The inflation story was supposed to be over. Central banks had hiked aggressively, demand had cooled, and the narrative had shifted decisively toward rate cuts. Back in January 2026, markets had priced in two reductions for the year. As of this writing, that has been pared to one—and is edging toward zero. That repricing is not an accident. It is the market slowly absorbing what this piece argues is a structural reality: that the comfortable disinflation of 2024–2025 has collided with three forces that do not resolve quickly—an oil market structurally broken long before Iran, a central bank constrained by the wrong tool for the wrong problem, and a fiscal position that quietly shifts incentives away from genuine price stability. The assets that win in this environment are not the ones that dominated the last cycle. This is the argument for why, and what to do about it.
The Structural Oil Supply Deficit
To understand where oil prices are today, you have to go back to a deliberate act of market warfare. In mid-2014, oil traded above $110 a barrel. By early 2016 it had collapsed below $30—the result of a Saudi decision to flood the market and kill US shale. The strategy largely failed to kill shale, but it devastated the balance sheets of every major oil company globally and triggered the first of three waves of capital expenditure destruction that now defines the structural supply picture. Long-cycle upstream projects—deep-water drilling programs, oil sands expansions, complex reservoir developments that take five to ten years from investment to first barrel—were cancelled en masse. The wells not drilled in 2015–2018 are not producing oil today. That is a direct and unrecoverable supply gap.
The second wave came from a different direction. From roughly 2018 onwards, institutional investors applied sustained pressure on oil majors to stop investing in new fossil fuel production. The argument was partly moral, partly financial: why allocate billions to oil fields that might be stranded assets in twenty years? Oil company boards faced a stark choice—invest in new production and be punished by shareholders, or return cash through dividends and buybacks and be rewarded. Almost universally they chose the latter. This was rational. But the collective result was a multi-year structural underfunding of the supply pipeline at precisely the moment demand was recovering.
The third wave arrived in March 2020. Demand collapsed overnight. Oil prices briefly went negative in April of that year—an extraordinary moment where producers were paying buyers to take oil off their hands. Capex was cut to the bone across the entire industry. The demand recovery came faster than anyone expected, but you cannot restart a five-year deep-water drilling project in six months. The supply response was structurally lagged—and that lag produced the 2022 energy crisis even before a shot was fired in Ukraine.
These three shocks created a compounding supply gap across nearly a decade. Oil fields naturally deplete at four to eight percent per year without continuous reinvestment. Underfund capex across three separate crises and you are not just failing to add new supply—you are allowing existing supply to erode. The industry is running on fields developed before 2015. Those fields are ageing and becoming more expensive to maintain. US shale partially filled the gap, because shale has short cycle times—wells can be drilled in weeks. But post-2022, shale investors are demanding returns over growth. The aggressive drilling mentality of the early shale era is gone. This is why a return to $60 oil is not a credible scenario even if the Iran conflict resolves tomorrow. Geopolitics is the accelerant. The fuel was already burning.
The Iran Conflict and Geopolitical Risk Premium
On top of this structural foundation, the Iran conflict has introduced three distinct additional pressures. The first is direct supply risk—roughly twenty percent of global oil and LNG transits the Strait of Hormuz daily. Even a partial, temporary disruption would constitute a supply shock the market has not adequately priced. The second is a permanent geopolitical risk premium. Even after a ceasefire, the market now knows disruption is possible. Risk premia, once activated, are sticky—think of it as earthquake insurance after an earthquake. The premium goes up and stays up even after the shaking stops. The third is OPEC+ strategic behavior, the most underappreciated dynamic in the whole setup. Saudi Arabia needs roughly $80–90 oil to balance its national budget. The conflict hands Gulf states cover to maintain elevated prices without bearing the political cost of engineering them. They have zero incentive to compensate for war-driven supply disruption by opening their own taps.
The China Variable
No oil thesis is complete without China, and China's role is genuinely two-sided. China is the world's largest oil importer. A recovery driven by stimulus or manufacturing rebound puts significant additional pressure on an already tight market. A Chinese slowdown acts as a natural deflationary counterweight—one of the few forces that could allow Western central banks to look through the energy shock without triggering a wage-price spiral. China is currently caught between both forces simultaneously: targeted stimulus pointing one direction, deep structural headwinds in property and domestic demand pointing the other. The China variable genuinely widens the distribution of outcomes for oil in both directions. It is not a marginal consideration—it is a primary one.
