Graham
This is the archived full research. Read the story
A 90-year-old maker of vacuum and pumping equipment for refineries that now earns most of its money building parts for U.S. Navy submarines.
The scores
Growth3/4 →
Growing fast, but a material part was acquired and the growth narrows to one buyer, so not compounding under its own power.
Business quality3/4 →
Navy qualification is a real structural barrier, but the same rules price the work from cost, so it buys duration rather than returns.
Financial position4/4 ↑
No debt, ample undrawn credit and customer deposits funding working capital; trailing free cash flow is negative, but that is growth spending, not a gap.
Execution record2/4 ↑
Volume delivered and price did not: gross margin missed its guided range, and the margin goal once dated fiscal 2027 now arrives in fiscal 2029.
Per share capture2/4 ↓
Share count up 8.0% in two years and 6.1% in one quarter, a share-funded acquisition, no buyback, and adjusted per-share growth half the total.
Potential2/4 →
The upside is bounded and merely expected: orders yet to convert, a margin plan already slipped once, and somebody else's build rate.
Risk profile3/4 →
One serious dependency, disclosed and survivable - Navy programmes through a few primes, ten customers at 60% of sales. The India bribery referral stays open.
Valuation1/4 ↑
About 54.1 times operating profit, 161% of a three-stage value near $48.48 at an 8% return: priced for perfection, beyond the 20% margin.
Evidence quality3/4 →
Everything central is filed and current, corroborated by four outside origins, but the forward half rests on management targets graded rather than verified.
The Verdict
We rate Graham Unattractive. It builds custom vacuum and pumping equipment, and most of its money now comes from parts for US Navy submarines, with a record backlog. The latest quarter's margin jumped, but management's own forecast has it falling back, shipyards run late and orders fell, so the business is mixed. Even after a 22% fall, the price counts on management's three-year margin plan arriving in full and sits above our value and our best case. The margin holding in the second half, or Graham showing what the defence work earns, would change the view.
The long caseThe Navy wants submarines built faster and is paying suppliers to get there; Graham must show that defense work earns a rising margin, not just a bigger backlog.
How we weighed it
The price against our value
Against it
Price at 161% of our value
The price sits above our value and above the bull case.
The business
Cuts both ways
Mixed
A mixed business: a record backlog, but defense work earns less.
Growth against what the price assumes
Cuts both ways
Sales grew, but the forecast implies the margin fades and orders fell.
What could go wrong
Against it
Defense work earns less, shipyards run late and a small purchase may disappoint.
Against competitors
Cuts both ways
No filed rival compares directly; against its own normal it is dearer.
How its group is behaving
Cuts both ways
It sits in none of our ranked groups; Navy shipbuilding moves it.
The latest news and report
In its favour
The last report brought the margin sooner than expected, which argues against us.
The evidence under this - 8 claims
- 2026-11-05editedQ2 backlog and its slower conversion
- The fiscal 2029 target excludes depreciation and amortisation and share awards, so it is not comparable to an earnings multiple; on the same measure the trailing figure is about 39 times, which makes the forward number look like real progress rather than a full price.
- The multiple is struck on a company whose backlog is at a record and whose end market has a legislated, multi-decade build programme behind it, which is not the usual reason an industrial trades at four times sales.
- Accounts advised by T. Rowe Price paid $83.36 a share for $50M of new stock in April 2026, which is a large, informed buyer taking the other side within 6% of this price.
- 2026-11-05editedre-struck at the 4 November close on the Q2 balance sheet
- The same two figures - 8% to 10% organic growth and a low-to-mid-teens adjusted cash profit margin - were stated in the fiscal 2024 annual report as goals for the fiscal year ending March 31, 2027, and in the fiscal 2025 annual report as being on schedule for fiscal 2027. The target has moved out two years while the numbers stayed the same.
- The margin has been flat at about 10.7% for the three years the target has existed.
- Return on invested capital on the company's own vendor-computed basis is 5.9% on trailing twelve months, against the above-20% figure the target names.
- The identical 24.5% to 25.5% gross margin range was guided for fiscal 2026 in the prior year's annual report and the year came in at 23.5%, below the bottom of it.
- Q1 fiscal 2027 delivered 25.0%, inside the range, but with defense at 58% of sales rather than the higher commercial mix the full-year guide assumes.
- The guide was set in June 2026 and reaffirmed in August with one quarter reported, so most of it is still forecast.
- 2026-11-05editedQ2 release reaffirmed sales and adjusted cash profit and moved the gross margin, selling cost and capital spending guides
- The company says it is now winning competitive defense solicitations at market prices, and radar and directed-energy cooling work uses commercially available intellectual property that escapes certified cost and pricing entirely. If that share grows, defense growth stops being margin-dilutive.
- Certified cost and pricing does not cap margin at a low level by itself: a supplier that cuts its own costs between negotiations keeps the difference, which is exactly what the automation and welding investment is for.
- Aftermarket work on the installed defense base is also defense revenue and grew 20% in Q1 fiscal 2027; spares and overhauls are not priced the same way as new-build production.
- Curtiss-Wright, on the same programmes with the same buyer, earned a far higher segment margin than Graham's whole-company margin, which shows the regime does not force a low margin on everyone in it.
