Earnings Call
Root, Inc. (ROOT)
Earnings Call Transcript - ROOT Q4 2020
Operator, Operator
Thank you for joining us for the Root, Inc. Fourth Quarter 2020 Earnings Conference Call. I will now turn the call over to Joe Laroche from Investor Relations. Please proceed.
Joe Laroche, Investor Relations
Good afternoon, and thank you for joining us today. Root is hosting this call to discuss its fourth quarter earnings results for the period ended December 2020. Participating on today's call are Alex Timm, Co-Founder and CEO; and Dan Rosenthal, Chief Financial Officer. Earlier this afternoon, 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 annual report on Form 10-K to be filed with the Securities and Exchange Commission next week. 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 upcoming 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 this 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 in 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 CEO.
Alex Timm, CEO
Thanks Joe, and good afternoon everyone. On today's call, I'll be underscoring a few highlights of Root's Q4 in fiscal 2020 performance and provide some additional context on three key drivers of our business: 1) The powerful competitive advantages enabled by our investment in proprietary technology in telematics; 2) How these advantages uniquely position us to manage risk as we expand our footprint and achieve scale; and 3) How our seasoned states will increasingly contribute to our management of the business; the deployment of our capital and profitability. I'll then turn the call over to Dan who will discuss our Q4 and fiscal 2020 results in detail. And then we'll open up the call for Q&A.
Dan Rosenthal, CFO
I'm happy to share that despite the unprecedented turbulence of 2020, Root is better positioned across almost every metric. We delivered a 37% increase in direct written premiums while simultaneously delivering a 26-point improvement in our direct accident period loss ratio. That is tremendous. Today, every large insurance carrier is eager to message to consumers and to the world that they're developing advanced technologies. In reality, they are not technology companies. Root is a technology company. So, let me tell you a little bit about our technology and our models and why we believe our models are so much better. First, Root's pricing model has been built from the ground up based on modern technology and data placing a fair percent of our model. The realization that each individual is a universe of one has allowed us to create a better pricing and underwriting model, in a word safe drivers. We put the customer in control while simultaneously delivering a superior modern experience in our mobile app. It is Root's proprietary telematics that enables us to deliver this consumer benefit and deliver profitable growth to the business. The foundation of Root's proprietary telematics is a transparent collection analysis of actual driving behavior off of a smartphone. This is the most powerful variable in our underwriting model. By collecting and synthesizing massive amounts of rich sensory behavioral data across thousands of driving variables, including identifying distracted driving, which is one of the biggest causes of car accidents today, we price based on actual causality rather than just correlation. The data we collect identifies the worst 10% to 15% of drivers on the road. We call it like it is, and we also write those drivers. This factor alone separates Root from any competitor. This allows us to then quote the remaining population fairly and at lower rates, which then allows us to build a more attractive book over time. While it may be very clear, this technology is difficult to build. While telematics in one form or another has been around for decades, mobile telematics has only recently become feasible due to advancements in mobile technology, as well as advances in machine learning. Today, we believe that Root has actually accumulated the largest proprietary data set of behavioral driving data tied to the associated client's data. We are the only insurance company to design our program to operate across our entire book of business. This has allowed us to build what we believe is the most powerful model in the industry. Almost all of our competitors outsource capabilities, which prevents them from achieving similar results. This is evidenced by the middlemen study that validated that Root is 10 times more predictive than a leading industry competitor and that was outlined in our prospectus. So, given all of this amazing technology, we often get the question of why we can't see it in the loss ratio. To isolate the positive impacts of our proprietary technology and models, we've created a matrix of seasoned states. As shown on today's letter, a seasoned state is one where: 1) The regulator has approved our data science-driven loss cost model and 2) We have been writing policies in that state for a minimum of one year and we've gotten two price filings approved. Applying this criteria, we started 2020 with only three states that we considered seasoned. By the end of 2020, that number was 20. As expected, our percentage of earned premium in these seasoned states also increased throughout the year and totaled 60% in the second half of 2020, which has enabled us to draw meaningful conclusions about these states from that data. What we've learned and what we see is that the seasoned states during the second half of 2020 ran a loss ratio a full 15 points below unseasoned states. This demonstrates the success of our approach and a materially positive impact of our improvements in segmentation. This is possible due to the increasing data that we have in these states, and this shows our data flywheel at work. This strong loss ratio performance in our seasoned states in 2020 validates that the Root pricing model is working. It is consistent with our expectations that each iteration of our UBI model will result in meaningful improvements in predictive power and segmentation, which then fuels our loss ratio trend. This is a trend that we fully expect to continue in the future. As the management team, we take very seriously the deployment of your money. And as such, we are constantly balancing growth versus profitability. Something special about Root, in addition to driving growth and profits, is that growth creates a natural moat around our business. This is what we call our flywheel. Basically, more data leads to better pricing, which drives