A bank can lower its efficiency ratio without cutting costs

PNC spent more on operating expenses in the third quarter of 2026 than a year earlier. Its efficiency ratio improved. Both statements appear in the same filing, and both can be true because the ratio measures expenses against revenue, rather than expenses alone.

The bank reported $3.751bn of noninterest expense, compared with $3.461bn a year earlier. Revenue rose from $5.915bn to $6.562bn, while the displayed efficiency ratio fell from 59 percent to 57 percent. A larger business absorbed its operating bill more readily even though that bill was $290m higher, according to PNC's third-quarter results.

An adjusted ratio adds another question: which expenses and which revenue remain in the calculation? PNC's first-quarter acquisition reconciliation provides an unusually clear answer. It removes integration costs from both sides of the fraction, and the difference is larger than the rounded headline percentages suggest.

What does the bank efficiency ratio measure?

The bank efficiency ratio measures operating expenses relative to revenue, usually expressed as a percentage. A lower result means less operating expense for each dollar of revenue under the chosen definition. It does not establish that total spending fell, that credit losses are low, or that a bank is financially safe.

For PNC's reported measure, the calculation is noninterest expense divided by total revenue. The denominator combines net interest income and noninterest income. Personnel, premises and equipment are operating costs; the bank's third-quarter expense table also separately identifies marketing and other expenses.

Interest expense is already reflected in net interest income. Adding it to the numerator while leaving the denominator unchanged would count the funding cost twice in different parts of the calculation. This is why the word noninterest matters, even when a short definition describes the ratio as costs divided by income.

A percentage also loses information. The same ratio can describe institutions with very different expense bills, revenue sources and business sizes. Reading the underlying dollars restores what the percentage removes: whether the bank is spending less, earning more, or doing both at once.

The ratio therefore answers a narrow operating question. PNC reports provision for credit losses separately from noninterest expense, and its first-quarter revenue less noninterest expense equals the disclosed pretax, pre-provision earnings. The remaining revenue is not automatically available to shareholders.

How do reported and adjusted ratios differ in an actual filing?

PNC's first-quarter 2026 filing reports an efficiency ratio of 61 percent and an adjusted ratio of 60 percent after excluding FirstBank integration costs. Recalculating from its disclosed dollar amounts produces 61.12 percent and 59.54 percent. The adjustment changes expenses and revenue, so subtracting the total integration charge from expenses gives the wrong calculation.

The reconciliation appears on page 18 of the April 15 financial results. All dollar amounts in the following table are millions; the percentage calculations use those displayed amounts.

First quarter 2026	Reported basis	Excluding integration costs
Noninterest expense	$3,768m	$3,671m
Total revenue	$6,165m	$6,166m
Expense divided by revenue	61.12%	59.54%
Ratio displayed by PNC	61%	60%

The expense adjustment is $97m. The revenue adjustment reverses a separate $1m integration cost recorded against revenue, raising the denominator by that amount. Together they account for the $98m pretax integration cost identified elsewhere in the release.

This is an accounting classification with a practical consequence. Moving the entire $98m to the expense side would no longer reproduce the bank's reconciliation, even if rounding happened to conceal the error in the final percentage.

PNC labels the measure excluding integration costs as non-GAAP and explains that it is intended to improve comparisons of operating expenses. That explanation identifies management's purpose; it does not erase the expenses from the reported accounts. The two columns describe different views of the same period.

Why does rounding change the apparent size of an adjustment?

Rounding can hide a meaningful part of the gap between reported and adjusted efficiency ratios. PNC displays its first-quarter figures as 61 percent and 60 percent, a one-point difference. Division using the underlying amounts gives a gap of about 1.58 percentage points, because each displayed ratio is rounded separately.

The distinction between percentage points and percent matters here. Subtracting one ratio from another measures a percentage-point change; describing that subtraction as a percent reduction would imply a different calculation with its own comparison base.

Precision also has limits. The underlying expense and revenue figures in PNC's table are rounded to millions, so the two-decimal results are reconstructions from published amounts, rather than access to the bank's unrounded accounting system. Their purpose is to expose the calculation, not to suggest an audit-level precision the source does not provide.

