Anti-Money Laundering and
Countering the Financing of Terrorism

Submissions on the third discussion document (Supervisory Framework)

Kiwibank Limited

30 November 2006

Contents

A. Introduction

B. Executive summary

C. Comments on specific proposals

D. Consultation questions

A. Introduction

1. This submission is made on behalf of Kiwibank Limited (Kiwibank).

2. Contact details for the purpose of this submission are:

Gavin Weekes
General Manager, Risk Management
Kiwibank Limited
Private Bag 39888
Wellington 5045

3. Kiwibank is a registered bank, and a wholly owned subsidiary of the State-Owned Enterprise, New Zealand Post Limited (NZ Post).

4. Kiwibank undertakes both retail and wholesale banking activities, predominantly within New Zealand. The Bank offers a full range of retail banking products and payment transactions through the NZ Post Retail network; electronic channels (eg. telephone and Internet); and direct representation (eg. Home Lending Advisers).

The Bank also:
a) conducts international wire-transfer services as an agent for Western Union;
b) undertakes the sale and conversion of foreign drafts as an agent for Travelex and Citigroup
c) accepts consumer bill payment transactions for other financial intermediaries (eg. American Express) and;
d) accepts retail deposit transactions for other financial institutions (eg. TSB Bank).

5. The Ministry of Justice (MOJ) on behalf of the New Zealand FATF Inter-Agency Working Group[1] , has produced a third discussion document "Anti-money laundering and Countering the Financing of Terrorism (AML/CFT) - Supervisory Framework", outlining preliminary proposals for the regulation, supervision and enforcement of proposed Financial Action Task Force (FATF)[2] inspired reforms required for New Zealand to meet its international obligations.

6. The MOJ has sought comments on the proposed supervisory framework.

B Executive summary

7. Kiwibank has considered the supervisory framework proposals of the FATF Inter-Agency Working Group and submits that the current proposals appear superficial and ill-advised in many areas. Consequently, we cannot consent to or endorse many of the current supervisory framework proposals, largely because a sound case for these proposals has yet to be articulated.

Our principal concerns stem from:

a) The supervisors have not been tasked with or given a mandate to promote tangible AML/CFT outcomes (ie the actual detection, investigation and prosecution of money laundering) through the production of viable financial intelligence.

b) The supervisors' brief does not promote a balance between enforcing compliance and the production of real AML/CFT outcomes.

c) The multi supervisor/regulator model is highly unlikely to be the most efficient and effective means of attaining any real benefit from the FATF inspired AML/CFT regime

d) The absence of any proposals on civil and criminal liabilities (likely to be established under accompanying FATF AML/CFT enabling legislation) limits any meaningful evaluation of the appropriateness of proposed supervisory powers.

e) The lack of proposals on the sharing of AML information appears to be a significant oversight.

f) Proposals based on the adoption of a risk based approach have merely added to the confusion surrounding the application of a risk based approach.

In summary, we believe the discussion document is a useful starting point but does not present a complete or viable supervisory framework. Kiwibank asks that before proceeding further with the current proposals, that the Working Group give further consideration to the Australian Austrac AML/CFT supervisory model, as we believe this offers the soundest means of correcting a number of deficiencies in the current proposals.

C. Comments on specific proposals

Kiwibank makes the following specific comments on the supervisory framework proposals:

8. Unconvincing Supervisory Framework

While the design principles and outcomes sought from the regulatory framework appear laudable, their linkage to the actual proposals appears tenuous. Additionally, few of the proposals (and particularly the preferred multi supervisor model) have been justified in terms of basic economics (ie effectiveness in delivering tangible AML/CFT outcomes (see 9 below) and clearly articulated and quantified cost- benefit analysis).

The current discussion document also provides little insight into how the proposed role and responsibilities assigned to the supervisor(s) will in practise deal with remainder of the FATF inspired AML/CFT reform agenda. A number of issues raised in respect to the Working Group's second discussion document[3] remain unresolved (eg. the application of a risk based approach) and current document provides limited insight into how the whole AML/CFT regime will operate.

