<div class="font_8">Starting in 2020, the Current Expected Credit Loss (CECL) accounting standard will require financial institutions to reserve for estimated lifetime losses on loans and leases as soon as they are originated. This presentation will provide analytical insight and practical recommendations to help lenders strategize and effectively prepare for the new rule. Learn how to:</div> <div class="font_8"> <div class="font_8"> <ul class="font_8"> <li> <div class="font_8">Account for future losses under alternative economic forecasts</div></li> <li> <div class="font_8">Conduct impact analysis on auto loan portfolios, institution reserves, and profitability</div></li> <li> <div class="font_8">Prepare a checklist for small, medium, and large size institutions</div></li> </ul> </div> </div> [toggle title="TRANSCRIPT"] <div class="transcript-scroll-box"> 00:00 This is a topic that I personally am very eager to learn more about. It addresses a new accounting standard related to loss expectation models. Cecil, as it's called, will go into effect later this year. It will require financial institutions to estimate and reserve for lifetime losses on loans and leases. 00:30 As soon as they are originated. 00:33 I would imagine that you folks will have a number of questions for our presenters, and I encourage you to ask those questions in the app. So to help us sift through those requirements, we have a distinguished pair of Moody's economists to give you the lowdown on Cecil and what does it mean for your company. They've prepared our hands checklists for different sized institutions and they've really broken down the rule into all of its individual parts, so you'll be able to get a better understanding of it. Our presenters today are sohini Choudhury, director and senior economist at Moody's Analytics. And Chris serinus, deputy chief economists also at Moody's Analytics. So he specializes in macroeconomic modeling and forecasting, scenario design and market risks research with a special focus on stress testing, and guest Cecil applications. Chris specializes in the impact of the economy on housing, credit, public policy and other areas and he also oversees development of Moody's Analytics, econometric models and forecasts. Please give a warm welcome to this medium Chris. 02:16 All right. Good morning, everyone. Thank you very much for the opportunity to be here with you this morning. So it's not every day that you have a couple of economists talking about accounting, 02:26 traditional were sworn enemies. But this new Cecil rule has a lot of economics involved in it. So we have certainly taken an interest in it. And it serves a lot of questions about economic scenarios and how they can be embedded in this rule. So what we'd like to do today is really discuss the seasons, and we'll have a fairly sizable impact on lending or potentially impact on the lending of various institutions. So we'll take you through a little bit of history or a little bit about understanding what the season is Right, so I'll start us off with Cecil a nutshell. section here. Just give you an overview of what Cecil is actually, let me take a quick poll. Does anyone know Is anyone familiar with Cecil? Cecil, before the introduction? Okay, so a few people have. Right. So I'll, I'll give you a brief intro to to see what the rule is what's changing kind of level set up throughout the audience here. And then I'll turn the floor over to so you need to talk about the seaso checklist in terms of what are the steps or what are the actions to take in order to get ready for seasonal supplies? There's wide variety of institutions. So we'll see. So basically, anyone who goes to a gap accounting and lends money is going to be subject to So what are some of the decision points, the key decisions that have to be made when you're putting together your Cecil process? And then finally, for fun, we put together a an impact analysis on auto lending, specifically, right. So one question that we very often get is what is he going to do to my institutions are loss reserves what's what is the potential impact here? The answer is that it's certainly gonna vary by institution or a number of factors that affect that particular calculation. But we decided to take a more industry level view, we have some data from Equifax where we can track autocratic broadly. And so we've done some analysis there to show that actually, seasons will have by RSS, it's a very 04:22 sizable impact on 04:25 on loss preserving that certainly the trickle down to an impact on origination volumes, Atlantic standards. So that's our agenda for that. We'll have time at the end for any questions. Well, let me go ahead and get started here with what Cecil is. Right. So Cecil, Cecil is a nutshell. So Cecil is a new accounting standard that was passed by the Financial Accounting Standards Board back in 2016. Or was released in 2016. Right, so formally, it's called faz B accounting standard update 2016 dash 13 topic 326 because that's a valuable, revise. have gone to some shorthand industry and just call this little season. And Cecil stands for current expected credit losses. And the bottom line is that this is a new accounting standard that will change how firms will have to report or estimate their own loss prevention. Right. So, the release of Cecil was intended