Technology Moat Assessment: What the Technical Barriers Section Tells Hard-Tech Investors

The Hong Kong Stock Exchange’s (HKEX) Listing Rule amendments effective 1 January 2025, specifically the enhanced Chapter 18C for Specialist Technology Companies, have fundamentally shifted the burden of proof from revenue history to technological defensibility. In the first three quarters of 2025, HKEX received 14 listing applications under Chapter 18C, of which 11 were from hard-tech sectors including AI infrastructure, advanced materials, and quantum computing (HKEX, Q3 2025 Quarterly Review). The core question for sponsors and investors is no longer “Can they grow?” but “Can anyone else do what they do?”. This article dissects the Technical Barriers section of a hard-tech prospectus — the single most scrutinised disclosure in a Chapter 18C filing — and provides a framework for assessing whether a claimed technology moat is genuine, defensible, or merely aspirational.

The Anatomy of a Technical Barrier under Chapter 18C

HKEX Listing Rule 18C.05(2) requires a Specialist Technology Company to demonstrate a “high potential for growth” and, critically, “a proprietary technology or intellectual property that creates a significant barrier to entry”. The Technical Barriers section of the prospectus is the primary vehicle for this demonstration. It is not a marketing document; it is a legal representation that the sponsor must verify under the SFC’s Code of Conduct for Persons Licensed by or Registered with the Securities and Futures Commission (paragraph 17.6, due diligence requirements for listing applicants).

The Three-Pillar Framework Mandated by the Exchange

The HKEX guidance letter GL117-23 (December 2023) explicitly outlines three pillars that a Technical Barriers section must address: (1) the nature of the proprietary technology, (2) the basis for the barrier (e.g., patents, trade secrets, network effects, regulatory approvals), and (3) the duration and sustainability of the barrier. A prospectus that fails to provide granular, verifiable evidence on all three pillars will face substantive follow-up questions from the Listing Division. For example, in the 2024 filing of a semiconductor design house, the Exchange required the company to disclose not just the number of granted patents but also the patent family coverage by jurisdiction (USPTO, CNIPA, EPO) and the average remaining patent life. The company’s initial disclosure listed 47 patents; after HKEX review, the final prospectus disclosed that only 23 were granted, with an average remaining life of 6.2 years, and 14 were pending applications. This reduction from 47 to 23 materially changed the barrier assessment.

Distinguishing Hard Barriers from Soft Advantages

A common pitfall in prospectuses is conflating a “competitive advantage” with a “technical barrier”. A competitive advantage — such as a strong brand, superior customer service, or first-mover status — does not meet the Chapter 18C threshold. The SFC’s thematic review of IPO sponsors in 2024 (published January 2025) found that 8 of 12 hard-tech prospectuses reviewed contained language that blurred this distinction, citing “brand recognition” or “customer loyalty” as technical barriers. The SFC explicitly warned that such language must be removed or reclassified as non-technical factors. A genuine technical barrier must be rooted in something that a competitor cannot replicate within a reasonable timeframe — typically 3-5 years for a hard-tech company — without incurring prohibitive R&D costs, infringing on IP, or overcoming regulatory hurdles. For instance, a proprietary algorithm for drug discovery that requires 5,000+ GPU-hours of training data and a specific chip architecture is a hard barrier. A “unique customer relationship” is not.

Decoding the IP Portfolio: Patents, Trade Secrets, and the “Patent Thicket”

The most common technical barrier cited in hard-tech prospectuses is intellectual property, specifically patents. However, the quality, breadth, and enforceability of the patent portfolio are what matter, not the raw count. The Technical Barriers section must be read with a forensic eye on three specific dimensions.

Patent Quality Metrics That Matter

A patent count without context is noise. The HKEX Listing Division, in its review of a 2024 AI chip applicant, requested a breakdown of patents by: (a) granted versus pending, (b) jurisdiction (US, CN, EP, JP, KR), (c) technology classification (CPC or IPC codes), and (d) forward citations. The company had claimed 312 patents. After HKEX scrutiny, the final prospectus disclosed that only 89 were granted, 47 were in the US, and the median forward citation count was 2.3 — below the industry median of 4.1 for comparable AI chip companies (based on USPTO data cited in the prospectus). The sponsor’s due diligence report, filed with the SFC, noted that the patent portfolio’s “technological breadth” was concentrated in a single sub-class (G06N 3/02, neural networks), leaving the company exposed to competitors in adjacent sub-classes. For investors, the key metric is the patent family’s technological coverage ratio — the number of distinct CPC sub-classes covered divided by the total number of patents. A ratio below 0.2 suggests a narrow moat.