We also note that there is no firm linkage of the current proposals to the expected outcomes of enhancing New Zealand's international reputation, nor to contributing to public confidence in the financial system.

Given the lack of rigor in the business case supporting the proposed framework, Kiwibank recommends that the Working Group:

a) Increase its level of engagement with industry stakeholders to substantiate a number of assumptions on which the current proposals are founded.

b) Anticipate producing a more substantial set of proposals on the supervisory framework (including possibly discarding some current preferences (eg. the multi supervisor model)).

9. No responsibility for real AML/CFT outcomes

Kiwibank struggles to see any clear linkage of the proposed supervisory framework to the primary AML/CFT outcome sought, namely that the supervisory system must be judged effective in its ability to detect and deter money laundering and terrorist financing.[4] Specific concerns include:

• Under the proposed framework supervisors are not held accountable for producing any real AML/CFT outcomes. Measurable AML/CFT outcomes that demonstrate an environment that is harsh on the money launderers/criminals (because good financial intelligence is produced that detects and leads to the confiscation of dirty money) are required to prove both the effectiveness of the entire FATF based AML/CFT regime and the effectiveness of the supervisors.
• Regulatory zeal is not a satisfactory substitute for actual AML/CFT outcomes (ie. prosecution of criminals). The current supervisory proposals appear skewed towards penalties being imposed on reporting entities for perceived procedural breaches as opposed to actual instances of money laundering. Kiwibank contends that regulatory effectiveness should not be assessed purely on the number of prosecutions/reprimands imposed on industry.
A clear statement(s) of measurable AML/CFT outcomes would help provide the missing balance between compliance and real-world effectiveness that is missing from the current proposals. Additionally, Kiwibank would welcome a clear guiding statement instructing supervisors to achieve the desired balance between effectiveness (ie. real AML/CFT outcomes) and compliance.


• The absence of measurable AML/CFT outcomes suggests that the proposed reforms and supervisory framework will never be reviewed. Given the level of complexity and assumption that underpins the proposed framework it appears unrealistic to expect that multi supervisor model can be introduced effectively and maintained efficiently. Consequently, if the current proposals are implemented these should be subject to review against pre-determined success criteria (ie measurable AML/CFT outcomes) after an appropriate period of operation.
• The proposed supervisory framework is likely to drive a "defensive filing" mentality in reporting entities. As the supervisors are not tasked or held accountable for reducing the actual incidents of money laundering, many reporting entities will flood the Financial Intelligence Unit (FIU) with low quality financial intelligence, in order to comply with regulatory expectations. Clearly this behaviour serves no-one well and needs to be avoided. However, there is nothing in the current proposals that prevent this result. Again measurable AML/CFT outcomes could be used to redress this probable and averse effect.
• The merits of adopting a risk based approach to the FATF AML/CFT regime are apparent to all. And the current supervisory framework proposals promote a risk based approach to supervision. However, there is currently little consensus between industry and government on how a risk based approach will be implemented. Kiwibank believes that a clear statement of desired actual AML/CFT outcomes will be instrumental in resolving this current impasse, around the risk based approach.

Kiwibank contends that the regulatory framework should encompass a focus on real AML/CFT outcomes as opposed to merely policing industry compliance with FATF obligations. The supervisory framework itself must reflect the desire to establish a credible deterrence of money laundering and terrorist financing.

10. Requirement to regulate does not necessitate multiple regulators

The requirement to regulate and supervise reporting entities is inherent (and probably unavoidable) if the governments goal is achieving compliance with the FATF AML/CFT guidelines.

However, Kiwibank contends that the Working Group's preferred multi supervisor model is not the most sensible or efficient means of achieving AML/CFT regulation and supervision.