to address a two middle to late loss provisioning that went on during the Great Recession. I'll talk a little bit about some of that history of why we're adopting since the first place, but that wasn't the intent to address too little too late. It's going to replace a couple of existing accounting standards specified as 114. This is how we provision for losses today standard, commonly known as incur loss and and another key point to make about Cecil is that it's going to apply to entity issuing credit as I mentioned, so this applies to small credit unions, large money center banks, finance companies in text, anyone who is lending money, essentially in Applying the gap accounting so even Caterpillar industrial company has to comply with the Cecil standard because obviously they do make the rule is going to go into effect later this year. Right? So starts separate routine 2018 for any financial reporting after that, it's going to be rolled out in phases right so the first group to be impacted by Cecil's will be sec filers, they'll have to comply with Cecil in 2020. And then try to unions other non sec filers, again, of course, will have to comply with Cecil the next few years. There's a bit of a rollout or phasing over time. Your standard is shown on the screen here itself was beautiful at Blue document if you go to the website and download a free copy of it yourself. It's about 300 pages of nice dense accounting materials. So you're feeling a little sleepy tonight or having trouble sleeping I'd highly recommend 07:03 taking a closer look. And certainly if you want to know a little bit more about the 07:09 Alright, so why are why we have? Right? I think that's the first question we should ask ourselves, right? Why change in accounting standard accounts typically don't like change. Investors certainly don't like change, they prefer to have continuity. I want to see how earnings are evolving toward for the year. Right. So why go through the trouble and this is as trouble as Caesars. There's quite a bit of effort that goes into implementing this. Why go? Why go through the pain of this? And the reason is, if you look at our current standard and current law, current law standard, it clearly has been highly corrosive. Right, so in the chart here, what you're looking at is loss allowance ratio. That's the blue line with left hand side access. And I'm graphing that against the employment rate back to 2008. You can see that these two series are almost perfectly correlated, very high correlation coefficient of point. Right. Furthermore, if we take a closer look at the series here, what you see is that allowances or loss allowance reserves, bottomed in 2006, right at the height of the market, just as everything was getting rocky in the academy, and then they skyrocketed in 2009 2010. And justice came up and so unreserved going right into the recession. And then we had to very quickly try to make up those reserves. And what you saw throughout the prices, just lenders every quarter, just trying to catch up, or keep up with the charge officers. So this led this concept of a too little too late type of process. And therefore the idea was, well, wouldn't it be great if we could move some of these loss reserves earlier on in the process. By doing so we might accomplish two things. First of all, we'll have a higher buffer at all points. cycles what happened more loss allowance capacity will eventually lead to some some smoothing smoothing allowances. Right. It doesn't mean you catch the next recession perfectly predict what's happened is it perfectly counters but at least the increase in terms of loss allowances, maybe a bit more buffered. And second of all, the hope is that if we are moving in some of those logical structures earlier on in the process, we're connecting them to accountants or to lending standards that the fact that we have to increase allowances in good times when things are profit perhaps well, that might have waited on too much overextend right. So let's hope and dream 09:46 whether or not we make that dream, a reality implementation 09:53 Alright, so what's actually changing the standard table here incurred loss versus Cecil justice you have a sense of what are the specifics. So today, basically the way we approach the last allowance is we originate loans. And then we wait for something that happened. So when something bad happens, we start to reserve losses. Right? This is a concept of a problem of freshmen have lost, right? So the loan has originated and maybe the borrower was delinquent or there's a declining credit score. There's some other type of trigger that causes us to say, you know what, there's a good chance that they're going to be lost on this provision against those losses. Cecil changes that and pushes the provision all the way to origination. So as soon as we originate the loan, we're already going to have to make an estimate of what the future losses will be. Now obviously, when we originate a loan, we don't expect it to actually default. But we know in large numbers, right, some fractions of the population is going to fall. There is some probability of default and some losses associated with So the idea here is to come up with an estimate what that probably is what that loss expectation is, and assign that to the origination process. The second concept that