A different case might be a condo development: Private investment (rolled over from selling another property and repaying THAT loan) and loans to buy the land, loans to build, sell 60% of the units, - and there might be enough to repay most of the loan.
In that case, since you already have most of the capital lying around, a balloon payment lets you use other cash but you’re not worried about a large payment.
Seems like a lot of commercial real estate is structured as a balloon loan to people who may not be able to make the lump sum payment, which is probably where the underwriting failure lies. Refinancing is a tool a savvy borrower can use as rates go down, but it’s not a strict expectation that you do that if you get a balloon loan, even on short term loans.
Before the US government got involved, the typical real estate loan was a 5 or 10 year balloon loan that had to be refinanced at the end of the loan period. Large part of them were interest only. So that's exactly the way it used to work.
That blew up during the Great Depression. Then the government stepped in and started subsidizing loans which is where we're at now, the mortgage market in the US is very different than in say Canada or Europe.
Interest-only works for commercial because the building is (hopefully) making money, so some of that can be set aside to handle the last payment. That doesn't work for residential, because single-family homes don't usually make money, so residential homebuyers don't typically have ~$200,000-$2,000,000 in cash at the end of the 30 year period. But if that was feasible, you would see some loan officers trying to win business through offering interest-only mortgages with a balloon payment.
As for why underwriting didn't catch this? Underwriting is a lot like actuaries, they've got a bias towards recent population-level data. So for the last 30 years the rates of commercial default have been predictable, and it's tough to push back to execs on the risk. You're claiming that a potential wide-scale change in the way the U.S. works should impact sales strategy today, and before Covid that risk was very abstract. Very few execs would side with their underwriters over their sales team, and there's been subtle pressure on underwriters to price commercial real estate as though systemic risk isn't real.