Financial Centres in a Two-Shock World

International financial centres (IFCs) are being tested by two forces pulling in opposite directions. The war that began in the Middle East on 28 February 2026 has produced the largest energy supply disruption on record and re-priced geopolitical risk across every asset class. At the same time, an AI-driven investment boom is channelling record capital into technology infrastructure, listings and private markets. The IMF's July 2026 outlook captures the split: global growth of roughly 3.0% in 2026 rising to 3.4% in 2027, with headline inflation rising to about 4.7% this year as energy and food prices bite. Energy importers and conflict-exposed economies are losing; economies embedded in the technology value chain are gaining.

For financial centres, five conclusions follow.

First, the basis of competition has shifted from cost to certainty. The GFCI 39 survey found that market participants now rank predictability of regulation above every other regulatory attribute, followed by speed of response, flexibility and quality. Tax rates and office costs still matter, but they no longer decide the outcome. Ratings fell across almost all 120 centres in the March 2026 index — an average decline of 1.82% — which reflects a general loss of confidence in the global environment rather than failures by individual jurisdictions.

Second, the top tier has compressed and the Gulf has arrived. Only a single rating point separates New York, London, Hong Kong and Singapore. Dubai and Tokyo entered the top ten, displacing Chicago and Los Angeles. Concentration at the top and rapid churn below is the pattern to plan against: winning a place in the top twenty is realistic for a well-run centre; displacing the top four is not.

Third, physical and operational resilience is now a competitiveness factor, not a back-office concern. Drone strikes forced evacuations and remote-working arrangements at the DIFC; Iranian drone strikes hit cloud data centre facilities in the UAE and Bahrain; roughly a hundred cyberattacks on Gulf institutions were recorded in the first 72 hours of the war. Centres that can demonstrate continuity of trading, custody, settlement and payments under duress will attract capital away from those that cannot.

Fourth, capital has not fled the Gulf — it has become more selective. ADGM reported 57% growth in assets under management in Q1 2026 with active licences passing 13,353, and DIFC recorded roughly 1,506 new company registrations in the first half of 2026, up around 39% year on year. But Saudi Arabia has entered a phase of fiscal discipline, with PIF's 2026–2030 strategy cutting capital spending by about 15% and re-orienting toward returns, and construction awards across the Kingdom falling below $30bn in 2025 from $71bn in 2024. The era of open-ended state-funded expansion is over; the era of returns-tested allocation has begun.

Fifth, the product frontier is digital money, tokenised assets and AI-linked finance — but the regulatory map is fragmented. Hong Kong has an operative stablecoin licensing regime with its first licences issued in April 2026, the UAE's federal payment-token rules are in force, MiCA applies in the EU, and the US, UK and Singapore regimes sit at different stages of effectiveness. The commercial opportunity is real; the cross-border operating burden is heavy.

Sixth, the startup and venture strand is real but frequently overstated in centre strategies. MENA startups raised about $1.7bn across 242 rounds in H1 2026, an 18% fall on the prior year, while Asian startups took $42.8bn in Q2 2026 alone — the strongest quarter in more than three years — with AI absorbing over 60% of it. Both markets are concentrating hard: fewer deals, larger cheques, a narrow band of favoured companies. For financial centres, venture activity contributes talent density, product supply and narrative, but it is not yet a material source of assets, listings or regulated revenue in the Gulf, and the missing growth-equity layer is the region's most consequential structural gap. Section 6 examines this in detail.

Seventh, dual listing works in Asia for a reason that does not exist in the Gulf. Hong Kong recorded 24 A+H listings in H1 2026 — more than all of 2025 — contributing over half of its HK$209.9bn in IPO proceeds. That works because dual listing bridges a closed mainland capital pool to international investors. Gulf markets have no such wall to cross, which is why a decade of cross-listing agreements has produced very few transactions. Section 7 sets out where the strategy does and does not pay.

The strategic implication for any centre — established or emerging — is that the winning proposition in 2026 is institutional credibility plus demonstrable resilience plus a narrow, well-chosen set of specialisations, rather than a generalist offer built on tax and real estate.

2.1 The war shock

Hostilities beginning on 28 February 2026 disrupted shipping through the Strait of Hormuz, through which roughly a fifth of global oil and gas moved beforehand. Transit volumes collapsed: ship-tracking data showed between eight and fifteen vessels crossing per day in early August 2026, against roughly 130 before the conflict. A US–Iran ceasefire in April and a memorandum of understanding in June both failed to restore normal passage, and by early August the US had reimposed a naval blockade following renewed attacks on shipping. Brent traded around the mid-$80s in August, some 16% above pre-war levels, with the IMF assuming an average near $89 per barrel for 2026 and energy prices running roughly 25% above pre-conflict levels.

