You’re not losing your mind. If it feels like things are crazy, that’s because they really are. As crypto seemingly fades into irrelevancy, its knee-jerk moves have insidiously become commonplace to metals, megacaps, and even traditionally boring sectors such as consumer staples and insurance.

Part of the reason, to borrow a phrase from the President, is because “everything is computer”. Gross hedge fund exposure and median short interest are near record highs as investors pile into the AI Trade and discard everything else.

The chart above from Goldman Sachs’s Prime Brokerage Desk depict stunning gross and net concentration in Semis, and the continued pile-on into the AI trade has caused cross-sector correlations to go haywire. To see that exemplified, look no further than the insurance sector. For the first time ever, the insurance sector (KIE) is negatively correlated with the S&P on a 40 day rolling basis.

The S&P Low Volatility Index has similarly become negatively correlated with the S&P.

It’s not like AI has magically become the ultimate enemy of insurers, banks, and consumer staples businesses in the span of two months, rather, these wild moves and cross-sector correlations are the result of positioning: people are pressing the pedal to the metal on the AI trade by paring down everything else, and when they degross, the unwind gets violent quickly. From Wednesday June 3rd through Wednesday June 10th, the Semiconductor Index was down 10% while Banks rallied up 5%. If you’re looking for an explanation, look no further than the fact Institutional positioning in Financials started the month at the 0th percentile (in relative terms since 2010), while Information Technology started the month off at the 99th percentile (source: Morgan Stanley Prime Brokerage Desk).

Given the recent volatility, I think it’s fair to say many investors are aggressively exposed with complicated positions. The question is, what should we do?

I think the answer is clear: keep things simple until everyone calms down.

Since April I’ve been trying to do as little as possible and avoid working myself up into a frenzy. If you interpreted every Iran or Tech headline at face value, you’d wind up in the insane asylum.

I think the same goes for deciding whether to buy the dip. A lot of recent market action seems to be driven by recency bias, people looking at where a stock was merely two or three days ago, and using the differential to decide whether to go “all in” or “all cash”.

I think the memory names exemplify this well. In the depths of April, Turboquant nearly made its way onto the terrorist watchlist - it was panic like never before. If you asked stunned memory bulls (including myself) if they’d be willing to take off some Micron at $550 or $600 if it ever returned, the answer would’ve been an clear yes. “Oh if you let me get out of this hole, I’ll never take risks like that again”.

Now, Micron has more than tripled to $900, yet people aren’t merely maintaining their positions, they’re aggressively piling in because it was $1,080 a few days ago. If you’re scooping up shares because you believe Micron will hit a $30 a quarter run rate by Q3 and memory is no longer cyclical, I tip my hat to you, but otherwise, I think it’s okay to say “You know what, it’s up 200% YTD, a little volatility is to be expected and I don’t need to buy this dip”.

This year the volatility has been immense. The crown has rotated from Gold and Silver connoisseurs to Small Cap Oil Experts to Korean Memory Superfans, over and over and over again. Rather than chase, I think the best thing to do is just purchase reasonably priced equities with decent growth prospects and capital return potential, and let time put in some work.

My Portfolio is relatively untouched since the last update in April, but I have been adding to a few existing positions and collaring a few I view most at risk. Here’s what I’ve been up to.

Read more

Too many cooks spoil the pot.

The actual adage is too many cooks spoil the broth, but either way, it’s one of my favorite sayings. This trade is perhaps a financial representation of the adage.

“And we’re off, SpaceX opens at $180 a share up 30%, a valuation of $2.2 Trillion” a Bloomberg anchor quips as he bangs his pen on the circular table. “Really incredible isn’t it, one of the largest IPOs of the year”.

Social media goes wild. One Twitter user adamantly declares he is switching to the S&P 500 Sharia ETF (which excludes Defense Contractors) to avoid his nest egg being invested in SpaceX. Another Twitter user angrily laments “Look at that opening pop, the investment bankers left Hundreds of Billions on the Table! They have no idea what they are doing!” (he has no idea that the opening jump is, in fact, intentional). A third user quickly changes his handle to “SpaceXExpert”, plows his Robinhood account all-in at the opening print, and promptly starts a paid community to exclusively cover all things relating to SpaceX. To top it all off, Defiance promptly files for a 2x SpaceX ETF at 4:05 PM.

You scoff and shake your head, but your smirk quickly turns emotionless as you realize S&P removed the four-quarter profitability test to ensure 1.25% of your SPY money is blindly plowed into SpaceX six months from now.

“Aw f**k”, you mutter.

Now, I’m not saying this is exactly how the SpaceX IPO will go next Friday, June 12th. But if you use your imagination you can see it happening, can’t you?

Prediction Markets seem to believe it’s likely SpaceX closes above the $2 Trillion Valuation mark at the end of the session, and Hyperliquid’s Pre-IPO Perpetuals have the stock at roughly $188 a share, approximately 40% higher than the 135 a share SpaceX is currently discussing with banks.

Hyperliquid and Polymarket are by no means accurate indicators, but both, combined with the general excitement for tech IPOs we have seen, lead me to believe the bulls are getting ready to rumble and eventually cram SpaceX into every major index ETF.

It all seems so silly, doesn’t it?

Unless lightning strikes and the IPO allocators at Robinhood pity you enough to toss you a share, or perhaps you’re an Institutional client with a salesperson who finds you to be good company, chances are you aren’t getting an allocation and will have to watch the spectacle from the backrow with a “mental” position.

But imagine if you could make money off this by picking up some SpaceX exposure at a discount to the underwriting price?

Alarm bells should be immediately ringing in your heads. How?

The answer is a vehicle the average speculator hasn’t even considered.

Mutual funds.

There are two mutual funds, BARAX (Baron Asset Fund Retail Class), and BGAFX (Baron Global Opportunity Fund) that have significant SpaceX exposure and are still open to investors. Roughly 24.9% of BARAX (as of April 30, 2026) and 16.3% of BGAFX (as of April 30, 2026) are invested in SpaceX, with the SpaceX shares held at a likely 1.25 Trillion valuation.

“The Fund’s largest holding, Space Exploration Technologies Corp. (SpaceX), appreciated 24.5% in the quarter. SpaceX announced the merger with X.AI Holdings Corp. (xAI) in February. As a result of the deal, the implied value of SpaceX increased to $1 trillion while xAI remained constant to its last financing round at a $250 billion valuation. The combined company is now valued at $1.25 trillion.” - Baron Funds Q1 2026 Letter

What this means in simple terms is that if SpaceX opens at a $1.75 Trillion or a $2 Trillion valuation, both funds will see a significant uplift in Net Asset Value as the position is marked at the closing price on IPO day.

Surprisingly, unlike Neuberger Berman’s NBSRX fund (9% of assets in SpaceX) that has closed itself to new investors (to avoid special situations traders swooping in to add SpaceX exposure for themselves before a big payday), Baron’s Funds remain open, which means that as long as the SpaceX mark within BARAX and BGAFX has not shifted higher since March 31st, you can buy into the Mutual Funds and essentially have a large chunk of your investment allocated to SpaceX at an implied $1.25 Trillion valuation just days before the company is expected to IPO at a $1.75 Trillion valuation.

Assuming the AUM of BARAX and BGAFX has remained unchanged since April 30th, BARAX could experience a NAV uplift of 10% if SpaceX closes at a $1.8 Trillion valuation, and 14% if SpaceX closes at a $2.0 Trillion valuation, and BGAFX could experience a NAV uplift of 7% at $1.8 Trillion, and 10% at $2.0 Trillion. Even if the funds have doubled in AUM since, as long as the SpaceX mark has not increased, there is still potential for a helpful NAV uplift next week when SpaceX goes public.

This trade goes right back to the saying “too many cooks spoil the pot”.

Investors in these funds have long awaited the payday of the SpaceX investment, but because the funds are not closed, new investors can swoop in, establish SpaceX exposure at a lower-than-IPO valuation, and dilute the existing investors’ potential profits, thus “spoiling the pot” and siphoning some of the windfall.

Unfortunately, the NAV uplift trade is quickly getting out of the bag.

Bloomberg wrote an article about the expected NAV uplift yesterday, and Retail investors have plowed $1.1 Billion into the Baron First Principles ETF after realizing 2% of it is allocated to SpaceX.

While the Mutual Funds have barely been discussed, I think it is only a matter of time before the opportunity is discovered by the same crowd piling into the ETF.

While an investment in these Mutual Funds may provide the opportunity to guarantee yourself some SpaceX exposure, there are also several risks.

First of all, it is impossible to determine what percentage of the funds are invested in SpaceX. As with huge inflows into the Baron First Principles ETF, if BARAX ($3.3 Billion AUM as of March 31st) or BGAFX ($839 Million AUM as of March 31st) experience inflows (perhaps they doubled in size) the weight of SpaceX could become significantly diluted, reducing potential NAV uplifts.

Secondly, it is possible that Baron Funds has increased its mark for SpaceX in NAV calculations (implying you are no longer getting exposure to SpaceX at a $1.25 Trillion valuation). Looking at recent performance, there was a bump in BGAFX that I believe was caused by its exposure to Semiconductors, but there has been no recent bump in BARAX, leading me to believe BARAX has not increased the mark of its SpaceX position, which is why I find it the more attractive of the two.

Additionally, BARAX has substantial equity investments in the broader market, meaning that an investment in the fund leaves you exposed to general market risk on top of SpaceX, which is not guaranteed to go public, and may not be well received by the market.

Importantly, unlike ETFs, there are no limit orders for Mutual Funds. If you create a subscription to purchase $100,000, you don’t know the NAV you are purchasing at until it is calculated after the market close and your order has already been filled. In practice, this isn’t generally an issue, but it does leave you exposed if Baron Funds decides to suddenly mark up the value of its SpaceX stake, or chooses to not immediately update the NAV when SpaceX goes public next week.

Finally, brokers such as Schwab and Fidelity may impose a $49 fee for selling a Mutual Fund if held for less than 90 days, adding transaction costs. It’s also possible that BARAX or BGAFX modifies its redemption fee to disincentivize flippers. Currently, there is no load or redemption fee on both BARAX and BGAFX, but that can always change.

Despite these risks, I find the setup for BARAX attractive given the SpaceX exposure and potential NAV uplift, and have allocated 1% of my Portfolio to it. Even if the fund has experienced significant inflows (diluting the SpaceX stake as a percentage of NAV), an investment still represents an opportunity to lock-in pre-IPO exposure to SpaceX at what might still be a sub-IPO valuation (assuming the mark has not moved).

While you could wait until next week to establish a position closer to the actual IPO, I think there is a possibility BARAX closes itself to new investors to preserve a potential SpaceX windfall, hence why I initiated my position this week.

This trade isn’t risk free, but if the SpaceX mark within BARAX is still $1.25 Trillion and the IPO is successful, I think the trade has potential and I will attempt to cash out my shares shortly after the IPO. Given how many moving pieces there are to this trade and the general market exposure present in BARAX, I think it makes sense to size conservatively.

Keep this one on the down low, too many cooks spoil the pot.

Read more

Introduction

Stubs are some of my favorite asymmetric market setups. When a company disposes of major assets, many traders often “trim out” to lock in a quick profit, leaving behind a remaining business that nobody has spent time on valuing. In this piece, I’m covering two setups I believe the market has overlooked: the Aurora Networks stub within Vistance Networks, the new bank being formed by Green Dot and CommerceOne following Green Dot’s sale of its fintech assets, as well as a value bank, OSB Group, that doesn’t quite fit the theme but is simply too cheap to ignore at 6.3x forward earnings and 0.85x Tangible Book Value.

Vistance Networks (NASDAQ: VISN)

I’ve owned Vistance Networks (formerly known as Commscope) since 2024, and am extremely impressed with how the company has conducted itself. Over the past two years, Vistance has flipped the script and transformed itself from a debt-laden melting ice cube into an asset-disposition name focused on maximizing shareholder value.

In August 2025, Vistance announced a landmark sale, the $10.5 Billion disposition of its Connectivity and Cable Solutions segments to Amphenol. Vistance used the proceeds to entirely eliminate debt, redeemed all preferred equity, and recently used the remaining proceeds to fund a $10 a share special dividend. In April, Vistance announced another significant transaction, the sale of its Ruckus Networks Division to Belden Inc for $1.846 Billion.

After these two milestone sales, many assume Vistance is a shadow of its former self with insignificant remaining assets, but I see substantial value in the stub: it’s Aurora Networks division, who is poised to gain from the surprising comeback of coaxial cable.

Aurora Networks specializes in infrastructure broadband operators use to upgrade existing coaxial networks: headend and access platforms, DOCSIS 4.0 nodes and amplifiers, and virtualized access products. In simpler terms, Aurora sells the “upgrade kit” cable companies use to extract higher speeds out of their aging coaxial networks.

Recently, cable internet providers (such as Comcast) have come under immense pressure from expensive fiber buildouts by AT&T and Verizon (that offer significantly faster symmetrical upload and download speeds), and blazing fast over-the-air 5G Home Internet (offered primarily by T-Mobile).

The bottom line is cable internet providers have realized $99 a month for 50 mbps internet via their coaxial networks is no longer an attractive offering, and that they need to increase speeds and performance to keep clients from leaving. The Cablemagedon has transformed into an immense tailwind for Vistance’s Aurora Networks segment: cable companies are unwilling to commit to the massive capex of a fiber buildout (now that 5G Home Internet is rapidly taking market share) and instead want to get the most out of their existing coaxial networks via upgrades to the new DOCSIS 4.0 standard (delivery speeds up to 10 Gbps down and 6 Gbps up via coaxial cable).

Aurora Networks is one of the only fully integrated providers of DOCSIS 4.0 products, and a surge in DOCSIS upgrade interest from providers such as Comcast and Vodafone has resuscitated this seemingly dead segment back to life.

Aurora Networks Q1 revenue surged 32.6% to $298 Million and Q1 EBITDA jumped 31.7% to $50.3 Million, driven by record demand for amplifiers and nodes as well as new multi-year contract wins. The results lend credibility to the management team’s bullishness in the 2025 Q4 call, where they indicated large contract wins would drive 2026 results higher:

“Yes, I don’t think we’re going to give the precise number, but it’s a meaningful dollar amount. It’s tens of millions of dollars of opportunity that comes with that win.” - Kyle Lorenzten, CFO, in reference to winning new contracts

The point is, Vistance’s Aurora Networks segment isn’t a melting ice-cube. The backdrop is extremely favorable for 2026, 2027, and 2028, as these multi-year cable upgrades continue, yet the market is seemingly ascribing Aurora Networks very little value at Vistance’s current market price.

Vistance is currently sitting on $250 Million of cash (after paying out the $10 special dividend), and is primed to receive $1.7 Billion in net proceeds from the sale of Ruckus later this year. With pro-forma cash of $1.85 Billion, the implied valuation of the Aurora Networks segment is just 4.1x EBITDA, extremely low.

The company has indicated it intends to pay out “a significant portion of the excess cash” from the Ruckus sale and analysts have already modeled a $5 Special Dividend, so I don’t think Vistance is likely to become a value trap where a company has a lot of cash it refuses to return - instead, we are likely to see a sizeable special dividend and for analysts to increasingly focus on Aurora as the remaining entity.

Bears have expressed concern about a lawsuit filed by lenders, who allege Vistance owes them at least $150 Million as a penalty for paying down debt early, but even if Vistance is found wholly liable for $150 Million (roughly $0.63 a share) it isn’t material to the story and the Aurora Networks stub would still be trading at an extremely inexpensive 4.7x EBITDA.

