Daqin Railway: China's Coal Artery at the Energy Transition's Edge

I. Introduction & The Case (5 min)

There is a stretch of double track in northern China where, if you stood beside it for twenty-four hours, you would watch roughly eighty-two loaded coal trains grind past you — and about fifty of those trains would each weigh twenty thousand tonnes. Not twenty thousand tonnes of train and cargo spread across a day. Twenty thousand tonnes in a single consist, more than two kilometres of steel hoppers, hauled by locomotives synchronised by radio because no human hand can coordinate that much inertia by feel.¹

That track is the 大秦铁路 Daqin Line, 653 kilometres from the coal fields around 大同 Datong in 山西 Shanxi province to the port of 秦皇岛 Qinhuangdao on the Bohai Sea.² In 2025 the listed company that operates it, 大秦铁路股份有限公司 Daqin Railway Co., Ltd., shipped 548.84 million tonnes of coal across its network — 19.9% of all coal moved by rail in China, and 26.3% of everything the national railway system hauls in coal.¹ One corporate entity, anchored on one line, moves roughly a fifth of the coal that travels by rail in the world's largest coal economy.

And here is the paradox that makes this a story rather than a spreadsheet. In 2025, Daqin Railway's revenue rose 4.04%, to RMB 77.65 billion. Its net profit attributable to shareholders fell 34.73%, to RMB 5.90 billion. Gross margin collapsed from 15.02% to 8.13%. The fourth quarter produced an outright operating loss of RMB 324 million — a loss, at a business that has printed cash more reliably than almost any other asset on the Shanghai exchange for two decades.³⁴

That gap between rising revenue and collapsing profit is the whole episode. It is what happens when an irreplaceable piece of physical infrastructure — a genuine cornered resource — starts running into a demand curve that is bending the wrong way, and management responds by buying growth it cannot generate organically.

There are two competing frames for Daqin, and both are defensible. Frame one: this is an infrastructure monopoly of a kind that essentially cannot be rebuilt, throwing off cash, run near its engineering ceiling, and now returning close to 80% of its earnings to shareholders. Frame two: this is a melting ice cube. Its single commodity is thermal coal, and China is systematically substituting electrons for coal trains — moving power from generation regions to demand centres over 特高压 ultra-high-voltage transmission lines instead of moving the fuel itself by rail.

The honest answer is that both frames are true simultaneously, on different time horizons, and the investment question is which one dominates first.

Here is the route this story takes. It starts with a Mao-era-lineage engineering project born of a 1980s energy crisis, becomes a listed company inside China's state rail monopoly in 2006, builds a physical position no competitor has been able to dislodge — and then, in the 2020s, finds that the threat is not another railway but the power grid. It ends with a management team that has chosen a genuinely unusual response for a Chinese state-owned enterprise: instead of empire-building to replace declining volumes, it has been shrinking its own share count and pushing its payout ratio to 75% and beyond.

Whether that is discipline or resignation is the question worth arguing about. Let's start where the track does.

---

↑ Back to Top

II. Origins: Building China's First Heavy-Haul Railway (1983–1992) (7 min)

On January 1, 1985, roughly seventy thousand construction workers began assembling in the 太行 Taihang and 燕山 Yanshan mountain ranges in the dead of a northern Chinese winter.⁵ They were there because the Chinese economy had run into a wall that had nothing to do with ideology and everything to do with logistics.

China in the early 1980s was an energy-starved industrial economy in the first flush of 改革开放 reform and opening. The coal existed — the basins across Shanxi, 内蒙古 Inner Mongolia, and 陕西 Shaanxi held one of the largest concentrations of accessible thermal coal on the planet. The demand existed too, on the eastern seaboard, where the factories and the power plants and the cities were. What did not exist was a way to connect them at scale. The existing rail network was a patchwork of mixed-traffic lines designed for a different century's freight volumes, and coal sat at pitheads while coastal power plants rationed load.

The state's answer was surveyed and designed beginning in 1983 and broke ground two years later: a purpose-built corridor running from Hanjialing near Datong east to the Bohai coast at Qinhuangdao, 653 kilometres across four provincial-level jurisdictions — Shanxi, 河北 Hebei, 北京 Beijing and 天津 Tianjin — with 37 stations along its length.² The full line opened to traffic on December 21, 1992.⁵

The technical choice made at the outset is the one that matters for everything that follows. Daqin was not built as a better conventional freight railway. It was built as China's first double-track, electrified, heavy-haul unit-train coal line — the first Chinese railway to run centralised microcomputer dispatching with fibre-optic communication along the entire route.⁶

The distinction between conventional freight and heavy-haul unit trains is worth pausing on, because it is the source of the moat. A conventional freight railway is a bit like a bus network: mixed cargo, mixed destinations, wagons assembled and disassembled at marshalling yards, trains stopping and starting. A heavy-haul unit train system is more like a conveyor belt. The same commodity, loaded at the same handful of origins, unloaded at the same handful of destinations, in fixed-composition trains that essentially never break up. You do not sort. You do not shunt. You just cycle the same trainsets around a loop, continuously, and the economics come from throughput per kilometre of track rather than from flexibility.

That design decision has three consequences that compound. First, cost per tonne-kilometre falls dramatically, because fixed infrastructure is amortised across enormous volume. Second, the physical plant is specialised to the point of being nearly useless for anything else — the track structure, the ballast depth, the electrification, the loading and unloading terminals, the signalling headways are all tuned to one job. Third, and most importantly, the whole system only works if you control the entire corridor end to end, including the port.

Which is exactly why it has never been duplicated. To build a competitor to Daqin you would need to assemble a continuous right of way through mountainous, densely populated territory, build a heavy-haul-rated formation on it, electrify it, connect it to producing mines at one end, and secure deep-water coastal terminal capacity at the other — and then find enough incremental coal demand to justify running it. Each of those is hard. All four at once, in the same corridor, has proved practically impossible.

The origin story matters less as history than as a description of a physical fact that still governs the business in 2026. Daqin is not a company that built a brand or a technology or a network of relationships. It is a company that occupies a specific groove in the earth, and the groove is the asset.

What turned that groove into a security you could buy came twelve years later.

---

↑ Back to Top

III. From State Asset to Public Company: IPO and Corporate Structure (2004–2006) (8 min)

Daqin Railway Co., Ltd. was incorporated on October 28, 2004, and its shares began trading on the Shanghai Stock Exchange on August 1, 2006 under the code 601006.⁷

It is easy to read that as a normal corporate event and miss what it actually was. This was not an entrepreneur taking a company public. It was the Chinese state carving a revenue-generating slice out of its rail bureaucracy, wrapping it in a joint-stock company, and selling minority participation in the cash flows while retaining unambiguous control. The distinction is not academic. It determines who decides what, and it determines whose interests get optimised when they conflict.

The ownership chain runs like this. Daqin Railway Co., Ltd. is controlled by 中国铁路太原局集团有限公司 China Railway Taiyuan Group, which holds roughly three-fifths of the listed shares.⁸ Taiyuan Group is in turn wholly owned by 中国国家铁路集团 China State Railway Group, the state entity that succeeded the old Ministry of Railways and that operates essentially the entire national network. So a public shareholder in Daqin is a minority participant in a subsidiary of a regional railway bureau of a national railway monopoly.

