers to take on more credit may be limited. We calculate that BNY Mellon could see a 26% increase in pro forma 2013 pretax profits. We expect that Northern Trust would benefit nearly as much and project a 24% increase in pro forma profits.

​

Our calculations make several key assumptions. First, we anticipate that a 100-basis-point increase in short-term rates would allow the trust banks to recoup 100% of waived money market fees. A smaller increase would allow the banks to recoup a smaller percentage of the fees.

Second, we assume that the banks' fully taxable equivalent net interest margins rise to normalized levels of 1.7%, 1.6%, and 1.8%, respectively, for BNY Mellon, Northern Trust, and State Street, with variations among them attributable to business mix and historical levels of risk taking. These levels are below peak 2007-08 levels, both because of structural changes at the firms and because we anticipate that global demand for credit is unlikely to return to previous levels because of already-high debt/GDP ratios in the developed world and sluggish economic growth.

Third, all of the trust banks have attracted significant amounts of deposits since the financial crisis, as clients seek safe places to store their excess cash, which the banks have then invested in low-yielding, short-term assets. In late 2011, the influx of deposits was so severe that BNY Mellon warned that it planned to charge some large depositors, although it quickly reversed itself after client outcry. While these deposits have not been profitable for the banks, we expect them and their associated assets to roll off as the economy improves and to be therefore unavailable for the banks to invest in higher-yielding assets. We use 2013 loans and normalized loan/deposit ratios to estimate the quantum of each banks' excess deposits: We estimate that excess deposits at year-end 2013 equaled $79 billion at BNY Mellon, $21 billion at Northern Trust, and $7 billion at State Street.

Finally, we assume that the benefit of higher rates resulting higher revenue will flow, in part, to employees. Using historical cost information, we estimate that 35% of incremental revenue will be distributed to employees as compensation and that 65% will flow to investors.

Digging into the numbers, we see two reasons BNY Mellon and, to a lesser extent, Northern Trust are likely to benefit more from rising rates than State Street.

First, the impact of money market fee waivers at BNY Mellon and Northern Trust has been much more significant on the downside and therefore likely to be more significant on the upside. BNY Mellon and Northern Trust waived $402 million and $108 million, respectively, in money market fees in 2013, compared with just $40 million at State Street. BNY Mellon's big money market business is, in part, an artifact of its predecessor, Mellon Bank's 1994 acquisition of Dreyfus, a leading provider of money market funds. For Northern Trust, the increased exposure stems from its proportionately larger wealth management business.

Second, we think BNY Mellon and, to a lesser degree, Northern Trust are likely to benefit more from rising rates, because we expect these two banks to see the greatest net interest margin expansion. BNY Mellon's and Northern Trust's net interest margins have experienced significant negative pressure from inflows of excess deposits, the proceeds of which are subsequently invested in highly liquid, low-yield assets. State Street, as more of a pure-play custodian, has experienced less of this pressure--we estimate that State Street's excess deposits total only 8% of BNY Mellon's. Evidence of this can be seen in the banks' net interest margins: State Street's was 1.37% in 2013 on a fully taxable equivalent basis, while BNY Mellon's and State Street's were only 1.13%. Further evidence of the weight of the excess deposits and the banks' resulting temporary conservative positioning can be seen on their balance sheets, with some 39% of BNY Mellon's assets invested in cash, central bank deposits, or deposits with other banks, compared with only 28% of State Street's.

Criticisms of BNY Mellon Are Overdone
BNY Mellon has fallen out of Wall Street's favor in recent years, as some have said that its cost-cutting has not been aggressive enough, that its client asset growth is lagging, and even that the 2007 merger of Bank of New York and Mellon Financial should be unwound. We think these criticisms are overdone and, in large part, represent misunderstandings of the underlying dynamics.

