HOCHTIEF: Data Centers Are 21% of Backlog and 32% of FY2025 Orders, and EUR 426 Prices the Boom Against a EUR 358 Base SOTP

That cash conversion carries the central construction-cycle caveat. Fast-growing construction managers often receive customer advances and progress payments before paying the full subcontractor bill. Contract liabilities and trade payables rise with activity, and cash conversion can outrun accounting profit. Slower intake reverses part of that benefit. Using H1 2026's LTM operational earnings of roughly EUR 914 million against EUR 2.37 billion of OCF, even a 25% reversal of the excess working-capital benefit would absorb roughly EUR 360 million of cash; a 50% reversal would be about EUR 730 million. Those are sensitivities, not company guidance. The balance sheet is strong enough for such a reversal, but investors should not capitalize peak cash conversion as perpetual free cash flow. The company's own H1 metric showed EUR 71.6 million net cash at June 30, versus EUR 97.3 million net debt at December 2025.

The Thiess transaction adds another wrinkle. CIMIC already consolidated Thiess after raising ownership to 60% in 2024, so the July 1, 2026 purchase of Elliott's remaining 40% creates no further step-up in consolidated Thiess revenue. It redirects the minority's share of profit to HOCHTIEF/CIMIC. CIMIC paid AUD 1.18 billion for that 40%, implying approximately AUD 2.95 billion for 100% of the equity. Thiess had A$5.9 billion of 2023 revenue, more than 15,000 employees and more than 60 projects when CIMIC raised its stake in 2024. The business has deliberately moved toward metals, critical minerals, underground mining, asset management and rehabilitation, although public disclosures reviewed here do not give a sufficiently current coal-versus-copper/gold revenue split to quantify the remaining commodity concentration.

Using an analytical translation rate of EUR 0.57 per AUD around the transaction date, the 40% purchase cost is approximately EUR 673 million. Against an assumed EUR 80 million of incremental annual earnings from 2027, that is roughly 8.4 times incremental earnings. The EUR 80 million is my own working assumption: HOCHTIEF's completion announcement quantifies nothing, guiding only to a slightly positive effect on 2026 operational net profit. The economics are reasonable. The capital intensity is not Turner-like: Thiess owns and operates mining equipment, so depreciation and replacement capex are structural costs. That difference is one reason a single group P/E obscures what investors actually own.

Governance deserves a discount rather than a footnote. ACS is the controlling shareholder and HOCHTIEF's CEO Juan Santamaría also leads ACS. HOCHTIEF's 16-person codetermined Supervisory Board contains eight shareholder representatives and eight employee representatives. Four shareholder representatives are explicitly independent of the controlling shareholder, which means the other four sit on the ACS-linked side of the governance structure; Pedro López Jiménez, Cristina Aldámiz-Echevarría, Ángel García Altozano and José Luis del Valle all have current senior or board roles at ACS or its affiliates.

The best available 2026 ownership indication in this research pass is approximately 80.32% for ACS, implying about 19.68% free float. HOCHTIEF's public shareholder-structure graphic itself still identifies its underlying voting-rights information as older, so I treat 80.32% as the best current estimate rather than pretending it has been independently reconstructed from a fresh primary notification. The economic consequence is clear even without the second decimal place. Minority holders do not control capital allocation, the dividend or strategic combinations. At a 65% operational-profit payout, the FY2025 EUR 6.60 dividend theoretically distributes about EUR 513 million across the issued share count, of which an 80.32% owner receives roughly EUR 412 million. HOCHTIEF confirms the EUR 6.60 dividend and 65% payout policy; ACS's own strategic plan calls for substantial shareholder remuneration at parent level.

The DAX promotion helps liquidity but adds no intrinsic value. STOXX announced HOCHTIEF's addition on June 3, effective June 22, 2026, replacing Porsche Automobil Holding. The DAX uses free-float market capitalization for constituent selection. With only about one fifth of HOCHTIEF economically floating, today's EUR 33.1 billion headline equity value corresponds to only around EUR 6.5 billion of free-float market value. Passive DAX demand can be meaningful relative to the float. It cannot make a construction contract more profitable.

The stock's rerating has outrun the earnings improvement. HOCHTIEF itself said on February 19, 2026 that its share price had risen 160% over the preceding twelve months, while 2025 operational net profit grew 26%. A market-data anchor shows EUR 218.60 on September 1, 2025, versus EUR 426 on September 4, 2026. At the current quote the company is worth about 42 times FY2025 operational net profit, 36.7 times FY2025 nominal profit and 30–32 times the new FY2026 operational-profit guidance.

My qualitative portrait lands on re-rating at a thematic and cyclical high. HOCHTIEF has genuinely improved the business: Turner is taking share in high-value work, contract risk has been reduced, margins are expanding, cash has been strong and the balance sheet is sound. The stock price has gone further. Today's buyer is paying in advance for Turner to retain a large part of its extraordinary data-center growth, for Thiess and the other businesses not to create offsets, and for a roughly 80%-controlled subsidiary to retain a premium public-market multiple.

The core bull/bear disagreement can be reduced to one variable. Bulls see data-center construction as a decade-long capacity build whose short-term growth rate may slow but whose absolute spending keeps compounding. Bears see a familiar capital-expenditure cycle inside an unfamiliar AI wrapper: hyperscalers may still spend heavily, yet construction starts can fall sharply after the current wave of campuses is committed. Because data centers already represented 21% of backlog at the end of 2025 and a much larger share of incremental ordering, the difference between those views is worth several hundred million euros of normalized Turner earnings and many turns of HOCHTIEF's P/E.

Vertical History, Financial Review and Price Narrative

HOCHTIEF's origins explain an enduring feature of the company: it has repeatedly shifted from simply supplying construction labor toward organizing increasingly complicated projects. Philipp Helfmann began as a timber merchant and building contractor in Frankfurt; his brother Balthasar joined him in 1873, just before Germany's Gründerkrise. By 1874 the business appeared in the Frankfurt address book as Helfmann Brothers. Philipp was a mason by training and Balthasar a mechanic. The enterprise was converted into a stock corporation in 1896. HOCHTIEF's own historical material records the 1873 founding, the 1896 conversion and the headquarters move to Essen in 1922.

There is no modern IPO event to reconstruct. The 1896 conversion pre-dates contemporary book-building, prospectus practice and electronic exchange records. The available corporate history does not provide a reliable modern-equivalent IPO offer price, funds raised or flotation valuation, so inventing one would create false precision. What matters for the capital-market record is that HOCHTIEF has spent most of its corporate life as a public joint-stock construction company whose ownership has repeatedly been shaped by larger industrial and financial shareholders.

