Quanta Services
A contractor that builds and repairs America's power lines, substations and data-centre wiring, with one of the largest trained workforces in the trade.
The scores
Growth3/4 ↑
Revenue grew 34.0% in the first half of 2026, but about 13 points were bought; about 20.7% without acquisitions is still well above the market.
Business quality3/4 →
Scarce trained crews win share of utility budgets, but margins never rose, utilities employ most line workers themselves, and its own 10-K concedes low barriers.
Financial position3/4 →
Net debt near $5.6B at 1.7 times EBITDA, maturities spread to 2041, but acquisitions are debt-funded and gross debt rose again in August.
Execution record4/4 ↑
Guidance raised twice in 2026, eight straight quarters above consensus, and no material project charge found across 2023 to 2026.
Per share capture3/4 →
Diluted shares rose 1.0% in a year from stock pay and share-funded deals, with no buyback in 2026 - outrun by growth, not reversed.
Potential2/4 ↑
Utility spending plans are large, but the big transmission and generation work is not in backlog and depends on regulators and customers' decisions.
Risk profile3/4 →
No customer above 8%, but 70% of revenue rests on regulated utility budgets and 60.6% of revenue is fixed-price.
Valuation1/4 ↓
At 51 times operating profit the price needs a 12.2% margin on 2026 revenue; no listed rival earns above 5.8%.
Evidence quality3/4 →
Central facts are filed or from EEI, NERC, BLS and Census, but the moat and data-centre share rest on inference and management's words.
If it works 0.8× · if it fails you keep about 38%.
The Verdict
We rate Quanta unattractive. It builds and repairs America's power lines, substations and data-centre wiring, and it is a solid business: its trained crews let it grow faster than utility spending. But that edge wins more work, not better prices, and its profit on each sales dollar has barely moved in five years. The price assumes margins about double what it has ever kept, so it is stretched. As a bet it is fragile: even if it works, today's money shrinks a little; if growth stalls, well under half is left. We think the business is good and the price is not.
The long caseAmerica needs far more grid after years of flat demand, and someone has to build it; Quanta has the trained crews and scale that utilities themselves lack.
The evidence under this - 8 claims
The 2025 actual of $204.1B came in below EEI's own September 2025 projection of $207.9B, so the association's projections have run ahead of spending. The Census Bureau's measure of power construction actually put in place grew only 3.6% in the first seven months of 2026 (C31). And investor-owned utilities are not all of Quanta's utility customers - cooperatives, municipal utilities and power developers sit outside EEI's figure.
The share gain is flattered by what Quanta buys outside the utility budget: data-centre electrical and mechanical work sold to technology customers is inside the Electric growth rate and is not utility capital spending, so the comparison with EEI overstates the share taken from utilities. And the Q2 2026 margin rise to 7.3% from 5.5% may be the scarcity finally reaching price; the next two prints can overturn the 'volume, not price' half of this claim.
The first half of 2026 broke the pattern: gross margin 15.2% against 14.2% a year earlier and operating margin 5.9% against 4.7%, with Q2 2026 at 16.2% and 7.3%. If that holds through a full year it is the first evidence of pricing in the series.
The organic figure is an estimate built from the 10-Q's rounded acquired-revenue amounts, and 'acquired' there means businesses bought since the comparable period, so a business bought in mid-2025 counts as acquired for only part of the year. The call says acquisitions closed in Q2 2026 contributed only about $11M of adjusted EBITDA, so the quarter's margin gain is not the new companies.
The cohort members are each only partly comparable - utility grid work is 28.5% to 48.4% of their revenue (C22) - and all three are smaller, so part of the premium is a scale premium the cohort cannot price. Quanta also grows faster than the cohort median. Neither closes a gap of 2.4 times.
The peer median is not the only defensible multiple: a business with Quanta's scale, growth and record could deserve a lasting premium, and at 30 times rather than 21.5 the required operating profit is about $3.5B, which guided 2026 revenue reaches at an 8.8% margin - above anything Quanta has earned, but within reach of the chief executive's stated ceiling if corporate costs shrink as a share of revenue. The estimate also uses trailing profit, which in a business growing 34% lags the current run rate by about half a year.
The acquired amounts are the 10-Q's rounded figures and 'acquired' means bought since the comparable period, so a business bought in mid-2025 counts as acquired for only part of the year; the organic rates are approximate. Segment operating income is struck before corporate costs, amortisation and stock pay, so it flatters both segments by construction.
Most of the corporate line is amortisation of acquired intangibles and acquisition costs, and the larger buys of 2025 (Dynamic Systems) and 2026 sit partly in Underground and Infrastructure, so a revenue split may put too much of that cost on Electric and understate its profit; it also may not, since Cupertino Electric, bought in 2024, sits in Electric.
The company
When a utility needs a new high-voltage line, a substation, or storm repairs, it has two choices: use its own crews or hire a contractor. Quanta is the largest of those contractors, with about 69,500 employees at the end of 2025.
It sells the work, not the equipment: engineering, rights of way, towers, wire, and live-line repairs. It also wires data centres and factories. Utilities hire it for the peaks their own crews cannot cover. PG&E, one customer, uses about 39,000 contractor workers beside 29,000 employees.
The evidence under this - 4 claims
Hiring more than 7,000 people organically in a year says the labour pool is deeper than 'four years to make a journeyman' implies, at least at Quanta's wage. If a well-paid contractor can grow its craft workforce that fast, so can a rival or a utility, and the scarcity is a matter of pay and time rather than a wall.
