1. Black-Scholes was originally designed to describe the price of assets like equity (which tend to move proportionally to their price). So applying to rates was already using it for something it was not meant to. And everybody knew it. A lot of people had developed normal models instead of lognormal models.
2. Black Scholes has been broken for years. Pricing with a volatility smile has been around since the 90s. It's a way bigger hack than what's needed to make black scholes work for rates.
3. If the problem is that R can go below 0, you can just model R+2%, and voila, you get a process that's never negative
4. The negative rates problem dates back to crisis, so of course people have already found ways to make it work.
Was it painful? Well it required tweaking the models for sure. But this is neither high-flying math nor out of "business-as-usual" activity.
1. Black-Scholes was originally designed to describe the price of assets like equity (which tend to move proportionally to their price). So applying to rates was already using it for something it was not meant to. And everybody knew it. A lot of people had developed normal models instead of lognormal models.
2. Black Scholes has been broken for years. Pricing with a volatility smile has been around since the 90s. It's a way bigger hack than what's needed to make black scholes work for rates.
3. If the problem is that R can go below 0, you can just model R+2%, and voila, you get a process that's never negative
4. The negative rates problem dates back to crisis, so of course people have already found ways to make it work.
Was it painful? Well it required tweaking the models for sure. But this is neither high-flying math nor out of "business-as-usual" activity.