Opening Remarks of FASB Member Hal Schroeder
(as prepared for delivery)
AICPA National Conference on Banking and Savings Institutions
“Financial Instruments: The Way Forward”
National Harbor, Maryland
September 9, 2019
(as prepared for delivery)
AICPA National Conference on Banking and Savings Institutions
“Financial Instruments: The Way Forward”
National Harbor, Maryland
September 9, 2019
Back in the spring, as the school year was ending, I showed my daughter a picture of what appeared to be an old, very thick textbook. Knowing the last thing she wanted was another textbook, I jokingly told her we were going to buy it. She then asked two logical questions: Why? and How much?
I won’t get the tone of her questions quite right, so I’ll just say there was a lot of eye rolling. Now, in my defense: She didn’t know the book’s history. Without any historical context, without opening the book to see its contents, she couldn’t appreciate the importance of this book.
What she saw was the tattered, brown-paper cover . . . blank, except for three handwritten letters. So, she was shocked to learn what it would likely fetch at auction.
The book was Luca Pacioli’s Summa de Arithmetica. In the press release1 announcing the auction, Christie’s noted its broad-ranging contents: “from double-entry bookkeeping to probability theory and computing, the mathematical principles of the most vital features of contemporary finance are all present.” In other words, this 525-year-old book is the “first practical how-to book on succeeding in business.”
During a multi-city tour, Christie’s invited the FASB to experience the book firsthand. As the curator opened the cover to read passages, to highlight various illustrations, I developed a deeper understanding of what was inside the book, why it was important and how it was constructed. No wonder this book has stood the test of time.
I tell this story, because it’s a great example of the idiom: You can’t judge a book by its cover. The same can be said for the FASB’s projects I’ll talk about today.
But of course, I must first remind you of the standard disclosure: Official positions of the FASB are reached only after extensive due process and deliberations. In other words, what I am about to say are my views and only my views. This disclosure also covers Shayne, who will be joining me shortly to answer questions.
The standard disclosure I just read appropriately puts an emphasis on the FASB’s “process” for setting accounting standards. And, that’s where I’ll start.
About the FASB Process
The Board and staff often talk about the standard-setting process, up to the point of issuing a new standard. We don’t talk as much about the equally important post-issuance phase. Once a major new standard is issued, we spend the time—often years before its effective date—educating stakeholders and monitoring company implementation efforts. One reason we do this is to re-confirm that a new standard’s wording is understandable, operable, and auditable. Another is, we want to ensure it will stand the test of time.
During numerous education and monitoring sessions—based solely on the questions and comments of the participants—it becomes readily apparent who’s gone past the cover and is deep into implementing a new standard. And, equally apparent, who hasn’t broken the book’s binding. (Yes, I know, in the mobile-device era, this phrase is dated, but it has the imagery.)
With the latter group, we often find ourselves reiterating and relitigating a range of questions. Personally, I enjoy the debates. However, most of their questions have already been asked and answered in the due-process documents and final standards issued in accordance with our Rules of Procedure.2
As much as I enjoy the debates, I’ll simply highlight a few points about what’s required, and how to do it.
Regarding the “what” of CECL specifically—in the past year the staff has received very few questions, many of which could be quickly answered by referring directly to the guidance. And, for other questions, we’ve made targeted improvements to enhance understandability and operability.
Regarding “how,” we’re committed to looking for ways to ease transition efforts. For example, in January, we published a FASB Staff Q&A about using WARM [weighted-average remaining maturity method]3 . . . an estimation method already familiar to many smaller financial institutions. It was followed in July by another that addresses developing estimates.4 In addition to those Q&As, the staff are hosting several implementation workshops around the country. One of which is being held here this afternoon at 4:25pm; check the conference agenda for details.
Considering that our technical inquiry service is free and accessible online,5 the limited number of questions is at odds with the view some have voiced that there are many unanswered questions and, therefore, CECL should be delayed.
And, despite some very public comments to the contrary, a thorough cost-benefit assessment of CECL was conducted—as is done on all standards in accordance with the FASB’s Rules of Procedure. We explain why we came to our decisions in the “Background Information and Basis for Conclusions” section of the standard. A summary6 of the assessment is also provided on the FASB’s CECL implementation page.