(The sensitivity table above errs on the side of caution and assumes Vistance will lose the $150 Million lender lawsuit)

Using management’s 2026 Aurora Networks EBITDA midpoint of $237.5 Million and an EV / EBITDA ratio of 6.0x, the implied share price of Vistance Networks is $13.58, representing upside of roughly 10% from the current price. Peers such as Harmonic Broadband (20x EV / EBITDA), Vecima (6.5x EV / EBITDA) and Teleste (8x EV / EBITDA) all trade at a significant premium to Aurora Network’s implied 4.7x EV / EBITDA multiple at $12.30 a share, which is surprising given Vistance’s multi-year contract wins from Vodafone and Comcast which serve as a testament to their superior offering.

I think you can make a strong case that the DOCSIS 4.0 upgrade cycle and increasingly attractive economics for upgrading coaxial in lieu of building fresh fiber will serve as a significant tailwind to Aurora Networks over the next several years. The Vistance management team has exhibited credibility, generated significant value for shareholders via asset dispositions, and recently announced a $100 Million share repurchase, which I believe is an excellent tool to utilize given the stub is trading at a steep discount to peers while the company is using no leverage (giving it significant flexibility to financially engineer the stock once the transaction with Belden is complete).

Vistance Networks isn’t risk free, the sale of Ruckus to Belden needs to pass regulatory scrutiny, and rising DDR4 memory prices are impacting Aurora Network’s margins. Still, the market is seemingly pricing Aurora Networks below < 5x EV / EBITDA, which I believe represents a rare opportunity to buy a “growth” name with a multi-year tailwind behind it at a deep-value multiple.

I’ve owned Vistance for years, and while I never like to chase, given the extremely attractive multiple the “stub” is trading for, Vistance has quickly become a high-conviction trade for me and I have been adding aggressively < $12.50 a share. Because the sale of Ruckus to Belden was announced just a few weeks ago, I think many traders simply haven’t realized how inexpensive the “stub” is, which would explain the sudden move down to $11 as traders took profits after the announcement, before the stock immediately recovered. I’m encouraged by significant increases in ownership from funds such as Carronade Capital (7.64 Million shares, +6.099 Million in Q1) and FourSixThree Capital (3.04 Million Shares, +1.055 Million in Q1), and I believe the market is slowly beginning to appreciate that there is more to the Vistance “stub” than appears at first glance. I am hopeful that as the market begins to treat Aurora Networks as a stand alone entity, the stock re-rates to $13.50 to $14.50 a share (which would imply 6.0x - 7.0x EV / EBITDA for the stub), and in the meantime, I think it’s worth considering an entry below $12, which implies just 4.5x EV / EBITDA at the midpoint estimate.

OSB Group (LSE:OSB)

Another name I have been adding to is OSB Group, an English bank specializing in “buy-to-let” (loans to landlords purchasing properties specifically to rent them out). I purchased OSB in March 2025 and it has certainly been a bumpy ride - the stock hit an all time high of 640p earlier this year, and has since tumbled to 500p as concerns over rising UK Interest Rates, a government crackdown on landlords, and general political instability mount.

These concerns have heavily impacted OSB’s shares, which are now trading at the steepest discount to Tangible Book Value since “Liberation Day”. What I like about OSB is that it is a “prove me” story: sentiment is routinely negative despite solid fundamental results, which continually provides the company an attractive window to return capital via accretive buybacks. The stock is currently trading at 502p (0.85x Tangible Book Value), 6.5x forward earnings, a 7% dividend yield, and importantly, OSB is repurchasing shares at the fastest pace on record.

Regulatory filings confirm OSB is rapidly deploying the 100 Million GBP buyback. The company has recently stepped up the pace to over 300,000 shares a day, nearly 0.08% of the float, and more than 30% of the daily volume. OSB is certainly making a statement with these aggressive repurchases.

OSB is well capitalized, their Q1 update indicates credit quality is stable, and I think OSB’s concentration in buy-to-let is in fact an underappreciated asset, as evidenced by larger UK Banks such as Natwest announcing new partnerships to muscle into buy-to-let. OSB has consistently generated a mid-teens ROE despite a high cost deposit base, which I believe makes it an attractive acquisition target for a larger peer with an inexpensive deposit base that who would want bolt on a specialist mortgage division without having to build an underwriting team and brokerage relationships from the ground up. A transaction makes even more sense when considering larger institutions all trade at a premium to Tangible Book Value, making it easier for them to use their shares as “currency” to take over OSB.

Until then, I believe OSB is attractive on a stand alone basis, and that recent drop provides a rare opportunity to enter a name that routinely exhibits shareholder friendly capital allocation and industry leading fundamentals, at a sizeable discount to its own Tangible Book Value.

The bear case is relatively straightforward: investors are concerned the retirement of longtime CEO Andy Golding signals trouble ahead, that UK yields continue to rise, that OSB is lending against overvalued collateral, and of course, that a nationwide rent freeze could slow demand for new mortgages and deteriorate the credit quality of existing ones. These concerns are sensible, but I think they are already priced into the stock and more importantly represent tail-risk scenarios rather than a plausible bear case.

What I mean by that is a nationwide rent freeze isn’t likely to happen, the UK 10 Year has already dropped 30 bps in the last couple of days as concerns over energy-incited inflation recede, the UK housing shortage is likely to persist (driving rents to fresh highs, benefiting landlords), and OSB’s Q1 update showed no signs of the credit issues investors fear. If anything, the setup has become unusually tilted to the upside: OSB is making steady progress towards its 2027 goals as it finishes technology upgrades, it is repurchasing huge quantities of shares below Tangible Book, and it has built enough excess capital to absorb a reasonable amount of housing weakness. At 0.85x Tangible Book, we don’t need the UK Housing Market to turn red hot for the setup to work, we simply need credit quality to remain stable and for the share count to continue expeditiously declining.

OSB is about 3% of my portfolio, and I continue to add steadily near 500p. Unfortunately, the company won’t be reporting earnings until early August, so it’ll be a while until the negative narrative can be firmly disposed of, but until then, OSB has 57.35 Million GBP remaining on its share repurchase program, which leaves it capacity to repurchase 200,000 shares a day until earnings (nearly 20% of the daily volume). As with all banks and insurers, we can’t ignore the left-tail risk of something going horribly wrong, which is why I think it makes sense to take a basket approach to these value setups and own several smaller positions rather than concentrating in one name. Nonetheless, I think OSB is trading at highly distressed levels despite solid fundamentals, a strong capital position, and ample capacity to return capital to shareholders. If you have an interest in adding value banks to your portfolio, I think OSB is certainly worth considering as a long-term setup. If inflation fears recede and yields decline, I believe OSB could quickly re-rate to 0.90x - 0.95x Tangible Book Value, which would imply 530p - 560p a share.

Green Dot Corporation (NASDAQ: GDOT)

I’ve initiated a small tracker position in Green Dot.

The Green Dot story is remarkably simple: the company is selling its non-bank fintech and payments assets to Smith Ventures, and merging its remaining bank assets with CommerceOne Bank (an affiliate of Smith Ventures, Bill Smith is on the board of CommerceOne). Upon consummation of the transaction, Green Dot shareholders are expected to receive $8.11 a share in cash, and 0.2215 shares of the new publicly traded bank. The question we need to figure out is what is the 0.2215 shares of the new bank worth?

The new banking entity has a seven-year MSA (Master Services Agreement) with the Green Dot Fintech platform, meaning nearly $4 Billion in relatively inexpensive, sticky, fintech deposits will be parked at the new bank until at least 2033. These deposits aren’t necessarily zero cost, the new bank will need to share about 40% of interest income with banking-as-a-service partners, but the positive of this MSA is that it ensures the new bank will hit the ground running with nearly $5.2 Billion in total deposits and just $750 Million in loans, providing ample room to grow.

I’ll keep my analysis here brief: I don’t anticipate the new bank will trade below Tangible Book Value. The new bank will be extremely well capitalized, have minimal credit risk (despite recent charge offs at CommerceOne, the new bank will have an extremely low loan-to-deposit ratio which minimizes the impact of any credit deterioration in the existing book), and will have ample capacity to repurchase shares to engineer a 1.0x Price to Tangible Book Value ratio. With implied Tangible Book Value of around $27.80 a share for the bank, GDOT shareholders’ 0.2215 shares should be worth about $6.15 each.

Green Dot currently trades at $12.78, which implies the market is valuing the $6.15 value of the new banking stake at just 0.75x Tangible Book Value (this excludes the probability of the deal falling through, which is also likely priced in, but impossible to identify definitively). The below table shows the implied P/TBV of the new banking shares given the GDOT share price. As you can see, the risk-to-reward gets considerably more favorable in the high 11s and low 12s, which implies the new bank is being valued at less than 0.67x Tangible Book, extremely unusual for a bank with a clean slate and significant excess capital.

It’s important to note the transaction is awaiting regulatory approval, and the market will likely be apprehensive when valuing the new bank. Still, given the fact banking mergers are being approved at the fastest pace since 1990, I think the deal is unlikely to fall through, and furthermore, that the bank is likely to be valued at or above Tangible Book value given the company will have enough excess capital to simply repurchase shares until it is.

Even if the deal somehow falls through, I think Green Dot has a good chance of finding another suitor and is otherwise worth $9 to $10 standalone, which caps downside to about 25% from currently prices. Green Dot’s ownership list boasts value investors such as No Street GP (7.5% stake), Western Standard (6.26% stake), Steel Partners (2.2% stake - although it appears they are reducing their position), and M3F (1.3% stake), which makes me believe the combined bank is likely to hit the ground running with an immediate and attentive Institutional shareholder base.

The risk-to-reward for this setup varies significantly with slight movements in price, given what we are really trading here is the value of the banking stub. I’ve initiated a small starter position in Green Dot at $12.78, and have left a lot of dry powder to add in the low 12s and high 11s where the math becomes considerably more favorable and implies the bank is trading at less than 0.63x Tangible Book Value (under $12). I think it makes sense to keep Green Dot on your watchlists - any pullback not tied to a deterioration in CommerceOne credit quality or regulatory approval could offer a highly attractive entry point, especially as we near the listing of the new bank.

Conclusion

Of the three setups, I find Vistance the most attractive, followed by OSB. OSB’s steady capital returns and steep discount are notable, but ultimately it is likely constrained to Tangible Book Value (or a few percentage points higher), while Vistance’s Aurora Networks stub could significantly re-rate if investor excitement over the refurbishment of coaxial cable networks builds. As I write this, significant investor attention is being given to Harmonic and its Broadband Solutions segment, which I believe could benefit Vistance as investors realize there is more to the “stub” than initially meets the eye.

My blog isn’t intended to be a “clone a portfolio”. Instead, I want to cover and present niche setups, everything from Georgian Investment Firms to Samsung Preferred and SK Hynix, and ultimately, for you, the reader, to determine whether the setup is attractive and if it is appropriate for your Portfolio. My next Portfolio Review will come out mid-June, where I cover the 50 positions I hold and the setups I find most attractive to allocate new capital to. I hope you’ve enjoyed this piece, there is nothing below the paywall, I just can’t figure out how to send an email without one.

Read more

It’s been a busy earnings season and I wanted to share a brief update on two existing value positions that delivered excellent earnings, as well as a new position in Korea.

1. For Startups (7089 JT Equity)

I featured For Startups as a new position in the April 2026 Portfolio Review as Idea 3, "Capital Light in Japan" (Entry Price: 1,224 Yen). For Startup’s Q1 Earnings report (released on Friday), was excellent, and I think its 30% top-line growth and significant guidance upgrade are likely to finally get it some attention. Before I get into the results, I want to briefly summarize the story.

For Startups is perhaps the most Japanese company you can imagine, a Venture Fund, a Database of startups, and a Headhunter all in one. What it truly represents, however, is a pure play way to ride the wave of exploding Japanese VC Activity in a capital-light vehicle.

For Startups has significant ties to leading Japanese VCs such as M3, Incubate Fund KK, Globis Capital Partners, Wil LLC, Strike Co, and Sparx Group (who are also shareholders in the company). For Startups specializes in placing C-level executives and technology employees across startups, and leverages its close ties to existing Japanese VCs and its proprietary startup database (Startup DB Japan) to identify companies in need of its services. The company also has a small co-invest portfolio that it invests alongside its VC clients, while the portfolio is not meaningful at 600 Million Yen (less than 10% of the market cap), the combination of its Startup DB, the co-invest portfolio, and its close ties to leading VCs has created a massive referral base: when VCs have portfolio companies that need to hire, they just call For Startups.

The company has been riding the wave of increasing startup activity in Japan, and earnings have steadily grown from 22 Yen in 2021 to 99 Yen in 2025, a CAGR of 45%. The CEO, Yuichiro Shimizu, has routinely highlighted he believes the shares are undervalued, and in addition to pulling 2027 guidance forward one year, the company repurchased 160,000 shares (roughly 4% of the float) in March.

For Startup’s incredible earnings report (released last Friday) has caught my eye. Revenue for the first quarter of 2026 was up a stunning 63% year over year, and earnings per share jumped 80% (to 55 Yen per share), demonstrating the company’s significant operating leverage and capital-light nature. Unlike many Japanese companies who trade at low multiples due to heavy investment needs to remain competitive and a general inability to return capital, For Startups is the exact opposite: its headhunting business is inherently capital-light, and the company continues to opportunistically pursue share repurchases while still generating 50% top-line growth.

The company’s investor presentation (which is astutely posted in English to attract US attention) has guided significant growth for 2026: revenue growth of 25%, and 145 Yen in Earnings per Share (38% growth).

(Source: Company Filings)

I think For Startups has all the markings of a “capital light compounder” aside from a migraine inducing multiple and the usual circle of GARP lovers who only get interested in 6% growth once it’s already trading at 30x earnings. I think the recent first quarter results and full year guidance upgrade to 145 Yen in EPS confirms the thesis I had last month: For Startups has positioned itself as an indispensable headhunter for the Japanese Startup industry, and will continue to ride the wave of elevated business investment. Quarterly headhunting placements are soaring, the company is increasingly focused on high ticket (salary) roles, and its Startup DB division is generating record annual recurring revenue.

For Startups excels in meeting my value investment criteria. The company specializes in catering to an under-served niche, has a clean balance sheet with a strong net cash position (approximately 330 Yen per share), has extremely aligned insiders who own 20% of the company, and is capital-light in nature, making it easy for the company to return cash to the shareholders.

With guidance for 145 Yen in earnings in 2026, I think the company is too cheap to ignore at 1,359 Yen, and that the combination of 30% guided earnings growth, a capital-light business, and the management team’s consistent history of execution, may warrant a 11x - 14x multiple, which would imply upside to 1,595 - 2,030 Yen a share.

The next two names are below a paywall for a very simple reason: both trade at deep discounts to Tangible Book Value, and are actively conducting highly accretive share repurchases. It’s a bit paradoxical, but the cheaper these stocks remain, the more value is created for long-term shareholders when management is able to repurchase stock at 5x earnings. If you have zero interest in Asian value investing, don’t subscribe. I can assure you both of these pitches are exceedingly boring, and you may even fall asleep at your desk if you read further.

Read more

Who would’ve thought?

After years of anticipation, IBKR finally did it. As of last night, Interactive Brokers finally enabled trading for the last bastion of value stocks, Korean Equities.

Up until 2024, Korean markets were a graveyard for Western value investors. Corporate giants (“Chaebols”) would flood the market with listed subsidiaries who were carved up, spun off, and folded back in on terms that conveniently benefited the founding family at the expense of foreign investors.

