EMCOR Group, Inc. (NYSE: EME) — The Value Lane of the Data-Center Build, Priced Like the Boom Holds

Independent Equity Research Report date: June 21, 2026 Price (as of 2026-06-18 close): ~$836.59 · Market cap: ~$36.9B · Enterprise value: ~$36.2B (net cash ~$0.9–1.1B) Shares out (diluted): ~44.7M · FY end: December · Sector: Industrials — Construction & Engineering (Mechanical/Electrical specialty contractor + facilities services)

⚡ Claude’s Take

This is the author’s own independent opinion and general information, not investment advice. The analysis that follows takes no position and carries no price target; this opening block is the single, clearly-labeled exception where a directional view is expressed.

Verdict: GREAT BUSINESS, FULL-BUT-NOT-ABSURD PRICE — HOLD at ~$837; for existing owners a HOLD, not a fresh chase; explicitly NOT a short. Fresh-buy / accumulation zone on a reset toward ~$640–$720 (≈22–24× a normalized ~$29–30 of forward earnings — a multiple that still pays a premium to most contractor peers and roughly doubles EMCOR’s own pre-2024 average). Directional fair-value zone ~$720–$900. Conviction: medium.

EMCOR is, on the evidence, the best-diversified and arguably best-capital-allocated way to own the U.S. mechanical/electrical (MEP) construction super-cycle. It is the larger, broader, cheaper cousin of Comfort Systems (FIX): ~$17B of revenue across five businesses, a genuine ~25%-of-revenue recurring facilities-and-industrial-services base that FIX lacks, a fortress net-cash balance sheet, a decade of disciplined buybacks that shrank the share count ~17% since 2022, GAAP-pure reporting (management refuses to add back amortization), and a CEO who says out loud that he “wouldn’t borrow to buy back stock.” The operating record is extraordinary: revenue compounded ~14% since 2020, operating margin doubled from 5.1% (2022) to ~9.2% clean (2025), diluted EPS went $8.10 → $28.19, and ROE reached 23.6%. The engine is unambiguous — data-center (“network & communications”) construction, where remaining-performance-obligations grew ~60% to $4.46B and total backlog hit a record $15.6B at a ~1.5× book-to-bill. None of the quality is in dispute.

The debate is entirely price against durability, and EMCOR sits in a more defensible spot than FIX on both counts. It trades at ~28× reported / ~32× clean trailing earnings and ~2.1× sales — the richest multiple in its own ten-year history (composite 95th percentile), but materially below FIX (~50× / ~4.7× sales) and Quanta (PWR) on a far larger, more recurring base. What the ~$36B enterprise value underwrites is that today’s ~9.2% operating margin (versus a 5–6% historical norm) is the new baseline, that low-double-digit growth persists, and that the premium multiple holds — three peaks stacked on one bet: that hyperscaler capex is a secular plateau, not a capital cycle. Management’s own framing is honest about the risk: “where we make our most margin… is where we take the most risk… fixed-price risk.” A reverse-DCF says the current price already discounts the base case (~9% margin, ~8% growth, 14× exit ≈ today’s price); the asymmetry from here is a fat left tail (bear ~−20%/yr if margins mean-revert toward 6.5% as data-center capex digests) against a thinner right tail (bull ~+12%/yr). The framing is momentum + quality-at-a-fair-to-full-price — not a bubble, not a bargain. I am not short it (real secular demand, net cash, a labor-scarcity moat, a diversified base, and powerful momentum make shorting a peak-multiple compounder a fast way to get carried out), but I would not pay up here; I’d wait for the multiple to come back to the low-$700s or below, where the diversification and balance sheet give a real margin of safety. Note the tells: the Street’s most bullish target ($1,123, Cantor) sits in the bull corner, while insiders sold exclusively via 10b5-1 plans with zero open-market buys, executive comp carries no return-on-capital metric, and reported FY25 EPS was flattered ~11% by a one-time UK divestiture gain run through the operating line.

One-line tag: The value lane of the data-center build — better-diversified than FIX, priced like the boom holds. Conviction: Medium. Flips bullish if: through a genuine data-center/macro softening EMCOR sustains ~9%+ consolidated operating margins and >1.1× book-to-bill (proving the step-up is structural, not cyclical), or the multiple resets to the low-$700s on no fundamental break. Flips bearish (toward avoid/trim) if: book-to-bill falls below 1.0 for two consecutive quarters or a hyperscaler guides capex down while the stock still trades >28× — the margin/growth/multiple trifecta starting to unwind with the price not yet adjusted.

📈 Stock Price Action — Five-Year Event Map

EMCOR is a five-year ~8.7-bagger that has just paused below its peak. The stock troughed at ~$96.26 (June 2022) in the rate-shock industrial selloff, ran almost without interruption to an all-time high of ~$943.75 (May 6, 2026), and trades at ~$836.59, about −11% off the high and well inside a 52-week range of ~$494.60–$943.75. It sits above its 21-day (~$837), 50-day (~$834), and 200-day (~$731) moving averages — a one-way uptrend that recently stalled, not a breakdown. Beta is ~1.36 with strongly positive alpha (~0.41). The move is overwhelmingly an earnings-and-backlog story compounded by AI-data-center re-rating; the price moves below are Fact, the attributed drivers Interpretation.

Cycle narrative. (1–2) The 2022 low was a macro/rate event, not a company event — EMCOR’s earnings were rising the whole way down, setting up the re-rate. (3–4) 2023–early-2024 is where the market re-categorized EMCOR from a cyclical late-cycle contractor into a structural electrification/data-center beneficiary, as backlog and margins inflected together. (5) 2024’s doubling layered the AI-capex narrative onto the Miller Electric acquisition (closed Feb-2025), deepening the electrical/data-center tilt. (6) The early-2025 −27% drawdown — coincident with the broad AI-capex “DeepSeek” scare — shows the stock’s two-sided sensitivity: a high-multiple momentum name de-rates fast on any threat to the capex story, with no change in the print. (7–8) The recovery to the May-2026 ATH was pure execution (FY25 results, Q1-26 +30% EPS, a record $15.6B RPO at ~1.5× book-to-bill), and the subsequent ~11% pullback is a multiple pause, not a thesis break. The opportunity/mispricing judgment belongs to Claude’s Take above; this map states only what happened and why.

1. Executive Summary

EMCOR Group is the largest U.S. mechanical and electrical (MEP) specialty-construction and facilities-services company — $16.99 billion of FY2025 revenue across five segments: U.S. Electrical Construction, U.S. Mechanical Construction, U.S. Building Services, U.S. Industrial Services, and (until its December-2025 sale) U.K. Building Services. It is a decentralized holding company of ~100 operating subsidiaries that design, install, and service the electrical and mechanical “guts” of buildings and industrial facilities — power distribution, lighting, low-voltage/communications, HVAC, piping, plumbing, fire protection, controls — plus a large recurring base of facilities maintenance and refinery/industrial-plant services. Like FIX, it is an installer and integrator, not an equipment manufacturer: it buys the switchgear, chillers, and generators it installs.

