NARI Technology: Grid-Control Franchise, Quality Already Priced

One label fits NARI best: a high-quality policy-cycle compounder under mix pressure. Real capex, new products, and overseas expansion still drive its growth, so it is no mature cash cow. It entered this period healthy, which rules out a cyclical-reversal story. And the stock trades near its recent lows rather than at peak thematic exuberance, so a valuation bubble is off the table too. Classifying NARI is easy; the hard part is its next stage, which turns less on whether demand arrives than on whether management can protect economics while following that demand into new categories.

Company vertical history

NARI’s roots reach back well before the listed company existed. SASAC’s 2026 profile traces the name to “NARI,” the English abbreviation of the Nanjing Automation Research Institute, and notes that over more than fifty years the brand grew from a small power-automation institute into a large State Grid technology platform with assets above RMB 100 billion. That institutional origin shapes nearly everything that followed. NARI was born inside China’s power-system modernization effort, not as a private entrepreneur’s bid to crack an open market. Its first job was the control problem inside an expanding grid: dispatch, protection, automation, and the software-hardware interfaces that keep a large power system reliable. The company exists because power systems grow both more valuable and more fragile as they become more interconnected.

The listed vehicle was established on 2001-02-28 and went public in Shanghai in 2003. The IPO was modest by later standards: 40 million A-shares at CNY 10.39, raising gross proceeds of CNY 415.6 million and net proceeds of CNY 396.4 million, on a 20x issue P/E. The pitch to investors was simple, a specialist in electric-power automation sold during a period when China’s grid was still industrializing at speed. The original NARI equity story was grid automation as infrastructure modernization, not “energy transition,” “AI,” or “storage.”

From there, the company’s development can be divided into four stages.

The first stage was specialization and validation, when NARI set out to prove that a research-institute pedigree could become a repeatable business model. Its identity then sat in dispatch automation, relay protection, and other control-heavy offerings. Competition existed, but technical credibility, integration capability, and relationships with grid operators mattered far more than consumer-style substitution. NARI’s early edge was proximity to the operating logic of the Chinese grid itself, which handed it data, engineering feedback loops, and an installed base no newer entrant could copy quickly. The model was much narrower than today’s, yet the DNA was already there: sell high-value software-hardware systems into mission-critical grid nodes, then deepen the relationship through engineering, lifecycle service, and adjacent products.

The second stage brought scale without full market recognition. Through the late 2000s and early 2010s NARI was already an important domestic supplier, yet the listed entity still failed to capture the breadth of the group’s best assets. That gap mattered: investors saw a capable automation company, but not yet its most valuable version. A 2010 private placement, which later disclosure shows raised about CNY 760 million net, kept capital-market funding in play as a tool for broadening the platform. The bigger structural shift, though, was still waiting.

The third stage was the decisive one: the 2017 restructuring and the 2018 follow-through. The 2017 deal proposed buying roughly RMB 26.48 billion of assets through a mix of shares and cash, including major group power assets and stakes in entities such as NARI Relay and Puri UHV, alongside a sizeable supporting fundraise. Analysts at the time were right to call it close to “rebuilding another NARI” inside the listed shell. This was real surgery, not a cosmetic SOE reorganization: it pulled the listed company much closer to the group’s real industrial center of gravity, reduced intra-group overlap, and turned NARI from a narrower automation play into a broader listed platform for critical grid-control and power-electronics assets. Many of its later strengths in UHV DC, converter valves, and system integration trace back to this moment.

The fourth stage is the current one: expansion from the grid core into a wider control stack. By 2024 and 2025 the company was running four major business clusters: smart grid, digital-energy integration, low-carbon energy, and industrial interconnection. The core grid business stayed the anchor while the growth frontier pushed outward into renewable-grid integration, storage, wind-power control, digital infrastructure, grid software platforms, and overseas markets. That mix is why the 2025 results look both strong and slightly less clean than in earlier years. The old moat still works; it is just being asked to carry more businesses that do not all earn the same returns.

A few nodes still define the company today. The first is the 2017 asset injection, which changed both scope and market perception: without it NARI would be a good company, with it the default listed expression of State Grid’s control-layer technology base. The second is the 2025 direct-shareholder transfer, in which Guowang EPRI took direct control of 56.90% of the company, replacing NARI Group as the direct controlling shareholder while State Grid and SASAC stayed at the top of the control chain. The practical effect is continuity, not revolution, but it confirms that governance runs on SOE architecture rather than entrepreneurial ownership. The third is the present overseas and storage push: investor-relations records show management targeting overseas revenue above RMB 10 billion in 2026 across Southeast Asia, the Middle East, South America, Europe and Africa, with a focus on dispatch, substation, distribution, flexible AC transmission and storage. That is the most important directional shift in the story, because it is where faster growth now sits.

The company’s history reads less like a standard corporate chronology and more like the progressive listing of a national grid-control capability. Instead of reinventing itself from scratch, NARI kept moving the listed boundary outward, from dispatch and automation to a fuller grid-control platform and on to a broader energy-control and digital-power stack. So its capital-market narrative has always been part era-tailwind, part asset-packaging. Investors who ignore the institutional backdrop miss the point: NARI is a technology company, but it is also a state industrial platform whose value rises whenever China decides the grid’s control layer matters more strategically.

Financial vertical review

The top line has been steady enough to head off most arguments about business quality. From 2022 to 2025, verified revenue climbed from RMB 46.83 billion to RMB 57.42 billion to RMB 66.23 billion. Attributable net profit rose from RMB 6.45 billion in 2022 to RMB 7.18 billion in 2023, RMB 7.61 billion in 2024, and RMB 8.28 billion in 2025. This is no one-cycle wonder; it is a business that keeps finding fresh demand in the same broad operating system. The more revealing point is that revenue has lately outrun profit. Revenue rose 14.53% in 2025 against an 8.79% rise in attributable profit, and over the first nine months of 2025 the gap was wider still: revenue up 18.45%, profit up 8.43%. That is the cleanest numerical expression of the current mix issue.

