Bert Hochfeld Positive on Upstart

OK - Upstart is not looked on fondly by many on this board. Saul has compared it to Fastly. John Wayne (whom we’ve not heard from for some time) initially provided incredibly insightful positive analysis on the company only to excoriate them repeatedly after their greatly disappointing 3Q21 report. At the risk of being considered a hypergrowth heretic I admit that I have maintained a small position in Upstart (3%).

Those of us who paid attention watched the stock get killed the day after the quarterly CC (August 9) only to soar by 17.75% at the close of the market today (August 10) based on what appears to be moderating inflation suggesting the Fed my ease up on boosting interest rates (I guess).

But we don’t invest in stocks here, we invest in companies. So I would like to make a case for Upstart.

My first encounter with Saul was several years ago (I think in 2012, prior to this board’s existence) when, as a MF adherent I read a post written by Saul on one of the discussion boards about a company named Westport Fuel Systems (WPRT). This company was a MF darling with many avid followers. Saul very calmly and very logically pointed out why this company was a bad investment. Westport was trading in the $40 range at the time and due to very positive MF analysis many folks (myself included) thought this company was destined to be a multibagger. Saul had the temerity to express a contrary opinion. His post was not warmly received. There were a lot of heated, emotional retorts to his post. My recollection is that I posted my gratitude for his insightful post. I sold my shares the next day.

So with all due respect to Saul, I have a contrary evaluation of Upstart. I do not see it as story stock like Westport. I do not believe that the comparison to Fastly is valid.

Westport was a story stock. If you’re not familiar with the term, it refers to stock issued by a company that purveys a promise, usually based on some technological breakthrough which they do not possess. But, they’re engaged in R&D with very positive results. Maybe there’s an issue with scale. Lots of technology can be demonstrated in the lab, but that’s very different than a marketable product. Maybe, the lab results aren’t even realistic (think Theranos and Elizabeth Holmes). Maybe there’s some other problem which the company typically either doesn’t acknowledge or downplays. In any case, the company is peddling a promise rather than a product. They have a great story and the tell it very convincingly.

Upstart has a real product. They have developed and brought to market an AI/ML based product that does a better job of performing risk assessment associated with lending than the competition (i.e. FICO). This is an inarguable fact. Some critics have challenged this fact, but that doesn’t invalidate it. Upstart is not a story stock. Most certainly, the company has some problems, but the problems are not made of undeliverable promises.

The comparison to Fastly might be closer to the mark, but again, I don’t believe the analogy to be valid. Unlike Fastly, Upstart is not trying to focus on a small but elite customer base. Quite to the contrary, they are making serious efforts to have a broad base of customers. For example, they have developed an interface for Spanish speaking potential customers, a greatly underserved segment of the lending market. They are expanding into auto lending, small dollar lending, SMB lending and in the not too distant future, mortgage lending.

However, at the present time, they have curtailed S&M activities as their greatest single problem right now is the source of financing in support of lending. Their banking partners (which admittedly have been spinning up very slowly) have reacted negatively to the current macro-economic environment. Upstart’s management recognizes this problem and they are taking steps to address it.

So, IMO, Upstart remains an unique, disruptive company in the world’s second oldest profession - lending, the foundation of capitalism. As with all market timing issues, I can’t predict when Upstart will establish long term partnerships with sources of financing. So, I am willing to maintain a small position in the company commensurate with my confidence level.

Bert’s public SA article goes into greater detail. It is linked here: https://seekingalpha.com/article/4532689-upstart-the-bad-new…

98 Likes

Hi brittlerock,

I really like Bert Hochfeld and admire his often good and helpful research. In my opinion, he has only one weakness - he is selling too late. Until recently, I had a tiny (<0,5%) watchlist position in Upstart. I sold it after listening to the earnings call. For me, it was a real pain to listen to. Why?

  • They changed from “we sometimes hold loans on our balance sheet“ to “we are a software company and no bank - that’s why we don’t hold loans on our books” and back to “we changed our mind and now hold loans again” in just 6 months!!! That’s unprofessional to say the least - how can you believe them going forward when the management flip-flops like that? No go for me.

  • Zero visibility. Revenue miss and huge guidance cut again. After announcing preliminary results only 4 weeks ago!!! They seem to have no idea about what’s going on in their business. Sure, it’s a really cyclical business, but you really see now that they are no bankers who saw several economic cycles in their career.