The financial consequences have been direct. UAE equity markets lost around $120bn in value in the first month, with Dubai's index down about 16% and Abu Dhabi's about 9%. Tens of thousands of flights were cancelled, damaging the Gulf's connectivity proposition. Goldman Sachs warned that Gulf economies could contract by 2–5% in 2026, with Qatar and Kuwait most exposed through Hormuz dependence. The IMF forecasts Middle East and North Africa growth falling from 3.7% in 2025 to about 0.7% in 2026, before a projected rebound to around 6.5% in 2027 — a V-shape that depends entirely on de-escalation.

2.2 The technology boom

Running against this is an AI-driven capital cycle. Economies at the centre of AI hardware supply chains — Taiwan, South Korea, Thailand, Malaysia — have outperformed. Hong Kong raised roughly HK$110bn in IPO proceeds in Q1 2026, its strongest quarter in five years and first globally, with full-year forecasts of HK$350bn and AI, biotech and semiconductor issuers driving the pipeline. Capital markets have found a theme powerful enough to offset war-related risk aversion, at least for issuers with technology exposure.

2.3 What this means for centres

Three structural effects deserve planning attention:

• Bifurcation of flows. Capital is simultaneously de-risking geographically and re-risking thematically. The same allocator reducing Gulf physical exposure may be increasing AI infrastructure exposure booked through a Gulf entity.
• Risk premia are now location-specific in a way they were not in 2019. War-risk insurance, business continuity costs, and staff hardship premiums are line items that change effective cost of doing business by centre.
• Neutrality has commercial value. Centres perceived as insulated from great-power and regional conflict — Singapore, Tokyo, Zurich, and to a degree Riyadh's inland geography — gain a measurable premium.

Best practice for financial centres has traditionally been described in terms of the GFCI's five pillars: business environment, human capital, infrastructure, financial sector development and reputation. That framework still holds, but the weighting has changed. Eight principles now separate the centres that are compounding from those that are stalling.

1. Regulatory predictability over regulatory generosity. The most important single finding from the GFCI 39 survey is that participants value predictability first. A slightly stricter rule that is stable and clearly interpreted beats a lighter rule that may change. Practical implication: publish multi-year regulatory roadmaps, consult properly, grandfather changes, and publish decision timelines with performance against them.

2. Speed as a published service standard. Authorisation turnaround, in-principle approval to licence, fund registration and visa issuance should be measured and disclosed. ADGM's Q1 2026 disclosure of 22 in-principle approvals and 29 new financial services permissions — a 45% year-on-year increase — is as much a marketing document as an operational report, and works because it is quantified.

3. Legal certainty through independent, common-law-based dispute resolution. The DIFC and ADGM courts, and the arbitration infrastructure around them, remain the single most-cited reason international firms choose Gulf centres over onshore alternatives. Emerging centres that skip this step do not achieve escape velocity.

4. Anchor institutions before real estate. Centres succeed when a critical mass of anchor allocators, exchanges, market infrastructure and law/audit firms is present. Abu Dhabi's draw is proximity to ADIA, Mubadala and ADQ; Hong Kong's is access to mainland issuers and capital. Buildings follow institutions, not the other way round. Master plans that lead with floorspace targets tend to underperform.

5. Cluster depth over breadth. A centre with genuine global depth in three activities beats one with shallow presence in twelve. Candidate specialisations: private credit and alternatives; family offices and succession structuring; reinsurance and specialty risk; Islamic finance; aircraft, ship and equipment leasing; commodity trading and trade finance; digital asset market infrastructure; and global in-house centres.

6. Talent mobility as an economic instrument. Visa regimes, portability of licences and qualifications, schooling and spousal work rights determine whether senior professionals relocate. This is the constraint most likely to bind in the Gulf: retaining and attracting senior expatriate executives in a conflict-adjacent environment is now the primary limiting factor on regional headquarters strategies.

7. Domestic capital mobilisation. Centres increasingly cannot rely on foreign flows alone. Singapore's Equities Market Review is instructive: MAS committed a multi-billion-dollar programme placing capital with fund managers investing in Singapore equities, moved to a more disclosure-based IPO regime, and required new Global Investor Programme family office applicants to deploy their minimum investment into Singapore-listed equities. Domestic pension, insurance and sovereign capital can be directed toward local market liquidity without heavy-handed mandates.

