CBRE Group, Inc. (NYSE: CBRE) — Sold Like an Office REIT, Still Priced Like a Compounder

Independent equity research. Report date: 2026-06-21. Price reference: $131.55 (NYSE close, 2026-06-18).

⚡ Claude’s Take

This block is the author’s own independent opinion and general information only — not investment advice. The analysis that follows (Sections 1–15) is written position-free: it takes no recommendation and names no price target. The only opinion and the only valuation zone in this article appear inside this fenced block.

Verdict: HOLD — own-for-quality, accumulate-on-weakness toward the low-$120s/$130s; not a short. Directional fair-value zone ≈ $130–160 (≈16–19× normalized Core EPS of ~$8.00–8.50, or ~15–17× EV/Core-EBITDA on a ~$3.6–3.9B normalized base). At $131.55 the stock sits at the lower bound of that zone — fairly valued, with the margin of safety still thin rather than generous.

CBRE is the rarest thing in this back-catalog: a high-quality franchise that the market has actually marked down rather than melted up. It is ~23% off its January-2026 all-time high, sold in a May-2026 rate-driven real-estate purge that the factor model confirms was indiscriminate — CBRE trades factor-nearest to office REITs (VNO, BXP, SLG) with a −0.44 interest-rate beta, so when ten-year yields spiked, the world’s largest capital-light services firm got dumped alongside the levered landlords it merely advises. That is the variant-perception kernel: the tape is pricing CBRE as a rate proxy, while the business underneath grew Core EPS +25% in FY25, guided FY26 to +20%, and is compounding a genuine data-center/infrastructure annuity (~14% of Core EBITDA and growing >20%/yr). The bull case is that consensus mis-sorted a recurring-revenue compounder into the most-hated corner of real estate. The catch — and why this is a HOLD, not a buy — is that the de-rating only took CBRE from rich to fair, not to cheap. On honest Core metrics it still trades at the 75th percentile of its own decade (85th on price/book), at a clear premium to twin JLL on every multiple, with reported ROIC (~6%) below its cost of capital because the roll-up keeps burying tangible returns under acquisition goodwill, a comp plan with no capital-return governor, negative tangible equity, and almost no insider conviction (one open-market buy in five years). You are buying a good business at a fair price in a soft cycle — not a mispriced bargain.

Framing: de-rated quality cyclical / abandoned rate-proxy — not a falling knife (the y3 track record is +18.8% annualized) and not a momentum melt-up. Conviction: medium. Bullish trigger: a sustained yield retreat plus a clean second/third quarter of accelerating Advisory transaction volumes (US property sales/originations) that proves the recurring + cyclical engines are firing together → re-rate toward 20×. Bearish trigger: higher-for-longer persists, Advisory stalls, and FY27 Core EPS prints flat near $6.25 — confirming the factor model was right that this is a levered transaction-cycle play, not a compounder. Tag: “Sold like an office REIT, still priced like a compounder.”

📈 Stock Price Action — Five-Year Event Map

Factual price history, not a recommendation. Price moves are FACT (AZI five-year daily series); attributed drivers are INTERPRETATION.

The arc. CBRE round-tripped a full real-estate cycle. From a post-COVID recovery base near ~$86 (mid-2021) it climbed to ~$109 on the 2021 transaction boom, then collapsed through 2022 into a $65.76 trough (Oct-2023) as the Fed’s hiking cycle froze capital markets. From there it nearly tripled — riding a transaction thaw and an AI/data-center re-rating — to an all-time high of $171.61 (Jan-29-2026), before a renewed Treasury-yield surge knocked it back to $131.55 (close 2026-06-18). It now sits −23.3% off its all-time high, inside a 52-week range of $124.64–$171.61.

Cycle narrative. (1–2) The 2021 boom and 2022 collapse are textbook CRE-services beta to rates: as policy moved from zero to restrictive, transaction-fee revenue — CBRE’s most cyclical profit pool — evaporated and the multiple compressed in tandem. (3) 2023 marked the transaction trough; the $65.76 low coincided with the slowest sales and financing volumes of the cycle. (4–6) The 2024–25 recovery stacked two stories: a cyclical transaction thaw and a secular re-rating on data-center development and the recurring GWS/project-management base, carrying the stock to an all-time high and the multiple to the top of its decade range. (7) H1-2026 is the pivotal recent event: a renewed Treasury-yield surge in May-2026 triggered a broad rate-sensitive-real-estate sell-off, de-rating CBRE ~−27% from its January peak even though the Q1-2026 print was strong — a move driven by the discount rate, not the business.

1. Executive Summary

CBRE Group is the world’s largest commercial real estate (CRE) services and investment firm: FY2025 revenue of $40.55B (+13.4%), ~155,000 employees, operations in 100+ countries, and roughly 90% of the Fortune 100 as clients. Following a January-2025 reorganization it reports four segments: Advisory Services (leasing, capital-markets/property sales, mortgage origination, valuation — the cyclical brokerage), Building Operations & Experience (BOE) (facilities and property management plus the Industrious flex-office business — recurring), Project Management (Turner & Townsend — recurring/backlog), and Real Estate Investments (REI) (CBRE Investment Management + Trammell Crow development — lumpy/principal).

The central analytical move is to read CBRE on net revenue (~$23.8B), not gross: ~41% of headline revenue is reimbursable pass-through cost in BOE and Project Management. On net revenue, Core EBITDA margin is ~13.9% (vs a meaningless ~8% on gross). The widely repeated “GWS is ~70% of revenue” is true only on gross revenue and badly overstates the recurring book’s earnings weight — by segment operating profit, Advisory (the cyclical brokerage) remains the single largest profit pool at ~47%, with BOE ~28%, Project Management ~14%, and REI ~8%. The “pivot to recurring” is genuine and improving earnings quality, but CBRE remains materially levered to the transaction cycle.

