Welcome to the Root Incorporated Second Quarter 2021 Earnings Conference Call. My name is John, I'll be your Operator for today's call. At this time, all participants are in listen only mode. Later, we will conduct a question and answer session. During the question and answer session, if you do have a question, please press star then one on your touchtone phone. I'll turn the call over to Christine Patrick. Christine, you may begin.
Good morning, and thank you for joining us today. Root is hosting this call to discuss its Second Quarter 2021 Earnings Results. Participating on today's call are Alex Timm, Co-Founder and CEO, and Dan Rosenthal, Chief Financial Officer. During the question and answer portion of the call, our Presenters will be joined by Matt Bonakdarpour, Chief Data Science and Analytics Officer, and Frank Palmer, Chief Insurance Officer. Last evening, Root issued a shareholder letter announcing its financial results. While this call will reflect items discussed within that document, for more complete information about our financial performance, we also encourage you to read our 10-Q, as well as our 2020 Form 10-K. Before we begin, I want to remind you that matters discussed on today's call will include forward-looking statements related to our operating performance, financial goals, and business outlook, which are based on management's current beliefs and assumptions.
Please note that these forward-looking statements reflect our opinions as of the date of this call, and we undertake no obligation to revise this information as a result of new developments that may occur. Forward-looking statements are subject to various risks, uncertainties, and other factors that could cause our actual results to differ materially from those expected and described today. In addition, we are subject to a number of risks that may significantly impact our business and financial results. For a more detailed description of our risk factors, once again, please review our Form 10-K, where you will see a discussion of factors that could cause the company's actual results to differ materially from these statements, as well as our shareholder letter released today. A replay of this conference call will be available on our website under the investor relations section.
I would also like to remind you that during the call, we will discuss some non-GAAP measures in talking about Root's performance. You can find the reconciliation of those historical measures to the nearest comparable GAAP measures in our shareholder letter released today, and on our filings with the SEC, each of which will be posted on our website at ir.joinroot.com. I will now turn the call over to Alex Timm, Root's Co-Founder and Chief Executive Officer.
Thank you, Christine. Good morning, everyone, and thank you for joining us on our second quarter call. Today, I will provide a summary of our second quarter business results. I'll turn the call over to Dan for a deeper dive into our second quarter financials. Before touching on those topics, I am extremely excited to share further detail on our exclusive partnership with Carvana that we announced last night. Our partnership has deepened as the two companies have gotten to know each other and understand each other's businesses and shared values. Mutual respect has been built as we are two companies focused on delivering the best customer experience through technology platforms, and after success with our initial partnership, it made sense to find ways of accelerating this opportunity by expanding our relationship.
This industry first partnership goes beyond what any other carrier or aggregator has done before, empowering Root to build a fully integrated insurance solution for Carvana customers. As a tech company with a full stack insurance carrier, Root is best positioned to deliver personalized, bindable quotes embedded into the car ownership experience. Meeting customers where they are at the point of sale creates a natural onboarding experience that is likely to result in meaningful uptake with an attractive customer profile and attractive acquisition costs. Carvana's investment in the company underscores their commitment to the partnership while providing Root with additional capital. We are hard at work bringing the new partnership with Carvana to life and will share our progress along the way.
Now, turning to the second quarter. We continue to operate on the three key drivers I laid out at the beginning of the year.
First, the powerful competitive advantages enabled by our investment in proprietary technology and telematics. Second, how these advantages uniquely position us to manage risk as we achieve scale. Third, how our seasoned states will increasingly contribute to our management of the business, deployment of capital, and profitability. A key milestone was the release of our latest underwriting model, UBI 4.0. This brings to market increased predictive power that allows us to give the best drivers rates they deserve. The enhancements to the model leverages more than three times the data of our prior model, allowing us to better identify extreme risks and more accurately predict loss. We now have UBI 4.0 running in 16 states, allowing over half of our current customer quotes to utilize this model, with continued rollout over the coming months.
The advantage of our technology is not just on display when a customer shops and binds with us. It's also apparent in our retention metrics. Through scaling our data sets and improving pricing algorithms, we have been able to better segment our risks, resulting in increasing retention among the best 25% of drivers while simultaneously decreasing the worst 25% of drivers that renew. This further validates that we're giving customers the policy and pricing commensurate with their risk, a concept that is central to our sense of fairness. We continued down the path of national expansion, having launched our entry into Wisconsin during the second quarter. We were also able to accelerate our entrance into Washington State following the insurance commissioner's ban on the use of credit score, a bias we believe should be removed from rate setting altogether.
Through our data science capabilities, we were able to move from research to filing in less than two months, an effort that takes most competitors in excess of a year. As incumbent carriers in the state struggle to reflect the change, we will be able to enter the disrupted market with a pricing model that reflects the elimination of credit score across all variables. Our flexible platform best positions Root to win in a future where this bias is consistently eliminated nationally. During the second quarter, our marketing environment proved to be a challenge. Across virtually all of our digital ad channels, we saw placement costs increase substantially, which has since become a documented trend across the entire consumer product landscape.