The Bear Case
A rigorous thesis acknowledges where it could be wrong. The genuine bear case for sustained elevated oil rests on three pillars. First, a Chinese demand collapse that overwhelms the supply-side thesis. Second, a diplomatic resolution that specifically includes Iranian sanctions relief—returning one to two million barrels per day of previously suppressed supply relatively quickly. This is the sharpest single counter to everything above and deserves to be taken seriously. Third, high oil prices are themselves recessionary—sustained energy costs above a certain threshold eventually destroy the demand that supports them. For the price level to return to pre-conflict lows however, you would need several of these factors to materialize simultaneously: clean diplomatic resolution including sanctions relief, OPEC+ choosing not to offset Iranian supply, no lasting risk premium in shipping costs, Chinese demand remaining subdued, and no second-round inflation effects already embedded in supply chains. Any single condition failing keeps the structural floor elevated.
The Rate Dilemma
The Fed's March 2026 dot plot projects a single 25 basis point cut this year and another in 2027—with Powell explicitly stating the cut is not guaranteed if inflation progress stalls. J.P.\ Morgan's chief economist has gone further, arguing the Fed holds all year and that the next move is actually a hike in 2027. The 10-year Treasury yield sits at 4.27% and the 30-year at 4.88%—long rates the bond market is keeping elevated independently of Fed guidance. The market is not pricing a cutting cycle. It is pricing a holding cycle with meaningful optionality to the upside. Any portfolio built on the 2025 narrative of rates heading toward neutral is already mispositioned for 2026.
This brings us to the rate dilemma—the intellectual heart of the thesis. Central banks are trained to fight one enemy at a time. Raise rates to kill inflation. Cut rates to stimulate growth. The playbook is clean, the mandate clear. What the Iran conflict has produced is far more uncomfortable: both enemies are present simultaneously, pulling policy in opposite directions, and the traditional tools risk making at least one problem worse regardless of which direction you move. Rate hikes are a demand-side tool. They work by making borrowing expensive, cooling spending, and reducing demand pressure on prices. They do absolutely nothing to fix a supply problem. You cannot drill an oil well by raising the federal funds rate. The Fed is being asked to use the wrong tool for the problem it faces. Hiking into a supply-driven inflation shock does not solve the inflation—it simply adds an economic slowdown on top of it. And yet the Fed cannot ignore inflation either. If expectations become unanchored—if workers demand wage increases to compensate for expected future inflation and companies raise prices preemptively—the supply shock transforms into a self-sustaining wage-price spiral. That is the 1970s scenario. The Fed will tolerate considerable pain to avoid it.
War Financing and Financial Repression
The war financing complication layers a second pressure on top of this. The question is not whether the Fed will explicitly monetize war debt—it almost certainly will not. The question is whether an institution that was already balancing multiple pressures before the conflict can maintain the degree of restrictiveness that pure inflation targeting would require, while a government running six percent of GDP deficits needs to roll and expand its debt at the lowest possible cost. History suggests that tension resolves gradually and quietly in favor of the borrower. Not through dramatic announcements—through a series of individually defensible decisions that in aggregate amount to the same thing. The Fed held rates at zero for years after inflation was clearly building in 2021. It was late to hike and then hiked into a banking crisis in 2023. It has been cutting even as inflation remained above target. None of these are the actions of an institution with an uncompromising mandate. The soft version of war financing pressure is already observable. It does not require a formal yield curve control announcement to matter for asset allocation.
The Fed Put in a Stagflationary Environment
If recession comes—and the combination of high energy costs and restrictive rates makes that increasingly plausible—the conventional assumption is that the Fed cuts, bonds rally, growth stocks recover, the cycle resets. That assumption is built on every recession since 1990—a period in which inflation was either low or falling when recessions hit, giving the Fed complete freedom to cut aggressively. That assumption does not hold in a stagflationary environment. If recession arrives while inflation is still running above target, the Fed faces its worst possible configuration. Cut rates and you risk reigniting inflation and unanchoring expectations. Don't cut and you allow the recession to deepen while political pressure becomes overwhelming. The honest answer is that it depends entirely on where inflation is when the recession arrives—and that conditionality is itself a market event, because it removes something markets have been pricing for thirty years: the Fed put.