8. Measurement discipline. The centres that improve publish comparable metrics on a fixed cadence — licences, regulated entities, AUM, funds domiciled, headcount, revenues and profits. DIFC's 2025 disclosure of 8,844 active registered companies, 1,052 regulated firms, revenues of $581m and net profit of $402m sets a standard: the centre operator is itself accountable as a business.

4.1 Where the money is going

Asset and wealth management is the fastest-migrating pool. ADGM's Q1 2026 AUM growth of 57%, with fund managers rising 24% to 179 and funds up 43% to 263, was driven by entrants including Man Group, Barings, Bain Capital and Capital Group; managers establishing there during the quarter represented over $4.4tn in global AUM. DIFC's wealth and asset management cluster grew about 35% year on year to 592 firms by mid-2026. Singapore retains roughly 2,000-plus single family offices. This is the most contestable pool of activity in the market, because it is people-light, capital-heavy and relatively easy to redomicile.

Public equity capital formation has concentrated in Hong Kong. Hong Kong topped global IPO rankings in 2025 and again in Q1 2026, with fundraising exceeding HK$140bn by late April and average daily turnover above HK$280bn. Listing reform — confidential filings, relaxed rules for specialist technology and biotech, changes to pricing and allocation, and consultation on weighted voting rights and secondary listings — has been the driver. National Stock Exchange of India and Shanghai have also ranked among global top-five venues, evidence that the listing market has genuinely regionalised.

Sovereign capital has become returns-disciplined. PIF's 2026–2030 strategy, approved in April 2026, restructures the portfolio into Vision, Strategic and Financial portfolios, targets six domestic ecosystems, and emphasises efficiency, governance and private sector participation alongside roughly 15% lower capital spending. PIF's share of Saudi construction awards fell from about 38% to 14%. Saudi Arabia projects a 2026 deficit of roughly $44bn, and bank lending growth in the Kingdom halved in the first five months of 2026 as project spending slowed. Capital-light technology and AI infrastructure investment has displaced megaproject construction as the priority.

Private markets and credit continue to absorb allocation. The combination of higher-for-longer rates, constrained bank lending in Saudi Arabia and elsewhere, and large sponsor demand supports continued growth in private credit, infrastructure and secondaries — activities that suit IFC structuring regimes and fund platforms.

Real assets and resilience infrastructure are a growing category. Pipeline bypass capacity, tanker fleets, desalination, grid hardening, data centre redundancy and logistics corridors are absorbing capital as a direct consequence of the war. ADNOC's purchase of five VLCCs for about $590m and Saudi Arabia's pivot to land-based export routes are early examples of a broader resilience capex cycle that centres can finance.

4.2 Implications for centre strategy

Digital money and tokenisation. Hong Kong's Stablecoin Ordinance regime is operative, with the HKMA granting its first two licences in April 2026 to HSBC and to Anchorpoint Financial, the Standard Chartered-led joint venture, from a field of around 36 applicants. HSBC's tokenised deposit service now operates across multiple jurisdictions and currencies, with UAE expansion planned. The UAE's federal payment token rules are in force alongside separate regimes at VARA, SCA, FSRA and DFSA. The strategic point for centres is that the value is not in issuing a token but in owning the settlement rail — the interbank layer, custody, and the bank-grade compliance stack around it. Fragmentation across jurisdictions is the principal drag: definitions of what a stablecoin is, who may issue and how reserves are held differ materially, and tokenised deposits sit in an ambiguous space between bank liability and novel instrument.

Tokenised real-world assets and trade finance. Trade documentation, receivables and commodity flows are the natural first application in the Gulf–Asia corridor, where physical trade volume is large and financing gaps are chronic. Realistically, the next phase is layered rather than substitutive: correspondent banking, instant payment systems, stablecoins, tokenised deposits and CBDCs operating side by side.

AI in and around finance. Two distinct opportunities: financing AI infrastructure (data centres, power, chips — where sovereign funds are actively deploying), and deploying AI inside financial firms for surveillance, KYC, credit and research. Centres can differentiate through regulatory sandboxes with genuine supervisory engagement, model risk guidance, and data regimes that permit cross-border training and inference within defined guardrails.

Private wealth and succession. Family businesses and foundations have been among the fastest-growing DIFC segments. This is a durable, high-margin cluster if supported by credible law, professional depth and a stable residency offer.

