Policy cycles × sector rotation
2010-2026: what the Fed did, and who led
For sixteen years, US sector leadership has rotated almost every time the Fed changed its policy mix: the three QEs each ran a different script, hikes come in growth-driven and inflation-driven flavors with opposite outcomes, and runoff and cuts each left a signature.
This page divides the period since 2010 into 13 policy eras and quantifies each era's leaders and laggards using dividend-adjusted total returns of the sector ETFs, with a mechanism read for every era. Methodology: era returns run from the last close before the era starts to its end (the live era updates with the latest session); eras before XLRE's 2015-10 and XLC's 2018-06 listings are honestly left blank.
QQQ weekly close (split-adjusted, log scale), data through 2026-07-24. Click a band to inspect the era.
2026 hold + new Chair
2026-01-01 → todaySPY +9% · QQQ +12%total returnFour straight holds at 3.50-3.75% (Jan/Mar/Apr/Jun); $40B/mo reserve-management bill buying; Warsh becomes Chair May 22
The hold in progress: energy has taken the baton at +35%, hard assets and industrials strengthening while communications and discretionary turned negative.
Sector leadership by era
Each column ranks the era's sectors best to worst; colors group sectors by behavior family. Click a sector to trace it across every era.
Values are era total return, dividend-adjusted. Same basis as the era analyses below.
See where rotation stands today on the live industry rotation page with live RRG and KOVA scores.
Which stretch of history does today resemble?
nearest 3 of 196 candidate months since 2010-01The candidate pool leaves out the current month itself (it is what the others are measured against) and the 2 months on either side of it: neighboring months share most of the same three-month window, so they always look like the closest match without telling you anything.
This is a historical analogy, not a forecast. The months below are the closest macro matches to the current reading; the table shows what each sector actually did in the 3 months that followed each of them. Three cases is a very small sample, and similar macro conditions have produced opposite outcomes before.
2 of 8 dimensions read "—" this month (the data is not published yet), so they take no part in the match either way.
Green outline marks a dimension in the same bucket as today; plain outline marks a different bucket; dashed marks no data on one side.
What the 3 months after each of those looked like
Compounded total return over the three months following each match, dividend-adjusted. Rows keep the fixed sector-family order and are deliberately NOT ranked: a ranking built on three cases reads as a finding when it is noise.
| Sector | 2018-042018-05 → 2018-07 | 2025-012025-02 → 2025-04 | 2022-012022-02 → 2022-04 |
|---|---|---|---|
| SPY | +6.8% | -7.6% | -8.1% |
| Tech | +8.7% | -8.9% | -12.5% |
| Comm | — | -6.5% | -19.9% |
| Disc | +7.6% | -14.8% | -11.8% |
| Fin | +2.3% | -4.9% | -11.3% |
| Indus | +6.9% | -4.9% | -5.2% |
| Hlth | +8.5% | -4.0% | -0.4% |
| Stapl | +7.0% | +4.2% | +2.7% |
| Enrgy | +5.2% | -7.4% | +15.1% |
| Util | +3.3% | +2.0% | +3.6% |
| Matls | +5.3% | -5.1% | +1.1% |
| REIT | +7.9% | +0.4% | -1.0% |
A cell reads "—" when that sector had not listed yet (XLRE before Oct 2015, XLC before Jun 2018) or when the three-month window runs past the end of the data. Nothing is back-filled with a proxy.
Regime slicer
months matched198/ 198Pick measurable macro conditions instead of a policy label, then read what each sector actually did in the months that met them. Every dimension is a bucket computed from the readings themselves; a month whose reading is missing on a selected dimension does not count as a match.