The Fed put is the market's belief—built on thirty years of consistent behavior—that the Fed will always cut rates and inject liquidity whenever markets fall sharply. It is not formal policy, but every significant drawdown since 1990 has been met with Fed accommodation: 1998, 2001, 2008, 2018, 2020. A generation of investors has been rationally trained to buy the dip, because the Fed will eventually make it worth their while. This consistent rescue has had a second-order effect: it has compressed the equity risk premium structurally. Investors take more risk than they otherwise would because the downside feels insured. This is partially why equity valuations have re-rated upward since the 1990s—not purely earnings growth, but the compression of risk premia by an implicit guarantee. The put is only available when inflation is low enough to give the Fed freedom of action. In a stagflationary environment, the Fed cannot cut fast enough or deep enough to rescue markets without reigniting the inflation it is trying to control. The put does not disappear—but it becomes delayed, conditional, and insufficient. For markets that have priced thirty years of reliable rescue, even that partial withdrawal is a repricing event.
Real Rates and the Repricing of Capital
The real interest rate—nominal rate minus inflation—is the variable that ultimately determines winners and losers across every asset class. When real rates are positive, capital has a genuine cost, long-duration assets are penalized, and savers are rewarded. When real rates are negative, the entire incentive structure inverts: you are paid to borrow in real terms, cash is a melting ice cube, and hard assets become stores of value because they are priced in the currency being quietly eroded. With the federal funds rate at 3.5–3.75% and PCE inflation at 2.7% and rising, real rates are currently positive—roughly 80 to 100 basis points. Not dramatically restrictive, but genuinely positive. The question is whether that remains true over the next 12 to 24 months, or whether the war financing dynamic gradually erodes it. The investment thesis of this piece is that the scenario where real rates turn negative—financial repression, the Fed blinking under combined war financing and economic pressure—deserves a significantly higher probability weight than current market pricing implies. Not as a certainty, but as a risk that is cheap to hedge and expensive to ignore. The asymmetry matters: the cost of being wrong about the muddling-through scenario when financial repression arrives is far larger than the cost of hedging for repression when muddling through is what we get.
Artificial Intelligence: Deflationary Promise, Inflationary Reality
Before turning to trades, there is one secular force large enough to cut across the entire framework: artificial intelligence. AI is simultaneously the most powerful deflationary force on the medium-term horizon and, within our investment window, a meaningful contributor to the inflationary pressures we have been describing. The deflationary potential is real—at scale, AI automates entire categories of service employment that have been structurally inflationary for decades. But broad AI deployment at the scale required to move macro needles is a 2028 and beyond story. We are in the long middle section between early demonstration and economy-wide impact. Within the 1 to 2 year investment horizon, AI deflation is a risk to monitor, not a force to build the base case around. Meanwhile the inflationary costs of AI are arriving now. Data centers are the fastest growing source of power demand in the United States and Europe, ending fifteen years of essentially flat electricity consumption. Natural gas is the marginal power source filling that gap—it cannot be replaced by intermittent renewables for the baseload requirements of round-the-clock computing. The technology that promises to eventually deflate service costs is simultaneously inflating energy costs today. That timing mismatch has direct investment consequences.
Markets are also pricing AI deflation significantly ahead of the deployment timeline—expecting productivity gains and earnings growth to manifest materially well before 2028. That gap between expectation and economic reality is a volatility generator. The repricing event does not require AI to fail. It only requires it to be slower than priced. In markets, being right about the destination but wrong about the timing is financially indistinguishable from being wrong. This AI expectation volatility is a risk to manage, not a signal to trade around—but it is material, and any equity allocation that ignores it is incomplete.
The Trades
Now to the trades. The organizing principle throughout is simple: own things whose value is anchored in the physical world, be cautious of things whose value depends on assumptions about the future that the macro environment is actively undermining, and treat cash as a conscious allocation decision rather than a residual.
