UnitedHealth and SOFI Earnings Review: A Case To Be Greedy and A Case To Be Fearful

Be greedy when others are fearful; be fearful when others are greedy.

Jul 30, 2025

It’s an irony that Sofi and UnitedHealth announced earnings on the same day, and I got the chance to write about them in the same earnings review.

Why is it ironic?

Because these two companies were in exact opposite standings until a year ago, and the reaction to them back then and today illustrates exactly what’s wrong about the markets and why most investors lose money.

I first bought SOFI in May 2023.

At the time, it was trading at around $5 per share. It was down 80% from its highs, yet the growth and execution looked amazing.

The narrative was that the company could go bankrupt. I am not even joking.

We were approaching the end of the Fed’s interest rate hikes. The rates were projected to settle around 5%-5.25% and stay there until the inflation decisively heads down. A few analysts on Wall Street thought that this would lead to an explosion in SOFI’s non-performing loans, ultimately leading to the demise of the company.

Yet, the data and the management said the exact opposite.

The average income of their borrowers was $164,000 and the average FICO score was 747 at the time. They also had a hard cut-off for FICO scores below 680.

This was more than a solid reason to be confident in SOFI’s loan portfolio for one simple fact—its average borrower was considered a “Prime Plus.”

As you see, even in the Great Financial Crisis, the 90+ days delinquency rate among Prime Plus borrowers was just 0.12% which peaked at 0.97% in Q3 2010. Yet, this wasn’t an ordinary crisis or a sudden economic shock; it was the US’s second biggest economic crisis in modern times.

In 2023, SOFI’s student loan delinquency rates were consistent with the above table:

Its personal loan portfolio, made up largely by Prime Plus borrowers, had a 90+ day delinquency rate of 0.40%, meaning it was performing at the Prime level.

On top of all these, management backed off from lending and put in place hedges, assuming that we would finish 2023 with 2% contraction in GDP and over 5% unemployment. At the time, the Federal Reserve was expecting 0.6%-1% growth in GDP and just over 4% unemployment, meaning SOFI’s estimates were incredibly conservative.

You could look at the data and say that there was a 5% chance that we would get a crisis worse than 2008. In every other scenario, SOFI was a bargain. That would lead any reasonable person to get at least a small exposure.

This was the picture I was looking at in 2023. I bought it with great confidence and conviction. My average price was around $7 a share.

In the next 12 months, the company only reported a triple beat after a triple beat.

Stock? It did literally nothing. It closed in September 2023 at $8 while the closing price at the end of September 2024 was $7.87, meaning it was in the red after a year.

If I sold back then, I would have missed the 300% rally it delivered since then!

My position had done nothing for a year, then it quadrupled in 9 months.

While SOFI didn’t get any attention for a year, UnitedHealth was a market darling.

It had missed earnings only once since 2008, and it was at all-time highs in August 2022 while the whole market was crashing.

The median P/E between 2021 and 2025 was 26, and it had become the largest position in the Dow Jones Industrial Index. It looked invincible:

Ironically, the roles have changed now.

People now think SOFI is a great company, and the stock is skyrocketing, while UnitedHealth is structurally broken. Everybody is flocking to SOFI and avoiding UnitedHealth.

Here are the facts:

SOFI was a great company back then; everybody was avoiding it, and it’s a great company now.
UnitedHealth was a great company while everybody was blindly buying it, and it’s still a great company despite everybody avoiding it.

This is exactly how you make money in the market. When everybody is betting on something, you take the opposite side of the bet and come out right.

There is no other way.

If you aren’t right, you don’t make money. If you bet with everybody and everybody is right, you don’t make any money.

You may be thinking, ‘betting against everybody and coming out right? It sounds hard?’

Well, if you haven’t conceived so far, here is the ugly truth for you—making money in the market is so damn hard.

I don’t know how most people have come to believe that making money in the market should be easy.

It’s so freaking hard. Only 2% of the hedge funds consistently outperform the market. Yet, most people still act like it should have been easy and there should be no pain involved.

This is bullshit.

You may ask, “Why does it need to be so hard?”

Let me put out another ugly truth for you—when you make money in the market, it’s not the treasury that pays the rewards, you pocket other people’s money.

There is no new money created in the market. Those who are right get the money from those who are wrong. It’s the freaking Wild West.

If you were short SOFI and your GenZ child was long, he pocketed your money. If you were short ULTA and your wife was long, she pocketed your money. That simple.

Naturally, everybody does their best not to let you pocket their money.

This is why it’s so goddamn hard.

So, that’s settled—contrarianism is a must, not an option.

If you are a long investor—and I recommend that you be only a long investor—there are only two ways:

The bar is too low, and you bet the company can do better.
The bar is already high, and you bet the company can do even better.