faster growth as prices get better, leading to more data. That growth makes us smarter, and with each new learning, we create better products at better prices. And this is the beauty of a machine learning-based business. Our seasoned state analysis is enabling us to learn and fine-tune our pricing model in each state before we aggressively drive growth. Once we've proven that our state unit economics are sound, we quickly and efficiently look at the overall market spend and the capital we are deploying in that state. This gives us confidence as we drive growth that we will also be driving profit. We look forward to sharing further details around the evolution of our telematics and our technology and the growth and performance of our business within these seasoned states in the quarters ahead. With that, I'll turn it over to Dan. Thanks, Alex. Let me start by saying how proud both Alex and I are of the entire Root team's accomplishments in 2020. Together, we grew the business substantially, improved our path to profitability, made significant improvements around our debt structure and reinsurance arrangements, and set the business up for long-term success with the completion of our initial public offering. Against any backdrop, and particularly in the context of a global pandemic and all the disruption and uncertainty it caused, those accomplishments in one year are nothing less than extraordinary. We continued to show progress against our financial objectives in Q4. You'll find these along with our GAAP financial results contained in the shareholder letter we published this evening. Highlights include for the fourth quarter of 2020, we grew direct earned premium 30% year-over-year to $155 million. Direct loss ratio totaled 76%, including $10 million of favorable direct prior period development primarily in the most recent quarters. Adjusting for the impact of that prior development, the direct accident period loss ratio totaled 82%, a 16-point improvement from the comparable period in Q4 of 2019. Direct contribution, a new metric that we're sharing with you, increased by $26.3 million to $13.5 million, with the majority of the improvement coming from loss. I will be talking a little more about direct contribution in a few minutes, as it's a really important profitability measure for the business. Adjusted gross profit increased by $18.1 million to $3.9 million. Now shifting to our full-year 2020 results, we were able to show strong growth despite the decision to pull back on marketing spend towards the end of the first quarter resulting from the global pandemic and surrounding macroeconomic and regulatory uncertainty. While this decision slowed our growth in the second half of 2020, overall we were still able to deliver a 37% growth in direct written premiums to $617 million. Direct earned premium grew by 71% to $605 million. Direct loss ratio improved 18 points to 82%, including $24 million of unfavorable prior period development. Adjusting for the impact of prior period development, the direct accident period loss ratio improved 26 points from 104% in 2019 to 78% in 2020. Direct contribution improved $76.3 million to $18.9 million, with the majority of improvement coming from loss. Adjusted gross profit improved by $75.2 million to a profit of $21 million. We ended the year with $1.1 billion in cash and cash equivalents at Root Inc. and outside of our regulated insurance entities, with an additional $255 million in cash and investments at our insurance subsidiaries. We feel great about our balance sheet, and it will enable all the progress yet to come. Since our last earnings call, we've met with hundreds of investors and held nearly 70 one-on-one meetings. Your passionate interest in understanding the Root business model and differentiation was remarkable. Many of you asked great questions around our loss ratio trends. To be responsive in our letter and as referenced by Alex, we provided additional cohort data around our seasoned state performance and the progress made in state management. I want to tie all that together to the 26 points of progress we made against the accident period loss ratio in 2020. First, we attribute 15 points of improvement to pricing and underwriting actions taken in 2020. We separate these into two buckets: our proprietary segmentation, which captures the power of our UBI and pricing algorithms to target superior risk segmentation. We believe improvements and further deployment of these models across two-thirds of our footprint delivered 8 points of annual loss ratio improvement. State management, which was a focus for us in 2020, delivered a further 7 points of improvement. We attributed another 5 points to positive 10-year mix as our renewal premiums increased as a percent of total earned premiums. We attribute the remaining 6 points to COVID on a full-year basis. Most of this came with lower claims frequencies in March, April, and May, with a slight uptick towards the end of the year as well. So, where are we going in 2021? We expect to continue to deliver meaningful improvements to the loss ratio through further seasoning of states and the launch of new iterations of our proprietary telematics model and pricing algorithm. In addition, our decision to enter fewer states in 2021 improves our loss ratio outlook. Net-net, we expect year-over-year improvement despite higher new writings and lapping the 2020 COVID-related favorability. I also want to lay out where we are going in the longer term. We believe that as our data grows and the flywheel accelerates, we will continue to extend our pricing advantage. With a developed and tenured book, we expect to deliver a loss and loss adjustment expense ratio in the low 70s. We expect expansion of fee income via cross-sell of our home owner's product where we collect an agency commission, as well as the embedded value of our telematics to grow a SaaS revenue stream. Minor variable cost efficiencies round out our long-term direct contribution target at 25% to 30%. We've added direct contribution to our ongoing reported KPIs. We as a management team focus on this metric and want to share it with you going forward. Our capital strategy and reinsurance programs are also vital to our business. Part of my and my team's job is to take our direct outcomes and manage the net results. As detailed in our prospectus, we put in place a comprehensive reinsurance program. Our counterparties include five of the top 10 reinsurers in the world, as well as a large pension