What does that mean practically? It means the things that would be board-level strategic decisions at an independent railroad — how much network capacity to build, what tariffs to charge, whether to accept a low-margin haul in service of national energy security, whether adjacent assets get injected into the listed vehicle — are substantially decided at the system level, above the listed company. Daqin's management runs the railway. It does not, in any meaningful sense, set the terms on which the railway operates.

The corporate structure produced one genuine curiosity, and it is worth flagging early because it reframes the capital-allocation discussion later. Daqin holds a 41.16% equity stake in 神朔铁路 Shuohuang Railway, the coal corridor operated under 国家能源集团 CHN Energy that runs from Shanxi across to the port at 黄骅 Huanghua.⁹ In 2024 alone, that stake paid Daqin RMB 2.206 billion in cash dividends — a material contribution to consolidated earnings that arrives as investment income rather than operating profit.⁹

Here is the thing to notice. That stake was not a deal. Nobody at Daqin ran a process, negotiated a price, or fought a competitive auction for it. It came with the box at formation — an artefact of how the state divided rail assets between the railway system and the coal system in the 1990s and 2000s. Which means that as of 2026, this company, more than twenty years after incorporation, has essentially never executed a conventional discretionary acquisition. Hold that thought. When we get to capital allocation and the question of whether management's restraint is a virtue or a constraint, the fact that the "M&A track record" consists entirely of an asset it was handed matters enormously for how much credit to give.

There is a passenger business too, and proportion demands honesty about it. Daqin ran 46.41 million passenger trips in 2024, up 8.3% year on year and a five-year high, and it budgeted for 45 million trips in 2026.¹⁰¹ It is a real operation, it recovered well from the pandemic trough, and it is not what you own this company for. Freight generated 71.09% of main operating revenue in 2025 and coal dominates freight.³ Passenger rail is a steady sideline attached to a coal machine.

Shuohuang, incidentally, is the second half of a picture most people miss: by April 2025 that line had hauled a cumulative five billion tonnes of coal since opening.¹¹ Between the line Daqin operates and the line it part-owns, this single listed company touches a startling proportion of the coal that moves on Chinese steel.

So what does the structure tell an investor? Two things, and they point in opposite directions. The good news is that state control has historically meant Daqin gets the coal — when national energy security requires maximum throughput, this line gets prioritised, and its volumes are protected by policy in a way a private operator's would not be. The bad news is the same fact viewed from the other side: a company whose volumes are allocated and whose tariffs are administered does not get to convert its monopoly position into monopoly pricing. It gets to be essential without getting to be extortionate.

Which brings us to the machine itself, and the question of exactly what kind of moat 653 kilometres of track actually is.

---

↑ Back to Top

IV. Anatomy of a Natural Monopoly: How the Daqin Line Actually Works (16 min)

Every spring and autumn, for a couple of weeks, China's thermal coal traders start watching a maintenance calendar the way commodity traders elsewhere watch weather. The Daqin line shuts sections down for scheduled overhaul, throughput drops, coal stocks at Qinhuangdao draw down, and prices at the port twitch.¹² That a routine maintenance window on a single railway is a tradeable event in a national commodity market tells you more about this asset's position than any market-share statistic.

Let's take the machine apart.

The physical conveyor

The core of the business is a loop. Loaded trains run east from the Datong-area loading points to the Bohai coast; empty trains run back west. Coal is loaded at high speed under gravity from silos, and unloaded at the port by rotary dumpers that invert entire wagons without uncoupling them. In 2025, the line averaged 81.6 loaded heavy-haul trains per day, of which 50.7 were 20,000-tonne consists.¹

Think about what that operating pattern requires. At those volumes on a double-track line, trains are running at headways measured in a handful of minutes. Everything is choreographed: the loading silos, the port dumpers, the traction power supply, the locomotive cycling, the crew changes. A twenty-thousand-tonne train is not simply a longer train — it is two or three locomotive groups distributed through the consist, coordinated electronically, because the coupler forces at the head of a single-headed train that long would tear it apart. This is why the moat is not merely "someone else would have to lay track." Someone else would have to lay track and rebuild two decades of operating know-how in synchronised distributed-power heavy-haul dispatch.

In 2025 the line moved 390.04 million tonnes of freight, down just 0.5% — a decline of 2.11 million tonnes.¹³ For 2026, management set a target of 400 million tonnes for the line.¹

That 400 million tonne number deserves a hard look, because it is the single most misread figure in this business. It is not a growth target. It is roughly the practical throughput ceiling of the asset as configured — the number the line has been bumping against, above and below, for years. When management "targets 400 million tonnes," it is targeting full utilisation of an asset that is already essentially full.

The analytical consequence is severe and under-appreciated: the core asset has no organic volume growth available to it. Every incremental tonne the group hauls must come from somewhere other than the Daqin line — other corridors, other commodities, or acquired traffic. Any narrative that treats Daqin as a business that can grow into rising Chinese energy demand is describing a company that does not exist. The line is a fixed-capacity toll road. The only questions are utilisation and price.

Where the money comes from

Freight was 71.09% of main operating revenue in 2025, and thermal coal out of the Shanxi–Inner Mongolia–Shaanxi complex is the dominant commodity within it.³ Group-wide, Daqin shipped 680.73 million tonnes of freight in 2025, down 3.6%, representing 12.9% of national rail freight volume, while freight turnover actually rose 1.8% to 375 billion tonne-kilometres — meaning the average haul lengthened even as tonnage fell.¹

That last detail is quietly informative. Longer average hauls at lower tonnage is the signature of a network taking on more through-traffic and joint-carrier movements: cargo that travels beyond Daqin's own metals, requiring settlement payments to other railway operators. It generates revenue and turnover, and it generates cost. We will see that show up violently in the 2025 margin.

Porter, applied honestly

Run the five forces on this asset and you get an unusually lopsided picture.

Threat of new entrants: close to zero. Not because of patents or brand, but because of right-of-way scarcity and capital intensity. The proof is empirical rather than theoretical, and we will meet it in Section IX: the Chinese state itself built a rival corridor with fresh capital and full political backing, and it has not displaced this one.

Supplier power: moderate, and mutual. Shanxi coal producers supply the cargo, and in a soft coal market they have more leverage over the marginal tonne than they do in a tight one. But the dependency runs both ways — for a Shanxi producer selling into coastal power markets, Daqin is the economic route to the sea. Neither side can walk away.

Buyer power: real and concentrated, but structurally blunted. The end buyers are coastal power generators and coal traders, a concentrated set with genuine bargaining ability. What they lack is an alternative. Nobody trucks 400 million tonnes of coal 650 kilometres. The bulk-logistics substitution set at this scale is empty.

Rivalry among existing corridors: present but limited. Other coal routes exist. None share Daqin's combination of distance economics, throughput, and terminal position at Qinhuangdao.

Threat of substitutes: this is where the case lives or dies. And the substitute is not another way to move coal. It is a way to not move coal at all — generating power at the pithead and shipping the electricity. That is Section V, and it is the reason the other four forces matter less than they appear.

Seven Powers, applied honestly

In Hamilton Helmer's framework, Daqin's advantage is a clean pairing of scale economies and cornered resource. The scale economies are the unit-train conveyor: cost per tonne-kilometre that falls with throughput and is unmatchable at lower volumes. The cornered resource is the corridor itself — a specific alignment of land, grade, electrification and port access that cannot be acquired at any sane price because it cannot be assembled at all.