While we were in the process of ramping customer acquisition, the prices we were seeing, particularly later in the quarter, simply got beyond our standards, and we began flattening or ramping spend down in certain channels to more acceptable levels. Given the magnitude of the cost increase this quarter, we recognize the value in diversifying our customer acquisition channels through technology and data science, building embedded insurance partnerships, and expanding our independent agent product. We see enormous and largely untapped opportunity in creating bespoke insurance offerings for relevant consumer verticals as we expand our network of partners. As a tech company with a full stack insurance carrier that owns the end-to-end consumer experience, we are in the unique position to offer a fully integrated digitally native experience.
Our exclusive partnership with Carvana is a significant step forward in demonstrating our value in this area, and we look forward to sharing the progress we are making in this channel in the quarters ahead. To extend our value proposition to more customers, we continue to build out our independent agency product. Nearly a third of customers shop for insurance with an agent, and this number has stayed steady over the past decade, even with the rise of direct channels. We are leveraging our technology to create convenience and transparency for both agents and consumers, and we'll share more on our work in this space in the coming quarters. Broadening our approach to customer acquisition channels is critical for us to manage your capital over the long term.
We are conscious of bringing the right risks onto the book for the right costs and believe using our technology advantage to tap into these enormous opportunities positions us to more efficiently deploy your capital over the long term. We believe we're on the path to truly unique opportunities within these channels, and we can't wait to update you in coming quarters. I'm thankful for the continued support of our investors, our team, and our customers. With that, I'll turn the call over to Dan.
Thanks, Alex, and good morning, everyone. The second quarter of 2021 reflected the environmental changes that the auto insurance industry experienced. You will find our full GAAP financial results contained in the shareholder letter we published yesterday evening, we wanted to give a few of the key highlights. On the top line, we grew direct written premium 24% year-over-year to $177 million. Our direct earned premium increased 19% year-over-year to $181 million. Our direct earned premium from seasoned states increased to 77% of total earned premiums. Shifting to profitability, direct accident period loss ratio was 88% for the second quarter, a 22% increase year-over-year against Q2 2020 when compared with the low loss environment in the previous year during the pandemic.
More importantly, we recorded a 23 point improvement from Q2 2019, demonstrating how significantly we've improved performance, stripping away the impact of the pandemic. This has been accomplished through the launch of new iteration of our proprietary UBI models and pricing algorithms, further state seasoning, and attention to bringing on the right risks at the right cost. The year-over-year increase in the loss ratio was primarily driven by 16 points of severity and 11 points of frequency as inflationary pressures increased costs and miles driven rose to above pre-pandemic levels in our book. Direct contribution was a $4 million loss for the quarter. The decline in direct contribution and related margin was driven primarily by direct loss ratio, as I covered above. While the industry, ourselves included, faced headwinds this quarter, we continue to make progress on the business overall.
While uncertainty remains around the ongoing impact from loss cost inflation, we believe that the underwriting and planned rate actions, coupled with our ability to update our loss cost model frequently, is allowing Root to react quickly. Additionally, we believe the prior rate increases taken as we saw these trends emerging will continue to earn into the book. As Alex detailed, we are focusing on diversifying our distribution channels, particularly through our embedded insurance offering and independent agent product. Our partnership with Carvana is the first significant step in unlocking the enormous and largely untapped opportunity that our technology platform brings to the partnership channel. As a whole, our actions this quarter set the stage for Root to drive growth at attractive unit economics and create long-term value for our shareholders.
The actions we are taking on both the marketing and partnership fronts not only better position the company for sustained growth, but open channels to more efficiently deploy your capital and strengthen our liquidity position over the long term. I am also pleased to announce that we have successfully placed a multi-year quota share effective July 1. Reinsurer interest was strong with oversubscribed demand and the ability to renew at improved terms. We remain excited about the strength of our reinsurance partners as a key component of our capital structure. There is additional stability as all of our treaties are now placed on a multi-year basis. I echo Alex's excitement about our prospects with Carvana as an embedded insurance partnership. Over time, we expect our relationship to drive a significant competitive advantage. We are laser-focused on efficiency and differentiation in our customer acquisition strategy.
We therefore expect to significantly reallocate and reduce our current marketing spend, impacting near-term growth, but positioning us for higher and more efficient growth into the future. As a result, we are revising our top line and profitability guidance for 2021. We now expect direct written premium growth for Q3 in the low to mid-single digits and a year-over-year decline in Q4. Also, on a year-over-year basis, we expect direct earned premium and GAAP revenues to grow high single to low double digits in Q3, followed by a decline in Q4. Looking ahead to 2022, we currently expect declines in top-line measures as we lap periods of high growth and significantly higher marketing spend, while we prioritize building out new channels at lower spend. We expect direct contribution to be in a modest loss position in the near term, as we make pricing adjustments to match the current industry environment.