The business has a real but narrow moat — economies of scale plus customer captivity concentrated in the multi-year BOE outsourcing and Project Management contracts (which grew and held through the 2022–23 transaction recession) — ringed by a large, low-moat, people-based brokerage where producers are portable and capital markets is commoditized. The disconfirming evidence sits in the returns: reported ROIC fell from ~9.5% (2021) to ~6.0% (2025) and ROE halved from ~24.7% to ~11.9% even as revenue grew +46%. Reported ROIC sits below a ~9–11% WACC — value destruction on the consolidated base — but this is an artifact of the acquisition roll-up: all book equity is goodwill + intangibles, and on tangible operating capital the franchise earns an estimated ~17–26%. The tension in one sentence: a good operating business earning mediocre returns on the rising pile of capital spent buying more of it.

Quality of earnings requires care. Core EPS of $6.38 sits 66% above GAAP diluted EPS of $3.85; the gap mixes legitimately non-recurring items (pension settlement, Telford remediation, transformation costs) with add-backs that recur because CBRE acquires every year (intangible amortization, integration costs that tripled to $303M). Core EPS is directionally fair but flatters sustainable earnings by an estimated $0.50–1.50/share. Capital allocation is mixed-to-slightly-below-average: disciplined buyback (share count down ~11% since FY21, no dividend, low SBC) offset by relentless goodwill-funded M&A (Industrious, Pearce $1.2B, Turner & Townsend; the earlier Telford Homes misstep), a comp plan with no ROIC/capital-return metric, negative tangible equity, net debt up to ~$5.7B, and effectively no insider buying.

On valuation, the H1-2026 drawdown did real work — from a decade-top multiple to ~20.6× trailing Core EPS / ~15–17× EV/Core-EBITDA, the 74.7th percentile of its own history — but CBRE is not cheap: it trades at a premium to twin JLL on every metric and at the 85th percentile of its own price/book. Management has guided FY2026 Core EPS to $7.60–7.80 (>20% growth), anchored on a “longer, slower, steady” transaction recovery (Q1-26 transactional revenue +22%) and a data-center/infrastructure annuity now ~14% of Core EBITDA. The market is underwriting roughly a base-case mid-cycle recovery — neither full bull normalization nor the bear’s higher-for-longer stall. The rate path is the swing variable.

2. Business Overview

CBRE traces to 1906 (San Francisco) and today is headquartered in Dallas. It is the scale leader of the global CRE-services industry by a wide margin, monetizing the built environment through four channels that span the full cyclicality spectrum — from pure transaction commissions to multi-year contracted facilities annuities to principal real-estate gains.

The four segments (FY2025, post-Jan-2025 reorganization). The table below is the analytical heart of the company: gross revenue, the reimbursable pass-through embedded in it, the resulting net revenue, and segment operating profit (SOP).

Advisory Services ($8,840M gross, ~$1,834M SOP). The classic brokerage and capital-markets business: office/industrial/retail leasing ($4,497M); capital markets — property sales advisory ($2,120M) and mortgage origination ($551M, under the CBRE Capital Markets brand, plus loan servicing); and valuation/appraisal ($815M). Revenue is commission-based and recognized at transaction close — the most rate-sensitive, most cyclical line in the company, and despite its modest gross-revenue weight it is the largest single profit pool (~47% of segment SOP) at a ~16% operating margin (pass-throughs are negligible here, so gross ≈ net). FY25’s strong growth (leasing +15.5%, advisory sales +20%) is the early-cycle recovery off the 2022–23 rate-shock trough, not structural growth.

Building Operations & Experience ($23,224M gross, ~$1,094M SOP). The new segment formed in January 2025, combining legacy Global Workplace Solutions (GWS) integrated facilities management ($20,645M), property management ($2,579M), and the fully-acquired Industrious flexible-office business. This is the recurring engine: multi-year contracts to run clients’ entire facilities estates, billed on a cost-plus/management-fee basis with large reimbursable pass-throughs (53.9% of segment revenue). Operating margin is a thin ~3% of gross revenue but ~10% of net revenue. The AI/data-center vector lives here — the BOE “digital infrastructure” and new “Critical Infrastructure Services” lines manage hyperscale/colocation/enterprise data centers, augmented by the November-2025 Pearce Services ($1.2B) acquisition of power/digital field-services.

Project Management ($7,657M gross, ~$561M SOP). Turner & Townsend (CBRE holds a majority stake), a global program/cost-management consultancy serving infrastructure, data-center, and corporate build-out programs. Backlog-driven and recurring, with a healthier ~16% SOP/net-revenue margin and structural exposure to the same data-center/infrastructure capex supercycle.

Real Estate Investments ($879M gross, ~$324M SOP). CBRE Investment Management ($155.5B AUM, mostly market appreciation; net inflows a modest +$1.6B) earning recurring management fees + carried interest, plus Trammell Crow development. The lumpiest segment: FY25 SOP included a $432M gain on disposition of real estate (vs $142M FY24) and absorbed a $132M Telford fire-safety remediation provision. Roughly $900M of embedded development gains sit across ~30 land sites, with data-center development the swing factor — and power availability the binding constraint.

Recurring vs transactional. Resilient/annuity lines (BOE facilities/property management, Project Management backlog, IM fees) now constitute the majority of net revenue and a rising share (~43%+ and climbing) of SOP; they grew through the downturn. Transactional/cyclical lines (Advisory leasing, capital markets, valuation; REI development gains and carried interest) remain ~50%+ of SOP and are the source of both the downside in 2022–23 and the recovery torque now. Verdict: a genuinely diversified services platform whose mix is shifting toward recurring revenue, but one that is still, at the profit line, half-levered to the real-estate transaction cycle.

3. Industry Dynamics

Structure: a scale-tiered global oligopoly over a long fragmented tail. By revenue: CBRE ~$40.6B > JLL ~$27B > Cushman & Wakefield ~$10B > Colliers ~$5.6B > Newmark ~$3.5B. CBRE is the clear #1, more than 1.5× its nearest rival, with a long tail of regional/local brokerages and facilities-management specialists beneath. The “big three” (CBRE, JLL, C&W) dominate large multinational occupier mandates and institutional capital-markets assignments, where scale, global coverage, and data are genuine entry barriers; the property/facilities-management and local-brokerage layers are far more fragmented and competitive.