The three-year operating record is one of the best in industrials. Revenue compounded ~14% annually (2020 $8.80B → 2025 $16.99B), with FY25 up +16.6%; consolidated operating margin doubled from 5.1% (2022) to ~9.2% (2025, clean of a one-time gain); diluted EPS rose from $8.10 (2022) to $28.19 (2025); ROE reached 23.6% and ROA ~15%. Free cash flow was ~$1.19B in FY25 on a net-cash balance sheet. The engine is the AI/data-center construction boom: “network & communications” remaining-performance-obligations (RPOs) grew ~60% to $4.46B at year-end 2025, total RPOs hit a record $15.62B at Q1-26 (+33% YoY) at a ~1.5× book-to-bill, and the two U.S. construction segments — 72% of revenue, 88% of U.S. segment operating income — carry low-teens segment margins.

The business quality is genuinely high and, importantly, more diversified and more recurring than FIX: U.S. Building Services (~$3.1B, facilities maintenance) and U.S. Industrial Services (~$1.3B, refinery turnarounds) together ~26% of revenue provide a counter-cyclical, aftermarket-heavy ballast that pure-construction peers lack. Capital allocation is a standout: ~$586M of buybacks in FY25 (shares 49.9M in 2022 → 44.7M at Q1-26), a dividend raised every year (and +60% to $0.40/qtr in 2026), >$1B of disciplined M&A (Miller Electric, ~$877M), GAAP-pure reporting, and a CEO (Anthony Guzzi, in the seat since 2011) who is explicit about not levering up for financial engineering.

The tension is valuation against durability. EMCOR trades at ~28× reported / ~32× clean trailing earnings, ~2.1× sales, and ~20× EV/EBITDA — the richest in its own ten-year history (composite 95th percentile on EMCOR’s own valuation), though well below FIX (~50× / ~4.7× sales) and PWR on a much larger base. The current enterprise value embeds three simultaneous peaks: peak margins (a fixed-price contractor is a structural price-taker over a full cycle), peak growth (data-center-led), and a peak own-history multiple — all correlated to one variable, hyperscaler capex. A reverse-DCF shows the price already discounts a healthy base case; the asymmetry favors the downside if 9.2% proves to be peak-cycle rather than the new normal. The body that follows evaluates EMCOR across the full framework and takes no recommendation and no price target. The short version: a superbly diversified, disciplined operator executing a genuine secular demand wave, priced as though peak economics are the baseline — but with enough diversification, balance-sheet strength, and (relative to FIX) multiple cushion that the risk is “full price,” not “bubble.”

2. Business Overview

What EMCOR does. EMCOR is a U.S. (and, until late 2025, U.K.) provider of electrical and mechanical construction and facilities services. Through roughly 100 decentralized operating companies it delivers: power transmission/distribution and premises electrical and lighting systems; low-voltage, voice/data, fire-alarm and building-automation systems; HVAC, refrigeration, plumbing, process and high-purity piping, clean-room ventilation, fire protection; site-based facilities operations and maintenance; and refinery turnaround, overhaul, instrumentation and renewable-energy services for industrial plants. It is an installer and integrator: like FIX, it purchases the branded equipment (Eaton, Schneider, Carrier, Trane, Cummins, etc.) it installs, so a meaningful share of project cost is pass-through materials/equipment, and its value-add is engineering, project management, skilled-trade self-perform labor, prefabrication, and field-supervision.

Decentralized holding-company model. EMCOR’s structure mirrors FIX’s Berkshire-style decentralization: ~100 locally-branded, locally-managed operating units, each with its own customer relationships, P&L, and trade specialization, coordinated by a lean corporate center that owns capital allocation, surety/bonding, insurance, safety, and shared best practices (prefab, virtual design/construction). CEO Guzzi describes the “core product” as field-labor supervision and leadership — the scarce input is not bodies but the foremen, general foremen, and project executives who can run complex jobs. This is the cultural core and the engine of both organic execution and the acquisition flywheel.

Five segments (FY2025). (Revenue from unrelated entities; operating income and segment margin per the FY25 10-K MD&A. The U.K. Building Services segment was sold December 1, 2025.)

*Consolidated operating income/margin includes the one-time ~$144.9M pre-tax gain on the U.K. divestiture (booked on the operating line). Clean of that gain, consolidated operating income is ~$1,568.5M and operating margin ~9.2% — the figure used throughout this memo.

The two U.S. construction segments are the story. Electrical + Mechanical Construction together are 72% of revenue and ~88% of U.S. segment operating income, both carrying ~12–13% segment margins and both explicitly driven by data-center construction. Electrical’s +51.8% revenue jump is largely the Miller Electric acquisition (~$1.1B incremental); its segment margin dipped 130bps to 12.1% almost entirely on non-cash acquired-intangible amortization (gross margin “relatively consistent,” per management). Mechanical’s margin expanded 30bps to 12.8% on mix and prefabrication execution. Building Services (facilities maintenance, mobile mechanical service, government site-based O&M) was flat at ~$3.1B and ~6% margin; Industrial Services (refinery turnarounds, shop services) declined slightly at ~2% margin (a lumpier, lower-quality, oil-&-gas-tied business).

Revenue model — project-heavy construction plus a recurring services base. The construction segments (~71% of revenue) are predominantly fixed-price or modified-fixed-price project work, recognized over time on the cost-to-cost percentage-of-completion method — lumpy, re-bid project-by-project, and structurally a price-taker over a cycle. But unlike FIX (~93% project), EMCOR carries a ~26%-of-revenue Building + Industrial Services base that is substantially more recurring and aftermarket: multi-year facilities-management contracts, mobile mechanical service, government O&M, and recurring refinery turnaround cycles. This is lower-margin (2–6%) but stickier and more counter-cyclical — a genuine quality differentiator versus a pure-construction roll-up, and a reason EMCOR’s earnings should prove somewhat more resilient than FIX’s in a construction downturn.

End-markets. EMCOR does not publish a single “% data center” revenue figure (an open question for precise sizing), but discloses RPOs by market sector. Backlog growth is led by network & communications (predominantly data centers) — RPO ~$4.46B at YE25, +60% — followed by institutional/education, water/wastewater, and healthcare, partly offset by a decline in high-tech manufacturing (semiconductor-fab completions rolling off). The mix is shifting from fabs to data centers. Geographically U.S.-centric (now entirely U.S. post-U.K. sale).

Verdict. A well-run, decentralized MEP contractor whose growth and margin are now driven by fixed-price data-center construction, but with a genuinely diversifying ~26% recurring facilities/industrial-services base that pure-construction peers lack. Understand EMCOR as a superbly-executed, more-diversified contractor — higher quality of business mix than FIX, lower headline margin, and (critically) a much cheaper multiple.

3. Industry Dynamics

Structure: enormous, hyper-fragmented, low-barrier at the small end. The U.S. commercial/industrial/institutional MEP-contracting market is ~$700B+ in annual revenue, “highly fragmented and competitive,” composed of thousands of local and regional contractors with “low barriers to entry in most markets” (the industry’s own framing, echoed in EMCOR and FIX 10-Ks). At ~$17B, EMCOR holds ~2% share and is far larger than its typically small, owner-operated competitors. The base-state industry is structurally mediocre — local, cyclical, price-competitive, commoditized — and a two-truck electrical shop is trivial to start. The investable question is whether the complex, large-project, multi-market slice that scaled incumbents serve carries higher barriers. The evidence says modestly yes — and increasingly so as data-center scope density rises.