In my eight years on the Board, I’ve come to appreciate the high-quality standards that result from its disciplined process. CECL may not be perfect, but it’s a vast improvement over the status quo. And, we’re doing everything we can to support your success.
So, I have a request: When you hear someone say the FASB didn’t follow its due process, please ask if they’ve gotten past the cover; past the headlines in media reports. Have they read the standard, in particular, the basis for conclusions?
Like any classic book, standards such as CECL should be read multiple times. The same is true for FASB Staff Q&As. With each subsequent reading, new insights are gained, leading to greater benefits and a more cost-effective implementation.
Final thought on process—disagreeing with a Board position is fine. I certainly have, in fact, about 10% of the time—just read any one of my dissents. But I’m troubled by those that make public statements, unsupported by the public record documenting the process.
Calls to “Stop and Study”
During the Past Year
Despite years of discussions and debates, there continue to be efforts to relitigate CECL. As you may recall, reconsideration of CECL was the focus of a roundtable meeting this time last year. It was hosted by three congressmen to facilitate a discussion of a “regional bank proposal,” the details of which I shared with you at last year’s conference. A few months later, a modified version of that proposal was formally submitted to the FASB. Again, following its established process, the FASB staff held dozens of meetings that included financial and nonfinancial entities, as well as other stakeholders.
The views heard in those meetings were very consistent with stakeholder comments at a public roundtable we hosted in January—splitting the provision, with a portion flowing through net income and the remainder in OCI, was not operational. This was not a surprise. Banks had provided the same feedback in 2013, when the Board had explored the same idea and its many variants.
We also heard from investors that, if the provision is split, they would need additional disclosures. However, those re-proposing a split did not support expanded disclosures. After hearing all sides, and consistent with its 2013 decision, in April the Board again voted to not move forward with any requirement to split the provision.
With no reduction in regulatory capital expected, the “stop-and-study” movement is now seeking a “legislative fix” with bills proposed in both the House7 and the Senate.8 In a separate action, a study-only directive has been attached to a House appropriations bill. That request is directed toward others including the SEC, and not to the FASB.
Those in the audience that know me won’t be surprised. I do have views on the matter. But as a reminder, these are my views and should not be interpreted as the Board’s views.
Calls for Quantitative Impact Study
“Stop-and-study” efforts call on the FASB to perform an economic or quantitative impact study on CECL. (Going retro, I’ve heard it called “delay-and-pray,” mimicking the approach used by some to deal with problem loans in past banking crises.) Those advocating stop-and-study allege dire consequences from CECL’s so-called procyclical effects on lending. However, independent studies9 already conducted do not support such concerns.
Consider data contained in a recent study10 published as a working paper in the Federal Reserve’s Finance and Economics Discussion Series (FEDS). The study, and I quote, “shows that a disproportionate share of the associated provision expenses occurs prior to the recession under CECL, rather than during it.”
The study quantifies CECL’s earlier provisioning, estimating that “roughly 62% of the trough to peak increase in allowances occurs prior to the recession [compared to] only 11%” under the incurred loss model.
Now, combine that finding with other independent studies showing banks that provision earlier are better positioned to lend in an economic downturn. Quoting from one such study: “Consistent with the pro-cyclical provisioning hypothesis we observe a greater reduction in lending during recessions by banks that delay expected loss recognition more compared with banks that delay less.”11 That conclusion was based on 24,788 bank quarters of data, stretching over a 16-year period (1993 to 2009) that included 2 U.S. recessions. Another study,12 based on 3,091 bank-year observations, across 27 countries over 11 years (1995 to 2006), yielded complementary results.
Yet, I expect critics will point to the FEDS working paper to show that CECL reserves would go up during a recession and would be higher than under the incurred model. I agree with the observation. CECL allowances would likely increase during a recession and peak at a higher point. In fact, using the same data, I calculate CECL-based allowances would have increased an estimated 29% from the beginning of the recession and peaked in the fifth quarter. However, those same critics are unlikely to mention that under the incurred model, allowances actually increased 264% during the last recession—8 times that expected under CECL—and, didn’t peak until the ninth quarter—taking almost twice as long to recognize losses.