The first long phase, from the Helfmann era through the mid-twentieth century, built HOCHTIEF's technical identity. It moved from regional building into civil works and international projects. Like other major German industrial companies, its history also passes through the Third Reich; that legacy belongs in a complete corporate history rather than being sanitized. Postwar reconstruction then put the company back into German civil and industrial infrastructure, followed by renewed international expansion. HOCHTIEF's official history divides its development into the eras 1873–1896, 1896–1921, 1921–1933, 1933–1945, 1945–1966, 1966–1989, 1990–2007 and the period since 2007.

The second economically important transformation came after 1990. HOCHTIEF wanted scale outside the saturated German construction market and moved into service-oriented and international construction. The critical assets accumulated in this era were Turner in the US, Leighton in Australia, later renamed CIMIC, and Flatiron in North American civil engineering. That collection produced a company whose country of domicile ceased to describe its sources of revenue. By 2026 HOCHTIEF itself characterizes the business around North America, Australia and Europe, rather than Germany.

The third transformation was ownership. ACS began building its HOCHTIEF stake in the 2000s and crossed control in 2011. Atlantia later became another large shareholder but sold a 14.46% block to ACS in September 2022 at EUR 51.43 per share for EUR 577.8 million. The long-run direction has been one-way: free float has been replaced by ACS ownership.

Under ACS, the portfolio also became cleaner. HOCHTIEF bought the remaining CIMIC minorities in 2022, paying about EUR 940 million for the takeover after a EUR 406 million capital increase, converting a listed Australian subsidiary into a wholly controlled strategic arm. HOCHTIEF's H1 2022 report explicitly showed how large the transaction was relative to the group's then balance sheet.

The fourth phase, beginning roughly in 2023, is the one investors are paying for today. Management recast the company around “strategic growth markets”: data centers and semiconductors, energy and nuclear, defense, critical minerals and infrastructure stimulus. This was initially an allocation story; by 2025 it had become an earnings story. Strategic markets represented more than 55% of FY2025 new orders. Data centers alone generated EUR 16.8 billion of orders, while Germany's backlog nearly doubled over three years to EUR 5.2 billion and the group ended 2025 with more than EUR 2 billion of defense backlog.

FlatironDragados is characteristic of the new ACS-HOCHTIEF structure. ACS's Dragados North America and HOCHTIEF's Flatiron were combined into a single North American civil-engineering platform, leaving HOCHTIEF with an equity-accounted minority interest rather than full operating consolidation. The transaction created a EUR 146 million non-cash fair-value gain in Q1 2025. That gain increased nominal earnings without increasing operational earnings, which is why an investor who simply screens HOT on reported EPS gets the wrong 2025 earnings base.

Thiess took the opposite route. CIMIC sold half of the mining-services business to Elliott in 2020, restored majority ownership to 60% in 2024, and bought the remaining 40% for AUD 1.18 billion on July 1, 2026. CIMIC described Thiess at the 2024 step-up as a business with long-term, lower-risk contracts and annuity-like cash flows; by 2026 it emphasized the shift toward metals, critical minerals, underground mining and rehabilitation.

Selected financial vertical checkpoints

† H1-period OCF; LTM H1 2026 pre-factoring OCF was EUR 2.37 billion. Operational figures exclude management-defined non-operational factors. Sources: HOCHTIEF FY2025 and H1 2026 disclosures.

The financial transformation shows up in the margin more clearly than in sales: revenue grew 15% in 2025 but operational PBT grew 29%, pushing the margin from 3.0% to 3.4%. By H1 2026 it was 3.8%. A 40–80 basis-point margin move sounds small until applied to EUR 40 billion of revenue. Every 50 basis points is approximately EUR 200 million of annual PBT before mix and tax effects. Turner is the principal reason that arithmetic has become favorable.

The quarterly disclosures let me rebuild the FY2025 nominal-to-operational bridge to a useful level:

The Q1 release identifies EUR 146 million as a non-cash Flatiron transaction gain. The annual arithmetic implies roughly EUR 33 million of net offsetting non-operational effects elsewhere during the year. The public HTML results releases reviewed for this report do not provide enough category-level detail to subdivide that residual safely into restructuring, impairment, M&A and tax lines, so the responsible reconciliation stops there rather than inventing labels.

This also explains why operational earnings make the better multiple denominator. The 2025 Flatiron gain was real under IFRS but did not represent repeatable construction economics. H1 2026 provides the opposite test: operational net profit of EUR 479.7 million exceeded nominal profit of EUR 467.6 million by EUR 12.1 million. A valuation framework that jumps between nominal and operational profit depending on which gives the lower multiple is analytically indefensible.

Cash requires a second adjustment. FY2025 EBITDA was EUR 2.215 billion and EBIT EUR 1.544 billion, implying about EUR 670 million of depreciation and amortization. Net operating capex and leases were EUR 665 million, almost identical to D&A. That suggests the group-level accounting charge is not obviously understating physical asset consumption. I estimate maintenance capex at EUR 500–550 million and growth capex plus expansionary leases at EUR 115–165 million in 2025. This is an analytical split because HOCHTIEF does not publish maintenance capex separately. On the midpoint, operational “owner earnings” are approximately EUR 935 million: EUR 789 million operational net profit plus roughly EUR 670 million D&A minus about EUR 525 million maintenance capex.

HOCHTIEF's EUR 1.41 billion of FY2025 post-capex operating cash flow looks much better than the roughly EUR 0.9–1.0 billion normalized owner-earnings estimate because fast growth released working-capital cash. At EUR 426 the stock is roughly 35 times my 2025 normalized owner earnings, versus about 42 times operational accounting profit. Cash accounting does not make the shares cheap. Normalized cash earnings remain expensive while the reported cash-flow yield is temporarily enhanced by the upcycle.

Capital-market narrative and price attribution

The longer price history makes the scale of the rerating clear. A market-data reference puts the share at EUR 71.56 in August 2021. It closed EUR 218.60 on September 1, 2025 and EUR 426.00 on September 4, 2026. At December 2025 HOCHTIEF itself reported a EUR 25.3 billion market capitalization; against its disclosed share count, that corresponds to roughly EUR 326 per issued share, subject to treasury-share treatment.

Price anchors and corporate events are drawn from HOCHTIEF, CIMIC, STOXX and market-data sources.

Of the rerating since the data-center boom became visible, I would attribute roughly 55–65% to the AI/digital-infrastructure theme, 25–35% to HOCHTIEF-specific execution and repeated guidance beats, and at most around 5–10% to DAX/passive-flow and other technical factors. These are analytical attribution ranges, not an econometric decomposition. The strongest evidence is timing: the company had already risen 160% over twelve months by February 2026, four months before DAX inclusion, while earnings had repeatedly beaten and data centers had reached 21% of backlog. The DAX promotion accelerated recognition and liquidity after most of the fundamental repricing had occurred.

