Two committees, one decision
A bank’s risk and strategy committees read the same data and reached opposite conclusions the board could not choose between.
Engagement
Decision framing, 2 weeks, then evidence review, 4 to 8 weeks, then board handover, 2 weeks.
The decision in front of the board
The bank was deciding whether to leave a lending segment. The risk committee read the portfolio data and concluded the bank should go. The strategy committee read the same data and concluded the bank should stay and reprice. Both papers cited the same tables. Neither contained an arithmetic error.
The board had sent the question back for more analysis more than once. More analysis was never going to settle it. The committees were answering different questions. Risk asked whether the segment could be held inside appetite through a downturn. Strategy asked whether the bank’s position in the adjacent segments survived the exit. Each answer was right for its own question. Nobody had put the two questions on one page.
What Plenary did
We wrote both questions at the top of one page, in the words each committee would recognise. Next we separated what the committees agreed on from what each had added on top. The agreed part was most of it: the loss history, the concentration, the funding cost. That left the disagreement in one place, a handful of assumptions about how the adjacent segments behave once the bank stops lending here.
We tested those assumptions against what could be checked, and recorded what could not be checked at all. Then we put one question to the board, the one it was entitled to decide, with the cost of being wrong in each direction. The risk committee’s reading went into the record in full and in its own words. A summary of a dissent is not a record of it.
What happened next
File status
Reopened at the exit review.
When the exit was reviewed
By then the board had voted to leave in stages. The risk committee’s objection sits in the file beside the decision, with the conditions that would call for a faster exit. The review opened at that page a year later. Nobody had to ask people who had since left what they had meant.
“I did not get the decision I argued for. I got my argument recorded properly, which is what the board needed from me.”
Ask which of these is closest to your file.
The gap is in what we can say, not in what we have done.