Dear Partners,

For the month of September, Caravel returned -2.40% compared to -1.48% for the benchmark (-0.35% for the S&P 500 & -2.62% for the SPTSX). This brings year-to-date total net return to +8.15% for the fund and +13.04% for the benchmark, respectively.

It has not been the start we'd hoped for to begin Caravel's second decade in business.  Losses in September  were concentrated within our equity long-short strategy (-3.0% gross), partially offset by gains in our merger arbitrage and credit investments. (+0.2% & +0.6% gross, respectively).

We try to be blunt in our communications during drawdowns, which we endeavour to circumvent entirely, but inevitably experience from time to time. We avoid excuses like the plague and hope to provide as much transparency as possible into our process and positioning. People usually pay much closer attention to red numbers – we get it. We have the same bias with regard to our investments.

The views we expressed in our June 2026 Letter regarding the forward global interest rate path, while carefully considered, have so far proved to be wrong. Policymakers have been less willing to look through the supply-side shocks and narrow economic growth drivers (geopolitical conflicts and AI CAPEX, respectively) at play than we had initially expected. As such, the market has flipped from pricing in about 0.75% of cuts to short-term interest rates in the US before the Iran war, to over 1.00% of hikes currently, including the 0.25% rate increase enacted last month.

Implied Fed Path 02/20/2026

Implied Fed Path 08/20/2026

You might be thinking, 'how much of a difference can a couple percentage points really make?' We believe examining the current fiscal situation of the US Government can help illustrate why this change has had such a profound impact on many subcategories of financial assets.

Floating Uncle Sam

The US Government's National Debt stood at USD $10 Trillion in September 2008, the month that Lehman Brothers went bankrupt and the impact of the Global Financial Crisis became inescapable. That number doubled over the following nine years to $20 Trillion in September 2017. Then it doubled again over the next nine years, and as of September 2026 sat at an eye-watering $40 Trillion. This trajectory is good for an 8% compound annual growth rate (CAGR). By contrast, the US economy as measured by GDP has only doubled once over the same period, going from about $15 Trillion in 2008 to just over $30 Trillion today. As strong as the US economy has been since recovering from the GFC, its debt has grown nearly twice as fast, and overall debt to GDP has gone from less than 70% to nearly 130% over this period.

GDP is commonly referenced in relation to overall debt, but it's important to remember that GDP measures, however imperfectly, the total size of a given economy. Government revenues are a fraction of this number. Historically, US federal revenues have been in a remarkably consistent range, between 15% and 20% of GDP, with the long-run average around 17%. This phenomenon is known as Hauser's Law, and it has been true across a wide variety of tax regimes.

Now let's look at how the cost of this debt has changed, using the 10-year treasury yield as a proxy:

As an exercise, let's assume the US government financed all its debt at the 10-year duration (reality is more complicated, but short-term and long-term interest rates generally move in the same direction over time). Let's also generously assume it can extract 20% of GDP annually in taxes.

Year	2008	2017	2026
Total Debt Outstanding	$10 Trillion	$20 Trillion	$40 Trillion
Cost of Debt (%)	4.0%	2.5%	5.0%
Cost of Debt ($)	$400 Billion	$500 Billion	$2 Trillion
GDP	$15 Trillion	$20 Trillion	$32 Trillion
Government Revenue	$3.0 Trillion	$4.0 Trillion	$6.4 Trillion
Interest as a % of Revenue	13%	13%	31%

So What?

We believe that global governments were given a false sense of security over the period from 2008-2021, over which time interest rates were in structural decline. This was due in large part to a massive, coordinated effort by the US Federal Reserve and other Central Banks to suppress them. The new Fed Chairman Kevin Warsh has repeatedly criticized the institution he now leads for its policy errors over this period, and we think his aims are noble. He wishes to use the tools available to him to bring inflation, which has exceeded the 2% target for five consecutive years, back in line.

The problem with this, as we see it, is twofold.

1. From a leverage perspective, the cat is out of the bag.

Governments, businesses, and consumers were too enticed by historically low interest rates and are now so over-levered, that small changes in the cost of debt can cause severe strain. To our Canadian readers, how many of you know someone under the age of 50 that is being squeezed, or is about to be, by the rate on their 5-year variable mortgage rate resetting?  This dynamic is present everywhere.

2. Inflation is not just a demand story.

The Fed raises interest rates when it thinks the economy's demand is outpacing its productive capacity. Certainly, demand and growth have been strong in the US, especially relative to the rest of the world, largely due to the feverish buildout of A.I. infrastructure. But inflation is also being driven by supply-side factors, including multiple wars, demographics, supply-chain reshoring, resource nationalism and export controls, etc. 

