Are Hong Kong Insurance Firms Ready for Regulatory Ramp-Up?
As Hong Kong starts to look more closely at money laundering in its insurance sector, insurers need to respond with a new approach and better tools.
Is Hong Kong's insurance sector low-hanging fruit for money launderers?
As a major financial center, Hong Kong has long attracted criminal elements looking to blend into the woodwork while moving illicit funds. The city’s booming insurance sector has lately added to the attraction. With 13 of the top 20 global insurance firms present, the value of premiums has grown at a double-digit pace and insurance penetration is now the second-highest in the world. In 2018, written premiums reached nearly USD$70 billion.
Many other factors make the sector ripe for picking, including high volumes of activity, a significant degree of intermediation and opacity between insurers and beneficiaries, proximity and connectedness to high-risk jurisdictions, and low levels of enforcement actions.
Despite its value and vulnerability, recent statistics would seem to indicate that Hong Kong’s insurance sector is not being overtly targeted. There have been few filings of suspicious transaction reports ("STRs"), for instance, and only small amounts of insurance assets have been restrained or confiscated by the Joint Financial Intelligence Unit (JFIU) in Hong Kong.1
Is there little reason for concern? Or is it possible that illicit transactions are slipping through the cracks and not being reported?
While the interpretation is up for debate, the Financial Action Task Force (FATF) money laundering evaluation of mainland China suggests the latter. Last April, the FATF’s report explicitly noted Hong Kong’s role as a node in money laundering schemes originating in mainland China.2
Hong Kong’s government is slowly starting to acknowledge the rising risk and recently has implemented significant changes in its oversight of the insurance sector. Here’s what insurers need to know about those efforts in order to step up compliance.
A Tougher Stance on AML
In 2015, the government created the Hong Kong Insurance Authority ("IA") to take over functions of the former commissioner of insurance and granted expanded regulatory authority to revoke licenses and levy fines. As part of a phased effort, the IA’s oversight responsibilities were later extended to include insurance intermediaries (brokers and agents) along with insurers. (In fact, this action follows a growing global trend in anti-money laundering ("AML") circles of holding individuals — such as controllers, directors, or managers of insurance entities — personally liable if their actions facilitated money laundering offenses.)
Although no significant penalties have been levied in Hong Kong yet, recent fines on European insurers show that the cost of lax AML compliance can be high (in one case totaling nearly USD$10 million3). This does not include the costs of remediation and intensified monitoring, which can often dwarf the original fine.
For Hong Kong insurers, the fact that mainland Chinese customers purchase more than one in four new life insurance policies complicates issues. Most of these policies are written in USD or HKD and offer broader investment options when compared to mainland insurance products. Mainland Chinese authorities’ control over this stream of customers has not been fully tested but the clampdown on UnionPay credit card transactions in 2016 and its potential implications for Hong Kong insurers that run afoul of mainland policy objectives loom large.
China has in fact taken a tougher stance recently to combat money laundering. In 2018, its central bank fined a mainland insurer for record-keeping deficiencies related to AML compliance.4 Further, the mainland’s industry regulator, the China Banking and Insurance Regulatory Commission (“CBIRC”), recently issued stricter rules to clamp down on money laundering.
Next Generation Detection
Now is the time for insurers and insurance intermediaries to review existing AML controls and consider making improvements. Taking a fresh look at mitigation efforts can both reduce risk and cut costs. Recent onsite inspections of Hong Kong insurers by the IA noted deficiencies in monitoring and detection capabilities, which opens these entities to regulatory risks.5
At the same time, the costs of running existing checks is often quite high, and ineffective. Some studies suggest that 98 percent of system-generated flags do not ultimately lead to an STR.6 Sifting through this heap of false positives is a drain on resources and distracts in-house compliance staff from investigating high risk cases.
Improved detection capabilities are a critical pillar of a cost-effective, robust AML program. This solution should be technology-enabled, but practitioner-designed. Big data and machine learning can increase the breadth and speed of AML checks while also reducing costs, but these tools can only be harnessed effectively when combined with a deep understanding of the people, products and common money laundering typologies.
Consider these four essential components of a next-generation detection pillar:
- Shifting from a transaction-based approach using static rules to a risk-based approach that blends customer information, cross-border linkages in funding source or coverage, policy type, and known typologies to tailor the amount of scrutiny applied (consistent with the Hong Kong IA's revised AML guidelines7)
- Incorporating big data concepts to integrate multiple structured and unstructured data sources, improve processing speeds and reduce the cost of storage
- Employing advanced statistics and modeling to introduce behavioral (e.g., applied to customers) and peer-based (e.g. applied to intermediaries) analytics that reveal unusual patterns, trends and relationships and result in improved detection accuracy
- Utilizing machine learning and potentially more sophisticated artificial intelligence to identify previously unknown risk variables and patterns to react to emerging typologies
The insurance industry has typically been a leader in applying technology to business challenges. It has been at the forefront by applying advanced analytics and big data concepts to improve the customer experience with sales and claims and has also made leaps in risk assessment and underwriting. One notable example is making real-time adjustments to car insurance premiums based on driving behavior.
It’s time to apply this same outlook to AML. The threat is real: With bad actors developing ever more sophisticated schemes and regulators beginning to pay more attention, companies have been put on notice to step up their AML game. At the same time, the benefits of a more lean, cost-effective AML program can flow directly to the bottom line.
1: Hong Kong Money Laundering and Terrorist Financing Risk Assessment Report (April 2018): https://www.fstb.gov.hk/fsb/aml/en/doc/hk-risk-assessment-report_e.pdf
2: FATF Anti-money laundering and counter-terrorist financing measures: People’s Republic of China Mutual Evaluation Report (April 2019): http://www.fatf-gafi.org/media/fatf/documents/reports/mer4/MER-China-2019.pdf
3: Insurer CNP fined 8 mln euros over compliance breaches (Aug 2018): https://www.reuters.com/article/france-insurance-fine/insurer-cnp-fined-8-mln-euros-over-compliance-breaches-idUSL5N1UR7EG
4: China Life Insurance fined by central bank (July 2018): https://www.insurancebusinessmag.com/asia/news/breaking-news/china-life-insurance-fined-by-central-bank-107431.aspx
5: Key Findings of AML/CFT Onsite Inspection Visits to Authorized Insurers Carrying on Long Term Business (May 2018): https://www.ia.org.hk/en/legislative_framework/circulars/antimoney_laundering/files/cir_aml_20180531.pdf
6: Anti-money laundering controls failing to detect terrorists, cartels, and sanctioned states (Mar 2018): https://www.reuters.com/article/bc-finreg-laundering-detecting/anti-moneylaundering-controls-failing-to-detect-terrorists-cartels-and-sanctioned-states-idUSKCN1GP2NV
7: Guideline on Anti-Money Laundering and Counter-Terrorist Financing--GL3 (Nov 2018): https://www.ia.org.hk/en/legislative_framework/files/GL3_eng_Nov2018.pdf