Islamic finance, leasing and specialist niches. Sukuk issuance, aircraft and ship leasing, global in-house centres and reinsurance remain under-served relative to demand and suit centres with focused regulatory build-out — GIFT City's development around banking, leasing, fund management and global in-house centres is the clearest example in Asia.

Resilience and specialty insurance. War risk, marine, aviation, political risk and cyber cover are repricing sharply. A centre that assembles genuine underwriting capacity in these lines will capture premium flow that is currently leaving the region.

Every financial centre master plan of the last decade has included an innovation district, an accelerator and a fintech licensing regime. It is worth asking directly what this activity is actually contributing, because the honest answer is more nuanced than the marketing.

6.1 The venture cycle has decoupled from the capital markets cycle

Middle East. MENA startups raised roughly $1.7bn across 242 rounds in H1 2026 on Wamda's count — an 18% decline in capital and a 28% fall in deal volume against H1 2025. MAGNiTT's stricter equity-only count puts it at about $1.35bn across 214 deals, down 22% and the weakest first half since at least 2022. But the aggregate hides a sharp divergence:

• The UAE consolidated. UAE startups took about $1.2bn across 83 deals, roughly 70% of all regional capital and a 125% increase year on year. Q2 alone brought $591m across 37 deals.
• Saudi Arabia contracted severely. Saudi startups raised about $259m across 80 deals, an 81% fall. Fintech took 68% of that ($176m across 13 companies). Critically, every deal that closed in the Kingdom during the half was early stage — no later-stage round closed at all.
• Capital became domestic. Regional investors supplied an estimated 81% of all MENA venture funding, up from 58% a year earlier and the highest share in more than five years, as the international investor share of capital more than halved.
• Concentration intensified. The ten largest transactions accounted for 58% of all funding in the half, with two mega-rounds worth a combined $480m cushioning the headline.
• The founder base remains narrow. Male-founded companies captured around 95% of capital, with female-founded companies taking roughly 0.14%. This is a talent-supply problem before it is anything else.

Asia. The direction is opposite and the magnitude far larger. Investors deployed $42.8bn across Asian startups in Q2 2026, the highest quarterly total in over three years, following $27.4bn in Q1. AI-focused companies took more than 60% of Q2 capital — just over $26bn — while deal counts fell to a multiyear low. China captured roughly 60% of regional funding in Q1; India followed with $3.8bn, its strongest quarter in a year. Southeast Asia raised about $12.8bn across 178 equity rounds in the first seven months of 2026 against $5.42bn across 255 rounds in the same period of 2025 — average round size rising from roughly $21m to $72m. Southeast Asian native AI companies alone drew $4.1bn, though a single $2.8bn Series D accounted for about 68% of it.

The pattern in both regions is the same even where the direction differs: fewer companies clearing a higher bar, with larger cheques. Headline funding growth in Asia and headline decline in MENA both overstate what is happening to the median company.

6.2 How startups actually affect a financial centre

Startup activity reaches a centre's economics through five channels. Only two of them appear in the metrics centres usually publish.

Channel three is where the genuine strategic value sits. The tokenisation rails, compliance stacks and payment infrastructure that centres want to sell as differentiators are largely built by venture-backed firms. A centre that hosts those builders captures the capability, not merely the tenancy.

Channel four explains Hong Kong. The AI, biotech and semiconductor listings driving record IPO proceeds in 2026 are the output of a venture cycle that ran a decade earlier. The Gulf has no equivalent pipeline yet: exits are overwhelmingly trade sales, and the venture-to-listing path barely exists.

6.3 An honest assessment

What startups are delivering. Real fintech and RegTech capability; a credible innovation narrative that supports GFCI FinTech sub-index performance, where Hong Kong ranks first and Shenzhen second, followed by New York, Singapore and London; talent that would not otherwise relocate; and infrastructure that incumbents buy. The DIFC ecosystem's fintech firms have collectively raised more than $3.3bn to date, and its innovation community now numbers well over a thousand firms — a genuine cluster by any international standard.

What they are not delivering. Material AUM, listings or regulated revenue in the Gulf. To put scale in perspective: total MENA venture funding in H1 2026 was roughly $1.7bn, while individual asset managers arriving at ADGM in a single quarter represented over $4.4tn in global AUM. Venture is a rounding error against the capital flowing through these centres. Strategies that treat startup counts as a proxy for financial centre success are measuring the wrong thing.

The structural gap that matters. Saudi Arabia's complete absence of later-stage rounds in H1 2026 is the single most important signal in the regional data. An ecosystem that can fund seed and Series A but not Series B and C exports its best companies to jurisdictions that can — typically the UAE, Singapore or the US — along with the eventual listing, the senior jobs and the tax base. Building an accelerator is cheap; building a growth-equity market is not, and it is the part that determines whether an innovation strategy compounds.