How it makes money
Quanta is paid for work done, partly under fixed-price contracts (61% of 2025 revenue) and partly under long-running service agreements that commit the customer to no set amount. Utilities are 70% of revenue. Out of each dollar it keeps about 6 cents of operating profit, because the customer is regulated and the work is usually bid on price.
The evidence under this - 4 claims
The 5.7% is struck after $498.8M of amortisation of acquired intangibles, 1.7% of revenue, which management adds back in every adjusted figure it publishes. Before that line the operating margin was about 7.5%. Both are the company's own numbers; this thesis uses the GAAP line because the acquisitions that create the amortisation recur every year (C08).
Backlog is a non-GAAP measure Quanta itself warns 'may not be comparable to the methodologies used by other companies', and 41% of it is an estimate of future orders under agreements that commit the customer to nothing. The chief executive also says data-centre work is often booked and billed inside a quarter without ever appearing in backlog, so backlog can understate one part of the business while overstating the commitment in another.
- 2026-10-29editedextended with the September 30 backlog from the Q3 2026 release
Where the money comes from, stream by stream
| Stream | Share of sales (twelve months to 2026-06-30) | How it earns it |
|---|---|---|
| Electric | 81.8% | Builds, upgrades and repairs transmission, distribution, substations, generation and data-centre power for utilities, developers and large power users. |
| Underground and Infrastructure | 18.2% | Gas utility distribution, pipeline, industrial and refinery services for gas utilities and energy companies. |
Who runs it
Duke Austin has run Quanta since 2016 and Jayshree Desai its finances since 2022. Together the directors and officers own well under 1% of the company, and the chief executive sold a large block in May 2026. Most of the bonus rides on profit before the cost of past deals, and the share awards on return on capital and adjusted earnings per share.
The evidence under this - 3 claims
The figure is already stale on the low side: Austin sold 155,992 shares on May 5, 2026 (C28), after the record date, so insiders now hold less than the 0.6% shown.
- Return on invested capital is struck on adjusted EBITA but over total assets less cash and non-interest-bearing current liabilities, so goodwill from deals sits in the denominator and buying expensively does cost the plan something.
- The EBITDA bonus and the adjusted EPS award both leave out the amortisation of what was bought (C09), so an acquisition paid partly in shares can lift both without the GAAP margin moving.
Why it wins, and who could beat it
Why customers come back
The scarce thing is trained line workers. It takes about four years to make one, and the whole country has about 131,000. When utility budgets jump, the contractor with the most crews gets the work.
That shows up as share. Quanta's electric business grew about 27.5% in the first half of 2026 without counting companies it bought, while utility budgets are projected to rise 17%. It does not show up as price: its gross margin was 15.0% in both 2021 and 2025.
The evidence under this - 4 claims
The industry split cuts both ways: the buyer already employs about 1.6 line workers for every one in a line-construction firm, which is the strongest evidence that the customer can do this work itself. A 10% rise over ten years is about 13,600 people, which Quanta's own claim of more than 7,000 organic hires in a year (C18) says is not the binding limit on one company.
The share gain is flattered by what Quanta buys outside the utility budget: data-centre electrical and mechanical work sold to technology customers is inside the Electric growth rate and is not utility capital spending, so the comparison with EEI overstates the share taken from utilities. And the Q2 2026 margin rise to 7.3% from 5.5% may be the scarcity finally reaching price; the next two prints can overturn the 'volume, not price' half of this claim.
The first half of 2026 broke the pattern: gross margin 15.2% against 14.2% a year earlier and operating margin 5.9% against 4.7%, with Q2 2026 at 16.2% and 7.3%. If that holds through a full year it is the first evidence of pricing in the series.
The threat
The biggest rival is the customer. Utilities employ about 62,000 of the country's line workers, more than all line-construction contractors together, and Quanta's own annual report warns they may do more themselves. One Midwest utility reportedly chose to build its own gas plant for lack of contractors. MasTec, MYR Group and Primoris compete for the same jobs.
The evidence under this - 4 claims
The industry split cuts both ways: the buyer already employs about 1.6 line workers for every one in a line-construction firm, which is the strongest evidence that the customer can do this work itself. A 10% rise over ten years is about 13,600 people, which Quanta's own claim of more than 7,000 organic hires in a year (C18) says is not the binding limit on one company.
It concerns generation construction, where Quanta already says it will not take combined-cycle risk on the contracts offered, not the transmission and distribution work that is most of the Electric segment. The utility is unnamed and the report is second-hand from an analyst.
The rivals: PCG, MTZ, MYRG, PRIM.
Who else does this, rival by rival
| Ticker | Rival | Where it competes | How far along | Priced against it |
|---|---|---|---|---|
| PCG | The utilities' own line crews - the in-house workforce of customers such as PG&EPriced against it: Different product | Electric (Electric Infrastructure Solutions) | On sale | Different product |
| MTZ | MasTec, Inc. (Power Delivery segment)Priced against it: Yes | Electric (Electric Infrastructure Solutions) | On sale | Yes |
| MYRG | MYR Group Inc.Priced against it: Yes | Electric (Electric Infrastructure Solutions) | On sale | Yes |
| PRIM | Primoris Services Corporation (Utilities segment)Priced against it: Yes | Underground and Infrastructure | On sale | Yes |
How big it could get
How big this could get
Investor-owned utilities spent $204.1B on their systems in 2025 and plan $238.8B for 2026, about half of it on transmission and distribution lines. Quanta's utility revenue was roughly a tenth of that total.
But government figures for power construction actually built grew only 3.6% in the first seven months of 2026, far below the plans. Some of the spending boom is equipment prices, not more work.
