Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Martin Ellingsworth on Outstanding Claims with the Price Writer Podcast Ep 20
12 Plays1 month ago
Welcome to our latest episode where we return to the world of insurance pricing this time with our special guest, Martin Ellingsworth, a seasoned expert in data analytics and insurance pricing. Hosted by Jeremy, this conversation covers everything from Marty's unique journey into the insurance industry to the current trends and challenges in pricing and risk management. Key Takeaways: -Marty’s journey from military operations research to becoming an insurance data analytics expert. -The evolution and importance of accurate data in improving insurance pricing. -The impact of external factors, such as used car values and economic shifts, on the insurance industry. -The role of technology and predictive analytics in shaping the future of insurance. -Insights into the personal auto insurance market and the challenges it faces. Have thoughts or questions about this episode? Leave a comment below or connect with us on social media. Don’t forget to like, subscribe, and hit the bell icon to stay updated with our latest content.
Recommended
Transcript
Introduction and Weather Talk
00:00:00
Speaker
Hey, Marty, and welcome to the show. Hi, Jeremy. Thanks for having me on. Thanks for being on today. you said it's a foggy day there in California. theres I'll save you the view out the out the window to see all the gray.
00:00:14
Speaker
but the The June gloom here in the coastal Southern California is you know traditional in this in the summertime. guess it keeps it cooler. We've been having autumn in summer at the moment in here in London, so not much better.
Career Journey: Military to Insurance
00:00:30
Speaker
So the first thing i want to ask you, Marty, is how did you get where you are today? Like most people that find themselves in insurance, it feels like it was by accident.
00:00:41
Speaker
I started my career in the military. When I left, my advisor, who's a lifetime mentor, says, Marty, you're an operations research you know ninja.
00:00:53
Speaker
but You find new data, you find deeper meaning. And if you go find giant piles of money, finding little slivers of improvement will give you a really good career.
00:01:04
Speaker
I took that advice and I went to healthcare and unfortunately healthcare in the nineteen ninety s looks a lot like it does today, but back then it was a right for mergers and acquisitions. And every place I went would either get merged, purged or gobbled or somehow reorganized.
00:01:25
Speaker
And so after about a decade of ah having new bosses every yeah other quarter, it seemed. It was like, where else could I go that has big money but better data?
00:01:37
Speaker
And as I was at one of my last roles, I found workers' compensation as the the entryway from healthcare to casualty in PNC.
Advising on Data Analytics and AI
00:01:48
Speaker
and the yeah And it's been 25 years of not looking back since then. PNC is much better behaved. way more contractual, has a lot of statutory complexity, but but by and large, the data are better simply because people in the US s aren't changing health plans every year on an annual basis as their employers shop for better cover or a better service or better price.
00:02:19
Speaker
And what do you currently do? So currently I'm doing advisory work for a variety of companies, looking at data analytics, AI strategies. And oftentimes I'm labeled the chief grief officer because I'll come in and say, your data is really not as good as you think it is. Probably half of it is really ah great examples of your best decisioning and the rest of it is garbage, but you don't really know which is which. Starting with all of your data is is usually a terrible strategy.
00:02:50
Speaker
yeah That turns into from chief grief officer to chief causing grief officer, ah where it's now that you know what you need to change, let's go change something.
Insights on Insurance Pricing
00:03:01
Speaker
And when it comes to pricing, ands it's an ever-evolving race to more accurate pricing. I've been involved with several lines of insurance.
00:03:14
Speaker
I started in workers' compensation back with ah Risk Data Corporation. And there, the idea was better. micro-reserving. So if you know your claim costs, then you should do a better job of knowing your pricing.
00:03:30
Speaker
And that unique American tradition called Thanksgiving is is one of my favorite jokes around insurance, where once a year, usually around budget time or bonus time, ah claims and underwriting and get together and blame each other for a terrible year or slap each other on the back for a great year.
00:03:51
Speaker
It's a fun family table to be at, but the, the process of learning from your operational costs and identity payments to better improve your pricing accuracy, you know, is a ah continuous and virtual, you know, a cycle of improvement.