- 2026-11-05editedQ2 fiscal 2027 gross margin rose with defense at 66% of sales; the claim is qualified, not retired
The company
Refineries, power plants and submarines all need equipment that moves gas, heat and fluid without failing. Graham builds it: steam ejectors and condensers that hold a vacuum in a refinery's distillation tower, and turbopumps, blowers and turbines that move coolant and air inside submarines, torpedoes and rocket engines. Since January 2026 it also makes industrial mixers.
Almost every unit is built to one customer's specification, and 82% of revenue is booked as the work is done rather than on delivery.
The evidence under this - 2 claims
How it makes money
It is paid per engineered order, and the Navy work arrives through the shipbuilders rather than from the government directly. Once installed, the equipment needs overhauls and spares: management says more than $1B of it is in service, and aftermarket sales were $36.9M in fiscal 2026.
The evidence under this - 3 claims
Where the money comes from, stream by stream
| Stream | Share of sales (FY2026) | How it earns it |
|---|---|---|
| Defense | 60.1% | Ejectors, condensers, pumps, turbines and blowers for Navy submarines and carriers, sold mostly through the prime shipbuilders under long contracts. |
| Energy & Process | 34.0% | Vacuum and heat-transfer equipment for refineries and chemical plants, plus spares and overhauls, and since January 2026 FlackTek's mixers. |
| Space | 5.9% | Turbopumps and rotating equipment for rocket engines and space programmes, designed and built at Barber-Nichols. |
Who runs it
Matthew Malone has run Graham since June 2025, after years running its Barber-Nichols unit; Christopher Thome has run its finances since 2022. Insiders own little, about 2% between them. Their bonus rides mostly on cash profit and new orders, their long-term shares on sales and return on capital. Nothing pays them for a better profit on each sale, the thing this thesis is waiting for.
The evidence under this - 3 claims
- Half of each performance-share grant rides on return on invested capital, which a lower-margin mix does depress, so margin is not wholly unpaid.
- No metric pays on gross margin, the line the thesis turns on, while bookings and revenue reward the volume C22 says is priced from cost.
Why it wins, and who could beat it
Why customers come back
Equipment going onto a nuclear submarine has to be qualified to the Navy's own specifications, and swapping suppliers mid-programme takes years. That is why management says more than 80% of defense revenue is sole-sourced.
The same rules set the price. A sole-source award above $2.5M must show certified costs, and the government computes the profit from its own estimate of those costs. So the barrier buys duration, not margin: growing the sole-sourced share lowers the gross margin rather than raising it.
The evidence under this - 5 claims
- The figure appears only in an investor day slide; neither the annual report nor the quarterly report quantifies the sole-source share, so it cannot be checked against a filed table.
- On the Q1 call the chief executive said the company is now also winning competitive solicitations, which by construction reduces the sole-source share over time.
- Curtiss-Wright's own annual report describes supplying the same equipment family to the same buyer on the same three programmes, so being sole-sourced on a given part does not mean being the only qualified supplier on the platform.
- The company says it is now winning competitive defense solicitations at market prices, and radar and directed-energy cooling work uses commercially available intellectual property that escapes certified cost and pricing entirely. If that share grows, defense growth stops being margin-dilutive.
- Certified cost and pricing does not cap margin at a low level by itself: a supplier that cuts its own costs between negotiations keeps the difference, which is exactly what the automation and welding investment is for.
- Aftermarket work on the installed defense base is also defense revenue and grew 20% in Q1 fiscal 2027; spares and overhauls are not priced the same way as new-build production.
- Curtiss-Wright, on the same programmes with the same buyer, earned a far higher segment margin than Graham's whole-company margin, which shows the regime does not force a low margin on everyone in it.
- 2026-11-05editedQ2 fiscal 2027 gross margin rose with defense at 66% of sales; the claim is qualified, not retired
- 2026-11-05editedQ2 backlog and its slower conversion
The threat
On submarines, Curtiss-Wright already supplies pumps, motors, turbines and valves to the same three Navy programmes Graham names, from a naval business with $1.50B of sales. In refineries, Ingersoll Rand's mechanical vacuum pumps do an ejector's job without burning steam, so a new or steam-short plant may not buy an ejector at all.
The evidence under this - 2 claims
The rivals: CW, IR, and 1 more that is private, in-house or listed abroad.
Who else does this, rival by rival
| Ticker | Rival | How far along | Priced against it |
|---|---|---|---|
| CW | Curtiss-Wright CorporationPriced against it: Same product | On sale | Same product |
| IR | Ingersoll Rand (the Nash liquid-ring vacuum line, formerly Gardner Denver)Priced against it: Small part of it | On sale | Small part of it |
| Private | Hauschild SpeedMixer, Thinky Corporation and Resodyn Acoustic MixersPriced against it: Private | On sale | Private |
How big it could get
How big this could get
The market is measured in submarines. The Pentagon's goal is one Columbia and two Virginia class boats a year; the shipbuilders managed one Virginia a year as of June 2025, half the goal, and the two boats delivered in 2025 were each more than three years late. Graham's $470.5M of defense backlog takes up to six years to convert. Nobody publishes what one hull is worth to a supplier like Graham.
