Sure. Matt, I'd be happy to provide some color around that. So I'll talk about auto, although there are a lot of similarities between auto and comp. Certainly, claim counts are down. But the majority of our premium is exposure based, meaning the premium is based upon sales receipts, miles driven, for work comp, it's payroll. So we really have to wait and see what -- when we get the exposure reporting from our insurers and through premium audits to see what that denominator looks like. It's -- while counts are down, if you have a fixed denominator, you can perhaps declare an early victory around frequency and on the other hand, with the denominator that is somewhat variable depending upon where ultimately the premium settle in, it's really too early to declare that frequency is down.So for instance, if the denominator is miles driven, we have to see what those miles driven were and take those claim counts and divide it by miles driven. So it's a little different perhaps than, say, personal auto, where the denominator might be a fixed annual premium, and you can -- with a pretty high degree of confidence, because the counts are down say that the frequency is down. But with a denominator that is one that does fluctuate depending upon the exposure base, it's a little bit harder to be precise about where ultimately, frequency will be. But certainly, claim counts are down for auto. Workers' compensation, very much the same thing. There -- the denominator when we look at frequency is payroll for the most part, we look at it a couple of different ways. But to the extent that we have to see what payrolls are ultimately reported. What premium audits look like, it's a little hard to tell whether those lower claim counts right now are an absolutely clear indicator that frequency is down.Certainly, claim counts are down. I will just add one comment, not specific to what you asked, but also there's the question about, well, how do COVID-19 claims impact things. And there, I would say that there's some pretty good analysis out there by NCCI and WCIRB in California that talks about their observations and analysis on frequency. And to make it a long story short, what it really indicates is that there's a lot of things that are driving frequency down. And on the other hand, there are some things that are driving frequency up. So with respect to COVID-19, while there might be some uptick from that, there's arguably a downward pressure on frequency from other things such as auto accidents, for example. So a bit of a moving target, but hopefully, that gives you some more color around how we're looking at frequency when it comes to auto and comp.