Profit pools and cyclicality. The transaction pools (leasing, property sales, mortgage origination, valuation) are deeply rate-sensitive. Global CRE investment volume peaked at roughly $1.3T in 2021, fell ~34% into the 2022H2–2023 trough as the Fed spiked rates and bid-ask spreads froze deal-making, and recovered ~+13% (2024) and ~+19% (2025) to ~$850–900B — still ~30% below the 2021 peak. The recurring pools (facilities/property management, loan servicing, valuation contracts, IM fees) are structurally steadier and have grown through the cycle. This bifurcation is the defining feature of the industry: half the profit is a deep cyclical tied to the cost of capital, half is an annuity.

Secular vectors. Two genuine tailwinds favor the scale leaders: (1) the long-run outsourcing of corporate real estate and facilities to integrated providers (occupiers consolidating fragmented vendor rosters onto one global platform — exactly CBRE’s BOE/Project-Management strength); and (2) the data-center / AI-infrastructure capex supercycle, which drives leasing, development, project management, and critical-facilities operation simultaneously. Structural headwinds: the office “zombie” Class B/C stock and hybrid-work-impaired demand weigh on leasing and valuation in that sub-pool, and “higher-for-longer” rates cap the transaction recovery.

Marathon capital-cycle read. Capital is not flooding into brokerage — it is a talent-cost business with no asset build, so trailing transactional earnings understate mid-cycle power, a cyclically favorable setup for the survivors. But capital is flooding into data-center development and into engineering/project-management roll-ups (deal multiples rising), creating an overpaying-for-growth risk in CBRE’s own M&A. Verdict: a structurally AVERAGE industry (B/B−). Moderate barriers; the durable economics sit in the recurring outsourcing/servicing book (scale + captivity), not in cyclical brokerage. The secular tailwinds are real and disproportionately benefit the #1 platform — but they do not make commercial-real-estate services a structurally high-return industry.

4. Competitive Position

The moat, named. In Greenwald’s taxonomy, CBRE’s advantage is a narrow, line-specific blend of economies of scale + customer captivity, concentrated in BOE outsourcing and Project Management — not a wide franchise moat. Per the §9 discipline, each claimed advantage is tied to a financial outcome that would deteriorate without it:

• Scale and data. CBRE operates the largest CRE dataset and research platform (~500 researchers) and covers ~90% of the Fortune 100. This wins enterprise multi-line, multi-geography mandates that sub-scale rivals cannot service and enables cross-sell across leasing, facilities, project management, and capital markets. Financial fingerprint: market-share leadership and the ability to bid global outsourcing contracts. But: it is a breadth/share advantage, not pricing power — net-revenue EBITDA margins are only ~14% and BOE earns ~10% on net revenue. Scale here buys volume and stickiness, not fat margins.

• Switching costs (BOE / Project Management). This is the real moat. Once CBRE runs a client’s global facilities estate — embedded in their operations, data, and systems — switching is costly and disruptive, and contracts are multi-year. Financial fingerprint: BOE SOP grew +22% and held through the 2022–23 transaction recession; if these were re-bid commodity contracts, that line would not be that stable. Project Management’s backlog provides similar visibility.

• Brand / talent (capital markets, T&T). CBRE’s brand helps win institutional mandates and recruit top producers. But: in brokerage, talent is portable (producers can leave with their client books) and clients re-bid each transaction — so this supports share, not durable economics.

What is not a moat: the Advisory brokerage as a whole. Leasing and capital markets are people-driven, re-bid every deal, and purely cyclical. The disconfirming evidence is unambiguous and quantitative: if the brokerage franchise were a true moat, reported ROIC would not have collapsed from ~9.5% to ~6% and ROE would not have halved while revenue grew +46%. The mediocre, near-/sub-WACC consolidated returns are the proof that CBRE is, in aggregate, “scale without pricing power.”

Versus peers. CBRE is the largest and most diversified of the group, with the deepest recurring book and the most data-center optionality. JLL is the closest twin (and the factor model agrees — 0.956 similarity); Cushman & Wakefield is more transaction-levered and balance-sheet-constrained; Colliers (CIGI) has pushed hardest into higher-multiple engineering and investment-management diversification and carries founder-led ownership. CBRE leads on scale but is not the highest-return operator — Colliers’ diversified recurring mix arguably earns a cleaner premium.

Verdict: a real but narrow moat in the recurring BOE/Project-Management book, surrounded by a large, low-moat, cyclical, people-based brokerage. “Fortress in transition” is fair: the shift toward recurring SOP is genuine and is improving earnings quality, but the destination is a good-not-great compounder whose returns on capital have not earned their cost since 2021. This is a scale leader, not a pricing-power monopoly.

5. Growth History and Forward Opportunities

History. Gross revenue compounded at roughly a 10% 5-year CAGR, from $23.8B (FY20) and $27.7B (FY21) to $40.55B (FY25). But the headline overstates the organic story on two counts: (1) ~41% of revenue is pass-through cost that grows mechanically with reimbursable activity, and (2) CBRE acquires every single year — $1,535M (FY25), $1,203M (FY24), $330M/$558M/$1,336M before that — so a large share of reported growth is bought. Net revenue grew +14.1% in FY25 to ~$23.8B; the genuinely organic, recurring growth engine (BOE facilities, Project Management backlog) compounded steadily through the downturn while the cyclical Advisory pool swung from boom (2021) to trough (2023) and back to recovery (2024–25).

Segment growth in FY25. Advisory SOP +22% (leasing +15.5%, advisory sales +20% — cyclical recovery); BOE SOP +22% (recurring + data-center facilities + Industrious/Pearce additions); Project Management SOP +12%; REI SOP +24% (lumpy, driven by the $432M disposition gain).

Forward opportunities. Management has guided FY2026 Core EPS to $7.60–7.80 (>20% growth at the midpoint), raised from an initial $7.30–7.60, attributing two-thirds of the raise to higher full-year expectations rather than the Q1 beat. Three forward vectors carry the story:

1. Transaction recovery (Advisory). Q1-2026 transactional revenue rose +22% (a cycle high) — US property sales +64%, mortgage origination +53%, global leasing +18%. Management frames a “longer, slower, steady” recovery that does not require rate cuts, only a 10-year yield holding in a 4–4.5% band; sales volumes remain “well below prior peak,” leaving runway if the cycle normalizes.