We expect quarterly operating loss to stay elevated in Q3 2021, then materially decline in Q4 2021 as we focus on our most efficient marketing channels, resulting in a full-year operating loss in line to slightly unfavorable to the high end of our initial outlook range of $555 million. Noteworthy is that our updated guidance does not include any impact from the exclusive partnership with Carvana that we announced today. We are excited about the opportunities before us. I echo Alex's appreciation of your continued support. With that, we look forward to your questions.
Thank you. We'll now begin the question and answer session. If you have a question, press star then one on your touchtone phone. If you wish to be removed from the queue, please press the pound sign or the hash key. If you're using a speakerphone, you may need to pick up the handset first before pressing the numbers. Once again, if you do have a question, press star then one on your touchtone phone. Our first question is from Matt Carletti from JMP.
Hey, thanks. Good morning. Dan, I just wanted to make sure I'm hearing you right on kind of the reasons behind the changes in the growth outlook, and that they are kind of purely customer acquisition cost focused, and that is not related to the near-term pressures that have been on the loss ratio, but more just directly related to what's happening with cost of marketing through performance channels.
That's a good question. This is Alex, and I'll let Dan add more detail here. For us, what we saw with the spike in loss ratio quarter-over-quarter from the increase in inflation, as well as the increase in costs and marketing channels, we just don't near term see that as an efficient time to actually be deploying capital. We think it's much better longer term to deploy that capital when we have a better sense for where we want those loss ratios to be and when we have a better sense for that marketing acquisition. Dan, would you add anything?
Yeah. Thanks, and good morning, Matt. Thanks for the question. Look, I'd echo what Alex said. You should not think of this as a massive strategy shift. This is really a prudent use of capital given the current industry environment, having to do more with the marketing channels and the loss ratio both. Across our marketing channels, what you see, Matt, is some things are working, and we're going to keep investing in those channels. Some are not, and what we're saying is we're either going to fix those or stop utilizing those channels. We have a couple new things. We're thrilled about the partnership with Carvana. It's the first of what we hope will be multiple embedded insurance partnerships, and as well as some of the other new channels that we talked about.
We feel like those channels, combined with the current ones that are working, will help us get right back on the growth trajectory at the right time.
Okay, great. Then just a follow-up, if I could. I heard you talk a little bit about kind of you're working with agents. I think Left Lane is what I've seen as kind of the name for it. There was some news, I can't remember if it was an announcement from you guys or just industry press, about Root Ready during the quarter and kind of embedded tech within cars and some manufacturers. Can you tell us a little bit about what you're doing there and how we should think about that?
Absolutely. Root Ready is a program that we've actually been working on for a few years, and we're proud of where we've gotten it. It really allows us to position the company much better for the future as vehicle technologies continue to improve and as vehicles become increasingly connected. Effectively what we do is we have access to databases that effectively when a consumer comes to our website or our app, and they provide some basic information, we can actually see if they previously had a connected vehicle or currently have a connected vehicle in which we already have driving data. In that case, what we will do is we will actually provide an immediate quote leveraging telematics data.
We still want the cell phone data because obviously from vehicle data you don't get texting and driving and other behavioral data. We will still collect that data over the following six months after the consumer binds to incorporate that new data in on renewal. This really allows us to basically provide an instant quote that is much more data informed than otherwise.
Great. Thank you for the color.
Our next question is from Elyse Greenspan from Wells Fargo.
Hi, thanks. Good morning. My first question, I guess it's just building upon the last topic. You guys mentioned diversifying acquisition channels and bringing on new partner relationships as well as building out your independent agent product. Could you just expand and give us a sense of ultimately just how big some of these endeavors could be over time?
Absolutely. We think that there is going to be a fundamental shift towards embedded insurance, where it makes much more sense for the consumer and is much better consumer experience to effectively, while they're purchasing a vehicle, to just have insurance included in that vehicle purchase. It makes it much more seamless, much less of a headache. Auto insurance is not necessarily a product consumers love shopping for. What we have found is that through our technology, having really built an insurance company from scratch over the last five years, that we have a very nimble and flexible tech stack that allows us to seamlessly integrate, really with a variety of partners. We think Carvana is obviously a major partner here. We've gotten to know the company very well.
We feel very culturally aligned with them, we think that there's lots of opportunity for them to continue to disrupt the space that they are playing in. We think that that's meaningful. In terms of independent agents, they've weathered the direct market shift very well. They have really shown resilience over the last 10 years. Their market share has been relatively stable at about 30%. We do know, and we believe, that there are many consumers, and there will continue to be many consumers, that prefer to go to an independent agent. We also believe that those independent agents need better technology from carriers, and that allows them to easily quote somebody so that you're not on the phone for 20 or 30 minutes with an agent just to get a simple auto insurance quote.
