The argument for owning First Republic (FRC), a bank that has often appeared to trade at robust (if not excessive) multiples, has long been that it’s a different sort of bank. Skeptical veteran investors can be forgiven for shaking their head at the “it’s different this time” argument, but First Republic continues to make the case that it really is a different sort of bank – one that can navigate this unexpected pandemic-driven recession better than the vast majority of its peers.
I liked First Republic after first quarter earnings, and the shares have risen about 16% since then – better than the average regional bank, but only a little better than the S&P 500. While that move has shrunk some of the undervaluation I saw, the shares do still look undervalued and I believe the long-term expected return here is pretty attractive, particularly if you believe that multiples can re-inflate somewhere down the road when investors return to the banking sector.
A Beat That’s Not Likely To Be The Norm
Banks are only just starting to report earnings, and large banks like Citi (C), PNC (PNC), and Wells Fargo (WFC) really aren’t useful comparables to First Republic. Nevertheless, I think it’s worth pointing out that First Republic could be one of the relatively few banks to post any sort of meaningful beat on net interest income.
Revenue rose about 12% from the prior year and a little less than 1% from the prior quarter, beating expectations by a few percent on an FTE-adjusted basis. Net interest income rose 17% yoy and 5% qoq, beating by close to 5%, as NIM compression was modest (down 15bp yoy and 4bp qoq) and First Republic was able to effectively redeploy a surge in balance sheet assets (average earning assets up 24% yoy and 6% qoq).
First Republic did underwhelm on the fee income line with a 7% to 13% miss relative to the sell-side depending upon whether you use a reported (down 11%/down 19%) number or an adjusted figure. Investment advisory income was weak, falling 9% yoy and 14% qoq.
Opex came in lower than expected, rising about 8% yoy but falling more than 4% qoq as the company saw a decline in marketing, travel, and professional expenses. In terms of the expense ratio, First Republic beat by more than 260bp. At the pre-provision profit line, First Republic saw 19% yoy and 10% qoq growth, beating expectations by about 7%, or $0.13/share. Tangible book value rose 12% yoy and 2% qoq.
Good Loan Growth, With Or Without PPP, And Still-Very-Healthy Credit
Smaller banks outperformed large banks in the second quarter in loan growth, with PPP driving better than 32% qoq growth in end-of-period C&I lending balances and better than 6% overall qoq loan growth. First Republic didn’t match that, but the bank nevertheless reported very healthy loan growth, with end-of-period loans up 22% yoy and 5% qoq (or 3% excluding PPP). With annualized Q2 growth of 12% even excluding PPP, I’d say First Republic is continuing to do a very good job of gaining lending share in its targeted high-net-worth markets.
This was done in the face of weaker C&I lending, as C&I lending actually declined 8% qoq. Capital call lending declined 18% qoq, and I’d expect that this could be weak for a while longer. Lending categories like single-family (up 7% qoq) and multi-family (up 3% qoq) helped compensate, and originations were up 11% qoq (excluding PPP). Yields remain pressured, with First Republic seeing loan yields down 49bp yoy and 22bp qoq.
Net of a $12.5M reversal on unfunded commitments, First Republic saw provision expense decline about 30% qoq to $43.5M. The bank built reserves by another $30M, and First Republic will likely be one of the relatively few banks to post a lower-than-expected provision expense.
While First Republic’s loan loss coverage of 0.6% looks low next to the giants (where loan coverage is around 3% to 4% for the most part), First Republic has a distinctly different loan book and credit quality experience. Charge-offs are still virtually zero, and the NPA ratio is low (13bp this quarter), and while I do expect charge-offs to accelerate significantly relative to First Republic’s historical experience, I still expect that they will be well below systemwide averages – an 0.08% net charge-off ratio in 2021 or 2022 would be exceptionally high for this company on an historical basis, but far below the systemwide cyclical low (around 0.7%).
First Republic’s healthy loan growth this quarter certainly should help build confidence in the outlook for mid-teens loan growth even as the economy continues to grind through the pandemic-driven recession. While the timing on PPP loan forgiveness could lead to some “noise” in quarterly earnings over the next four quarters, I’m not too concerned about that. Likewise, while I don’t think this quarter’s expense ratio is necessarily sustainable, I don’t think First Republic has a cost problem. What could go wrong for First Republic? The biggest risk would be a bear market in technology and/or a property value crash in the Bay Area. While the fallout to First Republic wouldn’t be as bad from a credit quality perspective (as the bank is at least partially protected by healthy loan-to-value ratios), it would almost certainly lead to lower loan demand and weaker growth.
I’m still expecting double-digit core earnings growth from First Republic, with long-term growth in the low-to-mid teens still possible in my view.
The Bottom Line
Between discounted core earnings and ROTCE-driven P/TBV, I still believe First Republic is modestly undervalued, with a long-term prospective total return in the low double-digits. While First Republic’s growth is likely to slow over time, the company’s core high-net-worth market still offers exceptional scope for growth and I think this is still a reasonable entry point for investors.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any 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.