Demand — a genuine secular super-cycle, currently supply-constrained. The dominant driver is the AI/data-center build-out, layered on advanced-manufacturing reshoring (semiconductors, EV/battery, GLP-1 pharma), grid electrification, water/wastewater infrastructure (EPA-driven), healthcare, and an aging installed base needing efficiency retrofits. Management is emphatic that the market has flipped from demand-limited to supply-limited: there is more work available than the industry can staff. A second, often-missed tailwind is scope density — modern AI data centers carry several times the MEP content per square foot of a conventional facility (higher power density, liquid cooling, redundant electrical), a price/content tailwind independent of unit growth that favors the most sophisticated contractors. EMCOR’s record $15.6B RPO at ~1.5× book-to-bill quantifies the demand.

Labor — the binding constraint and the crux. The skilled-trades shortage (electricians, pipefitters, and especially field supervisors) is the limiting factor. EMCOR grows revenue 2–3× faster than headcount via prefabrication, virtual design/construction (VDC/BIM), and cross-unit labor sharing — a structural productivity lever — but the supply of qualified supervision ultimately caps how fast anyone converts backlog. This cuts two ways: it is the industry’s ceiling, and — for whoever already has the assembled, trained, retained workforce — the source of pricing power and selectivity. When demand exceeds the supply of qualified labor, the scaled firms win the right to select and price work. That is exactly what EMCOR’s and FIX’s margins reflect.

Marathon capital-cycle read. In the Marathon framework, abnormal returns attract capital, capacity floods in, and returns mean-revert. Where is capital flooding?

• Into data-center construction (the demand side): yes — hyperscaler capex is at a historic peak. By capital-cycle logic the customers are over-building, which eventually digests and removes the demand tailwind. EMCOR’s record margins coincide with a generational customer-capex spike — the macro risk, and the one that would hit backlog directly given the data-center RPO concentration.
• Into MEP contracting capacity (the supply side that sets pricing): no — the bull’s strongest structural argument. You cannot conjure journeyman electricians, pipefitters, and project supervisors with capital; the trades take years to train and the apprentice pipeline shrank for a generation. The normal self-correction — competitors add capacity and compete the margin away — is blocked by the labor bottleneck. This is why margins have held longer than a normal construction cycle would predict. But it is a delay in mean-reversion, not its repeal: if hyperscaler capex rolls over (demand side) before the labor shortage eases, EMCOR still de-rates, because the incremental backlog is data-center backlog.

Cyclicality, regulation, and political risk. Construction is cyclical; volume can fall sharply in a downturn. Regulation is light and mostly neutral-to-favorable (state contractor licensing as a modest barrier; OSHA; HFC refrigerant phase-down as a retrofit driver; IRA/CHIPS reshoring incentives as tailwinds). The newer risk is political/infrastructural: several jurisdictions are debating data-center moratoria and power-interconnection limits — a direct backlog risk given the data-center concentration. EMCOR’s diversification (water, healthcare, institutional, industrial services) and its recurring base partly cushion this versus FIX’s higher technology concentration.

Verdict. Structurally mixed — a mediocre base industry (fragmented, low-barrier, cyclical) enjoying a genuinely excellent moment because a generational demand shock (AI/data centers + reshoring + electrification) collides with a hard, non-replicable supply constraint (skilled trades and supervision), handing scaled, well-capitalized, labor-rich incumbents temporary but real pricing power and selectivity. The durability of “good” depends on whether the labor constraint outlasts the data-center capex peak. EMCOR’s diversification makes it a structurally better-positioned participant than a pure data-center contractor — but it does not exempt it from the cycle. Do not confuse the current windfall with permanent structural attractiveness.

4. Competitive Position

The honest framing. EMCOR is an exceptionally well-run operator harvesting a genuine secular demand wave with a narrow-but-real and diversified moat built from scale + national multi-market coverage + a scarce assembled labor/supervision pool + prefab/VDC capability + reputation + a fortress balance sheet. It is not a wide-moat, network-effects compounder; the 10-K concedes low barriers in most markets. The advantage is relative and cyclically amplified — but it is broader and arguably more durable than FIX’s because of EMCOR’s larger scale, multi-segment coverage, and recurring services ballast.

Naming the mechanism (Greenwald taxonomy):

1. Scale economies + national, multi-market coverage (the strongest claim). EMCOR is the only contractor able to staff and self-perform across ~17 electrical and ~7 mechanical data-center markets simultaneously, and to cover ~70% of U.S. data-center sites on fire-life-safety. For a hyperscaler building everywhere at once, single-vendor national coverage, purchasing leverage, bonding capacity, and the ability to reallocate engineering and supervisory labor across units is a real, hard-to-replicate advantage — and one that matters specifically for the mega-projects that small shops cannot bid. This is bounded (it does not help on the local strip-mall job) but it is exactly where the data-center spend concentrates.

2. Assembled labor + field supervision as a quasi-intangible (the most durable element). In a labor-constrained industry, the trained, retained workforce — and especially the bench of qualified supervisors — is the scarce asset, and it cannot be bought with capital. EMCOR’s recruiting/training machine, cross-unit mobility, and “core product is field-labor supervision” culture let it grow when rivals cannot staff. This is the closest thing to a moat EMCOR has, but it is a relative advantage in a shortage, not an absolute barrier; it erodes if the trades labor market loosens.

3. Reputation / execution intangible (real for data centers). For a hyperscaler where downtime costs millions per hour and schedule certainty is everything, a proven MEP contractor’s execution track record is a genuine selection criterion, not pure price. Management’s tells of captivity are credible: hyperscaler end-users reach through their general contractors to request EMCOR specifically, and Guzzi notes “I don’t remember us ever being replaced on a site.” This lets EMCOR be selective about the work and price it takes.

4. Prefabrication / VDC (the genuine differentiator). Prefabricating MEP assemblies off-site compresses schedule and slashes on-site labor — directly attacking the binding labor constraint — and is the reason revenue grows 2–3× faster than headcount. It is capital-light at EMCOR (modest capex ~$115–125M/yr) but requires scale and an order book to justify, and embeds some switching costs via co-designed capacity. It is a smaller share of the model than FIX’s ~18% modular, but it is a real, growing structural lever rather than a cyclical one.

5. Balance-sheet-as-moat (underrated). EMCOR’s net-cash position is a deliberate competitive differentiator: management notes hyperscalers “are not looking to do business with leveraged companies” — counterparty strength and bonding capacity are bid criteria on multi-hundred-million-dollar jobs. A fortress balance sheet lets EMCOR commit capacity, prefab investment, and surety that thinner-capitalized rivals cannot.

6. Recurring services base (the diversification edge over FIX). The ~26%-of-revenue Building + Industrial Services segments — facilities maintenance, government O&M, mobile mechanical service, refinery turnarounds — are lower-margin but stickier, more recurring, and more counter-cyclical than project construction. They give EMCOR a more resilient through-cycle earnings stream than a construction-only peer, and a cross-sell relationship with the same building owners.

Customer captivity / switching costs (weak-to-moderate). Base construction is re-bid project-by-project; captivity is thin at the transaction level. But it is real at the relationship and density level — repeat hyperscaler relationships, multi-year facilities contracts, and the reputation/execution selection criterion. Net: moderate and improving, not a wide moat.