2. Data-center / AI infrastructure (across all four segments). Total infrastructure-activity revenue exceeded $3B in 2025 (~14% of FY25 Core EBITDA). The new BOE Critical Infrastructure Services line was $1.7B in 2025, guided >60% growth; integrated Data Center Solutions is guided to ~$2B in 2026, +20%/yr; data-center leasing revenue more than tripled YoY in Q1-26; and CBRE cites managing over 1,300 data centers, with a new Meta talent-placement partnership. (All management figures — INTERPRETATION/hypothesis, to be validated against the 10-Q/10-K.)

3. Outsourcing penetration + M&A roll-up. Continued enterprise outsourcing wins in BOE, Turner & Townsend’s global program-management expansion, and tuck-in M&A (management is explicitly “M&A-first” on capital allocation, “especially in data centers”).

Verdict: mixed-quality growth. The recurring/secular vectors (data-center, outsourcing, project management) are genuine, high-quality, and structurally advantaged to the scale leader. But a large part of near-term EPS growth is cyclical recovery torque (Advisory) that the 2024–25 re-rating already front-ran, plus bought growth that dilutes returns on capital. High-quality engine, cyclically- and acquisitively-flattered headline.

6. Financial Quality

Read it on net revenue. On gross revenue ($40.55B) margins look thin and uninformative (GAAP operating margin 7.4%, “gross margin” ~18.7%). On net revenue (~$23.8B) the economics are respectable and improving: Core EBITDA margin 13.9% (FY25) vs 13.0% (FY24). Margins do expand modestly with scale on the correct base — but this remains a thin-margin, people-and-contract services business, not a high-margin franchise.

The ROIC-vs-WACC crux. Reported GAAP ROIC is ~6.0% (FY25), down from 9.5% (FY21) — below a ~9–11% WACC, implying the consolidated enterprise barely earns its cost of capital. But that is a roll-up artifact. All book equity is acquisition goodwill ($7,051M) + intangibles ($2,972M) = $10,023M, exceeding equity-before-NCI of $8,878M. Reconstructing NOPAT (≈$1,404M) against invested capital: on full invested capital ~9.1%; ex-goodwill ~16.9%; ex-goodwill-and-intangibles ~26.4%. The underlying operating franchise earns well above its cost of capital on tangible capital; it is the acquisition capital that drags reported returns to ~6%. ROE is ~11.9% (FY25), down from a transactional-peak-flattered 24.7% (FY21).

Cash flow. Operating cash flow is lumpy and converts poorly in growth years: $1,559M (FY25, OCF/NI 1.22×) but a collapse to $480M (FY23, 0.47×) on a −$944M working-capital swing. FY25 carried an −$882M accounts-receivable/unbilled build (partly offset by +$570M payables); the model is structurally working-capital-hungry as revenue grows. CBRE’s reported “free cash flow” (~$1.7B FY25, ~86% conversion) excludes the ~$1.5B/yr it spends on M&A — which is its true recurring capital outlay, not optional.

Balance sheet. Cash $1,864M; total debt ~$9,991M (incl. $2,405M capital leases); net debt $5,722M — up from $3,073M (FY24) on the M&A spree. Net-debt/Core-EBITDA rose to ~1.7× (CBRE cites ~1.2× “net leverage” on its own definition; either way at/near the top of its ~1–2× target band). Tangible common equity is negative (TCE ratio −1.9%) — book value is entirely intangible. Current ratio 1.09× is thin. Note the debt figure embeds warehouse/repo lines tied to mortgage origination that are not “permanent” leverage in the usual sense.

Verdict: Real operating quality (~17–26% on tangible capital, improving net-revenue margins) wrapped in a mediocre aggregate capital base — ~6% reported ROIC, lumpy and poorly-converting cash flow, negative tangible equity, leverage rising fast. Scale-with-economics on the franchise; scale-with-dilution on the consolidated returns. Economics improve with scale only if you measure the right thing.

7. Capital Allocation

M&A is the dominant use — and the central question. CBRE deploys ~$1–1.5B/yr acquiring: FY25 brought Industrious (flex office, full ownership Jan-2025) and Pearce Services ($1.2B, Nov-2025); recent prior deals include the majority stake in Turner & Townsend, J&J Worldwide / CBRE Government & Defense, and Direct Line Global (data-center FM). The roll-up is the engine of revenue growth — and the reason reported ROIC sits near WACC, because each deal adds goodwill that the operating earnings must then earn a return on. The Telford Homes UK residential-development acquisition is a documented misstep, wound down in 2025 with recurring fire-safety remediation provisions ($132M FY25 + $33M FY24 + prior charges).

Buybacks, no dividend. CBRE pays no common dividend — capital return is entirely repurchase. FY25 bought ~7.05M shares at ~$135.52 for ~$956M; FY24 ~$644M; FY23 ~$650M. Share count fell ~11%, from ~332.9M (FY21) to ~295.7M (FY25), with a $9.0B authorization (incremental $5.0B added Nov-2024) and ~$4.9B remaining. The FY25 average price (~$135.52) sat below early-2026 levels — decent timing. CBRE repurchased more aggressively in the past (~$1,850M in FY22); the buyback has been throttled to fund M&A. SBC is low (~$120M FY25), so dilution is minimal — the repurchase is real shrinkage, not an offset to grants.

The comp demerit. Per the 2026 proxy, the annual bonus is ~50% Core EBITDA / segment operating profit + 50% strategic, and the long-term plan is Core EPS performance awards + relative-TSR RSUs. There is no ROIC, ROE, or any capital-efficiency metric anywhere in the plan. A roll-up paid on Core EBITDA/EPS growth with no capital charge is structurally incentivized to keep acquiring to grow absolute numbers even when reported ROIC hovers near or below WACC — the same governance demerit flagged on Martin Marietta. The relative-TSR component is the lone mitigant.

Insider behavior. Across the full 308-Form-4 corpus (5 years), there is exactly one code-P open-market purchase: director Oscar Munoz, 2,100 shares at $115.29 (Aug-2024, ~$242K). Everything else is grants (A), tax-withholding (F), gifts (G), and sells (S). Most recently, CFO Emma Giamartino sold 2,250 shares at $130.74 (May-18-2026) with no 10b5-1 notation on the form; directors took routine grants/gifts in late May. The insider signal is neutral-to-soft — no conviction accumulation by management at the de-rated price, which is mildly telling.