The evidence under this - 3 claims
- 2026-11-05editedQ2 backlog and its slower conversion
Who has signed
The Navy orders arrive through the shipbuilders: $101.5M of Virginia class and $75.0M of Columbia class follow-on orders in fiscal 2026. One defense customer is also paying $13.5M toward Graham's new Batavia building.
The evidence under this - 2 claims
The results
The numbers
Sales grew 16% over the twelve months to Sep 2026, about the same pace as the 17% in the year to Mar 2026.
Operating margin is 6.0%, up from 3.7% in the year to Mar 2024.
Running costs take 19¢ of each sales dollar, up from 17¢ in the year to Mar 2022.
Each dollar put into the business earns 8.2% a year after tax, up from 0.7% in the year to Mar 2023.
It turned 11% of sales into free cash, $28.9M over the last twelve months.
It holds $71.4M of cash against $0 of debt, $71.4M more cash than it owes.
The share count rose 12% in 4.5 years, and pay in shares costs 0.9% of sales.
All the numbers
Filed figures from the company's 10-K and 10-Q, read 2026-11-10. Fiscal years end in March.
| Mar 2022 | Mar 2023 | Mar 2024 | Mar 2025 | Mar 2026 | 12 mo to Sep 2026 | |
|---|---|---|---|---|---|---|
| Size and growth | ||||||
| Sales | $123M | $157M | $186M | $210M | $245M | $265M |
| Sales growth | +26% | +28% | +18% | +13% | +17% | +16% |
| Profit | ||||||
| Gross margin | 7.4% | 16% | 22% | 25% | 24% | 25% |
| Operating margin | −9.2% | 0.8% | 3.7% | 7.2% | 6.1% | 6.0% |
| Net margin | −7.1% | 0.2% | 2.5% | 5.8% | 5.1% | 5.0% |
| What growth costs, per sales dollar | ||||||
| Research and development | 3.1% | 2.6% | 2.1% | 1.9% | 2.6% | - |
| Selling, general and admin | 17% | 15% | 17% | 18% | 17% | 18% |
| Pay in shares | 0.7% | 0.5% | 0.7% | 0.9% | 0.9% | 0.9% |
| Capital spending | 1.9% | 2.4% | 5.0% | 9.0% | 6.5% | 3.7% |
| Cash | ||||||
| Cash from operations | −$2.22M | $13.9M | $28.1M | $24.3M | $15.9M | $38.7M |
| Free cash flow | −$4.54M | $10.2M | $18.9M | $5.36M | −$121K | $28.9M |
| Balance sheet | ||||||
| Cash and short-term investments | $14.7M | $18.3M | $16.9M | $21.6M | $6.58M | $71.4M |
| Debt | $18.4M | $11.7M | $0 | $0 | $13.0M | $0 |
| Net cash | −$3.64M | $6.51M | $16.9M | $21.6M | −$6.42M | $71.4M |
| Returns and shares | ||||||
| Return on capital | - | 0.7% | 5.3% | 11% | 9.3% | 8.2% |
| Shares (diluted) | 10.5M | 10.7M | 10.8M | 11.1M | 11.1M | 11.9M |
What growth costs
Growing has not made Graham much more profitable. Office and selling costs take about the same share of each sales dollar as before, so profit follows what each order earns over its cost, and the Navy work pulls that down.
The evidence under this - 3 claims
Selling and administrative cost fell from 17.7 to 16.9 cents in fiscal 2026 (C49), and the fiscal 2027 guide of 16.5% to 17.5% of sales (C13) leaves room for a little more.
- The company says it is now winning competitive defense solicitations at market prices, and radar and directed-energy cooling work uses commercially available intellectual property that escapes certified cost and pricing entirely. If that share grows, defense growth stops being margin-dilutive.
- Certified cost and pricing does not cap margin at a low level by itself: a supplier that cuts its own costs between negotiations keeps the difference, which is exactly what the automation and welding investment is for.
- Aftermarket work on the installed defense base is also defense revenue and grew 20% in Q1 fiscal 2027; spares and overhauls are not priced the same way as new-build production.
- Curtiss-Wright, on the same programmes with the same buyer, earned a far higher segment margin than Graham's whole-company margin, which shows the regime does not force a low margin on everyone in it.
- 2026-11-05editedQ2 fiscal 2027 gross margin rose with defense at 66% of sales; the claim is qualified, not retired
The last report (5 Nov)
Quarter ended September 2026, filed figures.
The second quarter of fiscal 2027 broke the pattern. Sales rose 6% to $69.8M, gross margin rose to 28.2% from 21.7% while defense grew to 66% of sales, and operating profit rose 37% to $5.8M. Management credits the mix of programs, the FlackTek acquisition and fewer low-margin material pass-throughs than a year before.
Orders fell 11% to $73.7M and the record backlog of $561.3M is converting more slowly, 30% to 35% within a year against 35% to 40% before. The sales and profit forecast held, which implies a second-half gross margin of about 24.5%, below the first half's 26.6%.
What it did to the thesisMargin arrived sooner than we expected, which weakens our argument that defense work pays less. The forecast implies it fades and orders fell, so the view stays Unattractive.
The quarter in detail
Sales growth from the previous quarter, filed figures.