Why have margins structurally risen — durable or cyclical? Consolidated operating margin went 5.1% (2022) → 7.0% (2023) → 9.2% (2024, on a like-for-like basis) → ~9.2% clean (2025); construction-segment margins are now ~12–13%, and management targets a sustainable construction band of 12.5–13.5% read on a rolling 12–24-month basis. Decomposition (from the transcripts): the step-up is ~30–40% pricing/selectivity (a cyclical, labor-scarcity-driven component — ~3–4 points) and the balance structural productivity (prefab/VDC, mix toward complex high-content work, scale leverage on fixed overhead). The honest read: the structural leg is real and partly durable, but a meaningful slice of today’s margin is cyclical pricing that mean-reverts when labor eases or demand digests. Management itself volunteers that “where we make our most margin is where we take the most risk — fixed-price risk,” and disclosed a mispriced FY24 fixed-price job (“we own it”). This is not a runaway pricing-power monopoly; it is a scaled, disciplined contractor earning peak-cycle economics it does not claim are permanent.

Verdict. A narrow-but-real, diversified moat — scale + national coverage + assembled labor/supervision + prefab + reputation + balance sheet — that is broader and somewhat more durable than FIX’s, amplified by the cycle. EMCOR will out-earn the fragmented field for as long as the labor constraint and demand hold; it is not a structurally-protected compounder whose peak margins are guaranteed to persist through a data-center digestion.

5. Growth History and Forward Opportunities

Historical growth — high and accelerating, increasingly high-quality. Revenue compounded ~14% annually from $8.80B (2020) to $16.99B (2025), accelerating to +16.6% in FY25 and +19.7% in Q1-26 (+16.8% organic). EPS growth has far outpaced revenue on operating leverage: diluted EPS $2.40 (2020) → $7.06 (2021) → $8.10 (2022) → $13.31 (2023) → $21.52 (2024) → $28.19 (2025) — a ~12× increase in five years, of which the post-2022 acceleration reflects the margin doubling. Growth is overwhelmingly organic: the Miller Electric acquisition added ~$1.1B of FY25 revenue, but >80% of the data-center RPO growth was organic, and management characterizes FY26 guided growth as “really all organic” (the Miller comp anniversaries, and the U.K. sale is a ~3% headwind offset by Danforth and other tuck-ins).

Backlog / RPOs — the leading indicator, at records. Total RPOs: $8.85B (YE23) → $10.10B (YE24) → $13.25B (YE25, +31%) → record $15.62B at Q1-26 (+33% YoY), at a ~1.5× book-to-bill (a record). U.S. construction is ~90–91% of backlog. Network & communications (data centers) RPO ~$4.46B at YE25, +60% YoY — alone larger than EMCOR’s entire company-wide RPO at the end of 2019. RPOs are conservatively defined (funded phases only; ~78–82% burns within 12 months), so true visible demand is larger than the headline.

Forward opportunities.

• Data centers / network & communications — the dominant driver, with no sign of slowing per management (“AI infrastructure driving unprecedented levels of activity”); scope-density and the breadth of hyperscaler/colocation build keep both electrical and mechanical content rising.
• Reshoring / advanced manufacturing — semiconductors (CHIPS), EV/battery, GLP-1 pharma; lumpy and currently a headwind as fab completions roll off, but a multi-year tailwind as new projects start.
• Water / wastewater — EPA-driven, a named RPO growth sector (Florida and elsewhere); diversifying and counter-cyclical.
• Healthcare, institutional/education, grid electrification — steady, less cyclical demand.
• Recurring services growth — Building Services O&M and mobile mechanical service cross-sell into the installed base; Industrial Services turnaround cycles.
• M&A — a disciplined, internally-funded tuck-in pipeline (Miller, Danforth, and 9 smaller FY25 deals for ~$182M).

FY26 guidance — raised at Q1. Revenue $18.5–19.25B and diluted EPS $28.25–29.75, both raised from the initial $17.75–18.5B / $27.25–29.25. Even the raised guide looks conservative versus the Q1-26 run-rate (Q1 EPS $6.84 annualizes well above the range), with ~30% of FY26 revenue still to be booked intra-year — a setup that has produced serial beats.

Verdict. High-quality growth — predominantly organic, backlog-supported (record RPO at 1.5× book-to-bill), margin-accretive, and diversified across data centers, reshoring, water, and healthcare, with a recurring-services cross-sell. The single dependency is data-center demand; the quality caveat is that a portion of the margin embedded in that growth is cyclical pricing. But on volume and backlog, this is among the highest-quality growth profiles in industrials.

6. Financial Quality

Revenue & margins. Revenue $8.80B (2020) → $16.99B (2025), +16.6% in FY25. Gross margin expanded from 14.5% (2022) to 19.3% (2025); consolidated operating margin from 5.1% (2022) to ~9.2% clean (2025) (10.1% reported, including the U.K. gain). Incremental operating margins ran ~20–24% in 2023–24 — strong operating leverage on a scaling fixed-overhead base. Construction-segment margins are now ~12–13%.

Quality of earnings — a one-time gain to normalize, otherwise clean. Two QoE points matter:

1. The U.K. divestiture gain. EMCOR sold its entire U.K. Building Services segment on December 1, 2025 for net proceeds of ~$256.6M, recognizing a ~$144.9M pre-tax gain booked on the operating-income line (including a ~$73.5M AOCI reclassification). Because the gain was largely non-taxable, the after-tax benefit was ~$137M ≈ ~$3.03 per diluted share ≈ ~11% of FY25 diluted EPS ($28.19). Clean FY25 EPS is ~$25.16 (management’s “adjusted” figure is ~$25.87, adding back acquired-intangible amortization, which we treat more conservatively). Reported FY25 operating margin of 10.1% flatters the clean ~9.2% by ~85bps. This is the single most important normalization for valuation: the headline trailing P/E understates the true multiple.
2. Otherwise high quality. EMCOR uses GAAP-pure reporting — management explicitly refuses to add back amortization in its core metrics, a genuine governance positive. SBC is immaterial (~$21M, <0.2% of revenue). Cash conversion is solid: OCF/NI ~1.02× in FY25 ($1,302M / $1,273M), though FY25 OCF actually fell YoY (FY24 $1,408M) on a working-capital build (AR +$438M as revenue and unbilled work scaled) — normal for a fast-growing percentage-of-completion contractor, and worth monitoring. FCF was ~$1.19B (capex a light ~$113M, <1% of revenue — the asset-light installer model).

Returns on capital. ROE 23.6% (2025), up from 13.4% (2022); ROA ~15%. These are high and rising, reflecting both the margin step-up and the buyback-shrunk equity base. On invested capital, EMCOR earns well above its cost of capital in the current environment — but note the equity base now carries ~$1.4B goodwill + ~$1.1B intangibles (largely Miller Electric), so tangible returns are even higher while book equity (and thus P/B) is inflated by acquisition accounting.