This matters when comparing banks that publish different numbers of decimal places. A one-point change on a whole-percentage presentation can conceal different movements within the rounding intervals. The source dollars and the definition are more informative than adding decimal places to a rounded headline figure.

The same discipline applies over time. Recalculating PNC's first-quarter 2025 expense of $3.387bn against revenue of $5.452bn gives approximately 62.12 percent. Its first-quarter 2026 reported ratio improved to approximately 61.12 percent while the expense total increased, another example of revenue growing faster than costs.

Does excluding merger expense create a like-for-like comparison?

Removing integration expense does not by itself create a comparison of the same business before and after a merger. PNC began consolidating FirstBank on January 5, 2026. Its adjusted first-quarter results still include the acquired business's ordinary operations, so the comparison combines a cost adjustment with a change in the business being measured.

PNC's acquisition disclosure says FirstBank's financial results entered consolidated operations from the closing date. Its expense discussion separately identifies acquired operating expenses and integration expenses. Those categories should not be collapsed into a single merger-cost subtraction.

Integration costs relate to bringing businesses together. The acquired business also has an ongoing operating bill and produces revenue. Excluding the first category leaves the second within the consolidated result, which is why an adjusted percentage is not automatically an estimate of organic efficiency improvement.

The distinction prevents an unsupported conclusion about merger success. A smaller adjusted ratio can show that the combined entity absorbs costs better under management's definition, but it cannot alone establish how the acquired bank would have performed independently or how the buyer would have performed without the transaction.

Those counterfactuals require additional evidence. The reconciliation is valuable precisely because it shows a narrower, observable fact: which recorded items the company removed, where they appeared, and how the resulting ratio was constructed.

What do PNC's newer results reveal about expense reductions?

PNC's third-quarter 2026 expense decline looks different on reported and adjusted measures. Reported noninterest expense fell $347m from the second quarter, while expense excluding integration costs and significant items fell $113m. The $234m difference comes from a smaller amount of excluded expense, rather than an additional reduction within the adjusted expense base.

The October 15 reconciliation supplies the bridge. Second-quarter expenses included a $140m contribution to the PNC Foundation and $121m of integration expense, for $261m of exclusions. The third-quarter expense exclusion was $27m of integration costs.

Expense measure	Second quarter 2026	Third quarter 2026	Change
Reported noninterest expense	$4,098m	$3,751m	-$347m
Excluded expense items	$261m	$27m	-$234m
Adjusted noninterest expense	$3,837m	$3,724m	-$113m

Neither comparison is a substitute for the other. The reported column captures the full expense movement in the accounts; the adjusted column isolates the company's chosen operating comparison. Calling the entire $347m decline a reduction in recurring operating costs would overlook the bridge between them.

The revenue side changes as well. PNC's adjustments remove $218m of second-quarter noninterest income but add back $34m in the third quarter. The disclosed items include Visa-related gains and valuation adjustments, a second-quarter securities sale loss, and integration costs recorded against revenue.

That reversal is why an expense-only adjustment cannot stand in for a fully adjusted ratio. Removing income reduces the denominator and tends to raise the ratio, while reversing a cost recorded against income raises the denominator and tends to lower it, holding other amounts constant.

Using PNC's adjusted expenses and adjusted noninterest income, with its reported net interest income unchanged, produces the following analytical comparison. These are Money & World calculations, not separately reported PNC adjusted efficiency ratios for these quarters.

Calculation from disclosed amounts	Second quarter 2026	Third quarter 2026
Reported expense / reported revenue	59.61%	57.16%
Adjusted expense / (net interest income + adjusted noninterest income)	57.64%	56.46%

For the second row, adjusted revenue is $6.657bn in the second quarter and $6.596bn in the third. The direction of improvement survives both calculations, but its size changes. The company-specific exclusions must travel with the percentage wherever that percentage is compared.

Why is the FDIC industry ratio a different benchmark?

The FDIC's efficiency ratio subtracts amortization of intangible assets from noninterest expense before dividing by net interest income plus noninterest income. PNC's reported formula divides noninterest expense by total revenue without that subtraction. A comparison with the FDIC therefore requires matching definitions as well as periods and reporting entities.

