2 December Gavin Quigan Manager, Superannuation Schemes Compliance Monitoring Financial Markets Authority PO Box 1179 Wellington 6140

Transcription

1 MCA NZ Limited PO Box 8811 Symonds Street Auckland 1150 SuperLife House 41 Charlotte St Eden Terrace 1021 T (09) F (09) E mca@mcanz.co.nz 2 December 2011 Gavin Quigan Manager, Superannuation Schemes Compliance Monitoring Financial Markets Authority PO Box 1179 Wellington 6140 By gavin.quigan@fma.govt.nz Dear Gavin KiwiSaver Performance Fees FMA Guidance Note I refer to the draft guidance note and attach a submission by way of a discussion paper. The views are a combination of my personal views and those of MCA. MCA s views are based on its experience as a wholesale strategic investment consultant. I note that I am a shareholder of a KiwiSaver provider (SuperLife) and MCA is a related company. As such we have a vested interest in the outcome of the final guidelines. SuperLife itself does not have a formal view on performance fees. However, it is in SuperLife s interests for the guidelines to let other KiwiSaver providers adopt generous performance fees, as this will ultimately make SuperLife more competitive. I am happy to discuss any aspect as appropriate. Yours sincerely Michael Chamberlain for MCA NZ Limited

2 Discussion note in respect of the Guidance Note on KiwiSaver Performance fees 2 December 2011 This discussion note reflects the views of Michael Chamberlain and MCA on the draft FMA guidance note on KiwiSaver Performance fees. It is noted that SuperLife Limited a related company to MCA is a KiwiSaver scheme provider. It is in SuperLife s interests to encourage the use of performance fees as this will make SuperLife more competitive. However, both Michael Chamberlain and MCA believe performance fees should be banned for funds within KiwiSaver schemes and for vehicles that KiwiSaver schemes invest in. Recommendations 1. Performance fees should not be permitted in respect of KiwiSaver schemes. We make this recommendation as there is no evidence that performance fees lead to better investment outcomes. Performance fees will most likely change the incidence of the fees and ultimately increase the fees otherwise payable. These are against the principle of a low cost retirement savings initiative. Also, as the taxpayer has an considerable investment in KiwiSaver and KiwiSaver is part of public policy it is an inappropriate use of taxpayer money to encourage a wealth transfer from members to managers. If performance fees are permitted: 2. There should not be a guidance note on performance fees. Instead, the FMA should apply the legislative test of not-unreasonable to the total fees of the scheme and not each of the component parts of the total fees. The test should be whether the fees as a whole are not-unreasonable. A performance fee basis that might appear to be unreasonable as a performance fee may be offset by other aspects of the fee basis that makes it appropriate for that provider. We believe that the FMA should look at the outcome and not be involved in the commercial decisions of a provider. 3. Total management fees should be measured relative to the cost of the provision of services and total fees, including third party fees, should be required to be under the higher of $100 a year and 1% of assets. From 1 July 2017, the 1% maximum fee under a KiwiSaver should reduce by 0.05% a year, until it is 0.5% of assets by 1 July If a guidance note on performance fees is to be issued, then: 4. The design of the fee basis should allow: There to be no cap as the fee should only relate to performance and increased performance due to higher skill levels should be able to be rewarded; A high minimum return before a performance fee can be paid. It should reflect skill. It therefore should be appropriate recognised market index return plus a margin (say 2%). The market index must reflect the actual style of the manager so the fee is paid for skill and not style; 2

