CEO of Wall Street Asset Management Firm Reveals Holdings: Bets on Nasdaq and BTC—The “Middle Ground” Is the Worst Choice
TechFlow Selected TechFlow Selected
CEO of Wall Street Asset Management Firm Reveals Holdings: Bets on Nasdaq and BTC—The “Middle Ground” Is the Worst Choice
The “Big Seven” are mispriced; the AI capital expenditure panic is overblown.
Navigating Web3 tides with focused insightsCompiled & Translated by TechFlow
Guest: Anthony Pompliano, Founder & CEO of Professional Capital Management
Host: John Pompliano, Associate at Pomp Investments
Podcast Source: Anthony Pompliano
Original Title: I Just Revealed My Current Portfolio…
Air Date: June 24, 2026
Key Takeaways
Anthony Pompliano and John Pompliano reunite for this episode to deeply dissect the brutal sell-off in the U.S. stock market’s “Magnificent Seven” (Mag 7), widespread panic over AI capital expenditures (CapEx), and why underlying inflation data are far safer and more contained than sensationalized media headlines suggest.
The guest is Anthony Pompliano, Founder and CEO of Professional Capital Management, a long-time active participant in the crypto investment space whose podcast ranks among North America’s most-listened-to financial podcasts.
We’ll also fully recap Anthony’s latest portfolio holdings, deliver an exclusive deep dive into the real political chess game behind newly appointed Federal Reserve Chair Kevin Warsh’s interest-rate decisions, and thoroughly unpack why Bitcoin’s “high volatility” is not a risk—but rather its core, ultimate advantage—on the decade-long journey ahead.
Key Insights Summary
The Real Logic Behind the Mag 7 Pullback
- “The Mag 7 were sold off primarily because they are fundamentally long-duration assets—and therefore highly sensitive to inflation.”
- “For the Mag 7, these AI companies won’t suddenly explode or go to zero.”
- “If inflation recedes in Q3 and Q4, Mag 7 valuation multiples will likely re-expand.”
- “Google’s valuation is now cheaper than Apple’s, yet its growth is faster. Why? The core reason is market concern over AI CapEx. This creates an excellent opportunity: buying the world’s best companies at lower valuations.”
- “Many value investors have underperformed for years because they’re too rigidly applying past frameworks to today’s businesses. You should respect history—but you can’t just point to old data and say ‘these companies are too expensive’ year after year for 15 years.”
- “It’s highly plausible that a company announces $10 billion in AI CapEx over two years, but ends up spending only $6 billion. $6 billion remains extremely high—but it’s 40% less than what the market feared. Markets are overly fixated on inflation direction right now, and I believe this has led to mispricing.”
Repricing AI CapEx and ROI
- “The real issue today lies in token costs. When I talk with CEOs, everyone says the same thing: ‘We’re spending too much on tokens, but we have no idea where the ROI is.’”
- “Everyone is converging on the same goal: achieving the same output using fewer tokens.”
- “What bottlenecks the broader adoption of this technology? Power, data centers, and chips.”
- “Just the other day, I saw news about a company shifting employee work hours to 1 a.m.–10 a.m., partly because model inference is cheaper, system load is lower, and responses are more accurate during that window.”
Investment Philosophy: Avoiding the Middle Ground
- “If you summarize investment approaches, there are broadly two paths. First, buy large-cap indices—a sound strategy in itself. Second, pursue highly asymmetric opportunities. The worst path is the middle ground: allocating capital to mid-sized companies with only modest growth and unremarkable return profiles.”
- “Today, the two most fascinating technologies globally are Bitcoin and AI.”
- “I hold Tesla. I believe Elon Musk’s potential monopoly in humanoid robotics and autonomous driving would be extraordinarily valuable.”
- “Once he leverages AI, machine learning, and computer vision to equip hardware with perception and cognition, he secures the embodied entry point into ‘physical AI’—a position I consider extremely valuable.”
- “Where do I want exposure? I maintain a long-term cash position and hold Bitcoin.”