Margins tell the same story. Brokerage summaries of the 2025 annual report put full-year gross margin at 25.8%, down 0.8 percentage point year on year, while the first nine months of 2025 showed gross margin down 2.79 percentage points. This is a real squeeze, not a collapse. Management’s own investor-relations answers on the 2025 half-year results were more specific: smart-grid and low-carbon revenue grew fast on new-generation dispatch deployment, UHV project delivery, and stronger storage and reactive-compensation work, while product mix, project timing, and some cost factors including raw materials weighed on margins. The reason is plain. NARI is expanding into categories that are strategically attractive but not always as profitable as the installed-base-heavy control businesses that built the franchise.

Cash conversion runs better than the income statement alone suggests. Verified operating cash flow was RMB 11.44 billion in 2023, RMB 11.09 billion in 2024, and RMB 12.77 billion in 2025, against attributable net profit of RMB 7.18 billion, RMB 7.61 billion, and RMB 8.28 billion respectively, an average of about 1.53 times net profit over the three years. Even the 2022 data in public summaries show operating cash flow of RMB 8.76 billion against attributable profit of RMB 6.45 billion. That matters for valuation: accounting profit here does not flatter the business. Working capital is heavy within the year, but cash collection over the annual cycle has stayed solid.

The quarterly pattern is worth a look because it exposes the seasonality and the operational rhythm. In 2024, fourth-quarter revenue was RMB 25.10 billion and fourth-quarter operating cash flow RMB 9.31 billion; in 2025, fourth-quarter revenue rose to RMB 27.65 billion and operating cash flow was RMB 8.05 billion. Both revenue and cash collection are back-end loaded, as they tend to be for project-driven industrial software-equipment businesses selling into utilities and large enterprises. That seasonality is part of why quarter-to-quarter optics get noisy, and why receivable and contract-liability trends should be read across full-year cycles rather than reacted to on a single quarter.

The balance sheet is sound, if not light. Total assets at year-end 2025 were RMB 98.59 billion, up from RMB 93.07 billion a year earlier, and net assets attributable to shareholders rose to RMB 52.65 billion. The 2026 first-quarter filing shows RMB 451.7 million of cash outflow for fixed assets, intangibles and other long-term assets in Q1, against RMB 337.9 million a year earlier; annualized, that points to fixed investment near RMB 1.8 billion, close to management’s 2026 target of RMB 1.821 billion. The business still needs steady R&D and factory-tooling support, but it is not the capital sink a commodity heavy-manufacturer would be.

Returns on capital stay among the stronger features of the case. Weighted ROE was 16.08% in 2023, 15.87% in 2024, and 16.26% in 2025. That steadiness matters: margin pressure has not yet eroded the structural advantage in NARI’s model. The company still out-earns most project-heavy industrials because much of the value sits in control architecture, software content, engineering know-how, and the high cost of failure for customers, not just in the hardware. Were that ROE range to slide toward the low teens for a few years, the bull case would weaken materially. So far it has held.

The main financial caution is receivables and customer concentration rather than leverage. The 2025 annual report says the top five customers made up 66.96% of annual sales, with related-party sales inside that group at 55.40% of annual sales. This is the business model, not a hidden flaw. Being wired into State Grid’s procurement system is both the great strength and the great governance and bargaining-power risk. Investors should rate the company strong on cash conversion and capital return, but only medium on customer diversification.

Price and valuation history

The stock’s capital-market history has run through several distinct regimes. The earliest was a specialist-automation identity, when the market valued NARI as a technically capable but fairly narrow industrial software-equipment supplier. The second opened with the 2017 restructuring and the later 2018 fund-raising, when investors had to reprice the listed entity as the main public carrier of a much broader group asset set. The third was the China energy-transition and UHV enthusiasm period, when dispatch, renewable integration, and power-electronics exposure all grew more valuable. The present regime is stingier: investors still acknowledge the company’s quality, but they no longer pay peak thematic multiples just because a policy slogan is running hot.

That shift shows up in the current quote. On 2026-06-26 the stock closed at CNY 22.74, near the low end of its 52-week range of CNY 21.08 to CNY 32.06. Reuters puts market capitalization around CNY 182.64 billion on 8.03 billion shares outstanding. This is the chart of a high-quality A-share industrial after investors have cut the theme premium and begun asking whether the next leg of growth pays as well as the last, not a euphoric policy bubble.

The market’s labels have shifted with it. NARI has been cast at various times as a growth industrial, as a quasi-defensive SOE compounder, and as the “safer” name in the power-equipment complex when more cyclical peers got too volatile. Today it trades around 22.1x trailing earnings on Google Finance and Reuters, a premium to more commoditized primary-equipment peers such as Pinggao or Xuji, a steep discount to what the market pays for Sieyuan, and well below the multiples on pure digital or export-momentum stories. That multiple says the market still respects the franchise but wants proof that margin dilution will not keep pace with revenue growth.

The valuation center slipped because the market now draws a sharper line between growth quality and growth quantity, not because the business worsened. A few years ago, “benefits from the new power system” was often enough. Now investors want to know whether the extra revenue comes from high-margin control nodes or from lower-margin categories like storage integration and more competitive outside-grid projects. NARI is plainly the stronger business than a plain cycle-exposed equipment maker. The open question is whether it deserves to rank as the highest-quality growth compounder in the space. The market currently says no, and that is a reasonable call.