00:04:08
Speaker
So my. yeah When you look at the U S market, the largest line by far is personal auto. I think it weighed in at the 2000, 2024 first quarter at about billion. So at a pure loss ratio for personal auto,
00:04:30
Speaker
that means ah hundred billion dollars in run the business and profitability Just let that sink in again A hundred billion dollars is a lot of responsibility for like non-indemnity payment.
00:04:44
Speaker
And the, so there, I think the, it's a noble cause for risk transfer where, know, without risk or risk transfer, people couldn't go about their daily lives.
00:04:54
Speaker
but They couldn't make financial plans. They couldn't expect to be fairly treated ah if there was for God, God forbid an accident, and especially on the roadway.
00:05:07
Speaker
The idea of improving people's lives through risk transfer and improving risk transfer through better pricing is is like my happy place of using data analytics and advanced intelligence for understanding how can we do a better job of understanding the risk and artist understanding the value of the risk.
00:05:30
Speaker
And the two of those together are what makes, yeah I think, risk transfer work. That's right. That is right. So you left health care and then arrived in casualty.
00:05:42
Speaker
What was your journey like through casualty and insurance? You know, it's not been a straight line. um And some people say I've been but more lucky than I've been good. And I'm okay with that characterization.
00:05:56
Speaker
Yeah, I started with Allianz and moved from Southern Cal to Northern Cal and worked at the Novato campus of Fireman's Fund. That institution does isn't even in the market anymore. They've sold the brand. They sold the property. Allianz had bought Fireman's Fund, and then it spun it down and spun it off.
00:06:14
Speaker
So you, I think Ace and Chubb now own what's left of the branding, but very little paperwork left and and no signage. They even gave the truck back, the old fire truck back to the historical society. So that was disappointing.
00:06:29
Speaker
Watching a hundred-year-old company yeah disappear from the market has been... a bellwether for me of if you don't have pricing accurate, and really risking the whole enchilada.
00:06:43
Speaker
and and There might be a multiple ways of looking at it. I think the entertainment book did well and got sold off. Commercial mid-market did well, got sold off. Personal lines was doing okay, got sold off, a lot of high net worth.
00:06:56
Speaker
But their workers' compensation book really just got slaughtered in a California overrating. And the when you're running a, I don't know, 240 loss ratio in a major commercial line,
00:07:10
Speaker
Wow. That's, uh, and that was, that's my characterization. I don't know where the final thermometer landed on that whole process, but it was ah lesson to be learned about watching what happens in the regulated market.
00:07:24
Speaker
ah When open pricing comes to town and when a claim cost controls relaxed, you can have pricing go down and claims go up and all of a sudden you're in trouble.
00:07:34
Speaker
Yeah. Yeah. I thought that was the only time that I'd ever see that again. And and behold, COVID comes along and used car values enter. yeah And everyone bets that cars go down in value.
00:07:47
Speaker
And all of a sudden, cars became popular because there's no new cars coming out. So all the new used cars are being bought. yeah And that inflection of ah change in value at risk totally blew up the personal auto market in the U.S., probably globally.
00:08:04
Speaker
Just because the cost of ah replacing you know a light kind and quality vehicle could have be 50% more than you expected
Impact of Used Car Values
00:08:15
Speaker
or more. And that's a shock to the system. And when your pricing systems, like here in the US, are tied to an embedded evaluation structure, you don't have the the resilience to be mark to market. And it's weird, it's like the only line that's like that.
00:08:35
Speaker
Homeowners changed to Cover-J back in the 90s after Hurricane Andrew. Scheduled valuable possessions have always been that way. Even agreed value for risk and in commercial auto is like an old thing.
00:08:50
Speaker
So that, but anyway, not to get too digressed, but you the once in a lifetime punch in the face happened twice for me. Once in unregulated California workers' comp and then once in COVID surprise used car values. And the yeah ah between those two goalposts of my career,
00:09:11
Speaker
In between, it's been a ah journey of predictive analytic. I was one of the first people to bring in text analytics to insurance. And you can see everywhere you go now, and text is data, pictures are data, geospatial is data.