Verdict: mixed-to-slightly-below-average. Disciplined buyback (−11% shares, good timing, low SBC, leverage held in band) is offset by relentless goodwill-funded M&A that suppresses consolidated ROIC to ~6%, a documented Telford misstep, a comp plan with no capital-return governor, negative tangible equity, and thin insider conviction. The market business is good; the capital deployed on top of it earns only adequate returns.

8. Changes and Headwinds — Last Two Years

Strategic / structural. The headline change is the January-2025 reorganization into four segments, creating Building Operations & Experience (folding Industrious into GWS facilities + property management) and elevating Project Management (Turner & Townsend) to a standalone segment. This was paired with a presentation shift (Q3-2025, SEC-driven) from net-revenue to gross-revenue reporting and a March-2026 reporting recast of data-center project management — presentation-only, but they complicate year-over-year comparability and warrant care.

M&A. Industrious (full ownership, ~$1B flex-office platform, expanding toward >300 locations by YE26) and Pearce Services ($1.2B, Nov-2025, digital/power infrastructure field services, contributing ~$90M EBITDA in FY26) extended the data-center/infrastructure build-out. Both validate the “M&A-first, especially in data centers” capital-allocation priority.

Guidance trajectory. Management has been in beat-and-raise mode: FY26 Core EPS guidance moved from an initial $7.30–7.60 (Feb-2026) to $7.60–7.80 (Apr-2026, >20% growth). Importantly, management flagged FY26 as front-loaded — Q1 exceeded plan by ~10%, partly on a pull-forward of Trammell Crow data-center land profits, with H2 facing tougher comps. A reader should not annualize H1 strength.

Headwinds. (1) The rate environment — the May-2026 Treasury-yield surge drove the ~27% drawdown and is the single largest swing factor for the transaction recovery; (2) office structural impairment — Class B/C “zombie” assets and hybrid work continue to weigh on leasing and valuation; (3) rising leverage — net debt up to ~$5.7B funding M&A into a still-soft cycle; (4) a BOE reporting artifact — roughly half the Q1 BOE “~25% growth” guide is a net-income-neutral D&A reclassification, not operating outperformance (validate against the 10-Q); and (5) CEO succession — no public succession plan for Robert Sulentic appeared on any of the last three calls, an open governance question for a founder-era-scale leader.

Verdict: the structural changes (recurring-mix shift, data-center build-out) strengthen the long-term thesis and earnings quality; the near-term changes (rate-driven de-rating, front-loaded guidance, rising leverage, succession opacity) are genuine pressures but cyclical/transitional rather than thesis-breaking.

9. Risk Analysis

The dominant risks are the interest-rate path (cyclical, high-impact, but mean-reverting) and the persistent sub-WACC reported ROIC (structural, the real challenge to the compounder narrative). Catastrophic loss risk is low: this is a capital-light, geographically- and segment-diversified services firm with no single-asset exposure.

10. Valuation Discussion — Embedded Expectations

Use Core metrics, not GAAP. GAAP P/E ≈ 34× ($131.55 / $3.85) is an amortization/integration artifact and overstates richness; ignore it. The honest anchors at $131.55:

• P/Core-EPS ≈ 20.6× (FY25 Core EPS $6.38); ~17× on the FY26 guide midpoint (~$7.70).
• EV/Core-EBITDA ≈ 15–17× (EV ~$50–57B / Core EBITDA $3,308M).
• Own-history percentiles (AZI, ~10-yr): composite 74.7th, P/E 72.2nd, P/B 85.9th (richest metric, 4.59×), P/S 65.9th.

The H1-2026 drawdown took CBRE from a decade-top multiple back toward its upper-median — elevated but no longer extreme. Crucially, though, it is not cheap on any lens: still ~75th percentile of its own history, ~86th on price/book, and at a clear premium to its closest twin.

Peer comparison (ROIC TTM, approximate):

CBRE trades at a premium to JLL on every metric — defensible on its larger recurring base, data-center optionality, and scale leadership, but it is a premium, not a discount. CBRE is dear vs its twin and moderately elevated vs its own history.

Embedded-expectations scenarios (FY27 Core EPS × exit multiple):

What the market is underwriting. At $131.55 / ~17× EV/Core-EBITDA, the market is paying for roughly a base-case mid-cycle recovery (~$7.5–8.0 FY27 Core EPS at ~17×). It is not paying for full bull transaction-normalization, but it is also not discounting the bear’s higher-for-longer stall. In every scenario, the multiple does more of the work than the earnings number — and the multiple is governed by the discount rate (the Treasury path). Embedded expectations are therefore reasonable-to-slightly-full: the de-rating priced out the froth but left CBRE valued as a quality compounder that must keep compounding to justify the premium to JLL. (FY26/27 Core EPS figures are ASSUMPTIONS built off FY25 $6.38 and CBRE’s beat-and-raise cadence; no price target is implied.)

11. Variant Perception

Consensus belief. CBRE is the scaled, best-in-class CRE-services compounder whose transaction-fee earnings are recovering off a 2023 trough, plus a secular data-center/recurring-revenue growth story that justifies a quality premium. The H1-2026 drawdown is a rate-driven pause in an intact recovery.

The factor-positioning evidence. The tape has thrown CBRE in with the office REITs: its factor-nearest peers after JLL (0.956) and Newmark (0.946) are VNO, BXP, SLG, HIW, CUZ, KRC — the most rate-sensitive, most beaten-up corner of real estate — and its interest-rate factor beta is −0.44. The −27% peak-to-trough drawdown is the office-REIT trade, not a CBRE-specific verdict. The variant question is whether that clustering is information (CBRE really is a levered transaction-cycle play and consensus’s quality premium is the error) or noise (a scaled, recurring-revenue compounder mis-sorted into the most-hated factor and sold indiscriminately on yields).