The dependency risk. Domestic investors supplying 81% of regional venture capital reads as resilience, and in the short term it is. Structurally, it couples startup funding to the sovereign fiscal cycle at precisely the moment PIF is reducing capital spending by around 15% and Saudi construction awards have collapsed. When the state is the market, a fiscal tightening becomes a venture winter with a lag.

6.4 Innovation-specific risks

• Sovereign-linked funding cycle. Venture supply correlates with state budgets rather than diversifying away from them.
• Exit drought and mark-downs. Without IPO or secondary routes, portfolio valuations become stale and eventually require write-downs on sovereign and corporate venture books.
• Flag-planting without substance. Subsidised licences and free desks generate registrations that produce no revenue, no employment and no capability, while inflating the metrics leadership reports.
• Talent cost inflation. Startups and incumbents compete for the same scarce engineers, raising the cost base of the whole centre.
• AI valuation concentration. With over 60% of Asian venture capital flowing to AI, a repricing in that sector would hit the listing pipelines of Hong Kong, Shanghai, Shenzhen and Singapore simultaneously.
• Regulatory perimeter drift. Fintechs scale into regulated activity faster than supervisory capacity grows — the same capacity problem identified in Section 3, appearing first at the innovation edge.

6.5 What good practice looks like

1. Build the missing growth-equity layer. A fund-of-funds anchoring Series B and C managers, on matched commercial terms, with a requirement that investment teams are physically based in the centre. This is the highest-leverage intervention available to Gulf centres today.
2. Create a credible exit route. A growth segment on the local exchange with proportionate listing rules, plus support for secondaries and continuation vehicles. Without an exit, the ecosystem is a cost centre.
3. Use procurement as the growth instrument. Require or strongly incentivise regulated incumbents and state entities to pilot and purchase from licensed startups. Revenue converts an accelerator cohort into a company faster than any grant.
4. Publish a sandbox-to-licence graduation record. Proportionate capital requirements and a transparent count of how many firms actually graduated to full authorisation, and how many are still operating five years later.
5. Replace vanity metrics. Stop reporting startups hosted. Report follow-on capital raised, revenue generated in-centre, jobs created, licence graduations, and five-year survival.
6. Broaden the founder base. A funding market allocating 0.14% of capital to female-founded companies is not primarily an equity problem — it is an unexploited supply of founders in a region where talent is the binding constraint.

Dual listing appears in almost every regional capital markets strategy in the Gulf, and has been the single largest driver of Hong Kong's revival. The two regions offer an unusually clean natural experiment, and the lesson is not the one usually drawn.

7.1 Asia has made dual listing work at scale

Hong Kong raised HK$209.9bn across 85 IPOs in H1 2026 — a 92% increase in funds raised and a 102% increase in deal count against H1 2025, its strongest first half in five years. The composition matters more than the total: the city recorded 24 A+H listings and 13 specialist technology IPOs in the half, both already exceeding their full-year 2025 totals, and together accounting for more than 70% of funds raised. On PwC's count, companies already listed on mainland exchanges contributed HK$121.7bn — more than half of all H1 proceeds. There were 19 A+H listings in the whole of 2025; there were 24 in six months of 2026.

The mechanism is specific. Loss-making AI, robotics and semiconductor companies list as A-shares on a mainland market largely closed to foreign investors, while simultaneously listing H-shares in Hong Kong, which is fully open to them. Approval for the Hong Kong leg can run as fast as around 30 working days. The flow is now starting to reverse, with the CSRC signalling support for Hong Kong-listed mainland companies pursuing A-share dual listings, and applications and pre-IPO counselling already underway. With over 500 applications in the pipeline, PwC has raised its full-year forecast to roughly HK$380bn.

The transferable insight is that A+H is not a listing product — it is a capital account workaround. It exists because there is a wall between a large domestic capital pool and international investors, and dual listing is the door through it. That is why it generates real incremental demand rather than simply splitting an existing order book.