But here is the qualifier that separates Daqin from, say, a North American Class I railroad or a European toll motorway with inflation-linked concessions. A regulated monopoly's moat protects the asset without conferring pricing power. Freight tariffs on Chinese rail are administered within a state framework. Daqin cannot look at a coastal power plant with no alternative and price accordingly, because the framework it operates inside exists partly to prevent exactly that. The moat guarantees that nobody takes the volume away. It does not guarantee that the volume earns an economic rent.

This is the single most important analytical point about the business, and it explains a puzzle that otherwise looks strange: how can a company with a near-unassailable physical monopoly post a fourth-quarter operating loss? Because a monopoly on a regulated tariff with rising costs and flat volume is not a licence to print money. It is a licence to be indispensable at whatever margin the system permits.

Which is a comfortable enough position — right up until the system decides it needs less of what you do.

---

↑ Back to Top

V. The Inflection: China's Energy Transition Meets a Coal Railroad (2020s) (12 min)

In 2025, something happened in China's power system that had not happened in a decade: coal-fired generation fell. Not its share — its absolute volume, down roughly 0.7% year on year, even as the country's total electricity consumption climbed past ten trillion kilowatt-hours for the first time in history.¹³¹⁴

Sit with that for a second, because it is the entire bear case compressed into one data point. Chinese electricity demand grew. Chinese coal burn for power did not. The wedge between those two lines is what threatens a coal railway.

The mechanism, explained plainly

There are two ways to get the energy in a lump of Shanxi coal to a factory in 广东 Guangdong. You can put the coal on a train, take it 650 kilometres to a port, load it on a ship, sail it down the coast, unload it, and burn it in a coastal power station. Or you can burn the coal where it was dug, and send the electricity down a wire.

For most of modern Chinese history the first option won, because long-distance electricity transmission was lossy and expensive. Moving a tonne of coal was cheaper than moving the equivalent electricity. What changed is ultra-high-voltage transmission — think of it as the difference between a garden hose and a fire main. At 800 kilovolts direct current or 1,000 kilovolts alternating current, you can push enormous quantities of power across thousands of kilometres while losing only a small fraction along the way. By 2025 China operated 46 UHV lines, with cross-regional and inter-provincial transmission capacity of about 370 million kilowatts.¹⁴

Every gigawatt of pithead generation exported by wire is coal that never boards a train. The railway is not being out-competed. It is being disintermediated.

And the second force is larger still. In 2025 China added roughly 452 million kilowatts of renewable capacity. Wind and solar together reached about 1.84 billion kilowatts of installed capacity — 47.3% of the national total, surpassing coal as the largest single source of installed generating capacity for the first time.¹³ Coal's share of actual electricity generated fell from 68.5% in 2020 to 59.8% in 2025.¹³

Renewables do not need a railway at all. A solar farm's fuel logistics chain is sunlight.

The evidence already visible in Daqin's numbers

This is not a forecast. It has been showing up in the operating data for three years.

Coal shipments on the Daqin line fell 7.09% in 2024, to 392.15 million tonnes.¹⁵ Across the group in 2025, total coal volume fell 6.8% to 548.84 million tonnes, and total freight fell 3.6%.¹ The core line itself held roughly flat in 2025 — down half a percent — which tells you where the pain landed: on the group's broader coal network rather than on the protected trunk route.¹³

And then the financial evidence, which is worse than the volume evidence. Revenue rose 4.04% in 2025 while operating costs rose 12.25% — cost growth running at three times revenue growth. Gross margin fell 6.89 percentage points. Operating cash flow contracted 46.47%, to RMB 4.99 billion.³⁴

Read those together and the conclusion is uncomfortable but clear: Daqin held revenue up in 2025 by taking on volume and business that cost more to serve than it earned. That is a very different situation from a business suffering a cyclical volume dip with stable unit economics. It is a business substituting low-quality revenue for high-quality revenue, and the cash flow statement — down nearly half — is the tell that the substitution was expensive.

Management's own explanation, given with the annual results, attributed the decline to declining coal volumes, an adjusted transport structure, a logistics market still being "cultivated," and freight rate fluctuations.¹⁶ The specific cost drivers disclosed were more revealing than the summary: increased payments arising from more through-and-joint transport, higher end-to-end auxiliary logistics fees from the new turnkey logistics business, and rising labour costs.⁴

Two of those three are self-inflicted in the sense that they are consequences of a chosen strategy, not of the coal market. That distinction matters for judging management, and we will return to it.

The case against panic

The bear case is directionally right and it would be a mistake to soften it. But it would also be a mistake to model a straight-line collapse, for three reasons that are mechanical rather than hopeful.

First, renewables are intermittent and hydropower is volatile. China's grid still needs firm, dispatchable capacity to balance a system with nearly half its installed capacity coming from weather-dependent sources. Coal plants increasingly run as peaking and balancing assets rather than baseload — fewer running hours, but still burning coal, and still needing it delivered. A drought year in the southwest that cripples hydro output translates directly into more coal burn and more rail volume.

Second, the coastal load centres are precisely where domestic renewable resource is weakest relative to demand density. UHV substitutes for coastal coal burn at the margin; it does not eliminate it.

Third, and most concretely: in the first half of 2026, Daqin line freight volume rose 7.96% year on year, which the company attributed principally to a recovery in coal transport, and management stated that coal volumes are secured for a period ahead.¹⁷ Whatever the multi-decade direction, the near-term line is not falling off a cliff.

The correct framing, then, is neither "structural collapse imminent" nor "cyclical trough, buy the dip." It is that the ceiling on this business has been permanently lowered while the floor remains firm for the medium term. Volumes will oscillate around a slowly declining trend rather than fall off it. Which means the question for the next decade is not whether volume declines, but what management does with the cash while it does.

They have two answers. One is to change what the railway sells. The other is to shrink the company. We will take them in that order.

---

↑ Back to Top

VI. Reinvention: From Hauler to Logistics Platform (10 min)

On March 19, 2025, a container train pulled out of Datong carrying 2,700 tonnes of ethylene glycol, bound for the sea under a single transport document covering rail and ocean legs together. It arrived six days faster than the same cargo would have moved under the old arrangement, at a freight cost roughly 4% lower.¹⁸

Ethylene glycol is not coal. A single-document multimodal container service is not a unit train. And that is precisely the point: this was Daqin trying, in a small and concrete way, to become a different kind of company.

What actually changed

In early 2024, 国铁集团 China State Railway Group restructured its freight commercial organisation, standing up forty railway logistics centres across the country. The 大同铁路物流中心 Datong Railway Logistics Center was one of them.¹⁸ The intent was to convert railways from carriers — entities that move a wagon from A to B and hand it off — into logistics integrators that sell a delivered outcome.

The distinction sounds like corporate wordplay and is not. A carrier sells line-haul. A logistics integrator sells door-to-door: it arranges the truck at the mine, the loading, the rail leg, the port handling, the vessel booking, the last mile. It takes responsibility for the whole chain, prices it as one number, and keeps the difference.

