peshkov/iStock via Getty Images
Dear Fellow Investors,
2025 tested our patience, but not our conviction. The fund ¹ returned -2% in the quarter, bringing the full-year returns to -10% ² . While this 12-month result is frustrating, it follows gains of 51% in 2023 and 27% in 2024. More importantly, although the fund was down, the underlying businesses were not.
In any given year, a business can compound intrinsic value while its share price moves in the opposite direction. That divergence is uncomfortable, though not unusual, and it is rarely permanent. When business progress continues and valuation compresses, latent pressure builds and sets the stage for a snap back. Eventually, the multiple stops compressing and you get paid for being right on the business. I believe our returns have been delayed, not foregone. Historically, our returns have come in chunks or bursts; much like a coiled spring or a drawn crossbow, the conditions are in place for another burst.
The quarterly letter is typically organized around our top holdings and significant new positions. In this letter, I want to use a different structure. I believe there are two distinct “setups” that several of our holdings fall into, and I want to frame them accordingly.
The first setup, which I’ve dubbed “Bamboo Trees,” includes companies that made significant progress toward a radical transformation in 2025 with very little share price appreciation to show for it. The second setup includes companies whose share prices declined significantly in 2025 due to fears of changes that may never come to pass. These companies continued to grow and execute, but the market priced in negative outcomes that I believe are unlikely to materialize. For these businesses, I believe the other shoe is not dropping and I believe the risk/reward is favorable, as fear dissipates.
Bamboo Trees
The Chinese bamboo tree takes five years to grow. It has to be watered and fertilized every day, yet it does not break through the ground for five years. Then, in five weeks, it shoots up ninety feet. The question is: did it grow ninety feet in five weeks, or five years? — Les Brown
Lifecore (LFCR): As discussed in prior letters, Lifecore is a contract drug manufacturer operating at approximately 20% of capacity. Due to the Food and Drug Administration certification process, it typically takes two years to onboard a new customer even when a drug is already approved and in the market, creating a significant lag between customer wins and revenue recognition. As a result, substantial business progress can be made that does not appear in the financial statements for years.
The key question for Lifecore is whether it can sell its excess capacity at attractive prices. If utilization increases from 20% to 100%, revenue and margins should expand meaningfully due to operating leverage. This combination could lead to 4X+ share price appreciation over time if capacity is sold at attractive prices and cost discipline is maintained.
Based on public disclosures and our discussions, I believe that Lifecore will sell its capacity. Over the last three months, two multinational customers signed technology-transfer agreements under which Lifecore will manufacture existing drugs. Each of these customers could represent an incremental 5–10% of capacity. This is in addition to a GLP-1 customer win announced in November, as well as contractual minimum volume step-ups with Alcon (ALC) beginning in 2027, which should represent another 5–10% of capacity. Taken together, there is a clear line of sight to volumes doubling by the end of next year.
These wins are not simply a matter of luck. The company’s CEO, Paul Josephs, previously ran business development at Mylan and has invested meaningfully in building out Lifecore’s sales organization. The company serves the biologics fill-finish market, which is growing at approximately 10% per year, and Lifecore is a direct beneficiary of President Trump’s push to onshore pharmaceutical manufacturing in the United States.
An additional tailwind for Lifecore is the situation at Catalent (CTLT), a competitor that was acquired by Novo Holdings (NVO) in 2024 for $16.5B. Public records indicate a recent FDA inspection at Catalent’s Bloomington, Indiana facility identified severe deficiencies, including pest infestation and cat hair in a sterile environment. The inspection resulted in an Official Action Indicated (OAI) classification, the FDA’s most serious designation, which prevents facilities from receiving approval for new drugs until issues are resolved.
In contrast, Lifecore has a 40-year track record of safe operations, which is an asset that does not appear on its balance sheet but is highly valued as large companies navigate onshoring the manufacturing of their crown jewels.
Lifecore’s cost structure has also improved materially. Layoffs, better procurement, reduced use of consultants, and tighter expense controls have lowered the fixed-cost base. The company entered 2025 with approximately 15% EBITDA margins and has guided to roughly 23% for the upcoming quarter. With operating leverage inherent to the business structure, EBITDA margins at scale could exceed 30%.
