Thesis: BEP/BEPC is the direct-asset AI-power play: it OWNS the renewable generation and signs the hyperscaler PPAs (MSFT 10.5GW/2026-30 framework, Google 3GW hydro) while BAM only collects the fee — 10% FFO/unit growth is contracted and real, but BEP sits ~1.5% off 52wk highs with the AI-power-PPA story already in the price; BEPC is the cleaner 1099/C-corp US vehicle but trades at an ~8% premium to economically-equivalent BEP units

Scope & provenance. This is the direct-asset companion to the 2026-05-26 BAM deep dive. That deep dive established the structural split: BAM owns the management contract; BEP/BIP own the underlying assets and sign the PPAs. This artifact takes the BEP side. Event truth (contracts, FFO, capacity) is sourced from SEC EDGAR 6-K filings, company press releases, and credible trade press — cited inline. Price truth is a narrow price fetch, 2026-05-29 (BEP/BEPC are not in any watchlist). The two are kept separate per workspace rules.

Financials (cite before body)

Provenance: Massive returned No results found for both BEP and BEPC (expected — BEP is a Bermuda exempted LP / foreign private issuer that files 6-K/20-F, not 10-K/10-Q; ratios endpoint also 403'd as premium). Figures below are from SEC EDGAR (BEP 6-K FY2025 results, CIK 0001533232; BEPC 20-F FY2025, CIK 0001791863), the company FY2025 results press release, and a narrow market-data fetch (2026-05-29) for market cap and price. This satisfies the operating-company financials gate via EDGAR primaries; no financials_waiver needed.

Read: This is a contracted-cash-flow utility, not a momentum name. FFO/unit grew 10% in a year when GAAP net income swung from a loss to +$712M (the loss reflected mark-to-market and development non-cash items, not operating distress). The balance sheet is investment-grade with ~90% non-recourse asset-level debt — the structural feature that lets BEP run high gross leverage without holdco fragility. The 10%+ FFO/unit growth target is reaffirmed and is largely contracted (escalator clauses + commissioning of the ~84 GW advanced pipeline), which is the crux of the bull case below.

Tape snapshot (price truth — 2026-05-29 PT close, intraday/pre-close)

Tape is a same-day intraday/pre-close snapshot, not an official close. Returns above are simple price ratios from internal OHLC anchors, not annualized/total-return (which would add the ~5% distribution). No RSI/SMA computed — BEP/BEPC are not part of the standard scan universe.

Notable tape divergence: BEPC trades ~$2.89 (≈8%) above the economically-equivalent BEP unit, and sits much further from its own 52wk high (−11.5% vs BEP's −1.5%). Historically BEPC traded at a premium to BEP because of index inclusion + the broader buyer base (no K-1). The fact that BEP has rallied back to within 1.5% of its high while BEPC lags 11.5% off is a relative-value signal worth flagging (see Bear/Setup).

Setup (entry / stop / target framing)

This is framing for a contracted-yield direct-asset position, not a momentum trade. Two vehicles, two answers:

BEP (LP unit) — the cleaner relative-value entry today

• Entry zone: $34–36 (a pullback to the 30d base / prior consolidation; current $37.09 is ~1.5% off the 52wk high and not a great risk-reward entry). Add-on conviction zone $31–33 (re-test of April base).
• Stop: thesis-stop ~$30 (breaks the FY2025 uptrend base and would coincide with a >20% drawdown from the high — would signal either a rate-shock re-rating or a contract/FFO disappointment).
• Target: $44–48 over 12–18 months on continued 10% FFO/unit growth + re-rating as hyperscaler PPAs convert to commissioned capacity; total return enhanced by the ~4.2% distribution yield ($1.568 / $37.09).

BEPC (C-corp share) — the cleaner US-investor vehicle, but pay up / wait for the premium to compress

• Use BEPC only if you want 1099 reporting (no K-1) and IRA/index eligibility. It is economically equivalent (1:1 exchangeable into a BEP unit, identical distribution).
• Do not chase the ~8% premium. Preferred entry is when the BEPC/BEP spread compresses back toward 0–3%, or on a BEPC-specific pullback toward $34–36 (which would also restore the historical premium logic).