The evidence under this - 3 claims
The 2025 actual of $204.1B came in below EEI's own September 2025 projection of $207.9B, so the association's projections have run ahead of spending. The Census Bureau's measure of power construction actually put in place grew only 3.6% in the first seven months of 2026 (C31). And investor-owned utilities are not all of Quanta's utility customers - cooperatives, municipal utilities and power developers sit outside EEI's figure.
The Census series is dollars of construction completed, while EEI's figure is capital spending including equipment, and equipment - transformers, breakers, turbines - is where prices and lead times have risen most (C32). The two can diverge without either being wrong, but a 3.6% rise in construction put in place is not a market growing at 17%.
Who has signed
Quanta had $53.44B of backlog at June 30, 2026, but 41% of it is estimated orders under agreements that commit customers to nothing, and most contracts can be cancelled at short notice. The very large new transmission lines are not signed yet: management says they reach the field from late 2027.
The evidence under this - 2 claims
Backlog is a non-GAAP measure Quanta itself warns 'may not be comparable to the methodologies used by other companies', and 41% of it is an estimate of future orders under agreements that commit the customer to nothing. The chief executive also says data-centre work is often booked and billed inside a quarter without ever appearing in backlog, so backlog can understate one part of the business while overstating the commitment in another.
- 2026-10-29editedextended with the September 30 backlog from the Q3 2026 release
Large transmission has been 'just starting' for a decade - only 322 high-voltage miles were completed in 2024 (C16) - and the projects that are not in backlog are by definition not committed. Upper single-digit growth in the core utility business is roughly what EEI's spending growth would give without any share gain, and it is well short of the 20%-plus growth the price is paying for.
Where the growth is coming from, business by business
Where the growth is coming from
| Business | Share of revenue | Growth: last quarter | Last half-year | Last year | Share in 3 years at its half-year rate |
|---|---|---|---|---|---|
| Electric (Electric Infrastructure Solutions) | 82% | +43.6% | +37.5% | +31.4% | 87% |
| Underground and Infrastructure | 18% | +30.7% | +20.0% | +18.6% | 13% |
| Whole company | 100% | +41.1% | +34.0% | +28.8% | 100% |
On the filed split through 2026-06-30, if each business keeps its last half-year's growth, the company grows about 34% next year and 35% in year 3, against 26.3% over the last twelve months. Electric (Electric Infrastructure Solutions) carries the case: it grows 38% against 20% for the rest, and would be 87% of revenue by then. A scenario, not a forecast: no business holds its rate.
How we read it: The business carrying the case: Electric's 37.5% half-year is about 27.5% without acquisitions, while Underground's 20% is all bought and shrinks without it. The score reads Electric, less its bought quarter: 3.
The results
The numbers
Sales grew 26% over the twelve months to Jun 2026, faster than the 20% in the year to Dec 2025.
Operating margin is 6.2%, about where it was (5.4%) in the year to Dec 2023.
Running costs take 9¢ of each sales dollar, about where it was (10¢) in the year to Dec 2021.
Each dollar put into the business earns 11% a year after tax, about where it was (7.3%) in the year to Dec 2021.
It turned 7.3% of sales into free cash, $2.39B over the last twelve months.
It owes $5.60B more than the cash it holds, about 1.8 years of the cash its operations bring in.
The share count rose 4.9% in 4.5 years, and pay in shares costs 0.7% of sales.
All the numbers
Filed figures from the company's 10-K and 10-Q, read 2026-10-04. Fiscal years end in December.
| Dec 2021 | Dec 2022 | Dec 2023 | Dec 2024 | Dec 2025 | 12 mo to Jun 2026 | |
|---|---|---|---|---|---|---|
| Size and growth | ||||||
| Sales | $13.0B | $17.1B | $20.9B | $23.7B | $28.5B | $32.9B |
| Sales growth | +16% | +32% | +22% | +13% | +20% | +26% |
| Profit | ||||||
| Gross margin | 15% | 15% | 14% | 15% | 15% | 15% |
| Operating margin | 5.1% | 5.1% | 5.4% | 5.7% | 5.7% | 6.2% |
| Net margin | 3.7% | 2.9% | 3.6% | 3.8% | 3.6% | 4.0% |
| What growth costs, per sales dollar | ||||||
| Selling, general and admin | 8.9% | 7.8% | 7.4% | 7.7% | 7.7% | 7.6% |
| Pay in shares | 0.7% | 0.6% | 0.6% | 0.6% | 0.6% | 0.7% |
| Capital spending | 3.0% | 2.5% | 2.1% | 2.6% | 2.1% | 2.4% |
| Cash | ||||||
| Cash from operations | $582M | $1.13B | $1.58B | $2.08B | $2.23B | $3.18B |
| Free cash flow | $197M | $703M | $1.14B | $1.48B | $1.62B | $2.39B |
| Balance sheet | ||||||
| Cash and short-term investments | $229M | $429M | $1.29B | $742M | $440M | $506M |
| Debt | $3.75B | $3.73B | $4.20B | $4.16B | $5.99B | $6.10B |
| Net cash | −$3.52B | −$3.30B | −$2.91B | −$3.42B | −$5.56B | −$5.60B |
| Returns and shares | ||||||
| Return on capital | 7.3% | 7.1% | 8.9% | 9.4% | 9.1% | 11% |
| Shares (diluted) | 145M | 148M | 149M | 150M | 151M | 152M |
What growth costs
Growth has not made Quanta much more profitable. Crews and equipment cost about the same share of every sales dollar each year, and office costs barely shrank after 2022. The extra profit that did appear was mostly eaten by the cost of the companies it bought.