Assuming Lifecore fills capacity longer term, the business should be capable of generating more than $100 million of EBITDA per year, making today’s roughly $500 million enterprise value appear modest in hindsight. In the nearer term, management is likely to raise both revenue and margin guidance.
Vistry (VTYPF): Vistry is a British real-estate developer transitioning out of traditional homebuilding to focus exclusively on partnerships with local governments for mixed-income developments. This partnership model requires less capital, carries less market risk, and is generally considered a superior business. As Vistry exits traditional homebuilding, significant capital should be freed up for share repurchases.
Vistry’s share price tumbled in 2024 as the company issued three profit warnings related to its legacy homebuilding business. While not ideal to have profit warnings, the fact that they are related to the portion of the business they are exiting is comforting. While the share price appreciated modestly in 2025, it will still have to more than double to recover to pre-profit warning levels.
Fortunately, the path to Vistry’s material share price increase is quite plausible. Company guidance implies margin improvement in 2026, and the British government is in the process of approving the largest housing stimulus program in the country’s history. Home volumes, margins, and revenues should be up, all while the company is actively buying back shares.
Vistray’s shares currently trade below book value while similar asset-light partnership businesses have traded for 5X+ book value. Likewise, Vistray shares are trading at sub-3X estimated medium-term EBIT while partnership businesses have previously traded at 12X+ EBIT. If execution continues, there is a credible path to shares re-rating on valuation metrics as the company becomes a pure-play partnership business. The progress towards the partnership model is tangible and the buyback happens every day. The share price chart for 2025 does NOT capture everything going on below the surface, change is happening.
The Other Shoe Is Not Dropping
Markets tend to overprice downside risk when uncertainty is high and narratives are easy to simplify. Potential negatives can be treated as inevitable outcomes, while steady execution and base rates receive far less attention. As a result, valuation multiples compress not because fundamentals deteriorate, but because investors price in low-probability adverse scenarios as if they were central cases. When those feared outcomes fail to materialize, the subsequent repricing can be swift and meaningful.
I believe this dynamic applies to several holdings, including PAR (PAR), Cellebrite (CLBT), KKR (KKR), and Burford. In each case, I believe negative sentiment drove multiple compression despite continued business execution. I believe that bad events that will not ultimately come to pass are being priced in. As it becomes clearer the fears will not be realized, the shares can rerate.
In the sections that follow, I will outline what the market is worried about for each company, as well as why I believe those fears are misplaced and why the other shoe may not drop.
Burford (BUR): We own shares of Burford primarily for the quality of their ongoing litigation finance business and high returns on capital deployed. As a bonus, Burford has a judgment against Argentina that could be very significant relative to market capitalization. As a refresher, this asset is Burford’s share of a judgment against Argentina for taking the shares of the oil company YPF (YPF) from shareholders with zero compensation.
The YPF case has been ongoing for more than a decade, during which Argentina has paid millions of dollars to delay compensating shareholders for expropriating the YPF shares. At a hearing in October 2025, Argentina’s counsel argued that the judgment should be vacated and the case retried in Argentina. During a line of questioning by Judge Denny Chin, investor concern grew that the award could be overturned and Burford’s shares declined.
In my view, this reaction overstates the risk. In the U.S. Second Circuit, cases are overturned roughly 6% of the time, and the presiding judge in the YPF case has an overturn rate closer to 5%. Judge Chin himself has previously ruled against Argentina in materially similar circumstances, as reflected in his 2018 opinion rejecting Argentina’s arguments around the YPF bylaws.
Based on these base rates and judicial precedent, I believe it is highly likely that the YPF ruling will be upheld in 2026. The market appears to be pricing in a far more adverse outcome than history would suggest.
Burford shares were down 30% in 2025, ending the year at approximately 10X cash flow (excluding YPF) for a growing and scalable business that will not require any additional equity capital to grow. This was a broken stock in 2025, but not a broken business.
PAR: PAR and its vertical market software peers remain deeply out of favor. PAR’s 2025 share price trajectory can only be described as brutal, erasing all gains from 2024 despite organic ARR revenue growth of 21% in 2024 and 15% (E) in 2025, with a projected modest reacceleration in 2026 without winning a mega Tier 1.