Position-sizing note: generation-side utility — sized as a yield/ballast position inside the ai-power-bottleneck cohort, not as a high-beta equipment name (BE/GEV/PSIX fill that role). Pairs naturally against the IPP leg (CEG/VST) which the perspective has graded weakening on regulatory overhang — BEP's contracted, non-merchant, non-recourse structure is the conservative expression of the same demand thesis.

Bull case

1. Direct ownership of the slow resource. Per the ai-power-bottleneck thesis ("you cannot add 500MW to the grid in 6 months"), the scarce asset is commissioned, contracted generation. BEP owns ~47,200 MW operating + ~200 GW pipeline and is the counterparty hyperscalers sign with at scale. Unlike BAM (fee on the AUM) or the merchant IPPs (CEG/VST, exposed to PJM capacity-price politics), BEP captures the spread directly via long-dated, escalating, investment-grade-counterparty PPAs.
2. The MSFT framework is real and verified (see Event Truth). 10.5 GW, 2026–2030 delivery — "almost eight times larger than the largest single corporate PPA ever signed" (BNEF: prior record 1.3 GW). This is the validation that the hyperscaler-PPA pipeline is a contracting reality, not narrative.
3. Counterparty quality is investment-grade hyperscaler. MSFT (AAA/Aaa), Google/Alphabet, Amazon — the strongest PPA off-takers in the market. Default/curtailment risk is minimal; these are 10–20yr take-or-pay-style structures with creditworthy buyers racing to secure power, which inverts the historical merchant-renewable buyer/seller power dynamic in BEP's favor.
4. Contracted 10%+ FFO/unit growth. FFO/unit +10% FY2025; management reaffirms 10%+ going forward, sourced from inflation escalators in existing PPAs, ~84 GW of advanced-stage development commissioning, and accretive M&A (Neoen privatization, Geronimo Power, Westinghouse nuclear partnership). Much of the growth is contracted, not market-dependent.
5. Repowering / brownfield economics. BEP's existing hydro + early-wind/solar fleet sits at already-permitted, already-interconnected sites — the scarcest commodity in the AI-power era. Repowering (replacing old turbines/panels at existing interconnects) sidesteps the 3–7yr interconnect queue that gates greenfield, generating high-IRR incremental capacity. Hydro in particular is the round-the-clock, dispatchable, irreplaceable-interconnect asset hyperscalers most want (hence the Google ~3 GW hydro framework specifically).
6. Capital recycling funds growth without dilution. Record ~$4.5B of FY2025 asset sales at ~22–25% IRRs recycles mature assets into the development pipeline — self-funding model that limits equity issuance, the historical knock on yieldcos.

Bear case / risks

1. It's already in the price. BEP is ~1.5% off its 52wk high after ~+59% trailing 12m. The AI-power-PPA story is well-known (the BAM deep dive flagged BEP/BIP as "arguably more compelling than the fee play" — consensus has caught on). Entry here pays near a high; the relative-value edge is in waiting for a pullback or using lagging BEPC's eventual mean-reversion.
2. PPAs ≠ FFO today. As the BAM deep dive noted, framework capacity gets built/financed over 5–10 years. The MSFT 10.5 GW is a 2026–2030 delivery pathway, not in-service capacity. Execution risk (interconnect, supply chain, cost inflation on the build) sits between the headline and the cash flow.
3. Rate sensitivity. BEP is a levered, yield-oriented infrastructure owner. A renewed rate-up shock compresses the multiple on the distribution and raises the cost of the asset-level (non-recourse but still floating-exposed) debt stack. The sister IPP leg (CEG/VST) already de-rated 36–38% off highs in May 2026 on a valuation-unwind, not a demand-collapse — BEP is structurally more insulated (contracted, non-merchant) but not immune to a rate-driven re-rating of the whole power complex.
4. The $10B figure is secondary-sourced (see Event Truth caveat). The capacity (10.5 GW), timeline, and framework structure are confirmed in the primary press release; the "$10 billion" commitment number is not in the GlobeNewswire/company primary release and traces to secondary reporting. Treat the dollar magnitude as widely-reported-but-not-primary-confirmed.
5. Amazon deal is small and old. The "reported AMZN deal" is a 601.6 MW PPA set across 7 projects from November 2022 (1,370 GWh/yr) — modest and predating the AI-power surge, not a fresh hyperscaler-scale catalyst comparable to MSFT/Google. Do not stack it as a 2024-2025 AI-power signal.
6. Complexity / Brookfield-ecosystem entanglement. BEP is externally managed by Brookfield (fee leakage to BAM), carries the LP/exchangeable-share dual structure, and consolidates JV/minority interests that make FFO-to-economic-interest mapping non-trivial. Key-person exposure to the Brookfield bench (Connor Teskey runs BEP and is the de-facto next-gen Brookfield operator per the BAM deep dive).
7. BEPC premium risk. Buying BEPC at an ~8% premium to the equivalent BEP unit means paying ~8% for the 1099/index convenience. If the premium normalizes toward parity, BEPC underperforms BEP on a pure-price basis even though they're economically identical.

