PNC FINANCIAL SERVICES GROUP, INC.

Our consolidated financial statements are prepared by applying certain accounting policies. Note 1 Accounting Policies describes the most significant accounting policies that we use. Certain of these policies require us to make estimates or economic assumptions that may vary under different assumptions or conditions, and such variations may significantly affect our reported results and financial position for the period or in future periods. The following details the critical estimates and judgments around the ACL:

We maintain the ACL at levels that we believe to be appropriate as of the balance sheet date to absorb expected credit losses on our existing investment securities, loans, equipment finance leases, other financial assets and unfunded lending related commitments, for the remaining estimated contractual term of the assets or exposures, taking into consideration expected prepayments and estimated recoveries. Our determination of the ACL is based on historical loss and performance experience, as well as current borrower and transaction characteristics including collateral type and quality, current economic conditions, reasonable and supportable forecasts of future economic conditions and other relevant factors. We use methods sensitive to changes in economic conditions to interpret these factors and to estimate expected credit losses. We evaluate and, when appropriate, enhance the quality of our data and models and other methods used to estimate the ACL on an ongoing basis. The major drivers of ACL estimates include, but are not limited to:

•Current economic conditions: Our forecast of expected losses depends on economic conditions as of the estimation date. As current economic conditions evolve, forecasted losses could be materially affected.

•Scenario weights and design: Our loss estimates are sensitive to the shape, direction and rate of change of macroeconomic forecasts and thus vary significantly between upside and downside scenarios. Changes to the probability weights assigned to these scenarios and the timing of peak business cycles reflected by the scenarios could materially affect our loss estimates.

•Current borrower quality: Our forecast of expected losses depends on current borrower and transaction characteristics, including credit metrics and collateral type/quality. As borrower quality evolves, forecasted losses could be materially affected.

•Portfolio composition: Changes to portfolio volume and mix could materially affect our estimates, as CECL reserves

would be recognized upon origination or acquisition and derecognized upon paydown, maturity or sale.

We also incorporate qualitative factors in the ACL that reflect our best estimate of expected losses that may not be adequately represented in our quantitative methods or economic assumptions, as discussed below and in the Allowance for Credit Losses section of Note 1 Accounting Policies.

For all assets and unfunded lending related commitments within the scope of the CECL standard, the applicable ACL is composed of one or a combination of the following components: (i) collectively assessed or pooled reserves, (ii) individually assessed reserves, and

76 The PNC Financial Services Group, Inc. –2025 Form 10-K

(iii) qualitative (judgmental) reserves. Through this approach, we believe the reserve levels appropriately reflect the expected credit losses in the portfolio as of the balance sheet date.

Reasonable and Supportable Economic Forecast

Pursuant to the CECL standard, we are required to consider reasonable and supportable forecasts in estimating expected credit losses. For this purpose, we have established a framework that includes a three-year forecast period and the use of four economic scenarios with associated probability weights, which in combination create a forecast of expected economic outcomes. Credit losses estimated in our reasonable and supportable forecast period are sensitive to the shape and severity of the scenarios used and weights assigned to them.

To forecast the distribution of economic outcomes over the reasonable and supportable forecast period, we generate four economic forecast scenarios using a combination of quantitative macroeconomic models, other measures of economic activity and forward-looking expert judgment. Each scenario is then given an associated probability (weight) to represent our current expectation within that distribution over the forecast period. This process is informed by current economic conditions, expected business cycle evolution and the expert judgment of PNC’s RAC. This approach seeks to provide a reasonable representation of the forecast of expected economic outcomes and is used to estimate expected credit losses across a variety of loans, securities and other financial assets. Each quarter, the scenarios and their respective weights are presented to RAC for approval.

The scenarios used for the period ended December 31, 2025 consider, among other factors, ongoing impacts of trade and fiscal policy, including tariffs, on the U.S. economic outlook. Given these factors, growth is expected to slow from current levels in the coming quarters. While recession risks remain elevated, our most likely expectation at December 31, 2025 is that the U.S. economy avoids a recession. We believe the economic scenarios effectively reflect the distribution of potential economic outcomes.