we have today is something known as a loss emergence period. Right? So as we're thinking about our losses, wherever we really think about the losses over the next 12 months, some type of period of time, where we believe that the bulk of the losses in the existence portfolio are going to be realized, right under C, so we change that we move to a full lifetime loss expectation. So we're not just looking a few years at times where we have to forecast our entire lifetime of those assets in time, so that involves a forecast of the future in terms of using reasonable and supportable scenarios, and then an extrapolation of those future losses as well. Right. So, complete loss of the current method is very most based approach seems more principles based. The current loss method we have today is more backward looking Looking at our current portfolio, try to infer back based on what has happened most recently, what the losses might look like in the future. Under Cecil is much more forward looking process, again, full lifetime as of origination, and then update that forecast at every point in time in the future. The current standard if your boss does require transparency in assumptions, that actually doesn't change, we still have to have very transparent assumptions. we're forecasting out what loss reserves will look like. We have to back up those assumptions with some type of data analysis and actually see what's known as a fire some additional disclosures. 12:37 And one very important key takeaway, as we're thinking about these 12:42 processes is that they're really designed for investors. Right, so fazzy sets the rules here, in order to provide investors with more information about portfolios and about the company's potential investors. They really don't have any stake in terms of macroeconomic stability. Or cyclicality of renewables that's not really their, their objective. They're really just interested in providing us investors with more information or timely information 13:09 about the potential risks 13:13 or with a particular. Alright. 13:18 Last point to make here and I'm glad it was touched on a little bit in the previous panel as well, always glad to hear about economic forecasts and scenarios. But I will leave you with a one assessment of this new Cecil processes where they change a lot of things in terms of modeling in terms of processing, the actual loss numbers, excuse me, we'll get into, but perhaps the single most important decision in terms of affecting the loss allowance assets themselves will be the choice of economic scenarios that are used in this process. Right. See some requires before we look at these, there's some forecasts, you could choose a very optimistic view of the future, right and roll that into your loss allowance. And that might be nice for a time, it certainly will expose you to some volatility in that last estimate, as the last part realized, you could take a more balanced approach as we did before, as we recommend and look at multiple scenarios, right? So every month, we are generating multiple scenarios for the economy, not just one Moody's Analytics Do you have when your time is headed, you may have a baseline. But we certainly consider the risks of possible outcomes. And we find that running systems analysis, looking at multiple scenarios is a better way to manage risks, but the uncertainty that we certainly know as we think about economic forecasts, so I'll leave it at that in terms of, you know, the single most important element here most important decision is the economic scenario, it could affect affect the level of losses, the volatility of losses, as well as the process. So it's definitely something that once you're thinking about very carefully as your design 14:57 I think so he's gonna get a little bit more into that as well. processes are here. 15:05 So, my, my portion of the talk will focus more on the actual process and knowing full well that most of you are probably not there yet, which is okay. But at least just a high level overview of what are the steps, what are the boxes that need to be checked off. So, what I'm showing on the slide is the basic steps right. So, you start with collecting data and people so, banks, public banks, who have started the process are realizing that this is one of the biggest challenges they have CSUN puts additional, you know, focus on requirements on data collection, then you have to collect data by vintages because losses, guess what depend on the vintage of the loads. So those kind of things. So data collection becomes very important. You should start collecting data. So that by the time csudh goes into effect in 2020, you have at least a couple of years of good data. Next is segmentation. So, CSUN says that you segment your portfolio by similar risk characteristics and for each of those cohorts, you project lifetime losses, how you segment we talked about this a little bit called step model selection, you select a model, like the previous governor was talking about modeling to forecast losses in the future. as Chris mentioned, Cecil is principle based it does not tell you which model to use. There are various models that you can use and I go over a couple based upon you know, things that are depends upon materiality or the portfolio, how large or how granular to go, things like these will