3 The performance should be measured over a minimum of 3 years (ideally 5 or 7) and then not payable until after a further 3 years, but only if the manager has continued to accrue performance fees in the intervening 3 years; Have a low base fee, i.e. excluding the performance fee. This should be less than 0.2% p.a. of assets; The calculation formula of the fee should be based on the dollar weighted internal rate of return and not a time weighted return; Where performance fees are paid, the manager should be required to make full disclosure to the members and to provide details of the calculation; Performance fees should not be payable when the returns are negative. Performance fees increase total fees and do not incentivise good behaviour 5. Generally, we believe that performance fees for traditional managed funds (including KiwiSaver) are wrong. History shows that performance fees increase total fees paid to managers and do not lead to higher or more consistent performance. Managers generally like performance fees as they paint a good picture, introduce complexity and ultimately lead to higher income. If they do not generate sufficient income the fund is closed down and a new fund introduced. 6. Performance fees also, do not encourage or reward better behaviour. The evidence is that they lead to inadequate income for some managers or excessive income for other managers. In each case, the outcome is poor for the member, as it can lead to perverse manager behaviour. Overall, performance fees lead to higher fees or higher risk. 7. The presence of performance fees have always been a practice that if you say it quickly it sounds right. The claim is that it s about alignment, only rewarding good performance, sharing the upside and downside etc. The argument is that better performance (alpha) deserves higher fees. 8. The problem with this is that most alpha (i.e. excess returns) is generated by beta practices (i.e. the market). Bond managers generate alpha not by making good decisions but by having a bias to long duration or credit. Equity managers generate alpha by adopting a style that has a bias to value shares, smaller company shares, non-index shares etc. (see the appendix). 9. Performance fees do not align interests. In reality, a manager normally makes more out of the standard base fee on new money, than they do from performance fees on existing money. Good past performance attracts new money (even though it should not) and this is more beneficial to the manager and acts as the better incentive. If performance fees are permitted and are to have any ability to work as intended, the fund should be closed to new money otherwise it acts against the interests of the member. What is not unreasonable? 10. KiwiSaver is about total fees that are not unreasonable. They therefore need to be easily understandable, low transparent. 3

4 11. We think that a fee is unreasonable if it is a fee (no matter how small) that a manager charges (directly or indirectly) that is not disclosed. Disclosed in this context means explained in a way that an average member can understand the quantum, incidence and nature of the fee, and the true impact it will have on their returns and accumulated wealth. This should be the disclosure standard. 12. If fees are high, but fully disclosed and disclosed in a way that meets the disclosure standard, they should not be unreasonable. To view them otherwise means that the regulators are looking to intervene with the markets. 13. A key issue with fees is that there is asymmetric information and the manager has the negotiation advantage. There is not good information of reasonable market practices and the average investor is at a constant disadvantage, as they do not understand percentages, compound interest and the impact of variables in a mathematical formula. 14. We suspect that no performance fee could be disclosed in a way that meets this test. It would be interesting to see how many 7 th form economics and mathematics students could explain, in an exam, the calculation of the typical performance fee practices adopted by current managers as currently explained. 15. Separate to the issue of fees, which are deductions from returns, is the impact of low (below market) returns because of poor practice. These are also equivalent to a cost. Any guidance note on fees must look at this aspect as well. 16. Often a manager s low return arises from investment arrangements where there are conflicts-of-interest. In the presence of the low fee to a fund may be due to a high fee to the manager. 17. A simple example of this is a bank s KiwiSaver scheme that invests in the bank s cash deposits. The management fee may be low (sometimes zero) but the interest rate may also be low as it is declared in advance by the bank. This should be classified as a fee. The returns to date reflect this but the fee is not disclosed. Cash options of bank schemes Gross returns after fees to 30 September year (%) 3 years (% p.a.) ANZ Cash ASB NZ Bank deposit National Bank cash OnePath cash fund SIL cash fund Westpac cash fund Average SuperLife (Tyndall) Source: Morningstar, SuperLife Limited 18. The banks have potentially made, on average, fees of an extra 1% a year. This table highlights the issue with fees that there is asymmetric information and the manager has the negotiation advantage. The banks have no incentive to be competitive in this area, as cash options are not the focus of why a person will choose a bank s scheme. 4

5 19. Cash returns are not the only area where conflicts arise. All areas where a bank operates has the potential to blur the line between high fees and low returns. This includes derivatives, currency contracts, the purchase of bonds through the bank s treasury operation etc. In each case performance fees, if allowed they will introduce greater layer of confusion and the potential for less understanding. What is a managers skill? 20. Most managers do not have a skill advantage. Managers with skill when they invest will look at two shares A and B and choose to buy A not B. If they have skill, their returns should be above market but often (about half the time) if they have skill their returns are below market. This may be because A is right on a 3 year horizon but wrong on 1 year horizon. Skilful managers (if they exist) typically move in to an investment too soon and out too soon. 21. International evidence suggests that managers do not have skill and managers with an apparent advantage short-term, do not have a permanent advantage. Performance fees therefore reward average behaviour. 22. Attached are three graphs that plot for global shares. Developed emerging markets less markets Market equally weighted less weighted Value shares less growth shares 23. The graphs show that a manager that adopts a process with a bias to any of the strategies, would on average have collected higher fees if the benchmark was the MSCI index. In each case however, there was no skill. If performance fees are to be allowed then besides disclosing the basis, at the time of payment the manager should have to be able to demonstrate with clear evidence that the emerging fee relates to skill. 24. Any formula therefore has to have: A benchmark relative to the true style of the manager reflecting what they do not say they do and this can only be set after the event A timeframe that reflects the manager s decision true frame Removes statistical good luck. Dollar cost averaging 25. Where regular contributions are paid, the mathematical impact of what is referred to as dollar cost averaging boosts the return. There is no manager skill involved. If performance fees apply then they should relate to dollar weighted returns and not geometric time weighted returns. 5