We believe an overly restrictive monetary regime risks catastrophic damage to the global economy, given the amount of debt the system is carrying combined with the components of inflation that appear more structural than cyclical. This belief underlayed our view that the Fed was more likely to let inflation run hot during this period of (eventually) peak A.I. demand and (hopefully temporary) geopolitical turmoil, than it was to risk a recession. At 130% Debt to GDP, the US needs the economy to grow faster than debt to get the fiscal situation under control. We believe inflation is undoubtedly part of this equation. A shrinking US economy would only make the situation worse, as the debt load isn't getting any smaller in nominal terms.

President Trump and Secretary Bessent seem to understand this risk quite well, as evidenced by their repeated overtures, threats, and policy actions aimed at lowering interest rates.

If the entire US national debt was refinanced at 5%, the government would pay interest of $2 Trillion a year. For fiscal 2026, which ended on September 30^th, that number was just over $1 Trillion, owing to all the legacy debt that had been issued at lower interest rates in the past. Trump and Bessent desperately want to dampen the blow of this refinancing, which is why we view the Iran War as a particularly tragic error in terms of its execution, if not its intent – recall the market was pricing in interest rate cuts before March, and the 10-year yield was over 100 basis points lower than it is today.

At $2 Trillion, US annual interest costs would amount to:

• ~120% of 2026 Social Security spending.
• ~200% of 2026 Medicare spending.
• ~215% of 2026 Defense spending.
• ~275% of 2026 Medicaid spending.
• ~37% of all Fiscal 2026 government revenues.

The only way out is through...

The US is hardly alone in its transgressions. As dire as this American situation might appear, it is worse for much of the developed world, which does not enjoy reserve currency status. Our belief remains that the primary fiscal objective of global governments should be to grow their economies faster than their debts, such that they may bring their debt levels under control in the future and hopefully enjoy manageable costs of borrowing as a result. The cost of this aim, in our view, is inflation running closer to 3-4% annually than the 2% maximum espoused by the Fed.

Old man yelling at cloud?

I give this lengthy preamble not to excuse our performance, but to explain our thinking. Our thesis that the world is in the early innings of a commodity boom has not been much shaken by the recent about-face in the path of interest rates, but we have been caught off-guard by the magnitude of the reversal in the stocks within some of the core sectors we have been focused to express this view. 

Commodity	Gold	Copper	Uranium
Price	$4,050/oz	$6.40/lb	$90/lb
% Change since 08/31/26	-9%	-3%	0%
Associated ETF	GDXJ	COPX	URNJ
ETF % Change since 08/31/26	-20%	-20%	-35%

Of course, we expect the share prices of mining equities, particularly those in which we typically traffic (<$10 Billion enterprise value), to move more in percentage terms than the prices of the underlying commodities. As such, we cannot say we are overly surprised by the performance of the junior gold miners index (GDXJ) over the past six weeks. We entered September with 1% net weight in gold stocks, and were short the commodity and the GDXJ ETF to hedge our exposure, given our cautious view on rates in the short term. Unfortunately, the average stock on the long side of the ledger is down 40% in the six weeks since the end of August. We did not anticipate moves of this magnitude against the backdrop of a 9% correction in the gold price, which at over $4,000 an ounce still represents a highly economic level for the stocks we own. Our portfolio weighting has resultingly swung from net long to net short (approximately -1.0% net short at present), or about a 2% loss.

Copper and Uranium have been more perplexing. Despite the commodity prices holding in much better than their precious metals counterparts, these stocks have not been spared. The unwind in uranium stocks has been especially frustrating given that the commodity price is actually up marginally since the end of August. Our losses in these areas have been more significant given our greater net long exposure. While we don't think they are permanently impaired, we are triaging rather than doubling down here to protect our capital.

We believe the pressure on small-cap stocks, both within and outside of natural resources, has had more to do with fund flows than fundamentals. Large pools of capital view small-cap stocks as undesirable during a fed hiking cycle, and so have rotated out of the sector. The tailwinds that propelled many of these stocks to 2x-3x gains or better since 2024 appear to have become headwinds in the short-term. We have tried to mitigate the impact of these moves, but to this point have not done enough. For the reasons we have laid out above, we believe there are practical, structural, and immutable constraints on how high interest rates can go without forcing another intervention, similar to QE 1.0 in 2008. Accordingly, we are trying to balance the long-term opportunities we see in these equities, given their recent weakness, with the short-term risks presented by fund flows and other macro factors. While still in the trade, we have reduced our net exposure to metals stocks (<10% of net assets) across gold, copper, and uranium.

As regular readers know, we love our sports analogies here at Caravel. Right now, it feels like we are facing a pitcher throwing 110 miles an hour. We are just trying to foul the ball back and wait for our next fat pitch to arrive. We have ~40% of net assets allocated to merger arbitrage, ~25% allocated to fixed income and converts, and ~10% in cash. When we feel the time is right, we expect to further increase our net exposure to our preferred stocks, which accounted for over half of our net gains of ~71% from April 2024 to August 2026.

We thank you for your continued support,

Jack and Glen

Managing Partners

Caravel Capital