Mad Hedge Biotech and Healthcare Letter
March 27, 2025
Fiat Lux
Featured Trade:
(NO SHERPA REQUIRED)
(MRK), (BMY)
Perched high above the timberline on Colorado's Mt. Elbert last weekend, I found myself short on oxygen and long on questions—namely, which pharmaceutical heavyweight deserves a spot in my portfolio: Merck (MRK) or Bristol-Myers Squibb (BMY)?
At 14,438 feet, the air thins out fast, but the thinking gets clearer. Clarity tends to arrive when your brain’s running at 60% capacity.
I’d stuffed my pack with company reports, earnings transcripts, and a few too many granola bars—one of which was being stalked by a very persistent marmot as I paused to catch my breath. I must’ve looked like an underprepared Everest hopeful, hunched over charts and trying to find altitude-adjusted alpha.
On paper, both firms dominate the oncology space and have made a career out of telling cancer where to shove it. But markets don’t care about reputations—they care about margins, pipelines, and who's going to make it through the next patent cliff without blowing out their kneecaps.
Let’s start with the money.
Merck posted Q4 2024 revenue of $17.76 billion, up 6.77% year-on-year. Its price-to-sales ratio sits at 3.74x—above the sector median, but still 14.7% cheaper than its own five-year average. It’s also beaten revenue expectations for 12 straight quarters. That’s not a hot streak. That’s clinical precision.
Bristol-Myers pulled in $12.34 billion last quarter with 7.5% YoY growth, but it trades at a much lower 2.51x P/S. That’s a discount—16.5% under the sector median. Ten out of twelve quarters beating the Street is nothing to sneeze at either. You get the sense both firms have their accounting departments on creatine.
Debt? Merck sits on $24.6 billion in net debt, but with a net debt/EBITDA ratio of 0.84x, it's practically sipping debt through a paper straw. Bristol-Myers, on the other hand, carries $40.1 billion with a 2.07x ratio. Still manageable, but not the kind of leverage that makes you sleep like a baby—unless you're the baby in question.
Dividends? Bristol-Myers pays more—4.14% vs. Merck’s 3.42%. That might earn it a second glance from income hawks, but when you zoom out, Merck still wears the financial crown.
Now here’s where things get messier.
Merck has a bit of a single-product addiction problem. Keytruda brought in $7.83 billion last quarter, making up a jaw-dropping 50.2% of total revenue. It's a blockbuster, yes, but when one drug makes up half your business, you start looking like a biotech version of Jenga. Merck’s top five products represent 75.7% of sales.
Bristol-Myers shows better balance. Eliquis is its biggest hitter, pulling in 25.9%, while its top five products account for 71.6% overall. Not exactly ironclad diversification, but a more even spread than Merck’s lineup.
Still, Keytruda is a monster. It outsold Bristol’s Opdivo by a whopping $5.4 billion in Q4 alone. That’s not a competition—that’s a beatdown. But both companies are running out the clock on their oncology flagships. Keytruda loses U.S. patent protection in 2028. Merck’s answer is a subcutaneous version—MK-3475A—patent-protected until 2039. Bristol’s already fired back with Opdivo Qvantig, a smart preemptive strike that could buy them time and market share.
Pipelines? Merck leads here too. BMY has 74 active R&D projects, 11 in Phase 3. Merck? Over 90 clinical-stage assets, 31 of them in Phase 3, and five are already under regulatory review. They’re not just defending Keytruda—they’re building the next dynasty.
Meanwhile, Bristol-Myers’ stock is flashing overbought signals like a Christmas tree. Merck, by contrast, trades below its VWAP, and Wall Street sees an 18.3% upside from here. Bristol-Myers? A yawn-worthy 1.36%. That's a rounding error, not an investment thesis.
Fast forward to 2029. I expect Merck to print a non-GAAP EPS of $11, led by Keytruda, Welireg, and a few wild cards currently in late-stage trials. Bristol-Myers might reach $6.80 EPS on $44 billion in revenue. Not bad, just... not Merck.
After sorting through this on the summit—between water breaks, altitude headaches, and one increasingly assertive marmot—the picture came into focus. Merck is the better long-term pick. They’ve got the product, the pipeline, the margin, and the momentum.
As I packed up and started the long descent, I dropped my guard for half a second and the marmot made his move—snatched my energy bar right off my pack. Bold little bastard. But honestly, he earned it.