- “Anduril’s model is to let startups pioneer technological breakthroughs first; once validated, Anduril acquires those drone companies. It excels at M&A and boasts a strong BD team capable of rapidly commercializing acquired technologies via government contracts. In short, it’s a ‘buy-then-commercialize’ platform.”
His Actual Portfolio
- “In private markets, I hold many software-based AI companies—like Replit, Lovable, and Micro One—that dominate their respective verticals. In public markets, I concentrate on physical AI and robotics.”
- “Across both public and private markets—and across both software and hardware—I maintain a relatively comprehensive AI exposure.”
- “I hold Tesla—not for its cars—but because I believe Elon’s potential monopoly in humanoid robotics and autonomous driving would be extremely valuable.”
- “He will ultimately merge SpaceX and Tesla—likely before 2030. Once that happens, he’ll consolidate his most critical projects into a massive integrated entity.”
Bitcoin Enters a New Phase
- “Unhappiness stems fundamentally from the gap between expectation and reality. So don’t set absurd expectations for Bitcoin. If you expect 25–30% annual returns and it delivers more, you’ll be pleasantly surprised. But if you expect 100% every year, you’re setting yourself up for disappointment.”
- “Governments will keep printing money—the core investment thesis for Bitcoin remains intact. What’s changed is simply the stage of the game.”
- “Earlier, Bitcoin was like high-school basketball—you were one of only a few standout players on the court. Now it’s more like college basketball: all players are stronger, the gaps aren’t as extreme, but the overall quality of play is higher.”
- “Retail investors tend to be more emotional; institutions tend to be less so. I know internet sentiment around Bitcoin is poor—but structurally, this is also a typical part of bottoming processes.”
Mag 7 Sell-Off and AI CapEx
Anthony Pompliano:
Many people start worrying when Bitcoin becomes volatile—but that’s entirely the wrong way to think about it. In fact, I prefer holding assets that go through periods of unpopularity. I don’t want to own perpetually hot assets—because if something stays in the market’s spotlight forever, it’s likely overcrowded and its returns already arbitraged away. Conversely, assets that periodically fall out of favor often possess greater asymmetry and higher potential future returns—so volatility itself is essential.
In this episode, John interviews me on topics including Kevin Warsh, the Fed, inflation expectations, interest rates, the AI CapEx debate, Mag 7, the S&P 500, Bitcoin, SpaceX, and more. I’ll lay out my current investment framework across both public and private markets—and name specific assets I’ve already allocated to my portfolio.
John Pompliano: Let’s begin with our first topic. Mag 7 recently underwent a broad sell-off. Do you see this as a valuation reset—or is the market fundamentally rethinking the entire AI trade thesis?
Anthony Pompliano:
I believe the Mag 7 sell-off stems primarily from their nature as long-duration assets—and thus their high sensitivity to inflation. Earlier, markets worried Iran-related conflict would push energy prices higher—and reignite inflation—so these assets were sold first.
I’ve repeatedly emphasized two points since 2025.First, tariffs won’t cause persistent inflation. Second, unless the Iran conflict escalates into a multi-year war, its impact will be purely a short-term price shock. We’ve indeed seen such a short-term spike in energy prices—but it won’t evolve into structural inflation. More broadly, powerful deflationary forces remain active across the U.S. economy.
Certainly, some will shout “inflation is back!” the moment it hits 3%. I disagree. Many people’s understanding of inflation has been warped by the extreme 9%+ surge earlier this decade—as if such spikes recur regularly. Yet historically, inflation exceeding 9% is exceedingly rare over a lifetime. If I recall correctly, the last time U.S. inflation exceeded 9% was in the 1970s—before I was even born. That means, in my entire life, it’s happened only once.
That surge was severe—but also easily understandable. Massive U.S. money printing and near-zero interest rates in 2020 made high inflation almost inevitable. But that artificial manipulation differs fundamentally from transient price shocks caused by Iran conflict or tariffs.
If you examine Mag 7, the market currently harbors two layers of concern.First, if inflation rises further, these long-duration, rate-sensitive assets face valuation compression—leading to price corrections. That logic holds. Second, investors fear AI CapEx is excessive—worrying that frenzied capital spending will erode free cash flow and reduce future shareholder distributions—thus lowering valuations.