determine what kind of a model we see Then scenario selection. As Chris said, CSUN requires a forward looking view of the economy. What kind of a scenario to select a baseline, most likely Outlook or multiple scenarios averaged out. That is a choice you have to make. Finally, an execution engine, an engine, which will take the outputs of your models, which could be losses, the probability of default and all of those and put it all together to give you a CCS estimate, that would be the execution engine. And then the last step would be management overlays are things that your model cannot capture, like loan origination principles, practices, competition, concentration risk things that your model has not been able to capture up near those on top of your estimated lifetime costs. So, this is sort of, there are more details, but the chart is already busy. So, I just kept it like these are the main ones. And I will discuss it one chart each for the three that have a black border around them. So segmentation, some examples of segmentation could be product, like maybe direct loans, indirect loans and just other products. Vintage, so vintage platters of known orangy dated in 2012, will behave very differently from what originated in a different point in the economic cycle. So, that matters, credit score, borrowers credit score, as everyone in this room probably knows, that is one of the important factors so you can segment your portfolio by crying, you're crying soft, that kind of stuff. Geography where state matters a lot in Florida versus New Hampshire. As we know, the economic outlook as well as historical performance has been very diverse across the country. Some countries suffer more in a downturn, some less. So it's important to factor that in collateral type, what kind of collateral you have behind the loan that matters, materiality, how important is this particular portfolio to your overall business? If it is, if it's a very important part, then you probably need to segment that 19:44 term, long term. And all of these features because they impact life and also you sort of have to recognize these things. And then historical loss patterns. If a specific port for has had very different historical losses, you want to segment that out and recognize that acknowledge that too many sub segments can result in too few known counts. So, this is something and caveat to keep in mind do not have thousands of sub segments is anyone doing statistics? No. So, it is a trade off and then going to the second part, so segmentation, now, you have to choose a model methodology and this is where, you know most people are and there is a lot of debate ceases started off by being principle based because the idea was to give your people the adopters more power, like, you choose what you want to do, but it started out to be quite confusing among people trying to because everyone is trying to compare what the other person is doing. Nobody's quite sure. So, again, what modern methodology you choose, depends Upon Here are just a few examples depends upon various other things. But obviously your institution sites, what a small lender can get away with a large lender with a more complicated portfolio and bigger dollar amount on their books will not be able to get away. So something to keep in mind, portfolio size and materiality same thing. So you might be a large measure, but you might have a couple of portfolios which are not very significant, they might be with a runoff, for example. So for those you can choose a different methodology which is more aggregated top down, instead of like a lone level Bottoms up. portfolio characteristics, is it a homogeneous portfolio or is it something which is changing over time, you don't want to necessarily plug everything together in the same pocket, what kind of data you have available, you can Choose a very granular method only if you have data to support that method. And this takes us back to your question of being able to collect good data, and then again index it on street, like knowing that the models will take a long time to drop. So you need to be able to support them, just some things you should consider when choosing. 22:28 Again, this is slide number two, but 22:32 it's like how 22:33 much information should I put in versus Nope, but it is what it is. I not go over line number nine. The idea is I have listed some of the commonly talked about methods that people are using in the industry and the industry. In general, like banks, credit unions of all different sorts of lenders, starting from the most Simple, which is the top row to the last row, which is possibly the most complicated, complicated in terms of data requirements. And so those would be the negatives, you know, it would require more data. But the positives going from top to bottom up, it will be able to capture more information has multiple purposes, you can use a model like that for stress testing, not just for CC. So those are the benefits as you move down the list. But on the other hand, you lose in terms of the data requirements, and also it's all a trade off. And it doesn't have to be the same method for your entire portfolio, as I just said, which is why segmentation is so important, depending on what so the first one, for example, is a method which is the simplest because it includes this forward