Business model and moat

NARI’s revenue is easiest to read as a layered control stack. The four main clusters are smart grid, digital-energy integration, low-carbon energy, and industrial interconnection. Sell-side summaries of the 2025 filing put revenue at about RMB 33.4 billion for smart grid, RMB 12.3 billion for digital-energy integration, RMB 16.7 billion for low-carbon energy, and RMB 3.0 billion for industrial interconnection, with the first and third clusters driving most of the recent growth. An official company action document cuts the mix more simply, by customer domain: grid-industry revenue of RMB 42.04 billion in 2025, up 6.12%, against non-grid-industry revenue of RMB 24.11 billion, up 33.09%. That split is the key strategic fact. The grid core still supplies the base load of earnings; the faster growth now sits outside it.

Those segments do not earn the same economics. In 2025, brokerage summaries put smart-grid gross margin at 30.12%, digital-energy integration at 22.62%, low-carbon energy at 20.21%, and industrial interconnection at 21.56%. So NARI’s most profitable business is still the one closest to dispatch, protection, and core grid control, while the strategic growth businesses are broader and faster but thinner on margin. That is why top-line growth and margin pressure show up together. NARI is choosing to follow demand into categories where it has technical adjacency but less scarcity value, not being pushed there because the old categories are broken.

The cost structure carries some real operating leverage, but it is no infinite-software model. R&D stays essential. The 2025 action plan reports total R&D spending of RMB 4.739 billion, or 7.16% of revenue, high enough to matter and a sign that competitiveness is not self-sustaining. NARI has to keep spending to defend its control-layer position, qualify new products, push domestic substitution in components such as IGBTs, and stay relevant as the grid turns more digital and more power-electronics-heavy. Its profitability comes from high-value engineering and installed-base entrenchment, not from the absence of investment.

The first real moat is embedded system position inside the grid. Dispatch systems, protection logic, substation automation, and DC control do not get bought like ordinary industrial goods; they live inside the customer’s operating architecture. Replacement is possible, but switching costs run high because the cost of failure is enormous and interoperability with the installed base matters. NARI’s disclosures say its products and services reach more than 100 countries and regions, yet the deeper moat stays domestic: the company is rooted in the engineering standards, workflows, and project ecosystem of State Grid. That is hard for a new entrant to copy.

The second moat is systems integration across adjacent layers. A competitor can be excellent at one product; NARI’s edge is stitching dispatch, protection, digital systems, UHV control equipment, renewable-grid integration, storage control, and lifecycle service into one engineered package. Its own descriptions of the smart-grid and digital-energy businesses make the point: it sells platforms and operating logic, not just components. That counts for most as a power system grows more complex. As renewable penetration climbs, grid operators need tighter coordination between forecasting, dispatch, flexible control, protection, storage, and demand-side management. NARI wins when complexity rises faster than the customer’s appetite for multi-vendor integration risk.

The third moat is policy-standard proximity, the awkward SOE advantage investors sometimes understate because it sounds too political. In NARI’s case it is real. The controlling chain runs from SASAC to State Grid to Guowang EPRI to NARI. That closeness buys access to industry standards, pilots, procurement logic, and demonstration projects, even if it guarantees no profit. When China decides to push new-generation dispatch, grid-friendly renewables, UHV, storage coordination, or electricity-computing pilots, NARI is usually in the engineering conversation. This is a structural moat, not a marketing one. The cost is that the same system caps pricing power, because the main customer sits inside the same state ecosystem.

Management and governance read through that same lens. Chairman Zheng Zongqiang is a career NARI and Guowang EPRI executive; public profiles show he ran NARI as general manager before becoming chairman, while CFO Li Fang came up through the group’s finance system. This is technocratic SOE governance, not founder-led ownership with owner-operator incentives. The upside is continuity, institutional memory, and little appetite for reckless leverage. The downside is a governance discount tied to related-party intensity, state capital-allocation logic, and limited outside-shareholder influence over strategy. The company has still kept clean unqualified audit opinions and a strong dividend posture, and in 2025 it launched a RMB 0.5–1.0 billion buyback for future equity incentives, which is at least directionally shareholder-friendly.

Brand, network effects, and pure cost advantage rank as secondary moats here, not primary ones. Customer reputation matters, though not in a consumer sense; data matters, though not as a platform monopoly. NARI’s real advantage is the difficult, unglamorous mix of installed base, control-system trust, engineering depth, and institutional position. That moat is genuine, and narrower than some bulls imply. It guards the core grid-control franchise far better than it guards newer businesses like storage integration, where rivals such as Sungrow, Huawei, and BYD drag the market toward harder price competition.

Industry and cycle

NARI’s industry is the control, intelligence, and power-electronics layer of China’s grid investment cycle. It sits between upstream components and downstream utilities, EPCs, and large industrial users. Profit pools along that chain are uneven. Commodity primary equipment can throw off large revenue, but control-heavy categories such as dispatch, protection, and DC control usually earn better economics, carrying more software content, more qualification barriers, and higher failure costs. That is why NARI has long out-earned many peers even when its revenue was not the largest.

The growth driver is now a blend of policy cycle and capex cycle, not a short inventory snapback. The national action plan for 2024–2027 lays out nine special actions covering system stability, high-proportion renewable transmission, distribution-grid quality, smart dispatch, grid-friendly renewables, storage, charging and virtual power plants. State Grid’s record 2025 investment of RMB 665.7 billion, plus its stated expectation for more than RMB 4 trillion of fixed-asset investment over the Fifteenth Five-Year Plan, give that policy a funding base. That is why the NARI thesis holds over three to five years even when quarter-to-quarter margins wobble. The cycle driver is the physical need to redesign how the Chinese grid operates, not restocking.

The company also runs several cycle exposures at once: a capex cycle, since grid investment drives tenders; a policy cycle, since approvals, standards, and pilot programs decide what gets funded first; and a technology-iteration cycle, since more renewables, storage, and power-electronics force higher spending on control architecture. It rides the pure macro cycle less than a normal industrial exporter would, because utilities and state-backed grid companies do not cut spending the way private manufacturers do. NARI is a policy-funded infrastructure technology business, not a classic steel-or-machinery cyclical.