Strongest bull. The 2024–25 re-rating was right: the recurring mix (BOE facilities + Turner & Townsend, the majority of net revenue) structurally lowers cyclicality, and data-center development/operation is a genuine secular tailwind the office-REIT factor clustering completely misprices. As rates ease, the Advisory transaction engine re-accelerates on top of a recurring base that already grew through the downturn — driving Core EPS toward $9–10 and re-rating the multiple back above 20×. The May-2026 sell-off handed a quality franchise a cyclical entry point.

Strongest bear. CBRE is a levered bet on a rate cut that may not come. The −0.44 rate beta and office-REIT clustering are correct: ~47% of segment profit is transaction-fee revenue that does not recover until financing costs fall, and higher-for-longer keeps that pool depressed. The 2024–25 multiple expansion to a decade-top already front-ran a recovery that hasn’t fully arrived; on Core EPS the stock is not cheap (premium to JLL, 85th-pctile P/B), net debt is rising to 1.7× into a soft cycle, and reported ROIC sits below WACC. If yields stay elevated, FY27 Core EPS stalls near $6.25 and the multiple compresses — the bear corner is −33%.

The 3–5 assumptions that matter most:

1. The rate / yield path — does higher-for-longer persist, or does the transaction engine thaw? (The swing variable for both EPS and multiple.)
2. Recurring-revenue durability — is BOE + Project Management genuinely lower-cyclicality, or does it slow in a weak macro? (Determines whether the JLL premium is justified.)
3. Data-center / REI growth is real and monetizable — a durable secular annuity vs lumpy, capital-intensive, timing-dependent optionality.
4. The 2024–25 re-rating was earned, not borrowed — structural mix shift vs momentum overshoot now reverting.
5. Capital allocation creates value — buyback-funded M&A at ~1.7× net debt compounds vs levering into a trough at sub-WACC ROIC.

What falsifies each side:

• Falsifies the bull: yields hold/rise, Advisory revenue fails to inflect for 2+ more quarters, and Core EPS prints flat-to-down — confirming CBRE is the rate proxy the factor model already knows it is.
• Falsifies the bear: a yield retreat + a clean Advisory transaction re-acceleration with recurring segments holding, driving Core EPS toward $9 and re-validating the premium multiple — proving the de-rating was office-REIT-contagion mispricing of a structurally-improved franchise.

12. Fact vs. Interpretation

13. Open Questions

1. Durable data-center margins. Management calls the data-center push “as profound as our move into outsourcing… much faster.” What are the post-build-cycle margins and the recurring vs one-time (development-gain) split? The ~$900M embedded REI development gains are timing-dependent and power-constrained.
2. The BOE “~25%” reclass. Roughly half is a net-income-neutral D&A reclassification — what is the true underlying BOE operating growth rate net of presentation changes? (Reconcile to the 10-Q.)
3. Data-center count. “Over 1,300 data centers” vs ~800 a quarter earlier — definitional change or genuine step-up?
4. CEO succession. No public plan for Robert Sulentic on the last three calls. Who, and when?
5. True normalized ROIC. Does the next leg of M&A (especially data-center) earn above WACC, or does it keep diluting reported returns? What is management’s own hurdle rate (absent from the comp plan)?
6. Telford tail. Are the fire-safety remediation provisions fully reserved, or is there further UK residential liability?

14. What Must Be True

Bull case — what must be true:

• The transaction recovery continues without requiring rate cuts (10-yr holds ~4–4.5%), with Advisory volumes inflecting for multiple consecutive quarters.
• The recurring book (BOE + Project Management) proves genuinely lower-cyclicality and keeps compounding double-digit through any macro softness.
• Data-center/infrastructure growth is durable and reasonably-margined, lifting Core EPS toward $9–10 by FY27.
• Falsification test: if yields hold/rise and Advisory revenue fails to inflect for two or more additional quarters, with Core EPS flat-to-down, the bull thesis is broken — CBRE is the rate proxy the factor model says it is.

Bear case — what must be true:

• Higher-for-longer persists; the transaction-fee pool (~47% of SOP) stays depressed; FY27 Core EPS stalls near $6.25.
• The premium multiple (vs JLL, 85th-pctile P/B) compresses toward 14× as the market re-sorts CBRE as a cyclical, not a compounder.
• Goodwill-funded M&A continues to suppress reported ROIC below WACC.
• Falsification test: if yields retreat and Advisory transaction volumes re-accelerate cleanly while the recurring segments hold — driving Core EPS toward $9 and re-validating the premium — the bear thesis is broken; the de-rating was office-REIT-contagion mispricing.

The hinge: both cases turn on the rate/yield path and on whether CBRE’s recurring-mix shift has structurally lowered its cyclicality enough to deserve a compounder’s multiple. The evidence cuts both ways — hence a de-rated quality cyclical, neither a clean falling knife nor a clean bargain.

15. Source Appendix

See the Source Appendix below (Appendix B) for the full enumerated source list. Primary sources: CBRE FY2025 Form 10-K and FY2026 Q1 10-Q (SEC EDGAR, CIK 0001138118); the FY2025 earnings release and Q1/Q4/Q3 earnings-call transcripts; the 2026 DEF 14A proxy; the Form 4 insider corpus; multi-year fundamentals and ratios; a five-year price series and valuation-percentile history; and a quantitative factor model. Management guidance and data-center figures are sourced to the earnings calls and labeled as such (hypothesis to be validated against filings).

APPENDIX A — Standard Diligence Questionnaire

Supplemental to the research memo. Report date 2026-06-21. Price reference $131.55. Fact/Interpretation/Assumption labels applied where material.

General

What thoughtful questions have other investors asked about this company? The recurring debate is whether CBRE is (a) a structurally-improved, recurring-revenue compounder whose data-center and outsourcing businesses deserve a quality premium, or (b) a levered transaction-cycle play — a rate proxy — that the 2024–25 re-rating over-credited. Specific questions: Is reported ROIC (~6%) a real value-destruction signal or a goodwill artifact (INTERPRETATION: it’s an artifact — tangible operating returns are ~17–26%)? Is “Core EPS” (66% above GAAP) clean? Is the data-center growth durable annuity or lumpy development gains? Why does management refuse to put a capital-return metric in the comp plan? Can the buyback and ~$1.5B/yr M&A both be funded without over-levering into a soft cycle?