fund. We've shared that our reinsurance program would be in place for at least the next several years because it enables us to both use reinsurance capital to fuel our growth and de-risk the balance sheet. This program has a meaningful impact on our cost of capital and multiple lines of our consolidated financial statements. We've also shared that our reinsurance program is made up of several layered treaties and is designed for flexibility. Earlier this year, we made the decision to delay the renewal of one of these reinsurance treaties. Because of positive loss ratio trends, we expect to receive superior terms by delaying the treaty from January 1st to April 1st. This drives higher GAAP revenues as we retain more premium in the first half, but it has a negative impact on operating income due to reduced seating commissions. The decision to delay caused a short-term noise in our quarterly financials. But as we have always said, we will make the right decisions for the long-term business rather than managing the quarterly results. While the modification to seating levels will impact the P&L, we foresee only a minor change to overall 2021 capital needs because our structure has efficient alternatives such as our Cayman captive to manage the higher level of retained premiums. I will close with a few more details on how we're thinking about the financial outlook for 2021 and then Alex and I will welcome your questions. First, we plan to more than double our sales and marketing investments in 2021 following a COVID-driven pullback in 2020. This investment in marketing fuels an accelerating growth trajectory throughout the year. For the full year, we expect rewritten premiums in the range of $805 million to $855 million and direct earned premium in the range of $685 million to $715 million. Driven by my prior discussion of loss ratio, we expect direct contribution in the range of $25 million to $35 million. The delayed implementation of one of our reinsurance treaties results in seeded earned premium dropping in the mid-50s as a percent of earned premium by the second quarter, and then scaling back to our target seeding level by the fourth quarter. This reduced seeding level, along with fee income as a percent of earned premium consistent with 2020 and a nominal amount of investment income, results in GAAP revenues expected in the range of $270 million to $300 million. Based on what we know today and our base case expectation of our reinsurance for the year, we expect other insurance expense to result in a small expense position in each of the first two quarters of the year given reduced seeding commissions. Transition to an offsetting contra expense in the second half of the year as seeded premiums resume prior levels. Our fixed expense base remains in line with 2020 as a percent of direct earned premium. Together, these assumptions result in operating income in the range of a loss of $555 million to $505 million. With that, Alex and I look forward to your questions.
Operator, Operator
And now we have our first question from Yaron Kinar from Goldman Sachs. Your line is open.
Yaron Kinar, Analyst
Thank you very much. Good afternoon, everybody. Just maybe a couple of questions on the loss ratio. So, first, how do you determine the various components of the loss ratio improvement? I thought it was fascinating the way you weighted it out. I guess, as an outsider, I'm just curious how you know that 6 points of the improvement came from COVID as opposed to from seasoned and from the underwriting actions and better segmentation.
Dan Rosenthal, CFO
Thank you Yaron, this is Dan, and nice to hear your voice. We obviously spend a lot of time talking to investors over the last couple of months who wanted to see additional data about the loss ratio. We're thrilled to disclose the data today. We think it tells a very powerful story. The chart that you see on Page 07 of the shareholder letter splits the 26 points of loss ratio improvement into four different categories, and we talk about each of those categories in the letter itself. I think as you noted Yaron, it's very powerful to show that 20 points of that improvement, 20 of the 26 we believe were not related to COVID but were related to the work we've undertaken around getting our pricing algorithm and telematics models into market working on state management directly state-by-state, as well as what you saw from the renewal customers increasing, which we showed later in the letter. You asked about COVID itself. We monitor very carefully COVID in a couple of different ways. We think we have an ability to do that that goes beyond most other carriers given how we're able to use our telematics to track driving. So, obviously we're able to monitor miles driven, and we look forward to providing future updates on that on future calls. We look at not just the quality of the miles driven but the quantity as well. So, we understand not only how many miles people are driving but what time of day and what the road conditions are in those miles, so again it gives us a good understanding of what's happening in the mileage itself. And through all that, we did see, as we disclosed on our Q3 call, a 15 to 17-point impact on claims frequency, mostly with a little bit of recognition of claim severity in that sort of mid-early-to-mid March period through April and May, and then we saw a slight uptick towards the very end of the year that was not particularly material. So overall, blended across the year, it was 6 points represented in the loss ratio. I think the other part that's relevant if you're looking and comparing to other carriers is if you look at our footprint today of the 30 states, we are not licensed today in New York or in Massachusetts, which are obviously high commuting cities that had significant COVID restrictions in place for certain parts of 2020, and we're not particularly active in the California market as you know, Yaron, given the inability to use telematics at this time. So, we think that that is part of the reason that as we track the miles and understood our footprint, that's part of the reason as well understanding the 6 point delta due to COVID.
Yaron Kinar, Analyst
Okay. And inside of the story that I was trying to pick on COVID, I was just trying to better understand how does that proprietary segmentation account for 8% of the improvement versus state management accounting for 7%. How are you able to mainly allocate between the buckets, I guess is what I'm trying to understand?