Luckily, that has all begun to change as the wave of Asian financialization continues. Japan spearheaded the way in 2022-2024 via numerous market reforms, in particular the Tokyo Stock Exchange’s name-and-shame approach which requires companies at a low P/B ratio to disclose concrete steps on how to narrow the discount and improve ROE. Japan’s efforts were wildly successful, and Korea was eager to replicate the same for itself. Lee Jae-Myung, the governor of Gyeonggi, ran for the Presidency in 2025 on a pledge to eliminate the Korea discount and to double the Kospi Index (since his inauguration on June 5th, 2025, the Kospi is up 136%).

While much of Korea’s red-hot market is ascribable to the performance of SK Hynix and Samsung Electronics, I think there are also significant regulatory tailwinds behind Korean equities, such as the recent reduction in dividend tax from 45% to 15-30%, a 20% excess surtax on Chaebol retained earnings that aren’t distributed to shareholders or reinvested in the business, and fortifications in the Commercial Act (July 2025) that enable minority shareholders to fight unfair M&A. The bottom line is, Korea is rapidly transforming into a shareholder-friendly jurisdiction.

What I like about Korean Equities in 2026, similar to Japanese Equities in 2025, is that there are many straightforward value setups. This isn’t a matter of finding the next Nvidia, or stumbling upon the next “AI Bottleneck”. I’m not smart enough, anyways. I’m just looking to follow my framework and find value that already exists, and is simply being unlocked and returned to the shareholders via Dividends and Repurchases.

I think in Korea, where nearly half of listed equities trade at a sharp discount to Tangible Book Value (at least up until recently), this approach is likely to yield dozens of attractive setups. They won’t be home runs, but I think even a small allocation to a diversified basket of Korean value names has the potential to yield attractive returns, particularly as hungry value investors slowly pour into the country. I’ve had a couple names on my buying list for months, here are three value setups that have caught my eye.

1. SK Square - 402340

Price: 805,000 KRW

Sector: Technology

Price to Tangible Book: 0.56x

Implied P/E: 2.6x (Look-Thru Basis to SK Hynix)

Shareholder Return Yield: 0.5% (Solely Repurchases)

SK Square is an investment holding company that was spun off from SK Telecom in November 2021 to take direct ownership of SK Group's non telecom assets. Its consolidated operating businesses (ecommerce, music, and advertisements) generated 1.41 Trillion Won of revenue in 2025, but are not meaningful. SK Square’s real gem is its 20.07% stake in SK Hynix.

SK Hynix shares comprise 95% of SK Square’s assets, making SK Square a pure-play way to get exposure to SK Hynix at just 0.55x the mark-to-market value of the 20% SK Hynix Stake.

The management team of SK Square is acutely aware of the discount to Tangible Book SK Square is trading at, and has slowly reduced non-core holdings and used the proceeds to steadily reduce the share count from 141 Million in 2022 to 132 Million currently.

These shareholder-friendly steps have helped reduce the discount, and SK Square recently achieved its November 2024 goal of trading at 0.5x of NAV. The company has attracted several activist investors such as Dan Loeb, and announced an updated goal on March 25, 2026 of trading above 0.7x NAV by 2028. I think it’s clear the company is indicating “More repurchases are on the way”.

At just 0.55x Tangible Book Value with shareholder-friendly capital allocation, I think SK Square provides a unique opportunity to gain exposure to SK Hynix at an extremely attractive valuation (2.6x Hynix’s Forward Earnings on a “look through” basis). You could argue SK Hynix is already somewhat interesting around 4.8x forward earnings, but getting exposure at a 45% discount via SK Square seems to be an even more attractive opportunity, provided you are bullish on the memory trade.

Of course, there is tracking risk, there can be no guarantee the price of SK Square will track the value of its SK Hynix stake, but as shown in the graph above, the discount does seem to be closing and I think IBKR enabling Korean Equities will add additional momentum as value investors discover the opportunity. SK Square may constitute a PFIC (Passive Foreign Investment Company) under US Tax Law, which could pose adverse tax consequences to US Investors. As always, you should consult a qualified financial advisor and tax professional before making any investment decisions.

2. IM Financial Group - 139130

Price: 19,110 Won

Sector: Banking

Price / Tangible Book Value: 0.51x

Forward P/E: 6.20x

Shareholder Return Yield: 6% (Dividends: 3.7%, Share Repurchases: 2.3%)

iM Financial Group, formerly DGB Financial Group, is the parent of iM Bank, a small Daegu-based bank that has recently become the first Korean regional bank promoted to nationwide commercial status in 30 years.

iM is a turnaround story. Over the last 3 years iM Bank has steadily grown its deposit base (expanding into Chungcheong, Gangwon, Honam, Jeju in 2024), significantly reduced bad loans, returned its Securities Division to its first year of profitability since 2022, as it embarked on its National expansion. iM’s impressive turnaround has attracted significant Institutional attention, several subsidiaries experienced credit rating upgrades in 2025, and iM Financial has become one of most “foreign-owned” banks in Korea. Currently, foreign investors hold 45% of shares outstanding, which is notably high for a Korean Regional Bank.

I believe there is a significant tailwind behind mid-sized Korean banks. High interest rates, stable credit quality, and rapidly improving capital allocation policies has driven a stampede into industry titans such as Woori, Shinhan, and KB, and I believe foreign investors are beginning to focus on smaller competitors that still have a national presence, such as iM.

Now, iM is shifting its focus to shareholder returns. iM’s foreign-friendly investor disclosures, improving fundamentals, and capital allocation differentiate it from many peers who trade at similarly low multiples but do not return capital. iM’s “Value-Up” plan has set clear, attainable ROE and Capitalization Goals, and the company is making steady progress on its 150 Billion Won Buyback goal by 2027 (currently 50 Billion Won, 1.5% of shares outstanding, is remaining).

I think iM’s significant foreign ownership, a 10% stake held by competitor OK Financial (who purchased shares on the open market at just 6,000 Won in 2021, providing a strong vote of confidence amidst the turnaround), and continual insider buying from the management team is indicative that iM Financial is committed to transforming itself from a “value trap” into a well-run bank that earns its cost of capital.

At just half Tangible Book Value, 6x forward earnings, and with a 6% shareholder return yield comprised of dividends and buybacks, I think iM Financial represents one of the only remaining opportunities to ride the Korean Banking wave at a 6x multiple paired with strong capital allocation and significant foreign interest.

Read more

What is it worth to you?

That’s a question I think of often when it comes to AI. In both markets and in real life, everything is relative. You buy a stock when you think it’s worth more than what it’s trading for, you buy a preferred when the risk adjusted yield in excess of what you might earn in cash, you buy a product when the value it provides to you exceeds the price; the larger the differential, the happier you are to do it, and the more you want of it.

Over the past couple of weeks, we’ve seen countless complaints on Twitter and tech news from angry customers asserting Claude has “dumbed down” its models or ChatGPT was throttling usage. While many were quick to jump on the companies and assert they were doing this because the business model was unprofitable, I see it differently: the complaining highlights how large the value differential is.

A junior software developer (at least in the US) might cost $75,000 - $150,000, and I think most would agree Codex or Claude Code is akin to having a developer on call 24/7. AI serves as a personal butler when it comes to researching, finding restaurants, maybe even locating cheap laptop deals. It’s not perfect, but things are moving quickly. With that in mind, the price of a $100 Claude Max or ChatGPT Pro plan is actually a bargain to power users. When a $100 plan gets throttled, users riot because they are losing way more than $100 in value.

Even if the labs hiked pricing to $500 or $1,000 a month, it would still be a bargain to the power users driving AI demand. The price of a token to a company using AI to handle things like phone support or data entry is still significantly underpriced.

This isn’t a mere anecdotal observation, the proof is in insatiable demand. Anthropic’s ARR has surged to $30 Billion from $9 Billion in just 3 and a half months. Nebius (a leading neocloud) is sold out of capacity. Rental Rates for older H100 Chips are up significantly, and in fact have risen since 2024. It doesn’t seem like things are slowing down at all.

My point is, the value AI brings to the table far exceeds the cost, and with that in mind it’s easy to see why the Hyperscalers and Frontier AI Labs alike are clamoring for more chips: the value of the chips, to them, significantly exceeds the price, and the value of the “tokens” they provide to their clients also significantly exceeds the cost. As long as that is the case, I think the AI trade has legs, and that AI adoption will continue accelerating.

The AI Megabasket I posted earlier this year has been a major driver of returns in my portfolio. I’ve made a few adjustments, and have cut several of the more speculative names. My Memory and Storage portfolio is just SK Hynix, Samsung, Western Digital, Micron, and Kioxia (in that order), and my Infrastructure basket is comprised of IES Holdings, Comfort Systems, and Powell Industries.

Has my portfolio been fully “AI Pilled”? A bit, but I also see a lot of value in inexpensive banking, insurance, and discount-to-nav situations with aligned management teams and aggressive capital allocation plans to address the discount.

The “Blackjack 21” index continues to set All Time Highs and is up 16.4% YTD on an equal-weight basis. While it might not seem like much next to the AI names, it is significantly ahead of the MSCI Global Financials Index (-0.83%), the S&P 500 Financials Index (-5.20%). I think components in this basket still have significant tailwinds: Japanese banks are poised to gain from rising yields, and the aggressive capital allocation of Halyk Savings Bank, NT Butterfield, Georgia Capital, and International General Insurance continue to drive returns and improve ROE.

Before I get into the list of all 45 Portfolio holdings, I wanted to discuss the significant opportunity Global Value presents. The MSCI EAFE and MSCI Emerging Indexes are finally putting on a show after more than a decade of significant underperformance relative to US Equities.

While some of the reversion is due to Korean Equities (which is almost purely an AI trade), I think there is a long-term tailwind from improving global capital allocation and the “Americanization” of foreign equities. Japan, Korea, China, and the EU are putting increasing amounts of focus on corporate capital allocation. Japanese companies who used to rewards shareholders with “Yutai” (gifts such as discount cards and food) have pivoted to share repurchases, and management teams have aspirational Price to Book Ratios, European companies (who were previously allergic to anything other than a mediocre dividend) are wading their way into buybacks. The bottom line is these companies are finally being run for the benefit of the shareholders.

While the broader indexes have surged as a result, however, many small cap names with improving fundamentals and shareholder friendly capital allocation have been entirely left behind due to their lower liquidity and limited Institutional interest.

I think these setups are exactly where retail investors have an edge: finding companies with strong capital allocation, aligned management teams, and inflecting fundamentals that are simply too small for larger investors to touch. If you think about it, the liquidity discount does make sense. If you are a large Institutional Investor looking at two identical companies at the same valuation, but one has $1 Million in ADV while the other has $10 Million in ADV, you’ll of course opt for the more liquid investment. Over time, however, I think you can generate significantly higher returns by exploiting this phenomenon and positioning yourself in names at a valuation discount that have aggressive capital allocation plans in place to close the gap.

I haven’t pitched Japanese Equities since last summer. All four Japanese picks have significantly outperformed the broader index as well as the S&P, and astoundingly, three out of four have further 30-60% upside to Tangible Book Value.

After spending a lot of time screening, I’ve found two new Japanese companies with a similar setup, as well as three other value names that I believe are wrongly discounted. Before we get into the existing portfolio, here are the new positions.

Trade 1: “Buy a Pile of Cash, Get a Business, Free”

Long MOMO US Equity

Sector: Technology

Conviction: Medium

Current Price: $6.29

End of Year Target: $7.00

Forward P/E: 6.5x

Price / Tangible Book Value: 0.66x

Price / Adjusted Tangible Book Value: 0.90x

Shareholder Return Yield: 16.4% (Dividend: 4.7%, Share Repurchases: 11.7%)

Hello Group (MOMO US) is a Chinese social media company operating two flagship apps: Momo, a social networking and live streaming platform, and Tantan, a dating app commonly considered to be the Chinese version of Tinder. Momo’s domestic Chinese business has been in a steady decline, paying users have plummeted from a peak of 9 Million to just 3.5 Million driven by competition from Douyin (Chinese TikTok) and a years long crackdown on underreporting of tipping and gift income earned by livestreamers.

Momo stock is down nearly 90% from its 2017 Peak of $50 and seemingly prices in imminent implosion, but I think there is more to the story here. Momo’s Q4 2025 report just a few weeks ago showed the situation is inflecting: Chinese paying users have stabilized after an all time low of 3.5 Million in Q2 and climbed sequentially to 3.7 Million in Q3 and 3.9 Million in Q4, the first two consecutive quarters of growth since 2021. Additionally “overseas revenue” from Momo’s global portfolio (Happn – European Dating App, Soulchill – MENA Live Streaming Platform) has been accelerating. The International Portfolio of apps, which just a few years ago comprised just 1.6% of Revenue ($230 M RMB), now comprises 19.3% of Revenue ($2 B RMB) and have been guided as within striking distance of $3 B RMB in 2026:

“Something like RMB 3 billion in overseas revenue for 2026 is a pretty achievable target.” – Peng Hui – CFO, Q4 2025 Earnings Call

MOMO has always been viewed with wariness by American investors due to its VIE (Variable Interest Entity) structure, common among Chinese ADRs. MOMO itself does not own its Chinese operations, it instead has contractual economic rights to receive dividends from its Chinese business. Due to the VIE structure, significant cash (1.5 Billion RMB) is required to remain on the Mainland as “a statutory reserve”, and any dividends remitted to the overseas entity is assessed an additional 10% withholding tax on top of Corporate Income Taxes. While this setup isn’t optimal, Momo has routinely demonstrated excellent capital management and a willingness to move cash out of the Mainland to reward shareholders. Over the past 5 years, Momo has moved 11.4 Billion RMB (approximately $1.58 Billion USD) for use in share repurchases and dividends.

Much of the 2024 distribution was used to pay back bank loans at the parent level as well as a Tax Settlement for prior year taxes. Nevertheless, MOMO’s history of paying dividends to its parent entity is demonstrative that the cash profits generated in China are not trapped.

Additionally, since 2021 MOMO has been using cash paid offshore to reward the shareholders. In fact, MOMO has paid out more than 100% of adjusted net income in the form of repurchases and dividends 3 out of the last 5 years.

I think the capital return intensity is what differentiates MOMO from typical value traps. Anyone with a screener can find names at < 7x earnings, but how many of the names that show up return 100% of income, or even 50%, to the shareholders? Very few.

MOMO’s core business is capital light, and the balance sheet is comprised entirely of cold, hard cash. Essentially, MOMO shares are a pile of cash with a gushing app business on the side.

MOMO isn’t risk free, a sudden Chinese crackdown on VIE’s could ravage the company (as well as other VIE’s such as BABA, NIO, etc), although it is important to note MOMO’s Cayman Shell isn’t totally empty, the European and Japanese app portfolio is not subject to Chinese regulation. There is also the prospect of a continued Chinese crackdown on livestreaming and dating apps, which would significantly harm Momo and Tantan.

While this obvious downside exists, I also think there is significant upwards optionality. MOMO’s portfolio of International apps is rapidly growing, and much of the portfolio is not yet monetized which will provide a further tailwind once it is:

“We expect Yaahlan to turn profitable within this year, and Amar should be behind it by half a year or so. The AI-driven product in MiraiMind remains in investment mode as we prioritize user growth and product capability. We don’t want to focus too much on reaching profitability -- the ability for that product. But of course, for each of the region we are in, we also have a payback period that’s required.” – 2025 Q4 Earnings Call

MOMO currently trades at $6.29, a steep discount to its $7.50 net cash balance. Even if you strip out the $1.5 Billion RMB in “statutory reserves” MOMO is required to keep onshore (which I think you should since that cash earns interest but is otherwise locked up), MOMO still has a net cash position of $6.13 per share.