We found that in our early testing in R&D, that we can actually build much better seamless products for the agent that the agent enjoys, and then also that allows the agent to get their consumers onto a mobile device and into a mobile setting. We've really found that our technology there can extend to serve those customers, too. That's 30%+ of the market, and it's a $270 billion market in total. We think that clearly is a very large opportunity. In terms of customer acquisition costs, there's a few things I'd say. One, I think all of these channels won't be as exposed to some of the short-term noises that we may be seeing right now in digital marketing channels.
Two, I think they will also be not only much more predictable, but we think they'll be very efficient. We will be acquiring the right customer at the right price over the long term.
That's helpful. Do you have a sense of the relationship with Carvana? I recognize you guys just announced it, but do you have a sense of the TAM there and just over time, as you guys agreed to enter into this relationship, how much of an opportunity that could be for Root?
Yeah, we're really excited about, Elyse, this is Dan, for the reasons that Alex articulated. This is a transformational partnership for Root, and we believe for consumers of autos and auto insurance. Alex mentioned we are like-minded companies. We both put digital first, and the good news is we've gotten to know each other. We've worked together now for one year, and we've seen in the early days of our partnership, the significant opportunity. As far as the TAM, I would point you to Carvana's disclosure of having a proven model with some of their other embedded services. For example, as they've disclosed, about 80% of their customers choose to finance with them. They also offer warranty and service embedded.
We think, as Alex has said, consumers today go to buy a car with batteries not included because they don't have the opportunity to purchase insurance at that point of sale. The TAM here that we think we can unlock together with Carvana is massive. Now, at the same time, we are making sure that we invest in the right way with them to provide and build a product that works with consumers. We think that will ultimately lead to better economics as a more vertically integrated player, and we've structured our agreement with Carvana to split those in a way that's fair. Elyse, we're not going to provide any more details at this time. We're going to hunker down, get to work, and build an incredible product for consumers.
That's great. One last one. You guys said premiums should be down next year, right? Recognizing that there's been some strong growth to start 2021. Would you expect growth to be down in all four quarters, or is it something where, as you see things today, perhaps, given some of these dynamics are going to impact the second half of this year, that you would return to top-line growth in the second half of 2022?
It's a good question, Elyse. In the guidance that we've put forward, we've alluded to the first half of the year for the reason that you suggested. We think that we are going to be disciplined in our marketing spend, especially at the tail end of this year and into early next year. You're exactly right. As you look at lapping the first and second quarter of this year, we are going to make sure that we're focusing our resources on investing in the new channels that Alex talked about, as well as some of the existing marketing channels that are working. Our guide is that that's not going to keep us on the same growth trajectory as we conserve our marketing spend and focus it on the channels that are working. That is the plan as we move forward.
Then as we build out those new channels, we'll be excited to come back and talk to the market about the growth opportunities that we see beyond.
Okay, thanks for the color.
Our next question is from Josh Siegler from Cantor Fitzgerald.
Hi, good morning, and thanks for taking my question. Can you describe the actual experience that a Carvana customer will have regarding Root? When in the process are they presented with the quote, and will they receive any telematics adjusted rate after a certain amount of time?
Right now, we really are in the early stage of developing the product and embedding the product in the experience. We will make it very seamless for consumers, and we're going to test along the way, and we think we're going to learn a lot. In terms of telematics, we will be integrated with the Carvana app, and we do expect that we will be actually even closer to the vehicle through this partnership and even closer to some of the new sources of telematics data at this time. We'll be excited to share as we develop the product into the future with all of you.
Great. That's helpful. Then, speaking on partnerships, can you speak a little bit about the pipeline of additional partnerships right now? Do you expect the future partnerships to kind of look like this exclusive deal with Carvana or maybe more like some of your prior partnerships?
I think really where we are is we've had partnerships in place for quite a while, both in the auto space, but then even more broadly with other financial service companies. We do think that in intercepting consumers in times that make sense to them is really important, and we have seen various degrees of success with partners. We've tested out lots of partners, some that work really well and some that don't. What we want to do going forward is really focus and double down on those partners where we think the product makes the best sense for consumers. We see it as we have a very healthy pipeline, and we want to make sure that we're focused.
Right now, we think that the focus really is on the Carvana partnership because that's really where we see, one, as Dan alluded to, the most aligned team in terms of technology, in terms of culture, and really intimately getting to know one another, but also in terms of the ability to really create a differentiated consumer experience.
The only thing I would just briefly add is not only from our perspective does it create a differentiated consumer experience, but does so with a really attractive customer demographic for us. These are not consumers who are going on Google or vertical search to buy car insurance. These are meeting consumers at the point of sale. We believe that that will have not only a more efficient customer acquisition cost for us compared to our performance channels, but also, frankly, a longer retaining customer by having met them at the point of sale. We're really excited, and I echo what Alex said. I think we cannot overestimate the real bond that the teams have had between Root and Carvana, really focused on creating the ultimate consumer experience. We're very excited about what's to come.
Got it. That's very helpful. Thank you very much.
Our next question is from Nick Jones from Citi.