Balance sheet — fortress, net cash. Cash ~$1.11B (YE25) against ~$468M of debt (almost entirely finance/operating-lease obligations), i.e., net cash ~$0.6–1.1B depending on lease treatment; the revolver is undrawn. Total equity ~$3.67B; current ratio ~1.2×. This is a genuinely conservative balance sheet for a contractor and a competitive asset (bonding, counterparty strength, capacity to invest and buy back through a downturn).

Working capital & the percentage-of-completion caveat. As a POC contractor, EMCOR carries large contract receivables (~$4.6B) and contract liabilities / billings-in-excess (~$2.3B deferred revenue). On exploding backlog, customer advances and billings-in-excess provide a working-capital tailwind to cash (as at FIX); when backlog growth flattens, that float reverses and OCF can lag NI. EMCOR’s FY25 working-capital build (a use of cash) suggests it is past the peak-float phase that flattered FIX’s 2025 cash — a more sober, less “free-float-flattered” cash profile, which is actually a quality positive (less of the cash is borrowed from the future).

Verdict. Economics genuinely improve with scale — operating leverage is real, returns are high and rising, the balance sheet is a fortress, and reporting is conservative/GAAP-pure. Two honest caveats: (1) ~11% of FY25 reported EPS is a one-time gain that must be normalized out (true trailing multiple is higher than headline), and (2) a portion of the record margin is cyclical pricing that is not guaranteed to persist. Subject to those, this is a high-quality financial profile.

7. Capital Allocation

The strongest single pillar of the EMCOR thesis after the operating record. Management allocates capital like owners, and the track record is excellent.

Buybacks — consistent and counter-cyclical. EMCOR repurchased ~$586M of stock in FY25 (~1.4M shares), ~$490M in FY24, and ~$128M in FY23; cumulatively ~28.5M shares / ~$2.97B since 2011. Diluted share count fell from 47.56M (FY23) to 45.15M (FY25) to 44.69M (Q1-26) — a ~17% reduction since 2022, a genuine per-share tailwind. ~$680M remains on the authorization (recently topped up by $500M). Critically, Guzzi is explicit that he “wouldn’t borrow to buy back stock” — buybacks are funded from FCF, not leverage — and the program has been opportunistic (heaviest when the stock de-rated). This is disciplined, value-additive capital return, not financial engineering.

Dividend — small but growing. Raised every year ($0.69 → $0.93 → $1.00/sh, then +60% to $0.40/quarter in 2026); payout is tiny (~3.5%), appropriately leaving capital for buybacks and M&A.

M&A — disciplined, internally funded, GAAP-honest. The ~$877M Miller Electric acquisition (closed Feb-2025, all cash) deepened the electrical/data-center franchise; purchase accounting put ~$327M to goodwill and ~$475M to intangibles (customer relationships 16yr, backlog 1.5yr, indefinite-life trade name). Plus Danforth and 9 smaller FY25 tuck-ins (~$182M). Management’s M&A discipline is a feature, not a slogan: “We don’t buy into hype… not private equity guys… not averaging multiples down”; it won’t pay 12–15× for single-market data-center shops, will lever only 1–2× for the right deal, and self-grades a “strong B+.” EMCOR reports M&A on a GAAP-pure basis (no amortization add-backs), so the dilution to reported segment margin from Miller’s intangible amortization is honestly disclosed rather than adjusted away.

Capex — light. ~$113M FY25 (<1% of revenue), mostly prefab fit-out and equipment; the installer model is asset-light, which is why FCF ~= net income and ROIC is high.

Governance flags (the demerits).

• No return-on-capital metric in executive comp. Incentive pay keys on adjusted diluted EPS plus an operating-cash-flow-to-operating-income ratio (a cash-conversion/QoE gate — a partial positive). But there is no ROIC / return-on-invested-capital hurdle anywhere in annual or long-term comp. EPS is growable via buybacks and M&A regardless of capital efficiency; absent a capital-charge metric, this is a real (if currently academic, given high returns) governance weakness.
• Combined Chair/CEO. Anthony Guzzi has been CEO since January 2011 and combined Chairman since 2018 (a Lead Director mitigates). Long tenure has coincided with excellent execution, but the role concentration and lack of a clear succession signal are watch items.
• Low insider ownership. Directors + officers own ~0.73% of the company (Guzzi ~170k shares). Alignment runs through comp and stock grants more than through large personal stakes.
• Insider transactions — no conviction buying. Across all Form 4s since mid-2024 there were zero open-market purchases (code P); activity is exclusively grants (A), tax-withholding (F), gifts (G), and 10b5-1-planned sales (S) by Guzzi, CFO Nalbandian, and several directors. Routine, but not a bullish insider signal — and worth noting against a richest-ever multiple.

Verdict. Management has allocated capital intelligently — disciplined, internally-funded buybacks that meaningfully shrank the share count, a small growing dividend, value-conscious M&A, GAAP-pure reporting, and a fortress balance sheet. The blemishes are real but second-order: no ROIC comp metric, combined Chair/CEO, low insider ownership, and 10b5-1-only selling. On balance, a clear positive for the thesis and a differentiator versus less-disciplined contractors.

8. Changes and Headwinds — Last Two Years

Strategic / portfolio changes.

• Miller Electric acquisition (Feb-2025, ~$877M) — the largest deal in EMCOR’s history, deepening the U.S. electrical/data-center franchise; ~$1.1B of FY25 revenue, margin-dilutive on reported basis only via intangible amortization.
• U.K. Building Services divestiture (Dec-1-2025) — sold the entire U.K. segment for ~$256.6M net, a ~$144.9M pre-tax gain (the FY25 EPS one-timer). Sharpens EMCOR into a pure U.S. operator and recycles capital into higher-return U.S. construction.
• Continued tuck-in M&A (Danforth + 9 smaller FY25 deals, ~$182M) and prefab/VDC capacity expansion.

Operating trajectory. FY25 record revenue $16.99B (+16.6%), clean operating margin ~9.2%, adjusted EPS ~$25.87 (+20%); Q1-26 revenue +19.7%, EPS +30%; RPO to a record $15.62B at ~1.5× book-to-bill; FY26 guidance raised. The two-year arc is unambiguous acceleration driven by data-center demand and the margin step-up.

Capital return. Dividend +60% in 2026; ~$586M buyback in FY25; $500M authorization top-up. Share count down ~17% since 2022.

Leadership / board. No C-suite turnover; routine board additions (Amy Dahl Dec-2024, Pat Roche Oct-2025). Guzzi remains Chair/CEO.

Headwinds / watch items.

• Data-center demand digestion — the central risk; a hyperscaler capex pause would hit the data-center-concentrated backlog directly.
• High-tech manufacturing roll-off — semiconductor-fab completions are currently a backlog headwind (partly offsetting data-center growth).
• Margin mean-reversion — a portion of record margin is cyclical pricing; a labor-market loosening or demand cooldown compresses it.
• Data-center moratoria / power-interconnection limits — an emerging political/regulatory risk to backlog (not addressed on the calls — an open question).
• Working-capital reversal — as backlog growth flattens, the POC float reverses and OCF can lag NI.
• Multiple de-rating — at the richest own-history multiple, the stock is sensitive to any AI-capex scare (demonstrated by the −27% early-2025 drawdown).

Verdict. The two-year changes strengthen the franchise (Miller deepens the data-center electrical position; the U.K. sale sharpens focus; capital return and backlog are at records) — but they also deepen the data-center dependency and arrive at a peak multiple. Net: a stronger, more focused, more concentrated company priced for the cycle to persist.

