Defense Stocks Are Down 22%–33% From Highs — A Blue Wave Can't Undo the Pentagon's $700B Backlog

The defensive trade of 2026 has not worked. WSJ flagged the week of April 23 as the worst of the repricing, with Northrop Grumman, Lockheed Martin, General Dynamics, and RTX all down as much as 11% in a single week. By mid-September, the damage has compounded. LMT trades at $537.25 against a 52-week high of $692 — a 22% drop — and NOC has fallen31% from $767. L3Harris is the worst performer, sitting at $250.74 against a $379 high.

The trigger was the morning Polymarket first crossed70% on its \"Blue Wave in 2026?\" market and never really looked back. With Democrats now priced at 81% to win both chambers (218+ House, 49+ Senate), the market is treating FY27 defense topline as if it will be cut — not held, not grown, cut.

Defense backlog is not a promise. It is a signed contract, funded by congressional appropriations already enacted. Of the $230.4B Lockheed Martin holds at June 28, $193.6B was already on the books at year-end 2025; the rest came from $65B in new orders booked in Q2 2026 alone. The single largest contributor is Missiles & Fire Control at $87.9B — PAC-3, GMLRS, JASSM, THAAD interceptors, the systems Trump personally praised for intercepting Iranian ballistic missiles in June.

General Dynamics reports the same dynamic from the other direction: $104.1B funded, $32.4B unfunded, plus $50.4B in estimated potential contract value, totaling $186.9B. Five straight quarters of orders-exceeding-revenue. Gulfstream deliveries of 41 jets in Q2 anchor the commercial side, but the defense book is what's doing the heavy lifting — and it can't be canceled. NOC's $105B is dominated by B-21 production acceleration that was agreed before the midterm polls were in sight.

[The backlog] already funds five to seven years of revenue for the average prime.

The cleanest expression of the disconnect is L3Harris. The company posted Q2 revenue of $5.88B (+8.4% YoY), an adjusted EPS beat of 28.4% over consensus ($3.63 vs. $2.83), and a record $42B backlog — yet the stock trades33% below its high. The Q2 print alone should have moved the bid by 10%. It didn't, because the buy-side is refusing to look at backlog in a blue-wave frame.

The $1.5T FY27 topline the White House requested on April 3 is actually $1.15T in base discretionary plus $350B in a third reconciliation bill. The base budget alone is a ~28% step-up from the FY26 baseline of roughly $900B — a number that survives any election outcome. The reconciliation $350B is the variable.

On June 10, Republican defense appropriators publicly conceded the $350B is unlikely to land. Senator Mitch McConnell, the defense subcommittee chairman, said it is \"safe to conclude there will not be another reconciliation bill,\" with Senator Susan Collins agreeing. The CSIS tracker from August18 notes that even the FY27 budget resolution to enable reconciliation3.0 only included $60B in mandatory defense funds — versus the $350B the Pentagon had penciled in. September 1 brought a continuing resolution funding the government only through December 11.

The market has been pricing this as catastrophic. It isn't.

Even in a worst-case where reconciliation delivers zero and the CR rolls into spring2027, the FY27 base budget still clears Congress at near-$1.15T. That's the floor, not the ceiling, because CR funding typically tracks prior-year enacted levels plus modest adjustments. The $230B backlog at Lockheed Martin and the $186.9B total contract value at General Dynamics were largely booked under FY25 and FY26 appropriations — money already spent by Treasury. A blue wave can change what FY28 looks like; it cannot un-book FY25.

The blueprint under which defense primes operate — multi-year procurement contracts, fixed-price development deals, options exercises on already-funded programs — is designed to be politically hard to reverse. A Patriot missile order cut doesn't just lose jobs in Alabama; it triggers contract termination liability and forfeits sunk R&D. The Government Accountability Office has repeatedly flagged that terminating major weapon system programs typically costs more than completing them.

This is why the 2026 record backlog matters even more than the headline number suggests. Lockheed Martin's PAC-3, Raytheon's Standard Missile and Tomahawk, Northrop's B-21, and GD's Virginia-class subs are all programs that have crossed the political Rubicon — funded, contracted, and on the production line. A new Congress can choose to buy fewer in FY28; it cannot retroactively un-buy what is already on order.

The January 7 Trump executive order restricting defense contractor buybacks and dividends adds a different kind of insurance. The order, signed at the height of the Iran crisis, capped executive compensation at $5 million and limited buybacks for \"underperforming\" contractors. By August 11, Senator Warren and Senator Lee reported the four largest primes had \"cut buybacks and dividend payouts dramatically.\" That capital is now being redeployed into capacity — exactly the kind of investment that, once made, anchors backlog for years.

The defense supply chain has two layers, and they get hit at different speeds. The primes — Lockheed Martin, General Dynamics, Northrop Grumman, RTX, L3Harris — have the contracted backlog. Their upstream suppliers are the more sensitive thermometer, because they ride on production rates rather than signed orders.

Howmet Aerospace, the castings and fasteners specialist, is the cleanest upstream read. Its August 6 Q2 2026 release notes that \"defense markets remain healthy\" with focus on missiles, hypersonic tracking, and propulsion. Howmet's commercial-aerospace book (8.5-year backlog for both commercial and defense castings) means it is hedged, but missile and hypersonic exposure makes it a leveraged play on the same prime backlog.

The transmission mechanism is structural, not political. Even in a maximally bearish scenario where FY28 procurement is cut 10%, the FY25-FY27 production line continues to draw parts. The aftermarket and sustainment revenue — roughly 40% of LMT's business, a similar share at RTX — is essentially unaffected by topline debates. This is why L3Harris sitting33% off its high while posting28% EPS beats looks like the most obvious dislocation in the group.

The short-term horizon (days to quarters) belongs to the political tape. Between now and the November 3 election, any poll showing Democrats expanding their lead will likely pressure defense multiples lower. The Polymarket-implied 81% probability already prices a clean sweep; if betting markets shift to 90%+, expect another leg down on the primes. The next two earnings prints from defense-adjacent tech (HEICO, Howmet) will set the tone for whether the supply chain confirms the prime-level strength or breaks it.

The long-term horizon (one to three years) belongs to the contracts. By mid-2027, the FY28 budget will be written — likely at $1.1T to $1.2T base, with reconciliation dead either way. Lockheed's $230.4B backlog will have converted to roughly $80B in revenue, leaving $150B still on the books at June 2027. The undrawn portion is, by construction, immune to the FY28 topline debate.

The thesis: the blue wave is a September-to-November story. The backlog is a 2027-through-2030 story. Markets that conflate the two are creating the buying opportunity.