Decades of market history have taught us that the first one is way easier than the second one. Every penny I ever made in the market, I made it on this premise.

I have never made a penny by betting on a 45% 5-Year CAGR while the market already priced 40%. It never happens.

I made money in SOFI because I bet on survival when the market was pricing the stock to death.

The market sometimes acts overly pessimistic. When it happens, you get a bit optimistic, take the other side of the bet and wait patiently until it plays out. This is your best chance to make money in the market.

So, read the rest of this post with this lens, and you’ll get the most out of it.

1️⃣ UnitedHealth Is Sick, But It’s Not Dying Soon

People think owning an exceptional company is easy; I think the exact opposite.

Owning an exceptional company is as hard as owning any other company.

It’s like having an exceptionally talented and successful child. They get you used to spectacular results, and once they stumble, you are afraid that they’ll go all the way down.

This is what happened with Netflix in 2022:

It plummeted 70% from its highs in just 6 months because it lost subscribers for three consecutive quarters from Q1 2022 to Q4 2022.

Most, and I mean most people, thought the business was broken and the industry was saturated. Twitter was filled with such posts:

This wasn’t the first time it lost subscribers, but it was the first time since 2011.

Its shareholders had become complacent, and they had full confidence that it would grow forever. When it posted a bad quarter, everybody thought we had passed the peak point for Netflix, and it would only go down after that.

If it wasn’t, why would the growth engine working for 10 years stop? It was surely broken.

We all know what happened to the stock. If you invested back then, your money would make 5x in just 3 years.

Same goes for META.

People were so sure that Meta, and all its apps, were dying.

Reason? It had reported a declining revenue for the first time ever and provided weak guidance.

If something that was always increasing suddenly stops, it should mean it’s terminally broken, right?

We all know what happened to META stock since then. If you bought at the 2022 dip, you would have made 7 times your money. Not bad.

In both cases, people saw the incredible past performance of these companies as proof that the business is broken rather than a source of confidence for the future of the business.

UnitedHealth is no different.

This is the first time they have missed earnings estimates for two consecutive quarters since 2008:

The result is no different. Everybody thinks the business is broken.

There is no question that the earnings were disgusting.

Here are the numbers:

Revenue of $111.61 billion vs Est $111.59 billion ✅
EPS of $4.08 vs Est $4.45 ❌

It didn’t just miss the earnings estimates, it missed them a big time. Its EPS is now down 40% YoY and 43% QoQ:

This is not something that came with signals; this is an acute problem.

If this were a tech business or a software company, you would either think it’s getting disrupted by a new revolutionary service, or it has lost some of its largest customers at once.

In these cases, you would rightly question the future of the business. Yet, in both these cases, you would also see the top line deteriorating.

Do we see deterioration on the top line for UnitedHealth? No.

The top line is growing.

The problem is acutely eroding profitability. Why? It’s because of the business model.

In most business models, you price by knowing your costs. The loss or profit is realized at the time of the transaction. In insurance, you don’t exactly know your costs; you have to predict them and set your prices based on those predictions.

Once the price is set and the policy is issued, the only lever you can pull is cost control, and it’s not infinite. If the costs and client activity increase above your estimates, your insurance-related loss increases, narrowing the profitability.

This is exactly what’s going on with UnitedHealth:

UnitedHealth’s medical care ratio has skyrocketed since 2023. It was consistently around 82% until 2022. In 2023, it decisively broke above 83% for the first time in years and hasn’t looked back since then.

What happened in 2023?

Well, 2023 is the last year UNH somehow accurately expected the medical inflation and priced accordingly, ahead of the costs.

This is the good news. We know exactly what the problem is, and the business knows it too.

What they need to do is simple:

They have to revise their predictions.
They have to raise the prices accordingly.
They should tighten their actuarial standards and avoid clients with high expected activity.

The good news is that there is nothing structurally broken in the market or the business. Customers are out there paying for its products, but the costs are out of control.

This closely mimics the problem META had in 2022. Costs were out of control, and earnings were taking a hit:

Take a look at META’s earnings plummeting from $3.64 in Q4 2021 to $1.64 in Q3 2022. Earnings plummeted 45% and the stock price plunged by 75%.

Stock price follows the earnings. Simple as this, and the narrative forms around it.

In 2022, you could see many stories why META was poised for a secular decline, antitrust investigations were suddenly popular again, etc…

But the core reason was simple: Costs were out of control, earnings took a hit, and the stock price followed.

META’s P/E ratio dipped to 10.9, incredible, right? How many people can now think META could possibly trade at 11 times earnings?

Now, remember what I said about the low bars above in the intro.

At the time, the bar for META was very low. The right play was betting on it to jump over that bar, regardless of the result. Those who did that reaped very handsome rewards.

Now the bar is too low for UNH. It’s trading at just 11 times earnings, at decade lows.