Great. Thanks for taking the questions. I guess one on Carvana. Can you maybe describe the nature of the agreement or the exclusivity? Does this mean Carvana can only offer Root, or are you kind of an exclusive technology partner that over time they can embed other auto insurers into the platform? Just maybe a little clarity there. Then, can you remind us how many drivers who are maybe swapping out used cars or just changing cars also consider changing auto insurance during those transactions? Thanks.
I can take the first question, Nick, and then turn it over to Alex. The exclusivity in the partnership is something we're really excited about. We are going to build really the first vertically integrated embedded insurance product exclusively for Carvana and working with them on an exclusive basis. We recognize that there are some states where we are not yet writing business, and there are perhaps other aspects where the right thing for the consumer is, to your point, to broker in another insurance carrier or what have you. The north star for us here is to do the right thing for the consumer. This is really an exclusive partnership where we are going to be investing a lot of time and energy and people resources in order to build out that really differentiated embedded insurance product partnering with Carvana. Alex?
Yeah. In terms of consumer shopping behavior, this is one of the unique moments of truth. This is when consumers do change vehicles, as you'd expect, they typically see want a change in price to their current insurance. That does tend to trigger shopping moments. We think if you can get in front of a consumer at this moment, that this is easily within the top three or four triggers for shopping in insurance that's out there. It's very big. I'll also say even in these triggers and in these moments, you also get access, as Dan was referring to consumers that may not otherwise shop for insurance because they're not necessarily going to Google and googling auto insurance to shop frequently or maybe ever. It also gets us access to consumers that we may not otherwise see through more traditional marketing funnels.
Great. Thank you.
Our next question is from Tracy Benguigui from Barclays.
Good morning. You mentioned that miles driven is rising above pre-pandemic levels for your book. We're hearing from other auto insurers that miles driven is approaching pre-pandemic levels. Is there something unique to your book that we should consider, or is there a data set you're using that could help bridge this difference?
Good morning. This is Frank. I can take that one. I've got two perspectives on that. First, as I look across the industry and have seen what other folks are reporting, there are some general telematics and traffic data that besides just Root that have said that we're at pre-pandemic levels or higher. We might also be looking at very granular data on our customers and being able to see in real time where those miles are going. It's not clear that in all of the reports that you might be reading, that they're actually looking at real time in the last month or two versus looking at miles driven compared to the pandemic over the last six months or in the quarter one. We feel that we actually have really early warning signs and can really more quickly and granularly see what's going on.
It also varies by state. We can actually tell in certain states where the miles are up versus other states. We believe that we've got very granular insights into how that driving happens.
Okay. That's really helpful. In response to this loss trend, how would you describe the magnitude of rate increases you may need to get closer to your long-term loss ratio target? How swiftly do you think you could get through the approval process?
Sure. A few things on that. First, I'd say that we had seen some signs of increasing trend before the second quarter, and so we have had differentiated maybe in the industry rate increases in first quarter and second quarter. Not necessarily as large as what we think we need now, but we've kind of been flowing rate into our book all year long. We think faster and sooner than what some of the industry's been seeing. As we look out over the next six months, I think that we will end up being able to do more rate increases than what most of the competitors will be able to do. Some of that is just, again, because of our real time looking and the fact that we've been able to see this sooner allows us to start that filing process earlier maybe than some of the industry.
Okay, maybe just a quick follow-up on that. I think you're right. Would that limit your ability for more on a regulatory standpoint as others are trying to do it for the first time?
First as we filed that, there's some states we haven't filed in at all, and we will be first into those states probably with rate filed rate increases compared to competitors. In states that we took rate increases in the first quarter, most of those or many of those states are ones where you don't have to get regulatory approval. They're file and use. We think that given the fact that we took smaller rate increases in first and second quarter, as we're coming back in the third and fourth quarter, we'll also be then looking for smaller rate increases than if we had had to bunch it up the way our competitors do. It's much easier to get, say, two fours than an eight approved in a particular state.
Okay. Very helpful. Thank you.
Our next question is from Mark Hughes from Truist.
Yeah. Thank you. Good morning. I'm curious on the severity and frequency, how much of that do you think is this broader inflation spike? Certainly used cars you mentioned in your release, the pricing has been up dramatically, but some of those factors may be temporary or maybe not. I'm just curious how you see these losses, how much of it is going to endure, how much might be transitory?
Sure. First we've only had, say one quarter's worth of data where we've really seen this spike up. Trying to figure out those trends are still opaque, I'd say. We don't really have a clear view. Are we seeing a bubble that's going to come down? On the miles driven, when you look at other countries, say China, we did see China had a big spike right after their pandemic kind of slowed down as people moved from public transportation to regular transportation or to driving. Then it came back down a little bit on the miles driven as people moved back to transportation, but hasn't yet come back down to pre-pandemic levels. That we could see a bubble. On the severity side, we talked about used car prices. You can see general inflation throughout the U.S. economy.
I don't see that dropping back down in the next 12 months to zero trend levels. I see the severity as probably more of a 12-18 month sustained increase. Again, we've only got three to six months worth of data, so the trends are pretty hard to tell there.