Downstream bargaining power, even so, stays high. State Grid is at once the best customer and the toughest. NARI wins large, technically demanding tenders because it is trusted and qualified, but it cannot assume private-sector pricing freedom, because the customer is big, informed, and often related through the state system. That is the structural trade. Upstream bargaining power is weaker, but it still counts: management and broker commentary both blame raw-material and component cost pressure for some of the margin softening in 2025. When the company bids system projects at fixed or semi-fixed pricing into a dominant buyer, small cost moves bite harder than they would in a business with stronger pass-through.

Regulation and geopolitics cut both ways. At home, tighter policy is mostly a tailwind: more rules around grid stability, renewable integration, storage, and digital coordination mean more demand for NARI’s products. Abroad, the picture is messier. Reuters reported in 2022 that Britain’s National Grid had started stripping out certain components supplied by a China-backed NARI unit over cybersecurity concerns. That episode left the domestic investment case intact, but it shows that overseas grid infrastructure can fall under national-security filters. NARI’s overseas push is real; it just will not get the frictionless operating environment abroad that it enjoys inside China’s state-grid ecosystem.

The newest structural demand leg is data-center power and electricity-computing coordination. The 2024–2027 action plan explicitly calls for projects that pair computing load with power infrastructure, and Reuters reported in June 2026 that China wants renewables to supply 80% of AI data-center power by 2030, with AI data centers potentially adding 300–500 billion kilowatt-hours of demand by then. NARI is a control company, not an AI model company, and it gains whenever power systems get more complicated, because AI and data centers need more stable, flexible and coordinated energy infrastructure. The theme is real, but investors should keep NARI’s place in the value chain straight: it sits in the power-control rails beneath the GPUs and server halls, not in the chips or the halls themselves.

Horizontal competitor analysis

NARI has direct comparables, though none match it perfectly. This is a “scenario C” industry: enough competitors to build a proper peer set, yet each one sitting in a different niche. Xuji Electric is the closest state-system comparator in control and power equipment. Sifang Automation is the purer secondary-equipment and protection-automation specialist. China XD Electric and Pinggao Electric are the better read on primary UHV and high-voltage equipment. Sieyuan Electric is the most important market-based comparator, because it shows what faster growth and international execution look like once a company is less boxed in by SOE structure. Together these names form the competitive map investors actually use.

NARI is the scale leader in the listed control-layer cohort. In 2025 it delivered RMB 66.23 billion of revenue and RMB 8.28 billion of attributable net profit. Xuji reported RMB 14.99 billion of revenue and RMB 1.17 billion of profit; Sifang, RMB 8.19 billion and RMB 0.83 billion; Pinggao, roughly RMB 12.52 billion and RMB 1.12 billion; China XD, RMB 23.76 billion and about RMB 1.27 billion. Sieyuan is the growth outlier, with 2025 revenue of RMB 21.54 billion and profit of RMB 3.15 billion, far smaller than NARI but much closer in profit intensity. That last comparison is the one the market watches most, because it shows how much more profit-dense a fast-moving private-sector peer can look.

Source basis: company disclosures for 2025 financials and Google Finance / Reuters for 2026-06-26 market data.

What sits behind those differences is more useful than the table itself. Xuji Electric is still a serious competitor, but the market values it more like a mid-tier grid-cycle beneficiary than a category-defining control franchise; revenue fell in 2025 even as margin improved, a sign that investors see part-cycle volatility rather than clean structural compounding. Sifang earns a premium multiple as a well-regarded specialist in secondary equipment and protection, with better margin than many peers, though it is much smaller and the stock already prices in substantial growth. Sieyuan gets the richest operational narrative, pairing domestic equipment strength with far stronger export and non-state-customer momentum, and investors pay up for that freedom. NARI sits between the security of the SOE core and the growth allure of the private exporter.

China XD Electric and Pinggao Electric tell the other half of the story. Both lean further toward primary UHV and high-voltage equipment, where the economics tend to look more cyclical and more tender-driven. China XD posted 2025 revenue of RMB 23.76 billion and profit of about RMB 1.27 billion, while Pinggao posted revenue of about RMB 12.52 billion and profit of about RMB 1.12 billion. Their market values, near RMB 82–83 billion for China XD and RMB 25 billion for Pinggao, read as lower-quality and more cycle-sensitive than NARI’s. That gap is justified: NARI’s profit engine leans less on a pure volume cycle, because software and control content weigh more heavily in its mix.

Customer reputation also differs sharply. Utilities buy NARI when the project is complex, the control layer matters, and a bad integration would be costly. They buy Sifang for specialist secondary-equipment strength. They buy Xuji when tender competitiveness and product fit line up, though Xuji’s market narrative has been less consistent. They buy Sieyuan when first-tier equipment quality meets faster delivery, export capability, and a more aggressive commercial posture. The key horizontal conclusion: NARI wins because it is the safest pair of hands in some of the grid’s most failure-intolerant corners, not because it is the cheapest or the fastest-growing. That is a powerful niche, and the reason its growth profile is steadier while its margin profile draws more scrutiny.

In ecosystem terms, NARI leads the grid’s control layer, filling the space where engineering complexity, systems integration, and policy alignment overlap. Its profit comes most directly from secondary equipment, dispatch software, protection, and high-end DC control markets. The profit pool most exposed over the next five years is the newer low-carbon and storage-control businesses, where competition from power-electronics champions and faster private-sector innovators runs stronger, not the legacy near-monopoly dispatch pocket. If the industry slides into a deeper price war, NARI’s position in the core gets relatively stronger even as its growth mix gets weaker. That asymmetry is central to the stock.