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Mixed — recovering off a cyclical low. Transaction volumes (Advisory leasing, capital markets) remain ~30% below the 2021 peak; FY25 Advisory growth (+22% SOP) is recovery torque, not a new high. The recurring book (BOE, Project Management) is near trend. So consolidated earnings are below mid-cycle on the transactional half. (FACT: global investment volume ~$850–900B vs ~$1.3T 2021 peak.)

Driven by external environment or internal actions? Both. The transaction cycle is externally driven (rates/cap rates). The recurring-mix shift, data-center build-out, and M&A are internal actions. The H1-2026 drawdown was entirely external (rate-driven sector sell-off; −0.44 interest-rate factor beta).

How stable are revenues? Bifurcated. ~50%+ of segment operating profit is recurring/contracted (multi-year facilities and project-management mandates, IM fees) and held through the 2022–23 downturn; the rest is transaction commissions that swing hard with rates. Net revenue grew every year 2020–2025; SOP fell in the 2022–23 transaction recession.

Outlook for products/services? Positive on recurring/secular vectors (outsourcing penetration, data-center facilities/development/project management); cyclically-dependent on the transactional half. Management guides FY26 Core EPS $7.60–7.80 (>20%) (FACT, mgmt — Apr-2026 call).

How big is this market — growing, shrinking, domestic or international? Global, ~$40.6B CBRE revenue is a fraction of a multi-hundred-billion-dollar global CRE-services + facilities-management market that is growing low-to-mid-single-digits structurally (faster in data-center/infrastructure). International (100+ countries); ~45–50% of revenue outside the Americas.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? Stable-to-intensifying at the margin — the top is a consolidating oligopoly (CBRE/JLL/C&W) but data-center and engineering/PM roll-ups are drawing capital and bidding up deal multiples (Marathon late-cycle signal in those sub-pools).

How profitable is the business (ROIC, ROE)? Reported ROIC ~6% (FY25), ROE ~11.9% — both well below 2021 peaks (9.5% / 24.7%). BUT ex-goodwill operating returns on tangible capital ~17–26% (INTERPRETATION: roll-up goodwill masks a high-return operating franchise). Net-revenue Core EBITDA margin ~13.9%.

How profitable is the industry — competitors, barriers to entry? Average. Top-tier scale/data/switching-cost barriers in outsourcing; brokerage has low barriers (people-based, re-bid each deal). Five meaningful global competitors plus a long fragmented tail.

Can the business be easily understood? Mostly — but the gross-vs-net-revenue distinction and the GAAP-vs-Core gap require sophistication; naive reading of headline revenue and GAAP EPS misleads.

Can it be undermined by foreign low-cost labor? Limited — services are local/relationship-based and on-site (facilities management, brokerage); some back-office/transaction-processing offshoring exists but is not a thesis risk.

Do brands matter? Yes, moderately — the CBRE brand wins institutional capital-markets mandates and enterprise outsourcing RFPs, and aids recruiting. But brand ≠ pricing power here (thin margins).

Nature of competition? Mandate-by-mandate competitive bidding in brokerage/capital markets; multi-year contract renewals with high switching costs in facilities/project management; talent recruitment for top producers.

Customers’ switching costs? High in BOE/Project Management (embedded operations, data, multi-year contracts); low in transactional brokerage (re-bid every deal). This split is the moat map.

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? Yes — the brand, the proprietary data/research platform, client relationships, and the ~$155.5B AUM franchise are under-represented relative to their economic value (the inverse of the goodwill problem). Embedded REI development gains (~$900M) are off-balance-sheet optionality.

Off-balance-sheet liabilities? Operating-lease commitments (capitalized under ASC 842, ~$2.4B capital leases); mortgage warehouse/repo lines tied to origination (transitory); guarantee/co-investment commitments in REI; Telford fire-safety remediation tail (partly reserved). (OPEN QUESTION on the Telford tail.)

How conservative is the accounting? Mixed. GAAP is conservative (intangible amortization, integration costs expensed); the “Core” adjustments are aggressive on recurring items (integration costs tripled to $303M are added back though M&A is perpetual). Revenue recognition (point-in-time for transactions, over-time for contracts) is standard.

How CapEx-hungry is the business? Physical capex is light (services firm). But CBRE’s true recurring capital outlay is M&A (~$1–1.5B/yr) plus REI co-investment — capital-intensive in disguise, and the reason reported ROIC is suppressed.

Capital Allocation & Management

How much FCF, and how is it used? ~$1.7B reported FCF (FY25, ~86% conversion), but ~$1.5B/yr goes to M&A. Residual to buybacks (~$956M FY25). No dividend. Philosophy: M&A-first (especially data centers), then REI co-invest, then repurchase.

Significant acquisitions recently? Yes — Industrious (full ownership Jan-2025), Pearce Services ($1.2B Nov-2025), majority of Turner & Townsend; earlier Telford Homes (misstep, wound down).

Buying back shares? Yes, aggressively and consistently — share count −11% since FY21 (~332.9M → ~295.7M), $9.0B authorization with ~$4.9B remaining. Good average-price timing.

Issuing large amounts to insiders? No — SBC low (~$120M FY25), dilution minimal.

Compensation policy? Bonus ~50% Core EBITDA/SOP + 50% strategic; LTI = Core EPS performance awards + relative-TSR RSUs. No ROIC/ROE/capital-efficiency metric (DEMERIT — a roll-up paid on growth with no capital charge). (FACT: 2026 DEF 14A.)

Motivations of management? Growth-and-scale oriented (consistent with the comp design); CEO Robert Sulentic long-tenured; no public succession plan (OPEN QUESTION). Insider buying essentially absent (one open-market buy in 5 years; CFO sold May-2026).

Valuation & Market Data

ADR, MLP, or K-1 issuer? No — ordinary US C-corp common stock (Class A), 1099 reporting. No K-1.

Dividend policy? None — zero common dividend; capital return is entirely buyback.

How profitable is the business? See above — mediocre reported ROIC, high tangible operating returns, thin reported margins on gross revenue / respectable ~14% Core EBITDA margin on net revenue.