00:09:27
Speaker
That type of analytical framework was what I brought with me from the military where I was in the the science labs. helping to instrument the fighter cockpits of the future and all the man machine, human machine interfaces and, and pilot safety systems.
00:09:45
Speaker
Now we have driver safety systems. Spoiler alert for that one. Having better situational awareness around what's happening outside the vehicle is the only thing that matters.
00:09:57
Speaker
correct Yeah. we We have spent billions of dollars trying to give people information about how the machine is working. that's they know't it That's not necessary to not hit stuff. Yeah. And for the most time for the most part, at least with a ah vehicle, you're you're not stressing it. its It's doing 50, 60 miles an hour.
00:10:21
Speaker
it comes to a slow, calm stop. It comes to a slow, Maybe a fast acceleration, but by and large, you're in a two-day work two-day world operating on a level plane where with engineered surfaces.
00:10:39
Speaker
No big surprises that cars should stay functioning. So back to pricing. When you think about the average household in America being one of 120 million households with either one, two, or three cars or more,
00:10:58
Speaker
That means that you could have three, four or $5,000 of auto insurance or more expenses, depending on how many drivers in the household. So the yeah it's a real relevant thing for everyone. And it's one of the only things you have to purchase. Yeah. It's like people talk about death and taxes, death taxes and auto insurance. and um you It's a compulsory line.
00:11:23
Speaker
Yeah, it is. You have to if That's a good one.
Evolution of Vehicle Pricing Systems
00:11:28
Speaker
ah You mentioned the increase in the used car prices, and that has caused a problem here in the UK as well. We've had quite heavy auto insurance inflation, but we've also had issues where customers have received lower counts on write-offs, losing their vehicles, because Many people say it's because the tables and the information that we use is not actually keeping up to date with what's happening in the real world.
00:11:59
Speaker
Isn't that a shame that you could probably Google on your phone the value of your car and get a ah better spot estimate of what someone down the street would pay for it? than a table and a system or a book or a mainframe that was loaded a year ago using two year ago rates with a factor for deprecation.
00:12:21
Speaker
So there's a, there's a lot on the table still to clean up what's available in the real time world versus the mainframe world, looking at, at, you know, tables and a archaic rating pricing system.
00:12:37
Speaker
So there's a, There is a lot of tradition that was built in the U.S. maybe in the 40s and 50s when there was 25 million cars and no one had enough credibility to really know what's going on and there really wasn't computers or internet.
00:12:54
Speaker
Now that's not
EV Market Changes and Insurance
00:12:55
Speaker
true. The list price of a vehicle for sale is pretty close to the sale price these days. Gone are the days where we had to enforce mandatory window stickers because people didn't know what they were getting under the hood without a label that says, here are the features that are on this specific vehicle.
00:13:16
Speaker
yeah And in the early 80s, 1981, they started to put the vehicle identification number, like your cell phone MEID or any other type but of unique product identifier.
00:13:31
Speaker
Now when they build it from the factory, exactly what the factory put on it. Yeah. then And you can price that in the market. So it's, I'd say that that type of build sheet level beta, when you use your phone and the preview view of who's calling, the telephone company always knew who that was.
00:13:53
Speaker
Yeah. They just made it available to you for you to view it. And it's not a color ID. Yeah. So with a vehicle ID, and you can actually get but the features, the values, maybe even the current location, all sorts of things you can know. But the the value of vehicle is just a characteristic.
00:14:13
Speaker
It doesn't have a big engine or a small engine. Yeah, absolutely. And Marty, what would you say is your mission for insurance pricing?
00:14:24
Speaker
but First, get back to work. yeah the yeah I left JD Power as they were reworking and getting ready for their next big yeah ah transaction about this time last year.
00:14:35
Speaker
So it's been a great What do you would call it? A year at sea? I've traveled the world. I've done some advisory work, but I'm swimming back to shore going, I'm still young enough to have a job. So ah whether it's advice or real job, the idea of getting back to pricing or claims or both, like that that table that says, yeah you need to price to your expenses and you need to manage your expenses.
00:15:01
Speaker
The real world. And that's the big thing. So I think my my my future is have pricing be more reflective of the real world. Because in a and a cost plus system of compulsory ownership of the of ah vehicle liability insurance, there's a there's not a lot of incentive to keep the cost down.