Yet I foresee two developments.First, inflation won’t be as severe as the market fears. Early signs already point that way. In the latest inflation data, 60% of the increase came solely from energy—meaning energy prices drove the headline number. Once energy prices decline, inflation will follow. Oil has already dropped below $80/barrel; if it falls to $60, what happens to inflation? I won’t declare definitively that “inflation has peaked”—but it’s very close. Whether it peaks now or in May/June data, my view is clear:Inflation in Q3 and Q4 will be lower than in Q2. If correct, Mag 7 valuation multiples will re-expand.
The second issue is AI CapEx. To me, it hinges on two key questions.First, does the demand we anticipate actually exist? Second, will this CapEx ultimately generate returns? If demand exists, returns logically follow—since users will adopt the infrastructure. Does demand exist? Simply ask: Are you using AI more frequently today than a year ago? Yes—absolutely.
Yet recently, markets and media have shifted focus: Is our current use of AI tokens efficient enough? Is the compute and energy consumed to call models truly justified? I feel this deeply. Many know we’re building a product called CFO Sylvia. During development, we uncovered a stark reality: costs are rising. Because users generate queries themselves, without intelligent token management, spending theoretically has no ceiling.
Initially, the team prioritized “getting the product shipped.” Only after launch did we realizethe real problem lies in token cost. So what did we do? We optimized token efficiency.
We made numerous concrete adjustments. For example, certain pages re-ran models on every refresh—an unnecessary waste. We disabled them outright, instantly slashing token consumption. Other features triggered periodic model calls—cool in theory, but major token drains. We cut them entirely and monitored user complaints. No one complained—so we kept them gone. Another big token saving.
We later refined our architecture and scrutinized cash costs more closely. At first, I assumed this was unique to us—so I focused inward. But when I spoke with fellow CEOs, they all echoed the same line: “We’re overspending on tokens—and have no clue where the ROI is.”
Hence, everyone is converging on the same goal:achieving identical outputs with fewer tokens. That’s why I’ve long argued on Twitter that internal corporate competitions—like “who calls the most models”—won’t last. The world will inevitably refocus on efficiency, effectiveness, and ROI—the metrics that truly matter.
This is precisely why I’m most bullish on private-model firms like OpenAI, Anthropic, and Grok. Clients like CFO Sylvia—or any enterprise customer—are demanding: “I want to use fewer tokens, but get the same results.” Individual clients become more efficient—while total addressable markets expand and product adoption rises—meaning model providers’ total revenue may still soar. To me, that signals product-market fit.As long as open-source models haven’t fully displaced them, these companies are running excellent businesses.
The next question becomes: What bottlenecks broader adoption?Power, data centers, and chips. You’ll notice constraints cluster precisely here.Just days ago, I read about a company shifting employee hours to 1 a.m.–10 a.m.—partly because model inference is cheaper, system load is lower, and responses are more accurate during that window. Most firms won’t do this—but it shows how seriously companies now weigh inference efficiency.
If demand persists, then today’s insufficient data centers, power capacity, and chip supply imply continued bidding for these resources.
So those fretting over AI CapEx ROI can legitimately raise a long-term question: Will overbuilding occur? Yes—overbuilding eventually emerges in every cycle. History proves precise control is nearly impossible. But in the near term, I see no sign of it. Instead, I see surging software demand, intense workflow specialization, robust demand for data centers, power, and chips—even to the point where orbital data centers are being discussed.
Thus, I believe market anxiety is misplaced—especially regarding Mag 7. Many focus on “what companies say they’ll spend,” but guidance ≠ reality. Just as projected revenue ≠ confirmed revenue, projected CapEx ≠ actual spending.
It’s highly plausible a company announces $10 billion in CapEx over two years—but spends only $6 billion. $6 billion remains enormous—but it’s 40% less than feared. Markets are obsessing over inflation direction, and I believe this has been mispriced. More critically, markets overtrust AI CapEx guidance. I suspect actual spending will fall significantly below current fears.
