Can actuaries develop a science of planning that provides a basis for the art of giving advice? Here, John De Ravin from the Actuaries Institute’s Retirement Incomes Working Group (RIWG), details the findings of a survey of 65 financial planners from a range of different business environments, and subsequent interviews with dealer groups, to uncover the state of play in industry practice.
Personal financial advice is not a leading area of actuarial employment, however some actuaries are in fact actively practising in this area. Others also contribute significantly to research in the area. The advice landscape is changing; already there are personal financial apps receiving a lot of attention. There is also a trend towards “roboadvice” in various guises. But “roboadvice” needs a mathematical and a behavioural foundation; actuaries should be well placed to make a worthwhile contribution to these developments.
With this in mind, Jim Hennington, Paul Scully and I, of the Institute’s Retirement Incomes Working Group (RIWG), have been conducting research into the current financial planning process. Key research areas include:
- What process is used to prepare advice to clients in the area of asset allocation?
- Who designs the approach to asset allocation advice and retirement capital management, and to what extent is the role of the individual adviser constrained by the employer/licensor?
- What happens if the client’s risk profile is not optimal for achieving the client’s retirement goals?
- What retirement duration is planned for and how is that duration determined?
- What methodologies are used to advise clients on retirement capital and spending?
- Where Statements of Advice include more than one projection of retirement assets, how are those other projections determined and characterised?
The RIWG conducted a survey of 65 financial planners from a range of different business environments, and to date we have also carried out four structured interviews with dealer groups to discuss the same questions asked in the survey. The results had some interesting messages.
Risk Profile Questionnaire usage
Many readers of Actuaries Digital will be familiar with the traditional processes of providing financial advice in relation to asset allocation: ask the client to complete a Risk Profile Questionnaire (RPQ) designed to measure the client’s risk tolerance, then discuss the results with the client and map the resulting risk tolerance to an asset allocation. If the adviser wishes to recommend an asset allocation other than the allocation that flows from the RPQ, there may be some compliance rules that govern the conditions under which a departure may be recommended. These compliance rules are intended to protect both the client (against poor advice) and the dealer group (against possible legal liability if the recommended portfolio does not produce the investment returns the client expected).
Firstly, it became clear, in our research, that RPQs are still dominant, at least as the initial stage of determining the appropriate asset allocation - 83% of respondents were making use of RPQs. They are of varying lengths, from fewer than six questions (7%) to more than 25 (5%), with most (61%) being between six and 15. But the asset allocation that emerges from “adding up the points” in the RPQ is not necessarily the one that is recommended to the client: 48% of respondents said that the asset allocation determined from the RPQ was only used as a starting point. A further 10% also said they tended to use other approaches.
“…across all advisers and clients, we would guess that the advisers hold a view that the asset allocation emerging from the RPQ is appropriate roughly half the time.”
Although the survey showed that RPQs are still a key tool in the industry, the structured interviews suggested that some dealer groups are moving away from RPQs as a critical component of asset allocation advice. For example, some dealer groups allow their more senior advisers, or advisers with a very longstanding relationship with the client, to proceed without an RPQ. Other dealer groups are moving towards a process by which clients are provided with some educational information about the way in which different asset classes tend to perform (and the consequences of the alternative allocations) and the clients are then asked to select their own asset allocation (after discussion with their adviser).
Compliance and variation – is it worth it?
A key point of interest is the impact of compliance and the extent to which a corporate approach is imposed on advisers. Asked how difficult it was, from a compliance perspective, to recommend an asset allocation that is not completely consistent with the RPQ, one of 58 respondents said “very difficult or not possible”. Equal numbers (19) opted for each of the following three responses:
- “There is extra compliance effort, but it is possible”;
- "I have some discretion to vary slightly from the RP result”; and
- “I can recommend any asset allocation I think is best for the client”.
Some of the comments by the planners who responded: “there is extra compliance effort but it is possible” were illuminating: one said, “why bother – too much compliance and you just get slammed at audit” and another, “there has to be a very good reason for the variation for compliance to be happy”.
