Upstart Holdings Inc (UPST) 2022 Q2 法說會逐字稿

Good day, and welcome to the Upstart Q2 2022 Earnings Call. Today's conference is being recorded. At this time, I would like to turn the conference over to Jason Schmidt, Head of Investor Relations. Please go ahead, sir.

Good afternoon, and thank you for joining us on today's conference call to discuss Upstart's Second Quarter 2022 Financial Results. With us on today's call are Dave Girouard, Upstart's Chief Executive Officer; and Sanjay Datta, our Chief Financial Officer.

Before we begin, I want to remind you that shortly after the market closed today, Upstart issued a press release announcing its second quarter 2022 financial results and published an Investor Relations presentation and credit FAQ. All are available on our Investor Relations website, ir.upstart.com.

During the call, we will make forward-looking statements, such as guidance for the third quarter of 2022 related to our business and our plans to expand our platform in the future. These statements are based on our current expectations and information available as of today and are subject to a variety of risks, uncertainties and assumptions. Actual results may differ materially as a result of various risk factors that have been described in our filings with the SEC. As a result, we caution you against placing undue reliance on these forward-looking statements. We assume no obligation to update any forward-looking statements as a result of new information or future events, except as required by law. In addition, during today's call, unless otherwise stated, references to our results are provided as non-GAAP financial measures and are reconciled to our GAAP results, which can be found in the earnings release and supplemental tables.

(Operator Instructions) Later this quarter, Upstart will be participating in the Goldman Sachs Communacopia Plus Technology Conference, September 13; and the Piper Sandler Growth Frontier Conference, September 14.

Now I'd like to turn it over to Dave Girouard, CEO of Upstart.

Good afternoon, everyone. Thank you for joining us on our earnings call covering our second quarter 2022 results. I'm Dave Girouard, Co-Founder and CEO of Upstart. Today, we reported a decline in revenues, which is obviously disappointing and unacceptable to us. I want to explain where this decline came from and what we're doing to address it.

It may be natural for you to question whether Upstart's AI-powered risk models aren't working as designed, but we're confident this isn't the case, that, in fact, our models continue to improve with respect to accuracy and risk separation. But there's no getting around the fact that a decline in revenues is a business problem that we need to address. And today, we'll share with you the actions we're taking to address it.

Today, Sanjay and I will discuss a variety of topics, including credit performance, loan funding, lending partner sentiment and some of the actions we're taking right now to make sure Upstart's future is bright. I also want to share with you the progress we have made in many important aspects of our business and how they're setting the stage once again for Upstart's growth in the future.

I don't want to spend too much time restating what you've already heard about the current economic climate. Given the nature of our product and our borrower, we do, however, have a unique lens into what's transpired in the last 2-plus years and what may transpire in the coming months and years. We believe we're at the end of a unique economic cycle related to the pandemic that included 2 distinct phases. The first phase was triggered by a pandemic-constrained consumer spending and unprecedented government stimulus throughout 2020 and early 2021. These together drove significant improvements in consumer savings levels and liquidity, which in turn led to dramatic overperformance of credit during the space. Our platform experienced about a 50% reduction in credit defaults compared to the pre-COVID time frame.

In the second phase, toward the end of 2021 and into 2022, this effect began to unwind as stimulus was discontinued and consumers begin to travel, dine out and spend once again. And as expected, default rates returned to pre-COVID levels or in some cases, even higher. While virtually all consumers benefited financially from reduced spending during the early stages of the pandemic, this cycle was concentrated in consumers who received government stimulus checks, a demographic, which is also more likely to be Upstart borrowers.

Our risk models largely captured these effects and performed admirably, though not perfectly throughout. But I'll get to that in a bit. We believe we're now at the end of the 2-phase cycle and an important question for all of us is what's next. Will efforts to slow inflation lead to recession and unemployment. While no one knows the future, we do expect a significant slowing of the economy and a worse-than-normal macro for the next year and beyond. We'll speak to that as well. Our job to all of this is to ensure the future of our platform and to protect Upstart's ability to pursue our mission for years to come.

Alongside our earnings release, we today shared some responses to important questions regarding credit performance on Upstart's platform. It goes without saying that measuring credit performance is vital and it's also nontrivial. Comparing one platform to another can be challenging, different products, different borrowers, different return targets, months on book, prepayments, hardship policies and more. There's no simple apples-to-apples comparison. We believe the essential measurement for credit performance is actual dollar returns compared to the lenders or institutional investors target at the time of origination, full stop. And today, we provided this information for all Upstart cohorts going back to the beginning of 2018.

The bottom line is this: our 70-plus bank and credit union partners who typically retain loans in the lower risk rates appropriate to their businesses have seen to date portfolios consistently meet or exceed expectations since the program began in 2018. And however, institutional loan buyers done against the target of approximately 8% gross return since Q1 2018, institutional buyers have so far seen 12 quarterly vintages overperform with 5 expected to underperform.

It's important to highlight that a loan buyer who invested equally in all cohorts since Q1 2018 would have experienced a positive return on all vintages thus far with an overall 9.8% gross annualized return. This compares to a return of less than 3% in the U.S. high-yield bond index over that same period.

Lastly, we believe it's not reasonable to expect above-target loan performance, irrespective of the economic cycle. So it's fundamentally important to separate the impact of macro conditions from imperfections in the credit model. The essential litmus test for model performance is separation of high- and low-risk borrowers. As demonstrated in the loss rate by grade and AUC metrics we shared today, our model is positively differentiated in this respect, and it continues to improve. In an effort to deliver unparalleled transparency and analytics, we will provide this detailed information to each of our lenders and loan buyers.