The evidence under this - 2 claims
- The first half of 2026 broke the pattern, with operating margin 5.9% against 4.7% a year earlier and cost of services below 85 cents (C39), so a full year at that level would be the first sign of operating leverage in the series.
- Its spending to grow sits outside these lines: about $250M a year of training (C18) runs through cost of services, and the acquisitions run through amortisation.
The first half of 2026 broke the pattern: gross margin 15.2% against 14.2% a year earlier and operating margin 5.9% against 4.7%, with Q2 2026 at 16.2% and 7.3%. If that holds through a full year it is the first evidence of pricing in the series.
The last report (30 Jul)
Quarter ended June 2026, filed figures.
Third-quarter revenue grew 57.4% to $12.01B and operating profit reached 8.5 cents per sales dollar, from 6.8 a year earlier and 7.3 in June. Backlog hit a record $63.7B and the company raised its 2026 forecast for the third time.
The release does not say how much of the growth was bought; that is in the quarterly filing. Nine months together kept 7.0 cents, so the last four quarters have not yet passed the 8 cents our second route needs.
What it did to the thesisIt met the test we set, so margins are rising and our confidence falls to low. The price is still far above rivals', and one strong quarter is not a full year, so the view stays negative.
The quarter in detail
Sales growth from the previous quarter, filed figures.
Earnings per share $4.24 against $3.31 expected; $2.96 as filed, the gap being what the company adds back.
The evidence under this - 4 claims
The release gives no growth-without-acquisitions figure: five acquisitions in nine months and eight in 2025 are in the results, so how much of the 57.4% was bought is unread, and Q3 is the seasonally strongest quarter, so 8.5% is not a full-year margin.
Guidance has been raised each time since February, and each raise has been bigger than the last, so it says the company beat itself, not that the price is cheap; the seven weeks since the July guide include two more acquisitions that it says add nothing material this year.
The organic figure is an estimate built from the 10-Q's rounded acquired-revenue amounts, and 'acquired' there means businesses bought since the comparable period, so a business bought in mid-2025 counts as acquired for only part of the year. The call says acquisitions closed in Q2 2026 contributed only about $11M of adjusted EBITDA, so the quarter's margin gain is not the new companies.
What management did with the money
It spent its cash on buying companies: about $930M in the first half of 2026, then four more deals worth about $1.24B, paid partly in new shares. It bought back no shares, and borrowed $2.0B in August.
Insiders in 2026Q2: 18 purchases and 40 sales, from their Form 4 filings.
The evidence under this - 3 claims
Management says the Q2 2026 beat was 'broad-based organic strength' and that the new acquisitions contributed about $11M of adjusted EBITDA in the quarter, so the pace of buying does not by itself explain the margin gain. The point here is the recurring cost, not that the growth is fake.
The price, and where we disagree
In short. Priced for perfection. The price assumes Quanta keeps about twice the profit on each dollar of sales it has ever kept, for a business that competes with its own customers' crews.
The shares rose about 61% in a year, and about half of that came from buyers paying more for each dollar of revenue, not from the business growing. They then fell 28.5% from May to July, and jumped 17.3% on the July results.
The September-quarter results on 29 October passed the margin test. Next the quarterly filing shows how much growth was bought, and the December quarter shows whether 8.5 cents is a new level.
The evidence under this - 4 claims
A chief executive diversifying a large holding after a year in which the shares rose more than 60% is ordinary, and he still holds about 557,000 shares directly, worth about $360M at $649. The pack's base rate calls the second-quarter selling ratio ordinary for this company. What is not ordinary is that none of it was scheduled in advance.
The release gives no growth-without-acquisitions figure: five acquisitions in nine months and eight in 2025 are in the results, so how much of the 57.4% was bought is unread, and Q3 is the seasonally strongest quarter, so 8.5% is not a full-year margin.
How we read the multiple
Debt included, buyers pay about $51 for each $1 of Quanta's yearly profit from operations. Rivals doing the same work cost about $21.50.
To be priced like its rivals, Quanta must:
- Earn about $4.8B a year from operations, 2.4 times today.
- At today's 6 cents of profit per dollar of sales, that takes about $78B of sales, double this year's forecast.
- Or keep about 12 cents per dollar on this year's forecast sales.
- The chief executive's own stated ceiling points to about 8 cents.
Against its own history. Far more expensive than its own usual level, while the margin that would justify it has barely changed, so the gap is enthusiasm rather than a better business.
The evidence under this - 4 claims
The cohort members are each only partly comparable - utility grid work is 28.5% to 48.4% of their revenue (C22) - and all three are smaller, so part of the premium is a scale premium the cohort cannot price. Quanta also grows faster than the cohort median. Neither closes a gap of 2.4 times.
The peer median is not the only defensible multiple: a business with Quanta's scale, growth and record could deserve a lasting premium, and at 30 times rather than 21.5 the required operating profit is about $3.5B, which guided 2026 revenue reaches at an 8.8% margin - above anything Quanta has earned, but within reach of the chief executive's stated ceiling if corporate costs shrink as a share of revenue. The estimate also uses trailing profit, which in a business growing 34% lags the current run rate by about half a year.
Quanta's return on invested capital is depressed by $7.87B of goodwill and $3.22B of acquired intangibles on its balance sheet at June 30, 2026, which organic rivals do not carry; on tangible capital its return would be higher. But that goodwill is what shareholders paid, so the lower return is the return they earn.
