Managing Profitability Under CECL Through Loan Pricing (Part 1)
BY BEN MURRELL, FI CONSULTING
Part 1: Defining a Conceptual Framework for Loan Pricing
As organizations implement CECL, a key question is how CECL estimates should factor into loan origination and pricing decisions. Linkage is important because CECL will require that expected losses be reserved at time of origination.
Correctly pricing in new reserves for loan losses will enable the bank to meet its internal return on equity requirement, but how exactly does a bank price risk into its loans? How might this change under CECL?
In this first of two papers we use a hypothetical example to illustrate the mechanics of loan pricing. In the second blog, we will use actual residential loan data to model how, under CECL, pricing decisions may need to change during different levels of expected portfolio losses.
The Building Blocks of Pricing
The primary financial objective of pricing is to originate a loan with a positive Net Present Value (NPV) and a Return on Equity (ROE) above hurdle rate. Market share, volume, and economies of scale are other strategic considerations when setting price. However, our focus is on loan level NPV and return on equity ROE as these financial metrics are much more directly related to a loan’s risk.
Financial Institutions consider many factors when determining price. Those critical to NPV are highlighted below.
We first discuss how each of these factors individually fit into a pricing framework. Then we illustrate the full framework under two different scenarios. The numbers that follow are illustrative.
Capital Structure: Banks borrow money to lend money. Exactly how much they borrow is a function of the prospective loan’s risk and their overall book’s risk, among other variables. A less risky loan allows for more debt to fund it. It’s not uncommon to see leverage in excess of 20:1 on a mortgage loan.
Cost of Capital: A bank’s cost of debt is a function of its funding mix (e.g. deposits, securitization, unsecured) and its overall asset and liability mix. A bank’s cost of equity represents the opportunity cost of allocating this equity to another asset. It serves as the hurdle rate and the discount rate in the end NPV. We include a 25% tax rate to make our analysis more realistic.
Operating Costs: In addition to its lenders, its shareholders, and the government, a bank also pays its staff to originate and service the loan. These expenses are often amortized over the life of the loan via FAS 91, adding complexity to a pricing model. We assume these are expensed as incurred, i.e. when the cash changes hands.
Note Rate: This serves as a starting point since we solve for final note rate using the other inputs.
Expected Life: For the sake of illustration, we ignore prepayments in this conceptual framework. Prepayment speeds are especially important within some asset classes, namely mortgage, but we focus instead on the impacts of losses and their timing. We also assume straight-line amortization.
Expected Loss: Expected loss can differ materially from loan to loan with direct implications for pricing. This is reflected under Provision Expense in the Income Statement. Under some CECL approaches, loss reserves could be expected to be recognized earlier than under the incurred loss model. This would pull expected NPV down.
Bringing It All Together
Banks operationalize pricing differently, but the framework we present here is broadly representative. Lenders may make pricing decisions at a loan level or at a segment, channel, or product level. Whatever level they choose, they will at some point consider the full Balance Sheet and Income Statement impact of a prospective loan or portfolio.
Balance Sheet: Applying the equity and debt ratios to a fictional $10,000 allows us to create a pro forma Balance Sheet. Note the Equity Draw of ($400) at the end of period 0. This represents the bank’s initial “skin in the game.”
Income Statement: We apply our individual drivers to the average balances in the pro forma balance sheet above to arrive at a pro forma income statement. Provision Expense is often a key driver of the loan’s cash flows. Because reserves may be recognized earlier under CECL, we’ll illustrate an alternate scenario with an increased loss rate as a proxy.
Result: Because this is a leveraged NPV analysis, the bank does not simply discount earnings over the initial balance of the loan. (96% of this loan is borrowed.) Rather, the Total Equity Flows (net of expenses and taxes) are discounted at the bank’s (or business line’s) Hurdle Rate of 12% to arrive at the NPV. This loan is expected to contribute to the business line’s profitability.
An Alternate Scenario
Changing any of these factors will change NPV. Should the pricing desk want to originate a loan with higher credit risk, the model is rerun with updated factors. This is what NPV looks like for the same loan, but with a 14% probability of default (+4 percentage points over baseline). While CECL adoption will not change the underlying risk of a loan portfolio, it could result in higher Provision Expenses in earlier periods and a lower overall NPV. Modeling a loan with an increased loss rate serves as an illustration of these impacts.
For this loan to be profitable (like the first), it’s note rate will have to be increased from 6.5% to 7% to cover the additional Provision Expense. Operationally, this is accomplished via goal-seeking. At a 7% note rate, this loan’s NPV is above zero and its ROE is above the business line’s Hurdle Rate of 12%.
To summarize, the pricing desk constructs a full Balance Sheet and Income Statement for a potential loan or portfolio then discounts at the business line’s hurdle rate. Pricing should be increased when NPV falls below 0 or ROE falls below hurdle.
An NPV analysis aims to answer the question: Is it valuable to the shareholder? An important principle is that the accounting rules apply. After all, shareholders see the financial statements. The timing of different Income Statement line items will have implications for a loan’s NPV. Under some CECL approaches, loss reserves could be recognized earlier than under the incurred loss model. In our next installment, we offer a data-driven analysis of how CECL adoption could affect loan NPVs and business line profitability.
The conclusion of this entry is found in Part 2. To learn more about the pitfalls your organization may face when implementing CECL, download a copy of FI’s White Paper titled Four Pitfalls to Avoid During CECL Implementation. For more information on FI’s CECL services check out our CECL page.