MOMO insiders, including CEO Yan Tang, own 30% of the company (currently worth $300 Million) and are highly incentivized to do the right thing given executive compensation of $6.3 Million USD is just a fraction of their ownership stake. Insider selling from RSU’s has averaged < $10 Million a year, a minimal amount relative to both ownership and the company’s sizeable repurchases.

Simply put, at $6.29 a share, you are getting $6.13 in cash a share, along with an International portfolio of cash cow apps run by aligned insiders at a 100% payout ratio, for just a few cents. It doesn’t get more deep value than that.

MOMO has historically been very active in repurchasing shares in the $6 - $7 a share range (they repurchased 16.8 Million shares in 2025 at $6.84 a share), and I think they are likely to be even more aggressive in the $5 - $6 range, which creates a compelling setup if you are looking for a short-term trade. Of course, Dating Apps and AI Companions will never be seen as a “premium” or “quality” business, so I am not expecting it to trade at 10x or 15x earnings, but even a small re-rate to 8x earnings would imply $7.71 a share, significant upside from the current price and still just 1.25x MOMO’s cash pile (and below Tangible Book Value of $9.75). With all things China, it makes sense to approach your sizing conservatively (< 2.5%), but at the current valuation I believe the market is ascribing MOMO’s app portfolio zero value, and that it is worth considering a starter position at these prices.

Trade 2: “Japanese Healthcare at 6x Earnings”

Read more

Quick updates on two high conviction names: Modern Dental Group (3600 HK) and CBL Properties (CBL US).

Modern Dental Group (3600 HK)

2025 P/E: 8.9x

2026 Forward P/E: 8.4x

Dividend Yield: 3.45%

Insider Ownership: 67%

Modern Dental Group released a positive profit alert last night, indicating 2025 Net Income surged to $600 Million HKD, a 48% increase over 2024’s $406 Million HKD.

I said it in the recent Trade Review and I’ll say it again: Modern Dental is one of my favorite HK Small Caps.

Modern Dental Group (3600 HK) is a global dental prosthetics (crowns, bridges, dentures, implant components, and aligners) manufacturer listed in Hong Kong, but its largest markets are Europe (48%) and North America (22%) where it operates 80 dental labs under local banners. It’s innovative approach of rolling-up local dental labs and inserting its offerings has enabled it to grow like a weed: it slashes overhead, reduces product costs, and increases distribution while avoiding a “made in China” stigma since the labs operate under local banners.

I believe Modern Dental Group trades at a deep discount to its intrinsic value due to lower liquidity and Western wariness over HK Small Caps - all too often “too good to be true” stories in Hong Kong turn out to be shams, casting a dark cloud and disbelief over many value setups.

Modern Dental Group, with its strong M&A track record, shareholder-first capital allocation (share repurchases in the 3-4 range, 2025 Dividend: $0.199 HKD, 2024 Dividend: $0.17 HKD, 2023 Dividend: $0.104 HKD), and significant insider ownership (67%), however, continues to shine through. The stock is up nearly 120% over the last three years as the good times roll, and Hillhouse Investment Management (an Asian investment firm with $60 Billion in AUM and a specialty in Healthcare Investments) recently acquired a 17% stake at $5.80 HKD from the founding Ngai family, signifying rare Institutional interest in a HK Small Cap. I believe Hillhouse’s January purchase marked a turning point: Modern Dental Group is now a value setup that is compelling enough for Institutions to buy-in at a premium.

The positive profit alert is a welcome surprise. It affirms Modern Dental Group is already generating significant returns from its acquisition of Hexa Ceram (Thailand), and Digital Sleep Design (Australia), both of which closed in 2025, and the company is likely to report earnings of $0.64 HKD for 2025 (a CAGR of 19.25% over the last decade) and a bump in the dividend when the Board of Directors convenes later this month.

As the company continues to grow, attracts Institutional attention, and returns cash to the shareholders, I think Modern Dental Group will prove that the “HK Discount” applied is unwarranted: if it were listed in the US or EU it would likely demand a much higher valuation. Based on the positive profit alert and estimating moderate growth of 5-7% in 2026, I believe Modern Dental Group shares are within striking distance of $6.50 HKD - $7.80 HKD, which still implies a very reasonable multiple of 10x - 12x for a business with its track record of expansion.

CBL Properties (CBL US)

2025 P/FFO: 5.8x

2026 Estimated P/FFO: 5.6x

Dividend Yield: 4.76%

Insider Ownership: 30%

CBL Properties, an operator of lower-tier Malls and Open-Air Centers, announced a transformative financing package last Thursday that blows a hole through the “bear case”. For years, CBL detractors have asserted the company deserves a low valuation due to the “music stopping” in 2027 when CBL’s sizeable Term Loan matures. It was a reasonable concern to have, no-one wants to own a melting ice cube and if CBL wasn’t able to successfully re-finance, it’d have to hand over the keys on more than a dozen properties that accounted for 25% of FFO.

After last weeks refinancing, however, those concerns have been put to bed. CBL refinanced its $634 Million 2027 Term Loan with two smaller packages, a $425 Million Non-Recourse Term Loan secured by 13 properties (7.40%, Matures in 2031), and a $176 Million Non-Recourse Mortgage secured by 3 open-air centers and one mall (SOFR + 4.10%, Interest-Only, Matures in 2031 with two optional one-year extensions). While the loans carry a slightly higher weighted average interest rate of 7.50% vs 6.42% on the previous Term Loan, the package is massively accretive to cash flow.

How? Now that CBL is on much more solid financial footing, lenders are willing to accept significantly lower amortization: the $425 Million Term Loan is amortizing on a 22.5 year schedule, and the $176 Million Mortgage Loan is interest only. As a result of significantly decreased debt paydown, CBL’s 2026 guide for principal amortization has dropped from $90 - $95 Million to $58 - $63 Million, implying an additional $32 Million in discretionary cash flow. Even after accounting for $4 Million in additional Interest Expense, the net impact to CBL is an extra $28 Million in Discretionary Cash Flow, a massive 42% increase to 2025’s $66 Million in Discretionary Cash Flow.

Simply put, CBL’s financial engineering pushes out the “debt wall” 5 years and sacrifices just $0.18 out of its $6.90 in Funds From Operation (due to $4 Million in additional Interest Expense) to boost Discretionary Cash Flow from $66 Million to $94 Million. When your stock is trading at just 5.50x Funds From Operation, incurring $4 Million in Interest Expense to free up $28 Million for dividends and buybacks is exactly what you should be doing.

CBL is no stranger to capital returns, in 2025 the company returned $75 Million ($2.50 per share) via dividends and $18 Million ($0.60 per share) via share repurchases on roughly $66 Million of Discretionary Cash Flow. With 2026 Discretionary Cash Flow likely to come in near $94 Million, CBL has plenty of additional firepower to raise the dividend and repurchase shares.

It’s important to note that (just like the previous Term Loan), CBL’s new Term Loan and Mortgage are entirely non-recourse - the company can simply walk away from bad assets if mall Real Estate plummets. Additionally, all collateral pledged for the new Term Loan and Mortgage already served as collateral for the previous Term Loan (meaning CBL did not have to add collateral), and in fact, the Northgate Mall in Chattanooga, TN was released as collateral, offering CBL flexibility to sell or borrow against this property.

Despite the progress CBL has made in delevering the balance sheet, pushing out the “debt wall”, and boosting Discretionary Cash Flow (not to mention 2025’s copious dividends and repurchases), the stock still trades at a massive discount to Simon Property Group, Macerich, and Tanger. While these peers clearly have higher class properties, the discount is astounding: even highly distressed entities trade at a more respectable FFO multiple (and they aren’t even repurchasing shares or paying a dividend). I am by no means calling for CBL to trade at 12x FFO or 8x FFO, but even a small bump to 6x - 7x FFO implies $44 a share at the midpoint, significant upside from the current $38 price.

Given the company was repurchasing shares around $35.62 in December, I think $37.50 (5% above their repurchase price) represents an attractive entry. CBL will likely never trade at a peer multiple due to its past history of implosion (this time I think things are different - the balance sheet is 98% non-recourse) and lower quality properties, but I think the elimination of the bear case, improving sentiment surrounding retail Real Estate, and a massive decrease in debt-paydown due to the refinancing package offers a compelling-setup at the current valuation. As long as CBL continues to return cash to shareholders, gradually delever, and maintain a non-recourse balance sheet, I think the setup will continue to work.

There is no need to subscribe, there is nothing below the paywall.

Read more

The S&P is down 1.34% YTD, practically unchanged. After a 25.95% SPX return in 2023, 25.00% in 2024, and 17.86% in 2025, this should feel like nothing, but Fintwit seems to be singing a different tune. To their credit, things really do feel different.

Underneath the headline -1.34% figure, whipsaw and dispersion are at the highest levels since 2008.

The whipsaw isn’t even limited to sector-vs-sector, sub-sector volatility is also raging. As an example, the broader Real Estate Index is up 6.29% yet Office REITs have plunged more than 21%. Financials are down 7%, yet Regional Banks were up nearly 14% at one point and are still clinging on to tiny YTD gains. To top it all off, traditionally sneered-at capital-intensive sectors such as Commodities, Chemicals, and Oil & Gas are leading the way with 30%+ returns.

Fund Flows are notably different. For the first time in years, QQQ inflows are flat while investors clamor into RSP (the Invesco Equal Weighted S&P ETF) and the Korea craze continues as investors remain undeterred by the largest drop since 2008.

While it is easy to blame these moves on the Iran War or Tariff Policy, most of them have been bubbling up since the beginning of the year and I think the widespread rotation and dispersion was bound to happen anyways.

The rose-colored glasses we had on for the last three years are coming off, and the AI bull case that powered a two-year “everyone is a winner” rally in everything aside from a few phone-support stocks (Five9, Concentrix, etc) seems to be increasingly two-sided. These days, it seems one name’s AI Bull Case is another’s Bear Case.

Pieces such as Citrini's Global Intelligence Crisis have gone viral, and I think these thought exercises are valuable: the World is changing at a light-speed pace, and it’s worth questioning assumptions and asking “what if’s” to see if your thesis is still compelling.

Just this weekend, Bloomberg reported Stripe and Circle are racing to build their own payments systems for autonomous agents. I’m sure you’re skeptical, so am I. But also consider in 2022 AI was a gimmick that made pixelated videos of Will Smith eating spaghetti, and now in just a few sentences, you can draw up a lifelike-video with near Hollywood quality. Tools like Claude and Shortcut can save minutes, even hours in Excel. Claude Code and Codex aren’t perfect, but they can turn one man into a one man army.

Just imagine where things are in a few years.

If Visa, Mastercard, and Amex are unwilling to let their payments systems be used for autonomous transactions (due to concerns over chargebacks or fraud), someone will create a system that will. In 5, maybe 10 years, you might forget to buy Valentine’s day chocolates and flowers, and perhaps you can load a Claude or OpenAI “Wallet” with 100 USDC or PYUSD (PayPal’s stablecoin) and simply tell it what you want it and where it should be delivered. Is this a certainty? Of course not. But there is a slim chance it happens, and the disruption risk it represents is easy to understand when many names trade at historical valuation highs and shareholders emphatically claim “it won’t happen”.

As an aside, even after Amex’s 20% December - March drawdown, Amex still trades at just a 2% discount to the 5-Year historical valuation average.

I’m the last guy to call a top, and in fact, I’m not going to. But from a stock-picker’s perspective, I think 2026 is shaping up to be fundamentally different: the playbook of piling into large-cap “quality” regardless of the multiple is no longer working, and I think this is when active management and conservative position sizing (keeping names below 10%) begins to shine, especially when it comes to steering clear of value traps such as troubled Software Oriented BDCs, no matter how cheap they appear.

Since I’ve started this blog, I’ve covered about 50 different trades, most of which are still open, across Financials (Banking and Insurance), AI Infrastructure and Memory, and Global Equities. Over the past two weeks, I’ve been making many adjustments to my portfolio: I’ve entirely closed a few winners, been stopped out on a loser, raised cash to 20%, and added 2 new names where I think the risk-to-reward is compelling despite the current turmoil.

To be entirely transparent with you, my Portfolio is heavily skewed towards Small Cap Value and International Equities. The S&P Regional Bank Index has a market cap of less than $600 Billion (a tiny fraction of the S&P 500’s $59.94 Trillion), and countries such as Sweden, Japan, Australia, Georgia, Canada, and Kazakhstan make up less than 10% of the World’s Market Cap, but a much larger portion of my Portfolio.

As a result, in this Trade Review I’m going to cover the trades I am in, what I’ve been trimming, what I’ve been adding to, and my conviction on each trade, but ultimately the weights are up to you to determine. I want this blog to be a place to see ideas and setups you might not usually see, when I write up Small Cap setups such as International General Insurance, CBL Properties, Chain Bridge Bancorp, and Georgia Capital, it’s to get these names on your radar. It’s entirely up to you to conduct your own due-diligence, determine if you find the thesis attractive, whether these trades are suitable for your portfolio, and of course what the appropriate weight might be.

The Trade Review is organized into:

1. Trade List and Performance Review

2. Trade History - Removals and Hedging

3. Trade History - Additions and New Positions

4. General Commentary and Conclusion

Read more

Kazakhstan

Kaspi reported earnings of $2.76, missing the consensus of $3.34 due to higher taxes and bank reserve requirements impacting its Fintech division, but there is more to the story here.

For several quarters Kaspi has been plagued by an absence of capital returns as it poured billions into its acquisition of Hepsiburada(the second largest E-commerce firm in Turkey). Now, Kaspi is finally turning its attention back onto the shareholders.

Letter from Mikhail Lomtadze:

Two topics have dominated my conversations with investors over the last year: our progress in Türkiye and our approach to dividends. I want to address these and the other questions we hear most often, so that you better understand the context behind the decisions we’re taking.

Both Türkiye and dividends connect to a single ambition: to build Kaspi.kz into a 100 million user company. We believe that every strategic decision we make – the launch of new products and services, investments in Türkiye, and dividend policy – is anchored to this goal.

Earlier today, Kaspi announced an 850 KZT dividend per ADS, which translates into roughly $1.71 at the current exchange rate. The company also stated it believes this dividend is now the baseline going forward.

Based on the business's current performance and cash generation, we believe this is sustainable for the remainder of 2026.

Absent any exchange rate fluctuations, $6.84 in annual dividends represents a roughly 9% yield, immediately vaulting Kaspi from a flailing “value trap” into a cash cow.

Even after lowering my 2026 EPS estimate to $11.75 USD (sharply below the $13.08 consensus), due to more stringent regulations on the banking sector and a higher effective tax rate, Kaspi at 6.5x earnings and a 9% dividend yield still represents a compelling setup.

Kaspi is a “superapp” in Kazakhstan, their proprietary payments, e-commerce marketplace, travel, lending, and grocery divisions dominate the country. Practically every Kazakh adult uses Kaspi. While the government has taken action to “open up” Kaspi’s QR payments division (to allow clients with bank accounts at other institutions to also receive payments), Kaspi’s position in e-commerce is unchallengeable, and remains highly linked to a growing Kazakh economy.

With a massive 9% dividend yield now in place, I believe the bottom for Kaspi is finally in. Kaspi is a growth name trading at a deep-value multiple, and I think the recent share buyback program paired with the long-awaited resumption of the dividend is finally enough to change the narrative. I have increased my position to 2% of my portfolio, and have orders to buy more in the $70-$75 range should it trade down in the coming days - Kaspi is a high conviction name for me. Based on the current multiple and implied 9% yield, I think a more reasonable valuation for Kaspi would be $90 a share, implying a 7.5% dividend yield and 7.5x 2026 earnings.