The valuation center shifted because business quality improved and investors changed the peer set simultaneously. HOCHTIEF used to be priced principally as a cyclical construction conglomerate. Turner now invites comparison with US AI-infrastructure beneficiaries. That is a real change in mix; it is also dangerous because the market has effectively capitalized an unusually fast capex cycle as a structural margin reset.

Business Model, Moat, Industry and Horizontal Analysis

HOCHTIEF's consolidated revenue is economically heterogeneous. Turner is a construction manager. CIMIC combines infrastructure construction, services, mineral processing and Thiess mining services. Europe is closer to conventional civil engineering and design-and-build work. Abertis and FlatironDragados are associates, so their operating revenue and debt do not belong in consolidated HOCHTIEF sales or enterprise debt even though their share of profit can appear in HOCHTIEF's earnings. That mix is the main reason consolidated sales, margins and leverage need to be disassembled before valuation.

Turner's attraction begins with contract structure. It works predominantly as construction manager or construction-manager-at-risk. HOCHTIEF has steadily pushed the entire group toward lower-risk contracting: at Q1 2026 management said approximately 90% of backlog related to lower-risk contracts; at September 2025 the proportion was above 85%. ACS's 2024 strategic plan similarly said lower-risk collaborative contracts already represented 80% of CIMIC's backlog at the end of 2023 and were intended to rise further.

That changes what a 4% margin means. On Turner's gross reported revenue, subcontracted work passes through the income statement, so the fee margin appears thin. On true net fee revenue, the economic margin would be much higher. HOCHTIEF does not disclose a clean Turner “net revenue” or fee-revenue denominator, so a fee-margin rebuild cannot be completed without inventing a number. The observable measure is Turner's operational-PBT margin: ACS reported 3.9% for H1 2026, with H1 PBT of EUR 551 million, while HOCHTIEF raised Turner's USD full-year PBT guidance to USD 1.40–1.46 billion.

CIMIC and Europe have more conventional contractor risk. Design-and-build and fixed-price exposure turns errors in estimating labor, material, productivity and schedule into contractor losses. Alliance and reimbursable contracts transfer more of that risk to clients. Thiess is different again: its clients outsource mine operation and fleet capability over multiyear contracts, creating recurring service-like revenue, but the contractor must fund heavy equipment. CIMIC's description of Thiess emphasizes long-term contracts and stable cash flows, while the 2024 disclosure shows a global fleet-dependent business spread across more than 60 projects.

Turner's moat is reputation plus execution capacity, reinforced by client repetition rather than proprietary technology. HOCHTIEF reported an 89.1% group repeat-client rate in 2025, weighted by work done. In mission-critical projects that matters because delay costs dwarf small differences in construction fees. A hyperscaler commissioning a multi-billion-euro campus cares about schedule, labor coordination, supply-chain access, safety and the ability to execute several locations at once. Those advantages compound as Turner completes more advanced-technology projects and carries the teams and subcontractor relationships into the next campus.

The second moat is scale without proportionate balance-sheet intensity at Turner. The business can manage tens of billions of gross project volume without owning all the specialist equipment that creates the finished building. That is a far better economic model than taking every trade onto the contractor's balance sheet. The cost is that revenue is a poor measure of value added, while working capital becomes unusually sensitive to the timing of project payments.

The third moat is the group's ability to combine development, engineering, construction and, selectively, capital. HOCHTIEF and ACS have decades of PPP experience, while the Abertis relationship supplies brownfield-concession exposure. HOCHTIEF's own strategy increasingly tries to extend that model into data centers and energy infrastructure, including development rather than simply earning a construction fee. That can raise returns when projects succeed but also converts a capital-light contractor into a principal investor at the margin. HOCHTIEF and ACS describe themselves as one of the largest PPP-development platforms by projects and invested capital over multiple decades.

The moat has limits. General construction has low formal entry barriers, bids remain competitive and no patent protects Turner from Skanska, DPR or other major managers. Customer concentration can also turn reputation into bargaining power for the customer: a hyperscaler issuing a multi-campus program can negotiate fees and contractual terms across an enormous volume of work. The moat is strongest in project selection, execution reliability and labor/supply-chain coordination. It does not confer software-like pricing power.

The end-market mix and what is actually disclosed

§ ACS reporting is used here only for the same underlying Turner and CIMIC operations and is not added to HOCHTIEF group totals.

The table exposes a disclosure gap rather than filling it. The company can quantify data centers because that has become the investment story, but it does not provide a complete end-market backlog bridge. Healthcare, education, semiconductors, other advanced-technology work, transport, energy and mining cannot all be reconstructed into mutually exclusive group percentages. No investor can prove from published data that the remaining 79% of FY2025 backlog is evenly diversified. What can be proved is that data centers are already a very large standalone category and represented roughly one third of FY2025 intake.

On region, the picture is clearer. ACS's H1 2026 disclosure put Turner's backlog at EUR 46.1 billion and CIMIC's above EUR 23.7 billion. Relative to HOCHTIEF's EUR 84.8 billion consolidated order book, those amounts imply approximately 54% North American Turner exposure, 28% Asia-Pacific CIMIC exposure and an approximately 18% residual dominated by Europe. Because FlatironDragados is equity-accounted, its backlog should not simply be added to HOCHTIEF's consolidated total.

How much of the profit acceleration is data centers?

HOCHTIEF does not disclose data-center segment profit. Any precise answer such as “57.4% of profit growth” would be spurious. My bounded estimate is that hyperscaler/data-center construction accounts for roughly 45–60% of the increase in group operating earnings from the pre-boom 2023-era base to the 2026 run rate.

The reasoning is explicit. Turner generated roughly two thirds of 2025 operational earnings under the segment mix supplied by company materials, and by H1 2026 Turner's EUR 551 million PBT was about 73% of HOCHTIEF's EUR 759 million group operational PBT. Data-center order growth and work done have been substantially faster than Turner's other sectors, yet ACS also identifies sports, pharmaceuticals and aviation as contributors to Turner's H1 growth. Assigning 55–70% of Turner's incremental profit to data centers and 75–85% of group incremental profit to Turner produces a roughly 40–60% range; I round that to 45–60% because DC project mix is apparently margin-accretive. This is an inference from disclosed segment growth, not management reporting.

That estimate is enough for investment analysis. Half of incremental group earnings coming from one end-market is material concentration even when no single hyperscaler breaches a conventional revenue-disclosure threshold.