Alex Timm, CEO
Yes. That's a good question, Yaron. This is Alex. Essentially, when we ship things in one state, they don’t all go out all at once. For instance, if we have a new model that we're going to deploy, let's say we deployed in Texas first. While we know that maybe another one of our states, maybe Ohio, doesn’t have that. What we can do is we will look at what the delta in the loss ratio did between those states before and after. It's the same thing even with COVID. COVID hit with all of this the mileage driven. We can look at that across a bunch of different states. So, when you can look at states versus state and you see certain state management actions and then you start to see the loss ratios actually change relative to one another. That really is how you control for all of those confounding factors. So, that gives us—then we load that in, and that gives us a good data set to begin to actually attribute where the loss ratio is coming from, and sort of via those principle components that are driving it.
Yaron Kinar, Analyst
Got it. That's helpful color. And my follow-up is along the loss ratio would be just looking at '21. I think you did some qualitative commentary that you should expect that the loss ratio would improve over the course of the year relative to 2020. Are you able to roll in and quantify or get some sort of range to bridge the gap between where we ended 2020 and the longer-term target of the low 70s range?
Dan Rosenthal, CFO
Yes Yaron, I'll jump in on that one. As you know, we haven’t guided direct loss ratio specifically. We've instead guided direct contribution, which really is how we manage the business and we think is an important metric for investors. Ultimately, there are some different offsets happening to the loss ratio in 2021 in terms of the outlook. There are headwinds in two ways. 1) We are anticipating the elimination of the abatement of sort of the COVID benefit being abated year-over-year. We are planning that that is not showing up in 2021. We also are planning negative 10-year mix and we show this in the letter. Because we are ramping up our growth and ramping up our level of new writings with our sales and marketing spend, that means that new writings will again be a larger percentage of our overall earned premium compared to renewals. Obviously, as you know, that will drive a bit higher loss ratio. So, we have headwinds from both of those things. That said, we believe we're going to have year-over-year improvement in the loss ratio itself due to continued benefits of the proprietary segmentation and state management work. So, those first two buckets that we talked about are going to improve and overcome the headwinds from 10-year mix and COVID abatement. And so, that's the way the loss ratio picture takes shape, although again we're not guiding direct loss ratio specifically for 2021.
Yaron Kinar, Analyst
Thank you very much.
Operator, Operator
And we have our next question from Michael Phillips from Morgan Stanley. Your line is open.
Michael Phillips, Analyst
Hi, everybody. I guess I want to touch on the comments in the letter and some things you said here about how the bits and fluctuations in the KPIs are key given the way you're changing kind of new marketing plans and in state expansion. Can you expand upon what things we should look for in terms of new term fluctuations?
Dan Rosenthal, CFO
Mike, this is Dan. I'm not sure, maybe if you could be a little more specific. There are some different puts and takes, although by and large the strategy is the same strategy consistent with what we talked about during the IPO. Are there specific KPIs that you're thinking about beyond the loss ratio we just talked about?
Michael Phillips, Analyst
Yes. I'm referring I guess more so to—and maybe I read about it in your letter. But I'm referring more to it sounds like you're slowing down a bit for the state expansion to focus on states where you got the regulatory approval. So, you can focus on the loss ratio. I guess, what that means going forward for this year in customer count and premium, you've given the guidance on premium but the customer counts. I mean, your sentence was—and those what's going to cause you term fluctuations in our KPIs as the result of our new investments in 2021. So, kind of just focus on that and which KPIs you're trying to—Do they work, Dan, or if not we can—.
Alex Timm, CEO
Yes, that helps. This is Alex. Maybe I will start with sort of how we're managing those big investments that we're making and then turn over to Dan, who can talk a better bit about those impacts. In terms of state expansion, we do want to be very cognizant of the balance between growth and profitability and ensuring that we're really making the right and appropriate trade off and that in states we have in testing mode versus in states that we have that are fully seasoned. That process is some of all time. As we add new states, we're going to learn our way into those states. Additionally, when we do that and how we do that can be dependent on regulatory timelines. As we add those states and the faster we build confidence in those states, we'll certainly be able to then accelerate growth in those states. That's going to be some of the things where you may see some noise, which like I said, there's just some inherent variability whether it be from regulators or how long it takes us to really build confidence in those states to push growth. The second is, we're continuing to push out new products. We're still in mostly mono-line auto. We now have launched our renter's product and we're scaling that. We're also launching and scaling our homeowner's product, which is going to also materially change the business, but those are still some investments that we're making. Lastly, what I'll touch on too is we're still investing in brand. We've been very happy with those results from the Bubba Wallace campaign around 'Progress Owes No Apology', which was really our way of explaining our brand in a very authentic modern way. That got almost half a billion impressions in the matter of a month. We are starting to see early signs of success with those brand initiatives. That's something we're going to keep continuing to invest in, and we believe long-term definitely pays off. We also understand you don't build a household brand in a matter of a couple of months. Those are some of the big investments that we're making today that we believe long-term will certainly pay off well for the business but predicting the exact timing can be difficult. Dan, do you want to talk maybe about some of the other key KPIs that may be impacted?