Current fundamentals and bull-bear divergence

The last four reported quarters show a business still growing, with growth just turning less profitable at the margin. In 2025, quarterly revenue ran RMB 8.90 billion, RMB 15.35 billion, RMB 14.33 billion, and RMB 27.65 billion, while attributable profit ran RMB 680 million, RMB 2.27 billion, RMB 1.90 billion, and RMB 3.42 billion. The 2026 first quarter then printed revenue of RMB 9.56 billion and attributable profit of RMB 721 million, up about 7.5% and 6.0% respectively on the year-earlier comparable. That is solid continuity, not reacceleration.

The composition of that growth matters more. In 2025, smart-grid revenue grew about 16% and low-carbon energy about 37%, while digital-energy integration was roughly flat and industrial interconnection grew only about 5%. Outside-grid revenue rose 33%, well ahead of the 6% growth in grid-industry revenue. Overseas revenue jumped 84.13% to RMB 6.04 billion, and management’s 2026 target is to clear RMB 10 billion overseas. The plan is clear enough: hold the policy-backed grid base, then layer faster non-grid and overseas categories on top. The open question is whether those faster categories can get better, not just bigger.

Right now the market is trading three things at once. The opening phase of the Fifteenth Five-Year Plan grid capex cycle. Overseas optionality, especially storage and power-control projects in places such as the Middle East, Southeast Asia and South America. And NARI as a beneficiary of the more complex power demand coming from AI infrastructure and electricity-computing pilots. All three rest on policy or management guidance, and none of them fixes the margin problem on its own.

The bull case rests on evidence, not slogans. Contract momentum: full-year 2025 new contracts reached RMB 75.87 billion, up 14.40%. Cash quality: operating cash flow of RMB 12.77 billion ran well ahead of net profit. A sticky grid base: grid-industry revenue still topped RMB 42 billion in 2025. Overseas acceleration: management has called overseas a “potential plate,” with 2026 revenue targeted above RMB 10 billion. Keep those lines moving together and today’s modest multiple could prove too low for a business still capable of high-single-digit to low-double-digit earnings growth.

The bear case is just as concrete. Margins already show mix dilution. Related-party concentration stays extremely high, so NARI can be operationally protected and economically squeezed at the same time. The strongest growth increasingly comes from more competitive categories, including storage control and broader low-carbon projects. And the stock is not cheap enough to shrug off execution risk: at a trailing P/E around 22x, with owner-earnings value closer to fair than cheap, there is no cushion for a couple of bad margin quarters. These are concrete worries, visible in the 2025 results and the segment economics.

So the central divergence today is simple: bulls think NARI is extending its moat into a broader energy-control stack, while bears think it is broadening faster than it can defend the economics. The stock no longer prices in blind optimism, yet it does not price in a proper margin of safety either, for the case where both sides are partly right: demand holds up, but profitability on the incremental revenue stays merely average.

Valuation analysis

Historical valuation

At CNY 22.74, the stock trades on about 22.1x trailing earnings and yields about 2.74% on Google Finance, while Reuters shows 8.03 billion shares outstanding and market value near CNY 182.6 billion. Those figures put NARI above the low-quality grid-equipment names but well below the peak multiples granted to faster private-sector growth stories. With the price sitting near the bottom of its one-year range, the market has already squeezed out some of the policy-theme premium. I read the current valuation as around the middle of NARI’s recent quality range, not at either extreme.

Peer valuation

Peer multiples back that reading. Xuji trades near 20x trailing earnings, Pinggao near 21.6x. China XD is around 58–62x, but still-low earnings on a cyclical rebound base flatter that figure; it signals no extra business quality. Sifang trades above 60x because investors are paying for specialist margin and growth, and Sieyuan above 40x because the market treats it as the best growth-and-export operator in the space. NARI at around 22x lands where it belongs: above the plain cycle names on quality, below the fastest compounders on growth glamour.

Absolute valuation

Cash-flow passthrough

The right starting point is owner earnings, not accounting earnings. Over the fully verified 2022–2025 period, operating cash flow ran about RMB 8.76 billion, RMB 11.44 billion, RMB 11.09 billion, and RMB 12.77 billion, against attributable net profit of RMB 6.45 billion, RMB 7.18 billion, RMB 7.61 billion, and RMB 8.28 billion. That works out to an average operating-cash-flow-to-net-income ratio of roughly 1.49x. The driver is collection, not aggressive accounting: NARI’s project business still pulls in cash reasonably well on an annual basis despite heavy within-year working-capital swings.

Maintenance-versus-growth capex is not disclosed explicitly, so the split below is an assumption rather than a reported fact. The 2026 first-quarter filing shows RMB 451.7 million of cash capex in Q1, which annualizes to around RMB 1.8 billion, and management’s 2026 target fixed-asset investment is RMB 1.821 billion. For valuation work I assume roughly 70% maintenance and 30% growth, reflecting a mature but still-evolving manufacturing and R&D base. On that basis, 2025 owner earnings come to roughly RMB 11.5–12.0 billion before the maintenance/growth distinction and about RMB 10.9–11.3 billion after subtracting full annual capex, or roughly CNY 1.36–1.41 per share. At the current price that is an owner-earnings multiple of roughly 16–17x, against a headline trailing P/E of 22x. The gap runs above 30%, so the scenarios below lean on owner earnings rather than reported EPS.

This scenario work is a research framework, not investment advice. The numbers anchor to current owner earnings, management’s 2026 investment plan, and broker forecasts for 2026–2027 profits around RMB 9.18–10.45 billion, but the scenario ranges are my own synthesis rather than company guidance.

Expectation gap

The market is pricing something modest, not heroic growth: NARI stays a high-quality national champion, 2026 earnings grow by roughly low double digits, overseas turns into a meaningful second engine, and the gross-margin damage from mix shift does not deepen much further. The expectation gap opens on one of two outcomes. Either NARI proves margins can stabilize even as low-carbon and overseas exposure rises, or it proves the reverse and investors decide the company is turning more project-heavy than platform-like. The next earnings prints matter less for raw revenue and more for segment margin, receivables, contract liabilities, and cash conversion.