7.2 The Gulf has the frameworks but not the flow

The regional record is thin relative to a decade of announcements:

• Americana Restaurants (December 2022) remains the landmark: a concurrent IPO across the Saudi Exchange and ADX raising SAR 6.8bn (about $1.8bn) at a market capitalisation above $6bn, with roughly SAR 2.3bn traded across both venues in the opening period.
• GFH Financial Group is the region's most multi-listed name, trading across Bahrain Bourse, Boursa Kuwait and DFM with approvals sought for Tadawul and ADX.
• Volume remains marginal. As of early 2024, ADX had only five dual-listed companies, all from the GCC, representing about 6% of listed companies and 7% of market capitalisation.
• Memoranda have outpaced transactions. Cross-listing agreements between Tadawul, ADX, Edaa and Bahrain Bourse date back to 2018 and earlier, with repeated public expectations of a first cross-listing that took years to materialise. Aluminium Bahrain and Batelco both explored Tadawul listings that did not complete. Aramco's international listing has been under discussion since 2016, courted openly by HKEX, and has still not happened.

Where cross-listing has worked in the region is in funds rather than equities. Hong Kong's CSOP Saudi Arabia ETF grew to roughly HK$10bn; Saudi Arabia's CMA approved the region's first ETFs tracking Hong Kong-listed equities; and in December 2025 ADX became the first Middle East exchange to cross-list US securities, bringing NYSE-listed KraneShares products onto the exchange with AED settlement during Gulf trading hours and no need for offshore accounts.

7.3 Why Gulf equity dual listings underperform

Six constraints, in rough order of importance:

1. No additional investor pool. GCC investors overlap heavily across GCC exchanges. A second regional line reaches largely the same money, so it divides an order book rather than expanding it.
2. No index gain. Global index treatment follows the primary listing. A secondary GCC line usually attracts no incremental passive demand — which is the main economic reason issuers list anywhere.
3. The structural condition is absent. Saudi Arabia and the UAE have no capital account wall to bypass: foreign investors already access these markets directly through the QFI regime and MSCI Emerging Markets inclusion. The arbitrage that powers A+H simply does not exist in the Gulf.
4. Operational friction exceeds deal value. Depository links exist, but share fungibility, FX, custody and settlement alignment remain heavy relative to the size of most candidate issuers.
5. Liquidity fragmentation risk. Issuers rightly fear wider spreads on two thin lines rather than one adequate one.
6. Ownership and share class limits complicate foreign and cross-border participation.

7.4 Where dual listing genuinely makes sense here

Cross-regional rather than intra-regional. A Gulf issuer listing in Hong Kong, Singapore or London reaches capital it cannot otherwise access; a Gulf issuer listing in a neighbouring Gulf market usually does not. The Gulf–Asia corridor is the live opportunity: Asian investors seeking energy, petrochemical and infrastructure exposure, and Gulf capital seeking Asian technology exposure.

Funds and debt before equities. ETF and sukuk cross-listings are already working, cost far less, and build the depository, FX and market-making plumbing that any future equity flow will need. This is the correct sequencing and it is being under-used.

As an exit route for the venture ecosystem. Linking back to Section 6, a domestic growth-market listing paired with a technology venue abroad is one credible answer to the region's missing exit path — provided the foreign leg is chosen for its investor base rather than its prestige.

India's IFSC route as a comparator. GIFT City's direct-listing framework allows Indian companies to reach international capital from a domestic venue — solving the same problem as A+H by a different route, and worth close study by Gulf regulators.

7.5 What centre operators should do

1. Answer the additive-pool question before promoting dual listing at all. If the second venue does not bring different money, the strategy destroys liquidity rather than creating it.
2. Fix fungibility rather than marketing. Straight-through transfer between depositories, settlement cycle alignment, FX handling and market-maker obligations on both lines.
3. Engage index providers explicitly on the treatment of secondary lines, since index eligibility is what converts a listing into demand.
4. Lead with ETFs and sukuk, where the region already has proof of concept, and use them to build the rails.
5. Target Asia, not the neighbours.
6. Publish honest post-listing data: secondary line turnover as a share of total, spreads on both lines, and whether the second line is still meaningfully traded two years on. Almost no exchange publishes this, which is why the strategy persists without evidence.

The wider risk to note is on the other side: Hong Kong's current strength depends heavily on a pipeline whose supply is determined by mainland policy rather than by Hong Kong. A change in CSRC posture on offshore listings would remove the largest single driver of its 2026 numbers.

The concentration to watch is correlated risk: a Gulf centre may simultaneously face physical disruption, cyber intrusion, energy-driven inflation, sovereign fiscal tightening and talent attrition in the same quarter. Risk frameworks built around single-event stress tests understate this.