In the last letter, I included a long appendix covering PAR ( link ). As a reminder, PAR was engaged in four RFPs with Tier-1 restaurant chains believed to be McDonald’s (MCD), Tim Hortons (QSR), Chipotle (CMG), and Papa Johns (PZZA).
Thus far, PAR lost Chipotle (which chose to remain with its current vendor) and won the smallest of the opportunities, Papa Johns. The Papa Johns win covers approximately 3,200 locations and represents roughly $14 million in ARR.
The Papa Johns win marks PAR’s entry into the pizza vertical, which requires some incremental software development. As a result, it is not expected to be a meaningful growth driver in 2026, as implementations are likely to begin in Q4 2026, but it should contribute meaningfully in 2027.
Papa Johns’ CTO, Kevin Vasconi, is highly regarded within the restaurant industry, having overseen the buildout of Domino’s (DPZ) technology and e-commerce platform, which was widely viewed as industry-leading. His decision not to rebuild Papa Johns’ technology in-house, but to instead partner with PAR, is a validation of PAR’s technology and a data point suggesting that “vibe coding” (AI generated software) may not displace PAR’s value proposition.
Combined with PAR’s successful implementation at Burger King, Papa Johns’ selection materially de-risks PAR’s positioning as an “upstart” which should bode well for future enterprise wins.
With respect to McDonald’s, PAR has been a hardware partner for over 40 years and is currently a software partner in Asia for loyalty. Inertia is a powerful force at McDonald’s scale, and PAR will need to continue de-risking the decision by demonstrating its ability to deliver a McDonald’s-specific solution capable of operating across more than 40,000 locations in over 100 countries. It appears McDonald’s is now a two-horse race between PAR and the McDonald’s internally developed solution, which has enormous technical debt. A decision is likely in 2026. A PAR win could provide a Tier 1 customer with significant incremental functionality and cost savings.
Tim Hortons currently uses a combination of legacy software and solutions from a PAR competitor. Tim Hortons is owned by Restaurant Brands, which also owns Burger King. All indications are that PAR’s implementation at Burger King has been successful, with Burger King adopting additional software modules since the initial rollout. A decision is likely in Q2 2026.
It is difficult to handicap the pipeline from the outside, as PAR does not publicly confirm the identities of its Tier 1 prospects or the particulars of any RFP process. That uncertainty has contributed to investor skepticism. My best estimate, based on scuttlebutt, building a mosaic from scraps of information, and industry knowledge, is as follows:
Winning either of the remaining mega-deals would be impactful for both growth and earnings. Growth is not dead at PAR. Large restaurant chains such as Papa Johns, McDonald’s, and Tim Hortons—companies that historically built and maintained their own software—are actively seeking to exit the software business and refocus on restaurant operations. This behavior runs counter to prevailing “vibe coding” fears and supports PAR’s long-term positioning.
While the fears of AI are palpable and have pushed PAR to its lowest valuation in years, it is also worth noting the business is the healthiest it has been in years. PAR’s win rate in 2025 was the highest in company history. PAR’s pipeline continues to expand. In addition to the Tier 1 opportunities listed above, PAR received 2 additional Tier 1 RFPs this month. PAR increased its attachment rate of additional products in 2025 to the highest level in company history.
On a recent call with investors, the PAR CEO said, “AI has been upside for us, not downside,” because PAR’s customers are eager to buy AI capabilities built into PAR, as there are fewer complexities, better integrations, and fewer security issues. PAR currently has two AI products in market.
Currently, the market has placed virtually all software companies in the AI loser bucket and pushed prices and multiples down with little to no differentiation between those at great risk and those not. One of the most tech-savvy CTOs in restaurant technology just scrapped his company’s proprietary system for PAR; McDonald’s, with all the resources in the world, is considering doing the same. Two more Tier 1 RFPs just came in this month. If the other shoe does not drop and PAR continues to add Tier 1 customers and cross-sell, there is a path to the shares trading substantially higher.
Cellebrite: Shares were down more than 16% in 2025 while annual recurring revenue (ARR) was up 19% (through Q3) and free cash flow grew 30%. The “shoe” that most investors are fearing is the potential for AI and “vibe coding” to lower the barriers to software creation. This fear led to valuation multiples compressing across all software in 2025 and into 2026.