Earnings per share $0.52 against $0.34 expected.
The evidence under this - 6 claims
- 2026-11-05editedQ2 fiscal 2027 gross margin added
- The identical 24.5% to 25.5% gross margin range was guided for fiscal 2026 in the prior year's annual report and the year came in at 23.5%, below the bottom of it.
- Q1 fiscal 2027 delivered 25.0%, inside the range, but with defense at 58% of sales rather than the higher commercial mix the full-year guide assumes.
- The guide was set in June 2026 and reaffirmed in August with one quarter reported, so most of it is still forecast.
- 2026-11-05editedQ2 release reaffirmed sales and adjusted cash profit and moved the gross margin, selling cost and capital spending guides
- Graham's own annual report says the fiscal third quarter is structurally the weakest because revenue is driven by direct labour hours and two holidays fall in it, so a straight-line comparison overstates the implied slowdown.
- The company has beaten its own revenue guidance before: fiscal 2026 was guided at $225M to $235M and came in at $245.293M, above the top of the range.
- Defense revenue moves on the timing of project milestones by the company's own account, so a strong Q1 can be a pull-forward rather than a run rate.
- 2026-11-05editedrewritten for the two quarters remaining after Q2
- 2026-11-05editedQ2 cash flow turned positive on deposit and receivable timing
What management did with the money
In April it sold 600,000 new shares for $50.0M and repaid $13.0M of debt; it now has no borrowings. In January it bought FlackTek for $36.2M, with up to $25M more owed if profit targets are met. The share count rose to 11.7 million from 10.8 million two years earlier.
The evidence under this - 3 claims
- 2026-11-05editedQ2 share count and cover-page shares
- 2026-11-05editedQ2 balance sheet and deposits
The price, and where we disagree
In short. Priced for perfection. The price is about 161% of what we think the business is worth, at an 8% yearly return.
The shares are down 22% since early September, to $69.52, after returning 85.8% over the past year on a record backlog and a three-year plan. That plan has already moved once: the growth and margin goal first set for the year to March 2027 now points to fiscal 2029.
The next check is the third-quarter results in early February 2027, against a full-year gross margin forecast now set at 25.0% to 26.0%. Meanwhile the shares sold short have nearly doubled since May, to 6.7% of those available to trade.
The evidence under this - 5 claims
- The same two figures - 8% to 10% organic growth and a low-to-mid-teens adjusted cash profit margin - were stated in the fiscal 2024 annual report as goals for the fiscal year ending March 31, 2027, and in the fiscal 2025 annual report as being on schedule for fiscal 2027. The target has moved out two years while the numbers stayed the same.
- The margin has been flat at about 10.7% for the three years the target has existed.
- Return on invested capital on the company's own vendor-computed basis is 5.9% on trailing twelve months, against the above-20% figure the target names.
- The identical 24.5% to 25.5% gross margin range was guided for fiscal 2026 in the prior year's annual report and the year came in at 23.5%, below the bottom of it.
- Q1 fiscal 2027 delivered 25.0%, inside the range, but with defense at 58% of sales rather than the higher commercial mix the full-year guide assumes.
- The guide was set in June 2026 and reaffirmed in August with one quarter reported, so most of it is still forecast.
- 2026-11-05editedQ2 release reaffirmed sales and adjusted cash profit and moved the gross margin, selling cost and capital spending guides
- The fiscal 2029 target excludes depreciation and amortisation and share awards, so it is not comparable to an earnings multiple; on the same measure the trailing figure is about 39 times, which makes the forward number look like real progress rather than a full price.
- The multiple is struck on a company whose backlog is at a record and whose end market has a legislated, multi-decade build programme behind it, which is not the usual reason an industrial trades at four times sales.
- Accounts advised by T. Rowe Price paid $83.36 a share for $50M of new stock in April 2026, which is a large, informed buyer taking the other side within 6% of this price.