looking information that we talked about that Chris talked about. Only qualitative. So you sort of do what you're currently doing in your zip code loss approach, and then qualitatively, you know, boost up or down your lifetime estimates based on what your outlook is for the future of the economy, instead of mathematically incorporated that information. And then going down the second rule, which is an Austrian model, solves that problem. It does incorporate the economic forecasts mathematically, but it sort of fools everything together. So risk from the loan versus risk from the borrower. Everything is put together requires this data, but is still okay. And then like, for example, the btn gt approach which is the fourth row it distinguishes the default risk from recovery risk, which has What is the energy philosophy for so it makes a distinction, but requires data and then discounted cash flow, which is a completely different timing of the recoveries is important data on that. But again, the basic idea of the slide is you can choose different methods based on simplest to most complicated. It's a trade off play depending on how material your portfolio is, and how much detail in half. 25:34 And then the third thing on my black bordered slide 25:39 was 25:40 economic scenarios. We talked about segmentation, or kind of model methods to choose an economic scenario selection. So CCS requires forward looking. It doesn't tell you what kind of scenarios should you use a single baseline Outlook or some sort of an upside And then average it out. So it does not record the use of multiple scenarios. But to cap a similar sort of a chart, how many scenarios are appropriate. So going from top to bottom, you can see the number of scenarios increased simplest to most simplest would be qualitative overlays. So just qualitatively, use Duff incorporate or leverage the forward looking information that you might have smallest form size, possibly Okay, hard to defend and quantified. And then the next one would be you use a single scenario. And then below the two below that will be multiple scenario approach. The benefit of multiple scenarios and this is what we are seeing most banks, at least to banks and credit unions, is that it lets you capture the details of the loan. distribution, the fact that credit losses are nonlinear, you get hit more in in a deteriorating economy. So including a downside scenario to your average out allows you to capture that tail risk. And theoretically it also creates this volatility moving from quarter to quarter. So just two benefits of why you would want to consider using more than one scenario. So you use multiple scenarios, assigning a probability rate unlikely to each of those scenarios and then calculate your CCI losses from each of those and take the average. That is, in my view, a more recommended approach than just leasing your reserves on a single outcome. Which really change quarter to quarter causes jobs in your allowances. And then finally syncing access. That's the question in everyone's mind. There has been lots of debate about this. Every day we see an article like American banker card. So what impact will cc's have? We've talked about that, in fact, in the next slide, but before we get there, one thing to understand is that, in fact, depends on a number of things. Here are just some longer term loans, automatically common senses, because remember one of the differences that Chris talked about, between the incurred losses incurred loss approach, you have a loss emergence period of 12 months. So you are introduced today, when the loan shows signs of deterioration you deserve for the next 12 months. In Cecil doesn't matter you orange Leave alone today, you reserve for the lifetime for the lifetime. So clearly, longer term notes you have to rgg that you have to reserve for. So that's the first not just a contractual term, but also prepayment lifetime assumptions, the life of the road so that's important age of the road the season. As the previous talked about this, most of the defaults occur in the initial few quarters or initial few videos. For a mature season low, the probability of default is lower. So that factors credit quality stage of economic cyclic, and what kind of false assumption how far forward looking if you think that we are heading into a downturn, then you have to reserve more today. So that increases your and this is Where all this talk about procyclicality is coming at if you think we are heading into a downturn that we have to resolve more than lifetime allowances will be more 30:13 current incurred loss method what you are doing today. So if you're trying to figure out how much more you have to resolve under Cecil 30:22 one of the factors is what are you doing today? Are you already doing a conservative kind of a reserving in which cases will have less impact? Or are you just reserving considering of any optimistic scenario in future? So what you're doing today, your practices today also matters in terms of how much more you have to reserve under Cecil? Here is a study he did you feel free to reach out to me for the white paper. This is all free. We publish on this topic. Leadership articles all the time. That's what we all need. Even though both of us have PhDs. We hardly know anything else to do. But this is one study that we did