When Everything Correlates

If your wealth manager built you a "balanced" portfolio after your exit — some stocks, some bonds, maybe a gold allocation and a slice of private credit — this was the week it all moved in the same direction.

Brent crude topped $112, up more than 40% since the Iran conflict began on February 28. The Nasdaq 100 fell into correction territory, down 10%+ from its peak. The S&P 500 posted its longest weekly losing streak since 2022 and logged its worst single day since the war started, dropping 1.7% on Thursday. The 30-year Treasury yield briefly touched 5%. Gold sold off. Bitcoin sits at roughly half its pre-war peak.

Stocks, bonds, gold — down together. Four weeks of war have stripped away the fiction that most portfolios are diversified.

Three stories this week. The thread connecting them: the assumptions behind how founders structure and deploy capital are being stress-tested from multiple directions at once.

The Week Your Portfolio Lost Its Hedges

Most people think of diversification as holding different things. Some tech stocks, some bonds, a property allocation, maybe a commodities position. Different labels, different asset classes, different risk profiles — at least on paper.

Diversification only works when those assets behave differently under stress. This month, they haven't.

Since the war began, the conventional hedging playbook has broken. Bond prices fell alongside equities, pushing the 10-year Treasury yield to 4.48% — its highest since July. Gold, which should rally when everything else drops, sold off as central banks signalled they'd respond to oil-driven inflation with higher rates. Short-term money market funds and cash equivalents are about the only places providing shelter, and traders are now pricing in zero rate cuts from the Fed for the rest of 2026.

Goldman Sachs Research put numbers to this in a note published this week. Their global portfolio proxy — roughly $300 trillion in financial assets — has declined about 5% since the war started. For a 60/40 portfolio, Christian Mueller-Glissmann (head of asset allocation) called the losses "relatively small" compared to historical drawdowns. Technically correct. Also beside the point for founders who expected their bonds to provide ballast.

Why the hedges failed — and why it matters

Two forces are hitting simultaneously, and both break the traditional stock-bond relationship.

Oil first. Brent passed $112 this week. The Strait of Hormuz — through which roughly one-fifth of the world's oil supply passes — has been effectively disrupted since early March. The IEA described it as the "greatest global energy security challenge in history." Wood Mackenzie's Chairman and Chief Analyst Simon Flowers warned that $200 per barrel is "not outside the realms of possibility in 2026" if the conflict extends. We're not there, but the trajectory matters more than the current level.

Here's the mechanism that breaks portfolios: when oil spikes, inflation expectations rise. When inflation expectations rise, bond yields climb. When bond yields climb, bond prices fall. So the allocation that was supposed to protect equity losses — the bond sleeve in a balanced portfolio — instead amplifies them.

This played out in 2022. The S&P 500 fell 18.1%. The Bloomberg Aggregate Bond Index fell 13%. Morgan Stanley's analysis put the 60/40 portfolio loss at 17.5% — the worst calendar-year performance since the Great Depression. The CAIA Institute's research showed stock-bond correlation flipped from its two-decade negative average to positive territory, with some measures hitting 0.70. That was the first year in over 40 when bonds failed to provide any diversification benefit during a major equity drawdown.

Now it's happening again, different trigger, same mechanism. Oil-driven inflation breaks the negative correlation between stocks and bonds. A "balanced" portfolio becomes a concentrated bet on a single macro variable: that inflation stays low enough for bonds to work as a hedge.

Goldman Sachs' multi-asset co-head Alexandra Wilson-Elizondo flagged the 60/40's declining reliability even before the Iran conflict, noting that "you get more amplification in drawdowns" when stock-bond correlations turn positive. Goldman's asset allocation framework now calls for splitting portfolios roughly into thirds: one-third exposed to innovation (equities, tech, AI), one-third protecting against inflation (real assets, commodities, infrastructure), and one-third for risk mitigation (hedging strategies, quality fixed income, cash). They've been shifting toward alternative risk premia, quantitative strategies, and private market exposures as additional diversification layers.

For founders with $10M-$100M: if your portfolio was constructed primarily from stocks and bonds — the standard output of most post-exit wealth management conversations — this month exposed a structural vulnerability, not just a bad stretch.

We've written before about portfolio construction for founders and the case against the 60/40. The argument was always somewhat abstract: bonds might not provide ballast in a different rate environment. Real assets and alternatives earn their fees in the crisis you haven't experienced yet.

March 2026 makes that argument concrete.

Energy equities went up this month while everything else dropped — Halliburton gained 4% on Thursday alone, Exxon over 3%. International equities in markets less exposed to the Strait of Hormuz disruption have held up better than US large-caps. Real assets — infrastructure, certain real estate, commodities — behave differently because they're tied to physical cash flows that respond to inflation rather than being destroyed by it.

None of this is exotic. It's just not what most founders end up holding after exit. The standard post-exit conversation goes: "Let's build a balanced portfolio with some stocks and bonds, add a bit of alternatives for diversification, and we'll review it quarterly." What "a bit of alternatives" means in practice is often 5-10% — not enough to move the needle when correlations flip.

Goldman's one-third framework is one way to think about it. The specific allocations matter less than the principle: if your portfolio only works in one regime (low inflation, negative stock-bond correlation), it's not diversified. It's a bet on that regime continuing.