00:15:22
Speaker
yeah There, there's more, there is more incentive to keep alignment, but the idea that the claim system is tied to real world claims costs and ah market prices yeah and underrating can be too.
00:15:35
Speaker
And that the ah structure of the contract can have more accuracy. as we've proven with Real Books of Business at the you know most recent rate-making conference back in March. yeah We took the theory, talked about how we got to where we were in rate-making.
00:15:54
Speaker
We took Real Data with Willis Towers Watson at the helm with one of their customer portfolio, Books of Business. yeah And when we applied the the data for pricing in time with the market value, they were able to show dramatically what's going on with the loss cost curves of used value.
00:16:12
Speaker
And, uh, and how does that fit where the, with the future of pricing? And the, I think for a lot of people, the big surprise was before the COVID shock to the world system, cost of vehicles really behave differently than they are behaving right now.
00:16:32
Speaker
Yeah. Yeah. Now it's 2024. And three years ago was when a lot of the crazy stuff in the world was happening with COVID.
00:16:44
Speaker
That's the same time almost to the months that plants shut down. fleets stopped leasing ah yeah all sorts of ah production and supply issues and became material.
00:16:57
Speaker
And now we're going into the, what I'll call the ghost fleet era where yeah all those cars not produced are also now not available in the used car market.
00:17:08
Speaker
And typically the three year release return is like the most desirable used vehicle. because it had to be held under maintenance.
00:17:18
Speaker
It was under a financial contract. People managed the return of it and they prepped and warrantied it almost yeah as a yeah service as they brought it back to market. So well with millions of those not available starting now and running for the next three year, two year, one year lease anniversaries,
00:17:45
Speaker
it's gonna put a lot of pressure on used car values to come. While it's been coming down recently, yeah it went up 50% and now it's gonna leak down and then maybe even hold hold on because of that that ghost fleet effects.
00:17:59
Speaker
I'd like to say we don't need to worry about it for pricing. Everybody just go back to work, enjoy the profits that are coming through and from the large freight increases. But i i'm for the sake of the industry and and my future mission,
00:18:14
Speaker
Don't do that. it's Instead, work a little harder with those that $100 billion dollars a windfall and put some expense pressure on understanding the the accuracy of the value of a breath.
00:18:29
Speaker
Yeah, so there's a lack of supply of cars moving through the system, isn't there? yeah and Yeah, and if you did have a lease, a lot of people are just buying them out now because the yeah They can hold on to what they know um versus having to go to market in still turbulent area.
00:18:50
Speaker
Now, used car supply is getting better, but not everyone wants the used car or can afford the used there're somehow the There's some households in America that have never and will never buy and a new car.
Mission to Improve Pricing Accuracy
00:19:03
Speaker
ah Their budget's going to be $10,000, $20,000, not thirty fifty thousand And I suppose EVs are pushing those prices up as well. Yeah. For a minute, and now now they're falling. so the but kind of the the only silver lining to the EV competition is even Teslas are getting cheaper.
00:19:27
Speaker
What that means for total losses of EVs, that was pretty interesting. yeah When everything was high, total losses were fewer because The threshold for a total loss calculation at 80% of market fair market value would pay for this much damage to be repaired.
00:19:48
Speaker
As the value went up, the claims costs looked like they went up. Didn't really happen. that just yeah The underlying value at risk went up. And so the the shadow of payment...
00:20:02
Speaker
repairability got wider. Now, as cars are dropping in value, especially EV cars, that's being compressed and the total loss threshold getting lower, you'll see um more totals where I could just get you a new car that's an EV instead of replacing what you have. And so it's got to be a weird dynamic that...
00:20:23
Speaker
It's not obvious if all she did was look at loss costs to set your prices, you miss the subtleties of the value at risk in the loss costs. And the as people are being told, raise your deductibles and save money, that just kicks away the lower tail of claims.
00:20:39
Speaker
So even if you only did that, the average claim cost goes up. Yes, you know that's right. yeah And then as your as your vehicles get more valuable, then you start to repair a few more, your claim costs go up.
