The One Number Goldman Can't Control — And Why It Matters for Your Portfolio

In April 2026, Goldman Sachs quietly reshuffled its "Long Duration" stock basket — a list of companies whose share prices depend on cash flows that won't materialize for years. The catalyst was the same one that keeps bond traders awake: Treasury yields were climbing back to levels not seen in nearly two decades, and the stocks most exposed to that climb were starting to bleed.

The 30-year Treasury yield broke above 5% that spring, hitting heights last seen in 2007. The 10-year hovered near 4.9%. The Federal Reserve — which sets short-term rates — had its hands tied. Long-term yields are set by markets, and markets were demanding more compensation for lending to a government that carries nearly $40 trillion in debt and paid $857 billion in interest during the first nine months of the fiscal year alone.

This is not a story about Goldman Sachs. It's a story about a number most investors never think about: how far into the future a stock's value lives. And it matters more right now than it has in years.

What "long duration" actually means

Duration is a bond concept that measures how sensitive a security's price is to changes in interest rates. Bonds that pay you far in the future — a 30-year bond, say — swing wildly when rates move because those distant payments are worth less in today's dollars when the discount rate goes up. Short-term bonds barely blink.

The same physics applies to stocks. A company that earns most of its money today — an oil major, a railroad, a defense contractor — has "short duration." Its share price is relatively insulated from what the 10-year Treasury does. A company whose value depends on profits five, ten, or fifteen years from now — a speculative EV maker, an unprofitable biotech, a high-multiple software company — has "long duration." When rates rise, those distant earnings get discounted harder, and the stock drops.

Goldman defines long-duration stocks as those "with expected cash flows concentrated further in the future and typically more sensitive to changes in interest rates". Their basket has historically included companies like Lucid Group, Rivian, Teladoc, and other growth names whose earnings are projected years out.

The rebalance in April was Goldman's admission that the macro regime had shifted and the basket needed to reflect it. But even a rebalance doesn't fix the underlying physics.

The regime has changed — and it may stay changed

For over a decade, investors bought growth stocks with distant earnings on the assumption that interest rates would stay low. That assumption came from a specific economic regime: falling inflation, globalized supply chains, aging demographics, and a Federal Reserve willing to accommodate nearly zero rates.

None of those forces are working the same way.

Inflation has stayed above the Fed's 2% target for five consecutive years. Core PCE — the Fed's preferred gauge — hit 3.3% in April 2026, its highest reading since November 2023. Deglobalization, energy transition costs, fiscal spending, and demographic shifts are structural, not cyclical. The old playbook that assumed inflation would simply drift back to 2% is no longer the consensus.

Meanwhile, the U.S. government is borrowing at an extraordinary scale. A 1 percentage-point increase in interest rates adds roughly $4 trillion in interest costs over the next decade. That debt has to be funded, and it puts structural upward pressure on long-term yields regardless of what the Fed does with its overnight rate.

The market has shifted from pricing in multiple rate cuts to pricing in a potential rate hike. The base case is an extended Fed pause — but the risk is upside, not downside.

In this environment, the discount rate applied to distant earnings is higher. That is arithmetic, not opinion.

Two companies, two durations, two outcomes

The contrast between long and short duration is not theoretical. It shows up in the returns.

Consider Lucid Group, one of the growth names Goldman has kept in its long-duration basket. Lucid makes luxury electric vehicles, operates at a massive loss, and burns through cash at a staggering rate. Over the trailing twelve months, Lucid generated negative $5.1 billion in free cash flow — and that burn rate worsened 68% year over year. The company carries $5.9 billion in total debt against $1.8 billion in equity, for a debt-to-equity ratio of 1.76. Its stock is down 60% year-to-date and 78% on a rolling annual basis. It pays no dividend, earns no profit, and has no pricing power in a sector where competition is intensifying and margins are compressing.

Now look at Exxon Mobil. Exxon extracts and refines oil and gas — the kind of business the economy runs on regardless of interest rates. It generates $30.6 billion in free cash flow over the trailing twelve months, pays a 2.5% dividend that it has grown for 23 consecutive years, and carries a net debt of just $31.8 billion against $266 billion in equity — a debt-to-equity ratio of 0.16. Its stock is up 38% year-to-date.

The performance gap is not luck. It is duration. Exxon's earnings are here, now, funded by tangible assets and pricing power in a commodity the world cannot function without. Lucid's value is a promise, priced at a multiple that depends on interest rates cooperating. They aren't.

Why this matters for your portfolio

You don't need to be a bond trader to understand duration risk in stocks. The question is simple: when you buy a company, are you paying for cash flows that exist today or for ones you hope will exist a decade from now?

In a low-rate world, that distinction mattered less. Distant earnings discounted at 2% looked almost as valuable as near-term earnings. At 5%, they look like a fraction.

This doesn't mean growth stocks are dead. It means you need to distinguish between companies whose growth is funded by real cash flow and companies whose growth is priced into the stock price ahead of the money ever arriving. Goldman's own "Rule of 10" screen — identifying S&P 500 companies with at least 10% annual sales growth expected through 2028 — includes 35 stocks. Some of them generate cash today and compound from a foundation. Others are trading at premiums that require everything to go right and rates to come down.

The leading indicator for this call isn't GDP — GDP tells you what already happened. It's the shape of the yield curve, the trajectory of ISM new orders, and the Federal Reserve's own assessment of inflation persistence. Manufacturing is expanding — the August ISM PMI came in at 54.6, marking the eighth consecutive month above the 50-line that separates growth from contraction. New orders softened slightly to 53.7, suggesting the manufacturing engine is steady, not accelerating. The economy isn't booming enough to justify falling rates, and it isn't breaking enough to justify cutting them.

The real question isn't which basket Goldman rebalanced

The real question is what you're paying for when you buy a stock. In a regime where the 30-year Treasury yields 5%, where inflation is structurally higher than it was ten years ago, and where the government's interest payments have exceeded federal spending on Medicare and the military over the first nine months of the fiscal year — distant earnings carry more risk than they used to.

That doesn't mean you abandon growth. It means you demand that growth be real, funded, and priced at a multiple that doesn't require perpetual cheap money. You look for pricing power — the ability to raise prices without losing customers — because that's the mechanism that lets a company grow through inflation. You check the balance sheet, because leverage that was tolerable at 2% becomes dangerous at 5%.

And you understand that the same interest rate that makes a 30-year bond fall is the same one that compresses the valuation of any stock whose value lives in the future. The duration doesn't care whether the security is called a bond or a growth stock. It's math, and it doesn't negotiate.