The shift shows up in one striking operating statistic. Turnkey contract logistics — 总包 business, where Daqin takes end-to-end responsibility — went from 2.7% of the Datong centre's volume in 2024 to 23.3% in 2025.¹⁸ That is roughly an eightfold increase in one year. In August 2025 the centre also launched a freight-rate analytics platform integrating pithead coal prices, road haulage costs and port data, which the company said helped develop sixteen projects generating over RMB 190 million in revenue.¹⁸

Alongside this sits pricing reform. Daqin has moved parts of its book away from purely administered tariffs toward 阶梯运价 ladder pricing — where the rate per tonne varies with committed volume — and 市场化竞价 market-based bidding for certain capacity.¹⁸ For a business with no volume growth available, price flexibility is the only lever left that can move margin.

The uncomfortable arithmetic

Now the part that a company blog post would leave out.

The turnkey logistics business is currently a margin drag, and it is a material contributor to the 2025 profit collapse. This is not an inference — the company's own explanation of 2025 cost growth named increased end-to-end auxiliary logistics fees from the turnkey business as a driver.⁴

The mechanism is simple and predictable. When you sell line-haul, your revenue is the rail rate and your cost is running the train. When you sell door-to-door, your revenue is the whole chain and your cost includes buying trucking, terminal handling and shipping from third parties at market rates. Revenue per shipment goes up a lot. Gross margin percentage goes down, potentially a lot — and if you are buying those services badly, or buying them to win volume rather than to make money, gross profit in absolute terms can go down too.

That is what 2025 looked like: revenue up 4%, costs up 12%, margin down nearly seven points. A significant share of the incremental revenue was low-margin bought-in service, and the company was paying tuition to learn a business it had not been in.

How to judge it

Three tests separate a real platform transition from an expensive distraction, and Daqin has not yet passed any of them.

Test one: does the incremental revenue eventually carry incremental gross profit? As of the 2025 results and the first quarter of 2026 — revenue up 4.32% to RMB 18.57 billion, net profit down 7.26% to RMB 2.384 billion — the answer is still no.¹⁹ Revenue growth with profit decline is exactly the pattern of a business buying volume.

Test two: is there any structural reason Daqin should win at contract logistics? This is the harder question. Daqin's moat is the corridor. In turnkey logistics it is competing against freight forwarders and integrated logistics firms that have no track but also no legacy cost base, and whose entire institutional skill is procurement and coordination. A railway's advantage in the rail leg does not automatically extend to the trucking leg. The company has a real asset to build around — control of the scarce line-haul slot — but "strategic adjacency" is a hypothesis until the margin data confirms it.

Test three: is it big enough to matter? Not yet, and this deserves emphasis. Non-coal cargo volume grew 13% in 2025, and the company added 11.97 million tonnes of non-coal freight in 2024.¹¹⁰ Against 680.73 million tonnes of total freight, non-coal remains a small slice. The logistics pivot is currently large enough to hurt the P&L and not yet large enough to change the business.

So the fair verdict is this: the strategy is a genuine, on-strategy response to a genuine structural threat — not diworsification into an unrelated field, and not a hidden gem waiting to be discovered. It is a rational bet with an unproven payoff, being funded out of current earnings, at a moment when current earnings are already under pressure. Investors are paying for the option in real time and do not yet know its strike price.

Which makes the other half of management's response — the part where they give the money back instead of spending it — considerably more interesting.

---

↑ Back to Top

VII. Capital Allocation Playbook: Dividends, Buybacks, and Discipline (12 min)

Most companies facing structural volume decline do something predictable and usually destructive. They go shopping. They buy an adjacent business at a premium, call it diversification, and spend a decade explaining why the synergies are just around the corner.

Daqin has done close to the opposite, and the numbers are unusual enough to warrant careful examination.

The 2025 return

For 2025, Daqin distributed RMB 4.4262 billion in cash dividends — 75.02% of net profit attributable to shareholders.²⁰ That figure came in two instalments: an interim dividend of RMB 1.612 billion paid during the year, and a final dividend that brought the full-year total to roughly RMB 0.22 per share.²⁰

Now compare that to the promise. Under the company's three-year shareholder return plan covering 2023 to 2025, Daqin committed to distributing cash dividends of no less than 55% of annual net profit attributable to shareholders.²¹ Chairman 陆勇 Lu Yong characterised that commitment publicly not as an aspiration but as a binding order.²² In 2023 the actual payout was 58.08%; in 2024 it was 57.31%, split between an interim RMB 0.129 per share and a final RMB 0.14.²²

So in the year its profit fell by a third, Daqin raised its payout ratio by roughly eighteen percentage points above the prior year's level and twenty points above its own floor. That is not a company managing its dividend down to protect the balance sheet. It is a company deliberately increasing the proportion of a shrinking pie that goes to shareholders.

The buyback, and why the mechanics matter

Then there is the repurchase programme, first disclosed on August 29, 2025, with an execution window running from September 23, 2025 to September 22, 2026, a planned size of RMB 1.0 to 1.5 billion, and a price cap of RMB 8.11 per share.²³

Daqin executed at the top of the range. Total consideration came to RMB 1,499,824,499.26 — essentially the full RMB 1.5 billion authorisation. And critically, the shares were not parked in treasury for future issuance. On June 3, 2026, all 284,284,938 repurchased shares were cancelled, reducing registered capital and cutting the share count from 20,147,177,716 to 19,862,892,778.²⁴²⁵

That last detail is the one that separates a real buyback from a cosmetic one. A repurchase that holds shares in treasury for later re-issuance to employees or in a deal is a wash. A repurchase followed by cancellation permanently shrinks the claim base. And the effect showed up immediately in a small, elegant way: because the final 2025 dividend was declared as a fixed total pool rather than a fixed per-share amount, the per-share dividend had to be adjusted upward — from RMB 0.14 to RMB 0.14169 — precisely because there were fewer shares left to divide it among.²⁶

Combine dividends with buyback spending and Daqin returned 79.22% of 2025 attributable profit to shareholders.²⁷ Cumulative cash dividends since the 2006 listing have exceeded RMB 115.6 billion, across an unbroken run of nearly two decades.²⁷

And management is not finished. On July 22, 2026, the chairman formally proposed a second repurchase — RMB 400 to 500 million of self-owned funds, again with all repurchased shares to be cancelled, over a six-month window from shareholder approval.²⁸ Total buyback spending across the two rounds would exceed RMB 2 billion.²⁷

The context that complicates the story

Here is where an independent reading has to introduce some friction.

None of this happened in a policy vacuum. In 2024 Chinese regulators intensified a 市值管理 market-value-management push directed at listed state-owned enterprises, with a specific and mechanical trigger: companies whose share prices traded below net asset value for twelve consecutive months were expected to formulate and disclose 估值提升计划 valuation enhancement plans. Daqin met that criterion over the period from April 2024 to April 2025.²⁷ Its buyback programme and elevated payout are executed within, and disclosed as part of, exactly that framework.²⁹

This does not make the capital return fake. Cancelled shares are cancelled; distributed cash is distributed. But it changes the attribution. It is one thing for a management team to independently conclude that returning 80% of earnings is the highest-value use of capital. It is another for a state-controlled company to increase distributions in a year when the regulator has told companies in its exact situation to support their valuations. The second is still good for minority shareholders. It is simply less informative about management's independent judgment than it first appears.

The test that matters

The capital-discipline story deserves both credit and scrutiny.

The credit is genuine. Daqin has not bought a coal trader, a shipping line, a logistics roll-up, or a renewable energy developer to "reposition for the transition." It has not built a second railway nobody needs. The one equity stake on its balance sheet came with the carve-out rather than from a deal room. Against the base rate for capital-rich, growth-constrained companies globally — which is to destroy value in pursuit of a growth narrative — this is meaningfully better behaviour.

