When Mortgage Rates Move on Paper but Not in Pricing Behavior

Completed bank mergers can create a visible increase in a lender’s average mortgage rate even when the lender has not raised the price offered to a comparable borrower. The question in the Federal Reserve study is therefore more precise than whether the average moved: did the merger change pricing relative to other lenders in the same market, after accounting for the different portfolios that were combined? This chapter uses the large–community pattern to answer that question and then places fees, spreads, and the absence of proven savings in the same frame.

The distinction matters in practical terms. A mortgage rate is attached to a loan with a particular risk profile, amount, duration, purpose, and borrower. When two banks merge, the post-merger institution may report a different mix of those loans. If the target had charged higher rates to its customers before the merger, simply bringing those loans into the acquirer’s books can raise the combined average. That movement is not irrelevant, but it does not identify a new pricing strategy by itself.

The large–community pattern looks worse than it is

The study separates three relationships: community banks merging with other community banks, large banks acquiring community banks, and large banks merging with other large banks. The categories matter because the institutions serve different segments and use different lending and loan-retention models. A single average across all mergers could blur the difference between a small bank absorbing a peer and a large institution taking over a relationship-focused local lender.

Community–community mergers provide the least dramatic rate pattern. Before completion, acquiring community banks charge rates about five basis points below local controls, while target community banks are about three basis points below controls. Those differences are modest level differences. Across the eight post-merger quarters, the rates of acquirers, targets, and controls remain stable. There is no clear divergence suggesting that the new community-bank combination used the transaction to reprice mortgage credit systematically.

Large–large mergers also look relatively stable. Apart from the first two pre-merger quarters, the acquirer, target, and controls offer broadly similar residualized rates. Post-merger rates remain largely unchanged, with a modest decline of roughly three to four basis points in the last four post-merger quarters. The point estimates are not a sustained upward movement of the kind expected if the combined institution systematically exploited greater mortgage market power.

The large–community case is more visually alarming because its starting portfolios are different. Before completion, the large-bank acquirer charges slightly less than local controls. The community target, however, charges rates roughly 20–30 basis points higher than both the acquirer and controls. Once the two portfolios appear under one institution, the average rate rises modestly and converges toward control rates. A reader looking only at the combined average could reasonably suspect that the larger bank raised prices after eliminating a local competitor.

But the same pattern follows from portfolio arithmetic.