So reconsider Mag 7 individually. Take Google: its recent price drop is intriguing.Google trades at a lower valuation than Apple—yet grows faster. Why? Market concerns over AI CapEx. If you examine fundamentals and data objectively, this presents a compelling opportunity: buying the world’s best companies at discounted valuations.
Look at their P/E ratios and other valuation metrics—they’ve clearly declined over recent years. Many now sit in ranges that reasonable investors wouldn’t deem extreme. As long as AI CapEx doesn’t become a genuine problem, these enterprises merit allocation.
So stay calm—don’t panic.At least for Mag 7, these AI companies won’t suddenly explode or go to zero. More importantly, ask yourself:Has anything fundamentally changed about these companies versus three weeks ago? Mag 7 rose 18% over the past year. Headlines obscure this entirely. Media portrays the Mag 7 story as over—as if the entire thesis collapsed. True, they’ve lagged YTD: the S&P 500 rose 9%, the other 493 stocks ~13%, while Mag 7 is flat or slightly down. Sounds like “Mag 7 is broken.”
But iftheir long-term value remains intact, shouldn’t you find them more attractive? Other stocks have rallied—while these giants’ valuations barely moved over six months. You effectively gain six months of growth and earnings validation—yet pay yesterday’s price. That’s the opportunity Mag 7 offers today.
Of course, this doesn’t mean allocating your entire portfolio to Mag 7—I’m not suggesting that. But if you’re a value investor, favor large-cap tech, or lean toward index strategies, I’d worry little about Mag 7. On the contrary, I find them increasingly appealing.
I also suspect much of this anxiety stems from boredom. We’re amid a generational bull market—many feel there’s no new narrative, so they invent problems and obsess over hypothetical failure modes. Risk management demands such thinking—but I prefer focusing on issues already proven and verified—not endlessly worrying about “what if it rains, a squirrel pops out, and a peanut falls from the tree.” AI CapEx isn’t a current problem. It might become one—but today, it isn’t.
John Pompliano: If it’s not a problem yet—how should success be measured? User growth? Revenue growth? Margins? Free cash flow? What metrics should enterprises use to judge whether AI CapEx is worthwhile?
Anthony Pompliano:
Set AI CapEx aside for a moment and consider a foundational fact: S&P 500 profit margins have risen 58% since 2011. Over more than a decade, these companies’ earnings quality improved dramatically. They deserve higher valuations—because they’ve become better companies.
Many overlook a monumental shift: commerce has transitioned from analog to digital. These firms own intellectual property, sell digital products, operate more efficiently, and boast superior business models—so their enterprise value should naturally rise.
Yet many still scare themselves with historical valuations—citing “look at historical ranges” or “remember the dot-com bubble.” But the dot-com era meant printing MapQuest directions to drive across town—a world without iPhones or AI. Comparing 2026 valuations to 1992 makes no sense.
Recently, I visited our San Francisco office and sat with Sylvia’s engineers. I voiced a need—and they fed it directly into Claude. Engineers understand underlying architecture, databases, file systems, and repair protocols—but fundamentally, they now wield magic. This world bears no resemblance to the past.
That’s why I say,many value investors have underperformed for years because they cling too rigidly to outdated frameworks for evaluating today’s enterprises. You should respect history—but you can’t just point to old data and say ‘these companies are too expensive,’ repeating it for 15 years.
So the question arises: Do you wait for a 20% market drop to crow “I told you it was overvalued”—or jump in and participate in this generational bull market? I observe the world’s best investors doing exactly that. Whether Warren Buffett or other top-tier investors—they recognize these are real companies solving real problems, generating real profits, and sustaining real growth. Therefore, they allocate.
Measuring success is simple: Are companies consistently generating profits? Yes—and many do. That’s why they’re valuable. Everyone frets over free cash flow—but remember: S&P 500 margins have risen 58% since 2011. Will they keep rising? Absolutely—100% yes. These companies improve continuously—like fine wine aging into greater value.
So instead of standing outside criticizing, acknowledge their ongoing evolution. You needn’t constantly fault Mark Zuckerberg, Jeff Bezos, Andy Jassy, or Sergey Brin for mistakes. Reality is, they’ve gotten many things right.