The scrutiny is equally warranted, and it takes the form of a single hard question: is this discipline, or is it the absence of alternatives?

Consider what Daqin actually could buy. It cannot build a competing corridor; there is no corridor to build. It cannot acquire another railway; railways in China are allocated within the state system, not traded. It cannot buy a coal miner; that is a different state pillar. The organic reinvestment opportunity is capped at maintaining an asset that is already at capacity. When you enumerate the options honestly, "return the cash" is not one virtuous choice among several — it is very nearly the only choice.

That does not make it wrong. Recognising that you have nowhere good to invest, and acting on that recognition rather than manufacturing a deal, is itself a form of discipline that many management teams fail. But an investor should be clear-eyed that they are observing a structural situation as much as a cultural one, and should not extrapolate this restraint into confidence that management would resist an empire-building opportunity if the state offered one.

The sustainability question is sharper still. A 75% payout on RMB 5.9 billion of profit is RMB 4.4 billion. A 75% payout on a profit that has fallen another third is RMB 2.9 billion. The payout ratio can be maintained indefinitely; the payout amount cannot outrun the earnings that fund it. And 2025 operating cash flow of RMB 4.99 billion was only modestly above the RMB 4.43 billion distributed — meaning that in a year like 2025, dividends plus buybacks consumed essentially all internally generated cash.³²⁰ That is not a crisis for a company with this balance sheet and this asset base, but it is not a repeatable posture through a sustained earnings decline either.

The right question to carry forward is not "will they pay 75% again next year." It is: when the 2026–2028 shareholder return plan is published, does the floor go up from 55%, or does it quietly stay put?

Which raises the obvious follow-up: who, exactly, decides?

---

↑ Back to Top

VIII. Who's Running the Railroad: Governance and Management Credibility (10 min)

On May 21, 2026, Daqin Railway's newly constituted eighth board of directors held its first meeting and elected Lu Yong as chairman and legal representative.³⁰ He had held the same role on the seventh board, whose term ran from May 19, 2023 to May 18, 2026, and his re-election came after the controlling shareholder nominated him, alongside 杨涛 Yang Tao, 张竑毅 Zhang Hongyi, 张利荣 Zhang Lirong and 张志辉 Zhang Zhihui, as non-independent director candidates.³¹

That sentence contains the entire governance story, if you read it carefully.

The dual-hat structure

Lu Yong, born in November 1967 and a senior engineer by training, is not primarily a listed-company chairman. He is concurrently party secretary and chairman of China Railway Taiyuan Group — the controlling shareholder — and chairman of the 大西铁路客运专线 Daxi Railway Passenger Line company.³⁰

Pause on what that means. The chairman of the listed company is the chairman of the entity that controls it. In a Western governance framework this would be flagged as an irreconcilable conflict: the person charged with representing all shareholders is institutionally the agent of one shareholder. In the Chinese central-SOE framework it is entirely normal, and disclosed, and nobody pretends otherwise.

The practical consequence is not that decisions are made badly. It is that certain decisions are not made at the listed company at all. Network investment priorities, the pace of freight pricing reform, whether rail assets held elsewhere in the China State Railway Group system ever get injected into this listed vehicle — these are system-level questions. The board of Daqin Railway Co., Ltd. implements them.

Day-to-day, the railway is run by general manager Zhang Hongyi, born in March 1968, who also serves as the listed company's party secretary and sat on the seventh board.³¹ The operating team around him includes chief accountant 裴丽群 Pei Liqun, deputy general manager and board secretary Zhang Lirong, and deputy general manager 胡静 Hu Jing.³²

Incentives, and what replaces them

Here is the analytical point that Western-trained investors most often get wrong about Chinese SOE subsidiaries.

There is no meaningful management equity ownership at Daqin. No large option grants, no founder stake, no personal wealth riding on the share price. Applying a standard alignment framework, you would conclude the incentives are broken.

But that framework is measuring the wrong thing. The accountability mechanism for these executives runs through the state system: SOE reform performance indicators, party discipline, career progression within the railway bureaucracy, and — increasingly relevant since 2024 — explicit market-value-management mandates that make share price performance a measured KPI for state-owned listed companies.

This produces a genuinely different behavioural profile, with both a good side and a bad side. The good side: you are unlikely to see the pathologies that equity-heavy compensation produces — earnings management to hit option strikes, aggressive leverage to juice returns per share, empire-building for the compensation-survey benefit of running a bigger company. The bad side: when system objectives and minority-shareholder objectives diverge, the system wins, every time, and there is no mechanism by which minority shareholders can make it otherwise. If national energy security requires hauling coal at an uneconomic tariff in a supply crunch, Daqin will haul the coal.

An investor in this company is, quite literally, a passenger. That is not a criticism; it is a specification.

The credibility test

The most useful place to test management is behaviour under stress, and 2025 provided plenty.

The company held its FY2025 and Q1 2026 results briefing on May 20, 2026, from 9:00 to 10:00 a.m., in online interactive format via the Shanghai Stock Exchange's Roadshow Center, with Zhang Hongyi, independent director 郝生跃 Hao Shengyue, Zhang Lirong, Pei Liqun, Hu Jing and board office director 丁毅 Ding Yi taking investor questions submitted in advance.³²

Judge the disclosure on its content. In explaining a 34.73% profit decline, management attributed the result to four factors: falling coal transport volumes, an adjusted transport structure, a logistics market still in a cultivation phase, and freight rate fluctuations.¹⁶ The disclosed cost detail was more specific: higher settlement payments driven by increased through-and-joint transport, higher auxiliary logistics fees at both ends arising from the turnkey logistics business, and rising labour costs.⁴

That is a mixed report card, and it is worth being precise about why.

In its favour: the cost breakdown explicitly names the logistics build-out as a driver. This is not a company hiding behind "challenging macro conditions." Two of the three named cost drivers are consequences of management's own strategic choices, and disclosing them that way is more candid than the norm. The company also continued a granular monthly volume disclosure practice — publishing Daqin line throughput monthly — which is a genuinely useful transparency habit that many operators would abandon during a decline.³³

Against it: the framing still leans toward describing what happened rather than owning what it cost. "Logistics market cultivation" is a soft phrase for "we spent money on a business that lost margin." There is no publicly disclosed quantification of how much of the 6.89-point gross margin decline is attributable to the logistics pivot versus coal weakness — and that is precisely the number an investor most needs, because one is a controllable investment with a payback period and the other is structural erosion. Nor has management published a target for when the logistics business turns margin-accretive.

On narrative consistency: the record is reasonably good. Management has not pivoted the story. The 400 million tonne line target has been stated consistently, and notably it was carried unchanged into 2026 after 2025 came in at 390.04 million tonnes — which is honest in one sense (no goalpost-moving) and unambitious in another (the same target, one year later, after a miss).¹ The dividend commitment made in the 2023–2025 plan was not merely honoured but exceeded in every year, which is the single strongest piece of evidence on this management team's follow-through.²¹²² And on the coal outlook, the company has been willing to state a directional view in public — telling investors in mid-2026 that coal volumes are secured for a period ahead, alongside first-half line volume growth of 7.96%.¹⁷

Net assessment: this is a competent operating team with a good record of doing what it said it would do on capital returns, working inside a governance structure that limits how much of the strategic agenda it actually owns. The credibility gap is not honesty. It is authority.