Japan

A lot of times I get asked if “it’s still a good time to add to Japanese positions”. It’s a good question, the Topix is up 50% in the last year and many Japanese Banks and value setups that previously traded at just 5-6x earnings now trade at 10-12x/ It’s hard to determine if the setups are still compelling at first glance.

An excellent guiding point, however, is if the company finds its own shares attractive. Unlike US-listed equities where share repurchases are reported quarterly, Japanese-listed equities announce share repurchase authorizations and buyback activity on a monthly basis, with detailed reports denoting the exact price and volume executed on an individual day reported at the end of the month.

While these disclosures mildly complicate buybacks since they tend to “tip the hand” of the company to investors (sometimes leading to buybacks being massively front run), they provide an excellent way to determine if a company still has an “appetite” for its shares after a run up in price.

One of the names we discussed last summer has been on an impressive 50% run, but February’s repurchase report indicates their appetite is just getting started and the setup is still highly compelling at just 0.61x Tangible Book Value.

Read more

Earnings

Two names we discussed, Comfort Systems and Ategrity Specialty Insurance released Q4 earnings yesterday.

Comfort Systems, a leading provider of mechanical, electrical, plumbing, and HVAC Systems, reported blowout results: $9.37 in earnings per share versus the $6.76 consensus and $2.646 Billion in Revenue vs the $2.337 Billion consensus. Backlog grew a record $2.56 Billion to $10.1 Billion as demand surged from existing clients and new data center projects across the board. Comfort Systems remains a top infrastructure pick of mine, along with IES Holdings , Powell Industries , Sterling Infrastructure , and Power Solutions International . While Comfort System’s blowout results likely captivated everyone’s attention, I think Ategrity Specialty’s quiet beat stole the show.

Insurance has been a chaotic sector this year. Declining Global P&C Rates and intensifying competition in E&S (Excess and Surplus Lines) and Reinsurance has triggered a broad sell-off. Interestingly, many small cap names posses the agility to dodge overall rough market conditions have been punished the most, as investors rally around “large-cap quality” and ferociously buy-the-dip on “the generals”, leaving smaller names out to dry. I think Ategrity has been mistakenly left behind.

Ategrity Specialty is (as the name implies) a specialty insurer focused on low ticket Excess and Surplus lines for small and medium sized businesses across the United States. Ategrity distinguishes itself via a concentration in Casualty (liability), and its brand new technology stack has enabled it to reduce turnaround times despite overall policy pipeline growth of 250% in 2025. Its fast turnaround times and focus on traditionally underserved small businesses and states like North Dakota, Ohio, and Nebraska has helped it steadily poach market share from peers, with 2025 gross written premium jumping 30%, nearly 20% above the broader E&S industry.

Based on early 2026 monthly E&S Stamping data published from states such as Florida, it’s clear Ategrity’s growth and outperformance is showing no signs of slowing. Ategrity’s January 2026 Premium in Florida is up 14% YOY to $7.06 Million , while the more well-known E&S giant Kinsale Capital Group is down 6% YOY from $28.9 Million to $26.9 Million.

Despite excellent financial performance and clear growth ahead, Ategrity has languished the past few months due to three primary reasons: concerns about its reserves, jitters over the E&S market, and overall low liquidity. What really caught my eye yesterday is the fact Ategrity put all three concerns to bed.

Unlike previous quarters, Ategrity showed no prior year development:

We again had no prior year development. Catastrophic losses were 3.2% of net earned premiums, down from 3.7% last year due to very few catastrophe events in the fourth quarter. Our reserves are in a very strong position overall, both in property and casualty. You heard Chris mention in the prepared remarks that the early indicators for the recent years are coming in very strong. And so we are highly confident in our reserves there.

Of course, every insurance executive will want to state their reserves are adequate, but several quarters of “no bad news” on the reserve front is enough to be “good news” given how downbeat overall sentiment was.

Ategrity also affirmed it is still in “growth mode”, with internal projects such as “Operation Heartland”, a focus on “flyover states”, and new brokerage packages in Florida and New England are projected to help further increase market share.

Yes. So there’s two parts to Project Heartland. There’s an appointment component, adding more partners, and we are nearing the end of that phase. It’s more about getting more wallet share from partners. So we’re just at the beginning of that phase.

That is a, we feel like the investments we have made in the Midwest, not just in distribution, but in terms of developing products and really making a unique play in our verticals, is really what’s allowing us to stand out. So we see a huge runway for growth.

Lastly, and perhaps most importantly, Ategrity is taking steps to address the undervaluation of the stock.

Ategrity is tightly-held, founder Stuart Zimmer of Zimmer Partners owns 38 Million shares, ($760 Million, roughly 80% of shares outstanding). Due to the low overall float, liquidity has been thin and market activity has been highly choppy, dissuading new Institutional Investors and leading many June 2025 IPO buyers to throw in the towel.

Yesterday, Ategrity announced a new $50 Million buyback, which might seem small relative to the $960 Million Market Cap but once you realize less than $200 Million of shares are floating, the scale of the buyback becomes apparent.

From the transcript, it seems clear that they are saying one day Stuart Zimmer might want to sell shares, but only at a good price and that given the valuation Ategrity is trading at, they will be happy to repurchase shares until a premium valuation is hit. Given Stuart Zimmer himself was purchasing shares in the high 20s and 19s, it seems apparent that the buyback will provide significant support where Ategrity is currently trading, and that the risk of a secondary offering is minimal at the current valuation. Perhaps a concern at $24 - $26, but doubtful at $20.

Ategrity’s $0.51 report yesterday set an excellent tone for 2026: underwriting income tripled from $5.9 Million in 2024 to $15.4 Million in 2025, and the company clearly sees its shares as undervalued. While I am not expecting a triple for 2026, it’s apparent Ategrity is a scale story: now that they have their foot in the door via nearly 600 broker partners, new growth is highly accretive to the bottom line as investments in AI and technology keep underwriting costs relatively flat.

I think Ategrity found itself in the wrong place at the wrong time as investors mercilessly dumped insurance stocks and the IPO became considered as a “flop” Yesterday’s earnings report provided a highly important “reset”, and with the stock now trading at just 11x 2026 consensus earnings of $1.83 (frankly, I think Ategrity is within striking distance of $1.90 to $1.95) paired with a buyback program that amounts to 25% of the float, I think the narrative surrounding Ategrity is about to flip: Ategrity is drastically undervalued relative to peers and the broader market, and is perhaps one of the only insurance “growth” stories for 2026.

With Specialty Peers such as Palomar and Kinsale all trading in the 14x - 18x range, I think Ategrity is simply too cheap to ignore at 11x given an industry-leading growth profile and highly-aligned insiders. Given the excellent report, I have increased my position to 2% of my portfolio, and added Ategrity to my global basket of high conviction banks and insurers.

A Software Shot in the Dark

Up until Tuesday, I had zero software names in my portfolio. Now, I have one.

Old habits truly do die hard, I’ve traded Bumble on and off since 2024 and I have once again bought a speculative position. I’ll keep this one brief since Bumble is pretty obviously a melting ice-cube and a loser, but I also think there is a price for everything.

Bumble currently trades at 4x 2026 GAAP projected EPS of $0.75. Users are fleeing (I anticipate losses of another 150k in Q4) and the company is priced for terminal decline. Not everything is as it seems, though.

A good chunk of Bumble’s $1.15 Billion in liabilities was a complex $399 Million Tax Receivable Liability, cash due to Blackstone and Bumble Founders due to complications surrounding a reorganization of partnership shares into C-corp shares. Subsequent to the quarter end, Bumble was able to “buy out” this liability for $186 Million in cash at roughly 44 cents on the dollar, reducing overall liabilities to less than $800 Million (and its cash balance to $120 Million). Looks like Blackstone wanted to “cash out”, given they’ve also filed Form 144’s to liquidate their shares.

Given Bumble’s history of perpetual disappointment I would by no means qualify management as competent (stock is worth less than 5% of what it was when it IPO’ed) but given current cash flow it looks as though Bumble would have to intentionally try to run the company into the ground. One thing that caught my eye last quarter was the impact of the Apple App Store ruling.

In April 2025, Judge Yvonne Gonzalez Rogers ruled that Apple willfully violated a 2021 injunction in the Epic v. Apple case, and banned Apple from charging any commission on purchases made through external links in iOS apps. Although this decision has been partially walked back after an appeals court ruled “Apple can charge a reasonable fee” for external payments (links that divert to Stripe, etc), this reasonable fee has not yet been determined.

As a result, there is a goldilocks period where Apps are temporarily able to redirect existing clients to external payment methods such as Stripe, and entirely cut out the 30% due to Apple. For names like Match and Bumble, they are still charging the same price on both Apple-native and external payments, but it is obvious that any customer switched to an external payment rail, even if enticed with a “one month freebie” or a discounted annual rate via a marketing deal, is vastly more profitable when Stripe’s cut is 5%, versus the app-store at 30%.

The earnings call itself hinted at this:

Okay. So, no, I haven’t seen the details, so I don’t have any insight to share there. I can comment briefly on, sort of our direct billing initiatives. As you might expect, we, as soon as permitted, we set up alternative payments, and we’ve been testing those strategies throughout Q3, and we’ll continue to do that in Q4. You saw in our cost of sales a meaningful improvement in Q3, and we would expect to continue to, while we will, refine strategies around alternative billing throughout Q4, we’d still expect that benefit to persist into Q4, and we should have a full quarter effect of cost of revenue benefits.

Even if the appeals court enforces a “reasonable fee” that is levied on external payment methods established during this grace period, it’s obvious that the net effect will still be positive for Bumble provided they are able to switch users in Q4 and Q1. A mere 20% move of users from a 30% commission payments rail, to a 5% commission payments rail, could significantly boost earnings and cash flow, yet analyst consensus for Q4 and Q1 does not seem to price in the possibility at all.

Could Bumble blow up? Certainly, although I think it might be the first company in history to “marketing spend” itself into bankruptcy. The possibilities for cost cuts are endless: $119 Million in R&D, $189 Million in Marketing, $141 Million in G&A.

The Bumble founders still have decent equity in the company, roughly 25 Million shares, a little more than 15% of the company. I think after this epic implosion, they will be incentivized to enact cost cuts across the board to save their 8 figure salaries and equity stake. A mere 10% reduction in R&D and G&A (which seems possible given Claude Code and the implementation of AI) could add nearly 20 cents in earnings per share.

I get the Software-pocalypse, a lot of new tools seem to pose an existential threat to many businesses. I think the differentiator here, however, is Bumble’s edge isn’t the matching of users, it’s the users themselves.

Even if you had 1,000 upstart dating apps that could match regional bank lovers and SK Hynix fans with eachother at pinpoint accuracy, it would only really work if you had hundreds of thousands of users on the app. While Hinge is likely the only “growth” story in dating apps (Match Group also seems cheap to me), Hinge, Tinder, and Bumble still have massive user bases, and I think at the current valuation Bumble has several options at its disposal to prevent a total implosion and start returning value to shareholders.

Given Bumble’s history of implosion, however (not even kidding they usually drop 30% on each earnings report), caution is warranted, so I’ve sized the position very small at 0.50% of my portfolio. At 0.50% it’s not large enough to cause havoc, but if you sprinkle in a few Chinese AI names like Minimax, or Georgian Discount -to-NAV setups like CGEO, they can really add torque to returns while being isolated enough to eliminate catastrophic implosion risk, a 60% drop on a 0.50% stinks, but isn’t the end of the world.

Overall, I am way more confident in Ategrity Specialty over the next 10 years than Bumble, but there is a price for everything and I think Bumble below $3 a share may warrant a small, highly speculative position. At 4x 2026 earnings, plenty can go (and has been going) wrong, but with expectations so low (just like the Ategrity setup) we only need a little bit to go right for things to work out.

Next weekend, I will be posting a Review of my US Portfolio. Until then, have a good one. There is nothing below this paywall but I can’t figure out how to publish without one.

Read more

They told us 2026 would be the “Golden Age”. What they meant, perhaps, was the “Golden Age of Disruption”.

This piece is a departure from the usual, a rant about AI, followed by earnings updates from Portfolio Holdings, and finally two names I have been adding to recently in Infrastructure and Insurance.

It’s not hard to see how fast AI is evolving. Three years ago, it was a fun gimmick: videos of Will Smith munching on spaghetti and screenshots of people trying to “trick” ChatGPT into answering ridiculous questions. Now, we have models generating 3-D Worlds, production class videos, custom dashboards that may have taken weeks or months to build, and as of yesterday, reviewing and analyzing legal contracts. Things have moved, quickly.

Everyone seems to be hard at work guessing when the Memory and Storage cycle tops, and there’s no doubt given the recent run of Hynix, Samsung, Micron, Western Digital, and Kioxia that there is bound to be another -10% or -15% day and everyone will quickly declare it to be “over” (we had that last week for Western Digital and Kioxia, and Monday for the Kospi, Samsung, and Hynix). But what if the cycle lasts longer than a mere year or two? What if the pace of new AI features, model development, and data generation continues to accelerate. Do you think that is reflected in the price of Hynix at 6x 2026 earnings, or Samsung Preferred Shares at 6.5x 2026 earnings? Personally, I don’t believe so, which is why I own them.

I think we are just beginning to see AI features in daily life. Investment Bankers are using Claude, Shortcut, among others, to generate rough models in record time. Companies like Oura are using agentic AI to entirely handle customer service. Angie’s List is slashing their engineering and marketing department in half as productivity explodes higher. Digital Banks like Mercury now let you upload invoices that will automatically be processed, paid, and it sends a nice confirmation email to the company’s billing email address as well.

It’s not hard to use a bit of imagination to see where things are going.

What’s next?

Tax Software that can connect to your expense accounts and brokers, classify transactions, and have a return prepared quickly? Microsoft “revitalizes” Copilot, and all of the sudden you can just hit the Start Menu, ask it to organize your Downloads Folder, send a few emails, and purchase something online? Insurance software that can communicate with new home buyers and package together Inspection Reports and questionnaires, saving agents countless hours?

All of this sounds far-fetched now, and even if it is implemented you are right to assume the road will be bumpy and littered with errors, but give it 5 years to a decade and I have a feeling you will see all of these new products emerge.

The question is, how to play it, and what to buy?

I’ll be blunt, I don’t know.

I’ve sold my Nvidia shares and Powerchip shares, and I own a basket of Memory and Storage names (Samsung Preferred, Hynix, Western Digital, Micron, and Kioxia) as well as AI “Buildout” Names (Comfort Systems - probably expensive here, Power Solutions International, Powell Systems, IES Holdings). Is there still upside? Hopefully.

What I feel more confident in, however, is simply steering my Portfolio clear of setups with limited upside and decent downside (had to close FRSH out today). One group fits this in particular: “Direct Lending Funds” and BDCs.

I think when you take a look at someone like Adobe or Thomson Reuters (I’m no longer short ADBE), it’s easy to argue the moves are not based on short-term fundamental implications (both are reporting decent results), but rather shifting sentiment regarding the terminal value of their businesses. What will happen in 5 or 10 years?

While both names are fine long-term (from a credit perspective), imagine how the lenders to smaller software firms are feeling as they watch the equity of Large Cap software names implode.

The allure of Private Credit is that investment marks don’t change day-to-day. But with software names imploding left and right, do you really think the value of these loans has remained stable? Furthermore, do you think the guys extending credit at SOFR + 5.00% will be happy to do it again when these loans mature? I think the more likely answer is: They’re going to want their money back.

I love buying financials names at a discount, but I am entirely steering clear of Private Credit Funds and BDC’s specializing in software. I think in the coming years, their concentration in “Software Lending” will no longer be considered an asset, but rather a liability.