The margin ceiling is his own and it cuts against the price rather than for it: an Electric segment at its 'utmost' 12% with corporate costs at 2025's 4.0% of revenue gives a consolidated operating margin near 8%, far below the 12.2% the price implies at a peer multiple (C11). The technology figure is verbal and unaudited; the only filed number is 13% for 2025.
Priced against its rivals
The competitors its own annual report names, each confirmed against the rival's filing. Measured on company value for each $1 of operating profit, cheapest first:
| Ticker | Company | company value for each $1 of operating profit |
|---|---|---|
| MYRG | MYR Group | 20.2× |
| MTZ | MasTec | 21.5× |
| PRIM | Primoris Services | 22.2× |
| PWR | Quanta Services | 51.0× |
The middle of the other 3 is 21.5. Quanta Services, Inc. is dearest of 4.
Rivals' figures come from a data provider, so read the spread of the group rather than any single pair.
The dearest of the rivals doing the same work, by a wide distance. That is this company's position, not the whole group's.
Also checked and set aside: Dycom Industries (a substitute, not the same product); EMCOR Group (a substitute, not the same product).
The bet: what it pays if it works
This bet is on one business: Electric, the grid and power work that is most of Quanta and nearly all of its growth. The smaller Underground business is counted separately, at what listed rivals are worth.
Even if Electric grows as analysts expect and reaches the best margin its own boss thinks possible, the whole company would be worth less than today. If its growth stalls, shareholders keep well under half: this is an expensive bet, and the price already pays for more than success.
| Electric (Electric Infrastructure Solutions) operating profit in two to three years | Electrics at their trailing operating profit needed | Valued like its peers | Like the cheapest |
|---|---|---|---|
| $3.37B | about 2 at 100% kept | 0.7× | 0.7× |
| $3.44B (the big win) | about 2 at 100% kept | 0.8× | 0.7× |
To pay 2× (a Fair bet) it would need about $9.58B of Electric (Electric Infrastructure Solutions) operating profit (about 5 Electrics at their trailing operating profit); to pay 5× (a Cheap bet), about $24.33B of Electric (Electric Infrastructure Solutions) operating profit (about 14 Electrics at their trailing operating profit), at the same multiple.
Each row is Electric (Electric Infrastructure Solutions) alone, plus the rest of the company (Underground and Infrastructure) counted at about $5.38B.
The steps it has to take
- Done Electric margin rose to 11.5% in Q2 2026 from 10.1%, with 2026 guidance raised twice
- In progress Q3 2026: GAAP operating margin at or above 8.0% with growth excluding acquisitions at or above 20% due 2026-10-29
- Not started Large 765 kV, 500 kV and 345 kV transmission work reaches backlog due 2027-02
- Not started Trailing four-quarter operating margin above 8%, with Electric growing faster without acquisitions than utility capital spending due 2027-07
- Not started That transmission work reaches the field due 2027-12
- Not started 2028 revenue near analysts' $52.5 billion with Electric's margin after corporate costs near 8% due 2029-02
The detail behind the label
The evidence under this - 5 claims
The acquired amounts are the 10-Q's rounded figures and 'acquired' means bought since the comparable period, so a business bought in mid-2025 counts as acquired for only part of the year; the organic rates are approximate. Segment operating income is struck before corporate costs, amortisation and stock pay, so it flatters both segments by construction.
Most of the corporate line is amortisation of acquired intangibles and acquisition costs, and the larger buys of 2025 (Dynamic Systems) and 2026 sit partly in Underground and Infrastructure, so a revenue split may put too much of that cost on Electric and understate its profit; it also may not, since Cupertino Electric, bought in 2024, sits in Electric.
The margin ceiling is his own and it cuts against the price rather than for it: an Electric segment at its 'utmost' 12% with corporate costs at 2025's 4.0% of revenue gives a consolidated operating margin near 8%, far below the 12.2% the price implies at a peer multiple (C11). The technology figure is verbal and unaudited; the only filed number is 13% for 2025.
The cohort members are each only partly comparable - utility grid work is 28.5% to 48.4% of their revenue (C22) - and all three are smaller, so part of the premium is a scale premium the cohort cannot price. Quanta also grows faster than the cohort median. Neither closes a gap of 2.4 times.
Where we disagree with the market
Quanta is a grid-and-data-centre winner whose forecasts keep rising, worth paying far more per dollar of profit than for its rivals.
The growth is real, but scarce crews win volume, not price. Paying double the rivals' price per dollar of profit assumes margins regulated customers will not pay.
The market is watching growth and forecast raises, and both are real: 27 analysts rate it a buy and none a sell.
The evidence against us is Q2 2026, when margins rose to 7.3 cents per dollar. If that keeps climbing past 8, the business is growing into its price and we are wrong.
What analysts say, and the evidence
The evidence under this - 5 claims
The share gain is flattered by what Quanta buys outside the utility budget: data-centre electrical and mechanical work sold to technology customers is inside the Electric growth rate and is not utility capital spending, so the comparison with EEI overstates the share taken from utilities. And the Q2 2026 margin rise to 7.3% from 5.5% may be the scarcity finally reaching price; the next two prints can overturn the 'volume, not price' half of this claim.
The first half of 2026 broke the pattern: gross margin 15.2% against 14.2% a year earlier and operating margin 5.9% against 4.7%, with Q2 2026 at 16.2% and 7.3%. If that holds through a full year it is the first evidence of pricing in the series.