The Australian Austrac model of a combined single regulator and financial intelligence unit appears to offer a better opportunity for progressing the FATF agenda. As FATF is an international standard for AML/CFT, emulating Australia in this instance is not bad strategy. The single supervisor model is:

• Simple to understand (when compared with the multiple supervisor determined by the reporting entities' industry affiliation) which should promote public confidence.
• Consistent with the approach of our nearest neighbour, therefore arguably enhancing our international reputation.
• Likely to be the most cost effective model given the limited AML/CFT resources available in New Zealand.

While removing the perception that reporting entities may be treated differently by different regulators/supervisors.

11. Multiple regulator model is misguided

Severe doubts are held around the practicality of a multiple supervisor model. These doubts include:

• The multiple supervisor model appears to be a default choice, arising from the perception that a single regulator model will be too costly to adopt. However, no financial estimates have been provided to verify that the preferred multiple supervisor option is in fact a lower-cost model.
• At face value, the multiple supervisor model appears more expensive to establish and maintain than a single supervisor. The FATF guidelines expect supervisors to possess competence and expertise in AML/CFT. Inherently, the establishment and maintenance of this competence across three (and possibly four) supervisory organisations appears to be a higher long-term cost structure. Additionally, the perceived advantages of leveraging existing expertise in supervisory and regulatory arrangements may be illusory, as none of the proposed supervisors are currently tasked with AML/CFT supervision. Essentially they will each need to develop this new competence. Scarcity in the New Zealand market of personnel with real AML/CFT experience is also likely to further escalate costs as supervisors compete to establish capability.
• The Working Group has not opined on the effectiveness of a single supervisor compared to multiple supervisors in the actual prevention and detection of money laundering. We believe an assessment of the potential effectiveness in achieving the primary objective of the proposed reforms is a more important consideration than relative cost. Kiwibank suspects that the Australian Austrac model (of a single combined supervisor and specialist financial intelligence unit) is likely to be a more effective model for the collection of useable financial intelligence.
  Kiwibank asks that before proceeding further with the current proposals, that the Working Group give further consideration to the Australian AML/CFT supervisory model. In Austrac, Australia has:
  i) A proven, coordinated, national financial intelligence gathering capability.
  ii) Human capability (experience and credibility).
  iii) A strong focus on detecting/deterring criminal behaviour; and
  iv) A single AML/CFT supervisor with which various industry groups can confidently and effectively interact.
• The activities of reporting entities are unlikely to fall into the neat supervisory categories envisaged in the discussion document, as many entities have mixed business lines. This potentially gives rises some reporting entities having to sustain the additional direct and indirect costs from interaction with two or more supervisors. This circumstance appears contrary to the desired AML/CFT regime outcomes of cost effectiveness and competitive neutrality. Kiwibank is an example of an entity potentially exposed to this multiple supervisor burden. As a registered bank, part of our operations would be supervised by the Reserve Bank. However due to agency arrangements Kiwibank has with Western Union, another (yet unnamed supervisor) is likely to want to exercise its mandate for the supervision of money transfer/exchange agents. Additionally, given Kiwibank uses the New Zealand Post retail network there appears potential for the Securities Commission to classify these operations as "non authorised deposit takers". We also note the potential for Kiwibank's insurance operations to fall under the jurisdiction of another yet unnamed supervisor.

Again, Kiwibank urges the Working Group to re-assess the merits of the preferred multi supervisor supervisory model.

12. Underdeveloped supervisory functions and powers proposals

The Working Groups' proposals around supervisors' powers and functions appear simplistic and underdeveloped.

