First Horizon’s mortgage sale: weigh the upfront loss against future earnings

A loss today and a forecast for next year

First Horizon’s planned mortgage disposal presents two separate questions: how much economic value is being surrendered now, and what evidence would show that the balance sheet becomes more productive afterward? An adjusted earnings figure answers neither question on its own. Investors need a transaction timeline, a reconciliation of the current charge and a clearly defined baseline for the expected benefit.

The company reported a third-quarter loss available to common shareholders of $94 million. Excluding its designated notable items, common earnings were $253 million. That $347 million difference is material even if management considers the transaction strategically sensible. Removing a charge from an operating comparison does not remove its effect on shareholders’ capital.

Keep the three dates separate

The approximately $4.5 billion mortgage portfolio was identified for sale during the third quarter. First Horizon said it entered an agreement in early October and expected settlement early in the fourth quarter. Those are different milestones. The third-quarter results include the held-for-sale reclassification; they do not establish that the buyer had already completed settlement when the results were released.

That distinction matters when assessing the subsequent quarter. A signed contract can reduce uncertainty about the planned exit while leaving execution and final accounting questions unresolved. An investor’s evidence log should separately record the reporting period, the agreement date and a later confirmation of completion. Until that confirmation exists, describe the transaction as pending.

The adjustment is not a cost-free reset

The presentation identifies a $473 million loss associated with the pending mortgage sale and a $45 million provision reversal. It also includes $3 million of restructuring expense and $5 million of Visa derivative valuation expense. Together these produce a $436 million pretax reduction; an $89 million tax effect reduces the after-tax total to $347 million.

This bridge prevents two misleading shortcuts. The largest sale-related charge is not the same number as the total after-tax adjustment, and the provision reversal should not be described as new customer revenue. The useful comparison keeps the components visible rather than treating the entire adjustment as one homogeneous expense.

Nor should investors assume that every excluded item has identical persistence. The future cost of originating, servicing and funding the retained business must still pass through subsequent results. A transaction can improve future profitability while imposing a real upfront loss. Both statements can be true.

Define the forecast before calculating a return

Management’s presentation estimates a 2027 earnings benefit of roughly $45 million after tax, or 4%, relative to the consensus forecast as of October 2. It says this estimate precedes additional benefits from partial redeployment into higher-yielding loans. Those qualifications define what the forecast measures. They do not promise that every future loan will earn the projected return.

Dividing a current accounting charge by an annual earnings estimate would produce an apparent recovery period, but it would not be a reliable transaction valuation. Such a calculation ignores the timing and risk of future cash flows, changes in funding requirements, differences between accounting recognition and cash economics, and the earnings baseline used by management. The forecast also does not guarantee a corresponding stock-price increase.

The better question is whether subsequent disclosures demonstrate the improvement on the stated basis. Investors should separate benefits from retiring expensive funding, changes in the retained portfolio and additional lending. Otherwise an unrelated change in interest rates or credit costs could be mistaken for proof that the transaction achieved its original economics.

Liquidity capacity and capital are different tests

The presentation attributes an approximately 45-basis-point CET1 impact to the mortgage repositioning and shows a 10.3% capital ratio. That transaction impact is not the bank’s total quarter-to-quarter ratio change: the financial summary reports a 12-basis-point net decline. The presentation also shows a loan-to-deposit ratio of about 88%. Lower lending relative to deposits can create room to change asset allocation. It does not mean the capital cost disappeared.

Future lending requires customers, underwriting and acceptable pricing. Replacing an asset with a higher-yielding loan is not automatically an improvement after credit losses and operating costs. Likewise, prospective share repurchases should be evaluated alongside capital requirements and business needs, rather than assumed to be an unconditional use of proceeds.

The next evidence should therefore be concrete: confirmation of settlement, an explanation of any changes to the transaction accounting, the funding mix after completion and the performance of incremental lending. First Horizon has described a plausible route toward a more profitable balance sheet. The investment test is whether the eventual results justify the value surrendered to pursue it.