- 2026-11-05editedre-struck at the 4 November close on the Q2 balance sheet
How we value it
- Method
- Cash the business will produce, in three stages
- Required return
- 8%
- Long-run growth
- 4%
- Moat
- An average moat
| Bear | Base | Bull | |
|---|---|---|---|
| Sales growth, years 1 to 3 | 7.9%, 5% | 15.7%, 10% | 19.6%, 12.5% |
| Normal operating margin | 7.9% | 9.9% | 10.8% |
| Return on new capital after year ten | 8% | 12% | 15% |
| Value a share | $29.20 | $48.48 | $60.75 |
- Price
- $78.02
- Priced on
- 6 Nov
- Price as a share of the base value
- 161%
- Starting growth the price pays for
- 36.3%
- Yearly return the price offers on the base case
- 6.4%
- Risk check: valued at 10%, the rate for a cyclical or indebted business
- $34.05
Where each figure comes from
| Figure | Source | Where exactly |
|---|---|---|
| Sales, the last twelve months | Filed report | Quarterly report (10-Q) for the period to 30 Sep 2026, filed 5 Nov 2026; the four quarters to that date. |
| Operating margin today | Filed report | Quarterly report (10-Q) for the period to 30 Sep 2026, filed 5 Nov 2026; the four quarters to that date. |
| Tax rate | Filed report | Quarterly report (10-Q) for the period to 30 Sep 2026, filed 5 Nov 2026; the four quarters to that date. |
| Sales per dollar of new capital | Filed report | Annual report (10-K) for the period to 31 Mar 2026, filed 8 Jun 2026: the year's sales over its operating capital. |
| Shares | Filed report | Quarterly report (10-Q) for the period to 30 Sep 2026, filed 5 Nov 2026; the diluted share count. |
| Cash and investments | Filed report | Quarterly report (10-Q) for the period to 30 Sep 2026, filed 5 Nov 2026; the balance sheet on that date. |
| Debt | Filed report | Quarterly report (10-Q) for the period to 30 Sep 2026, filed 5 Nov 2026; the balance sheet on that date. |
| Sales growth, years 1 to 3 | Analysts' forecasts | FMP's consensus for the years to 31 Mar 2027, 31 Mar 2028 and 31 Mar 2029: the average of 2 to 6 analysts' revenue estimates, 15.6%, 10.3%. FMP publishes how many analysts, not who they are. |
| Normal operating margin | Filed report | The best operating margin it has filed before amortisation of past deals, over five years and the latest quarter: a cyclical business is held to its record. |
| Return on new capital after year ten | Our method's rule | Set by our reading of the moat: the 8% rate with none, 12% with an average one, 15% with a strong one; the bear case earns the rate. |
| The bear and bull cases | Our method's rule | Our method's spread: the bear case takes half the growth and a fifth off the margin, the bull a quarter more growth and a tenth more margin. |
| Price | Market price | FMP's closing price on 6 Nov 2026. |
What each case rests on
Bear caseGrowth runs at half the analysts' path, 8% and 5%, the margin settles a fifth lower at 7.9%, and new capital earns only the 8% rate: the downturn test.
Base caseRevenue grows 16% and 10% over three years, the analysts' path (C57), then fades toward 4%, closing 30% of the gap each year; the margin moves to 9.9%, the analysts' year-three figure after stock pay.
Bull caseGrowth runs a quarter above the analysts' path, 20% and 12%, the margin reaches 10.8%, and new capital earns 15%.
What moves the base value
| Lower | Higher | |
|---|---|---|
| Growth in years 1 to 3, a quarter lower or higher | $44.89 | $52.37 |
| The normal margin, 15% lower or higher | $42.07 | $54.90 |
| Capital per dollar of growth, a quarter more or less | $47.58 | $49.03 |
| Return on new capital after year ten, at the rate or one step up | $40.07 | $51.85 |
| The rate, a point higher or lower | $39.82 | $62.90 |
| Long-run growth a point lower | $44.09 | - |
Value a share as growth and margin change
| Normal margin ↓ · growth → | 50% | 75% | 100% | 125% | 150% |
|---|---|---|---|---|---|
| 7.9% | $34.60 | $37.16 | $39.93 | $42.93 | $46.17 |
| 8.9% | $38.09 | $41.03 | $44.21 | $47.65 | $51.37 |
| 9.9% | $41.58 | $44.89 | $48.48 | $52.37 | $56.57 |
| 10.9% | $45.07 | $48.76 | $52.76 | $57.09 | $61.77 |
| 11.9% | $48.56 | $52.63 | $57.04 | $61.81 | $66.98 |
Growth is shown as a share of the base case's. Green: the price is at or under 80% of that value. Red: at or over 120%.
How we read the multiple
Counting cash and no debt, a buyer pays about 47 times the last 12 months' operating profit. Management's 2029 plan makes it look cheaper, but only on paper:
- If cash profit hits the $50M to $55M target in full, the price is about 14 times it.
- That target is before about $11M of depreciation and $6M of share awards and deal costs.
- So roughly a third of it never reaches an owner.
Against its own history. Dearer than its own normal. The business is bigger, but it kept no more profit from each dollar of sales until this quarter.
The evidence under this - 3 claims
- The fiscal 2029 target excludes depreciation and amortisation and share awards, so it is not comparable to an earnings multiple; on the same measure the trailing figure is about 39 times, which makes the forward number look like real progress rather than a full price.
- The multiple is struck on a company whose backlog is at a record and whose end market has a legislated, multi-decade build programme behind it, which is not the usual reason an industrial trades at four times sales.
- Accounts advised by T. Rowe Price paid $83.36 a share for $50M of new stock in April 2026, which is a large, informed buyer taking the other side within 6% of this price.
- 2026-11-05editedre-struck at the 4 November close on the Q2 balance sheet
- The same two figures - 8% to 10% organic growth and a low-to-mid-teens adjusted cash profit margin - were stated in the fiscal 2024 annual report as goals for the fiscal year ending March 31, 2027, and in the fiscal 2025 annual report as being on schedule for fiscal 2027. The target has moved out two years while the numbers stayed the same.
- The margin has been flat at about 10.7% for the three years the target has existed.
- Return on invested capital on the company's own vendor-computed basis is 5.9% on trailing twelve months, against the above-20% figure the target names.
Priced against its rivals
Only Curtiss-Wright sells the same thing to the same buyers as a large part of what it does. Ingersoll Rand does too, but as a small part of a much bigger company, so its price says little about this one. A price comparison needs at least two, so the price is judged against its own history instead.