Our concerns include:

• The proposals for the application of the risk based approach are both ambiguous and amiss. The discussion document indicates that a risk based approach is to be permissive rather than mandatory. However, it indicates that supervisors will establish default requirements for businesses not carrying out risk assessments and proposes establishing customer due diligence (CDD) requirements for low risk activities and customers. This implies:

a) The Working Group is proposing to make CDD mandatory for all activities irrespective of the risk associated with the activity or customer. This presupposition by the Working Group effectively negates the risk based approach. As detailed in our earlier submissions, Kiwibank firmly believes that the application of the risk based approach is the key to achieving economically viable compliance with the FATF AML/CFT regime. Neutering the risk based approach as currently envisaged by the Working Group will turn the implementation of the FATF regime into a very expensive, widely intrusive exercise that is unlikely to yield any of the mooted benefits.
b) Perverse thinking about the risk based approach. Essentially, the Working Group is suggesting that supervisors are unlikely to accept reporting entities risk based AML/CFT solutions that are less onerous than the default or minimum position mandated by the supervisors. Consequently reporting entities have no incentive to invest in a risk based approach, as they will be forced either directly or indirectly to adopt the supervisors' mandated minimum position.

• The discussion document indicates that there is already agreement on how reporting entities will meet their obligations under FATF Recommendations 15 and 16. This is presumptive. It is unclear to Kiwibank how many of the issues raised in response to the Working Group's second discussion document are going to be resolved. Furthermore, the practicality of the stated measures (ie internal policies, procedures and controls, ongoing employee training programmes and an audit function to test the system, designated compliance officers etc) has yet to be deliberated.
  We note that full FATF compliance still eludes many developed countries (including the United States) so it seems sensible to consciously carve out from New Zealand's AML/CFT reform programme the FATF requirements that are economically unfeasible for a low risk country. We suspect full compliance with recommendations 15 and 16 is only relevant to selected reporting entities with high risk money laundering and terror financing customers and activities.
• The discussion document does not give sufficient recognition to the role of the supervisor in promoting responsible corporate management for the prevention of money laundering and terrorist financing. The supervisor functions and powers appeared skewed towards executing a "policing" mandate. Consequently, "fear of prosecution" seems to be the only strategy available to supervisors as they attempt to increase financial sector management accountability. While not objecting to increased management accountability, Kiwibank sees nothing in the Working Group's proposals that advocates increased management judgement in delivering successful AML/CFT outcomes and equipping supervisors with a suitably constructive mandate to engage business accordingly.
• The Working Group's advocacy of a multi supervisor model combined with the proposed policing mandate embodied in the functions and powers, gives rises to concerns about consistent application across sectors. While the discussion document acknowledges the potential for inconsistency, no options or proposals are offered as to how this risk will be managed. The existence of an AML/CFT Advisory Group provides little insight into how in practice consistency will be promoted.
• Kiwibank cannot fully consent to or opine on the proposed supervisory/regulatory sanctions until clear proposals are presented on likely criminal and civil liability that will be established under new FATF AML/CFT enabling legislation. In the absence of any draft legislation, there are a number of considerations that have bearing on the nature and suitability of supervisors' powers. These include:

a) Will the emphasis of the legislation be on overall compliance, rather than individual procedural breaches?
b) Is there a de minimus defence (ie. money laundering does not include the proceeds of crime below a certain level)?
c) Is there a due diligence defence?
d) Has an organisation or individual committed an offence if the relevant activity is not a crime in the place where it occurred?
e) If there is to be an offence related to "suspicion" or "unusual transactions" what do these terms mean?
f) Who is guilty of an AML offence, (ie. the individual that comes into contact with the customer, senior management, the Money Laundering Reporting Officer or the corporate entity)?
g) Where does liability lie in agent-principal relationships?

• While on-site visits by supervisors are advocated in the discussion document, we suggest that many supervisors may be ill-equipped or apprehensive about this prospect.

13. No proposals on exchange of AML information

The supervisory framework is silent on a key element of any effective AML/CFT regime, namely the exchange of information or sharing of AML/CFT information.

FATF recommendation 14 envisages a "safe harbour" law or regulation for the exchange of AML information between supervisors and reporting entities. However, this issue is not addressed in the current discussion document. Nor is the related matter as to whether reporting entities should be encouraged to share information for AML/CFT purposes. We note that the US Patriot Act (Section 314) authorises financial institutions to share information for AML purposes. As the production of sound financial intelligence is one of the main objectives of an effective AML/CFT regime, Kiwibank believes supervisors should be specifically required to facilitate the production of high quality financial intelligence.