Take Saratoga Investment Corp as an example. It’s a name I used to own, and at first glance, it looks like a winner. A historical return of 16.2% since inception versus 15.7% on the S&P 500. Trades around 10x earnings, 0.92x Tangible Book, and a dividend yield of 12%. I don’t think anyone would argue it is an “expensive” name.

Once I take a look at the Portfolio however, things get less positive. Nearly 44% of the Portfolio is concentrated in software. Real Estate Software, Healthcare Software, Education Software, the list goes on.

Another publicly traded credit name, Blackrock TCPC (TCPC), recently recorded a 19% drop in Net Asset Value, driven by declines across E-Commerce (SellerX) and Software (Edementum) names, among others. Private Credit giants such as Blue Owl Capital Corp, Ares, and Apollo are down nearly 20% as people begin to worry about Investment Gone Bad.

My point is, with the Asset Managers themselves sounding the alarm, why take a risk by bottom fishing BDC’s or Publicly Traded Lending vehicles with sizeable Software Lending exposure?

At best, you have limited upside to Tangible Book Value (I don’t anticipate any of these names trading above Book Value due to the high management fees charged), and at worst, you are stuck with Portfolios with greater than 10% software lending exposure coupled with high fees, in an era where Large Cap Software Equities are getting halved left and right.

I get asked about “bottom fishing” BDC’s nearly every day, and I wanted to answer it clearly: I’m steering clear.

I believe that by merely staying away from falling knives exposed to AI, and instead adding a bit of exposure to names that stand to benefit significantly from the trend (such as Google, Samsung, Hynix, etc), you are doing yourself a favor.

Earnings

Several Portfolio Holdings reported earnings last week: OPY, CBNA, WSBF, NECB, and MBIN.

OPY - Oppenheimer stole the show last week with its strongest quarter since 2022: $6.51 in Diluted Earnings.

I really like Oppenheimer’s unique structure of a very stable Wealth Management business, coupled with a Capital Markets division that provides significant right-tail optionality when Small Caps and Biotech do well. Operating Income from the Capital Markets segment surged from a loss of $5 Million in 2024 Q4, to a profit of $52 Million in 2025 Q4. While I am not expecting similar results for 2026 Q1, given the strength of Small Caps, Biotech, elevated Equity Issuance, and record market valuations, I think Oppenheimer is firing on all cylinders, and is in a strong position to beat 2025’s $13.04 in EPS.

Unlike Banks and Insurers who come with significant left-tail risk when asset quality declines or catastrophes occur, Oppenheimer is a fee-based business, and they are free to do with their profits as they please. They are sitting on a clean balance sheet with no debt, and recently announced a $1 Special Dividend in December. While the name is choppy day-to-day, long-term I believe Oppenheimer will continue to be an outperformer, and that record results and a bit of time under the spotlight will demonstrate the compelling value it is generating.

CBNA - Chain Bridge reported earnings of $0.81, a decent beat relative to the consensus of $0.75. Perhaps most encouragingly, Chain Bridge recorded nearly $400 Million of deposit inflows in the quarter for GOP-linked Political Action Committees ($200 Million of which were diverted to other banks via the Intrafi network).

As a reminder, Chain Bridge is a highly unique bank. It has minimal credit risk as it simply does not lend out its Non-Interest bearing deposits, instead parking them in Treasury Bills and reserves. Chain Bridge is a beneficiary of higher for longer, and balances out other rates down names in my Portfolio (such as OP Bancorp and Waterstone Financial). I have a feeling this year’s Midterms will be vicious, and in an age where cash continues to pour into politics, Chain Bridge represents a one the only ways to “own the house”.

NECB – Northeast Community Bancorp reported $0.79, a bit lower than the consensus of $0.82. Asset quality remains pristine with no non-performing loans. NECB is beneficiary of higher interest rates, and stands to gain (just as CBNA does) from the prospect of reduced rate cuts this year. What I like most about NECB’s report is 40,000 shares repurchased below Tangible Book around $23 a share. With Tangible Book Value now upwards of $25.30 a share, I think entries in the high 23’s and low 24’s provide compelling value, you are getting the stock just under 8x earnings, and the share repurchase program is likely to provide support in the low 22’s, should it trade down there.

WSBF – Very strong quarter at $0.44. No estimates, but I am pleased with the results. Net Interest Income continues to grow steadily, and the mortgage segment posted the strongest results in years. The company still trades at nearly $2 discount to book, and repurchased another 140,000 shares to take advantage of this. I think the stock is well on its way to make $2 a share this year, which means it is currently trading at just 8.5x earnings, a very compelling valuation given the upwards optionality of the mortgage underwriting segment. Ultimately, I think the mortgage segment is worth a significant premium to book, and that WSBF will one day be acquired by someone looking to grow their own mortgage business. While WSBF’s mortgage segment is what most people care about, Net Interest Income has grown from $13.7 million in Q2, to $14.7 million in Q3, to $15.5 million in Q4. This steady improvement, as a result of declining deposit costs, will continue regardless of activity in the mortgage segment. WSBF still trades slightly below book value and closely comparable peers, and if mortgage activity increases, I think WSBF is within striking distance of $2.00 a share in earnings this year.

MBIN – Great results, and asset quality improved significantly. EPS $1.28 beat the consensus of $0.94 (although the $1.28 figure includes some one-time tax impacts). What was most impressive was the announcement of a new $100 Million share repurchase program, which should help change sentiment now that MBIN is back in “capital return” mode. Still, MBIN at 6x earnings was a bargain, but at roughly 8x I’m not so sure given lingering credit issues. I think the upside and downside are fairly balanced here at $42 a share.

Now, time for two new trades.

Read more

Morning fellas. Felt like sending this out via email rather than chat.

Read more

1. Introduction

2. What are Box Spreads / Portfolio Margin

3. Uncertain Taxation of Box Spreads and the BOXX ETF

4. Using Box Spreads to Borrow and Lend at Favorable Rates

5. Risks

6. Conclusion

1. Introduction

Box Spreads seem to have earned a mythical reputation recently. We hear constant whispers of tax advantages and attractive funding rates, yet concrete examples remain elusive.

(Pictured Above: Bloomberg Article on the BOXX ETF)

With ETFs such as BOXX (that employ Box Spreads to generate money market like returns) soaring in popularity, I wanted to write a brief piece on what SPX Box Spreads are, the uncertain taxation risks associated with Box Spreads and the BOXX ETF, and show a real life example of borrowing and lending $100,000 near the Overnight Bank Funding Rate (currently 3.64%) via SPX Box Spreads.

While Box Spreads enable market participants to borrow and lend to each other at highly favorable rates (currently 3.94% for lending, 4.07% for borrowing), it’d be remiss to ignore the fact tax treatment of Box Spreads and the BOXX ETF is uncertain.

There is considerable debate over whether Section 1258 of the Tax Code applies to income generated from loans facilitated by Section 1256 Box Spreads, and the future of current BOXX “loophole” may close - the IRS may one day issue a notice requiring BOXX profits to be reclassified as Interest Income, forcing investors to refile previous taxes and creating significant adverse tax consequences (here is an excellent article explaining that risk). There’s no such thing as a free-lunch.

Since I am not a Tax Professional, I am not taking a position on potential tax treatment - I will simply discuss what both sides argue. The tax implications of Box Spreads and the BOXX ETF is a matter for you and your qualified Tax Professionals to determine.

Additionally, the utilization of Portfolio Margin to borrow or lend via Box Spreads involves the use of advanced margin calculations and significant financial risk. Portfolio Margin is designed for experienced, sophisticated investors who fully understand the risks of leveraged trading and may result in rapid, substantial losses in excess of your account’s equity. Whether Box Spread borrowing and Portfolio Margin is appropriate for you is a matter for you to decide in consultation with qualified Tax and Financial Advisors.

Read more

Timing. Sizing. They have more to do with each other than you’d think.

What are the odds of “top-ticking” or “bottom-ticking” a name? Just look at the graph of cosine(x).

If you tried to “top-tick” cosine(x) on the way up at various points, good luck, nearly all of your attempts would end in tears. The same goes for trying to “catch a falling knife” on the downswing.

Individual stocks and the market as a whole follow a much more complicated regime than a simple cosine graph, but the one thing they share is that top-ticking and bottom-ticking is nearly impossible. My advice: don’t even try to do it.

Frankly, it isn’t just traders who try, companies do it as well with share repurchases. Take Designer Brands, for example. In Q3 of 2024 they spent $50.6 Million (a massive amount relative to the market cap) repurchasing shares at $6.59, leaving roughly $20 Million in capacity remaining on the existing authorization. The repurchase screamed “bottom tick” attempt. Shortly afterwards, the stock dropped to $5.00 a share in Q4, $4.00 in 2025 Q1, and into the low 2’s in 2025 Q2, before recovering to the present $3.60 - $3.80 range. DBI’s repurchase activity since 2024 Q3? Essentially zero.

People angrily shake their fists “If you spent $50 Million buying it back in the 7’s you should love buying it back at $3!” It’s a good point.

But ask yourself, how many times have you done the exact same thing? You buy a position, it dips to a level even more attractive than your original entry, you don’t add, and a few months later you wish you did. For me, way more often than I’d like to admit.

That’s why I think timing and sizing have a lot to do with each other. Once you understand that the odds you have timed the “bottom-tick” are slim, starting with a smaller position and leaving yourself with room to add at more attractive levels can come in handy. The worst thing that can happen is the name runs away from you, and you make slightly less than you would have if you bought your entire position at once.

If I’m looking to make something a 3% position, I’ll start at a third of the desired weight, 1%, and buy another third of the desired position (1%) over the next 5 days. I’m not a large trader and it isn’t a matter of liquidity, rather, it prevents me from rushing in at a top or blowing out at a bottom and making emotionally charged decisions. If you look back to Google at the low 160’s earlier this year and Google at 320 currently, bullish / bearish sentiment entirely dominated a few trading days. Spreading your buys / sells over a few days reduces the likelihood of making drastic decisions at emotionally charged moments (of course, you can always make exceptions for moments driven by earnings, regulatory issues, etc).

I like to use the last third of my “firepower” to add at more attractive levels if the price moves against my entry, or add to my position at higher levels upon a breakout. This might sound counterintuitive; why save capital to buy more of something once it is already moving higher? The purpose of retaining the last third is to avoid “dead money”. Doing this gives you the option to add at more attractive prices, or establish a full position by adding at higher levels once a breakout begins; chances are once the breakout begins it has room to run.

In a market that is increasingly momentum driven, this tactic helps ensure your capital is deployed when the risk to reward is most favorable, not just when you first got excited about a trade. If you have a focus on smaller cap US Equities, or Global Value names that tend to float around until the breakout finally begins, saving a third of your firepower to add at more attractive levels, or upon a breakout, ensures you don’t tie up capital in a value trap.

Now, three trades I’ve been adding to recently.

Read more

You’re so cute, I wish there were two of you.

What if there were?

In the United States, many large publicly traded companies have dual classes of listed shares, all of which trade at a relatively tight spread to each other.

Google Class C (GOOG US Equity - $290.59) vs Google Class A (GOOGL US Equity - $290.10), Zillow Class C (Z US Equity - $73.05) vs Zillow Class A (ZG US Equity - $70.82), FOX Class C (FOX US Equity - $60.05) vs FOX Class A (FOXA US Equity - $66.53), in pretty much every instance even including Malone tracking empire (Liberty, Formula One, Atlanta Braves), different classes of shares tend to trade within 10% of each other.

There are a few global instances where this isn’t the case, and interestingly enough two of them belong to today’s AI darlings. There are various reasons why these discounts won’t close (one is a non-freely convertible ADR), but if you are a long-term investor, a believer in AI, and have an Interactive Brokers account, here are a few ways you can maximize “bang for your buck”.

1. Long Samsung Electronics Preferred (SMSD LI Equity) instead of Samsung Electronics Common (SMSN LI Equity)

Contrary to popular belief, you can buy Samsung if you are American, just not in America. More on that later.

Samsung has two classes of shares, the Common - 005930 KS Equity (88% of shares outstanding, $3.1 Billion USD daily volume) and the Preferred - 005935 KS Equity (12% of shares outstanding, $200 Million USD daily volume).

In the United States, Preferred Stock and Common Stock differ meaningfully. Preferred Stock usually carries no voting rights and receives fixed (occasionally floating) dividends. In the event of bankruptcy, Preferred Stock stands higher up on the capital structure and are entitled to liquidation payments prior to Common stockholders.

This isn’t the case in Korea.

Samsung Common and Samsung Preferred are almost identical. The Preferred Stock carries no voting rights, but the Common and Preferred have the same exact claim on the company. The only minor difference is that due to Korea Securities Law Samsung must pay a higher dividend on Preferred Stock than Common Stock. In 2024 Samsung paid a dividend of 1,447 Won to Preferred holders vs 1,446 Won to the Common, and in 2025 paid a dividend of 1,473 Won to Preferred holders vs 1,472 Won to the Common. As you can see, the difference in dividends is just a token amount of 1 Won.

Given the slightly higher dividend, you’d assume the Preferred would trade somewhat close to the Common, but that is not the case. In fact, it trades at a 24% discount. Why? The elephant in the room: passive ownership via index inclusion.

Only Samsung’s Common Stock is allowed into indexes such as the benchmark Kospi 50, Kospi 200, and MSCI Korea. As a result, as investors hungry for Korean equities and AI related names plunge into the Kospi and MSCI Korea, Samsung Common has benefitted from a massive passive tailwind, while Samsung Preferred has experienced no benefit at all. As a result of this passive buying differential, the Common Shares have outperformed the Preferred by 15% year to date, and trade at the highest premium in 5 years, 31%.

I’m not calling for a widowmaker trade of Long the Preferred and Short the Common – if you want to shove a fork in a toaster there are easier ways such as owning a value ETF, being an energy investor, or adhering to red lights as a cyclist in New York City (you will be rear-ended).

What I am saying, however, is that if you are a long term believer in Samsung, the Preferred is certainly looking attractive.

The Common is trading at a forward valuation of 12x 2026 earnings and a dividend yield of 1.43%, while the Preferred is trading at 9x 2026 earnings and a dividend yield of 1.87%. Samsung at 9x earnings? Don’t mind if I do.

While you can’t trade the Common and Preferred directly on the Korean Stock Exchange, luckily both are listed as GDRs (Global Depository Receipts) on the LSE International Order Book under the tickers SMSN and SMSD, and trade in US Dollars. The volume in these GDRs is notably lower than on the native Korean Stock Exchange (about $35 Million for SMSN and $5 Million for SMSD), but there is still sufficient liquidity to buy a handful.

I’m not expecting this valuation differential to close anytime soon as the passive flows into Korea continue, but if you want some Samsung exposure, why not consider the Preferred at 9x earnings and get an extra 0.44% in dividends? Food for thought.

2. Long Taiwan Semiconductor (2330 TW Equity) instead of Taiwan Semiconductor ADR (TSM US Equity)

Taiwan Semiconductor is unique not just for its dominance in chip fabrication, but also for having one of the world’s only one-way ADRs. Historically, local Taiwan semiconductor shares (2330 TW Equity) have been convertible in small quantities into the ADR (TSM US Equity), but the ADRs have never been convertible back into local shares.

Currently, 21% of outstanding shares are ADRs, and any additional conversion of local shares to ADRs has to be expressly approved by the company. ADR conversions haven’t occurred for years, and as a result American tech investors hungry for chip exposure compete for a tiny slice of the pie, and the price differential between local shares and the ADR has grown larger.