Also checked and set aside: AMETEK (a substitute, not the same product); Kratos Defense & Security Solutions (a substitute, not the same product); Xylem (a substitute, not the same product); Honeywell International (a substitute, not the same product); L3Harris Technologies (a substitute, not the same product).
Private or foreign rivals that cannot be priced: Precision Castparts Corp. (PCC), Triumph Group (Triumph Aerospace), DC Fabricators, Croll Reynolds Company, Inc., Schutte & Koerting, Korting Hannover AG, GEA Wiegand GmbH, Edwards, Ltd., Concepts NREC, Collins Aerospace, Nikkiso Co., Ltd.; Donghwa Entec Co., Ltd.; Hangzhou Turbine Equipment Co., Ltd.; Oeltechnik GmbH; KEMCO; Chem Process Systems; Mazda (India), Holtec; Maarky Thermal Systems; Thermal Engineering International (USA), Inc., Hauschild SpeedMixer; Mazerustar Planetary Mixers; Thinky Corporation; Resodyn Acoustic Mixers, Westlake.
Where we disagree with the market
A record backlog and a Navy build-up make the margin plan look like earnings already booked.
The backlog is real, but the earnings are not arriving, because the fastest-growing work is priced from cost.
The story is told in backlog and adjusted cash profit. The filed operating line fell in fiscal 2026, and earnings per share rose only because the tax rate fell: at the old rate they would have been about $1.05, down 5.2%, instead of $1.12.
Q2 delivered the margin we said would change our mind, one quarter at a time. We would be wrong if gross margin holds near the first half's 26.6% through the second half while defense stays above half of sales, against a forecast that implies about 24.5%.
What analysts say, and the evidence
The evidence under this - 6 claims
- The lower rate was not a one-off trick: the 10-K attributes it to higher research and development credits and to discrete benefits on vesting restricted stock, both of which can recur while the share price is up.
- The One Big Beautiful Bill Act pushed the fiscal 2026 rate up by about 200 basis points and is expected to deliver about $8M of cash tax savings over two years, so the cash effect runs the other way to the reported rate.
- Management guides fiscal 2027 to 18% to 20%, so it is not presenting the low rate as sustainable.
- 2026-11-05editedQ2 fiscal 2027 gross margin added
- The identical 24.5% to 25.5% gross margin range was guided for fiscal 2026 in the prior year's annual report and the year came in at 23.5%, below the bottom of it.
- Q1 fiscal 2027 delivered 25.0%, inside the range, but with defense at 58% of sales rather than the higher commercial mix the full-year guide assumes.
- The guide was set in June 2026 and reaffirmed in August with one quarter reported, so most of it is still forecast.
- 2026-11-05editedQ2 release reaffirmed sales and adjusted cash profit and moved the gross margin, selling cost and capital spending guides
- Graham's own annual report says the fiscal third quarter is structurally the weakest because revenue is driven by direct labour hours and two holidays fall in it, so a straight-line comparison overstates the implied slowdown.
- The company has beaten its own revenue guidance before: fiscal 2026 was guided at $225M to $235M and came in at $245.293M, above the top of the range.
- Defense revenue moves on the timing of project milestones by the company's own account, so a strong Q1 can be a pull-forward rather than a run rate.
- 2026-11-05editedrewritten for the two quarters remaining after Q2
What could go wrong, and how we would know
The risks
- Defense work earns less. Management itself says defense carries a lower margin than the commercial business, and defense is now 60% of sales.
- Shipyards running late. The shipbuilders deliver half the boats the Navy wants, and their suppliers face worker and factory problems of their own.
- A deal that could disappoint. FlackTek depends on a newly launched mixer platform, and Graham added a risk factor warning its failure could force a write-down.
- An open bribery investigation. Graham reported misconduct at its India unit to the Justice Department and the SEC; the matter is unresolved.
The evidence under this - 6 claims
- On the same call the chief executive said newly won competitive defense solicitations allow healthier margins and market-bearing prices, so not all defense work is low margin.
- The company also attributes part of each gross-margin decline to material receipts rather than to defense mix alone, which is a timing effect that reverses.
What we still do not know
- What the defense work actually earns. No filing splits the margin between defense and commercial work.
- How much defense revenue is won in open competition, where Graham can charge market prices, rather than priced from its costs.
The evidence under this - 3 claims
- On the same call the chief executive said newly won competitive defense solicitations allow healthier margins and market-bearing prices, so not all defense work is low margin.
- The company also attributes part of each gross-margin decline to material receipts rather than to defense mix alone, which is a timing effect that reverses.
- The company says it is now winning competitive defense solicitations at market prices, and radar and directed-energy cooling work uses commercially available intellectual property that escapes certified cost and pricing entirely. If that share grows, defense growth stops being margin-dilutive.
- Certified cost and pricing does not cap margin at a low level by itself: a supplier that cuts its own costs between negotiations keeps the difference, which is exactly what the automation and welding investment is for.
- Aftermarket work on the installed defense base is also defense revenue and grew 20% in Q1 fiscal 2027; spares and overhauls are not priced the same way as new-build production.