Gold
Gold is the clearest expression of the central thesis. It requires no earnings growth, no management execution, no competitive moat. It is simply a store of value that cannot be printed, in an environment where the pressure to print is building. The investment case rests on three pillars that do not all need to be true simultaneously—any two suffice. First, eroding real rates: as the war financing dynamic gradually compresses the gap between nominal rates and inflation, the opportunity cost of holding gold—which pays no yield—disappears entirely. Second, currency debasement: even moderate, sustained dollar dilution compounds meaningfully over a 24-month horizon, and gold priced in dollars reflects that dilution. Third, tail risk insurance: gold is cheap optionality on the scenarios most destructive to everything else you own. The cost of carrying that insurance when unnecessary is low. The cost of not carrying it when needed is high. Gold's great bull markets have consistently occurred in negative real rate environments—the 1970s, 2002–2011, and 2020–2021. We are not yet there, but we are closer than at any point since 2021, and the direction of travel is clear. What kills this trade: inflation falls rapidly back to target, the Fed holds firm, real rates stay positive and potentially rise. That is the official Fed projection. It is also the scenario that requires the most things to go right simultaneously.
Energy Equities
If gold is the financial repression trade, energy equities are the inflation trade. They own the physical commodity driving the entire macro setup. When oil stays elevated—for the structural reasons established above—energy producers convert that elevated price directly into cash flow, often at margins extraordinary relative to their current valuations. After years of capital discipline enforced by post-2015 investor pressure, energy companies are not drilling aggressively into elevated prices the way they did in 2005–2014. They are returning cash to shareholders. This means the supply response is slower and more muted than historical cycles would suggest—which means the elevated price environment persists longer than traditional mean reversion models imply. Natural gas deserves specific emphasis. The AI data center buildout is a direct and growing source of natural gas demand largely independent of the broader economic cycle. Data centers need power continuously. That demand does not disappear in a recession the way industrial demand does. It is structural, growing, and largely unpriced in natural gas producer valuations at current levels. The energy equity trade sits at the intersection of two of the most powerful forces in the framework simultaneously—oil inflation and AI energy consumption. It works if either force is dominant. It works even better if both are. What kills this trade: Iranian sanctions relief returning 1–2 million barrels per day to market, or a Chinese demand collapse overwhelming the supply-side thesis.
Energy Infrastructure, Nuclear, and Uranium
The most structurally asymmetric trade in the entire framework—and the one most likely to be underowned—is energy infrastructure serving AI data centers: power grid operators, nuclear power developers, and uranium. Every AI outcome requires one input above all others: reliable, continuous, large-scale electrical power. The competition at the application layer is fierce and the winners are genuinely uncertain. The requirement for power infrastructure is not uncertain. It is a near-certainty that compounds regardless of which technology companies win. This is the picks-and-shovels logic applied to the infrastructure layer of AI—the people selling shovels to every miner regardless of who finds gold. Nuclear occupies a specific and increasingly important position within this. Data centers require power that is stable and available around the clock—characteristics that solar and wind cannot reliably provide without storage solutions that do not yet exist at scale. Nuclear provides exactly that profile, it is carbon-free which satisfies hyperscaler sustainability mandates, and it is experiencing a genuine renaissance through reactor restarts, small modular reactor development, and direct power purchase agreements between technology companies and nuclear operators. Uranium sits upstream of this entire dynamic with a structural supply deficit identical to the oil thesis—a decade of underinvestment following Fukushima, with supply pipelines thin precisely as demand is accelerating. What kills this trade: a breakthrough in grid-scale battery storage that makes renewables viable as baseload power, or a broad AI investment pullback driven by expectation disappointment.