Sometimes, the one who climbs higher sees further and waits patiently gets the prize. Merck just did all three.
After all, in investing—as in mountain climbing—peaks and profits favour those who don’t lose their breath or their nerve.
Global Market Comments
March 27, 2025
Fiat Lux
Featured Trade:
(HOW TO GAIN AN ADVANTAGE WITH PARALLEL TRADING),
(GM), (F), (TM), (NSANY), (DDAIF), BMW (BMWYY), (VWAPY),
(PALL), (GS), (EZA), (CAT), (CMI), (KMTUY),
(KODK), (SLV), (AAPL)
One of the most fascinating things I learned when I first joined the equity trading desk at Morgan Stanley during the early 1980s was how to parallel trade.
A customer order would come in to buy a million shares of General Motors (GM), and what did the in-house proprietary trading book do immediately?
It loaded the boat with the shares of Ford Motors (F).
When I asked about this tactic, I was taken away to a quiet corner of the office and read the riot act.
“This is how you legally front-run a customer,” I was told.
Buy (GM) in front of a customer order, and you will find yourself in Sing Sing shortly.
Ford (F), Toyota (TM), Nissan (NSANY), Daimler Benz (DDAIF), BMW (BMWYY), and Volkswagen (VWAPY), were all fair game.
The logic here was very simple.
Perhaps the client completed an exhaustive piece of research concluding that (GM) earnings were about to rise.
Or maybe a client's old boy network picked up some valuable insider information.
(GM) doesn’t do business in isolation. It has thousands of parts suppliers for a start. While whatever is good for (GM) is good for America, it is GREAT for the auto industry.
So through buying (F) on the back of a (GM) might not only match the (GM) share performance, it might even exceed it.
This is known as a Primary Parallel Trade.
This understanding led me on a lifelong quest to understand Cross Asset Class Correlations, which continues to this day.
Whenever you buy one thing, you buy another related thing as well, which might do considerably better.
I eventually made friends with a senior trader at Salomon Brothers while they were attempting to recruit me to run their Japanese desk.
I asked if this kind of legal front-running happened on their desk.
“Absolutely,” he responded. But he then took Cross Asset Class Correlations to a whole new level for me.
Not only did Salomon’s buy (F) in that situation, they also bought palladium (PALL).
I was puzzled. Why palladium?
Because palladium is the principal metal used in catalytic converters, it removes toxic emissions from car exhaust and has been required for every U.S.-manufactured car since 1975.
Lots of car sales, which the (GM) buying implied, ALSO meant lots of palladium buying.
And here’s the sweetener.
Palladium trading is relatively illiquid.
So, if you catch a surge in the price of this white metal, you would earn a multiple of what you would make on your boring old parallel (F) trade.
This is known in the trade as a Secondary Parallel Trade.
A few months later, Morgan Stanley sent me to an investment conference to represent the firm.
I was having lunch with a trader at Goldman Sachs (GS) who would later become a famous hedge fund manager, and asked him about the (GM)-(F)-(PALL) trade.
He said I would be an IDIOT not to take advantage of such correlations. Then he one-upped me.
You can do a Tertiary Parallel Trade here by buying mining equipment companies such as Caterpillar (CAT), Cummins (CMI), and Komatsu (KMTUY).
Since this guy was one of the smartest traders I ever ran into, I asked him if there was such a thing as a Quaternary Parallel Trade.
He answered “Abso******lutely,” as was his way.
But the first thing he always did when searching for Quaternary Parallel Trades would be to buy the country ETF for the world’s largest supplier of the commodity in question.
In the case of palladium, that would be South Africa (EZA).
Since then, I have discovered hundreds of what I call Parallel Trading Chains and have been actively making money off of them. So have you, you just haven’t realized it yet.
I could go on and on.
If you ever become puzzled or confused about a trade alert I am sending out (Why on earth is he doing THAT?), there is often a parallel trade in play.
Do this for decades as I have and you learn that some parallel trades break down and die. The cross relationships no longer function.
The best example I can think of is the photography/silver connection. When the photography business was booming, silver prices rose smartly.
Digital photography wiped out this trade, and silver-based film development is still only used by a handful of professionals and hobbyists.
Oh, and Eastman Kodak (KODK) went bankrupt in 2012.