Margin-of-safety recheck

At the current price, the margin of safety is thin. The stock sits above the value implied by the conservative scenario’s buy-range basis and only modestly below my base-case fair-value center. The most fragile assumption in the base case is margin stabilization in the low-carbon and outside-grid business. Cut that assumption to 70% strength, so owner earnings per share land nearer CNY 1.30–1.34 than CNY 1.38–1.45 and the multiple compresses toward 16.5x to 17x, and base-case value falls back toward roughly CNY 21–23. A mild disappointment would not be catastrophic, then, but it would wipe out most of the expected excess return.

If earnings stay flat for the next three years and the stock holds roughly the same multiple, total return would come mostly from dividends and modest buybacks, which today imply only a low-single-digit annual return. China’s 10-year government bond yield was about 1.73% on 2026-06-26, so flat earnings would still likely beat the bond, though by too thin a margin to call the current price a bargain. This is a good-company-bad-cheapness problem, not a broken business. My margin-of-safety verdict: not obvious.

Risk analysis

The first real permanent-capital risk is margin compression through product-mix drift. Probability is medium; impact is high. The transmission path is straightforward. NARI keeps winning faster-growing business in storage, low-carbon energy and outside-grid projects, but those categories do not earn smart-grid gross margins. Revenue keeps compounding while gross margin stalls in the mid-20s and net profit lags meaningfully. The market then stops valuing NARI as a control-platform compounder and starts pricing it as a broader, lower-quality project integrator, and investors feel it through slower EPS growth and multiple compression together. The indicator to watch is whether smart-grid margin holds near 30% while low-carbon and outside-grid margin keeps drifting lower, not total revenue.

The second risk is customer concentration inside the State Grid ecosystem. Probability is high; impact is medium to high. In 2025, the top five customers accounted for 66.96% of sales, with related-party sales inside that group at 55.40% of total sales. The concentration protects scale and trims credit risk in a narrow sense, but it weakens pricing power. NARI can be a strategic supplier and still be made to share the economics of system expansion with its controlling ecosystem. The telling indicator is whether related-party sales stay high while margins and collection terms worsen, since the concentration itself is structural.

The third risk is working-capital deterioration. Probability is medium; impact is medium. Annual cash generation has been good, but a business carrying tens of billions of receivables and heavy project seasonality can surprise investors if collections slip just as faster growth comes from grid-external customers. The transmission path runs through slower operating cash flow, higher net working capital, weaker owner earnings, and eventually a lower valuation multiple even while reported profit still looks healthy. The indicators to watch are receivables, contract assets, contract liabilities, and the ratio of operating cash flow to net income over the trailing four quarters.

The fourth risk is overseas execution and geopolitics. Probability is medium; impact is medium. Management wants overseas revenue above RMB 10 billion in 2026, targeting the Middle East, Southeast Asia, South America, Europe and Africa. It is a sensible growth vector, but it opens the company to currency risk, project-execution risk, local-partner risk, and security screening on sensitive infrastructure. The 2022 British National Grid episode is a live reminder that geopolitical filters can hit grid suppliers. The transmission path runs through lower-than-expected overseas revenue, weaker confidence in the “second curve,” and a smaller multiple premium for international optionality. The indicators to watch are overseas order conversion, disclosed regional spreads, and whether management keeps stressing risk controls such as export credit insurance and FX hedging.

The fifth risk is governance by absorption into the SOE machine. Probability is low to medium; impact is medium. The 2025 share transfer to Guowang EPRI left ultimate control unchanged, but it underlined that NARI is run as part of a state industrial platform. That usually means stability, yet it can also mean strategy shifts, asset moves, or capital-allocation calls made with system priorities in mind rather than minority-shareholder optimization. The transmission path would be subtle, working through lower returns on capital, more related-party complexity, or diluted strategic focus rather than fraud or leverage distress. The indicators to watch are related-party transaction intensity, acquisition logic, and whether dividend discipline weakens.

Catalysts and tracking indicators

Positive catalysts are fairly clear. The best would be a quarter or two where outside-grid and overseas revenue stay strong while gross margin stops falling, because that would settle the central doubt in the case. A second would be proof that overseas revenue above RMB 10 billion comes with healthy collections rather than by pushing risk outward. A third would be sustained acceleration in UHV approvals, distribution-grid digitalization, or electricity-computing pilots, since those categories play to NARI’s core control strengths rather than only its broader project capacity.

Negative catalysts are just as visible. A quarter with revenue growth still in the teens but another step-down in gross margin would keep the stock trapped. So would a rise in receivables without matching growth in operating cash flow or contract liabilities. A third would be signs that the overseas growth story is outrunning actual conversion, especially if NARI has to bid on thinner terms to win business. A fourth would be sharper security or geopolitical friction around overseas grid infrastructure.

These indicators matter because they separate the story from the accounting optics. The margin set shows whether the mix shift is under control; the cash set, whether reported profit still converts; the order and contract-liability set, whether NARI is winning business on terms that can still feed future revenue. The bond-yield and P/E lines matter because NARI is a quality industrial: when rates rise or style rotates, quality names can de-rate even when operations hold up.

Cross-synthesis summary

NARI’s journey proves one capability above all: it can turn state-backed technical relevance into durable listed-company economics. Many SOE technology stories fail because their advantages stay trapped inside the parent system, and NARI has largely escaped that trap. The listed company has become the public-market vessel for a real control-layer franchise in Chinese power infrastructure. Three reinforcing forces got it there. China’s grid kept growing more complicated. NARI kept occupying the parts of the grid where that complexity is hardest to manage. And the listed platform was widened step by step until it better matched the group’s real industrial core. That is why the company has posted years of rising revenue, mid-teen ROE, and operating cash flow consistently above net income.