9. Risk Analysis

Catastrophic-loss risk is very low (net cash, diversified, highly profitable, asset-light). The realistic risk is not impairment but multiple-plus-margin de-rating: a data-center capex digestion would compress margins, reverse the working-capital float, decelerate growth, and de-rate the multiple — partly simultaneously — from a peak valuation. EMCOR’s diversification and balance sheet make that left tail shallower than FIX’s, but it is the defining risk.

10. Valuation Discussion — Embedded Expectations

No price target; no recommendation. This section frames what the current price implies.

Where the multiple sits. At ~$836.59, EMCOR trades at:

• ~28× reported / ~32× clean trailing EPS (clean of the ~$3.03 UK gain; clean FY25 EPS ~$25.16),
• ~28× forward FY26 guided EPS midpoint (~$29),
• ~20× EV/EBITDA (TTM, current price; ~15.3× at the FY25 year-end price),
• ~2.1× sales, ~9.75× book (book inflated by Miller goodwill/intangibles).
• On EMCOR’s own ten-year history these are the richest ever: P/E 89.8th percentile, P/B 97.0th, P/S 97.9th, composite 94.9th. The stock has never been this expensive on its own metrics.

Relative to peers — cheaper than the premium names, premium to the cheap ones. (TTM peer data, at EME ~$836.59.)

EMCOR is materially cheaper than FIX (27.6×) and PWR (33.1×) on EV/EBITDA, in line with MTZ/STRL, and at a premium to MYRG/PRIM — on the largest, most diversified, most recurring-revenue base in the group. This is the heart of the “value lane” framing: the same secular trade as FIX, at ~70% of FIX’s EV/EBITDA multiple and ~45% of its EV/sales, with a more diversified and more recurring business mix. The relative-value cushion is genuine. The absolute-valuation caution (richest own-history multiple) is equally genuine. Both can be true.

Embedded-expectations / reverse-DCF (3-year, to FY28; FY25 base $16.99B revenue, ~9.2% operating margin).

For a ~10%/yr forward return, EMCOR’s market cap must reach ~$49B by FY28 — which requires, at the 9.2% margin held, roughly a 22% revenue CAGR at a 15× exit, a 15% CAGR at an 18× exit, or a 32% CAGR at a 12× exit. In every case the price already underwrites that 9.2% operating margin (up from 5.1% in 2022) is the new baseline, that growth stays at least high-single-digit, and that the exit multiple stays premium. The base case roughly equals today’s price; the asymmetry is a fatter downside (bear ~−20%/yr if margin mean-reverts toward 6.5% as capex digests) against a thinner upside (bull ~+12%/yr). Cantor’s $1,123 target (June 2026) sits in the bull corner. The single swing variable is whether ~9% consolidated operating margin is structural or peak-cycle.

Sum-of-the-parts sanity check. A rough SOTP — construction segments (~$1.5B segment op income) at ~16–20× and the recurring services segments (~$210M op income) at ~12–14× — lands in a similar ~$32–40B enterprise-value range, i.e., the market is paying a full-but-defensible multiple for the construction crown jewel and roughly fair value for the services base. It does not reveal hidden value; it confirms the construction segments carry the valuation.

Verdict. EMCOR is richly but not absurdly valued — the richest in its own history on an absolute basis, but meaningfully cheaper than the premium contractor peers on a larger, more diversified base. The price discounts the base case; there is little margin of safety at ~$837 and a fat left tail if the margin step-up proves cyclical. The relative-value cushion versus FIX is the most compelling valuation argument; the absolute own-history richness is the most compelling caution.

11. Variant Perception

Consensus belief. EMCOR is a best-in-class, diversified beneficiary of a multi-year data-center/electrification/reshoring construction super-cycle, with a fortress balance sheet, disciplined capital allocation, record backlog, and serial guidance beats — and it is “cheap relative to FIX/PWR,” so the premium multiple is justified and the beats continue. Sell-side is constructive (Cantor Overweight, $1,123).

Strongest bull case. The labor/supervision bottleneck blocks the normal margin mean-reversion (capital can’t conjure tradespeople), so ~12–13% construction-segment margins persist for years; data-center demand is a decade-long secular plateau, not a cycle; EMCOR’s scale, national coverage, prefab, balance sheet, and reputation let it keep selecting and pricing work; the ~26% recurring services base cushions any construction softness; disciplined buybacks compound per-share value; and at ~70% of FIX’s EV/EBITDA on a more diversified base, EMCOR re-rates further or simply compounds into its valuation. In this world the base/bull reverse-DCF scenarios hold and the stock works.

Strongest bear case. Today’s ~9.2% operating margin is ~3–4 points of cyclical, labor-scarcity pricing stacked on a ~5–6% structural base; when hyperscaler capex digests (a capital cycle, not a plateau), the data-center-concentrated backlog stops growing, fixed-price margins compress toward 6.5–7%, the working-capital float reverses (OCF lags NI), growth decelerates, and the richest-ever multiple de-rates — all at once, from a peak. The −27% early-2025 drawdown previewed the multiple’s sensitivity. Reported EPS is flattered ~11% by a one-time gain; insiders sell only via 10b5-1 with zero buys; comp has no ROIC hurdle; and the bull’s own Street target sits below where a 10%/yr buyer needs the stock to compound. In this world the bear corner (~−20%/yr) is live.

The 3–5 assumptions that matter most:

1. Is ~9% consolidated operating margin structural or peak-cycle?
2. How long does data-center / hyperscaler capex stay at this level?
3. Does the labor bottleneck outlast the capex peak?
4. Does the premium-to-history multiple hold, compress, or re-rate toward FIX?
5. Does the recurring services base + diversification actually cushion a construction downturn?

Falsification tests. Bull falsified if: book-to-bill falls below 1.0 for two consecutive quarters, or a major hyperscaler guides capex down, or construction-segment margins compress below ~10% on stable revenue — any of which signals the peak is cyclical. Bear falsified if: through a genuine data-center/macro softening EMCOR sustains ~9%+ consolidated operating margins and >1.1× book-to-bill, with OCF holding at/above NI — proving the step-up is structural.

Factor-positioning read. EMCOR loads as a crowded industrial-momentum / AI-infrastructure trade with genuinely high-quality fundamentals — not a falling knife. On a quantitative factor model (Base model): Momentum +0.73, Market ~+0.90, DividendYield +1.11; Value −1.15 (strongly anti-value), Quality −0.36 and Growth −0.23 (statistical strips — the fundamentals are high-quality). Risk-adjusted track record is strong (y1 +73%, Sharpe 1.83), but lifetime max drawdown is −70.6% (deeply cyclical historically). Factor-peers are FIX (0.95), PWR (0.93), STRL, IESC, NVT plus a cluster of momentum ETFs (PRN/AIRR/PDP) — confirming EMCOR is held as a momentum/AI-infrastructure vehicle. The tape is a one-way uptrend that just paused (−11% off ATH, above all moving averages). For consensus-offsides: the risk is that a momentum-crowded, anti-value, AI-infrastructure name de-rates violently if the capex narrative cracks — the early-2025 −27% drawdown is the template. This is input to the framing, not a price call.