This is an exploration tool. Eight dimensions produce hundreds of reachable slices, and the more of them you try, the easier it is to land on a coincidence that looks like a pattern.
| Sector | Avg monthly | Up-month share | vs SPY | Months used |
|---|---|---|---|---|
| SPY | +1.20% | 68%(135/198) | — | 198 |
| 1Tech | +1.68% | 63%(125/198) | +0.48% | 198 |
| 2Disc | +1.29% | 63%(124/198) | +0.09% | 198 |
| 3Indus | +1.25% | 63%(124/198) | +0.05% | 198 |
| 4Fin | +1.06% | 58%(115/198) | -0.14% | 198 |
| 5Hlth | +1.05% | 63%(124/198) | -0.15% | 198 |
| 6Comm | +1.04% | 61%(59/96) | -0.27% | 96 |
| 7Util | +0.91% | 65%(129/198) | -0.29% | 198 |
| 8Matls | +0.90% | 58%(115/198) | -0.30% | 198 |
| 9Enrgy | +0.88% | 56%(110/198) | -0.32% | 198 |
| 10Stapl | +0.87% | 59%(117/198) | -0.34% | 198 |
| 11REIT | +0.69% | 59%(76/128) | -0.55% | 128 |
Averages are the simple mean of monthly total returns (dividend-adjusted) over the matching months, not a compounded path. vs SPY compares each sector against SPY over the same months that sector actually traded, so a late-listing sector is never measured against years it did not exist.
Ranked sectors with fewer months than n did not exist for the whole slice: Comm 96/198, REIT 128/198. They clear the 12-month floor, so they are ranked, but compare them on vs SPY rather than on the raw average.
The dataset's last month (2026-07) is still in progress and is excluded from every slice: a half-month return averaged together with full months is a basis mismatch. The slice pool is 198 completed months.
Today's forward read
in-house live data · not the history aboveDifferent source, different clock. Everything above is 2010-2026 dividend-adjusted return for the 11 SPDR sector ETFs — that era has no KOVA scores, because the scoring system is recent. This block is TODAY's live in-house snapshot: every scored stock in the market, grouped by our own industry taxonomy (30 IBD sectors), which is NOT the 11 SPDR sectors above. Read the two as separate questions — what paid off historically, versus where the scores cluster right now — never as one series.
Loading today's snapshot…
KOVA median = the median score of that sector's scored constituents. >90 = share of them scoring above 90. 2-way EPS = share whose EPS is growing both year-over-year AND quarter-over-quarter, counted only over names where both legs are available. Sorted by KOVA median; a missing figure renders as “—”, never as a zero.
Post-crisis ZIRP
2010-01-01 → 2010-11-020-0.25% rates; QE1 ends Mar 2010; Jackson Hole hints at more easing (Aug)
Early-recovery beta: consumer discretionary and industrials led while financials and health care lagged.
QE1 ended in March 2010 with rates pinned at zero. The moment purchases stopped, the market delivered the flash crash and a 16% drawdown into summer: the first ladder-removal test. Bernanke hinted at more easing at Jackson Hole in August and the tape repaired.
Sector structure was classic early recovery: discretionary +21% and industrials +18% led (restocking plus a consumer rebound), QQQ (+16%) ran well ahead of the S&P (+9%); financials made just +2% under a bad-debt overhang and health care +3% under reform uncertainty.
TakeawayOwn beta and cyclicals early in a recovery, and do not fight the exit-taper tape: when QE stops, the market forces the Fed to speak.
QE2
2010-11-03 → 2011-06-30$600B Treasury purchases announced Nov 3, 2010, completed Jun 2011
QE2 was a commodity reflation trade: energy ran away with it while growth actually lagged.
On Nov 3, 2010 the Fed announced $600B of Treasury purchases. The liquidity did not flow to growth stocks: the dollar weakened, oil ran to $110, and inflation expectations jumped.
Energy led by a wide margin at +27%, with industrials +16% and health care +15% behind; QQQ (+9%) lagged the S&P (+12%) for the whole era. Same medicine, different soil: QE2 landed in a year of closing output gaps and booming emerging markets, so it became a hard-asset bull.
TakeawayDo not treat QE = buy tech as a formula. Ask what macro soil the easing lands in: when inflation expectations jump, hard assets are QE's beneficiaries.