I believe that Cellebrite is far likelier to be a beneficiary of AI than a victim. Anthropic is leading the vibe coding charge with Claude Code, which has changed how software is written. Ironically, when I prompted Anthropic’s AI tool Claude about the feasibility of replicating Cellebrite, it responded with “iOS security isn’t something you casually bypass. Cellebrite spends enormous resources acquiring zero-day exploits, reverse-engineering the Secure Enclave, and staying ahead of Apple (AAPL)’s patches. This isn’t a weekend project—it’s nation-state-level security research.”
Building hacking tools is also against the terms of use for most vibe coding offerings, but more importantly, the customer base for Cellebrite consists primarily of government agencies for whom security is far more important than cost.
For security reasons alone, a police department, the FBI, or the CIA is unlikely to replace Cellebrite with an unproven alternative. Many of these agencies were slow to adopt cloud computing and remain highly risk-averse when it comes to mission-critical technology.
In addition, Cellebrite operates in multi-vendor environments where agencies often rely on two or three tools simultaneously. At any given moment, one vendor may be unable to extract data from the latest operating system or bypass a new security patch. This is not a winner-take-all environment and even if new vibe coded alternatives spring up, the scaled industry leader with dominant market share likely retains their seat at the table.
Cellebrite has all the tools in place, from data collection to the management of the data. More importantly, it has the largest customer base to build tools for within the investigative vertical, taking advantage of AI-driven productivity increases while protecting sensitive data, complying with search warrants, and documenting evidence chain of custody. Some tools have been launched this month and we will learn more about the functionality and business case as the year progresses.
While the company has not yet given formal guidance for 2026, management indicated with confidence that growth would reaccelerate in 2026 as the Federal Government shutdown passes, meaning ARR should grow another 20%+ in 2026 with a long runway ahead in 2027 and 2028.
Although the share price may remain volatile, I believe the probability that AI meaningfully disrupts Cellebrite’s business in the medium term appears quite low. If vibe coding fears are realized, the multiples can always contract – but Cellebrite shares trade for a bit over 20X this year’s free cash flow for a business with recurring revenue and earnings that should grow approximately 20%. I believe that we should get paid going forward.
KKR: Similar to Cellebrite, KKR saw a share price decline of approximately 15% despite growing AUM 16%, fee-related earnings 16%, and adjusted net income 17%. ³ The two “shoes” that investors are fearing with KKR and the other alternative asset managers appear to be LP/investor fatigue that will lead to fundraising issues and potential issues within the private credit funds.
With regard to private credit issues, KKR’s Co-CEO Scott Nuttall addressed these concerns on their most recent conference call, saying, “From everything we are seeing, there is nothing alarming going on. Just the beginning of a return to a more normal default environment…our view is that forward credit fundamentals, both liquid and private, will remain attractive. Our clients feel the same way, which is why we are having a record credit fundraising year.”
On the overall fundraising front, Mr. Nuttall discussed the topic at the Goldman Sachs (GS) conference in December, stating, “We keep reading all these headlines about how it’s really hard to raise money, and we’re having a record fundraising year…What we’re reading about is not consistent with our actual results and our actual experience.” ⁴If the fundraising shoe is about to drop, management is doing a great job of denying it.
KKR should raise close to $20B from the wealth channel ((the company’s K-series product)) in 2026 and their Flagship North American Private Equity fund should raise another $20B+ in 2026. These developments make it appear unlikely that AUM growth will grind to a halt at KKR any time soon.
Hagerty (HGTY) Firing on All Cylinders
Four of our top five holdings (PAR, Cellebrite, KKR, and Lifecore) have been covered in the “Bamboo Trees” or “Other Shoe not Dropping” sections. Hagerty is also a top five holding, and I want to provide an update on the business, as I want you to understand what we own and why we own it.
Hagerty is one of our holdings that actually appreciated in 2025 (+36%) and ended the year as our second largest holding. Hagerty is a business that is firing on all cylinders.
State Farm: After years of delays, Hagerty is finally onboarding State Farm policies, turning the State Farm opportunity from one that required investments in people and systems of tens of millions of dollars for the period from 2021-2025 to one generating revenue. Over the next couple of years Hagerty should onboard over 1 million State Farm policies, which is quite material relative to their current 1.6M policies in force.