Large-Caps vs. Asymmetric Opportunities
John Pompliano: Excellent point. When viewing Mag 7, most investors assume they’ll be worth more a decade from now. Yet the market also offers crypto assets and other avenues. How do you balance short-term and long-term allocation?
Anthony Pompliano:
I pay little attention to short-term volatility. “Short-term” doesn’t mean daily or weekly—I rarely even look monthly. My default time horizon is 10 years.
Within that framework, I prioritize one objective: earning more money and accumulating more deployable capital. Once I buy an asset and commit to holding it long-term, I focus on generating new capital. Then I decide whether to allocate fresh funds to new opportunities.
This doesn’t mean I’ll literally hold every position for a full decade. If core theses change or assumptions break, I adjust. But overall, my trading activity is minimal.
Summarizing investment approaches, there are broadly two paths.First, buy large-cap indices. Exact figures escape me—but Nasdaq’s 10-year CAGR is roughly 18%, possibly higher. That’s extraordinary. An asset compounding 18–20% annually over a decade is exceptionally rare to beat consistently. So,index investing is itself an excellent strategy. Second, pursue highly asymmetric opportunities.
The worst path is the middle ground. Don’t allocate to mid-sized companies offering only modest growth and unremarkable returns. Such assets lack both the cash flow, resilience, and business durability of large-cap leaders—and the explosive, asymmetric upside potential of smaller opportunities.
SoI favor a “barbell” approach—avoiding the middle. To me,the two most fascinating technologies globally today are Bitcoin and AI. Of course, both permeate other industries. In finance, Bitcoin is embedded into more and more products. BlackRock even prominently features its IBIT ETF on its homepage—a mainstream signal if ever there was one.
Now consider AI. Take SpaceX: it executed history’s largest IPO, drawing immense attention. But what *is* SpaceX? A space company? An AI company? Something else?
You could argue its near-monopoly in space launches is valuable; Starlink’s satellite internet service is valuable; its orbital data center vision is valuable. Yet reviewing its IPO roadshow materials, one slide details its total addressable market (TAM). Beyond rockets and satellites, enterprise AI appears—and dominates the TAM size. I forget exact numbers, but enterprise AI comprises the vast majority. In essence, SpaceX *is* an AI company—that’s why markets assign it such lofty valuations.
It later finalized—or advanced—its acquisition agreement for Cursor, partnered with xAI, and signed multiple new contracts. I recall at least three contracts signed within the past three weeks—each commanding hundreds of millions in monthly compute revenue. Suddenly, the company added a computing business that didn’t exist two years ago—and may already run at a multi-billion-dollar annualized revenue rate.
This circles back to AI CapEx ROI. For xAI and SpaceX, do these investments add value? Absolutely. So from a portfolio perspective, I ask myself:Where do I want exposure? I’ll hold cash long-term—and hold Bitcoin. Beyond that, I want exposure where growth and innovation are genuinely happening—and in my view, that’s overwhelmingly tied to artificial intelligence.
This doesn’t mean software alone matters. Hardware and power are equally critical.For instance, I hold Tesla—not for its cars—but because I believe Elon’s potential monopoly in humanoid robotics and autonomous driving would be extraordinarily valuable. I also believe he’ll merge SpaceX and Tesla—likely before 2030. Once merged, he’ll integrate his most critical projects into a massive conglomerate. Its foundation relies on enabling hardware to “see” and “think.” If he achieves this via AI, machine learning, and computer vision, he captures the embodied entry point into “physical AI”—a position I consider extremely valuable.
I also hold Robbo Strategy, a publicly traded closed-end fund with significant exposure to private robotics companies. Why is it interesting? Because it transforms the “physical AI + robotics” theme into a publicly accessible proxy for private-market exposure. Tesla gives me public-market exposure; Robbo Strategy adds private-robotics exposure—combined, they form a systematic bet on physical AI and robotics. I see this as a massive opportunity.
Masayoshi Son recently declared physical AI the next trillion-dollar opportunity. Reviewing the technologies and applications under development, I largely agree.