Dubai (DIFC). Entered the GFCI top ten in March 2026 and has continued to grow through the conflict, with roughly 1,506 new registrations in H1 2026 and regulated firms up around 16% to 1,134. Its strengths are cluster depth, legal infrastructure and an unmatched regional talent pool. Its vulnerabilities are physical exposure, aviation dependence, and real estate concentration. Priority: convert demonstrated crisis performance into a formal resilience proposition — published continuity standards, redundancy commitments and independently verified recovery capability.

Abu Dhabi (ADGM). The fastest-compounding centre in the region by AUM and licences, with proximity to very large sovereign pools as its structural advantage and an explicit ambition to reach the global top five. Priority: supervisory capacity must keep pace with licence growth, and the centre should deepen in two or three activities rather than pursuing full-spectrum breadth.

Saudi Arabia (Riyadh / KAFD). The strategy has shifted from spectacle to pragmatism. The regional headquarters programme and KAFD build-out continue, but the binding constraint is senior expatriate attraction and retention in a wartime economy, and the fiscal envelope is materially tighter. Riyadh's inland geography is a genuine resilience advantage relative to coastal Gulf centres. Priority: lead with legal and regulatory certainty and with domestic capital markets depth, not with construction milestones.

Qatar and Bahrain. Both are highly Hormuz-exposed. Qatar's LNG dependence makes it structurally vulnerable to the same chokepoint that threatens its revenue base; Bahrain's markets fell about 7% in the first month of the war. Both should prioritise niche specialisation — Qatar in energy-linked finance and sovereign co-investment, Bahrain in its long-standing banking and insurance base — over generalist competition with Dubai and Abu Dhabi.

Astana (AIFC). Improved three places to 65th in GFCI 39 despite a lower absolute rating, showing that relative gains are possible in a falling market. It is a useful model for emerging centres: English-law framework, focused product set, patient institutional build.

Hong Kong. Ranked third globally and first in FinTech in GFCI 39, and the world's leading IPO venue. Its proposition is institutional connectivity — mainland access combined with common law, convertibility and international capital. Risks are regulatory bandwidth under listing volume pressure, document quality, and the persistent geopolitical discount applied by some Western allocators. Priority: sustain listing reform while visibly tightening quality controls, and complete the stablecoin and tokenisation build-out that gives it a genuine first-mover position in Asia.

Singapore. Fourth globally, one rating point behind Hong Kong. Its advantages are perceived neutrality, ASEAN reach and regulatory credibility; its weakness has been equity market liquidity, which the Equities Market Review is directly addressing through capital placement, listing reform and family office deployment requirements. Priority: convert wealth management scale into capital markets depth, which is the gap Hong Kong currently exploits.

Tokyo and Osaka. Tokyo re-entered the top ten at tenth, up from fifteenth, and Osaka rose to 26th from 36th. Japan's asset management reform agenda, corporate governance improvement and currency dynamics have made it a credible destination again. Priority: sustained follow-through on asset management entry, English-language administration and tax clarity.

Shanghai, Shenzhen and Beijing. Shanghai rose to sixth and Shenzhen ranks ninth and second in FinTech. Their strength is policy proximity and industrial ecosystem depth; their constraint is capital account convertibility. Priority for international firms is to treat them as complements to Hong Kong rather than substitutes.

India (GIFT City IFSC). Backed by a unified regulator (IFSCA) that has been progressively liberalising fund management rules, global in-house centre regulations, dematerialisation and payments frameworks. Combined with NSE's global top-five IPO position, India is the most credible new entrant to the top tier over a five-to-ten-year horizon. Priority: build professional services depth and secondary market liquidity, which lag the regulatory framework.

Asia's shared exposure. The region is a large net energy importer. The IMF expects Asian growth to moderate to about 4.4% in 2026 with emerging Asia inflation rising to roughly 2.6%, and warns that a longer or larger energy shock would materially worsen this. Asian centres benefit from the AI cycle while carrying the energy shock — the exact inverse of the Gulf's position.

Immediate (0–12 months)

1. Publish a resilience standard and report against it. Recovery time objectives for trading, clearing, custody and payments; annual sector-wide cyber and physical continuity exercises; independent verification. Make this a marketing asset, not a compliance document.
2. Map and reduce third-party concentration. Cloud, data centre, connectivity and market data dependencies, with named alternates and tested failover.
3. Set and publish authorisation service standards. Time-to-licence, time-to-visa, time-to-fund-registration, reported quarterly with variance explanations.
4. Fund supervisory capacity ahead of licence growth. A centre that grows regulated entities 30% a year and supervisory headcount 5% is accumulating a tail risk it cannot see.
5. Stand up a talent retention package. Family provisions, evacuation guarantees, schooling capacity and licence portability. Treat senior talent attrition as a board-level KPI.