The peer median is not the only defensible multiple: a business with Quanta's scale, growth and record could deserve a lasting premium, and at 30 times rather than 21.5 the required operating profit is about $3.5B, which guided 2026 revenue reaches at an 8.8% margin - above anything Quanta has earned, but within reach of the chief executive's stated ceiling if corporate costs shrink as a share of revenue. The estimate also uses trailing profit, which in a business growing 34% lags the current run rate by about half a year.
What could go wrong, and how we would know
The risks
- Margins could really be rising. Q2 2026 was the best margin in years; a lasting move past 8 cents per dollar would undo our case.
- Utilities doing it themselves. Utilities already employ most line workers and could hire more rather than pay contractors.
- Fixed-price mistakes. Fixed-price work rose to 61% of revenue; a costly error on one large project lands entirely on Quanta.
- Plans running ahead of building. Utility spending plans rose 17% while power construction actually built grew 3.6%.
- Customer bills. Regulators could slow spending if electricity bills keep rising.
The evidence under this - 6 claims
The organic figure is an estimate built from the 10-Q's rounded acquired-revenue amounts, and 'acquired' there means businesses bought since the comparable period, so a business bought in mid-2025 counts as acquired for only part of the year. The call says acquisitions closed in Q2 2026 contributed only about $11M of adjusted EBITDA, so the quarter's margin gain is not the new companies.
The first half of 2026 broke the pattern: gross margin 15.2% against 14.2% a year earlier and operating margin 5.9% against 4.7%, with Q2 2026 at 16.2% and 7.3%. If that holds through a full year it is the first evidence of pricing in the series.
The industry split cuts both ways: the buyer already employs about 1.6 line workers for every one in a line-construction firm, which is the strongest evidence that the customer can do this work itself. A 10% rise over ten years is about 13,600 people, which Quanta's own claim of more than 7,000 organic hires in a year (C18) says is not the binding limit on one company.
The Census series is dollars of construction completed, while EEI's figure is capital spending including equipment, and equipment - transformers, breakers, turbines - is where prices and lead times have risen most (C32). The two can diverge without either being wrong, but a 3.6% rise in construction put in place is not a market growing at 17%.
The 2025 actual of $204.1B came in below EEI's own September 2025 projection of $207.9B, so the association's projections have run ahead of spending. The Census Bureau's measure of power construction actually put in place grew only 3.6% in the first seven months of 2026 (C31). And investor-owned utilities are not all of Quanta's utility customers - cooperatives, municipal utilities and power developers sit outside EEI's figure.
What we still do not know
- How much of Quanta's revenue comes from data centres; the company gives only a verbal 15% to 20%.
- Why utility spending plans and government construction figures disagree so widely.
- Whether Q2 2026's margin came from better prices or a lucky mix of projects.
- Whether utilities are hiring their own line workers faster than contractors are.
The evidence under this - 4 claims
The margin ceiling is his own and it cuts against the price rather than for it: an Electric segment at its 'utmost' 12% with corporate costs at 2025's 4.0% of revenue gives a consolidated operating margin near 8%, far below the 12.2% the price implies at a peer multiple (C11). The technology figure is verbal and unaudited; the only filed number is 13% for 2025.
The Census series is dollars of construction completed, while EEI's figure is capital spending including equipment, and equipment - transformers, breakers, turbines - is where prices and lead times have risen most (C32). The two can diverge without either being wrong, but a 3.6% rise in construction put in place is not a market growing at 17%.
The organic figure is an estimate built from the 10-Q's rounded acquired-revenue amounts, and 'acquired' there means businesses bought since the comparable period, so a business bought in mid-2025 counts as acquired for only part of the year. The call says acquisitions closed in Q2 2026 contributed only about $11M of adjusted EBITDA, so the quarter's margin gain is not the new companies.
The industry split cuts both ways: the buyer already employs about 1.6 line workers for every one in a line-construction firm, which is the strongest evidence that the customer can do this work itself. A 10% rise over ten years is about 13,600 people, which Quanta's own claim of more than 7,000 organic hires in a year (C18) says is not the binding limit on one company.
What would change our view
- Unattractive→NeutralThe multiple falls toward the cohort while organic growth holds
It needs all of:
- ○The price falls to about $32 for each $1 of yearly operating profit, from about $51.about ≈$407.30 today, 39% below the last close
- ○Growth without bought companies stays above 15% in the next quarter.checked at the next results, due in Feb 2027
- Unattractive→NeutralThe business grows into the price: margins rise for a full year
It needs all of:
- ○Operating profit over the last four quarters passes 8 cents per sales dollar, from 6.2.checked at the next results, due in Feb 2027
- ○The electric business keeps growing faster than the 17% rise in utility spending, without counting purchases.checked at the next results, due in Feb 2027
We would also be wrong if
- A large utility moves line building to its own crews.
- One fixed-price project costs more than $200M in a quarter.
What is coming
- 6 November 2026 Q3 2026 10-Q: The split of the 57.4% into bought and not bought, the master-service-agreement share of backlog, total debt and any change in estimates on fixed-price work.
Every dated event ahead
- 6 November 2026 Q3 2026 10-Q: The split of the 57.4% into bought and not bought, the master-service-agreement share of backlog, total debt and any change in estimates on fixed-price work.
- February 2027 Q4 2026 results, 2027 guidance and the 2026 10-K: Whether the first-half margin gain holds for a full year, how much of 2027's guided growth is bought, and the year's acquisition spend and amortisation.
- April 2027 2027 proxy statement: Insider holdings after the 2026 sales, the 2026 bonus payout against its margin target, and whether pay metrics move toward GAAP or stay on adjusted earnings.
Since we wrote this
- 29 OctEarningsMargin hit 8.5% and guidance rose; conviction falls to LowThe test the thesis set was met, so confidence in Unattractive drops from Medium to Low; the price, still well above rivals', keeps the view.