Today, we're in a funding-constrained environment, which is the primary cause of our revenue shortfall. I want to share some thoughts on this situation and actions we're taking to address it. First, as we have said recently, our goal is to operate as a marketplace for credit over the long run. We want loan transactions to take place when they make sense for the borrower and the lender. And certainly, lending is a category which would expect to experience some volatility over time due to macroeconomic factors.

Having said that, in the last few months, lenders and institutional credit investors reacted more quickly and abruptly than we anticipated. Despite the fact that our bank partners have seen consistently strong credit performance, meaning portfolios performing at or above plan across quarterly cohort, several of them have paused or reduced originations due to fear about the future of the economy. To be clear, these lenders and institutional investors have not left Upstart platform but have temporarily paused or reduced their originations.

As we shared in our credit performance FAQ today, we believe our models are well calibrated to the current economic environment and in fact, include a generous accommodation for a recession over the next 18 to 24 months. And given funding constraints, we believe the opportunity for lenders to generate strong returns on Upstart is unusually high right now, yet the reaction of lenders is often binary in nature, more so than we would have anticipated. As a result, we've concluded that we need to upgrade and improve the funding side of our marketplace, bringing a significant amount of committed capital on board from partners who invest consistently through cycles. We're currently evaluating a variety of opportunities to do just that. So we expect this will take some time to bring to fruition.

Furthermore, while we continue to believe that it doesn't make sense for Upstart to become a bank, we've decided it may make sense to, at times, leverage our own balance sheet as a transitional bridge to this committed funding. I acknowledge that this is a shift relative to what we planned and communicated earlier this year but a changing and volatile environment suggests we need to be flexible and responsive in our approach.

We're taking this step for a few reasons. First, there's an obvious information asymmetry where we understand better than anybody how our model is performing today and how well it's calibrated for the current economic environment. Secondly, we believe the opportunity to generate outsized profits on our platform is unusually high right now. And third, we can bring a level of stability to our business that's important to our longer-term goals while we work to put these committed capital structures in place. Sanjay will share some more about this in his remarks shortly.

I want to also highlight that we're building a business that can survive and thrive through a variety of market conditions to make sure we achieve these ambitious, long-term goals. Our fixed costs are low, and our gross margins are strong so we continue to invest in our road map and in our future through a variety of macro environments.

We continue to make rapid progress in the newer parts of our business, and we're optimistic that this progress is setting up the next stage of growth for Upstart, which I'm sure you're all looking forward to. First, we continue to add new lenders to our marketplace, with a total of 71 banks and credit unions as of today, up from 57 when we last spoke to you in May.

Despite the cautionary outlook in the financial services industry, forward-thinking banks and credit unions continue to choose Upstart. We now have 640 dealerships using Upstart auto retail software. And just a few weeks ago, industry analyst automotive market data declared that Upstart was the nation's fastest-growing auto retail software provider in the second quarter. Subaru and VW were the latest OEMs that announced support for Upstart auto retail, joining Toyota, Lexus, Mitsubishi and Kia as well as top franchise dealers from 37 brands including Ford, Honda and BMW.

We also expanded our auto retail lending product out to 29 dealerships and saw the first $10 million in retail loan originations in the second quarter. In just the last couple of weeks, we merged our machine learning model for automated income verification, originally developed for our personal loan product into our auto retail lending flow. We expect this improvement to more than double the percent of applicants for whom we can now automatically verify their income.

I'm also pleased to announce that we quietly launched our small business loan product at the end of June, well ahead of schedule. We've already seen some more than 40 small business loans originated, totaling more than $1 million in principle in just a few weeks. That team is quickly ironing out operational issues with an eye toward rapidly expanding this product in the coming months and years. Lastly, the small dollar loan team launched support for Spanish-speaking applicants, another giant step towards serving those left out of the country's mainstream financial system.

Some of you have questioned whether Upstart veered too quickly into lending to riskier borrowers in 2021 in order to grow in our post-IPO phase. But I believe we have done exactly what we set out to do and what we said we would do. Upstart's mission is and has been to leverage modern technology and data science to improve access to affordable credit. There are tens of millions of Americans who deserve access to reasonably priced credit from our nation's banking system, yet are denied access through no fault of their own. We're unique among our fintech peers and that we aim to tackle this problem directly.

The terms nonprime, near-prime and subprime. These are words the industry invented to describe people that our current systems don't understand. The truth is that the vast majority of these Americans are entirely credit worthy. Upstart's mission is to identify those borrowers and provide them with access to affordable credit, and we haven't wavered from that challenge.

How does growth fit in? We approach our business as a waterfall of priorities in a way analogous to structured credit. Upstart's highest priority, our A bond, if you will, is credit quality. Our goal is to reliably deliver the return the lender or investor expects for a specific allocation of risk. Our B-bond or next highest priority is unit economics or gross profits. We don't strive for loan transactions that lose money for Upstart and generally seek to avoid them.

And finally, whatever is left over goes to platform transaction growth or residual, so to speak. In truth, growth isn't a specific target for us. It's a plug based on our waterfall of priorities. The reasons for this ordering are clear without strong credit performance and solid unit economics, growth over the long term would be unsustainable.

To close, I want to acknowledge that we've experienced some setbacks in our business, but our fundamental economic engine is strong. Our risk models are better than ever, and I'm confident that we'll be on the growth path again soon. We're taking decisive action to bring committed capital to Upstart. And to those who say that we should focus on the traditionally prime market, I say that there are plenty of others focused on that. Improving access to credit for all Americans is too important to go ignored and Upstart has the right stuff to get it done. Thank you.