Another fascinating area is defense tech convergence. I’ve openly disclosed holding Anduril stock. Its business model is distinctive. Drone industry challenges include: teams that build breakthrough tech often lack commercialization skills.Anduril’s model lets startups pioneer tech first; once validated, Anduril acquires them. It excels at M&A and boasts a strong BD team that rapidly deploys acquired tech into commercial contracts. In short,it’s a “buy-then-commercialize” platform. I like it because it again centers on the same themes: defense, AI, robotics, and machine learning—all packaged into drones. And they’re advancing beyond aerial drones to ground and maritime platforms.
Thus, my portfolio follows a clear thread. In private markets, I hold many software-based AI firms—like Replit, Lovable, and Micro One—dominant in their verticals. In public markets, I concentrate on physical AI and robotics. So across both public/private and software/hardware, I maintain a relatively comprehensive AI exposure.
Bitcoin remains a core conviction. I believe Bitcoin will continue serving as a check against federal overprinting, monetary policy distortions, and fiscal policy failures. As long as this logic holds, U.S. debt expands, the dollar dilutes—and Bitcoin performs well long-term.
Those panicked by Bitcoin’s volatility fundamentally misunderstand. I welcome assets passing through “unpopular” phases. I avoid perpetually hot assets—because popularity implies overcrowding, and overcrowding implies arbitrage has already priced in returns. Conversely, assets cycling between favor and disfavor often harbor higher future return potential. So volatility isn’t a flaw—it’s a feature I deliberately seek.
Consider Tesla, Robbo Strategy, Anduril, and Bitcoin—compared to holding Walmart stock alone, they’re inherently more volatile. But that’s precisely what I want.I want thematic volatility—and I bundle these names into a portfolio that simultaneously bets on one major theme across both public and private markets.
SpaceX and Other AI Exposure & Portfolio Construction
John Pompliano: Your allocations skew more toward consumer-facing applications—or more toward infrastructure (“picks-and-shovels”) plays?
Anthony Pompliano:
Take Tesla: many view it as a consumer company selling self-driving cars and humanoid robots. But I see it more as a “picks-and-shovels” company.
Why? Because Tesla’s true value isn’t the hardware itself. Hardware is merely its monetization channel. Its deeper value lies in computer vision, machine learning, AI models, real-world data—and how it integrates these models and data into workflows that hardware executes.
Viewed this way, it’s already approaching infrastructure—because what you truly own is model capability. That’s why OpenAI and Anthropic embed into so many products.
When people hear “picks-and-shovels,” they imagine traditional, low-level infrastructure—not models. Yet how much model revenue stems from pure consumer use cases? A great deal. So in my view, Tesla *is* a picks-and-shovels company—though the market mislabels it as purely consumer-facing due to direct B2C transactions. I disagree.
Kevin Warsh and the Fed
John Pompliano: Let’s shift topics—Kevin Warsh. Do you expect rate cuts this year? Prediction markets still lean toward “hold” for upcoming meetings.
Anthony Pompliano:
Let’s apply first principles.Three conditions must align for a rate cut: First, inflation must fall further. Second, the Fed Chair must personally believe rates should be lower. Third, the economy must withstand lower rates.
These factors interlink. Lower inflation typically implies the economy can absorb lower rates; and if inflation falls *and* the economy endures, you ultimately need a decisive, execution-ready leader to press the button.
My view: a rate cut will occur sometime in 2026. Some market participants even anticipate hikes—but that contradicts my core view: inflation won’t be as high—or as persistent—as widely feared.
A crucial distinction: I don’t believe the U.S. will become “cheaper” again—but “affordability” and “inflation” are distinct concepts. Inflation is an economic metric for central bankers and professional investors; affordability is what ordinary Americans care about. People don’t care whether prices rise 2% or 3%—they care whether groceries and bills feel unaffordable, making life harder.
So you can’t tell ordinary people, “The dollar’s purchasing power fell 30% over five years”—then comfort them, “But inflation is only 2%.” Investors and central bankers cheer—signaling controlled inflation, healthy economics, and future rate cuts. But ordinary people ignore this; they care about affordability—not academic definitions of inflation.