Criticism 1: BNY Mellon has not cut costs enough. Some have argued that BNY Mellon has not responded to the threat of low interest rates aggressively enough and its profits have suffered disproportionately has a result. In particular, they point to State Street as a firm that has taken the necessary but painful cost-cutting measures: It has eliminated about 3,000 jobs since 2011 compared with about 1,500 at BNY Mellon. We think this argument ignores evidence that State Street was in a significantly worse cost position to begin with in 2011.

In 2011, State Street was rightly criticized by investors, most notably by Trian Partners, an alternative investment management company, for having let costs grow faster than sustainable revenue. State Street's average operating expenses were 0.0466% of average assets under custody, some 44% higher than operating expenses of 0.0323% at BNY Mellon. While State Street's announced job cuts may not have materialized (the firm's total employee count fell only 310 between 2011 and 2013), the firm's efforts to contain costs in its custody business have clearly been successful, as business division operating costs per unit of client custody assets fell 19% between 2011 and 2013.

Despite this improvement, BNY Mellon remains the low-cost provider by this measure, spending only 0.0306% of average assets under custody in 2013, compared with 0.0377% at State Street. Other measures of cost-competitiveness show reinforcing results--pretax operating margins in the custody businesses of the two banks were similar in 2013. Moreover, firmwide average compensation per employee is lower at BNY Mellon, at $117,100 in 2013 compared with $129,100 at State Street, and has fallen 0.4% since 2011 compared with a 0.5% increase at State Street.

Criticism 2: BNY Mellon's assets under custody growth has lagged peers. On the face of it, this criticism is plainly true: Assets under custody (or assets under custody and administration, in the case of BNY Mellon), grew just 37% in total between 2009 and 2013 compared with a cumulative 87% at Northern Trust and 72% at State Street.

However, this ignores the fact that the banks' client assets under management are very different. BNY Mellon is the largest custodian for U.S. Treasuries, and as such, has a disproportionately large fixed-income custody business. Because of this focus on fixed income, BNY Mellon's growth in assets under custody lagged peers in 2009, 2012, and 2013, years that equity markets were unusually strong. Therefore, we believe that its underperformance does not indicate any weakness or decline in the business. While we expect this slower growth to continue, as equities typically grow faster than fixed income, we also expect this underperformance to be less pronounced in future years, as the outperformance of equities in 2013 is unlikely to continue. We project that assets under custody will grow an average of 5% annually at BNY Mellon, versus 7%-8% at Northern Trust and State Street.

Perhaps more important, we haven't seen evidence that this different business model has led to, or is likely to lead to, firmwide underperformance, thanks in part to BNY's diversified business model, which incorporates a large asset management business. Since 2008, BNY Mellon's revenue growth has been slightly ahead of peers, and we expect growth to remain largely in line thorough 2018.

Criticism 3: Bank of New York's 2007 merger with Mellon Financial has been a failure, and the firm should be broken up. This is a complicated argument. We agree that the deal was overpriced and destroyed shareholder value, but the all-stock deal makes it hard to say definitively who overpaid. It doesn't help that the deal was crafted at the peak of the market and created $11 billion of goodwill; as a result, annual returns on equity for the combined companies haven't exceeded 8%, compared with the 15% average returns posted by Bank of New York before the merger.

However, we continue to think that the deal made strategic sense: It brought together an institutional-focused investment management firm and a custody bank, which naturally have a significant amount of customer overlap. We think that this one-stop shopping creates an attractive value proposition for customers and probably leads to revenue synergies though cross-selling. We see evidence that the underlying business remains strong as measured by returns on tangible equity, which we expect to increase to the upper end of historical levels as interest rates rise. Excluding the bubble years of 2006-07, returns on tangible equity at Bank of New York ranged between 25% and 35%; we estimate that midcycle returns on tangible equity for the combined business will be 30%-32%.

We think investors should see the value destroyed by the overpriced merger as a sunk cost--it cannot be recouped by breaking up the firm--and focus instead on the moatiness of the underlying business, especially given the company's current focus on organic growth.