What is a credit mark-to-market model?
A credit mark-to-market model is a special kind of math used by banks and other money lenders. They use it to figure out how much money they might lose if someone they gave a loan to has money problems later on.
How the model works
The model looks at the chances that the person or company who got the loan (called the “counterparty”) will have a harder time paying back what they owe at different times in the future.
There are two main things that can cause the bank to lose money according to this model:
- If it looks like the counterparty is having money troubles, other banks won’t want to lend to them as much. This makes it more expensive for the counterparty to borrow money. The difference between what it costs them and what it costs someone with good credit to borrow is called the “credit spread.” When the spread gets bigger, it means the loan is worth less to the bank who gave it.
- The other way the bank could lose money is if the counterparty has such big money problems that they can’t pay back the loan at all. This is called a “default.” In the model, a default is the worst case scenario.
Figuring out the losses
To use the mark-to-market model, the bank will look at the counterparty’s finances and try to estimate the likelihood of these two things happening at various points in the future.
They’ll consider stuff like:
- How much debt the counterparty already has
- How much money the counterparty is making
- What the counterparty’s assets are worth
- What’s going on in the overall economy
The bank then plugs these guesses into the math of the model to calculate the expected losses from both a wider credit spread and a potential default.
Why banks use this model
Banks like this model because it gives them a more complete picture of the riskiness of their loans than just looking at the chance of a default. By considering the possibility of credit spread widening too, they can get a sense of potential losses even if the counterparty doesn’t end up defaulting.
Accounting rules
The other reason a lot of banks use mark-to-market models is because of accounting rules. The rules say that banks have to report the current value of their loans based on market conditions, not just what they’re owed.
Using a mark-to-market model helps them do this. It lets them show that a loan has lost value if the counterparty is looking riskier, even if they haven’t missed a payment.
Limitations of the model
Even though a lot of banks rely on these models, they’re not perfect. Here are a few potential issues:
It’s all guesswork
At the end of the day, the model is only as good as the assumptions that get fed into it. If the bank guesses wrong about the chances of the counterparty running into money trouble, the loss predictions will be off.
Unexpected events
The model looks at the counterparty’s current financial situation to make predictions. But unexpected things happening in the future could make those predictions wrong. Think of something like a global pandemic shutting down the counterparty’s business. The model couldn’t predict that.
Interconnections
The model mostly considers each counterparty in isolation. But in the real world, their financial health could be tied to other companies, or even the whole economy. A big event that hurts a lot of companies at once could make the model’s predictions too optimistic.