And now I'd like to turn it over to Sanjay, our Chief Financial Officer, to walk through our Q2 financial results and guidance. Sanjay?

Thank you, Dave, and thanks to everyone for joining. The environment we're operating in has continued to evolve rapidly since our previous call. Industry data shows a general rise in delinquencies across all segments of unsecured credit, disproportionately impacting the higher risk tiers that is comprised a significant component of our borrower base. The impact of this dynamic on the credit performance of Upstart loans can be seen in the supplemental credit performance information that was released today together with our investor materials.

The macro uncertainty and the impact of economic stress on consumer delinquencies have led to a decrease in available funding for loans on our platform which has become the operating constraints of the business. While today's results are in line with the preliminary numbers we preannounced on July 7, I will quickly call out the key financial headlines.

On the top line, origination volumes and revenue from fees were both down from last quarter and below our internal expectations, driven primarily by funding constraints in the capital markets. While profitability was also below guidance, we began to systematically improve unit economics in the second half of the quarter and have pivoted to optimizing for in-quarter cash flow generation, which will carry over into our third quarter contribution margins.

Following our recent share repurchase authorization, we have repurchased approximately 4.4 million shares of Upstart, totaling $150 million in repurchases. Additionally, we sold a meaningful amount of the loan assets from our balance sheet in Q2 in order to fortify our cash position. With these dynamics in mind, here now is a more detailed summary of our numbers.

Net revenues in Q2 came in at $228 million, up 18% year-over-year. Revenue from fees constituted $258 million of that amount representing 113% of overall revenue and up 38% year-over-year but down sequentially 18%. Net interest income was a negative component of net revenue this quarter as we entered into multiple loan sale transactions, some of which incurred a negative fair value impact and as the valuation marks of our remaining loans continue to be negatively impacted by the rising interest rate environment.

The volume of loan transactions across our platform in Q2 was approximately 321,000 loans, up 12% year-over-year and representing over 233,000 new borrowers. Average loan size was up 5% over last quarter, largely owing to auto loans representing a higher proportion of the mix. Our contribution margin, a non-GAAP metric, which we define as revenue from fees minus variable costs for borrower acquisition, verification and servicing was flat sequentially at 47% and 200 basis points ahead of guidance.

Without the inclusion of the fledgling auto loan volume, our contribution margin for core personal lending would have risen to 51%. As we optimize our fees and marketing spend for lower near-term volumes, we expect that unit economics will continue to show meaningful sequential improvement.

Operating expenses were $260 million in Q2, down 5% sequentially. We reduced our sales and marketing by 21% sequentially as we downsized our marketing campaigns to reflect our constrained funding supply.

Engineering and product development grew 14% sequentially and remains our priority area of investment. Although by the end of the quarter, we had slowed down hiring significantly and concentrated most of the remaining hires into key technical roles. Growth in general and administrative spend grew 8% sequentially. Taken together, these components resulted in Q2 GAAP net income of negative $29.9 million. Adjusted EBITDA of $5.5 million contracted 91% Q-on-Q. Adjusted earnings per share for Q2 was $0.01 based on a diluted weighted average share count of 93.3 million.

We ended the quarter with $790 million in unrestricted cash, mildly up from $758 million in Q1. Our balance of loans at the end of the quarter was $624 million, of which $484 million represented R&D loans, principally in the auto segment. While our balance of core personal loans at $140 million was only marginally down from Q1, we did sell a significant number of the loans that had accumulated on our balance sheet subsequent to the end of Q1, but prior to our earnings call in May.

Earlier today, we published some key data regarding the credit performance of Upstart loans. Just to recap a couple of the key points. Our models continue to provide around 5x the amount of risk separation than a credit score, and the statistical accuracy of our model continues to improve. This has not changed. Most vintages from 2021 will underperform their return targets. This volatility comes on the heels of vintages significantly overperforming targets for 12 consecutive quarters. Despite this latest volatility, an investor who invested equally across all Upstart cohorts would expect a 9.8% unlevered gross annualized return.

Notwithstanding the performance of the credit, we must confront the fact that the largely uncommitted nature of our third-party funding has proven inadequate to the task of navigating the current market turbulence. And we have turned our efforts towards building a more resilient funding model over time.

Despite not having suffered any adverse loan performance, some banks are moving to limit their overall exposure to unsecured lending. Investors who have earned significant excess returns during the benign cycle over the past few years are now anxious over the state of the economy and wary over the future prospects of less affluent borrowers who have been the most impacted by the termination of the stimulus.

Despite significant conservatism in our current underwriting and the prospect of historically high returns, investors have been reluctant to reenter the fray. Consequently, our intention is to significantly increase the fraction of forward committed capital deployed on our platform through partnerships with investors that are comfortable investing through cycles with an eye towards longer-term outcomes and in exchange for predictable future access to yield.

As Dave has said, this will not happen overnight. In the interim, we are prepared to be more proactive with our own balance sheet operations if we deem it necessary to provide a level of stability for the business in this transitional period as well as to demonstrate our own confidence in the models to the funding markets. Please note that this does not represent a change in permanent strategy, and we continue to maintain the view that it is not in our long-term interest to run a large balance sheet or to become a bank. However, in the context of the current extenuating circumstances, we will be flexible in determining whether a temporary change in tactics around balance sheet usage would be in the best interest of supporting the business through to its next state.