- 5 OctThesis reviewThe bet now names what it ridesThe owner asked that every bet name the macro trend behind it, after BWXT's bet was missing its chief executive's case for nuclear over gas.
- 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 writtenUnattractive, medium convictionA strong grid builder priced at 51 times operating profit, more than double three rivals, after a year when the price rose faster than the business.
Also checked 1 time with nothing new to act on.
Sources 28
Sources - 28
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 filing2026-02-19Item 1 business description, segments, acquisitions, customer-type mix, largest-customer and top-ten concentration, competition and in-house utility crews, headcount and union share, contract-type mix, segment income and corporate costs, amortisation and stock compensationbacksC01C02C05C07C08C09C10C18C21C22C24C25C32C37C38C39C40C45C46
- S02primarycompany filing2026-07-30Q2 2026 income statement, segment results and acquired revenue, backlog and MSA share, balance sheet, cash flow, debt, contract liabilities, unapproved change orders, acquisitions, share count and repurchasesbacksC01C03C04C07C08C09C10C24C25C26C27C29C32C40C46
- S03primarycompany release2026-07-30raised full-year 2026 guidance, the four acquisitions and their consideration and expected contributionbacksC03C08C09C12C25C26C27C30
- S04primarytranscript2026-07-30management's answers on margins, technology share, self-performance, labour and the timing of large projects; the analyst questions (UBS, Jefferies, Evercore, ROTH, Mizuho, Citigroup) are a separate origin from the answers, and one of them - Jefferies on a Midwest utility self-performing - is the only outside report of a customer building its own capability
We self-perform 80% to 85% of our work, which is what allows us to deliver on time and on budget at scale. [...] We've always said that the electric segment has the ability to operate 10 to 12 on the utility side, 12 would be the utmost margins that you would see. [...] We have ratepayers on the utility side. It's very much regulated. [...] we've seen a large utility in Midwest recently indicate for -- at least for their gas generation RFP that they're electing to self-perform because there's inadequate resources. [...] I do think some of the utilities, a few of the utilities are trying to self-perform. [...] 15% of the business is technology, call it, 15% to 20% right in there. [...] we're not in backlog on any of the bigger work yet. Even the generation work, most of it is, I would say, 95% of it is not in backlog either. [...] kind of what I would say upper single-digit growth in the business as it sits [...] It takes about 4 years to make a craftsman and a journeyman. [...] we spend about $250 million a year in training.
backsC06C18C23C26C35 - S05primarycompany release2026-02-192025 GAAP and adjusted earnings per share, and the initial 2026 guidancebacksC02C09C12
- S06primarycompany release2026-04-30the first 2026 guidance raise, backlog of $48.5 billion, and the Investor Day statement of more than doubling adjusted earnings per share by 2030 and a $2.4 trillion addressable marketbacksC04C12C13
- S07primarycompany release2026-08-04$2.0 billion of senior notes due 2029, 2033 and 2036 and their stated usebacksC26
- S08secondaryvendor data2026-09-26Portfolio-management research pack (market tier, filed statement series, filed-competitor cohort)content/PWR/research-pack.jsonprice and returns, the price-move decomposition, analyst consensus and its drift, analyst revenue estimates, earnings surprise history, short interest, the six-year valuation history, the annual revenue and margin series, revenue per employee from 10-K headcount, and the filed-competitor cohort multiplesbacksC10C12C24C28C29C30C37C38C39C46
- S09primarycompany filing2026-09-26every open-market sale in the year, the price, the holding after it, and the unchecked 10b5-1 box on all 35backsC28
- S10primaryindustry data2026-05-27Edison Electric Institute, Electric Perspectives - investor-owned utility capital expenditures (article of 2026-05-27) Edison Electric Institutewhat Quanta's largest customer class spent in 2025 and plans for 2026 to 2030, measured by their trade association
the industry's annual capital expenditures hit a record $204.1 billion last year-a 14th-consecutive year of record-high investment. In 2026, capital expenditures are projected to jump 17 percent to $238.8 billion [...] The $1.4 trillion that investor-owned electric companies will invest in the grid between 2026 and 2030 is an increase from the previous projection of $1.1 trillion
backsC14 - S11primaryindustry data2025-09Edison Electric Institute, Industry Capital Expenditures 2015-2029 chart pack (updated September 2025) Edison Electric Institutethe transmission and distribution share of investor-owned utility capital spending in the 2025 projection, about half
Distribution $66.0B 32% … Transmission $37.6B 18%
backsC14 - S12primaryregulator2026-01the load growth utilities are building for, from the grid's reliability regulator
Over the 10 -year period, aggregated assessment area summer peak demand is forecast to rise by over 224 GW. This is 69% higher than last year's 10-year growth projection of 132 GW. [...] New data centers for artificial intelligence and the digital economy account for most of the projected increase
backsC15 - S13primaryindustry data2025-07Grid Strategies and Americans for a Clean Energy Grid, Fewer New Miles (July 2025) Grid Strategies LLChow little high-voltage line is being built against what is needed; the revised FERC figure of 888 miles cited in a later revision could not be read and is not used
In 2024, only 322 miles of new high-voltage transmission lines were completed, the third slowest year of new construction for 345 kV and above transmission over the past 15 years. For comparison, nearly 4,000 miles were built in 2013 alone.