Yeah. Then Dan, on the reinsurance, the renewed multi-year agreement, anything you could tell us around the economics? Is the model similar to what you had discussed in earlier periods? I think there was also an element of that where the loss sharing mix shifted over time. Just any updates there on those new agreements.
Thanks, Mark. All positive on the reinsurance front. As we've talked about before, we manage the business on a direct basis, and then Alex says to my team and me to manage the balance sheet in the appropriate way, and we're really pleased with the renewal of the 7/1 treaty. It is a similar cohort-based, multi-year treaty to what we've been doing. Oversubscribed, improved financial terms, and we're very pleased with not only the terms and the size, but the partners who continued and increased as well their allocations. We can talk more about that in the weeks and months ahead, but nothing to see here. All positive, all consistent with the model, all moving forward.
Thank you.
Our next question is from Josh Shanker from Bank of America.
Yeah, thank you. I wanted to follow up on that reinsurance question. In a quarter like 2Q 2021, where your direct loss ratio plus LAE is 100%, what are the economics of a quarter like that to the reinsurers? Are they taking also a loss ratio of 100% and they're paying you a ceding commission, and they have their own expenses? Can you sort of talk about how their economics work in the contract?
Yeah. It's something that we considered as we approached, obviously, the discussions with the different reinsurers. Josh, I was really pleased the reinsurers took a long-term view and saw the improvements that are happening in our loss ratio through the use of our differentiated technology, and some of the work that Frank is doing on segmentation and state management. That's showing up in the actuarial results. They're committed for the long term through these cohort-based treaties. The structure's the same as we've talked about in the past. We get paid a ceding commission upfront for securing the business, and then we pay the reinsurers a margin over time. They ultimately focus on the fact that in a cohort-based treaty, they are benefiting from the lifetime value of the customer. They're not so much focused on this month's results or next month's results.
It really is a long-term approach, looking at the trend of our losses, the improvements that Frank and his team are making, as well as the overall personal auto sector. Reinsurers, as you well know, Josh, want a diversification in their risk book, and they're pleased to continue renewing with us. I was not only pleased with the fact that we were oversubscribed in the treaty, but again, the terms improving as well.
On your growth guidance, I know you don't want to talk about the Carvana product yet because it's still in development and whatnot. The timing may not be clear, but the growth guidance is X, any possible benefit from the Carvana relationship, does it include some Carvana in it? How should I think about your outlook for growth and that deal?
It does not include Carvana in it.
It does not include any Carvana.
Correct.
The timing of when we could see some contribution, you're not prepared to say just yet?
We've talked about that with our partners at Carvana, and our message is we're getting in and starting to work. The product teams, Josh, are tremendously excited by some of the opportunities that are in the works. Stay tuned. We're not prepared to give a specific timeline. Again, these are two teams that move very fast, that have digital native approaches. I would say stay tuned, and we'll come back to the market together when we have more to talk about.
Thank you for the answers.
Our next question is from Phil Stefano from Deutsche Bank.
Yes, thanks, and good morning. Congrats on the Carvana partnership. I feel like there's been a number of questions around the strategy. Maybe just a quick housekeeping item on that. Apologies if I missed it. Was there a coupon or a rate that was mentioned on the preferreds, or is there a way you could help us to think about that?
Yeah. We filed our 8-K this morning, Phil, with all of the details. There is no coupon on the convertible preferred.
Okay, perfect. Thank you. In the shareholder letter on the UBI 4.0 section, you noted that there's a belief that it will help in the identification of new segments for expansion. I was hoping you could just flush this out a little more. Is that a roundabout way of saying the independent agent markets, or are there different new segments that you're trying to point us to when you talk about this?
Yeah. Hi. Thanks for the question. This is Matt. That's referring to something we're really excited about with UBI 4.0, which is the ability to score drivers with far less data. Historically, we've been very conservative in the criteria we set to make a user eligible to be scored. We wanted to ensure that we had enough data to create a high-fidelity score. Now that we've collected tremendous amount of trip data and loss data over time, we've been able to fine-tune our telematics risk score to basically do more with less. There was a reasonably large population of drivers who don't drive very much, who we were unable to score because of those conservative eligibility rules. Now we're able to confidently score them with a predictive score and really open up that quoted population to getting fair rates through our telematics score.
Is there a way to think about how this will be evidenced? Is it more people coming into the funnel? Is it better conversion through the funnel? How do we think about, as there's more iterations of the UBI technology, how do I think about the right metric to judge the success of this?
Yeah. In the shareholder letter, you'll look at the good driver proportion over time. We will see a dramatic increase over time because we're able to provide segmentation on this bucket that we were unable to provide segmentation on before. Said another way, there's a large proportion of good drivers that we were not segmenting before UBI 4.0, and now we're able to provide them adequate rates that reflect their driving experience.
Got it. Thank you.
Our next question is from David Motemaden from Evercore ISI.