Twist + euro crisis
2011-07-01 → 2012-09-12Operation Twist (Sep 2011); US downgrade + euro debt stress; rates on hold at zero
Twist pressed the long end: a quality-growth plus defensives barbell, with financials and materials dragged by Europe and China.
August 2011 brought the US downgrade and a full euro-debt flare-up; the Fed answered with Operation Twist (sell short, buy long) to press the long end, and extended the zero-rate pledge. Duration repriced.
Technology +23% (the iPhone-supercycle Apple era) paired with staples +17% and utilities +13% in a quality-growth plus defensives barbell; financials +4% were hostage to European banks and materials −4% reflected China slowing. QQQ's +21% vs the S&P's +12% was almost entirely the duration re-rating.
TakeawayWhen policy presses long rates lower, far-cash-flow assets (quality growth) and bond-like assets (defensives) can outperform together: the barbell is not a contradiction.
QE3 open-ended
2012-09-13 → 2014-10-29Open-ended MBS+Treasury buying; taper tantrum May 2013; taper Dec 2013 → ends Oct 2014
The open-ended bull: health care led at +73%, financials and industrials followed, energy trailed.
QE3 was the only open-ended round: monthly buying with no fixed total. The May 2013 taper tantrum hit, but the rally resumed because the economy was genuinely improving.
The S&P returned +44% for the era; health care +73% led (a biotech innovation cycle plus post-reform certainty), financials +55% (housing recovery, steeper curve), industrials +54%; energy trailed at +21% as shale supply capped oil. A textbook case of fundamentals catching the liquidity: leadership went to sectors with their own cycle, not to mere liquidity beneficiaries.
TakeawayIn mid-to-late QE, ask who has their own story: liquidity is the ticket in, but sector-level cycles (biotech and housing back then) decide who leads.
Zero-rate wait
2014-10-30 → 2015-12-15QE over, rates still zero; strong dollar + oil collapse; liftoff telegraphed all year
The strong-dollar era after QE: oil halved and killed the commodity chain while consumer and growth stayed clean.
Purchases hit zero in Oct 2014 and hike expectations built. The dollar index rose 25% in a year and oil fell from $100 to $35. Not a single hike had happened, yet the commodity chain was already executed by the dollar: energy −26%, materials −6%, industrials −1%.
On the other side, cheap oil was a consumer tax cut: discretionary +19% led and QQQ's +14% ran far ahead of the S&P's +6%. The Aug 2015 yuan devaluation added a global volatility shock mid-era. The era's driver was not the Fed's hand but its mouth: forward guidance alone did the tightening.
TakeawayMost of the pricing happens in the expectation phase before the policy turn: watch the dollar and commodities; they move faster than the hikes themselves.
Slow hikes 0.25→1.25%
2015-12-16 → 2017-09-30Liftoff Dec 2015; only one hike in 2016; gradual pace, no balance-sheet action yet
Slow hikes met a global recovery: technology led at +41% with financials strongly outperforming; hiking was no bear-market switch.
After the Dec 2015 liftoff the Fed managed only one more hike in 2016: but global manufacturing synchronized into recovery in late 2016 and the post-election Trump trade lit reflation expectations.
Technology led at +41%, with industrials +40%, financials +38% (about 10 points over the S&P's +28%, on margin repair and deregulation hopes) and materials +36% broadly strong. The era falsified hikes = sell stocks: when hikes happen because growth accelerates, growth, cyclicals and financials win together.
TakeawayWhy the Fed hikes matters more than whether it hikes: growth-driven hikes reward financials and cyclicals; inflation-panic hikes are what compress everything.
QT1 + faster hikes
2017-10-01 → 2018-12-24Balance-sheet runoff starts Oct 2017; four hikes in 2018 to 2.25-2.5%; Q4 2018 crash
The price of double tightening: hikes plus runoff left only defensives standing, ending in the Q4 2018 crash.
Runoff started Oct 2017, four more hikes landed in 2018; Powell called rates far from neutral in October, then called QT on autopilot at the Dec 19 press conference, and three sessions later the S&P completed a near-20% drawdown into the Christmas Eve low.