Given the volatility of the current funding environment and the difficulty in forecasting the timing of changing macro sentiment, we feel it is prudent to limit our guidance for now to the coming quarter and withdraw prior full year guidance. With that, for Q3 of 2022, we are expecting revenues of approximately $170 million, representing a year-over-year contraction of 26%; contribution margin of approximately 59%; net income of approximately negative $42 million; adjusted net income of approximately negative $9 million; adjusted EBITDA of approximately 0 and a diluted weighted average share count of approximately 85.5 million shares.

Our gratitude and admiration once again to all the folks at Upstart, who are remaining resilient through the choppy waters that we are currently navigating as a company and who remain as focused and with as much conviction as ever about the purpose and opportunity before us.

With that, Dave and I are now happy to open the call to any questions. Operator, back to you.

(Operator Instructions) And we will go first to Simon Clinch with Atlantic Equities.

I was wondering, Dave, or Sanjay, if you could talk a little bit more about what -- how you actually go about sort of refocusing your institutional buyer -- investor base to sort of longer-term investors. And I guess how long that might take and a sense of the steps that you need to take to achieve that goal?

Sure. This is Dave. I'll give a quick answer, and then Sanjay may want to chime in. Sure. So essentially, the nature of our agreements today, by and large, are at-will agreements with the volume that anybody any particular entity is originating or purchasing is decided on a month-by-month basis. And we're talking about instead about structures where there's committed funding over a significant period of time, many months or even years. And really, that's in return in some form for access to yield over that period of time in some form of economics that make sense for those entities.

So we don't have more specifics to share than that, other than certainly, I think a lot of marketplace businesses in many different types of industries take actions to secure effectively secure inventory on their platforms one way or another. And we've decided this is just necessary for us. And so we're beginning the steps toward taking -- getting that done.

Yes. I'll just add, Dave. This is Sanjay. I think that we have demonstrated and we'll be able to demonstrate certainly as we go through this cycle, pretty attractive long-term yields for anyone who's willing to hold an investor cycle and Dave cited some of those numbers and we've got some of those in the releases we provided. And so I think there's a class of capital provider out there for whom access to that would be attractive. And those are sort of more arrangements that will take a while to put into place.

But I think predominantly, what we have today are capital providers who are vintage by vintage and in some sense, may depend on either leverage or liquidity for the ABS markets, which creates more volatility. So I think now that we have some proof points which demonstrate what yield looks like through a cycle, we will use that to enter into negotiations and arrangements with, one is that are more of the style of wanting predictable stability in terms of access to yield.

And so I think that's all we really have to share at this time. As Dave said, these aren't going to happen overnight. They're pretty complex relationships, but I think we're already convicted that, that's the direction that will provide stability for our platform to get to the next level of volume.

Okay. All right. That's useful. And just as a follow-up, Dave, I think in your opening comments, you mentioned your view that we're heading for a significant slowdown in the near-term recession in the next 7 months. And I guess, a slower economic growth outlook beyond fiscal '23. And I was wondering if you could expand on your thoughts there and I guess, sort of what you're seeing and what gives you that much weaker outlook than perhaps I've heard from others.

Yes, sure, Dave. I wouldn't say that it's my outlook per se. I'm definitely not a macro forecaster and Upstart does not try to hold limited skills, kind of a crystal ball about the next phases in the economy. What I was trying to state, though, is that we try to build in what you would think of as some form of market consensus where the market thinks the economy is going to go with a degree of conservatism sufficient for banks and investors and credit unions, et cetera, to feel comfortable in the platform.

So we necessarily take what you might consider a conservative viewpoint on them, only because it's a good starting point for those who are on the platform with capital at risk. It doesn't necessarily mean it's my personal outlook or Upstart's personal outlook It really is just trying to reflect a reasonable and a conservative take on where the economy could be in the next couple of years.

And we'll go next to Mike NG of Goldman Sachs.

I just have to 2. First, could you talk a little bit more about the fee revenue as a percentage of originations. It's quite strong in the quarter. I was just wondering if there are any specific drivers that led to that increase, whether those are price increases or the loan mix? And then second, just on the guidance for the third quarter for $170 million of revenue. Could you just talk a little bit about the mix between fee revenue and net interest income, and any notable points around fair value adjustments?

Sure. Michael, this is Sanjay. Sure. On the first question, can I presume you're sort of talking about take rates. When you say strength, you're talking about -- Okay. Yes, I mean, I think it's as simple as -- and I think we've signaled this in the past. We've typically been optimizing for not interim cash production, but sort of long-term volume and how we price.

And so our take rates have generally always been at a level where we are able to produce more volume that will lead to model acceleration and learning and will lead to future value in the form of repeat loans. And obviously, in a situation like we're in today, where we are funding constrained and we're much more focused on in-quarter cash generation. We've sort of set our fees at a more optimal level, if you will. And so we've priced them higher, and that has the effect of creating a more resilient in quarter P&L. And so that's, I would say, is an artifact of how we're managing the business through the choppiness that we're experiencing in the market.

And the second question, really, was around the guidance and fair value in particular, I believe. I mean I think most of our revenue is an expression of transaction volume and fee revenue. There still is some downside in terms of fair value. We disclosed in our investor materials, the size of the balance sheet we're holding, which is pretty much on par with last quarter. And to the extent there's more interest rate exposure, if the rates continue to rise, those will necessarily depress asset values, create some fair value exposure but we're not making, I would say, large assumptions one way or the other about macro variables. We're really, I guess, trying to express the direction of transaction volume and the consequence of our revenue.

