Design a bank deal for the wait, not the announcement

I believe the most expensive months of a bank acquisition are the ones between signing the merger agreement and receiving the last regulatory approval, and that most buyers write their agreements as though those months will be few. My position, after two acquisitions as a buyer and twenty-three transactions as an adviser, is that a buyer should plan for the long end of the review, loosen the restrictions it places on the seller during the wait, and answer the questions reviewers will ask before the application is filed rather than after.

Approval times for US bank mergers have varied a good deal over the past decade, lengthening in some periods and shortening in others as the agencies' policies changed. I do not try to forecast them. A bank buyer needs approval from the federal agencies that supervise it, and often from state regulators. The review considers competition, the financial and managerial strength of the combined bank, its record of serving its communities and the effectiveness of its compliance programmes. Any of these can extend a timetable, and some depend on facts at the seller that the buyer only partly controls.

Eleven months of standing still

In February 2019 we agreed to acquire Dunmarrow Bancorp, a US$5.6 billion bank in Salt Lake City. Our plan assumed closing in five months. During the review a supervisory finding about Dunmarrow's anti-money-laundering controls had to be remediated, and the deal closed in January 2020, eleven months after announcement.

The merger agreement had been written for five months. Like most, it restricted the seller's conduct before closing: Dunmarrow needed our consent to hire above a certain level, to make loans above a modest size outside its existing policy or to change its deposit pricing. Those terms protect a buyer from a seller that might otherwise alter the business being bought. Over eleven months they also froze a bank in a competitive market. Competitors told Dunmarrow's clients, accurately, that it could not commit to new credit without approval from a buyer in Phoenix, and told its bankers that their future was uncertain. Twenty-three of sixty-one commercial bankers left before closing and deposits fell 12%. The cost savings in our model arrived on schedule. Much of the revenue did not.

Let the seller keep competing

When we agreed to acquire Kessling Bancshares in July 2024, we wrote the agreement for a twelve-month review and loosened the interim covenants on purpose. Kessling could hire bankers within an agreed budget, renew and extend loans within its own credit policy and set deposit rates without asking us. We kept consent rights over large new exposures, capital actions and anything that would change its risk profile materially. We also agreed that Kessling would run its own retention awards for its commercial bankers, funded within the purchase price, so that the people deciding who mattered were the managers who knew the clients rather than our integration office in Phoenix.

This is uncomfortable for a buyer. Our chief credit officer argued that we would inherit loans we would not have made, and he was right: we later added about US$14 million to loss provisions on loans Kessling made under those terms. There is also a legal discipline to respect. Until closing the two banks are separate competitors, so a buyer cannot direct the seller's pricing or lending, and integration teams must handle client and pricing information under strict rules about what may be shared. Looser covenants mean trusting the seller's management more, not supervising it more closely.

The return was visible. Kessling's deposits fell 3% between announcement and closing, against Dunmarrow's 12%, and four of its fifty-two commercial bankers left, against twenty-three. In the economics of the deal those differences were worth several times the extra provisions.

Settle the hard questions before the filing

Reviewers ask predictable questions. Will the combined bank hold too large a share of deposits in some local market? Has each bank served the low- and moderate-income communities where it operates? Are the compliance programmes adequate for the larger organisation? A buyer that waits for these questions to be put to it negotiates its answers under time pressure, with the seller's bankers watching the calendar.

For Kessling we identified two New Mexico counties where our combined deposit share was likely to draw competitive objections, and offered to divest seven branches there before we filed. Our chief financial officer argued that we were conceding a negotiating position and US$610 million of deposits we might have been allowed to keep. We may well have been. I judged that a buyer which has already solved the obvious problem is reviewed differently from one that has not, and that the months saved were worth more than the branches.

We also published a five-year community lending and investment plan with the application, and here I made a mistake. We wrote the plan with our own staff, and a coalition of community organisations in Albuquerque submitted a comment letter arguing, with some justice, that nobody had asked them what their neighbourhoods needed. Meeting them added about six weeks to the review and produced a better plan. I would now hold those conversations before announcement, under confidentiality where necessary, rather than in response to a letter.

There are serious arguments against my approach. A buyer that pre-commits a divestiture may give up value a reviewer would never have demanded. Looser covenants expose the buyer to a seller that behaves badly during the wait. A plan built for a long review may even invite one. I accept all three risks. My view is that buyers consistently underprice the cost of a frozen seller, because it never appears as a line in the deal model. It appears a year later as weaker revenue, and by then everyone calls it integration difficulty.

I would put three questions to any board approving a bank acquisition. What happens to the seller's deposits and bankers if approval takes twice as long as management expects? What may the seller do during that period without the buyer's consent, and is it enough to keep competing? Which of the reviewers' questions could we answer before we file? If management cannot answer the first, the price is probably wrong.