Only defense survived the era: discretionary +4% (Amazon alone), utilities +2%, health care +1%; industrials −13% and materials −15% (trade war), financials −12% (flattening curve), energy −18%. Markets can absorb hikes; hikes plus runoff drain liquidity twice over: QT's first lesson.
TakeawaySingle tightening trades on fundamentals; double tightening trades on liquidity: when hikes and runoff stack, hold defensives as ballast and do not bet against the drain.
Pivot: cuts + QT end
2018-12-25 → 2020-02-19Powell pivot Jan 2019; three cuts (Jul/Sep/Oct); QT ends Aug 1, 2019; repo bill-buying from Oct
The Powell-pivot melt-up: technology's epic +81%, with even utilities up +40%.
A single word, patient, completed the pivot on Jan 4, 2019. Three insurance cuts followed (Jul/Sep/Oct), QT ended Aug 1 (announced Jul 31, two months early), and bill purchases began in October to repair the repo market. Liquidity flipped from drain to pump.
A falling discount rate lifted both ends of the duration spectrum: technology +81% led while bond-proxies like utilities +40% and real estate +43% also surged: the cuts-era signature of growth and defensives rising together. Energy +9% finished last; shale oversupply kept it out of the party.
TakeawayCuts without a recession are melt-up fuel; the era's laggard (energy then) usually has its own supply problem: do not bottom-fish cheap.
COVID: 0% + unlimited QE
2020-02-20 → 2022-03-15Rates to 0 (Mar 2020); unlimited QE Mar 23; taper announced Nov 2021, ends Mar 2022
Unlimited QE in two acts: a growth mania, then the inflation trade: energy +46% edged technology +45% over the full window.
Two emergency cuts to zero plus unlimited QE on Mar 23, 2020, compounded by fiscal transfers: the most violent policy combination on record. Act one (2020) was the stay-home growth mania; act two (2021) was inflation waking up and commodities taking the baton.
Measured from the pre-crash high (Feb 20, 2020) to the eve of liftoff (Mar 15, 2022): energy +46% edged technology +45% at the top, with materials +42% third; utilities +8% and communications +14% trailed. One window holding two leaderships shows unlimited liquidity does not pick sectors: it lifts the longest duration first, then hands the baton to hard assets once inflation shows up.
TakeawayRead extreme easing in acts: own duration (growth) in act one, rotate to hard assets when inflation prints: the gear change between acts is the single most valuable decision of the era.
Fast hikes + QT2
2022-03-16 → 2023-07-260.25% → 5.25-5.5% in 16 months (incl. four 75bp); QT2 runoff from Jun 2022
525bp in 16 months: energy was the only true shelter and real estate the most direct casualty.
From Mar 2022 the Fed hiked at the fastest pace in four decades (four 75bp moves) and added QT2 runoff in June. Under the supply shock, energy led at +27%; real estate −11% finished last: rates transmit to cap rates most directly; financials −2% absorbed SVB's failure in Mar 2023.
Note the window starts Mar 15, 2022: after the first leg down: and contains H1 2023's AI rally, so technology shows +22% for the era. A reminder that window boundaries shape the print: tech's calendar-2022 drawdown was far uglier than this number.
TakeawayIn an inflation-driven hiking cycle, hard assets are the only hedge; the longest-duration assets (real estate, unprofitable growth) take the most direct mathematical damage.
5.5% plateau
2023-07-27 → 2024-09-17Terminal rate held 14 months; QT continues, Treasury runoff cap cut from $60B to $25B/mo from Jun 2024
The surprise on the 5.5% plateau: high rates did not kill the AI mega-caps: communications +33% and financials +30% led.
After reaching 5.25-5.5% in Jul 2023 the Fed held for 14 months. Textbooks say high rates compress valuations; reality was an AI capex supercycle whose earnings upgrades outran the discount rate: communications +33% (META/GOOGL), financials +30% (soft landing plus margins), technology +25%.
Energy +5% fell behind as oil slid, and defensives broadly lagged. The plateau's real signal sat at the two ends: the market voted soft landing through financial leadership and voted disinflation through energy stalling: paving the way for the Sep 2024 cut.