Past success came from era tailwinds, yes, but also from technological and institutional fit, not luck. There was a period in China when selling into grid expansion was enough for a power-equipment company. NARI’s stronger record came from selling into the grid’s intelligence layer: dispatch, control, protection, stable operation, renewable integration. That is why the 2017 restructuring mattered so much: it moved more of those capabilities into the listed vehicle rather than creating them. Investors should resist two lazy readings, that NARI is merely a policy stock, or merely a software-like quality stock. It is both a policy beneficiary and a real engineering franchise, and neither view alone is enough.

Those success factors still hold today, though not unchanged. The core control moat remains, and so does the policy and capex tailwind; State Grid’s spending plans and the national new-power-system policy make that plain. What has changed is where the incremental growth comes from. The most dynamic demand is shifting toward low-carbon energy, storage, digital coordination, overseas projects, and the power demand created by AI infrastructure. NARI can play in all of it, but the economics there are more mixed than in legacy dispatch and protection. The company is stretching its moat, not losing it. That makes for a less comfortable phase, because stretching a moat can either build a larger platform or expose where the moat stops.

Horizontally, NARI’s advantage is being the preferred operator of the grid’s nervous system, not simple technological superiority in every category. Sifang may look cleaner in some secondary-equipment niches. Sieyuan may look faster in exports and richer in pure growth optics. Primary-equipment peers may give more torque to simple UHV capex headlines. NARI’s edge is that it sits where system complexity turns into operational risk, and utilities choose it because mistakes are expensive. That advantage is the hardest to copy and the easiest to underestimate. NARI’s weakness, though, is structural too, not temporary: the closer the customer is to the State Grid ecosystem, the better NARI’s access and the weaker its pricing autonomy. That tension will not vanish; it can only be managed.

That management challenge is what the valuation is really about. Today’s share price looks like a market compromise, neither rewarding pure past success nor lavishly pre-spending future success. Investors give NARI credit for being the safest large listed way to own Chinese grid intelligence, and they refuse to pay up until they see new growth turn into clean owner earnings. That stance is mostly right. The likeliest market misjudgment is about the speed of margin normalization, not the existence of demand. The market may be too impatient there, since some mix pressure is just the cost of entering adjacent categories before they mature. Some bulls are too relaxed about it, as if any revenue tagged “new power system” deserves a high-quality multiple. It does not.

Over the next year, the variable that counts is segment margin, especially smart grid versus low-carbon energy, plus whether operating cash flow keeps outrunning net income. Over three years, it is mix quality: can outside-grid and overseas revenue scale without dragging group returns down. Over five years, it is whether NARI becomes the default system integrator for a more software-defined grid spanning storage, virtual power plants, AI-linked electricity coordination and international power-control projects. If it does, today’s valuation will look undemanding. If it does not, NARI stays a good company that mostly treads water for shareholders while doing more work for average returns.

Two conditions would make NARI a better investment. The first is operational: gross margin stabilizes even as low-carbon and overseas business keeps growing, showing the moat is stretching successfully rather than thinning out. The second is valuation: the stock drops into a range that offers a genuine discount to conservative owner-earnings value. A sensible investor should also know what would break the thesis. If smart-grid margin slides sharply, if receivables outrun revenue for a sustained stretch, if overseas growth arrives with poor collections, or if ROE falls well below its long-run mid-teen range, the claim that NARI is still a high-quality policy-cycle compounder would have to be re-examined.

Bull and bear reasons

Bull reasons

• NARI remains the dominant listed proxy for China’s grid control layer, with 2025 grid-industry revenue above RMB 42 billion and strong positions in dispatch, protection and UHV-related control equipment.
• State Grid’s record 2025 investment and its stated expectation for more than RMB 4 trillion of fixed-asset investment in the Fifteenth Five-Year Plan support a multi-year demand runway.
• Cash earnings remain stronger than headline earnings, with 2025 operating cash flow of RMB 12.77 billion versus net profit of RMB 8.28 billion.
• Outside-grid and overseas businesses are now real, not hypothetical: non-grid revenue rose 33% in 2025 and overseas revenue rose 84.13% to RMB 6.04 billion.
• ROE stayed in the mid-teens through the mix shift, reaching 16.26% in 2025, which suggests the franchise still has structural earning power.

Bear reasons

• Revenue is outgrowing profit because lower-margin categories are taking a larger share of the mix; 2025 and 9M25 both showed clear margin pressure.
• Related-party and customer concentration are very high, with related-party sales inside the top five customers equal to 55.40% of total annual sales in 2025.
• The fastest-growing areas, especially low-carbon energy and storage-related business, are also where competition is tougher and moat protection is weaker.
• Overseas expansion offers growth, but also introduces execution, FX and geopolitical risks that the domestic core business largely avoids.
• At roughly 22x trailing earnings, the stock is not expensive in a bubble sense, but it is also not cheap enough to offer a clear margin of safety if margins disappoint.

Pre-mortem

The first plausible 50% down script is a pricing-and-mix failure in 2027–2028. State Grid keeps spending, but the mix shifts harder toward distribution upgrades, storage coordination and grid-external projects where more vendors can compete. NARI keeps winning contracts, yet consolidated gross margin slides from roughly 25.8% in 2025 toward 23%–24%, owner earnings stall around CNY 1.20–1.25 per share, and the market stops paying a quality multiple. Compress that multiple from roughly 20x forward owner earnings to 12x to 13x, and the share price could fall into the low teens without any revenue collapse. The business would still look busy. Shareholders would still lose a great deal of money.

