Just when you thought it was safe to go out at night again, another reason not to deploy capital is slouching into Bethlehem. We’ve written a lot here at CrunchedCredit about the Damian-like progeny of Dodd-Frank and Basel, but we’ve let this one slip through the cracks. And, boy, oh boy! – We need to pay attention to this thing. We’re talking FASB.

Okay, so what’s this all about? The story starts in Norwalk, Connecticut back in the 1970’s. The accounting industry at that time, chartered a private institution known as the Financial Accounting Standards Board (FASB) to establish financial accounting and reporting standards for public and private companies complying with Generally Accepted Accounting Principles (GAAP). A powerful organization was born. FASB still sits today in leafy Norwalk, Connecticut and generally beavers away in relative obscurity, tinkering with GAAP standards, both large and small. Periodically, however, the Board tosses a Zeus-like bolt of lightning from on high masquerading in the clothing of dry, dusty guidance to auditors which fundamentally changes how business is conducted. (Btw, let me tell you, Norwalk makes a pretty crappy Olympus).

This is one of those cases. The new rule will fundamentally change how lenders report impairments to the loans they make. Here we have a new Rule which might (I say might) arguably, make sense inside the academic bubble of the FASB, but will create ruin and frustration in the real world.

Why? First, because the musings of FASB matter a lot. The FASB standard is effectively federalized by the SEC through the SEC’s governance of publicly traded companies. Consequently, changes to GAAP ripple through the prudential rule-making activities of all federal (and state) agencies overlooking the US banking system. These changes will change lender behavior, and we are concerned that these changes will not be good.

As readers of this column undoubtedly know, banks have, since time immemorial, policies and procedures to identify potential losses associated with their lending books and create appropriate reserves. While practice varies between institutions, broadly speaking, when evidence of an impairment has been observed, the loan loss position of the bank is bolstered. This is called an incurred loss model. FASB’s new rule, known with only the elegance that an elaborate bureaucratic institution can so blithely achieve, has been designated “(ASU) Update 2016-13—Financial Instruments—Credit Losses (Topic 326): Measurement of Credit Losses on Financial Instruments,” or as it is known to the consigliere, CECL[1]. CECL was issued in 2016 as a final rule and becomes effective in 2021 with early adoption permissible in 2019.  Given the nature of the banking regulatory establishment, early adoption is likely and that means that we need to start worrying about this now.

Perhaps I’ve buried the lead here, but the CECL will require reporting entities to begin booking losses on loans at the time the loan is made instead of when impairment to the loan is observed. So, under CECL at the very moment the loan is made, a reserve equal to the difference between the cost basis of the loan and the expected net recoveries needs to be recognized. Maybe in some alternate universe this makes a great deal of sense, but in the real world it feels a lot like going to the hospital before driving your car because maybe you’ll crash. While it might mitigate the injury, it surely means making the trip considerably harder.

Under the auspices of the American Bankers Association, a number of major banks recently addressed a letter to Secretary Mnuchin asking the Treasury to conduct a review of the economic effect of the rule. One can hope that that efforts will bear fruit, but I am not sanguine.

So, at this point, I’m assuming that it is more likely than not that CECL will become effective given the deference accorded the FASB by the SEC and the prudential regulators. Moreover, if the Treasury indeed stirs itself to inquire about economic consequences, FASB is certain to mount a robust defense of its Rule. After having been pilloried by the great and the wise in the aftermath of the Great Recession for being a patsy for the financial industry (not true, but that’s a different story), the pendulum has well and truly swung. Since the Great Recession ended, the Board has displayed, in my view, a marked anti-financial industry animus in a response to the opprobrium heaped upon them by the chattering class. While a great deal of water has gone under the dam since I last dealt directly with the Board, in the fight around true sale and consolidation, many in Norwalk brought a distinct antipathy to the views of industry experts who endeavored to explain how structured finance actually worked. (We are still burdened today with GAAP rules around those issues that, in many respects, make little sense.)

So what happens when CECL goes into effect? All the major banks are undoubtedly preparing for this eventuality now by developing policies and procedures to assess future expected credit losses at the time of loan origination. This strikes me as a difficult task. How’s that crystal ball today? What’s the data-set upon which to make these assumptions? The Board helpfully says, “The measurement of expected credit loss is based on relevant information about past events, including historical experience, current conditions and reasonable and supportable forecast that affect the collectability of the reported amount.” Now that’s constructive.

While coming up with an effective model will, in and of itself, be a daunting exercise, it’s entirely clear that all our 7,000 prudentially regulated banks and the hundreds and hundreds of non-bank lending institutions reporting financial results under GAAP are not going to come up with the same model nor come out with the same levels. We’ve seen the chaos that flowed from the High Volatility Commercial Real Estate (HVCRE) initiative from a few years back and that’s certainly going to be replicated here. How does that do the banking system any good, one might ask?

So, what’s going to happen? It is going to create a great deal of pricing volatility across the lending community, even more than we have now, as each institution rolls out its own methodology of assessing expected losses. Second, all loans will get more expensive since lenders will have to earn some sort of return against those early losses. Note that losses are now going to be incurred up front while earnings will be incurred over time. That seems like a mismatch that shouldn’t happen.

Third, this will incent short-term lending over long, as the longer the term, the more likely that losses will exist and more likely the losses will be large. Sidebar: Absent a federalized guarantee through the GSEs, what’s the impact of all of this on the classic 30-year residential mortgage? Can’t be good.

It’s going to create real disconnects between GAAP sensitive and GAAP insensitive lenders. Will there be reconciliation with this new rule with IFRS? Will non-domestic banks have an advantage here? How about lifecos reporting under SAP? Finally, it is dangerously pro-cyclical. When markets turn down and losses begin to grow, the models will undoubtedly project larger expected losses and this will be yet another reason for lenders to make fewer loans. That does not seem like good policy.

It all seems to lead to a diminution of capital formation. This is yet another reason to aggressively manage down the loan book.

Our friends in Norwalk shouldn’t be producing academically coherent rules that have material negative externalities in the real world. Even the government is supposed to test the economic impact of its major regulations (a laughable assertion, actually). FASB needs to as well. We’ve seen the Board make this mistake before by not understanding how rules will impact the real world and they’re making it again here.

[1] CECL governs the accounting of all financial assets measured on an amortized cost basis, but at least here, I only care about loans. These changes to lease accounting will have an equally significant impact on how business is done.