Average Annual Total Returns (%)
Average Annual Total Returns (%)

Dear Shareholder:

Performance Overview

The FPA Crescent Fund – Institutional Class (“Fund” or “Crescent”) gained 3.09% for the quarter and 17.65% in 2025.

Performance versus Illustrative Indices (%)1
Q3 2025Trailing 12-month
FPA Crescent3.0917.65
FPA Crescent – Long Equity4.3024.97
MSCI ACWI3.2922.34
S&P 5002.6617.88
60% MSCI ACWI / 40% Bloomberg U.S. Agg2.4216.22
60% S&P 500 / 40% Bloomberg U.S. Agg2.0313.70
Portfolio & Market Discussion

Crescent’s top five performers contributed 8.13% to its trailing twelve-month return while its bottom five detracted 2.39%.

Trailing Twelve-Month Top and Bottom Contributors (%) as of December 31, 20252
Top ContributorsPerformance ContributionPercent of PortfolioBottom ContributorsPerformance ContributionPercent of Portfolio
Alphabet3.195.0CarMax-0.640.9
Citigroup1.452.4Charter Communications-0.631.2
TE Connectivity1.432.5Comcast-0.462.0
JDE Peet’s1.201.5Int’l Flavors & Fragrances-0.361.8
Safran0.871.7Jeffreries-0.311.2
8.1313.1-2.397.1

If you spend time in our office, you will inevitably overhear our FPA colleague Brian Selmo comment that “Securities ask different questions at different prices.”  For securities trading at elevated multiples, the questions are likely to be difficult, if not impossible to answer, and possibly even multiplicative in nature.  In other instances, for out of favor names, one is simply determining if the future could be better than the present, with conditions potentially turning positive as the cherry on top.

By focusing on the less challenging questions, we largely avoid the speculative areas of the market where we deem the reward for taking risks insufficient relative to the potential returns.  This does not imply that we completely avoid mistakes, but we believe it helps limit the damage when we do err in judgement. Over the last 30 years, the end-product of this process has been a strategy that has generated equity-like returns while placing equal importance on both capital preservation and capital appreciation.  Ultimately, this approach requires an unwavering commitment to being “value aware,” a philosophy that underpins our research process. 

In doing so, we consider both quantitative and qualitative analysis as we underwrite potential new investments and monitor existing positions.  It is not enough to be a great company, nor is it sufficient to simply trade on a superficially low p/e multiple to garner our interest and attention.  Instead, we are looking for rare cases where both quality and “value” intersect, even if near-term headwinds may delay the potential payoff.

Hence, when it comes to deploying capital into new names, it is rare that you will find our recent purchases on the 52-week high list.  Instead, we thumb our noses at the momentum factor and regularly troll the 52-week low list for potential opportunities.  It is in that vein that we let value guide our focus.  So, before you pose questions on our upcoming conference call, we don’t have any unique insight into American exceptionalism or whether we are in an AI bubble.  Well, you can still ask, but we don’t think we know the answer.

However, we do believe that the investment community is currently casting its gaze away from various market constituents that offer asymmetric risk-reward for those willing to look forward three to five years.  We primarily speak of small- to mid-cap global securities, and, up until recently, healthcare stocks.  As you can see from the image below, new purchases over the past two years are clear evidence that we are willing to back up this view with our actions and not just our words.    

Venn Diagram - Purchases over the last two years

While few of the above are household names, we believe each holds a strong competitive position in their respective industry, if not market leadership.  Furthermore, we would go so far as to say we believe we purchased them at absolute values that we believe will help us achieve equity-like returns over the coming years, and at bargain-like valuations compared to the S&P 500. 

As for the why, we know not the answer, but we have our suspicions.  With the recent outperformance of large US technology companies, not to mention the adoption of passive investing, there may be a shrinking pool of active investors with the interest, resources, and capital to conduct in-depth research on 1) smid-cap names, and 2) ideas that necessitate deferred gratification longer than one to two quarters. And as it relates to the active investors hoovering up assets, in our experience it is rare to find a pod shop on the share registers of, say, Azelis, Nippon Paint, Sodexo, or NOV.

Lastly, there’s a good reason why the commentary above does not necessarily jive with all of our holdings, say such as our top four holdings by weight Alphabet (purchased 2010), Meta purchased (2018), Analog Devices (purchased 2011), and TE Connectivity (purchased 2012). We use these names to remind you that our average holding period is in excess of five years, and for those in the top ten, demonstrably longer.  We purchased each of the aforementioned “GARPY” names initially when they were making headlines for the wrong reasons, and we would argue their present standing does not betray our ethos as price-disciplined investors, but rather is indicative of our research process and strategy working as intended… sometimes the high case does actually play out!3

Now, as an example of a tough question currently being posed by the market, we submit Microsoft, a company we actually owned from 2010 to 2020.  To begin, we hold management in the highest regards, particularly the CEO and CFO, who created tremendous value for us during our ownership.  When we originally purchased Microsoft back in 2010, the valuation was less than 10x after-tax earnings after subtracting the cash on the balance sheet.  The question we had to ask at the time was whether Microsoft was a melting ice cube, or could the company deliver an equity-type rate of return even if earnings never grew again.  To our regret, we sold Microsoft too early, failing to appreciate how technological developments, particularly those of AI/cloud, would transform the business.