If it can jump even this very low bar, the return to shareholders will be immense.

Let’s value it, shall we?

I’ll put an even lower bar and assume a mid-single-digit growth rate for the next decade, stable operating margin, and a final ROIC of 15%.

As you see, even at a mid-single-digit revenue CAGR for the next decade and without any expansion in operating margin whatsoever, we get a fair value of $490.

Assume a 30% margin of safety, you get an entry price of $343, at 40%, it would be $294.

At the current price, you are getting a freaking 47% margin of safety!

In my book, this is where you should be greedy.

Now, some people may ask, ‘Why are you feeling so strongly about UNH? Why aren’t you looking elsewhere?’

The same reason why I picked and defended SOFI back in 2023.

What measures the success of an investor is the long-term result. I think this should also be what measures the success of a publication about investing.

Our portfolio here, which is open to premium members, is on its way to delivering a nearly 50% annualized return in the last three years.

This is what measures our success.

I would have made my money on SOFI, Hims, NBIS, MELI, or others anyway. But I did that publicly, tried to explain, and battled others because this is also what measures our success as a publication, and we have a commitment to be successful.

This is how we deliver our returns, this is the only way I know, and beyond that only way that has been proved to exist so far.

This is why I am battling.

Anyways, the bar for UNH is now too low.

You may bet on it to jump a very low bar, or seek others that have higher bars and bet on them to jump even higher. You’ll have to pick one to make money for sure.

In my experience, the former generally works if you are patient, while the latter rarely works, and it almost never works consistently enough.

Your choice.

2️⃣ SOFI: Poised To Compound At 25% For Decades!

Let me be straight—it’s all heading in the right direction!

I was one of the first people on Substack and Twitter back in 2023 who actively propagated for SOFI. I remember a few more names, such as

Data Driven Investing

and Breed Freeman. Most people, on the other hand, thought SOFI would be a failure.

It’s truly fascinating to see that this company has executed beyond my most optimistic expectations in the two years since I first bought it in 2023.

For the eight quarters I have owned it, it has only posted a double or triple beat, not a single miss on either the top line or the bottom.

It hasn’t surprised this time either and delivered blockbuster numbers:

Revenue increased an incredible 44% YoY to $858 million.
EPS increased eightfold from $0.01 to $0.08, growing 33% QoQ.

These are truly amazing numbers, but I think what’s under the hood is even more amazing for SOFI.

Let’s think this through, shall we? SOFI is a digital bank with a full suite of fintech products. What do you think is the most important metric for such businesses?

It’s the member growth.

If you are rapidly adding new members, you’ll have a dual growth engine provided that you effectively upsell and cross-sell to your existing customer base. The new members will become an upselling or cross-selling opportunity too in a few quarters. This is why it’s the metric I most carefully track—and SOFI crushes on this!

It has grown members every quarter since it became public:

It added more than 800K members in each of the last two consecutive quarters. This is the fastest rate at which the company has grown its members in any two consecutive quarters ever.

This degree of growth normally comes with certain challenges in financials:

Customer acquisition is expensive, so the costs may get out of control.
Customer quality is always a source of concern; low-quality customers may lead to a jump in charge-off rates.
Service quality issues are endemic in the industry.

I think the real magic of SOFI is that it’s been navigating all these so well that they almost look non-existent. Despite the stellar growth and still high-interest rates, SOFI’s charge-off rates kept declining consistently:

Its personal loan charge-off rate has steadily declined since Q2 2024, and the student loan charge-off rate is still under 1% despite the jump in this quarter. Both personal loan and student loan portfolios have an average FICO score above 740 and a weighted average income above $130K.

This is some high-quality loan portfolio that you may see in elite banks like Chase.

So, the quarter was amazing in terms of both headline metrics and what’s under the hood. Beyond that, the business is giving some strongly bullish signals for the future.

First is that it’s becoming increasingly a capital-light, fee-based business.

In 2021, its fee-based revenues made up only 26% of all revenues. This didn’t change much in the next two years. In 2023, its share of fee-based revenue was still just 27%.

This was an important reference point for those who called it ‘just a bank.’

Yet, they didn’t realize that SOFI was just building its customer base with lending products, and the fee-based revenue would skyrocket when it ramped up upsells and cross-sells.

We started to see this in 2024 as the fee-based revenue share jumped to 37%.

Last quarter, for the first time, its fee-based revenue share climbed over 40% reaching 44%:

This is one of the most bullish things I see about the business.

We knew that it wasn’t just a bank, but we now see it in the data, too.

This will lead to two things going forward:

Margins will expand.
Margin expansion will lead to a multiple expansion.

This is the recipe for stellar shareholder returns, as analysts will have to revise both their earnings estimates and their exit multiples.

Second, the technology platform is now back to growth.