Hi. Good morning. I guess just this feels like a big strategy shift here just from the distribution side. I guess I just wanted to understand sort of the mix of your business and your PIF and how much of it comes from each channel, whether it be performance marketing, direct partnership. Is there any way you can give us a sense for how big of a shift in your book of business this will be?
Yeah. We really do not view this as a strategy shift. We have been working on our embedded insurance product for actually over four years and have been experimenting with various partners. We've been working with Carvana over the last year, and they really got to know us and our company intimately. We got to know each other on a cultural level intimately as well. I think that's really where we started to see a potential for the acceleration of our strategy. With independent agents and agents, it was always something we wanted to. We've always believed that we can, through our technology and through telematics, provide a disproportionate value proposition to customers that are coming through that channel as well. That said, we also still really believe in our direct model.
We will grow with our direct model when it makes sense to grow in our direct model, and we won't when it doesn't. In that way, we believe we're going to create the best long-term shareholder value. When we look over the long term, we do believe that our marketing channels will continue to invest in the marketing channels that are the most efficient and the distribution channels that are the most efficient. We would expect there to be a large shift of volume to these new channels that are emerging. We truly do believe that the future of auto insurance looks much different than what it does today. We think that it'll work better for consumers through partnerships like Carvana and the embedded platform. Dan, would you add anything?
No, I think that's right. I think that as we move forward, you're going to find that some of these channels are accretive, really, in multifaceted ways. I'd echo what Alex said. This is not a massive strategy shift for us. This is about investing capital in the channels that are working and doing the right thing for shareholder value over the long term. Where we see channels that are not working at a time where you can see it in Google and Facebook's earnings, you can see it in what other carriers are doing in terms of vertical search and some of the different channels out there, it doesn't make sense to invest that capital. We think our investors would prefer that we be prudent stewards of that capital and focus on building out the new channels that we're tremendously excited about.
This partnership with Carvana is really transformational for us. We think it's groundbreaking. We think it's going to change the way that consumers buy cars and buy their car insurance. It's important for us to prioritize that based upon what we've learned in partnership with them over the last year. This is only the beginning, and we're excited to move forward and really build out that shareholder value for the long term.
Understood. That's helpful. I guess just wanted to just shift to the debt and capital position. I believe there's some debt that's maturing in October. Could you just remind me of your plans for that? Also, do you have an update just on the capital levels in the Ohio subsidiary and the Cayman captive, and I guess your plans for the proceeds from the investment from Carvana? Are those going to be injected into either one of those subsidiaries to support capital levels there?
David, thanks for the questions. On the first question, we do have a $100 million funded term loan with a group of eight banks maturing in October. We are in discussions right now with those banks, and we'll come back to the market as we have a further update. As far as the capital levels overall, there are two aspects that are important here. One is obviously the investment from the Carvana partnership is important to us on several levels. It's a strategic enabler that really solidifies our already strong financial position and then gives us the flexibility to take full advantage of the opportunity in front of us.
Second, we're noting throughout our shareholder letter and on the call today that we are going to drive a material reduction in our marketing spend in the coming quarters, that we are going to focus on long-term shareholder value. When you combine those two things, the reduction in our marketing spend and the investment from the Carvana partnership, that will meaningfully extend our capital runway expectations. We have more on the debt front, reinsurance front, or what have you, we'll come back and talk to the market. We sit here outside of the insurance entities as of June 30th with $825 million of unencumbered capital. We feel like we're in a good position. When you add that to the Carvana proceeds from their investment, coupled with the reduction in our marketing spend going forward, we're in good position for years to come.
Great. Thank you.
Our next question is from Gary Ransom from Dowling & Partners.
Yes, good morning. Most of my questions have been answered, but I wanted to understand what you were doing with Carvana before this deal. You said you had been working with them. Were there some sales through the partnership? Was it just more rudimentary type of technology? Could you help us out on what you were doing before versus what you've talked about for the future?
Absolutely. Before, we were really just displaying Root to a subset of customers, a small subset of customers that were coming through the Carvana experience. It also was not an embedded experience. It was really just testing to see what those take rates could be and how interested consumers may be in the product.
Thank you. One other question on retention, just overall, I know you talked about retention being better for the better drivers, but even within those better drivers, has there been any change as the frequency has come back and driving has come back? Anything you're seeing there that might be helpful for the outlook?
We look historically and compare to seasonal trends. We see that this year's seasonal trends as it pertains to retention mimics the pre-COVID seasonal trends. The trend is looking the same throughout the year. We have implemented product changes that have shown, through rigorous experimentation, improvements to our retention. While the trend is the same, we have seen some shifts in those curves. As those changes get implemented on top of our entire customer base, we expect to continue to see improvements to those retention numbers.
All right. Thank you very much. That's all I have.
Our next question is from Ryan Tunis from Autonomous.