The discussion document is also silent on a related matter, that of privacy. The proposals do not recognise the need to strike a balance between the supervisors' right to investigate and test systems, and the law abiding customer's right to expect a degree of privacy about his or her financial affairs. Public confidence in the current proposals is likely to be diminished if the Working Group does not opine on this matter.

14. Financial intelligence Unit

As previously mentioned in Sections 10 and 11, Kiwibank endorses the Australian Austrac model of a combined single supervisor and specialist financial intelligence unit, and believes the Working Group should give further consideration to adopting this model in New Zealand.

The combined single supervisor FIU model appears to offer the best opportunity to generate and collate high quality financial intelligence that will lead to the prosecution of money launderers. If those who are tempted to commit financial crime believe that the chances of it being detected, investigated and prosecuted are worth the risk to them, then the entire FATF AML/CFT strategy will be ineffective and resources wasted. As expressed in section 9 the detection, investigation and prosecution of money laundering are all real/tangible AML/CFT outcomes and they are catalysed by the production of sound/high quality financial intelligence. Consequently, the supervisory model should be structured to facilitate the production of this financial intelligence.

Additionally, the combined single supervisor FIU model avoids the potentially messy and wasteful scenario of multiple supervisors all issuing their own AML/CFT standards and guidelines and the FIU issuing its own unenforceable guidelines.

Current proposals to grant the FIU/Police new powers to enforce the filing of Suspicious Transaction Reports (STRs) appear to be a response to inadequate consideration of the role of the supervisors. If supervisors were correctly instructed and focused on ensuring reporting entities produced high quality financial intelligence, there would be no need for the FIU to separately/jointly regulate the filing of STRs. Again, the combined single supervisor FIU model appears to tidily resolve this issue. And if some rigorous thought was given to AML information exchange arrangements (refer section 13) then conferring statutory "express production" powers may also prove unnecessary.

Kiwibank views formal feedback by the FIU as an essential part of an effective AML/CFT regime, however data on the filing of STRs is not instructive. Most businesses are focused on, measured on and understand tangible outputs. STRs are an input. Consequently Kiwibank believes that a stronger AML/CFT partnership would be forged if the FIU reported outcomes (ie money-laundering incidents detected, investigated and prosecuted). This information would provide a clear indication about the effectiveness of suspicious transaction reporting and how industry's efforts are helping to investigate and prosecute money laundering.

15. Two staged implementation a flawed idea

Kiwibank believes that the proposed and preferred two-stage implementation option for supervisory reforms is superficially appealing but full of contradictions when fully considered. Specifically:

• A two staged implementation is not consistent with the overall FATF agenda goal of creating an environment that is hostile to money laundering and financing of terrorism, as it would theoretically create a window of opportunity for criminals to channel there money laundering activities away from the financial institutions and casinos into the Designated Non-Financial Businesses and Professions (DNFBP) for an indefinite period.
• Arguably the DNFBP sector should be the first target of the new supervisory regime, as it not subject to many of existing AML/CFT regulations and disciplines that are accepted practice in the banking sector, and for larger financial institutions and casinos. While the current AML/CFT regime has been found wanting against the FATF AML/CFT guidelines, it has had some successes in detecting money laundering. However, very little is known about the DNFBP sector and merely because of this uncertainty it could be considered higher risk.
• Due to factors of size, complexity and legacy systems larger financial institutions (ie. the major Australian owned banks) will have a significant change management burden to carry in respect of the proposed FATF AML/CFT reforms. In the interests of both fairness and maintaining consumer confidence in the banking sector, Kiwibank suggests that an early imposition of the supervisory framework to the banking sector would be unconscionable. Given the complexity of the proposed AML/CFT changes themselves and scarcity of resources to implement these changes in reporting entities, realistic timeframes (possibility 3 or 4 years) must be permitted to allow reporting entities to prepare internal systems and process. Only then should the new supervisory regime commence.