Industry cycle

HOCHTIEF sits at the intersection of several capex cycles. Public transport and German infrastructure are multi-year policy cycles. Mining services depend on commodity investment but can be buffered by long-term operating contracts. Conventional commercial construction reacts to rates and real-estate demand. Data centers are currently a technology-capex cycle with a physical-infrastructure bottleneck. The group is more diversified than a pure data-center contractor, yet Turner's incremental profit makes digital infrastructure the marginal earnings driver.

The policy backdrop away from AI is genuinely supportive. Germany's EUR 500 billion infrastructure fund began its first full deployment year in 2026, while HOCHTIEF's German backlog had reached EUR 5.2 billion by FY2025. Defense work exceeded EUR 2 billion of backlog, and the group has been selected for nuclear and grid-related programs. Those businesses create an earnings floor if US digital construction normalizes, although their margins and timing will not necessarily replace Turner's peak contribution one-for-one.

For hyperscaler capex, my valuation assumptions deliberately sit below a straight-line extrapolation of today's boom. From the 2026 high base I assume aggregate hyperscaler infrastructure capex grows about 18% in 2027, 10% in 2028 and 5% in 2029, then settles toward roughly 4–6% nominal through-cycle growth. HOCHTIEF itself cites industry expectations for data-center and compute capex to multiply dramatically by 2035, but a long-run rise in spending does not prevent annual construction starts from falling 30–40% after an overbuild.

My bear stress is a one-third fall in new data-center starts, not an implausible collapse in AI usage. I assume 45% of Turner's peak PBT is directly or indirectly associated with data-center economics at the current margin. A 33% decline in starts would be cushioned for roughly a year by backlog, then reduce Turner's PBT by about USD 210 million at full run-rate sensitivity. Using EUR 0.86 per USD as the analytical translation assumption, that is about EUR 180 million of PBT and roughly EUR 135–140 million after tax, or around 13% of the midpoint of HOCHTIEF's 2026 operational-net-profit guidance. The cash effect could be larger because slower starts would also reverse advance-payment working capital.

Horizontal portrait

Skanska is the cleaner European construction comparator because its economics remain those of a disciplined general contractor and developer; it does not deserve Turner's fee-model margin interpretation. VINCI is useful mainly for testing the value of concessions and infrastructure ownership, not for applying a single multiple to HOCHTIEF. Fluor provides a US reference for project-risk discipline in large engineering and construction. EMCOR, Quanta and Sterling Infrastructure are better thematic comparisons for the American data-center build-out, since they sit in the specialty electrical, mechanical or site-infrastructure profit pool. Their self-performed value added means their revenue margins are not directly comparable with Turner's gross construction-management revenue.

ACS is the most important comparison and also the least independent. ACS consolidates Turner through HOCHTIEF, so an ACS disclosure about Turner is the same economic activity, not corroborating evidence from a second company. At H1 2026 ACS reported EUR 26.17 billion of total sales, EUR 1.62 billion EBITDA, EUR 510 million operational net profit and EUR 105.9 billion backlog; Turner was explicitly the principal growth engine.

A HOT minority share gives cleaner and more concentrated access to Turner, CIMIC and Thiess than an ACS share. An ACS shareholder, by contrast, also owns the parent's direct Dragados operations, direct 30% Abertis interest, Iridium and other corporate assets, while enjoying the control economics over HOCHTIEF. ACS's strategic-plan disclosure shows the distinction particularly well: ACS owns 30% of Abertis directly and another 20% indirectly through HOCHTIEF. The HOT shareholder participates only in the latter stake.

That creates an unusual niche. HOCHTIEF is neither a normal European contractor nor a pure US AI-infrastructure stock. It is a controlled listed subsidiary containing one exceptional capital-light construction manager, one more capital-intensive Australian infrastructure/mining platform, and two important off-balance-sheet associates. Customers may prefer Turner because of execution reputation; investors prefer HOT because it is the most direct listed route into that business. The second proposition has driven the multiple much faster than the first proposition has driven normalized earnings.

Current Fundamentals and Bull/Bear Divergence

H1 2026 is close to a textbook peak-momentum report. Sales rose 9.6% reported to EUR 20.14 billion. Operational PBT rose 29.1% to EUR 759 million, so the margin expanded 60 basis points to 3.8%. Operational net profit rose 35.2% to EUR 479.7 million. New orders reached EUR 31.5 billion, 25% higher on an FX-adjusted basis, and the EUR 84.8 billion backlog was 22.9% above the prior-year reported amount. Intake equaled 1.5 times work performed.

The H1 beat is better than a one-off accounting beat. Nominal profit was actually lower year on year because the 2025 comparison contained the Flatiron gain, while operational profit and cash grew sharply. OCF over the latest twelve months reached EUR 2.37 billion pre-factoring, and the company moved from EUR 1.41 billion net debt at June 2025 to EUR 72 million net cash at June 2026 on its reported definition.

The guidance trajectory matters more. FY2025 began with EUR 680–730 million operational-net-profit guidance, was raised to EUR 750–780 million and finished at EUR 789 million. FY2026 initially targeted EUR 950–1,025 million. On July 27 the range became EUR 1,025–1,100 million. Turner's own guidance went to USD 1.40–1.46 billion of operational PBT. Repeated upward revisions are evidence that management was not merely dressing an unchanged business in AI language.

The market is trading three real fundamental improvements at once: backlog growth, a favorable shift toward construction-management and collaborative contracts, and a higher share of advanced-technology work carrying stronger fee economics. The market is also trading a fourth thing that is not a fundamental improvement: scarcity value. There are few listed ways to own the largest US construction manager while AI hyperscalers are racing to build physical capacity. Scarcity value can persist for years, but it is a multiple, not a cash flow.

The order book deserves a breakdown before any celebration. FY2025 data-center orders of EUR 16.8 billion were nearly one third of total intake. The year-end data-center backlog was about EUR 15.2 billion. In H1 2026, management no longer gave an exact euro DC backlog but said trailing-twelve-month DC orders had more than doubled. That means concentration was moving upward, not downward.

The disclosed customer list also makes the underlying driver obvious: Meta projects in Canada and the US, NTT in Berlin, CoreWeave in Pennsylvania and Turner's participation in a Stargate-associated campus. The upside is that these are capital-rich customers pursuing strategic capacity. The downside is that the same technology-capex decisions can synchronize. Five apparently separate construction projects can still be one macroeconomic bet if all were authorized because the same group of hyperscalers expects explosive AI-compute demand.

Working capital is the second disagreement. Bulls can point to EUR 2.37 billion LTM OCF and conclude that HOCHTIEF's accounting is conservative. Bears can correctly answer that contractor cash flow is pro-cyclical. Both statements can be true. The data support excellent cash discipline today; they do not prove EUR 2.4 billion is a normalized annual cash-generation level.