Upstart is a pure turnaround speculation and I really wish they’ll make it. I think their good intentions of giving all people access to fair credit are honest and no marketing thing.

IF their AI really proves to be MUCH better than FICO in every market environment + they get better as a management team, there will be plenty of time to get in again.

Right now, it’s no place to put my money in. As jon wayne said already, they are hyper-negative growth right now and therefore don’t belong on this board anymore.

Hannes

32 Likes

it wasnt just a story stock.It was a classic pump and dump stock by FinTech-Community.

Read the 10-Q. The most interesting data point is that on a 6M YTD basis 89% of origination # loans still comes from Cross River Bank and Finwise. Q1 same data point was 90% so this implies Q2 was still ~88%.
This is despite Upstart increasing bank partners by 70% YTD (from 42 to 57 to 71) and despite origination # loans decreasing 31.5% sequentially (per IR slide 13).
This data shows that Upstart’s real bank partners were hitting the brakes just as hard if not harder per bank than the company’s institutional loan buyers!
This also shows that the incremental value of adding more bank partners into the B2C Upstart Referral Network is very low. This data suggests the real bottleneck for Upstart is better credit quality borrowers that bank partners want. The more bank partners is just a smaller slice of the same pie for each partner. Another mouth to feed. This is a fundamental problem with the company’s business model!

And I don’t consider the management to be very competent, it may be that they are technically (KI/ML) very good but have no idea about there core business and macro events. For example:

“During this transitional phase, we may choose at times to leverage our own balance sheet to bridge to this newer model.” CEO speak he’s really saying our business model DID NOT WORK so we are trying a new one.

How are you going to rate UPST now? Is it a bank or a pure software company? I don’t think UPST really knows either…

11 Likes

I read one write up where the author said one of the problems is that both CEOs are from Google. No fintech experience, which I believe to be true. Seems as though they are figuring it out as they go along. I guess the lesson here is when something seems to good to be true it probably is. Many were dazzled with the company, its potential and its numbers, along with myself, Tom Gardner, Bert, and others. I finally sold out, at a big loss. I wish all holders the best, I hope they pull it off, but this one got thrown in the too complicated pile for me.
Jille

8 Likes

Upstart investment thesis is based only and only on how their ML algorithms perform.
If you have conviction that ML algorithms over long term will do better, you should buy.
The stock is reasonably priced with tons of upside potential.

I work in the ML field. Although this will be a 2-3 year journey, in my opinion upstart models will outperform competition by a mile.

Patient investors who are accumulate will do very well. Buy on dips. Sleep well. Get rich slowly.

30 Likes

The key issue to appreciate re Upstart is the accuracy of their credit assessment vs Fico. See the Q1 presentation and the separate Q2 credit report. This is what distinguishes Upstart from the rest of the field.

The second issue is that Upstart did not appreciate the separate impact of funding in a period of credit contraction. Their response is to accept that and try to get better commitments re funding. And ensure it can continue to measure credit results by using its own balance sheet for a temporary one-off period.

Credit tightening will probably mean credit reductions for the next several years and Upstart’s earnings will end up being cyclical to a greater or lesser extent. But the market opportunities are massive if the credit assessment outperformance is maintained.

I sold out Upstart shares last year at a profit and have just started buying the stock again.

5 Likes

I honestly do not understand the rationale for being invested in Upstart at this point. The management team clearly demonstrated that they do no understand an incredibly important part of their business - investor demand for their loans. For a while I thought they were lying to us when they claimed that their business does not depend on ABS issuance much but now I know that they were just really ignorant of how that market works. From my perspective there is absolutely no excuse for that. 6 months into this correction they still have little to no idea for how to solve their main problem. Their solution for getting permanent capital sounds great on paper but naive and unrealistic in the current enviroment. I find that astonishing. Maybe they should just realize and admit that there is no good short term fix. For me, they have a long way to go to prove that they are capable running this business in a way that generates good risk adjusted returns. There are plenty of other fiancial companies that can make big profits when times are great - think RKT. UPST may argue that their models are better, RKT may argue that the way they process mortgage applications is superior from a technology perspective. No matter what, results is all that matters and UPST results speak for themselves.

10 Likes