- Curtiss-Wright, on the same programmes with the same buyer, earned a far higher segment margin than Graham's whole-company margin, which shows the regime does not force a low margin on everyone in it.
- 2026-11-05editedQ2 fiscal 2027 gross margin rose with defense at 66% of sales; the claim is qualified, not retired
What would change our view
- Unattractive→NeutralThe price falls back within 20% of fair value
It needs:
- ○The share price falls to about $58.18, back within 20% of what we think the business is worth.about ≈$56.53 today, 36% below the last close
- Unattractive→AttractiveThe price falls 20% under fair value
It needs:
- ○The share price falls to about $38.78, 80% of what we think the business is worth.about ≈$35.48 today, 60% below the last close
- Reopens the researchThe defense margin is finally disclosed
It needs:
- ○Graham discloses what the defense work earns, and it matches or beats the commercial business.checked at the next results, due by 5 Feb
We would also be wrong if
- Gross margin lands inside the guided 25.0% to 26.0% for fiscal 2027.
- The share count is flat or lower two years from now.
What is coming
- 6 November 2026 Q2 fiscal 2027 results: Whether gross margin holds at or above the guided 24.5% with defense above half of sales, and whether orders recover from Q1's 23.9% fall.
Every dated event ahead
- 6 November 2026 Q2 fiscal 2027 results: Whether gross margin holds at or above the guided 24.5% with defense above half of sales, and whether orders recover from Q1's 23.9% fall.
- February 2027 Q3 fiscal 2027 results: How much of the full-year revenue guide lands organically after FlackTek laps, and whether selling costs stay inside 16.5% to 17.5% of sales.
- June 2027 Fiscal 2027 year end, results and 10-K: The first FlackTek earn-out year against its roughly $3M cash-profit plan, the year's margin against guide, and the first step of the fiscal 2029 targets.
Since we wrote this
- 11 NovThesis reviewUnattractive, high conviction: the view holds even in our best caseConviction re-graded on the new anchors: how robust the view is, not whether anything is still open.
- 9 NovThesis reviewStill Unattractive: priced at 156% of a value near $49.90Re-valued in three stages at an 8% return, the method now drawn on every thesis: the price sits beyond the 20% margin of safety.
- 5 NovEarningsGross margin jumped to 28.2%; orders fell 11%; view heldThe margin test was met, but the forecast implies a weaker second half and orders slowed. The price fell 22% and still assumes the whole plan.
- 5 OctThesis reviewThe long case now sits under the verdictAfter the five-name pilot, the owner asked for the long case on every thesis.
- 4 OctThesis reviewOwner's checklist: revenue streams, cost of growth, management, catalystsThe owner's checklist found the method silent on who runs the company, what growth costs, where the money comes from and what is dated ahead; this pass fills those four blocks from the proxy and filings.
- 26 SepThesis reviewNext test corrected: Q1's 25.0% margin already broke the streakThe pressure branch spoke of a fifth straight quarter of gross margin below 24%, but Q1 fiscal 2027 came in at 25.0%. A miss now means falling back below 24%. View held.
- 9 SepThesis writtenUnattractive, medium convictionWritten a month after first-quarter results showed sales up 28.6% and earnings per share down 21.4%, with the shares up 85.8% in a year.
Also checked 3 times with nothing new to act on.
Sources 23
Sources - 23
Every claim above cites one of these, or states the reasoning that derived it. The quote is frozen at the date it was read: a link can rot, and an agent fetch returns a reading of a page rather than the page. Backs counts the whole claim ledger, so it can name a claim this page does not surface.
- S01primarycompany filing2026-06-08the income statement for fiscal 2026, 2025 and 2024; sales, orders and backlog by market; the competition table; customer concentration; the fiscal 2027 outlook; the acquisition terms; the India investigationbacksC01C02C03C05C07C10C12C13C17C25C26C28C29C33C34C35C38C40C43C45C48C49C54
- S02primarycompany filing2026-08-06the Q1 fiscal 2027 income statement, balance sheet and cash flow; orders and backlog; the FlackTek revenue contribution; the April 2026 share issue; the diluted share countbacksC01C05C08C09C10C11C13C17C18C27C29C35C38C40C42C48
- S03primarytranscript2026-08-06Graham Corporation Q1 fiscal 2027 earnings call transcriptdata/cache/transcripts/GHM-2027Q1.json (FMP earning-call-transcript, quarter label 2027Q1, call held 2026-08-06)management on why defense carries the lower margin and on wanting a 50-50 mix; the statement that submarine and carrier work is sole-sourced under certified cost and pricing; three analysts' questions on space, orders and margin
So the lion's share is in the submarine and aircraft carrier. That's sole sourced and under TINA cost and pricing. And then the other up-and-comings we've seen the ability to differentiate and get higher margin.