Thiess strengthens 2027 reported earnings with relatively little execution risk from the transaction itself because the operation was already consolidated. The AUD 1.18 billion purchase is fundamentally an acquisition of the remaining minority claim on earnings. At the approximately EUR 673 million analytical translation used in this report, the assumed EUR 80 million earnings uplift would be an attractive high-single-digit multiple.

The financing effect should still be recognized. HOCHTIEF's EUR 71.6 million H1 net-cash figure is dated June 30; the Thiess purchase completed immediately afterward. A simple pro-forma subtraction of approximately EUR 673 million would turn that into around EUR 600 million of net debt before subsequent cash generation, financing actions or other July movements. This remains very modest relative to HOCHTIEF's scale, but using June net cash and simultaneously crediting the investor with the full Thiess earnings uplift double-counts the transaction.

Abertis is more leveraged economically. ACS's 2024 strategic plan reported Abertis net debt/EBITDA of 6.3 times for 2023 and envisaged around EUR 600 million of annual Abertis distributions through at least 2038. HOCHTIEF's 20% economic share of that planned distribution is roughly EUR 120 million a year. The toll-road debt stays inside Abertis and should not be added to HOCHTIEF's reported net debt; it should instead depress the equity value assigned to the 20% stake.

I value Abertis separately on that basis. Capitalizing roughly EUR 120 million of prospective distributions at an 8–10% required equity yield gives an indicative EUR 1.2–1.5 billion value for HOCHTIEF's stake before adjusting for future concession investments. That sits close enough to the economics of HOCHTIEF's original investment to be a sensible base rather than assuming toll-road equity deserves a construction multiple. The reported 6.3x Abertis leverage metric is stale, from 2023, so this range carries more uncertainty than the Turner valuation.

FlatironDragados also belongs outside the core multiple. HOCHTIEF is economically exposed to the civil-engineering platform but no longer consolidates the entire operation as if it were a wholly controlled business. The EUR 146 million 2025 remeasurement gain reinforces the accounting distinction. My SOTP values the interest at EUR 1.2–2.2 billion across scenarios rather than allowing equity-accounted earnings to inflate a consolidated Turner/CIMIC P/E.

The governance debate is more one-sided. Four of eight shareholder Supervisory Board representatives are independent of the controlling shareholder; the four ACS-linked representatives occupy key board and committee roles, and the CEO is shared with ACS. That arrangement can be perfectly legal and operationally productive while still warranting a minority discount. HOCHTIEF shareholders are buying participation, not control.

An ACS squeeze-out does not look imminent merely because ACS is near 80%; there is still a meaningful free float and HOCHTIEF has just joined the DAX. Yet the direction of ownership since 2011 has been increasing ACS control. I would assign zero value to a hypothetical future takeover premium. A future parent transaction could produce a premium, but it is not a rational component of intrinsic value.

The strongest bull evidence is the sequence of order growth, margin expansion and guidance upgrades; the strongest bear evidence is that the share price has capitalized those facts much faster than normalized earnings have risen.

Valuation Analysis

The market-cap arithmetic is simple: at EUR 426 and 77.7113 million issued shares, HOT's headline equity value is EUR 33.1 billion. Against FY2025 operational net profit of EUR 789 million, that is 41.9 times earnings. Against nominal EUR 902 million it is 36.7 times. Against the midpoint of 2026 operational guidance, EUR 1.0625 billion, it is 31.2 times.

† H2 2025 plus H1 2026. Calculated from company disclosures and the EUR 426 September 4 close.

The earnings multiple has plainly rerated. HOCHTIEF reported 26% operational-profit growth in 2025 while saying the stock had appreciated 160% over the preceding twelve months. Even without a perfect ten-year daily multiple series, that difference proves that multiple expansion, rather than earnings alone, supplied most of the market-value increase. I would describe today's valuation as top-decile-like relative to the pre-AI HOCHTIEF regime, but I will not invent a percentile without a consistent historical multiple series.

Cash-flow passthrough

FY2025 pre-factoring OCF was 2.30 times nominal net income; post-operating-capex cash flow was 1.56 times nominal income and 1.78 times operational profit. Over the latest twelve months to June 2026, OCF stayed extraordinarily high at EUR 2.37 billion and post-capex operating cash flow at EUR 1.66 billion.

I would not use EUR 1.66 billion as owner earnings. Fast backlog growth is pulling cash forward through working capital. My 2025 maintenance-capex estimate of EUR 500–550 million versus EUR 665 million total operating capex and leases leaves EUR 115–165 million as growth-oriented expenditure. Because 2025 D&A was roughly EUR 670 million, normalized operational owner earnings fall around EUR 0.90–1.0 billion. At the midpoint, roughly EUR 935 million, current price is around 35 times normalized owner earnings.

The cash/earnings gap works in the shareholder's favor but does not clear the valuation hurdle. A 2.9% owner-earnings yield on normalized 2025 economics is thin for a contractor with customer-cycle, project and controlling-shareholder risk.

Associates and SOTP discipline

Abertis receives a separate equity value of EUR 1.1 billion in the conservative case, EUR 1.4 billion in base and EUR 1.8 billion optimistic. The range is anchored to the roughly EUR 120 million annual share of Abertis distributions implied by ACS's plan and discounts the high concession leverage. Those three values imply required equity yields of about 10.9%, 8.6% and 6.7% on that EUR 120 million, so the scenario band is deliberately wider on both sides than the 8–10% range used for the standalone indication above.

FlatironDragados receives EUR 1.2 billion, EUR 1.7 billion and EUR 2.2 billion respectively. That is the least certain major component because current standalone debt, distributable cash and complete HOCHTIEF-attributable earnings are not exposed in the public HTML disclosures reviewed here. I use a deliberately broad range rather than smuggling its equity-accounted income into the core multiple.

For the consolidated operating businesses, I normalize away part of peak data-center profitability. The base case assumes roughly EUR 1.05 billion of sustainable attributable owner earnings from Turner, CIMIC/Thiess, Europe and corporate activities after excluding the associates. That is above the 2025 core run rate because it credits the Thiess minority buyout and structural growth, but below a straight extrapolation of Turner's 2026 guidance.

Absolute valuation scenarios

This is valuation-scenario analysis within a research framework, not investment advice.

The 24x base multiple is intentionally much richer than a conventional general contractor deserves. It credits Turner for a capital-light fee model, customer repetition, high-value advanced-technology execution and a longer structural data-center runway. It does not credit the business with a software or semiconductor multiple. The optimistic 28x core multiple already assumes the market remains willing to treat Turner more like a scarce AI-infrastructure beneficiary than a construction manager.