consider. So what you're trying to see is what would be reserves are under incurred loss. And under CC, which is life for a single cohort of auto loans, which was originated in 2015. q1 2015 to one New York crime 60 month term, so this is a cohort of auto loans. And so at the reporting date is December 2015. So if we stand at December 2015, the blue bars show realized losses, losses have already happened. Those are the blue lights. The green bars are Under incurred loss, but the assumption is you're reserving for the next 12 months. Cecil changes this because in addition to the green bars, you will also have the orange bars, because you have to reserve for the life of the note, which in this case is five years. So if you just look at this chart and think so the impact of Cecil would be the green plus orange area 32:30 divided by the green area 32:34 has already happened. It's realized, remember we are standing the reporting date is end of 2015 December 2015. So blue bars are the dollar amounts in millions of realized losses per month. And then the green is the reserves allowances under incurred loss for the next 12 months. And agreed plus orange is allowances Because it's life. So the ratio of the green plus orange divided by the green is how much more? What additional proportion? Do you have to reserve for cc? For this particular example, I think it's 2.2 point two types. So under CCS, you would have to reserve them. 33:24 Now think of this as a single cohort of nodes, 33:30 we can consider, 33:32 let's stack the losses for all cohorts of loans at a single point in time. So now consider your standing in the middle of 2018. And find do this exercise for every cohort of loan that exists so one would be a loan originated in 2015. To one in this particular rescore bank with six Keep dogs with certain outstanding balance. Another good be alone origins or a cohort of dogs originated in the same vintage with a different dog. So, consider basically perform that exercise in the previous job for every single mode of auto loan, standing in the middle of 2018 and then add up the numbers to get which is the black seven, the total lifetime loss that is like one thought exercise. So, that total lifetime loss will be all vintages that exists and are still active in the middle of 2018 for all types of borrowers, all loan terms and total outstanding balance and there you get a total lifetime loss and we did that 34:54 and this is the ratio. 34:57 So compared to incurred loss 35:02 The lifetime losses would be doubled. 35:09 For all of this does not include. So this includes both banks and finance, auto loans, no leases. That's the study that we did. I'm happy to share the white paper, which explains this. in more details, we used a weighted scenario like we use certain default assumptions, one being that the forward looking information is coming from a baseline and upside and a downside averaged. And the reporting date is middle of 2016, which is when the study was conducted. And this is the ratio so now you have to reserve a number compared to what you're reserving not in. 35:56 That is the function 36:00 Of course, going back to my previous slide, it does depend upon all of these other things. But on average 36:29 we have some great questions in here. 36:32 So let me see this one actually, that I think it's just a clarification kind of question is not actually related to auto but it was just was the expectation for mortgage companies to forecast for the full 30 year term? 36:48 Is that what they have to do? So if we consider 36:53 prepayments 36:55 so the real effective term is going to be closer to 710 years. 37:01 And also add to that that when in 710 years, most of the losses happen in the 37:07 first couple of years. So 37:10 that 10th year itself starts to taper 37:12 off the additional losses or allowances from ceases versus incurred loss tapers off, the more far we get. Got it? 37:22 Okay, so a few questions here. First off regarding disclosure requirements, what are the key changes or new requirements? 37:32 Yes, was primarily there are some disclosure on the vintage analysis that are currently not required. So looking at that defaults, losses by vintage and recording those out 37:45 Why do the guidelines not prescribing one methodology everything that it's harder to regulate when you have sort of this choice? 37:57 That's a question for me, but I do Wasn't 38:03 off record. As we look back at the history 38:05 here of this thing, there is a sister accounting principle for IFRS. Nine, which is for the rest of us rest of 38:17 the rest of the world not including the United States being prescribed some metal, like the 38:27 CSUN wanted to go against that, because 38:33 I mean, the idea is going to be principles bases that apply to all right, so the smallest is cutting into money since a bank JPMorgan, right. So, given that framework, it's difficult to figure out a single method and have it applied to everyone. So the idea was, you know, what, generally speaking what current expected credit losses, the expectations and how difficult or how complex or how accurate those estimates are. According to size, 39:02 that's actually not all that different from. Right. So now here's kind of a practical implementation question. How many seasonal models? Do we need one for the whole portfolio? Or many? And then how often are we were required to update that? Question. So 39:19 again, it's principle based, so it's gotta choose your own adventure. 