For founders sitting in post-exit portfolios, the question this week forces is straightforward: if I remove the labels and just look at how each position would behave with oil at $120 and the 10-year at 4.5%, do I actually have diversification? Or do I have six different names for the same bet?

Most honest answers are uncomfortable.

86% of Family Offices Have No Succession Plan. During a War.

JP Morgan's 2026 Global Family Office Report landed in February, surveying 333 single-family offices across 30 countries with an average net worth of $1.6 billion and average AUM of $1.1 billion.

Then a war started.

A few numbers worth sitting with.

86% of family offices have no succession plan for key decision-makers. That's the most striking finding — and the one that aged worst. Geopolitics was cited as the top risk by 64% of respondents, yet 72% had zero gold exposure and 89% held no cryptocurrency. Families identified the risk. Most didn't position for it. The ones that built governance infrastructure and succession plans before February had a playbook when markets cracked. The 86% who didn't are improvising.

For sub-$100M founders running lean operations, governance often feels like bureaucracy. It isn't. It's the infrastructure you need when things break — and March demonstrated exactly how quickly things break.

Average annual operating cost: $3M. Rising to $6.6M for offices managing $1B+ and dropping to $875K for offices under $250M. At $2.45M for a $500M office, that's roughly 0.5% of AUM before a single investment is made. Eighty percent outsource some aspect of portfolio management. For founders asking whether a family office makes economic sense, JP Morgan's data provides fresh benchmarks — and fresh ammunition for the lean model.

65% plan to prioritise AI investments. 57% have zero exposure to venture capital or growth equity. More than 70% have no infrastructure investments at all, even though AI's entire value chain depends on data centres, energy, and connectivity. JP Morgan's Christophe Aba put it directly: much of AI's future value is still being created in private markets, where the top ten AI companies are already valued at roughly $1.5 trillion.

The gap between ambition and allocation isn't just an institutional curiosity. It's playing out in real time in the pre-IPO market.

Pre-IPO Tech: The $2 Trillion Access Problem

Family offices are racing to position themselves before what could be the largest wave of tech IPOs in history.

SpaceX could file its IPO paperwork as early as next week, targeting a valuation of $1.5-1.75 trillion — potentially the largest public offering ever. OpenAI's latest fundraise expanded beyond $120 billion, positioning it for a Q4 2026 listing that could value the company near $1 trillion. Anthropic, valued at $380 billion after its February round, has hired Wilson Sonsini to prepare for a potential listing. Combined, these three represent more than $2 trillion in private market value.

Institutional players are already in. Australian family office BFA Global invested more than $50 million in SpaceX, reportedly sourced through relationships within Elon Musk's inner circle. According to Crain Currency, the number of families increasing private allocations outnumbers those reducing them by 2.5 to 1.

For founders in the $10M-$100M range, the access question is the story. The Fundrise Innovation Fund (VCX) — one of the few publicly traded vehicles offering exposure to pre-IPO AI names — surged 2,500% in its first week of trading, reaching 16x its net asset value. That's not a rational price. It's a measure of how desperate investors are for any access at all.

Tom Tunguz published an analysis quantifying the structural challenge: at a standard 15% float, SpaceX, OpenAI, and Anthropic would need to raise $432-576 billion from public markets in a single quarter. For context: from 2016 to 2025, the entire US IPO market raised $469 billion. The likely outcome is tiny floats of 3-8%, creating extreme scarcity. S&P Global, FTSE Russell, and Nasdaq are reportedly considering fast-track index inclusion rules that would make roughly $12 trillion in passive assets forced buyers within days of listing.

For founders who aren't institutions and don't have inner-circle access: this is a structural reminder. The biggest wealth-creation events of the decade are happening in private markets, and the gap between what institutions can reach and what sub-$100M individuals can reach is the defining problem of this wealth range. The architecture you build — the relationships, the governance, the advisory team — determines what opportunities you can access.

Paying 16x NAV for a proxy vehicle isn't a strategy. It's a symptom of not having one.

Radar

US bank capital rules loosening. New regulatory proposals could free roughly $60 billion in bank lending capacity. If finalised, this accelerates the structural shift we've been tracking: banks reclaiming leveraged lending market share from private credit. PitchBook data already shows banks' share of buyout financings above $1B recovering from 39% in 2023 to over 50% in 2025. For founders evaluating lending relationships, the competitive landscape is shifting — historically, that means more competitive terms for borrowers.

Entrepreneurship through acquisition is gaining institutional attention. INSEAD's ETA conference runs May 9 in Fontainebleau. UCLA's conference is April 2. Wharton and HBS have expanded their search fund programmes. Stanford's latest data: search funds have produced a mean IRR of 35.1% and average ROI of 4.5x. For founders looking to deploy capital into operating businesses rather than financial instruments, the model is becoming a legitimate institutional asset class. Worth watching whether major family offices start allocating to search fund strategies alongside their PE commitments.

SpaceX IPO filing could come any day. Reports from CNBC and Bloomberg suggest the S-1 could land as early as next week. When it does, it recalibrates the entire 2026 IPO landscape — pricing, timing, and investor appetite for every other listing in the queue.

The assumptions that felt safe six weeks ago — that bonds hedge equities, that a balanced portfolio means a protected one, that you have time to build the governance and access infrastructure — look different after a month of war, a correlation breakdown, and a $2 trillion IPO wave building offshore. The founders who started the structural work early aren't smarter. They just aren't improvising right now.