The current market price essentially asks investors to live close to the optimistic case. At EUR 426, the market is 19% above my base value and only 12% below the EUR 484 optimistic value. Very little of the thesis has to go wrong before returns become mediocre.

Expectation gap

The next expectation gap is unlikely to come from headline revenue. Investors will care about four numbers: Turner's order growth, data-center order growth, Turner's PBT margin and operating cash conversion. A 15% group revenue beat accompanied by slower DC intake would probably matter less than flat sales with another large increase in high-quality Turner backlog.

The most important unpublished number would be the share of Turner's backlog and PBT coming from each hyperscaler. The company currently reports enough to show that data centers are important but not enough to determine whether the risk is “many customers buying the same asset class” or “a handful of customers dominating profit.” Those are different risks.

A second expectation gap could come from cash. If new orders fall toward 1.0 times work done from H1's 1.5 times, working capital should stop providing the same tailwind. Earnings can remain sound while OCF falls several hundred million euros. A market that has begun treating EUR 2.4 billion as normal annual OCF could interpret that mechanically as deteriorating quality even when it is merely the mathematical reversal of growth cash.

Margin-of-safety recheck

Current price is approximately 79% above my EUR 238 conservative intrinsic-value case. On that test the margin of safety is zero.

The most fragile base assumption is sustainable core owner earnings of EUR 1.05 billion. Cutting that to 70%, while keeping the 24x base multiple and associate values unchanged, reduces the estimated equity value to approximately EUR 260 per share. The sensitivity is large because the market value now rests primarily on Turner's earnings rather than on tangible assets.

If 2026 operational earnings are flat for the following three years and the company continues to pay roughly 65% of them, the midpoint guide implies operational EPS of about EUR 13.67 on the issued share count. The theoretical dividend is around EUR 8.89, or about a 2.1% yield at EUR 426. With no earnings growth and no further multiple expansion, that yield is the bulk of the expected return. Under any base-date risk-free yield above 2.1%, the correct conclusion is: there is no margin of safety at this buy price.

This is a good operating asset at a difficult purchase price. Waiting has a real opportunity cost if Turner's data-center earnings continue compounding and the market keeps a 30x-plus multiple; the cost of not waiting is accepting a valuation where a normal construction-capex downturn can erase several years of operating progress.

Margin-of-safety sufficiency verdict: none.

Risks, Catalysts and Cross-Synthesis

The largest permanent-loss risk is data-center normalization. I assign it medium probability and high impact over three to five years. The observable indicators are data-center order intake, total new orders/work-done ratio, Turner's PBT margin and the disclosed DC share of backlog. The transmission mechanism is stronger than a simple revenue decline: fewer starts reduce high-fee work, shrink Turner's mix-driven margin advantage, reverse working-capital inflows and remove the narrative that justified moving HOT from a contractor multiple toward an AI-infrastructure multiple. Data centers were already 21% of backlog and 32% of 2025 new orders.

The second risk is working-capital reversal. Probability is medium-high at some point in a normal construction cycle; earnings impact is medium but cash-flow and sentiment impact can be high. With LTM pre-factoring OCF of EUR 2.37 billion against roughly EUR 0.9 billion LTM operational earnings, several hundred million euros of cash inflow is plainly not a permanent accounting annuity. The alert is new orders falling below work done for multiple quarters while contract liabilities and supplier payables unwind.

The third risk is project execution outside Turner. Probability is medium and potential impact high. Europe, CIMIC civil work and mining services contain more fixed-price, asset and operational exposure than Turner's fee model. A single poorly priced megaproject can consume years of segment profit. HOCHTIEF's deliberate shift toward lower-risk contracts, including approximately 90% of backlog classified as lower risk in Q1 2026, reduces the probability but does not eliminate it.

The fourth is valuation compression. Probability is high over a five-year horizon and impact high. The shares are roughly 31 times midpoint 2026 operational guidance and around 35 times my normalized 2025 owner earnings. A fall from 31x to 20x with earnings unchanged would remove roughly 35% of equity value before any operational disappointment. This risk requires no recession. It requires only that investors decide construction-management earnings deserve a construction-management multiple again.

The fifth is parent/minority governance. Probability of day-to-day value destruction is low-to-medium; impact around a large related-party transaction could be high. Indicators are ACS ownership changes, related-party asset transfers, the dividend payout, financing between group entities and board changes. Four of eight shareholder representatives are not independent of the controlling shareholder, while the CEO also runs ACS.

Thiess commodity exposure is a secondary risk rather than the thesis driver. The operation has diversified into critical minerals and underground work, and its long-duration contracts reduce spot-price sensitivity, but mining fleet utilization ultimately follows customers' capital budgets. The exact current coal share is not sufficiently disclosed in the sources reviewed here, which prevents a more precise commodity stress.

Catalysts and tracking dashboard

Positive catalysts include another Turner guidance increase, data-center intake remaining above work done without further concentration, new nuclear/grid awards, stronger German infrastructure conversion, and evidence that the Thiess minority purchase delivers the expected earnings without stressing cash. The most powerful positive surprise would be proof that Turner's non-data-center advanced-technology profit is accelerating fast enough to replace DC normalization.

Negative catalysts are more asymmetric at today's valuation: a data-center order decline, hyperscaler campus deferrals, Turner PBT-margin retreat, OCF normalization that the market mistakes for accounting weakness, a large fixed-price loss in CIMIC or Europe, or ACS increasing control in a way that further reduces free float.

Current financial thresholds derive from HOCHTIEF and ACS disclosures. I have not captured a primary-source exact calendar date for the 9M 2026 release, so “early November” is an expectation based on HOCHTIEF's reporting cadence rather than a confirmed date; the comparable 9M 2025 report was November 6.

Cross-synthesis

Viewed vertically, HOCHTIEF's proven capability is not simply pouring concrete. Across repeated changes of ownership, geography and project type, the company has survived by moving toward coordination of increasingly complicated infrastructure. Turner is the clearest modern expression of that capability: clients entrust it with enormous gross project value while Turner commits relatively little physical capital compared with a self-performing contractor. CIMIC brings a different skill, the ability to operate transport, services and mining contracts at scale. HOCHTIEF Europe preserves deep civil-engineering competence. The group has gradually assembled a portfolio in which project expertise can be applied to whatever infrastructure category currently has capital behind it.

Past success came from both era tailwinds and management choices. German reconstruction, global megaproject spending, Australian mining and today's digital infrastructure all supplied external demand. Yet the current earnings acceleration cannot be dismissed as luck. Management reduced high-risk contract exposure, bought CIMIC minorities, recombined Flatiron with Dragados, re-consolidated Thiess ownership and put Turner at the center of the advanced-technology strategy. Operational PBT margins moved from 3.0% in 2024 to 3.4% in 2025 and 3.8% in H1 2026 while backlog and cash rose.