This year alone, Taiwan Semiconductor’s ADR (TSM US Equity) has outperformed the native listing by 4%, and the differential between the ADR and the Native Listing has risen to a record 30% before retreating to 24% this month.

(Note 1 share of TSM US Equity = 5 Shares of 2330 TW Equity, so a ratio of 6.2 means the ADR is trading at a 24% Premium to the local listing).

Once again, I am not calling for a widowmaker trade of Long 2330 TW Equity Short TSM US Equity Trade.

However, if you are a long-term believer in Taiwan Semiconductor, why not get a larger slice of the pie for the same amount of cash?

IBKR’s currency conversion is seamless (a few basis points), and while you do need to pay a 0.3% Financial Transaction tax when you sell and 0.15% in annual custody fees, the local listing trades at 20x forward earnings with a dividend yield of 1.30%, while the ADR trades at 24x earnings and a dividend yield of 0.98%. The lower valuation is attractive, the elevated dividend yield covers expenses, and you stand to gain from superior returns if the differential ever narrows to the historical range of 10% - 15%.

It’s also important to note tax treatment for dividends from ADRs and the local listing is the same for US residents. One concern you may have is that if China ever invades Taiwan, treatment will be different for ADRs vs local shares, but I can assure if China indeed invades Taiwan, what happens to your local shares will be the least of your concerns once you are on the front lines.

3. Miscellaneous – SK Hynix GDR

While the Hynix ADR in the United States doesn’t trade (aside from a guy bidding 100,000 shares at a penny), you can still own SK Hynix as an American. SK Hynix has a Global Depository Receipt listed on the Frankfurt Stock Exchange under the symbol HY9H that trades in EUR. Since the GDRs are freely convertible, just like SMSN and SMSD, it tends to trade close to the local listing.

Hynix has had an incredible run up this year, but I still find it interesting at 7x 2026 earnings. Furthermore, recent rumors that Hynix is contemplating a sponsored US ADR get me even more interested – a sponsored US ADR would enable a massive increase in passive ownership.

Korea itself is rapidly becoming investor friendly. The President literally promised to “double the Kospi”, and regulatory crackdowns on abusive related party transactions, and a government push for companies to return capital (similar to the Tokyo Stock Exchange’s shame list of companies who refuse to provide clear-cut capital return guidelines) have driven the stock market to record highs.

Hynix and Samsung moving forward with sponsored US ADRs would be a genius move. Imagine this from their perspective, pump your own markets 100% and then when things get shaky, list your two largest companies on the Nasdaq and watch the American passive money waltz on in. If Pinduoduo can be in the Nasdaq 100, so can Hynix (if they follow through with an ADR and list on Nasdaq).

The purpose of this isn’t to slam the table for Hynix as a buy, but rather, to let you know it is purchasable.

Finally, bonus round.

Read more

It’s been a busy start to earnings season. Five out of six names covered in previous articles reported excellent results, and more are set to report this week. I wanted to briefly discuss each report, some recent events (Argentina Elections and a Potential China Trade deal), and the setups I find most attractive.

At the time of writing (Sunday 7:39 PM), preliminary results indicate Javier Milei’s party, La Libertad Avanza, is headed towards a decisive victory.

I’m long a small basket of Argentinian Financial Institutions, all of whom were trashed this year over fears that Milei’s pivot to Free Markets would be overturned at the ballot box. Grupo Galicia, Banco Macro, and Banco BBVA Argentina have heavily underperformed the MSCI Argentina, and if Milei’s party indeed secures 40%+ of the vote I think the combination of US Support (via a $20 Billion Swap Line to support Argentina’s currency) and political certainty for the next two years will provide an excellent backdrop for these banks to recover much of their 2025 YTD losses.

I certainly don’t expect them to bounce back to 0% YTD within a day, but the names should be up 15%+ given the lethal bear case is now entirely removed. The reaction today is likely to be “kneejerk”, so if you are interested in Argentinian banks, I think it make sense to wait and see how things settle out this week.

Nonetheless, I think the “Argentina Comeback” trade is one that will last for a while given Milei’s free market reforms, and that regardless of today’s knee-jerk move, the positive trend among Argentinian financials, energy (YPF), and airports (CAAP), will continue for some time.

It also seems a “Sweetheart” trade truce between the US and China is on track. Both sides are already drawing attention to the meeting and praising the “tough but understandable stance” of the other negotiators, so that they can both play off an agreement as a win. Will anything of substance happen? Probably not, but mere reassurance is enough to get markets happy again.

Most names within the “Tariff Slime” basket have already recovered from the April 4th lows, but I think increased trade certainty could benefit names heavily associated with product sourcing from China and Southeast Asia such as Gigacloud Technology, Sonos, and Jakks Pacific, as well as tech names such as ACM Research (semiconductor cleaning equipment) given the prospect of a chip-war detente.

Between Argentinian Banks and “Tariff Slime”, I think Argentinian assets provide a more favorable long term setup, but given the event-driven nature of today’s market and a massive fixation on “themes”, I think both groups are worth paying attention to today.

Earnings Analysis and Setups

1. Customer’s Bancorp

Multiples: 8.9x 2026 EPS, 1.17x Tangible Book Value

Reported: $2.20 vs $1.90 Estimate

Takeaway: Strong Beat and Guidance Upgrade Boosts 2026 Outlook

Customer’s delivered a significant beat. The quarter was marked by excellent credit quality, accelerating loan growth, and increasing deposit activity driven by its newly hired banking teams.

The regional bank industry has been plagued by credit “cockroaches”, which makes Customer’s beat even more impressive. Customer’s dispelled any concern over its sizeable NDFI (Non Depository Financial Institution) loan book, stating:

Most of the recent attention this past quarter has been on the lender finance space. For Customers, this is one of the oldest specialized lending businesses we’ve been in. It’s one we’ve been in for over a decade, and we’ve not only had zero losses, but zero loan defaults. In this business, this is typically lending to a private credit fund where the collateral is a broadly diversifies pool of loans out to middle market companies. There’s significant overcollateralization. We have low advance rates and there is a -- it’s very diversified. So our single obligor exposure is very low, kind of mid single-digit. So when you put all those things together, it’s translated into, again, zero losses, zero defaults.

Customer’s also highlighted the success of its proprietary CUBIX offering. CUBIX is a private payments rail (similar to Zelle, but for Insitutions) that allows clients to transfer hundreds of Billions to other Customer’s clients. The primary use case for this is Institutional crypto activity. CUBIX is integrated across all major exchanges and most trading desks, and comes with an added element of security relative to pure stablecoins like USDC or Tether, since CUBIX can only be transferred to other Institutional CUBIX clients, and thus can’t really be stolen or hacked by third parties.

Customer’s has brought in nearly $3.9 Billion in zero-cost deposits (which they store in cash to minimize liquidity risk) via CUBIX, and growth has been accelerating as crypto trading volume hits record highs.

Customer’s provided very rosy guidance in terms of organic deposit growth, stating that its new 2025 hires are likely to give a “25% lift on the $300 - $400 Million” in quarterly deposit growth, starting in the middle of next year.

What I like about Customer’s is that time is on its side and that its successes build upon each other. Unlike other regionals stuck with low-rate securities or deteriorating office loans, Customer’s winning technology solutions such as CUBIX, and its highly successful new banking teams (poached from Flagstar, Signature, and SVB) contribute to the bottom line immediately and help it secure new opportunities and clients. Simply put, it’s a growth story, not a value story.

Despite Customer’s history of success and its impressive 46% YTD return, it still trades at a steep discount to the broader regional banking index (9x 2026 EPS, 1.17x Tangible Book Value vs 10x 2026 EPS, 1.47x Tangible Book Value for the index). Customer’s has differentiated itself from peers via its proprietary offerings such as CUBIX, its minimal credit risk, and its consistent growth track record, and as analysts slowly bump 2026 estimates up from $7.50 to $8.00 and as investors warm to the name, I think additional upside is present. A baseline case would be for it to trade at the Regional Bank peer multiple (implying $80 a share) and an optimistic case would be for it to trade at 11-12x earnings (implying $88 - $96 a share).

Read more

A lot of times success isn’t determined by how you trade, but rather what you trade. As markets become increasingly thematic, baskets and stories (Uranium, Rare Earths) seem to be more powerful than ever. I think desperation is a driver behind a good chunk of the move – many Americans, and in fact global investors, feel left behind and thus a need to break out of the rat race by generating massive gains. Companies with strong stories and narratives, combined with quick moves, help scratch this itch.

Are these investors putting the cart before the horse? Given a good chunk of the meme squad hasn’t generated a single dollar of actual revenue, it’s reasonable to say yes. Nonetheless, the price of a stock is simply what people are willing to trade it at 4 PM, and there are a few lesser-known names across AI Infrastructure, Critical Metals, and Defense, and Capital Markets that I think are worth a look. The purpose of this article is to briefly discuss them and bring them to your attention.

Bleichroeder Acquisition Corp (BACQ)

Bleichroeder Acqusition Corp is a SPAC intending to merge with Merlin Labs, an autonomous aviation contractor. Merlin Lab’s primary product, Merlin Pilot, combines proprietary hardware and software solutions with GE’s Flight Management System to fly a variety of aircraft, and can substantially reduce the need for large crews as Merlin Pilot can interface with Air Traffic Control and handle re-routing and collision avoidance.

Merlin has already won substantial long term business contracts. In 2024, Merlin secured a $105 Million contract from the US Air Force to retrofit C-130 Transport Aircraft with its Merlin Pilot system, and is in further discussion to expand this to the KC-135 Tanker Fleet. I think Merlin’s autopilot offering, which runs in partnership with GE’s Flight Management system (currently used on more than 14,000 military aircraft), has a strong opportunity to play a role in military readiness as drones and autonomous aviation play a larger role.

The SPAC transaction isn’t expected to close until early 2026, but what I like about BACQ is the favorable risk to reward of the trade. Spac Holders have the right to “redeem” their shares from the SPAC and receive cash. BACQ is currently trading at 10.88, and the redemption value is somewhere in the 10.40 – 10.50 range. As a result, your downside (up until the deal closes) is 30 to 40 cents, as you can simply “redeem” your shares upon the close of the transaction if the SPAC is trading below cash value. As a result, at 10.88, I think a favorable setup is in place as a move to $12-$14 would provide a significant gain relative to the 30 to 40 cents of risk you are taking.

Willow Lane Acquisition Corp (WLAC)

Willow Lane Acqusition Corp intends to merge with “neocloud” provider Boost Run. Boost Run specializes in “Bare Metal” GPU hosting solutions where a tenant rents the entire server. Boost Run does not own data centers, rather, it partners with Tierpoint Data Centers to offer capacity across Charlotte, St. Louis, and Pennsylvania. Boost Run’s founder, Andrew Karos, is well connected and previously ran high frequency trading firm Blue Fire Capital before it was purchased by Digital Asset Giant Galaxy Digital (who ironically, is also building data centers).

Boost Run has been profitable since inception in 2023, and has not needed outside Series A, B, or C funding.

Boost Run has never raised Series A, B or C funding. They’ve grown organically while maintaining profitability. This transaction values Boost Runat roughly $614 million on a post-money basis, with over $112 million of cash added to Boost Runs balance sheet, assuming no redemptions by Willow Lane shareholders. Boost Run’s 2026 revenue midpoint of $180 million implies an attractive 3.4 times forward EV to revenue multiple, significantly below the comparables. We are very excited about this transaction.

The company is cash flow positive, and is forecasting significant growth in 2026.

Our financial course demonstrates what we believe is a powerful model. Pre-cash flow positive since year one, and we are committed to this principle. 2026 estimated revenue is $180 million at the midpoint. This represents approximately a 470% growth year-over-year. We generate strong profitability within the neo-cloud industry with adjusted EBITDA margins of approximately 80%.

I hesitate to do “relative valuation” within the Neocloud industry, since practically anyone trades at sky high multiple (I sold Nebius at $30), but with Boost Run trading at an implied 5x EV / Revenue, it seems to be much cheaper than peers.

There is almost no coverage of Boost Run as of now, but I think that is likely to change quickly. Boost Run has an institutional call with DA Davidson later today, and I think heightened awareness will significantly benefit the stock.

While Boost Run is trading around 4 dollars higher than the SPAC redemption value (so significant downside is present), I think it provides an interesting opportunity to gain exposure to a neocloud provider at a sub-industry multiple.

Read more

1. The Collapse

2. The Comeback

3. Names to Consider

The Collapse

Ambulance chasers. Roofers. What?

The Florida insurance market was pushed to the brink of collapse in 2022 due to a highly litigious practice known as assignment of benefits. What started as a cottage industry in the early 2010’s quickly grew into a parasitic billion dollar combination of inflated claims, fraud, and millionaire sportscar-driving lawyers known for tying insurance companies up in court for years over roofing claims.

On face value, assignment of benefits is a simple tactic designed to make the insurance claims process easier. An individual claimant, whether for health, auto, or homeowners insurance gets immediate relief (surgery, auto repairs, emergency home restoration services) in exchange for assigning the financial value of their claim to a third party company, who then works out the cost with the insurer and receives the settlement.

The obvious issue with assignment of benefits is it creates a perverse incentive for repairs to be as expensive as possible. If you are a third party and know an insurer is going to pay you, you’re going to charge as much as you can legally get away with for repairs.

Florida roofers quickly realized they could turn roofing damage into a billion dollar industry by aggressively soliciting business. Highways were dotted with advertisements for “free roofs” and the practice soared in popularity from 2015-2022. After all, who wouldn’t want a free new roof?

Given the frequency of hurricanes in Florida, it was an easy tactic to get away with. Roofers simply claimed that regular wear and tear on older roofs (wear and tear is not a covered peril) was in fact hurricane damage, and if insurers wanted to fight this they’d have to go to court against attack dog lawyers who would also demand significant attorney’s fees when the case concluded.

The lawyers almost always won due a concept known as “social inflation” (jurors love sticking insurance companies with massive settlements and siding with the “little guy” as a way of lashing out). Given the cost to even fight a claim ranges from $50,000 - $100,000 in legal expenses, insurers just paid out. To the roofers, it was free money.

The scale and breadth of the shenanigans was widely apparent. News articles and academic studies discovered assignment of benefits claims were typically 30-50% more expensive due to inflated, sometimes fraudulent costs, but insurers had no tools to stop the slow motion train wreck.

Like the story of the scoprion and the frog, eventually parasites get large enough to kill their host. By 2022, after suffering Billions in losses, most insurers were high-tailing it out of Florida and most of those who remained, collapsed.

FedNat insurance, a Florida specialist, saw income plunge from $35 Million in 2014 and 2015 to -$120 Million in 2021, and went into liquidation in 2022.

American Coastal (previously known as United Insurance Holdings UIHC), had its United Property and Casualty subsidiary fall into insolvency, and the stock collapsed from $30 to just 62 cents as net income swung from $40 Million to -$400 Million.

Heritage Insurance faced similar disaster as income plunged from $30 Million to -$60 Million and the stock was “written off” after losing 90% of its value.

By late 2022, it wasn’t a mere corporate issue. State lawmakers began to panic.

Florida’s government runs an insurer of last resort called Citizens. The insurer is only open to homeowners that cannot find insurance elsewhere (at a somewhat reasonable price), and was originally intended to fill very small gaps left by private insurers.

As Gulfstream, United, Fednat, Southern Fidelity, Lighthouse, Avator, and Weston went bankrupt and Farmers, AAA, Allstate, and Progressive fled the state, more and more homeowners flocked to Citizens and it quickly became overloaded. After significant damage from Hurricane Michael (2018 - Category 5, $8 Billion in damage) and Hurricane Ian (2022 - Category 5, $50 Billion in damage), it was clear to even the lawmakers that Citizens was on its last legs.