Equities with Pricing Power
The equity market is not a single trade in this framework. It is a spectrum, and where you sit on that spectrum matters more than whether you are long or short equities in aggregate. The organizing principle is pricing power—the ability to raise prices at least in line with input cost inflation without losing meaningful volume. In an inflationary environment, pricing power is the single most important determinant of margin preservation. It concentrates in specific characteristics: genuine brand loyalty with limited substitutes, high switching costs, commodity producers who are price-takers on the upside, infrastructure operators with contractually inflation-linked revenues, and dominant platforms where network effects make competition structurally difficult. It is notably absent in highly competitive markets with commoditized products, consumer-facing discretionary businesses where demand is elastic, companies with high labor intensity where wages are rising faster than productivity, and high-multiple growth companies whose valuations are built on future earnings projections that inflation is quietly eroding. This is not a call to be bearish on equities in aggregate. It is a call to be highly selective. The index—market-cap weighted toward the largest technology and consumer companies—is not the right expression of this thesis. The composition matters enormously.
The Dollar
The dollar faces two contradictory pressures throughout the investment horizon. Near-term, rate differentials favor dollar strength—capital flows toward yield, and with the Fed holding at 3.5–3.75% while other major central banks ease, the mechanical force supports the dollar. Longer-term, if war financing eventually forces money creation at scale, the purchasing power of the dollar faces genuine pressure—the force gold is already partially pricing. The most likely sequence is dollar strength through 2026 as rate differentials dominate, then increasing pressure in 2027 as debasement dynamics accumulate. Dollar strength in the near term is therefore an opportunity to build positions in dollar-denominated hard assets at relatively attractive entry points before the debasement pressure arrives. In emerging markets, the distinction between commodity exporters and commodity importers is the most important single variable. Oil exporters and resource-rich economies are partially insulated. Commodity-importing emerging markets face the classic crisis configuration—dollar up, oil up, local currencies under pressure, capital outflows accelerating.
Cash
Finally, cash deserves explicit treatment because the conventional wisdom—that cash is what you hold when you have no view—is wrong in the current environment. At 3.5–3.75% nominal, cash and money market instruments are offering real positive returns for the first time in over a decade. That is a genuine asset class decision. With real rates positive, cash is not melting. The elevated macro uncertainty means the distribution of outcomes is wide, and cash preserves optionality to deploy into whichever scenario materializes more attractively priced. If risk assets reprice downward before the inflation thesis resolves, cash is the asset that lets you buy that dislocation. The alternative cash is explicitly competing with is long duration government bonds. The 10-year at 4.27% and the 30-year at 4.88% look attractive in nominal terms. In a world where inflation stays at 2.7% or above and the Fed cannot cut as aggressively as historical cycles imply, the real return on those instruments is thin and the mark-to-market risk if inflation re-accelerates is substantial. Cash dominates long bonds in the current framework.
Portfolio Summary
To summarize the portfolio logic in plain language: own gold, energy equities, energy infrastructure including nuclear and uranium, equities with genuine pricing power in commodity and infrastructure categories, and cash at current rates. Reduce or avoid long duration government bonds, high-multiple unprofitable growth equities, commodity-importing emerging market currency exposure, and consumer discretionary businesses without pricing leverage. Watch as the key variables: Iranian diplomatic developments as the primary oil thesis risk, Chinese demand trajectory as the inflation wildcard, real rate evolution as the signal for whether the muddling-through scenario is holding or financial repression is arriving, and AI deployment pace as the medium-term challenge to the inflation thesis from below.
Conclusion
Markets are priced for peace—for an orderly disinflation, a cooperative central bank, and a geopolitical shock that resolves without structural consequence. That pricing reflects the world of twelve months ago, not the world of today. The structural supply deficit in oil predates Iran by nearly a decade and cannot be resolved by diplomacy alone. The Fed is deploying the wrong tool against a supply-side problem, constrained by a fiscal position that quietly but persistently tilts incentives away from genuine price stability. The AI revolution—the most powerful deflationary force on the medium-term horizon—is within our investment window an inflationary consumer of the very commodity at the center of the crisis. Against this backdrop, the assets that win are those anchored in the physical world. The thesis has a natural horizon of 12 to 24 months. Beyond that, AI deflation becomes increasingly material, the war financing dynamics either resolve or entrench, and the landscape requires fresh assessment. The window is open. It will not stay open indefinitely.
This represents an analytical framework and investment thesis, not financial advice. All scenarios carry genuine uncertainty. The honest investor acknowledges the counter-arguments, sizes positions accordingly, and maintains the intellectual flexibility to revise when the evidence changes.