And we'll go next to Andrew Boone with JMP Securities.

As we think about you moving more loans on to the balance sheet, can you help us understand the guardrails that you're thinking about? I understood it's still early here, but how do we think about just what's the potential for using the balance sheet? And then as you talked about lenders and just the attractiveness of yields that are available right now, can you talk about just how you're educating your partners to be able to step back in? How can you proactively have them come back?

Andrew, this is Dave. On the balance sheet usage, I would just say, first of all, we will most certainly be prudent in usage of our cash in any way. As Sanjay said earlier, it is not our intention to become a large balance sheet lender whatsoever. Our long-term strategy hasn't changed. We aren't becoming a bank. But certainly, we see ourselves in a transitional phase where we are recognizing the need for permanent or more committed capital on the platform.

And as a bridge to that, we want to have the freedom to do the right things at the right time to get from here to there. It doesn't really change our overall philosophy, but we -- nor do we think it's great at this time to have sort of a litmus test of not using our balance sheet whatsoever with a lot of cash in our balance sheet, and we want to use it to the advantage of the business over the long haul.

So but for sure, we're a company that has always been very much capital efficient as a private company, what we raised very modestly compared to others how we used it. We've been profitable most of our time as a public company. So I think we have the genetics of the company that likes to be cash efficient, and we certainly will do nothing to put our operational capacity at risk or our business at risk with our balance sheet irrespective of whether or not we choose to use some of it within the marketplace.

Sanjay, do you have anything to add to that?

Yes. I mean, I guess, sure, just -- Andrew, I'll just reiterate. It's still our intention in the long term to be a platform that decisions third-party capital. We don't want to be in the business of being a balance sheet. Now as you've seen, we're sort of signaling a contraction in our guidance. We recognize that we need less (inaudible) funding, more committed funding and they get us through to that point.

I think what we're expressing is not so much an intention as a need for flexibility in making that transition. But as Dave said, we've always been very, very careful stewards of the capital that are on our balance sheet. We've always run a very lean company. So this is really more about just making sure we have what's in maybe on a stability of model in order to make that transition.

And then the second question you asked is how are we engaging the capital markets and the funding markets in order to provide them miscomfort. And there's a couple of ways. I mean, certainly, the amount of information that we've provided to the broader public in the form of sort of FAQs and the blog posts and the additional investor information we've released today, we have a much deeper level of information that we take to the funding markets. And in fact, in the same way that we hold sort of a conference broadcast in order to discuss business results with the equity markets, we're going to have a similar construct, with the funding markets.

And so we do have, I think, as much or more information about anybody as the current direction of delinquencies in very real time and we're going to do our best to express that. And then part of this goes back to your question on balance sheet in some sense that the markets would like a signal of confidence in the way that the loans are currently being priced and in the macro assumptions that Dave talked about and using our balance sheet to some extent as a signal, I think, can provide a lot of comfort to the funding market. So that's something that's not lost on us given the asymmetry of information that we have around how the models are calibrated and how the loans are currently trending.

We'll go next to Ramsey El-Assal with Barclays.

Can you give us some color on what you're seeing most recently in the business quarter-to-date in July?

Ramsey, this is Sanjay. Are you referring to any particular aspect of the business, meaning the financials, the sort of volume, the credit performance or just a general sort of?

Kind of -- it was quite general.

Okay. I mean I think the best expression of what we're seeing to date in the business is, I said, is sort of reflected in our guidance, if you will. So I think that, as we've said, there's a continuing contraction on the top line, which is evident to anyone. And I think that's probably the headline for what we're managing through right now.

And would you characterize that as having gone down and you're sort of seeing some stability in performance at this point? Or is it still something where you have relatively limited visibility as trends are sort of maybe unstable and still sort of on the move?

And when you say performance trends, are you referring to credit performance or...

I am referring to credit performance, but also, I guess, in addition, perhaps like the demand environment?

I don't think there's too much to comment on with respect to the demand environment outside of what we're signaling with guidance. I think that's probably the best reflection we have of it.

With respect to credit trends, I would say that the macro environment remains very fluid, obviously, and it's something that I think is changing month by month. And so I would characterize that as continuing fluidity. I would say with respect to how our model is sort of consuming and predicting the future, I think there's been significant recalibrations in our model since the beginning of this year. So when you therefore look at how the loans are performing against how they're being priced, I think there's pretty big changes in those curves. Maybe starting as sort of recently as January, February.

And so due to that, I think that the model has very much recalibrated to where the macro is. And as Dave said, in terms of how it's thinking about the future, there's significant conservatism and the assumptions around what will happen in the macro. And we don't have any specific knowledge or ability to forecast the macro than you or anyone else has. But with what we have and the trends we're seeing, I think that you could say that the assumptions are very conservative.

And we'll move on to our next question from Pete Christiansen with Citi.

To Dave, as it relates to the relationship with the CFPB, can you just walk through some of the changes that have been there? I know there's a bunch of nuances. But if you can give your take of how that relationship is moving? And then my second question as it relates to the 3Q guide, are there any assumptions that there'll be ABS issuance in the quarter?

Sure. Thanks, Pete. I'll take the first question. I'll let Sanjay handle the second one. We continue to have what we consider to be a great relationship with the CFPB. We've had that relationship since the very early days of the company through 3 different administrations. So we have a lot of history with them. We consider it constructive. We've always been very transparent and forthright with them.