So is inflation truly problematic? I don’t think so. Can the U.S. economy withstand lower rates? Yes—simply because American consumers remain resilient, spending data looks strong, equities rally, and household wealth expands. Wage growth faced pressure earlier—but that’s healing, aided by falling energy prices and Iran conflict resolution.
If inflation data truly turn downward in Q3, Kevin Warsh is exactly the type who acts decisively. He’s not the sole decision-maker—FOMC votes collectively—but my judgment is clear: once inflation trends downward, they’ll hit the button and deliver a rate cut.
I don’t foresee aggressive triple cuts—but a single cut is entirely plausible. It’s more about signaling: “We’re willing to act.”
What would change my view? Persistently elevated inflation—or clear U.S. economic weakness. But I see neither. So I believe a rate cut—some form, sometime before year-end—is probable. Prediction markets will reassess odds as data evolves.
Of course, unchanged rates aren’t catastrophic either. I don’t believe current rates materially harm the U.S. economy. But hiking *now* would be premature, unwise—and risk creating new problems. My stance: I lean toward cuts, but unchanged is acceptable. Roughly, I’d assign a 60% probability to a year-end cut.
John Pompliano:
So do you see his current posture as “we must do something”—or more “let’s observe for months, gather more data”? Intriguingly, the same data yield divergent conclusions. For instance, credit card delinquency rates might suggest consumers aren’t so strong.
Anthony Pompliano:
That’s a credit-card issue—not a consumer issue. This distinction is vital.
John Pompliano:
If I recall correctly, average U.S. credit card rates are now near 23%.
Anthony Pompliano:
If you can’t clear your balance monthly, high rates will crush you. So does that mean “consumers are failing”? Not necessarily. U.S. consumer spending remains robust.
Some argue high credit card delinquencies prove consumption falters. I disagree. Flip the question: If credit card rates were 5%, would delinquencies be this high? Unlikely.
I don’t support government-imposed credit card rate caps—because forced intervention would prompt issuers to slash credit lines. For these firms, >20% rates reflect risk-pricing models—they’re simply assigning credit based on mathematical calculations.
Back to Warsh: carefully review his press conference, statements, and post-meeting commentary. I’d describe it as “economic sleight-of-hand.” What does that mean? He held rates steady—verbally declaring “rates unchanged”—but he’s fundamentally restructuring the Fed’s operating framework.
He’s forming working groups, revising inflation measurement methodologies, explicitly rejecting past practices, and abandoning forward guidance. Don’t underestimate these moves—even one such action by Jerome Powell would dominate headlines for days. If Powell suddenly announced “We’re ending forward guidance,” markets would erupt.
Yet leadership transition grants Warsh natural leeway to overhaul many things—especially since he’s doing so *without* touching rates. It’s like walking a tightrope: avoiding rate changes buys him greater flexibility elsewhere. Had he cut rates immediately, he might not have secured space for these reforms.
So he and colleagues chose: delay rate decisions, gather more data—and aggressively reform everything within their authority. He’s laying foundations, shaping culture and environment—and I suspect he’ll optimize personnel, ensuring key monetary policy roles are filled by those he deems best suited.
Honestly, I don’t find his stated positions unreasonable. Some dislike them, others disagree—but to me, the logic holds. For instance, he advocates using inflation metrics immune to short-term price spikes—and argues against constant future forecasting, which traps policymakers in self-made guidance. All reasonable to me.
Overall, I’d rate his start solidly: 8 or 8.5 out of 10—not perfect, but far above passing—roughly a B+.
Bitcoin Over the Next Decade
John Pompliano: Finally, Bitcoin. Over the next 10 years, what are your bear and bull cases?
Anthony Pompliano:
Peter Schiff recently conceded on national TV that he no longer believes Bitcoin will go to zero. Consider him the “last holdout of the zero-case”—and if even he abandons it, most accept one truth:The probability of Bitcoin going to zero is extremely low.
Will it crash? Yes—entirely possible. But Bitcoin’s final critical barrier—institutional adoption—has already fallen. With that milestone passed, perceptions will gradually shift.