backsC16 - S14primarygovernment data2023-10US Department of Energy, National Transmission Needs Study (final, October 2023) US Department of Energythe government's range for how much interregional transmission has to grow by 2035; a July 2026 draft update exists and was not read
interregional transfer capacity must grow by 25% to meet future moderate load and clean energy growth, by 114% to meet moderate load and high clean energy growth, and by 412% to meet high load growth futures by 2035.
backsC16 - S15primarygovernment data2026US Bureau of Labor Statistics, Occupational Employment and Wage Statistics API - electrical power-line installers and repairers (SOC 49-9051), May 2025, by industry US Bureau of Labor Statisticswho employs the country's line workers: 131,070 in all (series OEUN000000000000049905101), 62,470 in electric power generation, transmission and distribution - the utilities themselves - 38,990 in utility system construction (OEUN000000023710049905101) and 8,140 in specialty trade contractors (OEUN000000023800049905101). The HTML occupation pages answered 403; the API is the second route
{"status":"REQUEST_SUCCEEDED","responseTime":108,"message":["No Data Available for Series OEUN000000022110049905101 Year: 2023","No Data Available for Series OEUN000000022110049905101 Year: 2024"],"Results":{ "series": [{"seriesID":"OEUN000000022110049905101","data":[{"year":"2025","period":"A01","periodName":"Annual","latest":"true","value":"62470","footnotes":[{}]}]}]
backsC17 - S16primarygovernment data2026-08-27US Bureau of Labor Statistics, Occupational Outlook Handbook - Line Installers and Repairers (modified 2026-08-27) US Bureau of Labor Statisticsthe projected growth of the trade, and a median wage of $95,320 in May 2025. Read through an agent fetch after curl was refused, so the quote is a model's reading of the page and is used for the rate only
Employment of electrical power-line installers and repairers is projected to grow 10 percent from 2025 to 2035, much faster than the average for all occupations.
backsC17 - S17primarygovernment data2026-09-01power construction put in place, public and private: $103.0 billion in January to July 2026 against $99.4 billion a year earlier, up 3.6%, and a July annual rate of $181.5 billion, up 5.3%
Power 15,287 15,313 14,983 14,616 14,453 14,487 103,015 99,394 3.6
backsC31 - S18primarycompany filing2026-02PG&E Corporation Form 10-K for the fiscal year ended December 31, 2025, Item 1 Human Capital PG&E Corporationa named Quanta customer's own account of how much of its work is done by contractors: about 39,000 contractor individuals from about 1,200 firms against about 29,000 regular employees
The Utility's contractors and subcontractors include approximately 39,000 individuals from approximately 1,200 contractor companies.
backsC20 - S19primarycompany filing2026-02-24Primoris Services Corporation Form 10-K for the fiscal year ended December 31, 2025 Primoris Services Corporationthe rivalry named from the other side, and the Utilities segment at $2,691.7 million of $7,574.9 million of 2025 revenue
competitors in our utilities markets include Quanta Services, Inc., Dycom Industries, MYR Group, and MasTec, Inc.
backsC22 - S20primarycompany filing2026-07-30Power Delivery revenue of $4,176.1 million in 2025 and $1,245.8 million in Q2 2026
Our Power Delivery segment primarily serves the energy, utility and data center infrastructure industries through the engineering, construction and maintenance of power transmission and distribution infrastructure
backsC22 - S21primarycompany filing2026-07-29Transmission and Distribution revenue of $2,002.4 million in 2025 and $524.0 million in Q2 2026
Our T&D segment provides a broad range of services on electric transmission and distribution networks and substation facilities
backsC22 - S22primarygovernment data2024-07-10US Department of Energy, Large Power Transformer Resilience Report to Congress (July 2024) US Department of Energythe equipment bottleneck Quanta has integrated into by making transformers and breakers itself
lead times for acquisition of an LPT have become exceptionally long, with 36-month lead times being commonly quoted and maximum lead times reaching as much as 60 months
backsC32 - S23primarygovernment data2026-05Lawrence Berkeley National Laboratory, Queued Up: 2026 Edition (end-2025 data, May 2026) Lawrence Berkeley National Laboratorythe interconnection queue shrinking for a second year, against about 1,400 GW of generation and 890 GW of storage a year earlier (2025 edition). Agent-read page; emp.lbl.gov answered 403 and OSTI 503
As of the end of 2025, there were ~8,200 projects actively seeking grid interconnection in the U.S., representing 1,312 GW of generation and approximately 749 GW of storage.
backsC33 - S24primarygovernment data2026-09US Energy Information Administration, Short-Term Energy Outlook, September 2026 US Energy Information Administrationthe demand the grid is being built for, measured in electricity actually sold, and the note that Texas has paused connecting new data centres for a regulatory audit
We forecast that commercial sector electricity sales will increase by 3.3% in 2026 and 2.7% in 2027, accounting for 63% and 56% of the increase in electricity sales in those years.
backsC34 - S25secondarynews2025-01-27the objection stated on the day of the first repricing: that more efficient AI models would need less power and a smaller buildbacksC30
- S26tertiarynews2026-06the objection stated during the May to July 2026 decline: valuation after the multiple expanded, a downgrade, and insider sellingbacksC30
- S27primarycompany filing2026-04-10executive officers and their tenure, beneficial ownership as of March 26, 2026, the 2025 annual incentive plan's metrics, weights, targets and payout, the 2025 long-term incentive plan's metrics and the 2023-2025 cycle's payoutbacksC42C43C44
- S28primarycompany filing2026-10-29Q3 2026 revenue, operating income, earnings per share, backlog and remaining performance obligations, the raised 2026 guidance, cash flow, liquidity actions and two acquisitionsbacksC04C47C48C49C50