And authority, in this system, is also what has determined who Daqin competes with — and who it does not.

---

↑ Back to Top

IX. Competitive and Structural Landscape (10 min)

In December 2019, China opened a railway that was supposed to change the geography of its coal industry. The 浩吉铁路 Haoji Railway runs 1,813.5 kilometres from 浩勒报吉 Haolebaoji in Inner Mongolia south through Shaanxi, Shanxi, 河南 Henan, 湖北 Hubei and 湖南 Hunan to 吉安 Ji'an in 江西 Jiangxi — the longest heavy-haul railway ever built and opened in a single construction programme, with a design capacity of 200 million tonnes a year.[^34]

Its purpose was explicit: create a 北煤南运 north-coal-to-south corridor that would carry coal directly to central and southern China by land, bypassing the Shanxi–Qinhuangdao–coastal-shipping route that Daqin anchors.

It has not worked out that way. Haoji carried more than 100 million tonnes for the first time in 2024, and crossed 100 million tonnes again on December 15, 2025 — fourteen days earlier than the prior year, and its second consecutive year above that threshold.[^34]³⁴ Six years after opening, a line built for 200 million tonnes was running at roughly half its design capacity.

At the time it opened, Caixin published a detailed analysis arguing the project looked destined to struggle — a view that has aged well.³⁵

Why the state's own competitor underperformed

This is the most instructive competitive fact in the entire story, and it is worth unpacking, because it is empirical proof of a moat rather than an assertion of one.

Haoji had every advantage a new entrant could want. Unlimited state capital. Political mandate. Modern engineering with no legacy constraints. A genuine strategic rationale. And it still could not fill itself.

The reasons illuminate what Daqin actually owns. Haoji's route runs through terrain that makes it expensive to operate. Its southern destinations are regions with substantial hydropower and, increasingly, renewables, so the coal demand it was built to serve was already being competed for by non-coal generation. It lacked the dense collection network at the origin end that Daqin has spent thirty years building around Datong. And it had no equivalent of Qinhuangdao — a port whose coal handling capability and coastal shipping connectivity are themselves scarce assets, giving Daqin's route access not just to a coastal destination but to the entire coastal shipping network beyond it.

The lesson generalises: in bulk logistics, the line-haul is the visible asset and the ends are the real one. Origin collection density and destination terminal capacity are what make a corridor economic. You can build track with money. You cannot conjure thirty years of accumulated origin relationships and a purpose-built port position.

Shuohuang: partner, not rival

The other significant coal corridor out of Shanxi is Shuohuang, which runs to Huanghua port under CHN Energy's operation. On paper, it is Daqin's closest competitor: same origin basin, same coastal destination logic, comparable heavy-haul capability, cumulative haulage past five billion tonnes.¹¹

In practice, the 41.16% equity stake makes it something closer to an affiliate.⁹ Daqin participates in the economics of the route that would otherwise be its sharpest competitor, and the two together carry roughly a quarter of China's rail coal volume. It is a structure that would be unthinkable under Western antitrust law and is entirely unremarkable inside China's state rail system.

For investors, the practical implication is that Shuohuang's cash dividend is a partial hedge against volume loss on Daqin's own line: if coal shifts between the two corridors, Daqin captures a meaningful share of the economics either way. It is a real, if under-discussed, structural asset — and one that arrives as investment income, which makes reported operating margins a slightly incomplete picture of the company's true earning power.

Why trucks and other rail never took the share

The remaining competitive question is why alternative logistics never eroded Daqin's position materially.

The answer is distance economics, and it is brutally simple. Rail's cost advantage over road haulage grows with distance and with load density, and coal is the densest, most distance-tolerant, least time-sensitive cargo there is. Over 650 kilometres with a 20,000-tonne consist, the gap is not marginal — it is a different order of magnitude. Trucking competes for coal movements at the short-haul margin, feeding into rail rather than replacing it.

The result is a competitive position that should be understood in two layers, and this is the framing that matters most:

Within Chinese rail coal logistics, Daqin's position is close to unassailable. The state tried to build around it and produced a half-utilised railway. Roads cannot compete at this distance and density. Its closest peer is partly owned by it.

Within the Chinese energy system, Daqin's position is genuinely threatened. Every UHV line, every gigawatt of solar in the northwest, every year of coal's declining generation share erodes the total addressable volume that this near-perfect competitive position gets to compete for.

The competitive threat is not another railway. It is a wire. And you cannot out-operate a wire.

---

↑ Back to Top

X. Bull vs. Bear: Why This Wins From Here, Why It Might Not (12 min)

Every investment case eventually reduces to an argument about which force compounds faster. For Daqin the argument is unusually clean: a durable, cash-generative, physically irreplaceable asset returning ~80% of its earnings, versus a structural decline in the demand for what that asset does. Let's stress-test both sides properly.

The bull case, tested

The asset genuinely cannot be replicated, and this has been empirically demonstrated. This is not a moat inferred from margins. It is a moat proved by a failed state-funded attempt to build around it. That is a higher standard of evidence than most competitive-advantage claims ever meet.

Pricing reform is real optionality on a fixed-volume asset. For a business at throughput capacity, price is the only margin lever that does not require volume. Ladder pricing and market-based bidding introduce genuine flexibility into a historically administered tariff regime.¹⁸ If even a modest share of capacity migrates toward market-determined rates, the earnings effect on a fixed cost base is significant, and it requires no additional tonne to move.

The falsification test: pricing reform in a regulated system can go both ways. Market bidding in a soft coal market with competing corridors can mean lower realised rates, not higher ones. And 2025 is not encouraging on this point — the company cited freight rate fluctuations as a negative contributor to results.¹⁶ The optionality is real; its direction is not yet demonstrated.

Shareholder returns are institutionalised, not discretionary. 75.02% cash payout, 79.22% including buybacks, nearly RMB 2 billion of repurchases across two rounds with every share cancelled, and a documented commitment mechanism that has been exceeded every year of its life.²⁰²⁷²¹ Formalising it inside the valuation enhancement plan makes the return policy a disclosed, monitored obligation rather than a management preference.²⁹

The falsification test: a policy tied to a regulatory mandate can be untied when the mandate's trigger is no longer met. If the share price rises above net asset value, the specific regulatory pressure that produced the valuation enhancement plan relaxes. Would the payout ratio drift back toward the 55% floor? Nobody knows, and the answer will come with the next three-year plan.

Defensive characteristics inside a volatile market. A low-beta, high-payout, state-controlled infrastructure asset behaves more like a bond than an equity in a market not overpopulated with such instruments.[^37]

The falsification test: bond-like instruments are attractive because their coupon is stable. A "coupon" that is 75% of an earnings stream declining at double digits is not a coupon. Investors attracted by the yield profile need to be underwriting the earnings stream, not the ratio.

Asset injection optionality. There is standing speculation that further rail assets could eventually be injected into listed vehicles as sector reform advances. It is exactly that: speculation. Not announced, not confirmed, and appropriately treated as unpriced optionality rather than as part of the case.