Dan Rosenthal, CFO
Yes. I think Mike, no I understand your question and I appreciate it. We've talked about there are four fundamental things that matter in the business as we manage it on a day-to-day basis; growth; loss ratio; retention; and customer acquisition cost. Alex just touched on most of those four but just to hit them quickly. On growth, I think we're trying to guide to is that there are going to be accelerations throughout the year. As we increase this marketing spend, you're going to see that show up in written and earned premium over the course of the year with acceleration along the way but starting-off smaller. On the loss ratio, I think I've touched on that in terms of that trend. Obviously, part of the decision we've made is to approach state expansion a little bit more moderately although we expect to be at 85% of the U.S. addressable market by year-end. So again, quite significant growth, quite a significant footprint. And retention for us is consistent with what we reported in our perspectives. The guidance accounts for continued seasoning of states, which could include pricing changes in a couple of places and temporary retention changes; not significant material near-term fluctuations. That's a little bit about how we expect retention to progress through the year. Then CAC, Alex talked about the brand. We talked about this Mike in our third quarter earnings call that we were going to invest into the IPO and the Bubba Wallace campaign and frankly, testing out some of our marketing and branding to prepare us better to be in that larger footprint as we go throughout 2021 and beyond. We're expecting our customer acquisition cost levels to be a bit higher in the fourth quarter and then in the early part of this year. You'll see that scale in a more efficient manner as the year continues. We'll see the impact of the marketing investments we're making in the back half pay-off even as we get into 2022. That's a little bit of how we expect the year to progress across those four key KPIs.
Michael Phillips, Analyst
Okay, no—thanks. That's very helpful. I guess this question will be geared more towards the states that you call the seasoned states and maybe one specifically that you show on Page 09, and your bigger states Texas, Kentucky, Pennsylvania, and Arizona. Do you think given where you're where you showed the loss ratios are at this year, do you think your current level of pricing in those seasoned states is where you need it to be or is there more work to be done on just the absolute level pricing in those seasoned states?
Alex Timm, CEO
Yes. The absolute level of pricing that we feel good about is adequate in the seasoned states for sure. That being said, we have UBI 4.0 in the harbor, and it's going to be shipped out sometime in the first half of this year. It is a material improvement showing roughly a 30% improvement in predictive power over our prior model, so that's substantial. We're also continuing to find other segmentation benefits. We'll keep rolling those out into those states. When we are comfortable with the target loss ratio that the state is running at, which like I said most of those seasoned states we are, what we do is we actually float those benefits to conversion and allow us to grow faster and reduce customer acquisition costs even further. That's the plan there. But in those seasoned states, we feel very good about our rate levels.
Michael Phillips, Analyst
Okay, thanks. Last one, more generically for me guys, is just if you look back over your history so far, mobile-based UBI telematics is fairly unique and certainly probably a lot more difficult than other forms of telematics. What do you think has been the hardest piece to get right for mobile-based telematics? Is it data integrity, data quality, or what would it be—how would you classify what's been the biggest challenge for you coming using mobile-based technology for telematics?
Alex Timm, CEO
Yes, that's a very good question. It's something that it’s hard to answer. Because really the thing you have to nail about mobile telematics is everything. It's hard to say there's only one thing, right. You've got to understand the engineering components of each of the smartphones and each of the makes of smartphones. You have to understand the quality of data you're pulling off the phone, you have to then understand the physical events going on in the vehicle. That alone is very difficult to actually understand what a heartbreak looks like, what texting and driving looks like. All of that is very difficult. But what makes it much harder and what very few people are then able to do is to then take all of that data and all of those insights in technology and correlate it to actual underlying claims data. That's the benefit we get as a carrier. We're not using any sort of implied claims or anything like that; we are just tuning on the actual underlying data. That's why our model has become so much more predictive than really any of our competitors because we can do and build all of this technology fully in-house. If I were to say one thing that's difficult, the most difficult thing is if you want to do this right, you have to be a technology company and an insurance company. That's a very difficult business to build.
Michael Phillips, Analyst
Okay. Yes, thanks Alex, I appreciate it.
Alex Timm, CEO
Thank you.
Operator, Operator
Okay, we have our next question from Tracy Benguigui from Barclays. Your line is open.
Tracy Benguigui, Analyst
Thank you. Maybe we could touch on the favorable results development. I remember last quarter you experienced unfavorable and now it's favorable, and I'm wondering if it's a little bit of Goldilocks and porridge. Maybe you took too high of a charge last quarter and you were able to release some this quarter. Should we be at just about the right spot on a going-forward basis?