12. Fact vs. Interpretation Table

13. Open Questions

1. Precise data-center revenue exposure. EMCOR discloses RPO by sector but not a clean “% of revenue from data centers.” Best estimate: data centers are ~30%+ of total RPO and the dominant growth driver, but the exact revenue share is not disclosed.
2. Margin decomposition. How much of the ~9% operating margin is durable structural productivity (prefab/VDC/mix/scale) versus cyclical labor-scarcity pricing? Management implies ~3–4 points is pricing/selectivity; not independently verifiable.
3. Data-center moratoria / power-interconnection. What is EMCOR’s exposure to jurisdictions debating data-center moratoria or grid-interconnection limits? Not addressed on the calls.
4. Customer concentration. EMCOR does not disclose a >10% customer (unlike FIX’s 12.8%); what is the actual top-customer / top-hyperscaler concentration within the data-center backlog?
5. Succession. No clear successor signal for the combined Chair/CEO (Guzzi, in role since 2011/2018).
6. Working-capital normalization. As backlog growth flattens, how much does the POC float reverse and pressure OCF relative to NI? FY25’s AR build suggests this is partly underway.
7. Through-cycle margin floor. What is the realistic trough operating margin in a genuine construction downturn — 6%, 7%, 8%? The 2022 print (5.1%) predates the structural prefab/scale gains; the true new floor is unknown.

14. What Must Be True

Bull case — what must be true, and its falsification test. For the bull (compounding into or beyond the valuation), EMCOR must sustain ~9%+ consolidated operating margins (12–13% construction-segment) through any data-center demand normalization, keep book-to-bill above ~1.1×, convert backlog at high-single-digit-plus revenue growth, and hold a premium multiple — i.e., the margin step-up is structural (prefab/scale/mix/labor-scarcity persistence), not a cyclical pricing spike. Falsification test: book-to-bill below 1.0 for two consecutive quarters, a hyperscaler capex-down guide, or construction-segment margins compressing below ~10% on stable revenue — any one breaks the “peak is structural” claim and the bull case fails.

Bear case — what must be true, and its falsification test. For the bear (a ~−20%/yr de-rating), today’s margin must be substantially cyclical, data-center capex must digest within the next 1–2 years (a capital cycle, not a plateau), and the richest-ever multiple must compress as growth and margin roll over together. Falsification test: through a genuine data-center/macro softening, EMCOR sustains ~9%+ consolidated operating margins and >1.1× book-to-bill with OCF holding at/above NI — proving the step-up is durable and the diversification cushions the construction cycle, which breaks the bear case.

The honest investment-committee read: the bull and bear hinge on the same variable — the durability of the margin step-up, which is itself a function of how long data-center capex and the labor bottleneck persist. At ~$837 the price pays for the bull base case with little margin of safety. The diversification, balance sheet, and (vs FIX) multiple cushion make the left tail shallower than a pure data-center contractor’s — which is why this is a “full price, wait for a better entry” situation, not an “avoid/short” one.

15. Source Appendix

See the separate Source Appendix (EME_source_appendix.md) for the full citation list (primary filings, transcripts, and data sources). Key primary sources: EMCOR FY2025 Form 10-K; Q1-2026 Form 10-Q; FY2025, Q4-2025, Q3-2025 earnings-call transcripts; DEF 14A (2026-04-21); Form 4 corpus (2024–2026); aggregated fundamentals/valuation, own-history valuation percentiles and price history, and a quantitative factor model. Comparison drawn against the closest public peer, Comfort Systems USA (FIX).

APPENDIX A — Standard Diligence Questionnaire

Supplemental to the research memo; supports, not summarizes, the analysis. Fact / Interpretation / Assumption labels applied where it matters. Report date 2026-06-21; price ~$836.59.

General

What thoughtful questions have other investors asked?

• Is the ~9% consolidated operating margin (vs 5–6% historically) structural or peak-cycle?
• How concentrated is the data-center backlog, and what happens to RPO if hyperscaler capex digests?
• Is EMCOR genuinely cheaper than FIX, or is the lower multiple just a lower-margin business correctly priced? (Fact: cheaper on EV/EBITDA and EV/sales on a more diversified base; Interpretation: a real relative-value cushion.)
• How durable is the labor-scarcity advantage, and what is the realistic trough margin in a downturn?
• Why does executive comp lack a return-on-capital metric?

Cyclicality & Earnings Nature

Cyclical high or low? Near a cyclical/structural high. Operating margin doubled 5.1%→9.2% in three years; the question is how much reverts. (Interpretation.) External environment or internal actions? Both: a generational external demand shock (AI/data centers) and genuine internal execution (prefab/VDC, scale, selectivity, capital allocation). (Fact + Interpretation.) How stable are revenues? ~71% project construction (lumpy, re-bid, cyclical) + ~26% recurring facilities/industrial services (stickier, counter-cyclical). More stable than FIX’s ~93%-project mix. (Fact.) Outlook for products/services? Strong near-term: record RPO $15.62B, ~1.5× book-to-bill, FY26 guide raised to $18.5–19.25B rev / $28.25–29.75 EPS. (Fact.) Market size & direction? ~$700B+ U.S. MEP market, growing on data centers/reshoring/electrification/water; EMCOR ~2% share, U.S.-only post-U.K. sale. (Fact.)

Business Quality & Competitive Moat

Industry more or less competitive? Fragmented and competitive at the base, but the complex large-project data-center slice is supply-constrained (labor), handing scaled incumbents temporary pricing power. (Interpretation.) How profitable (ROIC/ROE)? ROE 23.6%, ROA ~15% (2025); high and rising; asset-light installer model. Tangible returns even higher (book inflated by Miller goodwill/intangibles). (Fact.) Industry profitability / barriers? Low barriers at the small end; modest barriers (scale, bonding, labor, reputation) for mega-projects. (Fact + Interpretation.) Easily understood? Yes — a decentralized installer/integrator of electrical & mechanical building systems plus facilities services. Undermined by foreign low-cost labor? No — on-site U.S. construction and field service; not offshorable. (Fact.) Do brands matter? Locally (operating-company brands/relationships) and at the execution-reputation level for hyperscalers; not consumer brand. (Interpretation.) Nature of competition? Project-by-project bidding (price + reputation/schedule certainty) for construction; relationship/contract renewal for services. (Fact.) Switching costs? Thin at the project level; moderate at the relationship/density/aftermarket level (repeat hyperscaler work, multi-year facilities contracts). (Interpretation.)

Financial Condition & Balance Sheet

Assets not on the balance sheet? The assembled, trained labor/supervision pool and customer relationships/reputation — the real moat — are not capitalized. (Interpretation.) Off-balance-sheet liabilities? Surety/performance bonds and operating-lease/contract obligations (normal for a contractor); no unusual exposures identified. (Fact.) How conservative is the accounting? Conservative/GAAP-pure — management refuses to add back amortization; immaterial SBC; honest disclosure of the U.K. gain and Miller amortization drag. POC revenue recognition is inherently estimate-driven (a watch item) but well-controlled. (Fact + Interpretation.) CapEx-hungry? No — capex <1% of revenue (~$113M FY25); FCF ~= net income. (Fact.)