00:20:54
Speaker
So the now maybe the damage isn't any more severe than it used to be, um but your costs look that way. And if you've got inflation for labor sprinkled on top, then...
00:21:05
Speaker
If it takes more hours to do the work, more hours times more pay per hour, that goes on too. So it looks like a death spiral or a happy dream for a lost cost-traced business because you're just going to multiply you know your lost cost times a factor to get another factor to get your rate, and rates just keep going up.
00:21:28
Speaker
So it's it's insidious that you tell people, take more risk and pay less for your rate. But when you do that, loss costs inch up. And when your car gets more valuable, your loss costs get up.
00:21:41
Speaker
yeah And yeah yeah what you shouldn't do is cancel your collision on a vehicle that's holding its value. Because if it's actually worth 4,000 or 10,000, maybe you don't want to go bare on collision coverage because that's a pretty big financial shock.
00:22:00
Speaker
Yeah,
Future of Personalized Insurance Pricing
00:22:01
Speaker
absolutely. Yeah, no, it is a big shock. And what would you say is your vision for the future of insurance pricing? I think the risk of unit one is the the idea of getting the the most accurate price at the vehicle household level.
00:22:20
Speaker
Yeah. How it's driven, where it's driven, why it's driven, when it's driven is all more transparent in terms of its risk envelope.
00:22:31
Speaker
yeah And if you can't do anything besides go to work and have an awful commute, then that's your risk. If you behave better in that risk stream, you should still be rewarded with with a better operating rate.
00:22:49
Speaker
But if you've got more flexibility and you can choose when you drive and you don't drive much, that observability is a game changer environment. And we're seeing the the first decade of that with usage-based insurance and telematics. yeah um It's a lot creepy with people knowing where you are, when you are who'd you meet or where'd you go.
00:23:15
Speaker
but your Your phone already does that. So yeah it's so yeah it's same yeah so using your phone data to get through all that. So it's like, I'm on the one hand, little curious and on the other hand, how much can I save?
00:23:35
Speaker
Yeah, mean, smart watches as well are doing the same thing. All these devices already know where we are and it's sending that information back. But it's interesting how people are trusting their phone companies, but feeling nervous about insurance companies.
00:23:50
Speaker
And it's. If you went back to the 90s when credit scoring came in, yeah was like, well, you're going to use credit scoring in my auto insurance. Yeah.
00:24:02
Speaker
But we've been using it for making loan decisions for 30 years before that. like, oh, whatever mine. It's oh, and we use it to help you get your phone loan. You've got a thousand dollar new handset, whatever they're called these days.
00:24:18
Speaker
Some are even more expensive than that Yeah, they agree. Yeah, man. So it is curious that yeah when a consumer goes, wait, what? You're doing that? You know, that the the information when used in a different way becomes feeling ungoverned.
00:24:36
Speaker
yeah And yeah while you're, you never seen what your smartphone or your smartwatch or your smart car could really do, for some reason, the new new of love ah vehicle information seems more creepy than the comfortable new of everything else.
00:24:56
Speaker
Yeah, I think so. I think so. That's a very valid point.
Careers in Insurance
00:25:01
Speaker
So Marty, thanks ever so much for being on the show today. Is there anything else you'd like to add?
00:25:07
Speaker
Yeah, I would. the I would encourage all the generations who probably wouldn't ever hear this, but to maybe hear about something like this, say insurance is a the dullest granddaddy type of industry around, but it's complex, it's rich, it's meaningful to the everyday lives of of every person.
00:25:30
Speaker
And risk transfer is critical to making the economy work and making everyone's life more livable. I'd say if you hadn't thought of an insurance as a career, ah think again. Mm-hmm.
00:25:47
Speaker
I would agree with that. It's an ever ongoing challenge and Kurt certainly keeps us all busy and interested. yeah and And even when we make a mistake, the the necessity of having insurance to to make the world work ah gives us a chance to come back and do it better tomorrow.
00:26:05
Speaker
Yes, it definitely does. Thanks ever so much for being on the show today, Marty. You have a good rest of your day. You too, Jeremy. Thank you for having me. Brilliant. Thanks for being on. Bye-bye.