"History doesn't repeat… but it rhymes." — Mark Twain
◉ THE PRESENT
The consensus forecast for September CPI stands at 3.7%, up from 3.4% in August and right on the street estimate. Core is expected to hold at 2.4%. The entire projected jump is gasoline and heating oil — the kind of inflation that happens when tankers are dodging missiles in the Strait of Hormuz and Brent sits above $106. The S&P 500 opened at 7,798, less than half a percent from its all-time high, and the 10-year yield is parked above 5.2%. The last time a CPI print this hot landed this close to a market top, the month was March and the year was 2000.
Sep CPI 3.7% YoY (est.) | Core 2.4% (est.) | MoM +0.6% (est.) | S&P 500 7,798 | 10Y 5.24% | Brent $106 | Fed 3.75–4.00%
◉ THE ECHO — AUGUST 25, 1987
The Market Crashed First. Then the Inflation Number Landed.
The S&P 500 touched 1,552.87 on a Friday in late March. It was the kind of all-time high that didn't feel like one — no champagne, no cable-news countdown clocks. The economy was running at full speed. Unemployment sat at 4.1%. GDP growth topped 4%. Greenspan's Federal Reserve had been raising rates since June 1999, pushing the federal funds rate from 4.75% all the way to 6.00% by March 21st, and the market hadn't flinched. Higher rates, in this environment, were treated as confirmation that everything was fine.
Oil was the thing nobody was watching. WTI crude had tripled in eighteen months, climbing from roughly $11 a barrel at the end of 1998 to $34 by early March 2000. OPEC had been cutting production through 1999, methodically draining a market that had been swimming in cheap crude during the Asian financial crisis. Gasoline hit $1.52 a gallon in March, up fifty-three cents from the year before. The energy component of the Consumer Price Index would finish the year up 14.2%. None of this made the front page of the Wall Street Journal's markets section. Tech stocks were the only story that mattered.
Then came Friday, April 14th. The NASDAQ dropped 10% in a single session, closing at 3,321 — it had been above 5,000 five weeks earlier. For the week, the index lost 25%, the worst stretch in its history. The Dow shed 616 points, its largest single-day point loss ever at the time. Trading floors were quiet by the close, the stunned kind of quiet where people stare at screens and wait for someone to explain what just happened.
Three days later, on Monday morning, the BLS released the March Consumer Price Index. Headline: 3.8% year over year, the hottest reading since 1991. Month over month: 0.7%. And the core number — the one that strips out food and energy — came in at 2.4%. Almost exactly where it had been for months. The economy wasn't broadly overheating. Gasoline was. The distinction should have changed everything. It changed nothing. Greenspan looked at 3.8% and hiked one more time on May 16th, pushing rates to 6.50%. It was the last increase of the cycle. Eight months later he was cutting — fifty basis points on January 3rd, 2001, followed by another fifty on January 31st. By then the S&P 500 was already in a decline that wouldn't end until October 9, 2002, when it hit 776.76. Down 49% from the March 2000 peak.
◉ THE RHYME — WHAT'S IDENTICAL
Same CPI gap. Same energy illusion. Core reads 2.4% both times, but headline screams something different — and the Fed listens to the wrong number.
◉ THE DIVERGENCE — WHAT'S DIFFERENT THIS TIME
The oil shock has a different source. In 2000, crude tripled because OPEC squeezed supply during a demand recovery — a deliberate policy choice that could be reversed with a phone call to Riyadh. In 2026, Brent surged past $100 on actual conflict: Houthi missile strikes on Saudi airports, Iranian tanker disruptions in the Strait of Hormuz, and a war premium that won't disappear because a cartel votes to open the taps. War-driven supply shocks carry fatter tail risk.
Rate levels are far apart. Greenspan took the funds rate to 6.00% before the market cracked and 6.50% after. Warsh is at 3.75–4.00%. In real terms — the rate minus headline CPI — the Fed in March 2000 was running at roughly +2.2%. Today it's barely positive. Less real tightness could delay the economic slowdown, or it could simply let the imbalances build longer before they snap.
Tech earnings are real this time. The NASDAQ in March 2000 was packed with companies that had never turned a profit and never would. The AI trade in 2026 sits on top of Nvidia, Microsoft, and Alphabet — firms printing tens of billions in free cash flow. That doesn't make them immune to a repricing, but it means any decline is likely slower and shallower than the 78% NASDAQ wipeout of 2000–2002.
The consumer is weaker now. Unemployment was 4.1% in March 2000 and household balance sheets were clean. In October 2026, Michigan consumer sentiment sits at 46.3 — second-lowest ever recorded — and credit card delinquencies have been climbing for six quarters. The 2000 economy had a cushion to absorb rate hikes. This one is running on fumes.
◉ THE RECKONING — WHAT HAPPENS NEXT
Here is what happened after that 3.8% CPI print landed on April 17, 2000.
The market bounced. That's the part everybody forgets. After the April 14th bloodbath, the S&P 500 spent the summer clawing back. By late August it was trading near 1,500 — within 3% of its all-time high. Greenspan's final hike to 6.50% on May 16th had come and gone without breaking anything that showed up in the monthly data. GDP was still positive. Unemployment barely moved. If you had sold in April, you spent the summer watching your old positions recover and questioning every decision you'd made.
Then it turned. Corporate earnings started missing in the fourth quarter. The NASDAQ, which never recovered its March high, resumed its slide. The recession that officially started in March 2001 had been quietly building for months, hidden inside the same data that looked fine on the surface. The S&P 500 rolled over for good and kept falling until it hit 776.76 on October 9, 2002. The people who sold during the summer bounce — when the market felt safe again — were the ones who kept their money.
The signal was in the CPI all along. When headline inflation runs 140 basis points above core, and the gap is energy, and the Fed keeps hiking anyway, the economy is getting crushed from both ends. Consumers pay more at the pump. The Fed makes borrowing more expensive. Something breaks — usually the labor market — but never right away. It takes months. In 2000 it took eight months from the last hike to the first cut. The market didn't bottom for thirty.
The edge: In 2000, the market gave you a five-month window between the hot CPI and the final top. The smart money didn't sell on April 14th. They sold during the summer bounce, when everyone else decided the correction was over. If the 2000 playbook runs in 2026, the question isn't whether the S&P can hold 7,800. It's what happens after it does.
◉ TOMORROW’S WATCH
Thursday's September retail sales report. If consumers pulled back while gasoline prices surged, it mirrors July 2000, when real spending growth flatlined just as Greenspan delivered his final hike — the first crack that nobody noticed until the recession was already eight months old.