Faced with the prospect of a total insurance industry collapse, lawmakers had to act.

The Comeback

In late 2021, Florida banned aggressive roof advertisements, and determined that claims for damaged roofs would be limited to actual cash value rather than the cost of a new roof. While this was well intentioned, it did nothing to stop roofers from inflating the “cash value” of old roofs to get away with larger payouts.

In early 2022 as the situation worsened and public outcry mounted, the state established an emergency $2 Billion reinsurance fund to help insurers find reinsurance coverage (a giveaway to private insurers), hoping they would pass some of the cost savings to policy holders. They didn’t.

Faced with an imminent insurance collapse, in late 2022 and early 2023, state legislators chose the nuclear option.

They entirely banned assignment of benefits on new policies, eliminated one way attorneys fees making policy holders liable for their own court costs, and reduced the statute of limitations on property damage claims to just one year.

Assignment of benefits and most fraud ended instantly. Property claims declined 40% in 2023, and insurer’s legal expenses plunged from 10% of premiums to just 3% as they no longer needed to fight frivolous claims.

The effect was immediate and insurance profits exploded higher. Increasing margins were compounded by the fact there were only a few surviving insurers left in the state, meaning there was no immediate competition to take advantage of the gold rush.

Growth accelerated as Florida took steps additional steps to wind down its Citizens Insurance operations. Private insurers were allowed to assume Citizens policies, as long as they promised to not raise premiums by more than 20%. These Citizens “takeouts” are wildly profitable to existing private insurers as they come with lower brokers fees and private insurers don’t have to pay for reinsurance during the first year, providing a significant boost to net income upon policy assumption. Citizens polices have since dropped from a peak of 1.4 Million in 2023 to 800k in June 2025.

Despite the incredible megatrend, the Florida Comeback has gone widely unnoticed. Aside from a few scattered posts and Jonathan Cukierwar’s 20 page write-up on ACIC, it’s hard to find any mention of the turnaround.

Heritage Insurance (HRTG) is up 22x from $1.12 to $25.30. American Coastal is up 41x from $0.29 to $12.07. Universal Insurance Holdings is up 3.5x from $8.39 to $29.84. HCI Group is up 7.1x from $27.65 to $196.94.

While the runup has been incredible, I think further gains are to come.

For starters, we have miraculously made it through hurricane season without any major windstorm incidents. I don’t want to jinx things, but if we get to November 1st without any hurricane damage I believe the entire sector will drift higher as the one year statute of limitations implies wind damage claims for the 2025 and early 2026 period will be non-existent and loss ratios will plunge.

Simply put, you can’t blame roofing damage on a hurricane if there are no hurricanes.

Additionally, the entire sector stands to gain from improved reinsurance pricing.

Insurers never bear the entire risk of their policies since a single hurricane could wipe them out. As a result, all of them purchase reinsurance coverage from larger entities such as Swiss Reinsurance, Reinsurance Group of America, Everest Group Re, Axis Capital, Arch, and Renaissance Insurance among others, to split and cap their risk.

Given the vast increase in profitability in Florida policies, not only are the primary insurers like UVE, HCI, ACIC, SLDE, AII, and HRTG benefiting, the reinsurers themselves are making a fortune since reinsurance payouts are significantly lower. As a result of increased profitability, and no elimination of reinsurance capital due to sizeable hurricane losses, it’s becoming increasingly clear that reinsurance rates for the gulf region are set for double-digit declines in 2026.

Lower reinsurance costs and stunningly low loss ratios will provide a massive tailwind for the entire Florida sector in 2025, 2026, and 2027, but I believe the market is not yet pricing this in for a few select names.

Your first thought might be improved profitability will generate increased competition among the Florida specialists. That’s true. However, homeowners insurance is a very sticky product. Switching insurers is real pain as brokers require intricate wind mitigation reports and extensive documentation to accurately price risk. Unlike reinsurance which is an active, competitive market (insurers are always shopping for the best rates), even if homeowners insurance rates decline 1-3%, to someone paying just $4,000 or $5,000 it might not even be worth the hassle to switch. As a result, the good times should continue to roll for the Florida specialists all the way into 2027. Here’s how I’d play it.

Names to Consider

Before I get into the individual trades, I wanted to do a brief overview of the sector, and what the individual insurers are known for.

Heritage Insurance has sizeable operations in Florida, and growing exposure to New York and California. Both states have rapidly improving market conditions.

Universal Insurance Holdings has minimal geographical diversification, and is thus highly dependent on Florida insurance rates. While they handled the 2022 implosion better than others, they are still more than 30% below their 2018 high.

American Integrity has a unique focus on new build homes, and intends to significantly expand operations in South Florida, where housing is booming.

American Coastal has deep underwriting expertise and is known for its focus on low-rise apartment and garden style condo buildings. It only covers damage and does not cover liability, lowering its exposure to massive “social inflation” slip and fall claims. It’s also a heavy reinsurance purchaser, and stands to benefit significantly from lower reinsurance costs.

HCI Group is known its advanced risk modelling and expertise on Citizens “takeouts”. Its underwriting model subsidiary, Exzeo, is slated to IPO in 2026. While it may seem expensive on a P/E basis, Exzeo is likely to trade at a significantly higher multiple given the capital-light nature of the business. HCI also has a great track record of Office investments in the Tampa area, which lends credibility to the sharpness of the management team.

Slide Insurance recently went public in June and is a new entrant to the Florida market. Slide touts extensive AI capabilities (which I am not sure of), but it does pursue an interesting hybrid model via an in-house brokerage in addition to partnerships with external brokerages, which lowers its policy acquisition expenses.

Trade 1: Long Heritage Insurance (HRTG)

I think Heritage Insurance is attractive at roughly 6.5x consensus 2025 earnings. Frankly, I think if we make it through to November without a substantial hurricane, it may easily able beat the $0.47 Q3 earnings consensus (which I think has not been adjusted to account for no hurricanes occurring in the period). In an optimistic scenario, Heritage may earn as much as $4.50 this year and next, which would put it at a sub 6x multiple.

While Heritage does trade at a relatively rich 127% premium to Tangible Book Value, I think its strong earnings, geographical diversification, and 2026 growth potential justify a premium to book.

As mentioned on the recent earnings call, Heritage has significant growth opportunities in California and New York, two regions it pulled back from in 2022. Another insurer I follow, Kingstone, frequently pounds its chest on pricing improvements in the New York Region.

While you can’t blame them for celebrating their impressive comeback, Kingstone ironically attracts competitors like Heritage, who used to underwrite in the Long Island region through its Narragansett Bay subsidiary by showing how profitable the state has become. Heritage has the technical expertise and distribution to re-enter this heavily profitable geography, and given they explicitly mentioned they are opening up New York in 2025 and 2026, I think they will make quick progress to seize this opportunity.

Simply put, Heritage has engineered a huge turnaround and stands to gain significantly from lower reinsurance costs in 2026 and an improved 2025 loss ratio provided no hurricane occurs. It’s geographic diversification poises it to gain from improved Florida, California, and New York conditions, yet it trades at an extremely low industry multiple. The lower multiple is likely due in part over damaging claims made in a 60 Minutes Piece. Interestingly, the company has fully denied the allegations and filed a large lawsuit against the adjuster who made these claims in early 2025; either way, if Heritage ever does indeed need to pay a civil penalty, it is unlikely to be material to the company. It would be wrong for me to discuss Heritage without mentioning this risk, which is why I wanted to bring it up.

I think an attractive entry is within the $23.50 to $25.50 range. The stock trades wildly and moves up or down a dollar or two, despite the $750 Million market cap. If you recall, just a few weeks back we had the opportunity to get in a $17.50, and Heritage also traded up to $29.50 in late September before falling 6 dollars in the span of three days. It’s a volatile stock. Given the volatility, I think it’s prudent to consider a starter position at the upper end of the $23.50 to $25.50 range, and leave yourself firepower to get active at the lower end. If you prefer a more conservative entry, selling the December 20 strike puts for 90 cents, which would imply an entry point of $19.10, is also something toconsider. Absent a massive hurricane or terrible regulatory news, I don’t think Heritage trades that low this year and that strong Q3 and Q4 results may propel it to an ideal $30 within the next year.

Trade 2: Short Slide Insurance $12.50 November Puts (SLDE)

There was a recent short report on Slide Insurance, which has spiked the implied volatility of Slide’s options. Slide trades around 6.4x 2026 earnings at $15.99, which is standalone attractive, but I think an even better risk to reward setup is shorting the 12.50 November Puts for 54 cents. That would imply an entry of under $11.96 a share, and I think Slide is unlikely to return to the 12s given its recently announced $75 Million buyback, which should support the stock within the $12.50 to $14 range.

I’d take Heritage over Slide given the geographical diversification, but a sub 12 entry point on Slide is attractive and I am otherwise happy to collect the 52 cent premium if it stays where it is.

Trade 3: Long American Coastal (ACIC) within the $10 - $11 Range

American Coastal trades around 7x earnings and 8x 2026 EPS so it is a little less compelling than HRTG given the valuation and 100% Florida concentration, but I am attracted to its deep underwriting expertise and its capital light Managing General Agent unit, Skywater Underwriters. The MGA takes no risk and operates under a fee-based structure where it underwrites policies and transfers them to a third-party insurer, AmRisc, which is an affiliate of the Chairman of ACIC (who also owns more than $100 Million worth of stock).

The stock is down from its high of $14 last year. I think the $10 - $11 Range represents a compelling opportunity. American Coastal is an excellent underwriter and stands to gain significantly from lower 2026 reinsurance rates. Still, it’s extremely important to pick an attractive entry point for insurers and banks and I think the $10 - $11 Range (7-8x earnings) provides an “easier” entry point with more room for error, than the current $11.90 - $12.10 range. While this trade isn’t immediately actionable, small cap insurers move quickly, and if ACIC enters the $10 - $11 range, I will be adding to my position.

Reinsurance

An interesting way to play the Florida comeback is via reinsurers themselves. While Everest Group, Axis Capital, Arch, etc are all trending higher as the market realizes the prospects of a FL Hurricane are decreasing as we inch closer to the “finish line”, I am not too bullish on reinsurers long term due to the commodity like nature of reinsurance and indications of lower reinsurance pricing in 2026. While reinsurers themselves may experience a bumper profit in Q3 and Q4 of this year, softer reinsurance pricing in 2026 will serve as a headwind, and a tailwind to the primary Florida specialist insurers. Still, reinsurers are likely to beat Q3 and Q4 estimates and present a short term “earnings” setup as they “surprise”, but earnings trades are difficult to structure.

As we inch closer to the end of hurricane season, the entire FL Specialist sector stands to benefit from the first clean hurricane season in years. The setup is simple, if we can make if just four more weeks with no major hurricane activity, the entire sector is likely to blow through Q3 and Q4 2025 earnings estimates.

The group trades at a deep discount to the broader insurance sector and market as a whole, and I think a compelling opportunity is present in names like Heritage Insurance, and (at a lower valuation) American Coastal and Slide Insurance, to gain exposure to growing insurers at extremely low multiples. The broader FL Specialist sector has, for years, been written off as toxic and dangerous, but with litigation reform, lower reinsurance pricing, and stunningly low loss ratios driving profits to record highs, I think the sector is finally ready to let the good times roll.

Read more

1. Introduction

2. Trades

3. Sizing

Introduction

Don’t overthink it.

A lot of times, the best setups are simple. After all, the value of a stock at 4 PM is simply the price others are willing to trade with each other at. If your thesis is hard for others to understand it’s unlikely the market will see the same value you do.

I’m mixing things up for this week’s piece. Some of the best trades I’ve made this year weren’t predicated on a long winded thesis of “secret situation the market doesn’t understand”, but rather, “the stock just looks cheap”. Whether it was Heritage Insurance at four times earnings, or Brighthouse Financial at a 60-40% chance of being bought, or even perennial value trap OSB bumbling around at 5x earnings with a big buyback, the setups were simple.

This brings me to the land where cheap is, in fact, the norm: Chinese ADRs.

Despite the incredible run of Baidu and Alibaba, many of their smaller peers have entirely missed the recent rally. A few at 3-4x earnings with sizeable capital return programs are increasingly catching my eye, especially given their liquid options chains.

One of the largest risks when investing in Chinese ADRs is “zeroed” risk, that you wake up one day and discover your entire investment is gone, hence why the entire group tends to trade at a rock-bottom valuation.

Liquid options chains present an opportunity: the ability to dabble in these 3x - 4x names with strong capital allocation programs while hedging nearly all of your downside risk.

Many of these trades are still in the formative stages and I put on most yesterday, but I think the combination of rock-bottom valuations, strong capital allocation programs, aligned insider ownership, and options overlays that substantially eliminate downside risk present a few unique setups with 31% upside potential (if things go right) with just 11-12% downside risk. Without further huffing and puffing, let’s get into it.

Read more

1. Introduction

2. Trades

3. Final Thoughts

Introduction

I’ve written up some wild trades before. A crypto Closed End Fund trading with 100% upside to Net Asset Value, a Jordanian Insurer at 4x Earnings, a Georgian Financial Conglomerate with 70% upside to Net Asset Value repurchasing 15% of shares a year, and a Japanese Publisher with a concealed securities portfolio repurchasing 15% of the float every year.

I don’t seek out these setups just because they are crazy or niche, rather, I’ve learned some of the best risk-to-reward setups lie in names that are off limits to larger Institutional Investors.

As a Retail Investor you have two distinct edges, no investment mandate and a significant execution advantage. Unlike Institutional Investors who have risk, liquidity, and portfolio constraints, as a Retail Investor you can trade whatever you want, whenever you want. One day you can be an AI Expert and the next day you can specialize in Women’s Jeans. I’m not saying this is the best approach to take, but the freedom to trade anything is an advantage as it allows you to pursue the most favorable risk-to-reward setups.

Secondly, and more importantly, Retail Traders have a massive execution advantage. Kenneth Griffin and Douglas Cifu are your friends. Allow me to explain.

For Institutional Traders, the price and quantity they see on the exchange is the price and quantity they get. If a market is 200x15.05 to 100x15.10 and they want to buy 100 shares, 100 shares fill at $15.10 and then the ask evaporates, making it more difficult to buy shares (hence why Institutional Investors gravitate towards large caps with plentiful liquidity).

It’s an entirely different setup for Retail Traders.

A slew of market makers, Jane Street, Citadel Securities, Virtu Financial, gleefully pay your broker for the opportunity to fill your order at a better price at near unlimited quantity, because in general Retail Traders are wrong on a short-term timeframe. In the same market (200x15.05 to 100x15.10) where an Institutional Trader struggles to buy more than 100 shares, you might be able to buy 1,500 or even 3,600 at $15.0879. The market maker is happy to sell at $15.0879 because their model thinks the true price is $15.075, but you got a great deal and were able to execute significantly more efficiently than a larger a Institutional Investor.

What this means is if you cast a wide enough net and wait patiently for attractive entries, you can dance in and out of excellent risk-to-reward setups that are simply off-limits to Institutional Investors due to execution constraints.

I’m changing things up for this article. I’m focusing on four new names I’ve been buying starter positions in this week, and following the “early-stage” format of the 2023 IGI Pitch. Maybe they’ll go on to become larger positions, maybe the stories fall apart and they don’t.

All of them seem to be rapidly inflecting and have significant 2026 tailwinds, yet are priced as if everyone is skeptical, or more likely, unaware of the stories. Of course it’s possible the narratives fizzle out, but at the valuations they are trading at I think they are too compelling to ignore.

Read more