Is net income diverging from cash from operations? Periodically yes — OCF/NI was 1.22× (FY25) but 0.47× (FY23) on working-capital swings; the model is WC-hungry in growth years (FY25 −$882M AR build). Core EPS diverges materially from GAAP (66% higher).

Risks & Downside

What would cause the stock to decline? Higher-for-longer rates stalling the transaction recovery (the dominant risk; −0.44 rate beta); office structural decline; an over-priced/poorly-integrated acquisition; a data-center growth disappointment; multiple compression as the market re-sorts CBRE as a cyclical.

Risk of catastrophic loss? Low — capital-light, diversified by segment/geography/client, no single-asset concentration. Leverage is moderate (~1.7× net debt/Core EBITDA) and the debt embeds transitory warehouse lines.

Chance of a total loss? Negligible — profitable, cash-generative, investment-grade-quality balance sheet despite negative tangible book (the negative TCE is a goodwill artifact, not distress).

Recent News & Events

Has the business environment changed recently? Yes — a May-2026 Treasury-yield surge drove a ~27% peak-to-trough CRE-sector sell-off (CBRE −23% off its Jan-2026 ATH) despite a strong Q1; the rate environment is the live swing factor.

Significant acquisitions? Pearce Services ($1.2B Nov-2025); Industrious full ownership (Jan-2025).

Change in accounting policies? Presentation changes — Jan-2025 four-segment reorganization; Q3-2025 shift to gross-revenue reporting; Mar-2026 data-center PM recast. No substantive accounting-policy change.

Recent changes — new markets, facilities, management? New BOE Critical Infrastructure Services line; expanding data-center/digital-infrastructure footprint; Meta talent-placement partnership; no parent-level management change (CFO Emma Giamartino, CEO Robert Sulentic; succession unaddressed).

APPENDIX B — Source Appendix

Report date 2026-06-21. Primary sources prioritized over secondary. Access date 2026-06-21 unless noted. Management commentary labeled as hypothesis to be validated against filings.

Primary — SEC Filings (CIK 0001138118)

1. CBRE Group FY2025 Form 10-K — consolidated statements of operations, balance sheet, cash flows; segment data (Advisory / Building Operations & Experience / Project Management / Real Estate Investments); MD&A (revenue, pass-through costs, net revenue, Core EBITDA/Core EPS reconciliations); Item 1 business description (~155,000 employees, ~90% of Fortune 100, 100+ countries). SEC EDGAR.
2. CBRE Group FY2026 Form 10-Q (Q1-2026) — quarterly segment results, transaction-revenue growth, leverage. SEC EDGAR.
3. CBRE 2026 DEF 14A (proxy statement) — executive compensation structure (annual bonus ~50% Core EBITDA/SOP + 50% strategic; LTI = Core EPS performance awards + relative-TSR RSUs; no ROIC/ROE metric); board composition. SEC EDGAR.
4. Form 4 insider-transaction corpus (5 years, 308 filings) — one code-P open-market purchase (director Oscar Munoz, 2,100 sh @ $115.29, 2024-08-29); CFO Emma Giamartino sale (2,250 sh @ $130.74, 2026-05-18); routine director grants/gifts (May-2026). SEC EDGAR.
5. 8-K material-event corpus (trailing ~18 months) — $5.0B incremental buyback authorization (2024-11-21); Industrious / new BOE segment (2025-01-14); quarterly earnings prints (2025-02-13, 04-24, 07-29, 10-23; 2026-02-12, 04-23); Pearce Services ~$1.2B (2025-11-04); reporting recast (2026-03-24); investor presentation (2026-06-15).

Primary — Earnings Calls

6. CBRE Q1-2026 earnings call (Apr-23-2026) — FY26 Core EPS guidance raised to $7.60–7.80 (>20% growth); Q1 transactional revenue +22% (US property sales +64%, mortgage origination +53%, leasing +18%); data-center/infrastructure figures; capital-allocation commentary. Management guidance — hypothesis.
7. CBRE Q4-2025 / FY2025 earnings call (Feb-12-2026) — FY25 results; initial FY26 guide $7.30–7.60; segment commentary. Management guidance — hypothesis.
8. CBRE Q3-2025 earnings call — data-center/infrastructure sizing (~10% of Core EBITDA at the time); transaction-recovery framing. Management guidance — hypothesis.

Quantitative Data Sources

9. Multi-year fundamentals (company filings / aggregated financial data) — multi-year income statement, balance sheet, cash flow; profitability ratios (ROE, ROA, ROIC); enterprise value and valuation multiples; company profile. Third-party aggregated; reconciled to filings.
10. Five-year daily price series — adjusted/unadjusted OHLCV, EMAs, beta, alpha. Used for the five-year price-action event map: ATH $171.61 (2026-01-29), 5-yr low $65.76 (2023-10), latest close $131.55 (2026-06-18), 52-wk range $124.64–$171.61.
11. Valuation-percentile history (own ~10-yr history) — composite 74.7th, P/E 72.2nd, P/B 85.9th (4.59×), P/S 65.9th percentile; ttm EPS $4.38, BVPS $28.69.
12. Financial news / recent-events feed — recent-events triage (May-2026 rate-driven CRE sell-off; CFO sale; data-center coverage; Q1 outperformance).
13. FactorsToday factor model (factorstoday.com/api) — stock loadings (Market +1.20, Sector Real Estate +0.76, InterestRate −0.44, Growth −0.31 in All-Factors model); leaderboard (y3 +18.8% ann, Sharpe 0.57; m6 drawdown; lifetime max DD −94.3% = GFC); related-stocks (JLL 0.956, NMRK 0.946, then office REITs VNO/BXP/SLG/HIW).

Peer / Industry Cross-Reference

14. Peer public filings — Jones Lang LaSalle (JLL), Cushman & Wakefield (CWK), Colliers International (CIGI), and Newmark (NMRK) annual/quarterly reports, used for industry structure, global transaction-volume context, comparative valuation, and the diversified-CRE-services framing.

Secondary — Industry / Trade Press

16. Trade and financial press on the data-center/AI-infrastructure CRE theme, the May-2026 real-estate-sector sell-off on rising Treasury yields, and peer comparisons (JLL, Cushman & Wakefield, Newmark). Used for context/triage; primary filings govern all quantitative claims.