After stagnating for three quarters, it delivered a 7% QoQ growth.

Though most people tend to ignore it, I think this is a very important business for SOFI. It enabled founders to build new fintech businesses in no time, and some of them grew pretty big, like Dave and MoneyLion.

They have signed three new clients this quarter, and more are expected to come at an increasing pace going forward.

Third, the business is attacking growth on all fronts.

Noto briefly mentioned international expansion plans in the earnings call and said international payments and crypto will launch soon.

Crypto currently generates $1 billion annualized revenue for Robinhood. Suppose SOFI can reach half of that in the next two years. In that case, it may lead to a 5.5% improvement in annual revenue growth, assuming that the rest of the business will grow 25% annually as the management previously expected. And this will be just the beginning.

Overall, SOFI delivered one of its most phenomenal quarters.

It was so good that it closed yesterday up 7% despite its habit of plunging after posting stellar results. This time it was so good that there wasn’t a way of denying.

Yet, this doesn’t automatically make this a ‘buy.’

Even the best business in the world could be an abysmal investment at the wrong price.

So, let’s run the numbers, shall we?

The management previously expected 25% annual growth until 2026 and 20% annual growth beyond that until 2030.

Upon that, the company launched a new loan platform business and announced that crypto will be back on SOFI. International payments will also be launched this year.

Given these developments and obviously accelerating growth, I think it can grow the top line 30% annually for the next 2 years and 25% beyond that until 2030.

This will give us nearly $10 billion in revenue in 2030.

Assuming a conservative 25% net margin, we will get $2.5 billion net income.

Assuming 8% annual dilution, we will get 1.6 billion shares, which gives us an EPS of $1.56.

At 25 times exit multiple, we’ll be looking at a $39 per share stock price.

Discount it back to the current time at an annualized rate of 10%, and we’ll get a $24 per share fair value. This is the level the stock tested yesterday.

This means that the market is now fairly valuing SOFI stock, and those who are entering at these levels shouldn’t expect much more than the market returns going forward.

Assuming a 30% margin of safety, my buy price would be below $18 levels.

As long as the stock doesn’t pull back below that level, I won’t add to my position despite the stellar performance and amazing future outlook.

For the first time in the last two years, the market is fairly valuing it, which means that there is not much juice to squeeze in this one.

I would suggest looking for other opportunities instead of chasing this one.

🏁 Conclusion

This is an ironic post because, two years ago, I was propagating for actions exactly opposite to what I am defending now. I was saying:

Buy SOFI at just 7 times 2026 earnings.
Wait for a pullback in UnitedHealth.

Today I am saying the exact opposite.

Though ironic, it emphasizes once again that the price is the central concept of investing.

You don’t actually make money because you picked a very high-quality business, or you lose because you picked a shitty one. There are people making money by specializing in shitty companies, even in companies going bankrupt. They look at the price, think that the parts will be worth more when the company goes bankrupt and gets liquidated. There are people who have become insanely rich this way.

We like high-quality businesses because they make an extraordinarily hard game a bit easier.

How come? Because they are more predictable.

You can look at the price, forecast earnings a few years down the line, and decide to buy or not. This saves us from a lot of work, uncertainty, and perhaps having to specialize in domains that takes years.

I said it’s easier. Not easy. Still, to make money, you have to come out right on your bet. That is hard, and you only have two paths if you are a long-term investor:

You pick businesses with low bars ahead and bet they can jump over them.
You pick businesses with very high bars ahead and bet they can jump even higher.

I have made almost all my money in the former, and whenever I attempted to try the latter bet, I either lost money or made so little that it wasn’t worth the stress.

SOFI was that low-bar business two years ago. Now the bar ahead is high.
UnitedHealth was that high-bar business two years ago. Now the bar ahead is low.

We didn’t make money because the business we picked was named SOFI, but we made it because we acted with this mindset.

If you act this way, is there a guarantee that the business will jump over the low bar?

Of course not.

Just like humans, businesses also get old and sick; sometimes, they can hardly walk on a flat surface, let alone jump a low bar.

The good thing with investing is that you don’t need to be right on everything.

If you are right 6 out of 10 times, and you don’t lose everything when you are wrong, you can end up pretty wealthy over the long term.

Picking businesses with bars lower than their potential is the only way I know that works dependably over the long term.

You aren’t going to be right every time, but if you exercise this with discipline, you’ll be right enough times.

Thanks for reading Capitalist-Letters!

Please share your thoughts in the comments below.

👋🏽👋🏽See you in the next issue!

I am not saying this to show off, but to make the point that a few people could love SOFI as much as I do. It made me come out right on my contrarian bet, and also made me a lot of money along the way.

I truly love the company, and I think it has an amazingly bright future and decades of compounding ahead.

Yet, this doesn’t mean you should blindly buy the stock.

No matter