We’ve been doing this for pretty much approaching three decades now, at the lower end of the market. And the marketing there is direct marketing, stimulus response, very information based. And so, the marketing machine that we’ve built, which has been enhanced by technology here is definitely leaning into that opportunity. And I do want to say that the — we continue to get good traction in the subprime and prime parts of the marketplace, even as we certainly relative to 10 years ago, have a lot more marketing going on at the top of the market. So, there’s quite a bit going on, and we feel good about the traction there.
Moshe Orenbuch: Got it. Maybe just to kind of — as a follow-up, what would it take for you to see, either in the portfolio or in the market for you to do less marketing?
Richard Fairbank: Yes. The way this tends to happen is it happens in one little segment, one micro segment at the margin in response to things that we see going on there. I use this phrase a lot, trimming around the edges. And you’ve heard me use that for many, many years. And this is something that we always do or something we’re expanding around the edges. The net feel of these days is we’re doing more trimming around the edges than expanding around the edges, but it is — so it’s less about at the top of the house saying, we just believe we should do — obviously, at the top of the house, we’re looking at all the macro things, but we’re linking what we see in the macro level to what we’re seeing right there in real time or the earliest we can see from our credit metrics.
And then, using the technology we’ve continued to invest so heavily in to have a more and more granular diagnosis. And at an earlier time than ever before diagnosis of where anything is deviating from the trajectory that we would expect. And then one is sort of the diagnosis of deviation. And the second thing, of course, is trying to get sort of a root cause, understanding of what may be driving that. And this is something that we continue to put a lot of energy into and it has led us to trim some — there are some things that we have seen degrade a fair amount around the edges. They’re fairly small in the overall size of things, but we’re certainly glad when we see them. And then, what we try to do is to link data that we see to behavior that — excuse me, to sort of an explanation of what’s going on from a customer and credit dynamic to be able to be — it makes total sense.
So therefore, as things play out, it’s less likely — and let’s say, we go more into a downturn, it’s less likely on the card side that you would see a big pullback. The kind of things you’d see is more trimming around the edges, more reduction of the credit lines that are given, and that would be more how it would play out.
Operator: John Pancari with Evercore ISI.
John Pancari: On the — regarding the reserve build in terms of the drivers of the reserve build this quarter, I know you cited loan growth, you cited the macro backdrop, and you sited credit normalization. Is there any way to help parse out how much of the build of $1 billion is attributable to loan growth versus macro versus credit normalization?