The second plausible script is a failed second growth curve. Management pushes overseas revenue above RMB 10 billion and wins more Middle East and other international work, but collections lengthen, execution proves harder, and domestic margin pressure refuses to ease. Around the same time, the market decides the electricity-computing and storage themes ran ahead of actual monetization for NARI. Revenue still grows, but the story multiple evaporates. A stock priced as a dependable national champion gets re-rated as an ordinary project-heavy SOE industrial, and a fall from the low-20s to the low-teens is again realistic.

Final research conclusion

NARI is a real franchise. It sits at the control center of a grid that is growing more electrified, more digital, and harder to operate, which has secured it a durable role in one of the few Chinese industrial capex cycles with clear policy funding behind it. The stock deserves more respect than plain primary-equipment cycle names: the business is steadier in quality, more embedded in customer operations, and better at turning reported earnings into cash. The caution is about whether the newest demand looks enough like the old moat to deserve a premium multiple, not about whether demand exists.

At today’s price, the stock is ownable but not compelling. The current valuation already credits quality, but gives little yet for a cleaner second growth curve. That keeps the downside from looking reckless, while leaving little margin of safety if the revenue-profit gap stays stubborn. What would change my mind for the better is simple: a couple of reporting periods where low-carbon, outside-grid and overseas growth continue while gross margin and cash conversion hold steady. What would change it for the worse is just as simple: if the company keeps posting strong revenue on progressively weaker unit economics, the market will be right to pay less for every extra yuan of sales.

【Company-profile scores】

• Fundamental quality: high
• Growth: medium
• Moat: strong
• Financial soundness: strong
• Management credibility: medium
• Valuation attractiveness: medium
• Risk level: medium
• Suitable investor type: long-term growth

【Investment rating】

• Rating： Hold
• One-line thesis： A high-quality grid-control franchise with strong demand visibility, but current valuation leaves little room for further margin dilution.
• Three price signals：
  • Ideal buy price： see line below
  • Acceptable hold price： CNY 21–26 per share
  • Clearly overvalued price： CNY 31–34 per share
• Current-price classification: acceptable hold
• Whether to wait for a better price: yes. A buy becomes more attractive below CNY 18, provided smart-grid margin stays near 30%, owner earnings remain above accounting earnings, and receivable growth does not outrun revenue. The opportunity cost of waiting is missing a steady but not spectacular quality industrial.
• Target holding horizon： 3–5 years
• Expected annualized return: conservative about 0%–2%; base about 7%–10%; optimistic about 13%–16%
• Max-loss risk： about 45%–55% in the pre-mortem case where margin compression and derating happen together
• Reassessment-trigger signals:
  • consolidated gross margin below 25% for two consecutive quarters
  • smart-grid gross margin below 28% for two consecutive quarters
  • trailing twelve-month OCF / net income below 1.0x
  • receivables growth more than 10 percentage points above revenue growth for a full year
  • overseas growth slows materially while overseas margin or collections worsen

【Ideal Buy Price】16–18 CNY Basis: at least a 20% margin of safety below the value implied by the conservative owner-earnings scenario of roughly CNY 20–22 per share.

【Valuation Range】

• current: 22.74 (close as of 2026-06-26)
• bear (conservative · ideal buy zone): [16.0, 18.0]
• base (fair · acceptable hold zone): [21.0, 26.0]
• bull (optimistic · above the clearly-overvalued line): [31.0, 34.0]

Key data tables

That pattern captures the investment case: steady top-line growth, slower but still-healthy profit growth, and cash conversion that keeps beating accounting earnings. The margin issue is real, but it has not yet broken the underlying economics.

These quarterly figures show why the stock should not be judged on a single quarter’s cash flow. The business is highly seasonal, with working capital building early and cash collection catching up later in the year.

The peer table makes the market’s ranking clear. NARI is the large, trusted system franchise; Sieyuan, the premium growth operator; Sifang, the high-multiple specialist. Xuji, Pinggao and China XD sit closer to narrower product or cycle exposures.

Research uncertainties

• The exact maintenance-capex versus growth-capex split is not disclosed in public filings I could verify directly, so the owner-earnings split used in valuation is an explicit assumption built from Q1 capex cash flow and management’s 2026 fixed-asset investment target.
• Official public disclosure is clear on annual revenue, profit, cash flow and customer concentration, but less clear in directly searchable form on full-year segment-by-segment 2025 profitability; some segment figures above rely on broker summaries of the filed report rather than directly extracted company tables.
• The oft-cited market-share figures in dispatch automation, converter valves, DC control and storage control are directionally consistent across company materials and sell-side research, but current-year official quantified market-share disclosure is limited.
• Overseas backlog disclosure remains less granular than domestic activity disclosure, so the analysis of Middle East and broader international potential depends partly on management targets and third-party reporting rather than a fully segmented order book.

Sources

Primary sources used in this report include NARI’s 2025 annual report summary, 2026 first-quarter report, 2025 third-quarter report, investor-relations activity records, and the 2025 “quality and efficiency” action plan; SSE announcement pages; the company’s website; State Grid and SASAC materials; NDRC and National Energy Administration policy documents; and 2003 IPO listing materials.

Supplementary market and peer data came from Reuters, Google Finance, Trading Economics, ChinaBond, and public annual-report summaries or company-announcement reproductions for peer companies. Selected broker notes were used mainly to bridge segment economics, peer framing, and forecast ranges, and are identified as such in the body where relevant.

Other tickers mentioned

• 000400.SHE: Xuji Electric, used as the closest State-grid-system peer in power control and equipment
• 601126.SHG: Sifang Automation, used as the specialist secondary-equipment and protection comparator
• 601179.SHG: China XD Electric, used as a primary UHV equipment and cycle comparator
• 600312.SHG: Pinggao Electric, used as a high-voltage switchgear and UHV primary-equipment comparator
• 002028.SHE: Sieyuan Electric, used as the strongest private-sector growth and export comparator
• 600131.SHG: Guowang ICT, mentioned in the discussion of the State Grid listed-company ecosystem and governance context

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