6 Taxpayers interests 26. The taxpayer has considerable money invested in KiwiSaver and has an interest in ensuring that it is not wasted. Performance fees have the ability to transfer money from members (and therefore the taxpayers interests) to the manager. It is absolutely critical that is it only paid where return outcomes are improved through the application of observable skill. Mathematically this cannot happen overall as if one investor buys or sells well another must sell or buy poorly. Therefore the net outcome will be lower net returns and less favourable outcomes. 27. Because KiwiSaver is a public policy issue, performance fees should be banned. In fact if fees are regulated then there should be maximum fees levels set. It is better not to have a guidance note on performance fees 28. As a general principle, we believe that it is inappropriate to have a guidance note in this area. 29. The presence of a guidance note sends out a message that if you comply, your fees are not unreasonable. This is the wrong message; a guidance note will be able to be gamed. 30. We think that what is reasonable/unreasonable today, may be unreasonable/reasonable tomorrow. It is unlikely that a guidance note will be dynamic enough to meet the changing environment. It will be unhelpful commercially to either the scheme providers and the FMA. The focus should be on total fees not individual components 31. Any guidance note should be based on principles and not rules, i.e. not have any prescriptive requirements. The only relevant test is whether the total fees charged to members are not unreasonable. This is the legislative requirement. 32. In terms of the Act, we think that it is the total fees that are important and not the individual components that make up the total. To set rules around one aspect may lead to the wrong behaviours. 33. If performance fees are to be allowed then the question has to be answered is the reasonable test relative or absolute, or a combination of each test? If it is a relative test then the test becomes a test of are the fees within ±X% of the average. The effect of this is that fees will creep up as if you are below average you will be able to raise your fees and remain reasonable and thereby increase the average. The workings of performance fees where the performance fee is looked at independently, as opposed to as part of the total, is that from time to time there will be a boost to the average and this has the potential to raise the total fees for the future. 34. As noted above, a better test is relative to what was disclosed, the clarity of the disclosure and the understanding of the members. Performance fees are less likely to meet the disclosure standard. 6

7 Other specific comments on the draft guidance note 35. Notwithstanding the above comments, in terms of the draft guideline as drafted, we note: 36. Performance fees cannot share downside unless manages fees can go below zero. They therefore can never be equitable to members. 37. The risks taken by the manager is that they do not have a successful business. Performance is only one aspect of this risk. However, as KiwiSaver is about distribution and net returns, there can be little alignment of risks with performance fees. 38. If performance fees are performance fees, there is no logic of having an annual cap. If a cap applies, it should apply to all fees not just performance fees. 39. Defining the hurdle rate is difficult. This is because the short-term correlation between equity returns and cash or inflation is weak. Even relative to the normal market index, the correlation short term can be weak. The NZX50, for example, includes property shares yet few managers have property shares in their NZ share portfolio. 40. It is not possible to determine whether one manager benefits from the payment of a fee (including a performance fee) to an unrelated party. It would be interesting to know how many managers went to a game of the Rugby World Cup and did not pay for their ticket. 41. High water marks are difficult because the formula can be manipulated. Cash flows result in different outcomes for different investors. This is particularly true for volatile asset clauses. Given that the KiwiSaver returns to individual member is still dominated by cash flow and the markets have been volatile, you will need to also provide guidance on formula as arithmetic and geometric return formula give different outcomes. 7

8 Appendix MSCI performance The return of the MSCI emerging markets index less the developed markets index (the normal index) is: 50% 38% 25% 13% 0% -13% -25% -38% -50% 1 year emerging markets returns less MSCI The return of the MSCI equally weighted index less the market cap weighted index (the normal index) is: 25% 20% 15% 10% 5% 0% -5% -10% 1 year equally weighted less MSCI The return of the MSCI value index versus the MSCI growth index is: 15% 10% 5% 0% -5% -10% -15% 3 year value less growth markets returns (MSCI) The value versus growth graph is based on three year returns to highlight that for considerable periods of time one style can be favoured over another and that to outperform for such periods is style and not skill and should not be rewarded. Similar patterns to those above apply to small shares versus large shares, non-property versus property shares etc. The return advantage is the working of the market and not the skill of a manager. 8