The Trade Secret Problem

Many hard-tech companies, particularly in advanced manufacturing and materials science, rely on trade secrets rather than patents. This creates a fundamental tension with the disclosure requirements of a Hong Kong listing. Under HKEX Listing Rule 2.13, an issuer must provide “sufficient information” to enable a reasonable investor to make an informed assessment. Trade secrets, by definition, cannot be fully disclosed. The SFC’s Guidance Note on Due Diligence for IP (June 2023) acknowledges this tension and requires the sponsor to: (1) obtain an independent expert report verifying the existence and value of the trade secrets, (2) confirm that the company has robust internal controls to protect the secrets (e.g., NDAs, access logs, physical security), and (3) assess whether the trade secrets can be reverse-engineered. In the 2025 prospectus of a quantum computing startup, the sponsor’s expert report concluded that the company’s key trade secret — a specific qubit stabilisation algorithm — could be reverse-engineered by a well-funded competitor within 18 months using publicly available research. The company subsequently withdrew its application. This case underscores that a trade secret is only a barrier if the cost of reverse-engineering exceeds the expected return for a competitor.

The “Patent Thicket” as a Defensive Moat

A more sophisticated barrier is the “patent thicket” — a dense, overlapping web of patents that makes it prohibitively expensive for a competitor to enter a technology space without infringing. The Technical Barriers section should explicitly map the patent thicket, showing not just the company’s patents but the third-party patents that the company has licensed or cross-licensed. For a hard-tech company in the semiconductor lithography space, for example, a patent thicket might involve 200+ patents across 15 companies, with the applicant holding 30 core patents and having cross-licensing agreements with 5 key players. The barrier is not the 30 patents but the network of agreements that prevent a new entrant from accessing the ecosystem. The HKEX Listing Division has increasingly required companies to disclose the terms of material cross-licensing agreements, including duration, royalty rates, and termination clauses. A cross-licensing agreement with a 3-year notice period is a weaker barrier than one with a 10-year term.

The R&D Pipeline as a Forward-Looking Barrier

A static IP portfolio is insufficient for a Chapter 18C listing. The Exchange requires evidence that the technical barrier will persist and deepen over time. This is where the R&D pipeline disclosure becomes critical.

The “Technology Roadmap” Requirement

HKEX Listing Rule 18C.06(3) requires a Specialist Technology Company to disclose its “technology roadmap” for the next 3-5 years, including specific milestones, budget allocations, and key personnel. The Technical Barriers section must link this roadmap to the sustainability of the barrier. For example, a company developing solid-state batteries must show that its next-generation product — expected in Year 3 — will have a 30% higher energy density than current competitors, and that this improvement is protected by a new patent family filed in Year 1. The HKEX Listing Division, in its review of a 2025 battery technology applicant, requested a Gantt chart showing the timeline for each milestone and the specific patents or trade secrets associated with each. The company’s initial roadmap was qualitative (“we expect to improve energy density”); the final prospectus included a quantitative table with 12 milestones, 8 of which were linked to specific patent applications filed with the CNIPA.

R&D Spend Efficiency as a Proxy

The prospectus should disclose not just total R&D expenditure but also R&D spend efficiency — the ratio of R&D spend to new patent filings or to technology milestones achieved. The HKEX guidance letter GL117-23 suggests that a company with an R&D spend efficiency ratio above HKD 50 million per new patent family should provide a justification. In practice, for a 2024 AI chip applicant, the sponsor calculated a ratio of HKD 42 million per patent family, which was within the acceptable range. However, the company’s R&D spend as a percentage of revenue was 78%, which raised concerns about sustainability. The sponsor’s due diligence report noted that the company was burning cash to file patents, not to develop commercially viable products. For investors, the R&D spend efficiency ratio must be compared against industry benchmarks. For semiconductor companies listed on the Main Board in 2023-2025, the median ratio was HKD 28 million per patent family (HKEX, 2025 Annual Review of Specialist Technology Companies).