As many know, we had this form of a no action letter agreement that started way back in 2017, renewed in 2022. And a few months back, we requested to terminated early, really in the sense that it was mission accomplished. It had done what we hoped in terms of getting a lot of feedback from the CFPB on how to properly test for fairness in sort of modern-lending model. And so through a lot of period of times in the years, we built what we consider to be very sophisticated forms of testing that we do on behalf of all of our bank partners.

So we have a continued strong relationship with CFPB. That structure of no action letters, et cetera, is something that CFPB internally decided they want to move away from. So I think that's okay by us as we said, we felt in the early days of our existence and before we had really refined how to do fairness testing right. It was very useful. But today, we continue to believe we have state-of-the-art fairness testing. We do that reliably on behalf of all of our lending partners, and we do continue to have open communication with CFPB and would expect to do so in the future as well.

Pete, this is Sanjay. To your second question, there's no explicit assumptions we're making with respect to ABS issuance in our guidance. And we continue to issue regularly, obviously, the execution in the market right now is quite volatile. But we don't have an explicit assumption on what that looks like or dependency on it.

And we will go next to Arvind Ramnani with Piper Sandler.

Just a couple of questions. As you're deciding to use your own balance sheet or use your banking partners or some of your loans, what are some of the trigger points you will use to kind of make that determination?

Arvind, this is Dave. Let's just say, I mean, we don't have specific trigger points per se. What we just want to really have is flexibility. I think being able to transition from one state of our funding supply to another is one we want to make sure go smoothly and with some confidence and just in getting from here to there in a way that's not disruptive to our partners, to our employees, to anything else.

So we don't have any definitive trigger points other than we absolutely intend to be cautious and prudent with the use of our cash. We are confident that there is real profits available on our platform today. So for that sort of basic reason that makes sense for us to do so. But it is not our goal to build a giant balance sheet, and then certainly not our intention over time is to sort of switch toward that form of the business.

We do believe that it makes most sense for us, for our employees, for our shareholders, for all of our partners to have some flexibility in how we navigate through a pretty unique economic time that we are sitting in today.

Yes. That's helpful. And is that something that you will sort of plan to communicate to investors as you look to expand the balance sheet? Or is it kind of going to be part of the regularly scheduled earnings calls, if you plan to go that route?

This is Sanjay. I think it will make a component of our regular communications with the market. I don't necessarily foresee anything that's so extraordinary that would require an interim communication. But if there is, we'll certainly make it.

Just as an aside, I don't know if you've seen it, but we are sort of breaking out the balance sheet and the exact components of it in the investor materials now.

Yes. Yes. I did see that. And then just one of the things that you talked about certainly was how you continue to see your models better equipped to price loans. And I know in prior earnings calls, you've talked about some banks preferring to you Upstart versus like a FICO score.

But are you seeing similar validation through your partners that, that may sort of ease up the funding sources in terms of like sort of really validating that your models are better to price loans and to shift volume your way?

Well, Arvind, one of the things that we've kind of tried to make is pretty clear is that the 70-plus banks and credit unions who tend to originate and hold the primary end of the credit have all done really well through all parts of the cycle, if you will, and have actually performed at or above expectations. So I don't think we necessarily need any more than that. Some of them have stated it publicly, and we can see it in the data, and I've shared the aggregate data.

So I think that that's all a very good thing. But I think a lot of the issue out there really is about what may happen in the next year or 2 years. And everybody has the right to have a different opinion about that and take actions based on that opinion. So that's part of, of course, the challenge is -- it's about the future, not about exactly what's going on in the last year or 2 years.

And in that sense, that's why we have sort of said we want to move toward investor relationships that have a long-term approach across through-cycle approach toward investing, and that will be in the end, lead to a much stronger platform for Upstart.

We'll go to our next question from Vincent Caintic with Stephens.

I have 2. First question, so on the balance sheet usage. So I appreciate your comments on that. And if you could talk about your balance sheet strength. So you've got over $900 million in cash. And I guess if you were to leverage that, conservative leverage, maybe you could do $2 billion to $3 billion of originations to support the business in the interim. So I'm just wondering if you can maybe talk about some of the guardrails or sort of how you're thinking about the balance sheet usage.

And then relatedly, I think you spoke about in the prior press release about selling some of the loans that were on the balance sheet. If you could talk about how that performed. I see there's still about $600 million this quarter. How was the -- how did that go? And what was the, I guess, the par value?

Vin, this is Sanjay. So just to maybe put some parameters on our balance sheet. That's correct. We have sort of the $900 million in restricted and unrestricted cash. I would say we have about probably $400 million of loan equity on the balance sheet and about $600 million of assets. So maybe about $200 million of that is financed. That will -- we don't have an intention of getting into large amounts of financed loans. So numbers that you alluded to on your -- and I don't think it would be anything approaching that. I think it would be much more modest.

And this is an environment which (inaudible) average and candidly, it's not readily available these days at reasonable prices anyway. So I think of our loan balance sheet sort of flexibility has been denoted in values of maybe a couple of hundred million dollars.

With respect to loan transactions, I would say transactions, like the main force of gravity on those is what's going on with rates in the environment as you're transacting. In most of the transactions we've had have been older vintages, so vintages that have accumulated maybe in Q1 of this year and 6 months later, interest rates have gone up. And so to the extent that they have executed below par and we've indicated that, that has created some of the negative fair value pressure in our P&L. It's really a function of the fact that some of the more seasoned loans have just been impacted by the extreme environment this year.