Many loved Bitcoin for its glitter, asymmetry, and wild volatility—a “cowboy-era king asset.” But it’s no longer purely that. So some friends now buy Bitcoin less enthusiastically, thinking: “How’s Bitcoin doing? Up 30% yearly? I’d rather invest elsewhere for potentially higher returns.”
Meanwhile, many institutional investors say: “You’re saying this asset could compound ~30% annually for a decade—and it’s now relatively mature? I’m intrigued.”
In short, volatility compressing from ~80 to ~35–40 makes Bitcoin more attractive to large capital pools.
So for the bull case: you won’t see 10x or 20x super-bull markets—but you’ll get sustained compounding. My baseline expectation: Bitcoin delivers ~25–30% annualized returns over the next decade. Less spectacular than before—but still vastly outpaces most equity benchmarks.
Think about it: how many assets sustain 25–30% annualized returns for a decade? Very few. So I believe it will perform exceptionally well.
Yet I always remind people:Unhappiness stems from the gap between expectation and reality—so never set absurd expectations for Bitcoin. Expect 25–30%, and if it exceeds that, you’ll be delighted. But expect 100% yearly—and you’ll guarantee disappointment.
So maintain rational expectations. I believe Bitcoin will continue performing well, gaining broader adoption, and climbing steadily upward.Governments will keep printing money—the core investment thesis remains intact. What’s changed is simply the stage of the game.
Earlier, Bitcoin was like high-school basketball—you were one of only a few standout players on the court. Now it’s more like college basketball: all players are stronger, the gaps aren’t as extreme, but the overall quality of play is higher—and more compelling to watch. Further ahead, it may enter the NBA—perhaps the sovereign wealth fund era. We’re not there yet—but we’re already in college basketball territory, perhaps even Duke-level games.
John Pompliano: Compared to 2022, does this bear market feel different? Subjectively, retail sentiment seems worse—but institutional enthusiasm for Bitcoin may be higher than ever.
Anthony Pompliano:
Internet sentiment around Bitcoin is indeed poor.
John Pompliano:
Maybe it’s because we used to live in Miami and now reside in New York—so our perception differs—but I genuinely feel this time is different.
Anthony Pompliano:
Yes—internet retail sentiment is terrible. I see longtime acquaintances attacking each other with extreme rhetoric; some emotionally collapse and exit the market entirely. Institutional sentiment? Almost none of this occurs.
They may not be adding more capital to Bitcoin—but their overall demeanor is calmer and more methodical. They have investment committees, risk management protocols, and data-driven processes. So a general conclusion emerges:Retail investors tend to be more emotional; institutions tend to be less so. Of course, institutions comprise humans—and humans feel emotions—but intensity differs.
Thus, I describe the present as: poor internet retail sentiment—but decent institutional sentiment. I don’t see institutional euphoria or pervasive bubbles. They’re still building technologies, launching Bitcoin funds, developing custody solutions, and acquiring related teams.
For example, Franklin Templeton recently acted. OKX and ICE also just launched new products. Progress continues.
So when people say “investor sentiment is poor,” they often use internet chatter as a proxy. But today’s market isn’t confined to the internet. Previously, Bitcoin lived almost entirely online—I could gauge daily price direction just by reading internet mood. Not anymore—because a whole new player group entered the arena.
Perhaps the market is now split 50–50: half the influence and critical information remains online; half has shifted to institutions. Recall the past structure was 100–0—almost no institutional presence. Ten years from now? Could it be 80–20 or 90–10? Entirely possible.
This signals rising institutional importance—and their larger capital pools—while some retail investors surrender, exit, and pivot elsewhere.
So yes, I know internet sentiment is poor—and I dislike seeing friends fight. But structurally, this is a common phase in bottoming processes. Have we hit the bottom? I don’t know. But we’ve fallen substantially—and many signs suggest the bottom may be near—or already here. What comes next depends on market action.
Join TechFlow official community to stay tuned
Telegram:https://t.me/TechFlowDaily
X (Twitter):https://x.com/TechFlowPost
X (Twitter) EN:https://x.com/BlockFlow_News
深潮TechFlow