The distinction is explicit in the FDIC's second-quarter 2026 Notes to Users, across pages four and five. It is easy to miss when both measures appear under the same familiar label.

The FDIC's Quarterly Banking Profile reports 55.38 percent for all insured institutions in the second quarter. That is a banking-industry statistic. PNC's release describes the consolidated financial services group, so attaching the industry figure beside the company's ratio without qualification would combine a formula difference with a reporting-scope difference.

Size groups also differ substantially. Table III-A, on printed page nine of the profile, shows the following second-quarter results under the same FDIC definition.

FDIC asset-size group	Efficiency ratio
Less than $100m	76.47%
$100m to $1bn	64.64%
$1bn to $10bn	57.43%
$10bn to $250bn	56.59%
Greater than $250bn	53.74%

These groups provide context, not a causal estimate of what growing a bank would do to its costs. Institutions in different size bands also have different businesses. The table does not hold those differences constant or follow an otherwise identical institution as it grows.

The FDIC separately reports a 60.52 percent ratio for community banks in Table IV-B. A community-bank comparison and an all-industry comparison consequently start from different populations, even before introducing a company's own non-GAAP adjustments. There is no single percentage that settles whether every bank is operating well.

The period needs equal care. A quarter measures the revenue and expenses recognized during that quarter; a full-year ratio combines the year's amounts before dividing. Averaging quarterly percentages without considering the revenue attached to each quarter would give equal weight to periods that contributed different amounts of business to the annual accounts. The relevant annual calculation uses annual expenses and annual revenue under one consistent definition.

The comparison period matters. PNC's third-quarter expense bill fell against the immediately preceding quarter but rose against the same quarter a year earlier; neither statement contradicts the other or replaces the period label. A headline saying only that costs fell would lose information present in the company's own table. Keeping the dollars and both column headings beside the ratio makes that ambiguity visible before it becomes a conclusion about performance.

What remains outside the efficiency ratio?

The efficiency ratio leaves credit-loss provisions and income taxes outside its operating-expense comparison. It also does not measure capital adequacy, liquidity or the quality of customer service. A lower ratio establishes a relationship between selected costs and revenue; it cannot establish a complete verdict on profitability, resilience or the merits of a bank's shares.

PNC's first-quarter statement makes the earnings limit concrete. Revenue of $6.165bn less $3.768bn of noninterest expense leaves $2.397bn of pretax, pre-provision earnings. Its Form 10-Q separately reports a $210m credit-loss provision and net income of $1.772bn, demonstrating why the operating remainder is not the final profit figure.

Net interest margin answers another question again. It relates net interest income to earning assets, while the efficiency ratio relates operating expense to revenue; the two can move differently because their numerators and denominators describe different parts of the accounts. Our analysis of bank net interest margins examines that separate measure and its industry aggregation.

For a comparison to be reproducible, the record needs the reporting entity, period, expense definition and revenue definition. Any excluded items need their own bridge to the reported amounts. Once those are visible, the ratio becomes an interpretable calculation instead of a grade detached from its workings.

What are the short answers to common questions?

These measures answer different questions. Lower generally means less operating expense per revenue dollar under a consistent definition, while adjustments exclude specified items and net interest margin uses earning assets as its denominator. None of these measures alone establishes the safety or investment value of a bank.

Is a lower bank efficiency ratio better?

It indicates less expense per dollar of revenue under the same definition. It does not necessarily mean lower total spending, and it does not capture every cost or risk in the business.

Are reported and adjusted ratios interchangeable?

No. PNC's first-quarter adjustment removes expenses and reverses a revenue reduction. Both sides of that reconciliation are needed to reproduce its adjusted result.

Is 60 percent a universal dividing line?

No. The FDIC's second-quarter size-group results span both sides of that figure. Period, business mix, reporting scope and expense exclusions affect what a comparison means.

Does the ratio measure cash spending?

It uses reported income and expense measures. The FDIC's explicit removal of intangible amortization is one reason to inspect the accounting definition before describing the result as cash cost.