Dan Rosenthal, CFO
Thanks Tracy, this is Dan. If my kids are listening to the call, they will be thrilled that this might be the first part of the call that they actually understand with the Goldilocks and porridge reference. I’m not sure I remember enough of it to be accurate in sticking with your analogy. But here's what I would say: they really are not related. We talked about the first part back on our Q3 call that the new disclosure for the fourth quarter specifically, you're right on a direct basis we took 10 million in favorable prior development, the vast majority of this related to acts in the year 2020. Whereas what we talked about on the Q3 call was mostly related to prior years. It really is apples and oranges if you will. Overall, I would just note Tracy, we really feel great about our reserving process. We've invested considerable resources in it just like Alex was talking about on Mike's question. I think it's one of the places that for us, traditional insurance principles or the great reserving actuary combined with technology and data science has really made a difference in supporting the critical work around reserving. So, we’re really thrilled about the position we're in and feel good about it going forward.
Tracy Benguigui, Analyst
Excellent. And then my follow-up is really on the delay of implementing your reinsurance treaties. I'm just wondering if that would lead you to conserve more capital and not be able to grow at the same speed until you're able to close on that contract?
Dan Rosenthal, CFO
Yes Tracy, the good news is—this is Dan again. As you know, we've got over a billion dollars in cash and cash equivalents on the balance sheet, as well as several hundred million in the insurance company. We are well-positioned for the year. We're certainly not managing month-to-month; we're doing the right thing for the long-term. We put in place a comprehensive reinsurance program that we're proud of. Our counterparties include five of the top 10 reinsurers in the world, as well as a large pension fund. The way we've talked about our reinsurance program has been really consistent. It will be in place for the next several years because it allows us to use reinsurance capital to fuel growth through the seeding commissions, as well as de-risk the balance sheet. It’s really twin-twin positives. The program does have a meaningful impact on our overall cost of capital. The way we've talked about it is it is made up of several layered treaties and is designed for flexibility. Earlier this year, we saw the way that our loss ratio was trending in the fourth quarter and frankly into January. We made the decision to delay the renewal of the January 1st treaty, intending to delay it till April 1st. We are making great progress on the April 1st treaty. We have top commitments that indicate over-subscription at favorable terms. We're excited about it. It was not motivated by any short-term financial concern, and it will not at all impact our strategic plan and growth investments for 2021. In fact, it's quite the opposite. It's the right thing to do for the long-term of the business. The support from our reinsurers at improving terms shows that our loss ratio is trending well. As you know, reinsurers care about three things: loss ratio; loss ratio; and loss ratio. The fact that this is moving positively as we approach April 1st is again a nice positive reinforcement of the way our loss ratio is trending. We look forward to coming back on our Q1 call and talking more about where it stands.
Tracy Benguigui, Analyst
I appreciate the comment about the oversubscription because then that would that was like my follow-up question if there wasn't a meeting of the minds but that's good information.
Dan Rosenthal, CFO
Well, you're welcome. Thanks Tracy for the question.
Operator, Operator
We have our next question from Elyse Greenspan from Wells Fargo. Your line is open.
Elyse Greenspan, Analyst
Thanks. So, my first question. I recognize this is the first time you guys are giving 2021 guidance and obviously a bunch of moving parts. You guys obviously don't want to guide to a specific loss ratio. Is there a way to dive back into some of the prior questions that you could give us a sense? It sounds like the profitability that you're expecting for 2021, kind of when we adjust for some of the timing of the reinsurance noise is better than what it would have been three months ago. Can you just give us a sense on obviously as you're doing less state expansion, you talked about some of your plans? How it looks better perhaps today than it would have been a few months ago?
Dan Rosenthal, CFO
Yes Elyse, thanks for the question. This is Dan. You are right, just to reiterate a little bit on the loss ratio and tie it into your question. If you look at all the puts-and-takes in the loss ratio, what we're seeing is we expect a bit of year-over-year improvement overall in the loss ratio, although we're not guiding to a specific number. You can see in direct contribution we're guiding to a range of $25 million to $35 million that we're very pleased about and reflects the fact that we are moderating state expansion. So, we are going to have a bit of loss ratio relief from starting those states in the right manner and trying to put in place our pricing algorithm and telematics and appropriate rate plans early on. We are not trying to do every state this year, and that remains. The goal is to get to 85% of the addressable market by year-end, which we still believe is obviously significant. We will protect the loss ratio a bit and frankly, protect our capital that we are investing. As you go down the profitability lines, I think that was sort of the other part of your question around operating income. Most of that is tied to the sales and marketing expense and the ramp-up in that. There are—there are some puts-and-takes from the reinsurance. The fact is, we are going to be seeding fewer premiums by delaying the treaty from January 1st to April 1st. That means our GAAP revenues will increase because we are retaining a bit more premium, but it means we'll be paid a little bit less in seeding commissions. Obviously, think about that on that other insurance expense line as a controlled expense; that will be a bit lower. Again, we think that directionally this is absolutely the right thing to do for the business in terms of how we're managing our state expansion. From a capital standpoint, our plan around reinsurance is highly consistent with what we talked about during the IPO. Frankly, just taking advantage of the fact that our loss ratio continues to trend in the right direction and the overall market conditions.