Those capabilities remain present. What has changed is the price at which the market asks a minority shareholder to buy them.

Horizontally, Turner's strongest advantage over a conventional contractor is its risk-adjusted business model. A general contractor's backlog can conceal future losses because every fixed-price project is an unexpired estimate. A construction manager's backlog can contain much more pass-through revenue and fee-like economics. That is why comparing Turner's 3.9% PBT margin with a specialty contractor's margin without adjusting the revenue denominator gives a misleading answer. The flip side is that Turner's current high-value mix is linked increasingly to hyperscaler construction, so business-model quality does not eliminate end-market cyclicality.

HOCHTIEF's weakness versus a clean independent US compounder is governance. The shareholder is buying a 19–20% minority slice of a company whose CEO also runs the 80%-owner and whose board structure gives ACS considerable influence. Related-party combinations may be strategically sensible, as FlatironDragados arguably is, but the minority holder cannot determine the exchange ratio, the timing of a future asset transfer or the dividend. The 65% payout has so far been shareholder-friendly; it is also naturally attractive to a parent receiving most of the distribution.

The market is pre-spending future success rather than simply rewarding past success. FY2025 operational profit grew 26%; the stock's preceding twelve-month appreciation was 160%. H1 2026 then delivered another large earnings upgrade, validating some of that rerating, but the shares still sit above 30 times current-year operational guidance. The investor buying today needs two propositions to remain true: earnings must stay structurally higher after the data-center surge, and the market must continue paying a large premium for those earnings.

I think the market is most likely misjudging the distinction between a long AI-infrastructure runway and a smooth annual construction cycle. Both can coexist. Compute demand may compound for a decade while physical data-center starts fall by one third for 12–24 months because power, permits, financing, chip availability or simply prior over-ordering create a pause. HOCHTIEF's backlog would postpone the earnings impact, which could initially make the bear thesis look wrong. Working capital and margins would reveal it later.

For the next year, the critical variable is Turner margin plus order intake. The 2026 guidance already implies a large step up, so another revenue beat matters less than whether the USD 1.4–1.46 billion PBT base proves durable. For three years, the critical variable is data-center mix: can healthcare, life sciences, semiconductors, energy and transport take over as incremental earnings drivers when hyperscaler starts normalize? For five years, the question becomes capital allocation. HOCHTIEF wants to move further into ownership and development of digital and energy assets. That can make the group worth more than a construction multiple, but only if it earns investment-type returns without sacrificing Turner's capital-light model.

Abertis illustrates the discipline required. HOCHTIEF owns a valuable toll-road income stream but the asset also carries substantial project-level leverage. It should be valued as concession equity. FlatironDragados should be valued as an associate. Thiess deserves a mining-services multiple and higher maintenance-capex assumption. Turner deserves the highest multiple. Multiplying all reported group earnings by one premium number hides every one of those economic differences.

The minority structure then deserves one final subtraction. At approximately 80% ACS ownership and a roughly 20% float, HOT can enjoy DAX liquidity while still trading as a controlled company. Its June 2026 DAX promotion is technically significant because constituent weights are free-float adjusted, but the promotion did not make the company's claims on Abertis, Turner or Thiess larger.

Bull and bear reasons

The core bull case is unusually concrete:

1. Turner is still accelerating: H1 2026 parent reporting showed EUR 551 million PBT and a 3.9% margin, while HOCHTIEF lifted full-year Turner PBT guidance to USD 1.40–1.46 billion.
2. The EUR 84.8 billion group backlog equals roughly two years of revenue, and H1 intake was 1.5 times work performed, delaying the impact of any near-term macro slowdown.
3. FY2025 data-center orders of EUR 16.8 billion and subsequent doubling of trailing DC orders show that the AI-infrastructure narrative is backed by actual contracts rather than investor storytelling.
4. Contract risk has been reduced materially, with roughly 90% of Q1 2026 backlog categorized by management as lower-risk.
5. Buying the remaining Thiess 40% at AUD 1.18 billion appears earnings-accretive at a high-single-digit multiple on my assumed EUR 80 million of incremental earnings.

The bear case is equally specific:

1. Data centers were already 21% of backlog and 32% of FY2025 orders, meaning the incremental backlog is much less diversified by end-market than the headline EUR 84.8 billion suggests.
2. At EUR 426 the stock is roughly 31x midpoint 2026 operational guidance even after two major years of earnings acceleration, leaving large multiple-compression risk.
3. LTM OCF of EUR 2.37 billion substantially exceeds normalized earnings and will partly reverse if orders fall toward or below work done.
4. Roughly 80% ACS control, overlapping management and four shareholder-board representatives not independent of the controlling shareholder justify a minority discount.
5. A one-third decline in data-center starts could plausibly remove roughly EUR 135–140 million of group after-tax earnings after backlog protection fades, before the cash effect of working-capital reversal.

Pre-mortem: where this thesis can fail

The first three-year loss script starts in 2027. Hyperscaler capex remains enormous in dollars, but 2025–26 ordering turns out to have pulled projects forward. US data-center starts fall about one third as power interconnections and prior capacity additions slow campus launches. Turner's backlog keeps 2027 revenue respectable, so the first warning appears in new orders rather than sales. By 2028 Turner's PBT falls from the roughly USD 1.43 billion 2026 midpoint toward USD 1.1–1.2 billion, the group loses roughly EUR 150 million of annual attributable earnings, working capital consumes EUR 500–700 million, and investors lower HOT from around 30x earnings toward 18–20x. Even if the non-data-center businesses remain intact, the stock can trade in the low-to-mid EUR 200s.

The second script is less cyclical. Turner keeps performing, but 2026 proves to be near the maximum valuation scarcity premium. Growth decelerates from 30–40% to high single digits as the comparison base catches up. The market begins valuing HOCHTIEF against contractors rather than AI-infrastructure specialty names, while ACS increases its stake and free float declines further. EPS does not collapse; the multiple does. A move from roughly 31x 2026 operational earnings to 20x on EUR 14–15 of sustainable EPS gives a EUR 280–300 share before dividends. A 30% earnings disappointment on top of that produces the roughly 50% loss the pre-mortem is intended to expose.

Final research conclusion

HOCHTIEF has become a better company faster than its old construction label suggests. Turner has genuine execution advantages, a favorable construction-management model and exceptional positioning in the current digital-infrastructure build. CIMIC supplies diversification, Thiess adds long-duration mining-services earnings, and Abertis and FlatironDragados give minority holders valuable assets outside consolidated contracting. The repeated 2025–26 guidance upgrades and the movement from a 3.0% to a 3.8% group operational-PBT margin show that the transformation has reached reported earnings.