Key Personnel and the “Brain Drain” Risk

The Technical Barriers section must also address the risk of key personnel departure. Under HKEX Listing Rule 18C.09, a Specialist Technology Company must identify its “core technical personnel” and disclose their employment agreements, including non-compete clauses and termination provisions. The barrier is only as strong as the team that maintains it. In a 2025 prospectus for a biotech company developing gene-editing therapies, the core technical personnel were three scientists with 20+ years of experience each. Their non-compete clauses were limited to 12 months, and they had no golden handcuff provisions (e.g., restricted stock units with 5-year cliffs). The sponsor’s risk factor section explicitly warned that the departure of any one of these three could materially impair the company’s technical barrier. The HKEX Listing Division required the company to disclose a succession plan, which the company did — a plan that named two internal candidates but did not specify their qualifications. This was deemed insufficient, and the company had to amend the prospectus to include detailed CVs and a timeline for knowledge transfer.

The Sponsor’s Role: Verification and the “Reasonable Investor” Test

The Technical Barriers section is not a standalone disclosure; it is the product of a sponsor’s due diligence, which must satisfy the SFC’s “reasonable investor” standard. The sponsor must verify every claim in the section, and the SFC’s enforcement actions in 2024-2025 have made clear that a failure to do so can result in disciplinary action.

The Independent Expert Report

For hard-tech companies, the sponsor is required to engage an independent technical expert to opine on the validity and defensibility of the claimed technical barriers. The SFC’s Code of Conduct (paragraph 17.6(b)) requires the expert to be “independent of the applicant and the sponsor” and to have “relevant technical expertise”. In practice, this means a professor from a university or a consultant from a specialised firm, not a generalist. The expert’s report must be filed with the Exchange and is publicly available in the listing documents. In the 2025 prospectus of a robotics company, the expert report concluded that the company’s claimed “proprietary sensor fusion algorithm” was actually a standard implementation of a published IEEE paper, with no novel elements. The sponsor withdrew the application. This case illustrates that the expert report is not a rubber stamp; it is a substantive check on the company’s claims.

The “Red Flag” Checklist for Investors

When reading the Technical Barriers section, investors should apply a five-point red flag checklist derived from the SFC’s 2024 thematic review findings:

1. Vague Language: If the section uses terms like “proprietary,” “unique,” or “leading” without quantitative backing, it is a red flag. The SFC review found that 67% of prospectuses with such language had to be amended.
2. Patent Count Inflation: If the prospectus cites a total patent count without breaking it down by granted/pending and jurisdiction, assume the number is inflated. The median “inflation rate” in the 2024 review was 40% — i.e., 40% of claimed patents were pending or abandoned.
3. Missing Cross-Licensing Disclosure: If the company operates in a patent-dense field (e.g., semiconductors, telecoms, biotech) and does not disclose cross-licensing agreements, it likely has a weak IP position.
4. Short Non-Compete Clauses: If core technical personnel have non-compete clauses shorter than 24 months, the barrier is vulnerable to a brain drain.
5. No Technology Roadmap: If the prospectus does not include a quantified technology roadmap with milestones and budget, the company has not demonstrated that the barrier is sustainable.

Actionable Takeaways for Hard-Tech Investors

1. Prioritise patent quality over quantity: Focus on the ratio of granted patents to total claims, the average remaining patent life, and the technological coverage ratio (CPC sub-classes per patent) — a ratio below 0.2 indicates a narrow moat that is easily circumvented.
2. Verify trade secrets through the sponsor’s expert report: If the Technical Barriers section relies heavily on trade secrets, demand to see the independent expert’s reverse-engineering assessment; a cost-to-replicate estimate below HKD 50 million is a weak barrier.
3. Cross-reference the R&D pipeline with the IP portfolio: Every milestone in the technology roadmap should be linked to a specific patent application or trade secret; unlinked milestones represent execution risk, not a technical barrier.
4. Assess the sponsor’s due diligence rigour: Check whether the sponsor has filed an independent expert report and whether the report contains any qualifications or limitations; a clean report with no caveats is rare and warrants scrutiny.
5. Model the barrier’s decay rate: Using the disclosed patent life, non-compete duration, and R&D spend efficiency, estimate how the barrier will erode over 5 years; a decay rate above 20% per year suggests the moat is shrinking faster than it is being reinforced.