However, it seems that whenever one Parallel Trading Chain disappears, many more replace it.
You could build chains a mile long simply based on how well Apple (AAPL) or NVIDIA (NVDA) is doing.
And guess what? There is a new parallel trade in silver developing. Whenever someone builds a solar panel anywhere in the world, they use a small amount of silver for the wiring. Build several tens of millions of solar panels and that can add up to quite a lot of silver.
What goes around comes around.
Suffice it to say that parallel trading is an incredibly useful trading strategy.
Ignore it at your peril.
“Cheap is a dangerous word,” said technical analyst Carter Braxton Worth.
Washington D.C. - A seismic shift is underway in the landscape of financial regulation, as burgeoning calls for the deregulation of artificial intelligence (AI) within the sector gain momentum. While proponents tout the potential for innovation and efficiency, a growing chorus of experts warns that unchecked AI deployment could unleash unprecedented volatility and systemic risk upon global financial markets.
The debate, which has intensified following recent policy shifts, centers on the balance between fostering technological advancement and safeguarding the stability of the financial system. Critics argue that the rapid evolution of AI, coupled with insufficient regulatory oversight, creates a fertile ground for market manipulation, algorithmic flash crashes, and a potential erosion of market integrity.
The Deregulation Drive:
The push for AI deregulation is fueled by several factors. Firstly, the financial industry's embrace of AI has accelerated dramatically, with algorithms now playing a crucial role in everything from high-frequency trading to risk assessment and portfolio management. Advocates claim that stringent regulations stifle innovation and hinder the United States' ability to compete in the global AI race.
Secondly, a prevailing sentiment within certain political circles emphasizes minimizing government intervention in the market, viewing regulation as an impediment to economic growth. This ideology, combined with powerful lobbying efforts from tech and financial firms, has created a political climate conducive to deregulation.
"We must unleash the transformative power of AI to drive economic prosperity," argues a prominent industry lobbyist, speaking on condition of anonymity. "Excessive regulation will only serve to hamstring our financial institutions and cede our competitive advantage to nations with more permissive regulatory environments."
However, this perspective is met with fierce opposition from regulators, academics, and consumer advocacy groups, who express deep concerns about the potential consequences of unfettered AI deployment.
The Shadow of Algorithmic Risk:
One of the most pressing concerns revolves around the inherent complexity and opacity of AI algorithms. "These systems are often black boxes," explains Dr. Eleanor Vance, a leading expert in financial risk management. "We don't always fully understand how they arrive at their decisions, which makes it incredibly difficult to anticipate or mitigate potential risks."
This lack of transparency poses a significant challenge for regulators, who are tasked with ensuring market fairness and stability. The potential for algorithmic bias, where AI systems perpetuate or amplify existing inequalities, further complicates the regulatory landscape.
Furthermore, the interconnectedness of AI systems within the financial ecosystem creates the potential for cascading failures. A single algorithmic error or malicious attack could trigger a chain reaction, leading to widespread market disruption and systemic risk.
"We've already seen instances of algorithmic flash crashes, where automated trading systems triggered rapid and dramatic price swings," warns a senior regulatory official. "Without proper safeguards, these events could become far more frequent and severe."
Concerns of Market Manipulation:
The potential for AI-powered market manipulation is another major source of concern. Sophisticated algorithms could be used to exploit market vulnerabilities, engage in predatory trading practices, or spread misinformation to manipulate asset prices.
"Imagine an AI system designed to detect and exploit subtle patterns in market data, allowing it to front-run trades or manipulate prices with unprecedented precision," says a cybersecurity expert specializing in financial systems. "The potential for abuse is immense."
The proliferation of deepfakes and AI-generated misinformation further exacerbates these concerns. Malicious actors could use these technologies to spread false rumors or manipulate market sentiment, creating artificial volatility and profiting from the resulting chaos.
The Regulatory Void:
The current regulatory framework is ill-equipped to address the unique challenges posed by AI. Existing regulations, designed for traditional financial instruments and trading practices, are often inadequate for overseeing complex algorithmic systems.
"We're facing a regulatory gap," admits a financial regulator. "The pace of technological innovation has outstripped our ability to develop effective oversight mechanisms."
The development of new regulatory frameworks is further complicated by the lack of consensus on best practices and ethical guidelines for AI deployment in finance. International cooperation is also crucial, as financial markets are increasingly interconnected, and regulatory arbitrage could lead to a race to the bottom.