Bottom line: As independent studies show, banks that are better reserved heading into a recession are better positioned to lend during a recession. And, it is the banks that reserve later, and take longer to work through existing losses, that cut lending during a recession. Therefore, I believe concerns about CECL’s impact on lending are misplaced.
My comments today should not be interpreted to mean I’m anti-study. In fact, just the opposite. I believe well-designed, well-executed independent studies—a few of which I’ve cited—are an invaluable tool to standard setting. And I believe there should be more in the future, using real data as it becomes available.
What I’m opposed to is stopping CECL’s implementation. Here are a few reasons why.
No Change to Cash Flows
A key point to remember, under CECL: a good loan will still be a good loan; a bad loan will still be a bad one. This is because CECL doesn’t—in fact, it can’t—change the ultimate cash flows or a borrower’s ability to repay. And CECL doesn’t even change when to charge off a loan. It changes only the timing of when loss provisions are recognized in net income and, in turn, reserves built on the balance sheet.
Even absent CECL, investors are and will be making their own estimates of expected losses. As I’ve shown at this conference in the past, investors began ignoring GAAP allowances—effectively making their own, much-higher loss estimates—18 months before the beginning of the last recession. This is in stark contrast with the fact that GAAP-based allowances didn’t hit bottom—a multi-decade bottom—until just 12 months before the recession began.
What CECL does is provide greater transparency into changing credit risk. Instead of having to rely solely on their own estimates, investors will be able to start their analyses with managements’ more-informed estimates.
If there’s any doubt in my view, in the last few months, I’ve been talking with investors. What some have shared is that they’re already beginning to again ignore GAAP provisions and allowances and, to again, make their own loss estimates. As Mark Twain is often credited with saying: “History never repeats itself, but it rhymes.”
So, how do we reconcile the clear need for better alignment between accounting and changes in risk, with calls to stop a standard that provides much-needed transparency? I can’t.
On a Personal Note
In my former job as a portfolio manager, I’d have been okay with keeping opaque accounting. Frankly speaking, our investors earned better returns because we were estimating expected losses. It gave us a competitive advantage.
But as a member of the FASB, I’m not okay with the status quo. No competitive advantage is worth keeping accounting that:
- In “good times” masks warning signs of rising credit risk, and
- In “bad times” is ignored.
At this point, you may be asking: If he’s so pro-CECL, how can he support the recently proposed delayed effective dates?
It’s because some of you in this room made a compelling case—and our research confirmed—that implementation challenges are often magnified for private companies, smaller public companies, and not-for-profit organizations. Access to resources, education, and technology varies considerably among organizations of different sizes. We did not want those hurdles to block the path toward a successful implementation for anyone.
We also observed that when it comes to applying standards, private and smaller public companies can learn a lot from the experiences of larger public entities. More time between effective dates means more time to learn.
It’s also worth noting a practical implication—roughly 90% of financial services companies (by number) will get extra time to adopt the standard. However, I don’t believe investors will be significantly affected, because the 10% that won’t get extra time represent 90-ish% of the industry assets.
Not all smaller institutions agree. Minutes before the August vote, a community bank CEO called me to express his strong disagreement. He said his bank didn’t need the additional time.
In response, I told him what I’d heard: When implementing major standards, such as CECL, it’s more cost effective to take an integrated, wholistic approach that coordinates accounting changes with other changes needed to run a successful business—what I refer to as a “business approach”—which contrasts with treating accounting changes as simply a “compliance exercise”—incurring costs with minimal to no benefits for running a business.
I believe providing an opportunity to take a business approach is the strongest, most compelling argument for extra time.
Benefits to Taking a Business Approach
In fact, as we monitor progress among larger lending institutions preparing to implement CECL next year, we’ve observed the benefits of taking a business approach. In private meetings—including a few where we’ve had to sign non-disclosure agreements—we’ve been told that the effort to adopt CECL has resulted in widespread improvement of data quality, internal controls, estimation processes, and internal coordination and communication. As one banker told me, “improvements were needed; CECL simply accelerated the timing.”
One cautionary note: If the Board votes to provide the 90-ish% with extra time, I believe there’ll be an even greater expectation for high-quality implementation efforts.
