Interest Rates and the Cost of Capital — The single variable that determines whether your valuation holds or collapses

In 2022, the fastest Federal Reserve hiking cycle in 40 years compressed the S&P 500 P/E ratio from roughly 24x to 17x — not because earnings fell, but because the discount rate applied to future earnings normalized from near zero to 4%-plus, and valuations built on ZIRP-era assumptions had nowhere to hide.

That compression was not a market surprise. It was arithmetic. Every discounted cash flow model ever built has an interest rate assumption embedded in it, and for a decade that assumption was historically extreme. When it reverted, so did every valuation anchored to it.

Interest rates are not background noise. They are not a macroeconomic variable that belongs in a separate "macro" column away from the "real" analysis. They are the single most influential input in any intrinsic value calculation, and most investors treat them as a number to fill in after the work is done. That habit is expensive.

---

The concept in 60 seconds

A business is worth the present value of the cash it will generate over its life. To calculate present value, you need a discount rate — a percentage that converts future cash into today's dollars. The higher the rate, the less future cash is worth today. A company expected to earn $100M in year 10 is worth materially less when discounted at 10% than when discounted at 6%. The company has not changed. The rate has.

That discount rate is not arbitrary. It is built from observable market prices: the current risk-free rate (10-year US Treasury yield), the equity risk premium investors demand above that floor, and an adjustment for company-specific risk. These components combine into the WACC — Weighted Average Cost of Capital — which also blends in the proportion of debt versus equity funding, since debt (being tax-deductible) is cheaper than equity.

When rates rise, WACC rises. When WACC rises, every future dollar is worth less today. The valuation falls — not because anything about the business changed, but because the clock that measures time-value-of-money has been reset.

---

Mental model

Think of a long-duration business as a bond with no maturity date and uncertain coupons. The more distant the cash flows, the more the price is governed by the discount rate applied to the far-out years rather than by the next twelve months of earnings.

This is the duration problem. A mature utility or consumer staples company earns most of its distributable value in the near term — years 1 through 7. A growth-phase software company with near-term losses and promised profitability in years 8 through 20 is a long-dated instrument. For the utility, a 1% rise in discount rates is a moderate headwind. For the software company, the same 1% rise can eliminate a third of the calculated intrinsic value, because that third was sitting in cash flows 15 years out, now discounted much harder.

The third channel that makes this concrete is terminal value. In a standard DCF, 60–80% of the calculated intrinsic value typically lives in the terminal value — the lump-sum estimate of everything the business is worth beyond the explicit forecast period. The terminal value is computed by dividing terminal-year cash flow by the difference between the discount rate and the terminal growth rate. A narrow spread is brutally sensitive to rate changes. A company valued at a 10x terminal multiple at a 7% discount rate becomes an 8x terminal multiple business at 9%. The business has not changed. The math has.

The mental model to carry: duration is not just a bond concept. Every business has a duration profile. High-multiple, long-runway businesses are long-duration equity instruments, and they behave like them when rates move.

---

Worked example: Zoom Video Communications — the 2022 discount rate collapse

Zoom Video Communications (ZM) is the cleanest equity duration case in recent history.

Zoom IPO'd in April 2019 at $68 per share. By October 2020, the stock had reached $568 — an 8x rise in 18 months. The driver was partly genuine: Zoom's revenue exploded from $623M in FY2020 to $4.1B in FY2022. The business was real, the growth was real, and the adoption curve was exceptional by any measure.

But the valuation had a second engine running beneath the revenue growth: the risk-free rate in the United States had fallen to 0.5–0.7% on the 10-year Treasury. At those rates, a discount rate of 6–7% applied to Zoom's projected cash flows produced intrinsic value estimates that could plausibly support $400–$500+ per share, because the terminal value — anchored to cash flows a decade away — was enormous when discounted at near-zero base rates.

Revenue kept growing through 2021 and into 2022. By February 2022, Zoom had delivered $4.1B in revenue, its highest-ever annual figure. By late 2022, the stock was trading near $65.

The business did not deteriorate in proportion to the stock price. What changed was the risk-free rate. The 10-year Treasury moved from 0.7% in 2020 to 4.2% by late 2022. The equity discount rate applied to Zoom's future cash flows normalized from roughly 6–7% to 9–10%. Terminal value — which had been doing the heavy lifting at the peak multiple — collapsed under the new math.

This is the clearest available example of equity duration risk. A company whose revenue more than tripled over two years lost 88% of its peak market value. The operational story was not the culprit. The discount rate was.

The correct lesson is not that Zoom was a bad business. It is that the 2020 valuation was built on a rate assumption that was historically extreme, and no business is immune to the arithmetic of a rate normalization when its value is concentrated in distant cash flows.

---

Historical pattern

2010–2021: The ZIRP era and what it did to growth valuations

From approximately 2010 to 2021, US and European central banks maintained interest rates at or near historical lows. The Federal Funds Rate sat near zero for most of this period, interrupted only by a brief hiking cycle in 2015–2018 that peaked at 2.5% before reversing. During 2020, the 10-year Treasury yield touched 0.5% — a level not seen in the 150-year recorded history of US bond markets.

At near-zero risk-free rates, discount rates collapse. A discount rate of 5–6% applied to a long-duration growth company produces a valuation roughly double that produced by a 9–10% discount rate on the same projected cash flows. Price-to-earnings and price-to-free-cash-flow multiples that looked excessive by historical standards were, in a narrow mathematical sense, defensible under ZIRP conditions. This matters for understanding the period clearly: the bubble was not purely irrational exuberance. It was partly rational arithmetic applied to an irrational rate environment.