BEP vs BEPC — which vehicle

Verdict: For a US investor who can tolerate a K-1, BEP is the cleaner relative-value entry right now (cheaper, closer to fair value-vs-self but the unit itself is near its high — see Setup). BEPC is the cleaner vehicle for tax-advantaged/index buyers, but at the current ~8% premium it is a wait-for-compression buy, not a chase. BEPC was created specifically to attract the buyers who structurally won't touch an LP — that demand is the source of its premium, and also the reason its premium tends to persist rather than fully close.

Cross-reference: ai-power-bottleneck perspective

Perspective log: 2026-05-03-ai-power-bottleneck (id ai-power-bottleneck, status active, priority high). The perspective is generation-side (where the electrons come from) vs its sister ai-power-delivery (grid-to-transistor voltage drop). Per the BAM deep dive's own framing: "BAM is fee-stream-on-AI-power; BIP/BEP are direct-asset-on-AI-power; BN is balance-sheet-leveraged AI-power."

BEP fits the perspective as the contracted-generation-owner node — the conservative, non-merchant expression of the generation-side thesis. It complements rather than duplicates the existing cohort: where BE/GEV/PSIX are the equipment/prime-mover leg (contracted-backlog OEMs) and CEG/VST/TLN are the merchant IPP leg (graded weakening on PJM/FERC regulatory overhang), BEP is the contracted, investment-grade-counterparty, non-recourse-financed generation owner — structurally the lowest-beta way to be long the same demand. This is a coverage gap in the perspective's key_tickers (the perspective currently holds the equipment + IPP + miner-conversion legs but no pure contracted-renewable-PPA owner). Flagged as a mutation candidate below — not applied (producer-only).

Event truth — MSFT / Google / Amazon deal verification

(Separated from price truth. All verified against primary/credible sources.)

• Microsoft — VERIFIED (capacity/timeline/structure); $10B figure secondary. GlobeNewswire primary release (2024-05-01) + Nasdaq + The Energy Mix + APPA: "over 10.5 GW of new renewable energy capacity," delivery 2026–2030, US + Europe (expandable to APAC/India/LatAm). Explicitly a global renewable energy framework agreement (a contracting pathway, not a single PPA). "Almost eight times larger than the largest single corporate PPA ever signed" — footnoted to BNEF (prior record 1.3 GW). Signed jointly by Brookfield Asset Management + Brookfield Renewable with Microsoft; ~1 GW already contracted outside the framework prior. ⚠️ The "$10 billion" commitment number is NOT in the primary press release — it traces to secondary reporting (e.g., trade press summarizing Microsoft's framework commitment). Capacity + timeline + framework structure are primary-confirmed; the dollar magnitude is widely-reported-but-not-primary.
• Google/Alphabet — VERIFIED. BEP FY2025 results disclose a Hydro Framework Agreement with Google for up to ~3,000 MW (3 GW) — significant because hydro is the round-the-clock, dispatchable, irreplaceable-interconnect asset hyperscalers most want.
• Amazon/AWS — VERIFIED but small/old. 601.6 MW of committed PPAs across 7 projects (3 US, 3 Europe, 1 India), ~1,370 GWh/yr, announced November 2022 (BusinessWire/Amazon press). Modest and predates the AI surge — a real contract but not a fresh AI-power-scale catalyst. Do not count it alongside MSFT/Google as a 2024-2025 signal.
• Meta — NOT independently verified. The task referenced "reported AMZN/GOOGL/META smaller deals." I found primary confirmation for MSFT, Google, and the 2022 AWS deal, but no primary source for a BEP–Meta renewable PPA in this pass. Flagging as unverified — do not state a BEP/Meta deal as fact.