39:25 What so what's required there is no required, how many do you need? What's gonna depend on your specific portfolio? Right? very homogenous. Our portfolio might be one may be more than sufficient for what may be sufficient. If you have a lot of variation in your portfolio, then certainly we're talking about a lot of products beyond all right, obviously, one or more models to accommodate different products 39:50 as a suggestion need to take your existing portfolio right, and then apply those estimates, you know, to loans. Let's say that we're started in and originated in 2014 or 15 or 16. And then say how accurate are we to our actual losses? 40:08 Yeah, absolutely. So the autumn years, in terms of the second part of that question, are the ones who are going to be the real arbiters of whether those models are acceptable or not. Right. So there would be certainly asking for some evidence that, oh, you're using these models as the future losses? How did you get comfortable with these forecasts, the back testing and model validation and all that other good governance, that's, that's gonna be part of the processes. 40:33 And How different would this be then stress testing that companies that others might already be doing? Like what would be the add elements that will be required? 40:43 stress testing is really geared to that stress outcome, right? So it's really trying to break the bank in some sense or break institution. This is a sec as expected, but it's not intended to capture the unexpected costs. It's supposed to be your best asset. Another one credit losses we'll look 41:01 at today before 41:04 we might use the same models, but it was certainly 41:09 one of the main differences in the forward looking assumptions. So stress testing, as 41:17 everything breaks 41:19 500 basis point increase in unemployment rate 4% decline in GDP and the power 41:26 ratio. At that time the awareness here is choosing or reasonable. 41:33 That is one thing, right? It's not necessarily suppressing it to the limits, right? It's just what do you actually think it will be? Okay, so some questions here. That kind of factor in prices and decisions, right? How our financial institutions and thinking about Cecil in the context of pricing that is if loss provisions are expected to increase should this be incorporated into pricing decisions. 42:00 Yes, yes, absolutely sort of made make its way through the 42:04 pricing is not valid for up 42:07 after what we're seeing a number of institutions to move if they can is a number of different drivers. It's hard to answer that question specifically because of the principles based approach. You can choose different assumptions, choose different models and get different answers. So we're seen as institutions trying to optimize their SES lessons around those prices as issues as well. So we do expect to see shorter loans or higher quality loans things, steps that can be taken to lower that total lifetime. 42:39 A clarification question is a weighted average life and not the actual life of the loan? 42:47 Which 42:49 for which 42:53 I'm not sure 42:55 about follow? Yeah. Speak to speak to me After work, please. 43:03 Let me see what else we have here. 43:08 Okay, so this, I guess, kind of released that privacy rights, this larger provisions will be required for lower credit quality loans. Do you think that consumers with lower credit scores will be adversely affected by Cecil? Because they will appear to be less profitable? 43:27 That's a good question. 43:29 So there are a couple of things going on here. Right. So on the one hand, the Yes, loss condition will be larger, for those who either hand subprime or lower credit quality tends to be more stable in the sense of risk multiplies recession. So some of that volatility may go away, and that actually is a benefit. The other thing to note here is, as we move to Cecil, there may be more positive consequences on the institutions from an investor perspective right in five minutes. is a pretty good finance company and they're increasing supervisions and then I can see a stream of income that's a little less volatile. I'm valuing that more. So the cost of capital may actually start to kind of do an offset. So it makes sense. 44:16 And then also this other practical implemented implementation side, right? It goes into effect one, January 2024 44:24 SEC filings. 44:26 So are we going to start to see like in that once a first quarter, like first quarter earnings, just like massively Hi, boss. 44:38 first job, credit recording industry comes to 121. 44:45 And by then it would be, you know, we would have that for the SEC. filers, right, but then for smaller companies, it's maybe 44:58 better your public or private 45:04 Okay, so 45:06 even playing field 45:07 transition 45:21 that's where they're driving. People like 45:23 you know, and all the JP Morgan's Bank of America. 45:27 They're the results of 45:31 JP Morgan said like 5,000,000,035% primarily coming from credit cards. So those kinds of things are public information. 45:43 It's gonna be exciting. 45:44 It is gonna be exciting. All right, I think we are out of time here. Thank you very much. </div> [/toggle]