The Social and Economic Implications:
The potential consequences of AI-driven market instability extend far beyond the financial sector. A major market crash could trigger a global economic recession, leading to widespread job losses, social unrest, and a loss of public trust in the financial system.
Furthermore, the increasing reliance on AI in financial decision-making raises concerns about algorithmic bias and discrimination. AI systems could perpetuate existing inequalities, denying access to credit or investment opportunities to marginalized communities.
"We need to consider the social and ethical implications of AI deployment in finance," emphasizes a social justice advocate. "These systems should be designed to promote fairness and equity, not to exacerbate existing disparities."
The Call for Responsible Innovation:
Despite the risks, many experts believe that AI has the potential to revolutionize the financial industry, improving efficiency, reducing costs, and expanding access to financial services. However, they stress the need for responsible innovation, guided by robust regulatory oversight and ethical principles.
"We need to strike a balance between fostering innovation and mitigating risk," argues a financial technology expert. "This requires a collaborative effort between regulators, industry leaders, and academic researchers."
Key recommendations include:
- Enhanced transparency: Requiring financial institutions to disclose the algorithms used in their trading and risk management systems.
- Robust risk management frameworks: Developing new regulatory standards for algorithmic trading and AI-driven financial decision-making.
- Ethical guidelines: Establishing clear ethical principles for AI deployment in finance, addressing issues such as bias, discrimination, and accountability.
- International cooperation: Harmonizing regulatory standards and best practices across jurisdictions.
- Continuous monitoring and evaluation: Establishing mechanisms for ongoing monitoring and evaluation of AI systems to identify and mitigate potential risks.
The Future of Finance:
The future of finance hinges on our ability to navigate the complex challenges posed by AI. A failure to establish robust regulatory safeguards could lead to a period of unprecedented market volatility and systemic risk, with potentially devastating consequences for the global economy.
However, if we can embrace responsible innovation, guided by ethical principles and robust oversight, AI has the potential to transform the financial industry for the better, creating a more efficient, inclusive, and resilient financial system.
The coming years will be critical in determining whether we can harness the power of AI for the benefit of society, or whether we succumb to the algorithmic abyss.
Mad Hedge Technology Letter
March 26, 2025
Fiat Lux
Featured Trade:
(TECH FIRMS COULD BE OVERSPENDING)
(BABA), (MSFT)
I get it that there is a massive AI craze sweeping the tech industry and that these are the shovels to the potential gold rush in which could induce a revenue waterfall.
There have been many promises and like the fate of many promises – they aren’t kept.
Personally, I have not been convinced yet that this AI revolution will turn into some transformative movement.
Then there is the issue of whether humans will just revolt against AI once they begin to understand we are essentially training software to replace human interaction.
Talking to software engineers, the avalanche of firings in Silicon Valley has woken up their cohort.
Coders thought for a long time they were immune from firings and the gift that kept giving would continue unabated.
Now, software engineers are being terminated at record levels, and management has decided to pour money into building AI data centers.
Even China is getting in on the act.
Alibaba (BABA) itself — which in February declared it was going all-in on AI — plans to invest more than 380 billion yuan ($52 billion) over the next three years. Server farms are springing up from India to Malaysia.
Critics have also pointed out the persistent dearth of practical, real-world applications for AI.
Alibaba is mounting a comeback in 2025 thanks in part to the recent popularity of its Qwen-based AI platform, which it envisions boosting Alibaba’s core commerce business as well as cloud services.
American tech companies have already spent close to half a trillion dollars on AI data centers and there hasn’t been much revenue follow-through parallel to it.
Co-founder of Alibaba Joseph C. Tsai has said that American companies are overspending on AI data centers and less money can be spent than what is necessary to get the same result.
He said, “I’m still astounded by the type of numbers that are being thrown around in the United States about investing into AI.”
The latest news comes from Microsoft (MSFT).
They have quit new data center projects in the US and Europe that had been set to consume 2 gigawatts of electricity.
Microsoft’s retrenchment in the last six months included lease cancellations and deferrals.
Microsoft has said it will spend about $80 billion building out AI data centers this year, and that the pace of growth should begin to slow after that.
If investors don’t see anything meaningful in revenue possibilities soon, people will start to think this is beginning to feel like the Chinese ghost city problem.