Okay. I appreciate it. And one question, just following up on guidance. Just if you could talk about. So the $170 million in revenues, if maybe you can talk about the cadence of that? Like are you seeing an improving performance as we go through the quarter. And on the contribution margin. So I'm calculating 59%. So a nice expansion there. Just wondering what would be driving that maybe less marketing expense or more efficiency, if you could just talk about that.

Sure. Yes. I guess in terms of the guidance, we're not really telegraphing any directionality. I would say things are volatile right now. And so that's more of a level than a trend, if you will. With respect to the contribution margins, yes, I mean it's sort of a bit what we referenced earlier, which is when we're in a period of contraction like this, we optimize for in-quarter cash generation. And candidly, it's one of the important economic characteristics of the business, which is we can -- we're essentially suffering or sort of telegraphing a roughly 50% contraction between the forward guidance and what we did last -- or in Q1, and yet we can weather that with still guiding to breakeven EBITDA.

And the reason is because we have quite a bit of control in terms of our ability to set fees. We tend to be inelastic and below optimal fee levels in normal times, and so we can raise them to buffer volume contractions. And as you said, when we have less funding availability, our marketing programs, we tend to sort of keep the more efficient ones and discard the more experimental ones. So as a result, our take rates go up, our acquisition costs tend to go down and create some margin expansion, which in some sense tends to push against the volume contraction, and it allows us to be somewhat resilient as a business model.

And we'll go next to Jim Faucette with Morgan Stanley.

Thanks for all the detail and supplemental information. I'm wondering, when we look at kind of your expected returns on by quarter, et cetera, is that you're showing that you expect a pretty significant improvement on the cohort from Q1 '22 versus Q3 and Q4, even though the total -- or the target gross return is similar or even a little bit lower. Can you talk a little bit about the changes that were made for that Q1 '22 and that are driving your expected return higher versus what was being done in the second half of last year?

Jim, this is Sanjay. Sure. I mean, the simple version is you're seeing the model recalibrating to the changing environment. And in particular, I think starting in probably Q3 of 2021, the delinquency trends in the industry started to rise. And it's been disproportionately, I'd say, born by the, I would say, the less affluent borrowers, if you will. And so our models observe that and react to it and change pricing as that is happening.

And so in some sense, because that trend had happened between Q3 of '21 and early this year, maybe, call it, Q1 or Q2 of this year, our model has been reacting to that, adjusting, recalibrating. And on top of that, we do what you might think of as a manual overlay, which is we have to make some estimate of what we think the future macro holds because that's not something that's in the training data for our machines.

And so in addition to the model we're calibrating to loss trends as they change, we are making more conservative forward predictions what the macro will sort of do in the future. And we're to the point now where I would say our macro accommodation is fairly conservative in terms of what we're expecting to happen or maybe expecting it the wrong way, we're prepared to happen in the macro given inflation and unemployment, et cetera.

And so those are really the 2 variables that are changing. It's the model's estimate of loss depending as a function of the changing actuals and it's our forward prediction of how to prepare for macro -- continued macro volatility?

Got it. And then when you talk about looking to add committed capital and expanding that range of partnerships and agreements. What's your expectation for what that's going to do on cost of capital and what you would need to do, if anything? And if there is an impact, what you would need to do around fees and those kinds of things? Just trying to figure out like what you may have to give up in order to achieve that longer, better or stickier capital, I guess?

Yes, it's a great question, James. I would say that -- on the one hand, it will make notionally -- certainly to a capital provider who's committing capital forward, there's cost for that. I think it will make capital on the margin more expensive in good times, but maybe on the margin less expensive in times like now because they're investing through the full cycle.

And the other variable is that as we sort of talked about in some of the other questions, we are quite margin rich. And so we ourselves can trade off economics in the good times or economics in the choppier periods in the next economic cycle such that the investor is whole, and we're creating more stability for the platform. So I think to the extent the committed capital is more expensive in the benign period, we can offset that through our own business model and then make it up when there are down cycles as they inevitably will be.

So we view our margins to be a lever that we can use to make sure that the investors are stable and the platform is stable and the borrowers are getting some amount of stability. Even though, as you know, our mission is to fundamentally lend to for a proportion of the population, which notionally are riskier.

And we'll go next to Nat Schindler of Bank of America.

I think a lot of my questions have been asked already. But one thing I wanted to go over, well, 2 things. One, you said that you're going to go out to lenders and you have now a cross-cycle vision of your performance. Are we really cross cycle? Are you modeling that this is the bottom, and we've turned the corner on kind of the low-end consumer that is contrary to what most of our economists are saying at this point? So I wanted to just understand what you mean and whether or not you think that these delinquencies are going to get worse from here or better.

Finally, also, I'm a little confused on the back and forth of the balance sheet, no balance sheet, using the balance sheet, not using the balance sheet. One, why this oscillation? And two, how much of your -- how much do you really think that you're ever going to get to on their balance sheet? I mean, originally, you were saying you're going to be only about 5% and that's for experimental purposes. I think in Q1, you got up to 15%, and you said you would go down from there. While that was experimental with auto anyways. And now you're saying you're going to go onto the balance sheet again. What kind of level do you think makes sense?

One final question on top of that, and I'm sorry for the 3, but who are these investors that you haven't talked to? Who are you going out to, to kind of increase your supply? Just are they people that don't do personal loans? Are they people that are new to this market? Who is out there that you are trying to evangelize the product to on the supply?