The purchase price asks for too much of that success in advance. My base SOTP is approximately EUR 358 per share and my optimistic case approximately EUR 484. EUR 426 sits between fair-value and optimistic-value territory, while the conservative case is only EUR 238. The asymmetry is poor: the plausible upside from another strong AI-infrastructure cycle is modest compared with the loss if a normal capex pause and multiple normalization arrive together. The roughly 80% controlling shareholder makes the asymmetry slightly worse because minority investors should neither value a control premium nor assume capital allocation will always be optimized for the float.

I would change that judgment in either of two ways. Fundamentally, I would raise normalized value if Turner could show sustained non-data-center advanced-technology growth while keeping a PBT margin near 4% through a slowdown in data-center starts. On price, the stock becomes much more interesting after a large enough decline that the buyer is paid for the cyclical and governance risks rather than asked to waive them.

【Company-profile scores】

• Fundamental quality: high
• Growth: high
• Moat: medium
• Financial soundness: strong
• Management credibility: high
• Valuation attractiveness: low
• Risk level: high
• Suitable investor type: long-term growth and cyclical investors able to tolerate substantial valuation compression

【Investment rating】

• Rating: Watch
• One-line thesis: Turner's earnings quality is real, but EUR 426 prices much of the data-center boom while leaving no conservative-case margin of safety.
• Current-price classification: outside the three valuation bands, above the acceptable-hold band and below the clearly-overvalued band.
• Whether to wait for a better price: yes. I would require at least the low EUR 300s for a normal hold proposition and EUR 180–190 for a ≥20% margin of safety against my conservative valuation.
• Target holding horizon: 3–5 years for fundamental assessment; 12 months for the data-center/order-intake cycle.
• Expected annualized return: roughly -16% in the conservative case, about -4% in the base case and about +6% in the optimistic case after allowing approximately 1.5–2 percentage points a year of dividends; realized dividends will depend on earnings and the 65% payout policy.
• Max-loss risk: about 45–50% in the pre-mortem, triggered by a one-third data-center-start decline, Turner PBT falling toward USD 1.1–1.2 billion and a simultaneous derating toward 18–20x sustainable earnings.
• Reassessment triggers: Turner PBT margin below 3.3% for two reporting periods; group intake/work done below 1.0x for two periods; data-center backlog above 30% without materially improved customer disclosure; operational-net-profit guidance below EUR 950 million; or a material ACS ownership/related-party transaction that changes minority economics.

【Ideal Buy Price】180–190 EUR

Basis: at least 20% below the approximately EUR 238 conservative intrinsic-value case; this deliberately requires a deep discount because the conservative case already assumes normalized rather than recessionary earnings.

Acceptable hold price: EUR 305–410, centered on the approximately EUR 358 base SOTP and remaining within roughly ±15%.

Clearly overvalued price: EUR 535–560, beginning more than 10% above the approximately EUR 484 optimistic scenario.

【Valuation Range】

• current: 426.00 EUR (close as of 2026-09-04)
• bear (conservative · ideal buy zone): [180, 190]
• base (fair · acceptable hold zone): [305, 410]
• bull (optimistic · above the clearly-overvalued line): [535, 560]

Research uncertainties

Five blind spots materially limit precision.

First, HOCHTIEF's public shareholder page does not expose a fresh 2026 voting-rights breakdown in searchable text. The approximately 80.32% ACS figure is the best current indication identified in this research pass, but I have not treated the second decimal place as independently primary-verified.

Second, the FY2025 public results releases allow the EUR 113 million nominal-versus-operational gap to be reconstructed and explicitly identify the EUR 146 million Flatiron gain. They do not expose enough category detail to allocate the approximately EUR 33 million offset line by line among restructuring, M&A, impairment and other items.

Third, HOCHTIEF does not publish a complete end-market and customer-level backlog cube. Data centers can be quantified at 21% of FY2025 backlog, defense above EUR 2 billion and strategic-growth-market intake around 60% at H1 2026, but healthcare, education, transport, energy, semiconductors and mining cannot be reconciled into mutually exclusive percentages.

Fourth, current standalone leverage, dividends and distributable cash for FlatironDragados, and an equally current standalone Abertis debt bridge, were not available in sufficiently complete searchable disclosures used here. Abertis carries a wider SOTP range and FlatironDragados the widest. The latest clearly sourced leverage anchor used for Abertis is ACS's 6.3x net-debt/EBITDA disclosure for 2023.

Fifth, exact daily closing prices for every 2024–26 corporate event and a fully synchronized current peer-multiple dataset were not available from one consistent primary-source basis. I use verified price anchors and an explicit qualitative rerating attribution rather than fabricate event-day precision.

Principal sources

The primary evidentiary base is HOCHTIEF's FY2025 results release and financial table, which supplies sales, operational and nominal earnings, cash flow, orders, backlog, the dividend and the 21% data-center backlog share.

HOCHTIEF's H1 2026 release supplies the raised guidance, EUR 84.8 billion backlog, EUR 31.5 billion orders, cash-flow figures, net cash and the statement that trailing data-center orders more than doubled.

The Q1, H1 and 9M 2025 releases provide the Flatiron one-off bridge and the sequence of guidance upgrades.

CIMIC's Thiess releases cover the AUD 1.18 billion final-minority purchase, previous 60% ownership, 2023 Thiess revenue, project count and strategic commodity diversification.

ACS's H1 2026 results and strategic plan are used only where the parent supplies data for the same underlying Turner/CIMIC operations or parent-specific information on Abertis, capital allocation and consolidation. They are not treated as independent verification of Turner.

HOCHTIEF's corporate-governance and individual Supervisory Board pages establish board composition and ACS affiliations.

STOXX's June 3 announcement establishes HOCHTIEF's DAX addition effective June 22 and the free-float basis of the index framework.

HOCHTIEF's corporate history gives the 1873 foundation, founders and 1896 conversion into a stock corporation.

Other tickers mentioned

• ACS.MC: controlling parent whose filings provide consolidation context and direct Abertis exposure
• SKA-B.ST: European general-contractor reference for project-risk and working-capital economics
• DG.PA: VINCI, used as a concessions and infrastructure-ownership reference rather than a direct Turner margin peer
• FLR.US: US large-project engineering and construction reference for contract-risk discipline
• EME.US: US specialty-contractor reference for data-center mechanical and electrical exposure
• PWR.US: US specialty infrastructure reference for electrical and power-related data-center build-out
• STRL.US: US specialty infrastructure reference for the higher-value-added side of data-center construction

This report is based on public information and does not constitute investment advice. Markets carry risk; invest with caution.