China is usually not the type to overspend, and watching their development of AI for a fraction of the price is fascinating.
What does this all mean?
After a brutal correction in tech stocks in February, it could mean another leg down for tech stocks.
If it proves to be true in the short-term, tech stocks won’t deserve the premium they are fetching if they are in fact overspending on AI data centers.
Then throw into the blender that the government is fighting about trade, and there is a severe limit on what we can do in the short-term.
Global Market Comments
March 26, 2025
Fiat Lux
Featured Trade:
(WHY THE “UNDERGROUND” ECONOMY IS GROWING SO FAST)
There is no doubt that the “underground” economy is growing.
No, I’m not talking about violent crime, drug dealing, or prostitution.
Those are all largely driven by demographics, which right now are at a low ebb.
I’m referring to the portion of the economy that the government can’t see and therefore is not counted in its daily data releases.
This is a big problem.
Most investors rely on economic data to dictate their trading strategies.
When the data is strong, they aggressively buy stocks, assuming that a healthy economy will boost corporate profits.
When data is weak, we get the flip side, and investors bail on equities. They also sell commodities, precious metals, and oil, and plow their spare cash into the bond market.
We are now halfway through a decade that has delivered unrelentingly low annual GDP growth, around the 2% to 2.5% level.
We all know the reasons. Retiring baby boomers, some 85 million of them, are a huge drag on the system, as they save and don’t spend.
Generation X-ers do spend, but there are only 58 million of them. And many Millennials are still living in their parents’ basements—broke and unable to land paying jobs in this ultra-cost-conscious world.
But what if these numbers were wrong? What if the Feds were missing a big part of the picture?
I believe this is, in fact, what is happening.
I think the economy is now evolving so fast, thanks to the simultaneous hyper-acceleration of multiple new technologies that the government is unable to keep up.
Further complicating matters is the fact that many new Internet services are FREE, and therefore are invisible to government statisticians.
They are, in effect, reading from a playbook that is decades or more old.
What if the economy was really growing at a 3% to 4% pace, but we just didn’t know it?
I’ll give you a good example.
The government’s Consumer Price Index is a basket of hundreds of different prices for the things we buy. But the Index rarely changes, while we do.
The figure the Index uses for Internet connections hasn’t changed in 20 years.
Gee, do you think that the price of broadband has risen in a decade, with the 1,000-fold increase in speeds?
In the early 2000s, you could barely watch a snippet of video on YouTube without your computer freezing up.
Now, I can live stream a two-hour movie in High Definition on my Comcast Xfinity 1 terabyte per second business line. And many people now watch movies on their iPhones. I see them in rush-hour traffic and on planes.
I’ll give you another example of the burgeoning black economy: Me.
My business shows up nowhere in the government economic data because it is entirely online. No bricks and mortar here!
Yet, I employ 15 people, provide services to thousands of individuals, institutions, and governments in 140 countries, and take in millions of dollars in revenues in the process.
I pay a lot of American taxes, too.
How many more MEs are out there? I would bet millions.
If the government were understating the strength of the economy, what would the stock market look like?
It would keep going up every year like clockwork, as ever-rising profits feed into stronger share prices.
But multiples would never get very high (now at 20 times earnings) because no one believed in the rally, since the visible economic data was so weak.
That would leave them constantly underweight equities in a bull market.
Stocks would miraculously and eternally climb a wall of worry, as they did until February.
On the other hand, bonds would remain strong as well, and interest rates low, because so many individuals and corporations were plowing excess, unexpected profits into fixed-income securities.
Structural deflation would also give them a big tailwind.
If any of this sounds familiar, please raise your hand.
I have been analyzing economic data for a half-century, so I am used to government statistics being incorrect.
It was a particular problem in emerging economies, like Japan and China, which were just getting a handle on what comprised their economies for the first time.
But to make this claim about the United States government, which has been counting things for 240 years, is a bit like saying the emperor has no clothes.
Sure, there has always been a lag between the government numbers and reality.
In the old days, they used horses to collect data, and during the Great Depression, numbers were kept on 3” X 5” index cards filled out with fountain pens.
But today, the disconnect is greater than it ever has been, by a large margin, thanks to technology.
Is this unbelievable?
Yes, but you better get used to the unbelievable.
There May Be More Here Than Meets the Eye