Again the architects of the two-stage implementation option for supervisory reforms are faced with a dilemma. Does it make sense to start the supervisory regime in the sector/industry that inherently will be the slowest in reaching the FATF starting line?

16. AML/CFT supervisory framework

As Kiwibank views the production of quality financial intelligence concerning money laundering as a key tangible outcome of the entire FATF AML/CFT regime, we support the FIU being involved in framing new AML/CFT legislation. Obviously, the FIU has a key role in ensuring that the proposed legislation facilitates the prompt and accurate collection, collation and dissemination of this AML/CFT information.

Similarly, the supervisors also have a role in framing the legislation, however they must be tasked with/accountable for achieving real AML/CFT outcomes, to ensure new legislation is not skewed towards penalties being imposed on industry for procedural breeches as opposed to actual instances of money laundering.

Proposals for the FIU to be able to issue non-enforceable guidelines and separately retain oversight over suspicious transaction reporting appear to result more from the compromises necessary to create a balanced multi supervisor model, rather than any inherent requirement in the FATF guidelines themselves. Again, the Austrac model avoids the need for these "balance of power" compromises, which can be viewed as sapping the efficiency from the whole AML/CFT regime.

As previously stated (Refer sections 10 and 11) Kiwibank has profound doubt about the merits and workability of the multi supervisor model.

The need for the proposed AML/CFT Advisory Group is another compromise that arises from the multi supervisor model. In effect, the Advisory Group would not be necessary (to ostensibly promote consistency) if a single supervisory model was introduced. Before proceeding further, Kiwibank suggests that the AML/CFT Advisory Group concept be challenged so that there is clarity and understanding about:

• How the Advisory Group plans to achieve its stated goal of consistency of AML regulation and supervision across sectors?
• What real AML/CFT outcomes will the Advisory Group be tasked with delivering?
• What are the tangible accountabilities of this group? At present there are no proposed accountabilities for the Advisory Group, while the proposed name (AML/CFT Advisory Group) intimates a well intended but unaccountable entity in gestation.
• Who will determine the effectiveness of this group and against what criteria?
• How will the Advisory Group bridge the gap between its members and the reporting entities? The proposed composition of the group[5] effectively isolates it from the business sector. Kiwibank struggles to see how this supervisors' forum can achieve its proposed goals without having an understanding of AML/CFT issues from the perspective of the regulated reporting entities.

Again we note that the AML/CFT Advisory Group does not inherently add to the efficiency of the AML/CFT regime, but is obligatory because of the desire to pursue a multi supervisor model.

D. Consultation questions

The Working Group has requested input on specific questions related to broader policy issues. These consultation questions are cross referenced to the relevant sections of this submission, below.

Requirement to regulate

Refer sections 8, 9 and 10 above

Supervisory functions and powers

Refer sections 12 and 13 above

Financial Intelligence unit

Refer section 14 above

Supervisory models

Refer sections 10 and 11 above

Two staged implementation

Refer section 15 above

Proposed framework

Refer section 16 above

Footnotes

1. The FATF Inter-Agency Working Group is comprised of the following agencies: Ministry of Justice, Ministry of Foreign Affairs and Trade, Ministry of Economic Development, Reserve Bank, Customs, Police, Treasury, Securities Commission, Inland Revenue Department and Department of Internal Affairs.

2. The G7 countries established the Financial Action Task Force in 1989 as an international inter-governmental body whose broad purpose is to develop and promote national and international policies and legislation to combat money laundering and terrorist financing.

3. Anti-money laundering and countering the financing of terrorism: New Zealand's compliance with FATF recommendations. Second discussion Document. June 2006.

4. Third Discussion Document - Outcome Sought Page 10.

5. Reserve Bank, Securities Commission, Department of Internal Affairs, Ministry of Economic Development, FIU, Customs, Foreign Affairs and Trade, Inland Revenue.