2019: Hong Kong issued eight virtual-bank licences in one go — a big-bang bet that branchless, digital-first banking would pressure incumbents into innovation. Five years later: all eight still loss-making, combined deposit share 0.3%. The "low-cost revolution" turned out to be a high-cost experiment.
1. Why "no branches" didn't mean "low cost"
The pitch was clean: no branch network → lower cost → better rates for customers. Reality added the terms the pitch forgot:
| Expected saving | Actual offsetting cost |
|---|---|
| No branch network | Full banking-licence compliance stack |
| Digital-first efficiency | Customer acquisition costs in a saturated market |
| New-entrant agility | Same capital/liquidity requirements as incumbents |
| Scale through digital | Hong Kong's absolute market size is small |
2. Was issuing eight at once the mistake?
- Simultaneous licensing created eight competitors chasing the same thin wedge — fragmentation instead of a focused challenge to incumbents.
- Each bank faced the full capital/compliance burden while fighting for a market share that, combined, stayed at 0.3%.
- The policy goal — competitive pressure on incumbents — was achieved as an externality, not as a business model.
RegSir's take
The question isn't "did virtual banks fail" — it's "what were they for". If the goal was forcing incumbents to upgrade digital experience, it worked — as an externality the industry paid billions to produce. That's an expensive way to regulate. The more honest review is structural: eight licences in a market that couldn't support eight profitable digital banks meant the failures were predictable at issuance. The contribution is real — better apps, faster onboarding — but you can mandate those outcomes directly; you don't need eight loss-making banks to get them.
3. Compliance takeaways
- Deposits are safe — that's not the issue: virtual banks are licensed under the Banking Ordinance with full deposit protection; the question is viability, not safety.
- Licence count ≠ market outcome: when assessing new licensing regimes, model the post-entry economics — permission to operate isn't a path to profit.
- Externalities aren't business models: benefiting incumbents' customers isn't a revenue stream — the wedge has to be commercially capturable.
4. Official sources & reporting
| Document / report | Link |
|---|---|
| HKMA — virtual bank authorisation & supervisory framework | hkma.gov.hk ↗ |
| Banking Ordinance — licensing regime | elegislation.gov.hk ↗ |
| Deposit Protection Scheme coverage | dps.org.hk ↗ |