Newmark Group filed its 10-K on March 2, reporting $3.29B revenue (+20.3%), $298M FCF, and a business that by every operational metric is executing well. Capital Markets revenue grew 35.3%. The loan servicing portfolio hit a record $211.2B. Market share in U.S. debt origination went from 1.5% in 2015 to 9.7% in 2025.

The stock trades at 7.5x forward P/E.

CBRE trades at 18.5x. JLL trades at 18.3x. CWK trades at 10.5x. The CRE brokerage peer average is 15-18x forward. NMRK's discount is 2.5x the sector.

So either NMRK is one of the most mispriced mid-caps in financial services, or the market sees something the 10-K would prefer you didn't.

What the 10-K says

Howard Lutnick departed to become Secretary of Commerce. His son Brandon, now CEO of Cantor Fitzgerald, controls 57.8% of NMRK voting power. Barry Gosin remains as CEO. Orderly transition. Business as usual.

This is technically accurate. It is also the most sanitized possible version of events.

What the 10-K doesn't say

Brandon Lutnick isn't learning the family business. He's building an empire across at least four simultaneous fronts:

Tariff refund arbitrage. Cantor allegedly purchased tariff refund rights at 20-30 cents on the dollar while Howard Lutnick, as Commerce Secretary, championed those same tariffs. When SCOTUS struck the tariffs down on February 20, those refund rights became worth par. Raskin and Wyden are demanding Cantor's documents by March 9. Cantor denies executing trades but acknowledges "certain salespeople explored brokering tariff trades." Fortune reports a $100B secondary market in tariff refund rights.

Critical minerals. Cantor is lead agent on a $1.6B capital raise for USARareEarth. The Senate has written letters questioning the Commerce Department connection.

Crypto. Cantor holds a 5% stake in Tether — roughly $600M invested, potentially worth $25B at Tether's estimated $500B valuation. Brandon personally verified Tether's reserves. A venture called "Twenty One" holds 42,000 BTC.

CRE brokerage. That's NMRK. The smallest piece of the Cantor portfolio by ambition, if not by revenue.

This is the governance discount the market is pricing.

The factor decomposition

I ran NMRK through a full factor decomposition against its peers. The results explain the discount better than the 10-K can.

NMRK's idiosyncratic variance: 33.8%. For reference, CBRE is 27%, JLL is 29%, CWK is 38%. The entire CRE brokerage sector trades as a factor bet on commercial real estate recovery. None of these are stock picks. They're sector vehicles with different packaging.

Breaking down NMRK's thesis into independent factors:

Total beta: ≈60%. Total idiosyncratic without edge: ≈35%. Total idiosyncratic with possible edge: ≈5%.

The only genuinely differentiated factor is NMRK's claim that revenue growth was productivity-driven with flat headcount — while CBRE, JLL, and CWK are all hiring aggressively. If that's structural (better technology, better producer economics), it's real alpha. If it's cyclical (existing producers doing more deals in a recovery), it evaporates in the next downturn. One year of data can't distinguish between the two.

The share gain paradox

NMRK claims debt origination share went from 1.5% to 9.7% over a decade. That's impressive and MBA/MSCI-verifiable. But all four major CRE brokers are claiming simultaneous share gains:

• CBRE: "Consistently gaining market share... rebuilt spectacular team."
• JLL: "Investment sales notably outpaced global market volume."
• CWK: "Capital Markets grew impressive 26%... early stages of talent expansion."
• NMRK: Investment sales up 50% vs 21% industry growth.

This is mathematically possible only if the Big 4 are consolidating share from smaller regional brokers — which is exactly what's happening. It's an industry consolidation dynamic, not an NMRK-specific competitive advantage. The share gain story, while real, is shared.

The insider signal that isn't there

Zero open-market insider purchases (Form 4 code P) despite the stock at 7.5x forward earnings. CEO Gosin gifted 300K shares in December. Brandon Lutnick gifted 145K shares. Gifts are tax planning, not conviction signals. If management believed the recovery narrative at these valuations, they'd be buying.

The options market confirms the absence of institutional attention: total open interest of 4,371 contracts across four expirations. The P/C ratio of 0.07 looks bullish until you see 623 of 2,169 call OI sitting at the $30 strike — 113% out of the money, almost certainly stale. Nobody is positioning.

Why it matters — and why it doesn't

The honest assessment: NMRK is probably slightly over-discounted. Fair value is likely 10-12x forward (a Lutnick governance penalty of 30-40% from peers, not 60%). That implies $22-28 vs $14.28 — a 50-100% upside if the discount compresses.

But for the discount to compress, you need one of: (a) Brandon Lutnick demonstrating competence over 3-4 quarters, (b) the Raskin investigation fading without enforcement, (c) Cantor selling its controlling stake. None of these is observable, predictable, or within our analytical capability.

Meanwhile, 66% of your return comes from CRE sector beta that CBRE delivers more cleanly at 1.34 beta (vs NMRK's 1.90), and sibling company BGC offers the same Cantor/Lutnick governance complex with an actual idiosyncratic catalyst — its FMX exchange has captured 40% of UST electronic trading and is growing quarterly, on a path to reclassification from interdealer broker (8-12x) to exchange (25-30x) multiples.

The prediction

70% probability NMRK's forward P/E stays below 12x through December 2026. The governance discount persists because Brandon Lutnick's multi-front empire building keeps institutional allocators away, the Raskin investigation runs through midterms, and zero insider buying signals management isn't backing the thesis with their own capital.

Short interest at 17.4% — 8-10x the level at CBRE, JLL, or CWK — says informed money agrees.