Thanks. Good morning. First question on distribution and Carvana. A couple pieces to it. First, as you work with independent agents and/or Carvana, are you more willing to move away from kind of the core telematics drive first type offering? Also with Carvana, how does the underwriting work with that? I would think that Carvana would be holding you to some kind of mandatory acceptance rate. What's your flexibility and your ability to get the pricing and get the mix right underwriting with Carvana?
Really, our underwriting and our telematics will be deployed really in full force through the Carvana platform as well. We actually see it as an opportunity to extend our advantage as we get closer to vehicles, which we can then supplement the cellphone data from, whether it's the Carvana app or the Root app, with vehicle data and data directly off of the vehicle. In addition, there is no mandatory acceptance ratios that we have crafted with Carvana. We will do what's right for the consumer in every situation and make sure that we can address as many consumers as possible and make sure that consumers have a really great experience across 100% of their funnel. That may or may not mean that those consumers are going to be underwritten directly by a Root insurance entity.
Got it. Maybe for Dan, the pullback in marketing spend, I would have thought would have maybe benefited the P&L a little bit in the back half of the year. The guide still implies the low end of your previous guidance. I'm wondering, why is there not more of a benefit there? Is it that advertising will still be relatively elevated, or is there something else? Is it an expectation for a higher loss ratio or more tech investment? Just some color on why you don't expect the quarterly net income run rate to improve materially.
Yeah, thanks, Ryan. It's a combination of losses, which Frank talked about, and we obviously talked about in the shareholder letter. Just the fact that, we're sitting here already midway through Q3, and a lot of the marketing spend, obviously July is in the past, and some of the marketing spend is already committed as we look forward. As we assessed our channels and really decided to make a move to focus on making sure our capital is being invested in the channels that are working and turning off the ones that are not, it just takes a little bit of transition time to do that. I think you'll really start to see the impact of that show up in Q4. That impacts the operating income guide for the year.
Thank you.
Our next question is from Christopher Martin from KBW.
Hey, guys. Thanks for taking my call. It's kind of a follow-up from the last few questions as to taking it all in. In the coupling of moving to an independent agent distribution channel and then also looking at the embedded insurance channels, with all the work you've done to build your technology and your brand differentiation, how does that kind of fit with those types of growth for the future, where it's still, hey, here's what Root really stands for, but you're no longer really owning the customer in the same way and then shifting towards more of like a balance sheet and claims organization, as opposed to I'm getting this from Root, as opposed to I'm getting this from my agent, I'm getting this through Carvana.
I was wondering if you had any thoughts around that for the future growth potentials and the economics of it.
Absolutely. We do feel that this particularly leverages our engineering and data science work and really is an extension of our value proposition. Being able to power seamless, great experiences for consumers, whether or not they may see the Root brand or see a Carvana brand, we think it still leverages our core competencies that we've always been about, which is engineering, data science, and consumer experience. Everywhere we see the opportunity to build what is best for the consumer, we will do that. We do not believe that that necessarily precludes us from building a really strong brand. We think there's lots of fintech companies out there that embed their software that still have very strong brands.
We do not believe that that moves us away from what we are all about, particularly because what we view we're all about is, quite frankly, just building a better business through technology for the consumer. We think we can do that also through the independent agent channel as we see lots of technology gaps that exist in independent agency offerings. By the way, telematics gaps as well. The take rates on telematics in independent agency is historically very low. We believe we can change that, and we think that that's really meaningful for consumers to provide a large host of those consumers with better products at better prices. That's what we're going to continue to do, and we're going to continue to invest in that.
We still, again, believe in our direct channel, and we'll grow that when it makes sense, but we're going to be nimble, and we're going to build a diversified set of really demand generation platforms, and that's going to include the embedded insurance platform and the independent insurance platform, independent agent.
Got you. That makes sense. Thanks. That was my only question for now. Thanks.
Thank you.
Our next question is from Tracy Benguigui from Barclays.
Well, thank you for taking my next question. I recognize you've been pretty agile repricing at the 30-day mark when a driver doesn't go through that rehearsal period. As you enter more affinity relationships like Carvana, do you think you have that same ability to be just as agile? Where I'm getting at is that you have a good thought process behind your churn rate to get to your desired customer mix and rate adequacy. Would an affinity partner be okay with higher churn?
This is Alex. We view our telematics, as you said, as a way where we can clean up our book. What that means for some customers is that they take the test drive before they ever are insured with us. They go through because they want to see what rate they might be able to get. With other folks, we might see that they want a 30-day test drive, or to Matt's earlier points on being able to price people even earlier, it could mean even earlier than that. We view our telematics platform as something that is very flexible and extensible. What we're doing is we are building customized solutions with some of these affiliate partners that can work for their customer set while allowing us to still leverage that data in the best possible way.
Over time, what we believe that will do is build the best consumer experiences in the most data-informed way.
Okay, what you're saying will be nuanced for some of these relationships.
Absolutely.
Okay, great. Thank you.
Thank you, ladies and gentlemen. That concludes today's call. Thank you for participating, and you may now disconnect.