The bear case, tested

The core commodity faces a directional, not cyclical, ceiling. This is the strongest bear argument and it does not depend on forecasting. Coal generation's share of Chinese electricity fell from 68.5% to 59.8% in five years; absolute coal generation declined in 2025 for the first time in a decade; wind and solar capacity overtook coal capacity; 46 UHV lines carry electrons where trains used to carry fuel.¹³¹⁴ None of that reverses.

Profit is falling faster than revenue is rising, and cash flow is falling faster still. Revenue +4.04%, costs +12.25%, gross margin −6.89 points, operating cash flow −46.47%, a fourth-quarter operating loss, and a first quarter of 2026 that repeated the pattern with revenue up 4.32% and profit down 7.26%.³⁴¹⁹ Two consecutive reporting periods of revenue growth with profit decline is a pattern, not a blip.

The regulated tariff caps the upside. The moat protects the asset without conferring pricing power. Whatever operational excellence Daqin achieves, a meaningful share of the benefit accrues to the system rather than to minority shareholders.

The strategic response is currently value-dilutive. The logistics pivot is the right idea and is being executed at a cost that exceeds its current return.

Governance sits with rotating state officials. The chairman's primary institutional accountability is to the parent he also chairs. That is disclosed, normal, and unchangeable.

The activist stress test

Suppose a skeptical minority-focused investor got a meeting. What would they actually push on?

Disclosure of the logistics drag. The single most important undisclosed number is the margin contribution of the turnkey logistics business, separated from coal. Without it, investors cannot distinguish a controllable investment cycle from structural erosion. An activist would demand segment-level reporting for the logistics business — and would note that the absence of that disclosure is itself informative.

The payback discipline on the pivot. What is the target date for logistics to turn margin-accretive, and what happens if it is missed? A strategy with no published payback horizon is a strategy with no accountability attached.

Buyback sizing versus the mandate. RMB 2 billion of repurchases sounds substantial until measured against the share count: 284 million shares cancelled out of 20.1 billion is roughly 1.4%. Against a company with this cash generation and this capital-investment profile, a determined activist would argue the buyback is small relative to what the balance sheet could support — and that the sizing looks calibrated to satisfy a regulatory expectation rather than to maximise per-share value.

Related-party economics. A listed company controlled by, and transacting extensively within, a state rail system has settlement arrangements — through-transport payments, access charges, shared services — whose pricing is not independently negotiated. In a year when settlement payments to other carriers were named as a profit-decline driver, that is a legitimate question about where value accrues within the group.⁴

The Shuohuang stake. A 41.16% holding in a comparable railway, generating billions in dividends, sits inside the company with no standalone valuation disclosure and no evident path to monetisation or consolidation. Non-core assets with unclear treatment attract activist attention everywhere.

The honest synthesis

The two cases do not actually contradict each other. Daqin is a superb asset attached to a shrinking market, run by people who cannot fix the market and appear to know it. The bull case is about the quality of the cash flow and the discipline of its distribution. The bear case is about the durability of the cash flow itself. Which one dominates depends almost entirely on the rate of coal's decline — and 2026's first-half volume recovery of 7.96% is a useful reminder that the rate is not smooth, not fast, and not predictable from a trend line.¹⁷

What is predictable is the shape of the problem. And that shape has some general lessons attached to it.

---

↑ Back to Top

XI. Playbook: Durable Business and Investing Lessons (8 min)

Strip away the Chinese specifics and Daqin is a case study in four things that recur across markets and decades.

A regulated monopoly caps the downside and the upside

The instinctive investor reaction to "monopoly" is to reach for a pricing-power multiple. Daqin is the corrective.

There are two species of infrastructure monopoly and they are not close relatives. The first — an unregulated or lightly regulated one, like a private toll road with contractual escalators or a dominant terminal operator in a permissive jurisdiction — converts structural position into pricing power, and compounds at inflation-plus for decades. The second is regulated or state-influenced, and converts structural position into permanence rather than profit. It will still be there in thirty years. It will not be earning a rent that grows faster than its cost base.

Both can be good investments. They require completely different valuation frameworks. The first is a compounder with an inflation hedge attached. The second is closer to a very long-duration bond with equity-like volume risk — and pricing it on a compounder framework is how investors get hurt.

The practical test to apply anywhere: who sets the price? If the answer is the market, you may have pricing power. If the answer is a regulator or a state system, you have durability and nothing more. Daqin's 2025 results — a company with an unassailable competitive position posting a quarterly operating loss — are what that distinction looks like when costs move and prices cannot.

When there is nowhere good to reinvest, give the money back

This is the most transferable lesson, and Daqin is a useful case precisely because the alternative was so available.

The standard corporate response to a maturing core business is to buy growth. It is available, it is narratively satisfying, and it is usually value-destroying, because a company with a declining core is a structurally disadvantaged acquirer — it is buying from a position of weakness, into markets where it has no advantage, often against buyers with lower costs of capital and better information.

Daqin has done the boring thing instead: paid out three-quarters of earnings and cancelled 1.4% of its shares. It is not glamorous. It also has not lost anyone any money, which is more than can be said for most transformation strategies.

But the lesson comes with a caveat that is easy to miss. Returning cash is the right call when reinvestment opportunities are genuinely poor — not when management has simply failed to find them. The distinction requires actually enumerating the alternatives. In Daqin's case the enumeration is short and honest: no new corridors to build, no railways to buy, core asset at capacity. The restraint is real, but it is at least partly structural, and an investor should hold it as a fact about the situation rather than as a permanent trait of the people.

Valuing a melting ice cube with a dividend

The hardest analytical problem here is that two variables are moving in opposite directions: the earnings base is eroding while the proportion distributed is rising. For a period, distributions can rise even as earnings fall — which flatters the near-term picture and tells you nothing about the terminal one.

The right framework is duration-based rather than multiple-based. The questions that matter are: how long does the asset generate meaningful cash, at what rate does that cash decline, and what fraction reaches shareholders along the way? A high payout on a declining base can still be a perfectly rational holding if the decline is slow and the distribution is high enough — that is essentially the arithmetic of a self-liquidating royalty. It becomes a trap when investors capitalise the current yield as though the base were stable.

The specific error to avoid: treating a payout ratio as though it were a dividend policy. A 75% ratio applied to a falling numerator is not a stable income stream. It is a fixed share of a shrinking thing.

Incentive analysis when the real boss is the state

Standard governance analysis asks whether management's incentives align with shareholders'. In a state-controlled company, that question is malformed, because management's incentives are aligned with a system whose objectives partially overlap with shareholders' and partially do not.

Where they overlap: operational efficiency, safety, throughput, and — since the 2024 market-value-management push — the share price itself, which is genuinely useful to minority holders.

Where they diverge: national energy security, employment stability in the coal regions, tariff policy serving industrial competitiveness, and the strategic allocation of assets within the state system.

The useful reframe is to stop asking "are incentives aligned?" and start asking "on which specific decisions do the interests overlap, and on which do they diverge?" For Daqin the overlap is unusually broad on operations and capital returns, and unusually narrow on pricing and strategic scope. An investor who understands that map can hold the position with clear eyes. One who imports a Western alignment framework wholesale will misjudge both the risks and the protections.

Which leaves the practical matter of what to actually watch.

---

↑ Back to Top

XII. Epilogue: What to Watch (5 min)

There is a monthly ritual around this company that tells you something about what it is. Every month, Daqin publishes the freight volume moved on the Daqin line.³³ Not quarterly, not with a lag, not buried in a filing — monthly, as a standalone disclosure, because for coal traders and power planners across northern China it is a physical-market data point before it is a financial one.