Elyse Greenspan, Analyst
Okay. And then, we know how the growth actually looked pretty good and picked up in the quarter. Obviously, growth may be a little bit lower than you would have expected in 2021 right just due to your kind of state expansion plans switching a little bit. Now, we've also in the market heard about some other players trying to use UBI as a point-of-sale similar to what you guys are doing. I just want to get a sense, it sounds like the growth being lower is due to your own decision right, not to expand in certain days and not due to some other competitors in the space?
Alex Timm, CEO
Yes absolutely, you're totally right. This is completely been within our own control and is our own decision because we think it's prudent. We really haven't seen, although there's a lot more advertisement out there around the apps, every major competitor had an app that presumably they were rooted five years ago when we started the company. We haven't seen any material changes in that technology that has made us see any sort of increased competition or difficulty in actually acquiring customers.
Elyse Greenspan, Analyst
Okay. And then one last one from me. There's an ongoing event in Texas, I recognize that usually customer acquisition costs are not typically affected by auto losses whatsoever but do you guys have a sense of your exposure and if we should be thinking about some losses there?
Alex Timm, CEO
Yes, first I'll jump in here, and then let Dan talk a little bit about the financial impact. I do want to say, we do have employees in Texas and customers and partners. We are very aware of everything they're going through, and we have been affected by these events. We're thinking of them. We have proactively reached out to our customers to allow them to know about extended grace periods as they may not be able to pay their bills. We're certainly thinking about everybody down there. Dan, would you like to maybe cover some of the financial impact?
Dan Rosenthal, CFO
Yes, thanks Alex. It's well said about the events in Texas. Elyse, thanks for the timely question. We’ve seen insurance industry loss estimates ranging from $5 billion to $20 billion on risks affecting several lines of business across the entire state. Obviously, a significant event. But as you noted Elyse, much less so from an auto perspective. We're protected in a couple of ways. We've observed a slight uptick in claims being reported in Texas in the last two weeks. The losses don't look like they will materially impact our book. It's still early to estimate an aggregate loss on our book, but it's worth noting—and reinforcing that we have a robust catastrophe reinsurance cover in place, limits our exposure to $3 million per catastrophe event or just under about 50 basis points loss ratio impact to the overall annual result. In addition to that CAC coverage, we renewed our XOL coverage on January 1st of this year for 900/100, which limits our exposure to large claims. We have diversified our premium across our 30 states in a really good way. The reinsurance coverages really minimizes our overall risk tied to any one event in a single state.
Elyse Greenspan, Analyst
That’s helpful color. Thank you.
Operator, Operator
We have our next question from Phil Stefano from Deutsche Bank. Your line is open.
Phil Stefano, Analyst
Thanks, and good evening. In the shareholder letter, you talked about routinely changing the pricing, fine-tuning the models, being able to re-price the algorithms every nine months. I was hoping you could talk about like the state regulators in my mind don’t move as fast as you do. What are the frictional pressures of having new pricing so often but the regulators may not be able to keep up with you?
Alex Timm, CEO
Yes, that's a fantastic question and something certainly that I think we've pushed in the industry. One, I will say, like not all states are sort of created equally at all. There are certain states where theoretically you can file every day and change your pricing structure every day or every week. Certainly that's not something we tend to do, but we do want to file more frequently. That goes all the way to other states that want you to only change your prices a couple of times a year or maybe four times a year or once a quarter. We reach out to those states and we've got really good relationships, and we look to see, okay how do we actually build in flexibility to that rate plan specifically that then allows us to iterate around meaningful things. In the early days of state launch, that's usually where we have the biggest changes. Those are usually the ones we expect to take the longest. As we build these relationships and get more comfortable with what our models are doing, the updates don't take nearly as long going forward.
Phil Stefano, Analyst
Okay. And maybe a follow-up to a question that Elyse had earlier about the competitive dynamics in the industry. My mind is less so focused on the UBI and telematics and just more broadly it feels like some of the legacy personal lines insurers have been cutting price recently on the auto business in reaction to COVID and the auto frequency benefit. I mean, have you seen this in any way impacting the growth or the retention metrics that you have? Are you feeling this in the market yet?
Alex Timm, CEO
No, we have no data that suggests right now that we're feeling that in the market. It is a massive market, again it's a $266 billion market. We're still quite small. We believe particularly because our product is so unique with our telematics differentiation, that that does provide a pretty robust book of business against a lot of the more macro trends. If State Farm lowers their rates 3%, we're probably not going to be heavily impacted by that just because we are so differentiated in the market and we are still so small relative to the size of the market.
Operator, Operator
And now we have our next question from Matt Carletti from JMP. Your line is open.
Matt Carletti, Analyst
Yes thanks, good afternoon. I really have a question on the top line. As we think about the ‘21 guidance and kind of for lack of a veritable rebalance a little bit between growth and loss ratio. While I know you haven't provided any guidance on kind of a two or three-year view; I'm sure you have it internally. Would there be any change to where you'd expect to be from either a direct or written premium standpoint two or three years out from these changes or is this more of just a kind of steepness of the line and which you get there?