Capital Allocation & Management

FCF generation & use? ~$1.19B FCF FY25; deployed to buybacks (~$586M), dividend, and M&A; net cash retained. Owner-oriented, FCF-funded (no leverage for buybacks). (Fact.) Significant acquisitions? Miller Electric (~$877M, Feb-2025) + Danforth + 9 tuck-ins (~$182M); disciplined, won’t chase 12–15× data-center shops. (Fact.) Buying back shares? Yes — share count −17% since 2022 (47.56M→44.69M diluted); ~$680M authorization remaining. (Fact.) Issuing shares to insiders? Immaterial SBC (~$21M); routine grants. (Fact.) Compensation policy? Adjusted diluted EPS + an operating-cash-flow-to-operating-income ratio; no ROIC/return-on-capital metric (a governance flag). (Fact.) Management motivations? Long-tenured CEO/Chair Guzzi (since 2011/2018), disciplined operator; low insider ownership (~0.73%); alignment via comp more than personal stake. (Fact + Interpretation.)

Valuation & Market Data

ADR / MLP / K-1? No — U.S. C-corp common stock, NYSE; standard 1099 dividend. (Fact.) Dividend policy? Small but growing; +60% to $0.40/qtr in 2026; ~3.5% payout, ~0.15% yield. (Fact.) How profitable? Net margin 7.5%; ROE 23.6%. (Fact.) NI vs cash from operations diverging? OCF/NI ~1.02× FY25 (clean conversion) but OCF fell YoY on a working-capital/AR build — worth monitoring as the POC float normalizes. (Fact.)

Risks & Downside

What would cause the stock to decline? A hyperscaler capex pause / data-center digestion, margin mean-reversion, a multiple de-rate from the richest-ever level, or a broad AI-capex scare (the early-2025 −27% drawdown is the template). (Interpretation.) Catastrophic-loss risk? Very low — net cash, diversified, profitable, asset-light. (Fact + Interpretation.) Total-loss risk? Negligible. (Interpretation.)

Recent News & Events

Business environment changed recently? Yes — accelerating data-center demand (record RPO/book-to-bill), FY26 guide raised, Miller Electric integrated, U.K. divested. (Fact.) Significant acquisitions/divestitures? Miller Electric (buy, Feb-2025); U.K. Building Services (sell, Dec-2025, ~$144.9M gain). (Fact.) Accounting-policy changes? None material; GAAP-pure. (Fact.) Other recent changes? Dividend +60%; routine board additions; Cantor reiterated Overweight ($1,123 PT, June-2026). (Fact.)

APPENDIX B — Source Appendix

Report date: 2026-06-21. Primary sources before secondary. Facts reconciled to filings; third-party aggregated market data cross-checked and labeled.

Primary — SEC Filings (EDGAR; CIK 0000105634; mirrored locally to output/EME/sources/)

1. EMCOR Group FY2025 Form 10-K (filed early 2026) — income statement, balance sheet, cash flow; five-segment revenue/operating-income/margin; RPO disclosures by segment and market sector; U.K. Building Services divestiture (12/1/25, net proceeds ~$256.6M, pre-tax gain ~$144.9M incl. ~$73.5M AOCI reclass); Miller Electric purchase accounting (~$876.8M; goodwill ~$327M, intangibles ~$475M); buyback/dividend.
2. EMCOR Group Q1-2026 Form 10-Q (filed ~May 2026) — Q1-26 revenue $4.63B (+19.7%), EPS $6.84 (+30%); total RPO record ~$15.62B (+33%); segment detail; share count 44.69M diluted.
3. FY2024 / FY2023 Form 10-Ks — prior-year segment revenue/margin; multi-year revenue and EPS history.
4. DEF 14A (2026-04-21) — executive compensation metrics (adjusted diluted EPS + operating-cash-flow-to-operating-income ratio; no ROIC metric); CEO/Chair Anthony Guzzi; insider ownership ~0.73%; board.
5. Form 4 corpus (2024–2026) — insider transactions: zero open-market purchases (code P); grants (A), tax-withholding (F), gifts (G), and 10b5-1-planned sales (S) by Guzzi, CFO Nalbandian, and directors.
6. 8-K material events (trailing 24 months) — quarterly earnings; Miller Electric completion (2/3/25, Item 2.01); board additions (Amy Dahl Dec-2024, Pat Roche Oct-2025); buyback authorization.

Primary — Earnings-Call Transcripts

7. Q1-2026 earnings call (2026-04-29) — CEO Anthony Guzzi, CFO Jason Nalbandian: data-center demand (“no sign of slowing”), N&C RPO +60%, book-to-bill ~1.5×, margin/mix commentary (mechanical 10.9%, electrical 12.1% incl. Miller amortization), FY26 guidance raise, capital allocation, “wouldn’t borrow to buy back stock,” “core product is field-labor supervision.”
8. Q4/FY2025 earnings call (2026-02-26) — FY25 results, adjusted EPS ~$25.87, sustainable construction-margin band 12.5–13.5%, fixed-price risk framing (“where we make most margin is where we take most risk”), U.K. divestiture, M&A discipline.
9. Q3-2025 earnings call (2025-10-30) — backlog/RPO trajectory, market-sector mix (data centers vs high-tech manufacturing roll-off), margin progression.

Secondary / Market Data (public)

10. Aggregated fundamentals & valuation data — multi-year income statement, balance sheet, cash flow, profitability ratios (ROE 23.6%, ROA 15.0%, margins), enterprise value (EV ~$26.9B at FY25 year-end / ~$36.7B TTM at current price), valuation multiples; peer comp pulls (FIX, PWR, MTZ, STRL, IESC, MYRG, PRIM). Third-party aggregated; reconciled to filings.
11. Own-history valuation percentiles (as of 2026-06-18) — P/E 28.2× (89.8th pctile), P/B 9.75× (97.0th), P/S 2.12× (97.9th), composite 94.9th percentile of EMCOR’s own ~10-yr history; TTM EPS ~$29.65. Own-history context only.
12. Five-year daily price history — 5-yr low ~$96.26 (2022-06-17), ATH ~$943.75 (2026-05-06), current ~$836.59 (−11.4% off ATH); 21/50/200-day EMAs; beta ~1.36, alpha ~0.41. Basis for the Five-Year Event Map.
13. Financial news — recent items including Cantor Fitzgerald reiterating Overweight, $1,123 PT (2026-06-16).
14. Quantitative factor model — stock loadings (Base model: Momentum +0.73, Value −1.15, DividendYield +1.11, Market ~+0.90, Quality −0.36, Growth −0.23); leaderboard (y1 +73%, Sharpe 1.83; lifetime max drawdown −70.6%); related/factor-peers (FIX 0.95, PWR 0.93, STRL, IESC, NVT + momentum ETFs); stock-info (beta, alpha, relative strength). Third-party statistical estimates; positioning overlay only.

Frameworks Applied

15. Analytical frameworks — Greenwald Competition Demystified (moat-type taxonomy; barriers-to-entry; ROIC/share-stability tests) and Marathon Capital Returns (supply-side capital-cycle analysis; the demand-side data-center over-build vs supply-side labor-bottleneck read).

Note: This is independent research and general information, not investment advice. No position in the security is stated or implied.