2022: The fastest hiking cycle in 40 years

The Federal Reserve raised the Federal Funds Rate from near zero to 5.25% in less than two years — the fastest tightening cycle since the early 1980s. The 10-year Treasury moved from approximately 1.5% at the start of 2022 to over 4% by year-end.

The S&P 500 P/E ratio compressed from roughly 24x earnings at end-2021 to approximately 17x by end-2022. Corporate earnings grew during this period — they were not the source of the drawdown. The compression was almost entirely a valuation re-rating: the same future earnings stream, discounted at higher rates. For long-duration growth equities, the compression was far more severe.

The structural shift to watch

The relevant question is not whether rates return to ZIRP. It is whether your models are built to work across a range of rate environments, rather than being calibrated to the decade-long anomaly. Rate cycles of 20–30 years are normal across financial history. The ZIRP decade was the outlier.

---

Decision framework

This framework is designed for active use during company research — not for macro forecasting.

At the screening stage: identify where rate exposure sits in the business model.

• Is the company a bank, insurer, or leasing business with direct net interest margin exposure? Rising rates compress margins in the short run; understand which direction the company is exposed at current yield curve shape.
• Does the company carry large near-term debt maturities? Pull the maturity schedule. For a leveraged business, model the interest expense at current refinancing rates versus the coupon on outstanding debt. A $5B debt load refinancing from 3% to 6% is a scheduled $150M annual earnings headwind — not a forecast, a calendar event.
• Is the valuation high-multiple and dependent on cash flows in years 7–15? If so, the terminal value is load-bearing and rate-sensitive. Apply extra scrutiny before entering.

At the valuation stage: make the rate assumption visible and stress-test it.

• Use the current 10-year Treasury yield as the risk-free rate. Do not use a multi-year average anchored to a different regime.
• State the equity risk premium explicitly. If the 10-year is at 4.5% and the WACC is 7%, the implied ERP is 2.5% — know whether that is intentional.
• Run a sensitivity table: what is intrinsic value at WACC +100bps and +200bps? If the thesis evaporates at +100bps, that is a finding, not a footnote.
• Identify where in the DCF the value is concentrated. If more than 70% sits in the terminal value, mark that explicitly and understand what rate assumption the terminal multiple implies.

At the monitoring stage: treat rate moves as events.

• For any long-duration or leveraged holding, add the 10-year Treasury yield to the event-trigger list.
• A 50bps move in either direction on the 10-year is a scheduled model review, not necessarily an action. Re-run the sensitivity. If assumptions remain valid, note it and move on. If they do not, that is the signal.

---

Common mistakes

Mistake 1: Treating ZIRP-era discount rates as the permanent baseline

SoftBank's Vision Fund valued WeWork at $47 billion in January 2019. To produce a positive NPV on that valuation required cash flow projections and discount rate assumptions that were only defensible under near-zero rate conditions. The model was not wrong in isolation — it was wrong in a normalized rate world. WeWork filed for bankruptcy in November 2023. The gap between the 2019 peak valuation and the eventual outcome was not entirely a WeWork-specific story; it was substantially a story about what happens when permanent-ZIRP assumptions meet reality. SoftBank's Vision Fund wrote off billions across multiple holdings that shared the same structural flaw: terminal values built to work at 4–5% discount rates, not 8–9%.

Mistake 2: Ignoring the debt refinancing schedule

Signa Group, the Austrian real estate and retail conglomerate, accumulated significant debt during 2020–2021 at historically cheap rates — a mix of floating-rate instruments and short-duration fixed debt. The strategy was not unusual for the period; cheap money made leverage attractive across the sector. When rates rose rapidly in 2022–2023, refinancing costs tripled on the portions of the debt stack coming due. Signa filed for insolvency in November 2023, in one of the largest European real estate collapses in decades. The underlying properties did not disappear. The business model — which depended on the ability to refinance at manageable rates — became unworkable. Rate refinancing risk was a visible, calculable, scheduled problem. It was not stress-tested.

Mistake 3: Treating terminal value as a residual rather than the primary driver

Through 2021, the dominant debate about high-multiple technology companies centered on revenue growth rates and gross margin trajectory — the near-term operational story. Terminal value was a plug at the end of the model. This framing was backwards. For any company trading at 20x+ revenue, the terminal value was providing 70–80% of the modeled intrinsic value, and the discount rate applied to that terminal value was the most sensitive variable in the entire analysis. When the 2022 multiple compression arrived, it was almost entirely a terminal value problem: the same businesses, the same near-term revenue trajectories, but a terminal multiple that contracted 30–40% as rates normalized. Analysts who had built conviction around the operational story found themselves holding businesses whose operational story was intact but whose prices had halved.

---

How VI Stack uses this

Rate analysis enters the research process at two fixed points. At Gate 3 (The Forensics), the current 10-year Treasury yield is a direct input to every WACC calculation — not a historical average, not a multi-year mean-reversion estimate. The debt maturity schedule is reviewed as a standard part of balance sheet analysis, with refinancing cost modeled at current rates. The terminal multiple is anchored to the current rate environment. Every Gate 3 valuation includes a sensitivity table showing intrinsic value at WACC +100bps and +200bps.

In Block 4 (The Watch), the 10-year Treasury yield sits on the event-trigger list for any long-duration or leveraged holding. A 50bps move initiates a scheduled model review — not an automatic portfolio action, but a deliberate re-examination of whether the assumptions underlying the position still hold.

---

What's next

Macro 03 covers inflation regimes and real returns — how inflation erodes purchasing power, how different asset classes perform across high-inflation and low-inflation periods, and how to adjust valuation thinking when nominal growth rates are driven partly by price increases rather than volume. It is the natural follow-on to this module, because the interaction between rate levels and inflation expectations is where monetary policy and equity valuation connect most directly.