In the five years following our sale, from 2020 through 2025, Microsoft grew its top line at a 12% compound annual growth rate (“CAGR”), and its bottom line even faster!  But let’s now look at what one must believe to be constructive on the return prospects at the current multiple of 30x+ earnings. Of course, 30x earnings on its own is not necessarily the wrong multiple if Microsoft can continue to grow at a double-digit rate over the coming five years, and market analysts have noted that the valuation looks inexpensive five years out even if revenue were only to grow at a more pedestrian CAGR of 10% during this period.

Stepping back, we will note that Microsoft generated $293bn of revenue over the last twelve months. If revenue were indeed to grow at 10% CAGR over the next five years, Microsoft would theoretically generate $472bn in revenue in 2030, an increase of $178bn.  The bulls will say this is achievable based on growth in the core business and an expanding total available market (“TAM”).  However, to put the $178bn increase into perspective, Microsoft would need to add revenue greater than the aggregate sales of ERP leaders Oracle and SAP, who generated ~$100bn in combined sales over the past twelve months.  The bulls may say that our thinking is too narrow; vibe coding will also allow Microsoft to serve as the foundation or platform upon which corporations will create low-cost alternative solutions for CRM solutions (Salesforce: $40bn revenue), HR software (Workday: $9bn revenue), and new database products (Snowflake: $4.4bn revenue).  Moreover, as the bulls believe, from a horizontal perspective, agentic AI will enable Microsoft to offer services to automate and manage digital workflows (ServiceNow: $12bn revenue) across industries.  If that was not enough, the bulls will remind us that Microsoft has a consumer gaming franchise in Xbox that will grow over time (Nintendo: $12bn revenue).

To put the words into a picture, for Microsoft to grow at 10% over the next five years, one must believe the company will generate incremental revenue currently generated by the combination of:4

Expected Revenue of MSFT in 2030

We are not saying this can’t happen, but thankfully as benchmark agnostic investors, there is nothing forcing us to make the wager that it will. Speaking of wagers, we are regularly asked whether the higher-than-average multiples, particularly in US markets, make us inclined to believe a near-term market wobble lies around the corner.  To that we emphatically answer “Yes” from our square offices, but with the caveat that we don’t know from which corner.  And so rather than make top-down bets, we simply go about our business conducting bottoms-up fundamental analysis with the hope that it helps us find a handful of good ideas each year and keeps us away from the bad ones.

Speaking of bad, realized capital gains during 2025 were higher than we prefer, which resulted in a modest capital gain distribution in relation to the fund’s total return in December.  The good news is that 100% of it was in the form of “long-term” capital gains, which are taxed at lower rates than “short-term” gains, and that we remain in the top 2% of our Morningstar category for after-tax returns over the past 5, 10 and 15 years.

With stand-alone ETFs (like the Fund’s sister ETF, FPA Global Equity – ticker: FPAG) typically making only de minimis capital gain distributions and ETF share classes on the near to mid-term horizon, we know the bar is higher now than even five years ago. So, in late Q4 we onboarded an additional portfolio rebalancing tool that can have a positive side effect of helping us meet our goals to reduce capital gain distributions.

In 2025, we were able to dispose of assets with ~$182 million of gain without making corresponding capital gain distributions using our legacy tools we onboarded in 2021. In just November and December alone, we were able to dispose of assets with over ~$315 million of gain without making corresponding capital gain distributions as a byproduct of our new tool. As they say: so far, so good. We think our current set of tools coupled with our generally low-turnover approach should support our goal of low-single digit capital gain distributions attributable to securities that could have been disposed of through in-kind redemptions. 

With that said, it is important to note that even if the Fund converted to a stand-alone ETF tomorrow, it would still likely pay out capital gain distributions because Crescent often owns securities that cannot be disposed of through in-kind redemptions. These include private securities (like our recent shipping investments that have been a strong contributor over the past five years), derivatives (like our interest rate caps that performed well in 2022), high-yield bonds, some foreign equities, and the list goes on. To help shareholders get a better sense of how we are doing on our above goal, our future capital gain estimates will start to note what percentage of capital gains are from securities that can be disposed of through in-kind redemptions.

Closing

We appreciate the long-term partnership we’ve had with so many of our like-minded investors who maintain a similar “value aware” point of view and long holding period. To the old and the new, we seek to continually earn your trust through the inevitable market cycles for the capital you have entrusted to us.

Respectfully submitted,

FPA Crescent Portfolio Managers
January 29, 2025

FPA Crescent Fund Portfolio Highlights