Catalysts

• New hyperscaler framework/PPA announcements (Meta, Oracle, additional Amazon/Google tranches) — would extend the contracted pipeline and re-rate.
• MSFT/Google framework → commissioned-capacity conversion — the first GW of the 10.5 GW MSFT pathway coming in-service (delivery window opens 2026) converts narrative to FFO.
• Quarterly FFO/unit prints holding the 10%+ trajectory — the core thesis-confirmation metric each quarter.
• Westinghouse / nuclear partnership progress — the US-government nuclear partnership is an optionality leg outside the wind/solar/hydro core.
• BEPC/BEP premium normalization — a relative-value trade trigger independent of the fundamentals.
• Rate cuts — multiple expansion tailwind for the whole yield-infrastructure complex; conversely a rate-up shock is the main de-rating risk.

Mutations / follow-ups (producer-only — describe, do NOT apply)

1. Watchlist coverage decision. BEP/BEPC are in no watchlist (alt-managers has only BAM/BN/BX/KKR/APO/ARES/OWL/TPG/CG; the ai-power/ai-power-bottleneck cohort has no contracted-renewable-PPA owner). Recommend adding BEP (and optionally BEPC) to the ai-power / ai-power-bottleneck watchlist so it enters future scans. Not applied — lead to decide.
2. ai-power-bottleneck key_ticker candidate. BEP fills the perspective's contracted-generation-owner gap (complements the equipment leg and the weakening merchant-IPP leg). Recommend evaluating BEP for promotion to a key_ticker / watch_for entry in ai-power-bottleneck. Not applied. → Proposal filed 2026-06-04 as perspective log-entry 2026-06-04-bep-contracted-generation-owner-key-ticker-candidate (names BEP as the key_ticker symbol, BEPC as the exchangeable C-corp vehicle; trade-access + _securities.json family disposition flagged; tape refreshed 2026-06-04). README key_tickers still unedited — awaiting reviewer sign-off.
3. BIP sibling deep-dive still open (flagged in the BAM deep dive follow-ups) — BIP/BIPC is the infrastructure-side direct-asset play; this artifact only covers the renewable/BEP side.
4. Securities-family row — BEP (LP) + BEPC (exchangeable C-corp) are an economically-equivalent dual-listing pair; if added to watchlists, consider a _securities.json family row so capture dedup treats them as one economic entity.

Sources

Event truth (primary/credible):

• Brookfield + Microsoft 10.5 GW framework — GlobeNewswire primary release, 2024-05-01: https://www.globenewswire.com/news-release/2024/05/01/2873042/0/en/Brookfield-and-Microsoft-Collaborating-to-Deliver-Over-10-5-GW-of-New-Renewable-Power-Capacity-Globally.html
• APPA confirmation of MSFT 10.5 GW deal: https://www.publicpower.org/periodical/article/microsoft-enters-agreement-more-105-gw-renewable-energy
• BEP FY2025 record results (FFO $2.01/unit, Google 3 GW hydro, capacity/pipeline) — StockTitan recap of 6-K + company press release: https://www.stocktitan.net/news/BEP/brookfield-renewable-reports-strong-2025-results-and-announces-5-7lwyas68ktpf.html ; https://bep.brookfield.com/press-releases/bep/brookfield-renewable-reports-record-results-and-announces-5-distribution-0
• BEP FY2024 results (FFO $1.83/unit, liquidity, recycling IRRs) — company press release (Brookfield IR)
• BEP 6-K / 20-F filings — SEC EDGAR CIK 0001533232 (BEP) and 0001791863 (BEPC)
• Amazon/AWS 601.6 MW PPA, Nov 2022 — BusinessWire / Amazon press: https://press.aboutamazon.com/2022/11/brookfield-asset-management-selects-aws-to-enhance-its-renewable-energy-operations
• BEPC vs BEP structure (exchangeable, K-1 vs 1099) — OSC ruling + SEC FWP/424B3 filings (CIK 0001791863)

Price truth (internal):

• info, info (yfinance, narrow fetch 2026-05-29)
• *, *

Internal context:

• 2026-05-26-bam-deep-dive (parent BAM deep dive; structural split + BEP follow-up filing)
• 2026-05-03-ai-power-bottleneck (perspective log + README)