Don’t look at it as an extra year off. Treat it as an opportunity to improve your data quality, estimation processes, and internal controls. Start now. If you’ve yet to break the book’s binding, do it today.
CECL Feedback from Investors
One final comment on CECL, and it comes from investors: there’s growing concern about the lack of quantitative, CECL-related SAB 74 disclosures from the 10%. During our outreach, investors have told us they’re frustrated at having to work on their 2020 estimates with little to go on, other than a few high-level estimates.
At last year’s conference, I noted that when IFRS 9 took effect, many investors struggled to understand the accounting change. I heard significant frustrations voiced in Europe.
Again, having walked in their shoes, I encourage you—in the 113 days remaining until 2020 and roughly 215 days until the first earnings calls—to help your debt and equity holders readjust their eyesight. Otherwise, we’ll experience the same frustrations in the U.S.
Now, on to another issue—one that affects everyone in this room: reference rate reform.
Reference Rate Reform
The London Interbank Offered Rate, or LIBOR, is expected to be phased out in 2021. An estimated $350 trillion in loans, derivatives, and other financial contracts reference LIBOR.13
In July, the SEC issued a public statement14 noting that the transition away from LIBOR could have “a significant impact on the financial markets and may present a material risk for certain market participants.”
Reference rate reform has been a major priority for the FASB. We’re proactively addressing accounting obstacles that could hinder smooth transition.
For example, last year, we amended the hedging standard to include the Secured Overnight Financing Rate—or SOFR—as a permissible benchmark rate.
And last week, we issued a proposal that would make it easier to transition contracts and other arrangements to a new reference rate.
First, we proposed to simplify the accounting evaluation of a contract modification. For a contract that meets certain criteria, the change in its reference rate would be accounted for as a continuation of that contract, instead of the creation of a new contract. This accounting relief could be applied to loans, debts, leases, or any other type of contracts affected by reference rate reform.
We also proposed to simplify the assessment of hedge effectiveness—and to allow hedging relationships affected by reference rate reform to continue. In other words, you’d be allowed to preserve hedge accounting when updating your hedging strategies.
Optional and Temporary
Application of this relief would be optional on a hedge-by-hedge basis. We think it will minimize the disruption of reference rate reform on financial reporting and provide users with more useful information.
Companies and other organizations that elect to apply the guidance would do so prospectively—in other words, the new guidance could be elected for contract modifications and hedge evaluations that occur after we issue the final standard.
The guidance would be temporary, expiring on January 1, 2023. To my knowledge, if finalized, this would be the first time we’ve ever issued an accounting standard with a “sunset provision.”
The comment period on this proposal is open until October 7th. I encourage you to review it and get your comments to us as soon as possible. As always, your insights will help us develop a better standard.
One last thing: We know there’s additional work for us to do in this area. But, like you, we’re monitoring the markets for any new developments.
So, if you don’t hear anything new from us on this front, it’s not because we’ve been idle. It’s because we’re waiting for more information on what’s next. We refer to this as “moving in tandem with the market”—not getting too far in front or behind this evolving issue. But rest assured, we’ll let you know more as soon as we know more.
In addition to reference rate reform, we continue to improve the hedging standard we issued in 2017.
As many of you know, that standard added the last-of-layer hedging method. For a closed portfolio of fixed-rate, prepayable financial assets—such as mortgages—the last-of-layer method allows an institution to designate an amount that is not expected to be affected by prepayments, defaults, and other events affecting the timing and amount of cash flows. As a result, prepayment risk is not incorporated into the measurement of the hedged item.
Since issuing the standard two years ago, financial institutions and other stakeholders have asked us to consider additional improvements. Many stakeholders told us that the ability to hedge a single layer was useful, but not as helpful as it could be, and they asked if the single-layer model could be expanded to hedge multiple layers of a single portfolio.
We heard you. We decided additional guidance would be necessary. So, we added a project to our agenda to consider whether to expand the last-of-layer method to multiple layers.
Last fall, we conducted outreach with approximately 30 stakeholders, including financial institutions, hedging consultants, and accounting firms. Based on what we heard at those meetings, we understand that expanding the model to include multiple layers would be a valuable addition to your interest rate risk management tool kit.