Alex Timm, CEO
Yes. I’ll maybe talk first and then pass it over to Dan for some of that particular timing. We do not think that this change—in the small change in top line are rebalancing in our projection for 2021. We still think that this is a giant industry that we are still long-term going to be national and that we still see a massive opportunity to build market share via our differentiated products. We do not think that this changes whatsoever the long-term nature of value in this business. Dan, you can talk a little bit about maybe some of those nearer-term metrics and what we think how our strategy evolves might change?
Dan Rosenthal, CFO
Yes, thanks Alex and thanks Matt for the question. As I talked about at the top of the call, we have been out on the road talking to investors and it's been wonderful. The interest in Root, the interest in understanding our business plan has been remarkable. I've really enjoyed it and we're looking forward in the conferences ahead in the one-on-ones to getting back out there. The message that has come through loud and clear that is highly consistent with our strategy is to drive our investments towards sustainable, reputable, durable growth. That's our focus. We're really excited about the seasoned states and the progression in those states because for us that's sustainable, reputable, durable growth—we see it in the data. That focus is where we want to put our investments. We don't feel like this is a significant slowdown in growth. We're still going to add seven states this year, and we expect to be at 85% of the U.S. addressable market by year-end, which is a big deal. So, as far as what happens then in '22 and beyond, as you noted Matt, we're not providing specific guidance at this time, but our overall outlook remains really consistent. This is just a massive opportunity in auto. As we further enter states and season them, we're confident as Alex said that we're going to progress to our longer term goals.
Matt Carletti, Analyst
Great, thank you. And then just a quick follow-up kind of as we think about that a little bit slower state rollout. Can you talk a little bit about what that might mean for kind of the—as you're implementing UBI 3.0 and it’s rolling out, will 4.0 allow you to do that more efficiently because you're just focused on a smaller set of states—a more mature set of states? If so, what might we expect or what might that mean for test drive periods difference between pre-and post-telematics loss ratios, which I know you quantified a little bit in the shareholder letter, and I’d be curious if you talk about that for a second?
Alex Timm, CEO
Absolutely, I'll talk about it at a higher level and then pass it over to Dan for any further details. Focusing on a smaller set of states while still getting to 85% of the U.S. population is definitely a win-win because there's a material increase in the addressable market while being able to focus on fewer states. That allows us then to roll out these models much quicker and to really in a controlled sense measure how these models are performing in real-time in these states. Certainly, it will allow us to continue to iterate faster and to push faster on our pricing models. What you'll see too is as we've iterate on these models, things like the test drive period for instance actually have become shorter. We're actually able to identify good risks versus bad risks much sooner and more accurately. One of the things that our new model also does quite well is it does better with limited and sparse data. We know that there are phone models out there that for whatever reason, you may not get high-quality data off of. As we've advanced these models, we've gotten much better at that. Pre-telematics loss ratios and even optimizing the post-telematics loss ratios—I’ll leave that, Dan, if you’d like to comment on that?
Dan Rosenthal, CFO
No, I think you nailed it and maybe just in the interest of time, happy to catch up later. But you nailed it in terms of the evolution.
Operator, Operator
We have our next question from David Motemaden from Evercore ISI. Your line is open.
David Motemaden, Analyst
Hi, good evening. I just wanted to confirm it sounds like you still expect marketing spend to more than double in 2021. But you are entering into fewer states, and I guess we'll be acquiring fewer customers than originally planned. Have you changed at all the marketing expense expectation or are you just continuing with the same level of marketing spend as a sort of a, I guess, down payment on growth post-2021 or whenever you enter into the remaining states after 2021?
Alex Timm, CEO
Yes David, it's a great question and it's really the latter. We're continuing with the similar level of marketing spend; we are going to invest in brand spend. We talked about earlier how we've really seen it work through the fourth quarter, now we want to continue to test, even recognizing that the pay-off may be later in the year or into '22. We think it's really important as we continue to expand our footprint and outweigh any near-term pressure. Overall, CAC is going to remain a little bit elevated in the earlier part of the year. We do expect to see gradual efficiency throughout the year, and particularly as we do enter those new states we should see benefit from improved awareness and high-efficient social and digital ad placements. So, that's the plan that we are implementing in '21.
Operator, Operator
And now we will take our final question from Mark Huffs from Truist. Your line is open.
Mark Huffs, Analyst
Yes, thank you, good afternoon. Dan, did you quantify how much GAAP revenue might be impacted by the delay on the reinsurance agreement? Presumably, it's a little bit of a tailwind, but any sense of what that could be?
Dan Rosenthal, CFO
I didn't specifically but I'd point you to the guidance obviously with the revenue range from $270 million to $300 million for the year, which is reflective of the reinsurance plan. I talked in my remarks at the top a little bit about how we expect the seeding levels to flow throughout the year, and so that effectively can help you understand the direct impact on GAAP revenues.
Mark Huffs, Analyst
Thank you very much.
Alex Timm, CEO
Thanks, Mark.
Operator, Operator
This concludes today's presentation. Participants, you may now disconnect. Thank you for participating.