TakeawayA rate plateau is not an equity plateau: earnings upgrades can overpower the discount rate; watch financials and energy as the two macro voting machines.
Cutting cycle + QT end
2024-09-18 → 2025-12-31175bp of cuts (Sep 2024 → Dec 2025) to 3.50-3.75%; QT runoff ceases Dec 1, 2025
The soft-landing cut: growth led across the board while defensives went negative: the signature of cuts without recession.
Cuts opened with 50bp in Sep 2024 and totaled 175bp to 3.50-3.75% by Dec 2025; QT's end was announced Oct 29, 2025 with runoff ceasing Dec 1. Cuts plus expanding earnings reignited growth: communications +37%, technology +33%, discretionary +25%.
The information was in the laggards: staples −3%, real estate −6%, health care just +2%: recession hedges were systematically abandoned. Against 2019 (utilities +40% in a cutting era), the same policy produced the opposite fate for defensives: 2019 bought insurance, 2024-25 returned it.
TakeawayJudge which kind of cut the market is pricing by how defensives behave: defensives rallying means recession insurance; defensives falling confirms the soft landing.
2026 hold + new Chair
2026-01-01 → todayFour straight holds at 3.50-3.75% (Jan/Mar/Apr/Jun); $40B/mo reserve-management bill buying; Warsh becomes Chair May 22
The hold in progress: energy has taken the baton at +35%, hard assets and industrials strengthening while communications and discretionary turned negative.
Four straight holds at 3.50-3.75% across Jan/Mar/Apr/Jun 2026 (April went 8-4, the most dissents since 1992), with a split dot plot: nine of the eighteen submitted dots see at least one hike this year (new Chair Warsh did not submit one); Warsh took the chair on May 22; the $40B/month bill purchases that began in December are reserve management, not QE, slated through April. With easing expectations stalled, sticky inflation is back in the price.
The half-year scorecard from our data: energy +35% leads, technology +22% (the compute-power chain), industrials +18%, real estate +16%, materials +14% strengthening; communications −9% and discretionary −8% negative. Leadership has rotated from cut-beneficiary growth toward sticky-inflation beneficiaries: consistent with the committee's hawkish tilt. This era is live; readings update with the latest session.
TakeawayWhen easing expectations stall, rotation front-runs the Fed: track whether this gear change holds with KovaView's live industry rotation and KOVA scores instead of waiting for the next dot plot.
Cross-cycle patterns
QE has no single script
QE2 built a commodity bull (energy +27% alone at the top), QE3 built a broad bull (health care +73%, financials +55%), and unlimited QE lifted growth first then commodities. The script is written not by the purchases but by the macro soil they land in: output gap, dollar direction, inflation expectations.
Hikes are not the bear switch: double tightening is
Equities rose broadly through the 2015-17 slow-hike era (technology led at +41%, financials +38% with strong excess); the real liquidity killer was 2017-18's hikes-plus-runoff stack (S&P −4%, only defensives standing). Gauge tightening damage by whether both price (rates) and quantity (balance sheet) move together.
Cuts wear two faces: defensives are the lie detector
Utilities +40% in the 2019 cutting era (the market bought recession insurance); staples −3% and real estate −6% in 2024-25 (the market returned it). Same policy, opposite defensive tape: it tells you which script is being priced.
Duration is the through-line of all sixteen years
The direction of policy rates drives the seesaw between long-duration assets (tech, real estate, utilities) and short-duration hard assets (energy, materials): duration wins when rates are pressed down (2011-12, 2019, 2024-25); hard assets win when inflation runs (2010-11, 2021-22, H1 2026).
Window boundaries shape the print
Technology shows +22% for the 2022 hiking era only because the window opens after the first leg down and contains the AI rally: the calendar-2022 drawdown was far uglier. Always read this page's numbers against the era's start and end dates.
History offers no script, only rhymes. To see where the current rotation stands today, keep tracking the live RRG and KOVA scores on the industry rotation pages.