This project is picking up steam. In August, the FASB held an education session15 on expanding the last-of-layer method. At that public meeting, the FASB staff presented the results of its outreach efforts. The staff also discussed potentially providing further guidance on accounting for “fair value hedge” basis adjustments, for both the existing single-layer model and the proposed multiple-layer model.
The Board agreed with the direction of the project. In the coming months, we expect to make some decisions on this issue at a public meeting. I urge you to follow our progress—and be prepared to share your input on a future proposal.
Before I conclude my remarks, I want to thank you for your engagement in our process. Your input challenges us to set better standards—especially CECL.
Your involvement helps us prepare for, and respond to, changes in the capital markets and helps drive our work to ease the transition to new benchmark reference rates. And, your suggestions on how to improve existing standards—like hedging—allow us to work together to make financial reporting more decision useful.
We may not always agree on outcomes. But we will always engage in the conversation. And we will work with you to help ensure a successful transition to our standards.
And now, please welcome Shayne Kuhaneck, who will join me in answering your questions.
1Christie's NY to Auction Pacioli's "Summa de Arithmetica," June 12, 2019 https://www.finebooksmagazine.com/news/christies-ny-auction-paciolis-summa-de-arithmetica
2FASB Rules of Procedure, Amended and Restated through December 11, 2013, https://www.fasb.org/jsp/FASB/Page/SectionPage&cid=1351027215692
3FASB Staff Q&A, Topic 326, No. 1, Whether the Weighted-Average Remaining Maturity Method Is an Acceptable Method to Estimate Expected Credit Losses, January 2019, https://www.fasb.org/cs/ContentServer?c=Document_C&cid=1176171932989&d=&pagename=FASB%2FDocument_C%2FDocumentPage
4FASB Staff Q&A, Topic 326, No. 2, Developing an Estimate of Expected Credit Losses on Financial Assets, July 2019, https://www.fasb.org/cs/Satellite?c=FASBContent_C&cid=1176172971977&pagename=FASB%2FFASBContent_C%2FGeneralContentDisplay
5FASB Technical Inquiry Service, https://www.fasb.org/inquiry/
6FASB Understanding Costs and Benefits, ASU: Credit Losses (Topic 326), June 16, 2016, https://www.fasb.org/cs/Satellite?c=Document_C&cid=1176168233403&pagename=FASB%2FDocument_C%2FDocumentPage
7H.R.3182 - CECL Consumer Impact and Study Bill of 2019, https://www.congress.gov/bill/116th-congress/house-bill/3182/text?q=%7B%22search%22%3A%5B%22Gonzalez%22%5D%7D&r=1&s=2
8S.1564 - Continued Encouragement for Consumer Lending Act, https://www.congress.gov/bill/116th-congress/senate-bill/1564
9Refers to comprehensive studies conducted by researchers not associated with individual banks or trade associations.
10Loudis, Bert, and Ben Ranish (2019). CECL and the Credit Cycle," Finance and Economics Discussion Series 2019-061. Washington: Board of Governors of the Federal Reserve System, https://doi.org/10.17016/FEDS.2019.061.
11Beatty, Anne, and Scott Liao. 2011. “Do Delays in Expected Loss Recognition Affect Banks’ Willingness to Lend?”. Journal of Accounting and Economics 52: 1-20.
12Bushman, R. M., and C. D. Williams. 2012. “Accounting Discretion, Loan Loss Provisioning, and Discipline of Banks’ Risk-Taking,” Journal of Accounting and Economics 54: 1-18.
13Bloomberg, August 27, 2019, Wall Street Prepares for the End of a Crucial Benchmark, https://www.bloomberg.com/news/articles/2019-08-27/libor-undertakers-wanted-wall-street-braces-for-benchmark-s-end
14SEC Staff Statement on LIBOR Transition, July 12, 2019, https://www.sec.gov/news/public-statement/libor-transition#_ftn1
15Board Meeting Handout, Hedging—Last-of-Layer Method, August 21, 2019, https://www.fasb.org/cs/ContentServer?c=Document_C&cid=1176173211631&d=&pagename=FASB%2FDocument_C%2FDocumentPage