Nat, this is Sanjay. I'll take your first question and then Dave and I will (inaudible) on the last 2. The first question was about do we really think we're at past the bottom of the cycle. I think in a short word, no. But I'll just give you our view on a couple of different nuances. So first of all, I do think that what's happened to date. While in the aggregate, I think that maybe there's a view that it hasn't been too bad to date, and there's more to come. I think the picture is sort of a little bit different in a nuanced way depending where you are in the credit spectrum.

And if you look at the industry data, if you're an affluent borrower and you talk about like what has been the best sort of credit performance in the last year sort of mid-2021? And where are you compared to there? You're up mildly, you're up 30% and you're still below pre-COVID numbers. If you are on the less affluent side of the spectrum, you were up much more than that and you're well past where you were pre-COVID. So I think that the timing of all this is a little bit different. I think that the less affluent side of the spectrum peaked a little bit sooner and has come up a little bit sooner and maybe prime borrowers are still sort of in that schedule somewhere.

So -- and then I think the second component is we've largely -- to us, it's not just -- it's the rate of change of things and the very sudden changes in the delinquency trends in our core borrower we've sort of cut up to and we're making significant conservative assumptions about the future. So I think we're expecting further degradation in loss trends. But I think that the difference is compared to like mid-2021 when everything was rosy, and everyone still have a lot of stimulus in the bank account and delinquencies were still like 50% of long-term normal. We are now projecting it and prepared for it.

So there is still a world in which these vintages underperformed going forward. But it would have to be a very, very significant economic sort of step back in our view. And so what we really care about demonstrating to someone who will be forward-committing capital, in sort of a cross cycle kind of a way is like what's the cycle of the returns. And if you look at the sort of returns that we -- by cohort that we talked about in the investor materials, you can see that it's sort of troughed with the Q3, Q4 vintages. The Q1 looks like it's sort of working its way back. And I think that you'll see that when you see those Q2 numbers. Our belief is we will be able to tell a story that the returns have gone through the cycle and we're on the front edge of what's coming.

And of course, you can never fully predict the future, but just the preparedness of our models and the conservatism of the macro assumptions are such that I think we're ahead of the curve now. So that's the question with -- so that's the answer to the first question. I think your second question was about balance sheet usage. I don't know, Dave, do you want to chime in on that?

Yes, sure. And Nat asked, I don't know exactly how to put it, what's with the back and forth on balance sheet usage is maybe the short version of it. But Nat, look, I'd just say we're -- the company I founded -- cofounded 10-plus years ago. The thing that's made us successful over the years is being nimble and thoughtful and responsive when needed by the market. And we certainly have a long-term view that we are a market-based business, meaning to bring lenders and investors together with consumers, borrowers on the other side, and that's how we view ourselves long term.

Having said that, we came to the conclusion that having a sort of a litmus test that thou shalt not use the balance sheet at all, doesn't make sense, particularly when you see reaction to the supply side of our business happen in ways that almost the contrary effect to us, meaning credit is generally performing well. And yet for reasons that are obviously -- many of which are beyond our control, lenders or investors take decisions that they take.

So I just think in the interest of our shareholders and everybody else, it's prudent for us to use our balance sheet wisely. We don't intend to become a large balance sheet lender. But at the same time, we think it's important to state to the community out there that we're going to do the right things for the business, and we don't believe it makes sense to have a sort of ironclad litmus test that thousands shalt not use your balance sheet in the marketplace. And that's the conclusion we came to. We think it really is, as we've said numerous times now, a transitional step toward a place where we have committed third-party capital. And that's what we think also represents a significant chance to upgrade our business structure.

And so put those altogether, and we think this is the right step forward for Upstart. It's going to be part of building a much larger, a much stronger Upstart over time.

And we will go next to David Chiaverini with Wedbush Securities.

The first question is on the take rate, and it makes sense to increase the take rate in this environment. I was curious, how high should we expect the take rate to go over the next couple of quarters?

David, it's Sanjay. I think maybe one way of putting this is, given the guidance we've put out for Q3, you might imagine that we're doing all that we can to ensure that our P&L remains resilient.

Got it. And then somewhat similar question on originations for the third quarter. I guess, zooming out back to the second quarter, are you able to share what the June origination number was and how that compares to July?

Yes. We're not breaking out specific months, David. I would just -- I would say that the trend right now is volatile. So there's no real directionality in our numbers with respect to June versus July.

Got it. And then quickly on the buyback, can you talk about your appetite from here?

Sure, David. Well, as you know, we have Board authorization to go up to $400 million, of which we've used $150 million. I mean, it's sort of an ongoing equation between where the price is versus what we think is fair value. And now, of course, what other sources and uses of cash or I guess, uses in particular, we might have. And so there's always opportunities, whether it's buying back the stock and reducing the dilution or looking at some of the convertibles in the market that are ours or buying loans, which, as Dave said, we think can have a very lucrative return right now. So I would just say that we're monitoring that on an ongoing basis, and we're certainly interested in making efficient use of our cash balance on behalf of the shareholders.

And now I will turn the call over to Dave Girouard for any additional or closing remarks.

Thanks all for being with us today. Sanjay and I just want to make clear, we're not happy with our results. We're not a company that likes to have a declining revenue from 1 quarter to the next. But we are -- we feel doing the right things to make the company as strong and as powerful as it can be in the future. We are as committed as ever to the mission of improving access to credit for those who deserve it, and we are making steps to make the company stronger and better and do anticipate we'll be back on a growth track. So we appreciate all of them sticking with us through this, and we're going to get back to work at making Upstart the great company that it is.

And ladies and gentlemen, this concludes today's call. Thank you for your participation. You may now disconnect.