What should my loss rate be?

I advise a lot of younger lenders.  By which I mean startups, but sometimes the employees are pretty young too.  A question I often get from them regarding a lending product is “What should my loss rate be?”.  And they are looking for an answer like 1%, 4%, or 7%.  And that is a really difficult question to answer in the context of new products.

For well established products with substantial ABS markets, or a bunch of publicly traded players, you can benchmark loss levels fairly well, especially in prime.  For newer and more fintech products, loss benchmarking is much more difficult, particularly outside of the U.S..  And even where there is a known norm, things like pricing, channel, targeting, and user journey can all move losses by 30%.  Sometimes more.  And it’s different for every business.  Some business models can make a ton of money at a 20% loss rate.  Others are dead on arrival at 3%.

So I try to get them to think about losses not as a target number, but as a part of the integrated economics of the business.  A good loss rate is one that gives you healthy and resilient returns that cover the cost of capital and cost to acquire, while allowing you to hit your growth goals.

Let’s talk about those.

• Healthy returns – Interest and fees, less losses, less cost of funds, and less cost are positive and exciting.
• Resilient – The math works even in an economic downturn or if your estimates are off.
• Cover cost of capital – Applying an appropriate hurdle and discount rate.
• Cover cost to acquire – Preferably cover a multiple of it.
• Allow you to hit your growth goals – Have a big enough target market and conversion rate to enable growth.

None of this says anything about what level losses need to be.  Higher losses require higher revenues, sure.  But most higher loss businesses have less price sensitive customers and lower cost to acquire.  The opposite is usually the case for lower loss businesses.  They are more competitive and have thinner margins.

But isn’t a lower loss number always better?   All else equal, sure.  But all else is rarely equal.  I’ve told several lenders that I think their losses are too low.  If you increase your losses from 1% to 2% but through the impact on approval rate, conversion rate, and dilution of marketing costs you could grow the business by 40% (assuming the math above still holds), you absolutely should.

What’s the point of all this?  To try to show that loss levels shouldn’t be a goal in themselves.  They need to be seen as one input into a well understood integrated economic model of the business.

I talk about credit losses a lot, but I prefer to talk about them along with unit economics, pricing power, and product strategy.  If you want to talk more about this, let me know.  It’s literally what I do all day.  And please check out the other great advisors at LF Partners for advice on other topics.