The responses to some later questions are also relevant to compliance. On the question of who designs the retirement approach, 45% of respondents said that they had full or significant discretion over the methodologies they use, and another 46% replied, “my employer/dealer group provides a preferred methodology that I TEND TO stick to”. Only 9% replied that their employer/dealer group provided a methodology that they “MUST stick to”.
It appears that compliance is seen as a restricting factor for a minority of planners, but that many enjoy discretion in the approach to take. At this stage, there is no consensus on whether there should be one approach (at least to some issues) let alone what that approach should be. Perhaps it can be characterised as whether we can develop a science of planning that will provide a basis for the art of giving advice?
Advisers tended to think that the asset allocation that emerged from the RPQ was appropriate for the client “reasonably often” (39%) or “frequently” (29%), but some advisers responded “rarely” (5%) or sometimes (20%). Whilst we cannot exactly quantify terms such as “reasonably often”, across all advisers and clients, we would guess that the advisers hold a view that the asset allocation emerging from the RPQ is appropriate roughly half the time.
Where the asset allocation that emerges from the RPQ is considered by the adviser to be sub-optimal, the majority of advisers (54%) discuss the situation with the client and attempt to persuade the client to adopt a different asset allocation but try to stay within one risk band of the RPQ result. Another 16% have the discussion and try to stay within two risk bands of the RPQ result. Yet another 16% ignore the risk profile as long as the client is prepared to sign a statement confirming the asset allocation that is discussed and agreed on. But 14% stick closely to the asset allocation that flows from the RPQ.
Life expectancy, the planning horizon and withdrawal rates
Financial plans are generally constructed for a duration related to life expectancy, with 17% being for average or median life expectancy, 29% for life expectancy plus a small margin (one to five years), 19% for life expectancy/median plus a larger margin of six to ten years, and 7% being to a fixed age (generally 90 or 100). The survey did not request information about how the margin over life expectancy was determined. Some respondents noted that the standard planning software packages used population mortality unadjusted for future improvements in mortality and the survey did not question how a client’s own state of health was taken into account. Asked what they would do if their client could only achieve their target retirement lifestyle if high asset returns were achieved in future, 31% said that they would encourage the client to reduce their target and goals to align with what they can afford based on their asset levels. But 56% said that they would advise some combination of reduced target expenditure and a greater percentage asset allocation to growth assets. A further 13% said that they would invest according to the client’s risk profile and let the client make their own spending decisions.
Advisers were asked how they determined their advice about spending to their most recent retiree client. Just over a quarter (27%) stick to the SIS minimum pension withdrawal rates, and a further 11% use a “safe withdrawal rate” approach. More than a third (35%) use a bucket strategy approach, investing in different asset classes for the three periods of retirement. Of respondents, 15% said that they had used a “floor and upside” approach by investing in a guaranteed income stream with part of the client funds and investing the remaining investments in relatively riskier portfolio. Twenty-two respondents (40%) said that they simply accept what the client says about the client’s income needs and focus on portfolio construction to produce that amount of income.
Just over half (51%) of respondents said that their Statements of Advice (SOAs) do not contain alternative projections, they are solely based on the main estimate. But 16% said that they also included “optimistic” and “pessimistic” projections without much detail on how the assumptions for those alternatives were determined. However 15% of SOAs included alternative projections that were characterised in quantitative terms and 11% showed confidence intervals that were determined from stochastic models.
The next steps in the research project are:
- to continue an interview process with a select number of compliance and policy representatives of some of the financial advice firms;
- to prepare a paper for the 2018 Financial Services Forum; and
- ultimately, to make a worthwhile contribution to financial planning practice, particularly in a retirement incomes advice context.
More detail about the survey results, and many individual comments to the questions by survey respondents, may be found here.
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivatives CC BY-NC-ND Version 3.0 (CC Australia ported licence).
CPD Actuaries Institute Members can claim two CPD points for every hour of reading articles on Actuaries Digital.