backsC06C13C20C41 - S04primarycompany filing2026-06-18the three-year framework to fiscal 2029; the sole-source share of defense revenue; the adjusted cash profit margin by year back to fiscal 2021; FlackTek's expected revenue and profit; the customer grants to datebacksC02C07C15C19C30C41C46
- S05primarycompany filing2026-06-08management's framing of the fiscal 2026 result alongside the annual reportread, but no claim rests on it
- S06primarygovernment data2026-04-22GAO-26-109068, Navy and Coast Guard Shipbuilding: A Disciplined, Strategy-Driven Approach Is Needed to Achieve Ambitious Goals U.S. Government Accountability Officethe actual submarine build rate against the Navy's goal, the one-Columbia-and-two-Virginia target, and the finding that suppliers' workforce and infrastructure problems threaten on-time quality delivery
according to the latest data provided by the Navy, the shipbuilders were working at a one-per-year-pace as of June 2025-half the rate of the Navy's two per year goal. While the Navy reported accepting delivery of two Virginia class submarines in 2025, both were over 3 years late.
backsC23C24C44 - S07primarygovernment data2025-03-11GAO-25-108136, Navy Shipbuilding: A Generational Imperative for Systemic Change U.S. Government Accountability Officethe same shortfall measured a year earlier, so the gap is a standing condition rather than one bad year
In June 2024, we found that the Virginia class submarine program's rate of production to meet fleet needs remains at 60 percent of the planned rate of two submarines per year.
read, but no claim rests on it - S08primaryregulator2026-09-09Federal Acquisition Regulation 15.403-4 and 15.404-4 (acquisition.gov) FAR Council (GSA, DoD and NASA)the rule the chief executive was describing: above $2.5M a sole-source price is built from the supplier's certified costs, and the profit objective is computed from the government's estimate of allowable cost
The threshold for obtaining certified cost or pricing data is $950,000 for prime contracts awarded before July 1, 2018, and $2.5 million for prime contracts awarded on or after July 1, 2018. ... Contracting officers shall use the Government prenegotiation cost objective amounts as the basis for calculating the profit or fee prenegotiation objective.
backsC21 - S09primarycompany filing2026-02-12Curtiss-Wright Corporation Form 10-K, year ended December 31, 2025 Curtiss-Wright Corporationa listed competitor's own description of supplying pumps, motors, generators, turbines and valves to the Navy on the Virginia, Columbia and Ford-class programmes, and the size of that businessbacksC31
- S10primarycompany filing2026-02-17Ingersoll Rand Inc. Form 10-K, year ended December 31, 2025 Ingersoll Rand Inc.the successor to Gardner Denver describing its own vacuum and blower range, the competing product on the Energy & Process sidebacksC32
- S11primarygovernment data2026-08-28EIA weekly series WPULEUS3, U.S. Percent Utilization of Refinery Operable Capacity U.S. Energy Information Administrationan outside check on management's claim that North American refineries run near capacity, which is the demand behind the aftermarket business
series-description: U.S. Percent Utilization of Refinery Operable Capacity, period 2026-08-28, value 98, units %
backsC39 - S12primarygovernment data2026-09-09USAspending.gov federal award recipient search, October 2024 to September 2026 U.S. Department of the Treasury (USAspending.gov)the fact that Graham holds almost no federal prime awards of its own, so its defense revenue reaches it through prime contractors rather than direct from the Navy
results: [{amount: 4650469.01, name: BARBER-NICHOLS, LLC, uei: K4GJKMQD4961}]
backsC25 - S13primaryvendor data2026-09-09Graham Corporation research pack, frozen 2026-09-09price, returns, the twenty-day range, the analyst split, the FINRA short-interest series, the Form 4 transaction codes, the vendor multiple series and the filed competitor cohort resultbacksC33C36C37
- S14primarycompany filing2025-06-09the fiscal 2026 outlook as originally issued, and the fiscal 2027 date then attached to the margin goal; the risk factors this year's set is measured againstbacksC28C45C46C49
- S15primarycompany filing2024-06-07the first statement of the 8% to 10% growth and low-to-mid-teens margin goal, with fiscal 2027 as its datebacksC46C49C54
- S16primarycompany filing2026-06-17the completion of the chief executive succession and the retirement of the executive chairmanbacksC47C51
- S17primarycompany filing2026-08-06the announcement of the MK19 Mod 2 air turbine pump award, which with the MK48 option year totalled about $43M, and management's framing of the quarterread, but no claim rests on it
- S18primarycompany filing2026-07-14executive officers and their tenure, beneficial ownership as of June 26, 2026, the cash bonus and long-term award metrics, goals and fiscal 2026 payouts, the leadership transitionbacksC51C52C53
- S19primarycompany filing2026-11-05Q2 sales, gross margin, operating income, adjusted EBITDA, orders, backlog and its conversion, cash, and the updated fiscal 2027 guidancebacksC05C10C13C35C42C55C56
- S20primarycompany filing2026-11-05shares outstanding on 5 November, diluted shares, customer deposits, the MD&A explanation of gross margin and the year-ago material receipts, cash flowbacksC10C17C35C42C55C56
- S21secondaryvendor data2026-11-05FMP end-of-day closes for GHM via thesis_delta, read 2026-11-05content/GHM/updates/2026-11-05-earnings.jsonthe 2026-11-04 close of $69.52
69.52 (2026-11-04)
read, but no claim rests on it - S22primarycompany filing2026-11-09trailing revenue, operating profit, stock pay, tax, debt, cash and the diluted share countbacksC58
- S23secondaryvendor data2026-11-09the analysts' average revenue and operating profit for the next three fiscal years, and the closebacksC57C58