Medium term (1–3 years)

6. Choose three specialisations and resource them properly. Assess candidate clusters against existing anchor institutions, adjacent industry, regulatory readiness and talent supply. Decline the rest.
7. Build domestic capital mobilisation mechanisms. Directed but voluntary programmes placing sovereign, pension and insurance capital into local listings and funds, on the Singapore model, with clear performance conditions.
8. Complete the digital asset stack. Licensing, custody, bank account access, settlement rails and clear treatment of tokenised deposits — sequenced so that institutional participants can operate end-to-end within the jurisdiction.
9. Pursue mutual recognition aggressively. Bilateral equivalence for fund passporting, digital asset licensing and professional qualifications reduces the cost of multi-jurisdiction operation and is a genuine differentiator while global frameworks remain fragmented.
10. Develop specialty underwriting capacity. War, marine, political and cyber risk, supported where necessary by a government reinsurance backstop.
11. Build the growth-equity layer and an exit route. Anchor Series B and C managers through a fund-of-funds on commercial terms with locally based investment teams, and open a proportionate growth segment on the exchange. An innovation strategy without later-stage capital and an exit path exports its own successes.
12. Sequence cross-listing correctly: ETFs and sukuk first, equities later, Asia not the neighbours. Build depository, FX and market-making capability through fund and debt cross-listings before promoting equity dual listings, and pursue index-provider recognition explicitly.
13. Convert innovation activity into revenue through procurement. Require regulated incumbents and state entities to pilot and buy from licensed startups, and report on it.

Structural (3–5 years)

14. Diversify the centre operator's own revenue base away from real estate rents toward licensing, data, market infrastructure and services.
15. Institutionalise legal certainty. Independent courts, arbitration, insolvency regimes and enforcement track record are the slowest asset to build and the hardest for competitors to replicate.
16. Establish a mirrored operating location. For conflict-exposed centres, a genuinely operational secondary site in a different risk geography — not a paper arrangement — is becoming a condition of institutional participation.

Indicative KPI dashboard

The centres that emerge stronger from 2026 will not be those that had the lowest tax rate or the newest towers. They will be the ones that were predictable when the environment was not, that kept operating when infrastructure was attacked, that supervised competently while growing fast, and that chose a small number of things to be genuinely world-class at. The Gulf has demonstrated that capital will tolerate considerable geopolitical risk when institutional quality is high; Asia has demonstrated that reform velocity in listing and asset management regimes translates quickly into flow. Both lessons are available to any centre willing to act on them.

The risks are correlated and the window is short. The compression at the top of the GFCI means small differences in execution now translate into large differences in position within three to five years.

Principal sources consulted, August 2026:

• Z/Yen Group and China Development Institute, Global Financial Centres Index 39, March 2026
• IMF, World Economic Outlook Update, July 2026, and associated press briefing; IMF Regional Economic Outlook, Middle East and Central Asia, April 2026; IMF blog on Asia and the energy shock, April 2026
• US Congressional Research Service, The Strait of Hormuz: Security Developments and Impacts, August 2026
• ADGM Q1 2026 results announcement, May 2026
• DIFC 2025 annual results, February 2026; DIFC H1 2026 registration data as reported August 2026
• Public Investment Fund, 2026–2030 strategy announcement, April 2026; Saudi fiscal and construction award data as reported 2026
• KPMG and Deloitte capital markets reviews, Q1 2026 and 2025 full year; Hong Kong Financial Secretary and FSTB commentary, 2026
• Hong Kong Monetary Authority stablecoin licensing, April 2026; comparative stablecoin regulation analyses, 2026
• MAS and Singapore Equities Market Review announcements
• KPMG Chinese Mainland and Hong Kong IPO Markets 2026 Mid-Year Review, June 2026; PwC Hong Kong capital markets mid-year review, July 2026; Saudi Tadawul Group and ADX listing disclosures; ADX cross-listing announcements, December 2025
• Wamda and MAGNiTT, MENA venture capital H1 2026 reviews, July 2026; Crunchbase Asia venture data, Q1 and Q2 2026; Tracxn Southeast Asia funding and AI landscape data, 2026
• Al Jazeera, Reuters, The National, Arabian Business, Global Finance, Gulf International Forum, Deloitte Middle East economic bulletins, 2026

Figures are as reported at the time of publication and should be re-verified before use in a board or investment paper, particularly conflict-related data, which is changing week to week.