That habit is the right place to end, because it points directly at the metrics that matter.

Three KPIs, and only three.

One: Daqin line freight volume against the 400 million tonne target. This is the utilisation rate of the asset that produces the majority of the value. It came in at 390.04 million tonnes in 2025 and the 2026 target was set, unchanged, at 400 million tonnes.¹ Watch the monthly prints and the annual outturn against that number. Sustained readings materially below it mean the structural erosion is running ahead of the cyclical support; sustained readings at or above it mean the medium-term floor is holding. This is the volume side of the story in one number.

Two: gross margin, and specifically the relationship between revenue growth and cost growth. The 2025 gap — revenue +4.04% against costs +12.25% — is the entire margin problem in one comparison.³⁴ The question for every subsequent reporting period is whether that gap narrows. If revenue growth from non-coal freight and turnkey logistics eventually outpaces the cost of delivering it, the pivot is working. If the gap persists, the company is buying revenue, and investors should treat the top line as noise. This is the cleanest available proxy for whether the strategic response is creating or destroying value.

Three: total shareholder return as a share of profit, under the next three-year plan. 2025 delivered 75.02% in dividends and 79.22% including buybacks, against a 55% committed floor.²⁰²⁷²¹ The 2023–2025 plan has run its course. What replaces it — and whether the floor rises from 55% or stays there — is the most direct available evidence on whether the elevated payout reflects durable policy or a valuation-support exercise timed to a regulatory mandate.

Beyond the KPIs, two things to monitor without underwriting.

Freight pricing reform is the single biggest lever available for margin recovery without volume growth, because it is the only mechanism that improves economics on an asset that cannot carry more. Watch how much capacity actually migrates to ladder pricing and market-based bidding, and — critically — whether realised rates move up or down when it does.

And asset injection from China State Railway Group remains a genuine but entirely unconfirmed possibility as sector reform proceeds. It belongs on a watch list, not in a valuation.

The framing that survives all of this is fairly simple. Daqin Railway is not a growth story and has not been one for a long time. It is a test of a narrower and more interesting proposition: whether disciplined, aggressive capital return can outrun structural volume decline — whether giving the money back fast enough compensates for a shrinking base.

The line will still be there in thirty years. Roughly eighty-two heavy trains a day will very likely still be running on it, because there is no plausible world in which China stops needing to move bulk freight from the interior to the coast, even if the freight is not always coal. The open question is not whether the asset endures. It is how much of what the asset earns ends up in the hands of the people who own the shares — and how much is left to earn by the time that question is settled.

---

References

1. 大秦铁路：2025年煤炭发送量5.5亿吨 占全国铁路煤炭发送量的19.9% — 浙江省国资委, 2026-05-09 ↩↩↩↩↩↩↩↩↩↩↩↩

2. 大秦铁路：解码"西煤东运"653公里"突围路" — 中国网·铁路中国, 2025-02-13 ↩↩

3. Daqin Railway 2025 net profit down nearly 35%, targets 400 Mt transport for core line — Sxcoal, 2026 ↩↩↩↩↩↩↩↩↩

4. 大秦铁路2025年营收776.45亿元同比增4.04%，归母净利润59.00亿元同比降34.73% — 新浪财经, 2026-04-29 ↩↩↩↩↩↩↩↩

5. 百年瞬间：大秦铁路全线通车 — 共产党员网, 2021-11-22 ↩↩

6. 中国第一条双线电气化开行重载单元列车的运煤专线——大秦铁路正式通车 — 中国铁建 CRCC, 2018 ↩

7. 大秦铁路股份有限公司首次公开发行股票上市公告书 — 新浪财经, 2006-07-31 ↩

8. 大秦铁路股份有限公司2025年年度报告 — 上海证券交易所, 2026-04-30 ↩

9. 大秦铁路：目前公司持有朔黄铁路41.16%的股权，是其重要股东之一 — 新浪财经/证券日报, 2026-02-13 ↩↩↩

10. 大秦铁路：2024年营收746.27亿元 — 中证网, 2025-04-30 ↩↩

11. Shuohuang Railway Coal Haulage Tops 5 Billion Tons — CHN Energy, 2025-04 ↩↩

12. China's Daqin railway maintenance starts, but impact on thermal coal market limited — Sxcoal ↩

13. 电力行业发展向新向绿 — 国家能源局, 2026-06-12 ↩↩↩↩

14. 年用电量首破10万亿度 彰显中国经济"新"动能 — 新华网, 2026-01-19 ↩↩↩

15. Coal shipments on China's Daqin Railway down 7.09 percent in 2024 — SteelOrbis ↩

16. 大秦铁路2025年净利润下降超三成 预计2026年营收将微降 — 新浪财经, 2026-05-01 ↩↩↩

17. 大秦铁路：未来一段时间内公司煤炭运量是有保障的 — CCTD中国煤炭市场网, 2026-07-29 ↩↩↩

18. 货运突围看"大秦" — 人民财经网, 2026-02-11 ↩↩↩↩↩↩

19. 大秦铁路(601006.SH)：2026年一季报净利润为23.84亿元、同比下降7.26% — 东方财富网, 2026-04-30 ↩↩

20. 大秦铁路：2025年度现金分红总额为44.26亿元 — 证券之星, 2026-05-29 ↩↩↩↩↩

21. 大秦铁路股份有限公司三年（2023年-2025年）股东分红回报规划, 2024-04-27 ↩↩↩↩

22. 大秦铁路55%分红承诺落地 开中期分红先河 — 中国网·铁路中国, 2025-05-23 ↩↩↩

23. 大秦铁路股份有限公司关于以集中竞价交易方式回购公司股份的公告（公告编号：2026-011）, 2026-04-28 ↩

24. 大秦铁路：已于2026年6月完成15亿元回购并注销股份2.84亿股 — 证券之星, 2026-07-08 ↩

25. 大秦铁路：关于股份回购实施结果暨股份变动的公告, 2026-06-03 ↩

26. 大秦铁路股份有限公司关于调整2025年度利润分配方案每股分红金额的公告 — 上海证券报, 2026-06-24 ↩

27. 狂砸近20亿，大秦铁路拼了 — 新浪财经, 2026-07-27 ↩↩↩↩↩↩

28. 大秦铁路董事长提议4亿—5亿元自有资金回购股份 注销增厚股东权益 — 同花顺财经, 2026-07-22 ↩

29. 大秦铁路2025年度估值提升计划评估报告 — 证券之星, 2026-04-30 ↩↩

30. 大秦铁路（601006）披露选举公司法定代表人公告 — 搜狐, 2026-05-21 ↩↩

31. 大秦铁路股份有限公司第七届董事会第十八次会议决议公告（公告编号：2026-012）, 2026-04-30 ↩↩

32. 大秦铁路股份有限公司关于召开2025年年度暨2026年第一季度业绩说明会的公告 — 上海证券报, 2026-05-13 ↩↩

33. 大秦铁路4月运量同比增1.76%，环比降4.68% — 新浪财经, 2026-05-08 ↩↩

34. 浩吉铁路2025年货运量累计突破1亿吨 — 央广网, 2025-12-15 ↩

35. In Depth: Why China's Mammoth New Coal Railway Looks Destined to Struggle — Caixin Global, 2019-12-31 ↩