Operating Leverage: Over the last four years, Hagerty has driven EBIT margins from -8% in 2022 to +12% (E) in 2025. Going forward, with the addition of State Farm policies, savings coming from consolidating software platforms, and fiscal discipline, Hagerty should see significant operating leverage. As revenues grow faster than expenses, EBIT margins should march steadily higher to 20%+ in a couple of years.
Marketplace Growth: Hagerty’s primary business is car insurance, sold either directly to owners or through partnerships with major agencies or insurers such as State Farm and Progressive (PGR). The company has customer retention rates of approximately 90% and a net promoter score of 82, compared with an industry average below 40.
Customers are typically lost only when a car is sold. The time of sale is therefore the critical moment to retain a customer or keep a vehicle insured under Hagerty’s platform. Controlling the marketplace and being present at the time of sale materially increases the likelihood that Hagerty will retain the insurance relationship.
The marketplace business would be attractive solely for customer retention purposes, but getting a buyer and seller fee to simply facilitate the transaction is also an excellent business.
Hagerty entered the marketplace business in 2022 with the idea of holding live auctions in conjunction with their major car shows and building an online alternative to BringATrailer.com. These businesses gained considerable traction in 2025, with live auction hammer value growing 56.6% to $228.7M, private sales transactions growing 323.8% to $267M, and online marketplace hammer value growing 83.1% to $39.4M.
In the short term, Hagerty should benefit from 2025 being the first year in a decade where no major hurricanes made landfall in the US. Hagerty follows GAAP rules and industry standards to accrue for estimated losses throughout the year. If actual losses come in higher or lower than originally projected at the beginning of the year, a “true-up” occurs in the Q4 earnings. Given the absence of major catastrophe events this year, Hagerty will likely “true-up” the accrued charges in Q4, reducing the overall loss expense and producing full year earnings in excess of 40 cents versus analyst estimates in the mid 30s.
In 2026 and 2027, Hagerty should benefit from continued State Farm onboarding and organic growth such that total revenue growth, combined with operating leverage driving margin improvements, results in earnings growth of 25%+ per year.
Short Side
We ended the quarter short two companies facing significant litigation with the potential for treble damages ((i.e., 3X the actual amount)) for their actions and potential liabilities far in excess of their market capitalizations. We are also short two major indices, a consumer products company that has a significantly above-market multiple while growth has stalled, and a mature and expensive food-related business.
We also had a position via puts and calls on the 20-year bond that, in a scenario where Treasury rates rose significantly (e.g., government shutdown, tariff repeal), would have offset some of the multiple compression we would have seen across our portfolio. This “insurance” was not needed and the options have expired.
Outlook
The world feels off-kilter. As discussed above in the letter, we own several companies that grew attractively and have bright prospects yet had their share prices decline. At the same time, a number of AI, power, or Quantum computing companies that are pre-revenue in “fake it until you make it” mode have seen their share prices go vertical on the backs of press releases and unfunded projects.
I think 2026 is a year where we will see real productivity gains from AI implementation and we will see it in our portfolio companies on the expense side. The equity markets should also benefit from a declining rate environment.
These tailwinds are being confronted by an unstable geopolitical landscape, as American foreign policy shifts and the world deals with the “Donroe Doctrine” and the unknown effects that may ultimately manifest as a result.
I feel less certain about the world at large than the particulars of our portfolio. I believe AUM will go up at KKR, Lifecore will sell excess capacity, Burford’s YPF ruling will not be overturned, PAR will close on another Tier-1, Hagerty will onboard State Farm policies, and Cellebrite will not be replaced by vibe coders.
Waiting for the bamboo shoots to show or waiting for the other shoe NOT to drop is painful and it certainly lacks the dopamine hit of consuming TikTok – but ultimately, I believe that we should be rewarded.
Sincerely,
Scott
References
- Greenhaven Road Capital Fund 1, LP, Greenhaven Road Capital Fund 1 Offshore, Ltd., and Greenhaven Road Capital Fund 2, LP are referred to collectively herein as the “Fund” or the “Partnership.”
- See end notes for a description of this net performance.
- Figures are trailing 12 months through Q3
- https://www.bloomberg.com/news/articles/2026-01-20/kkr-on-track-to-beat-20-billion-target-for-americas-buyout-fund?embedded-checkout=true
Editor’s Note: The summary bullets for this article were chosen by Seeking Alpha editors.
