The inability to enter variables such as assets, age, gender, risk tolerance and desired income into financial planning software and receive an answer that gives clients their best possible combination of products is a major hindrance.cifotart/iStockPhoto / Getty Images
This is the fourth and final article in a series examining the decumulation product landscape in Canada and how advisors can explain the options to clients. Read Part 1 here, Part 2 here and Part 3 here.
The market for decumulation products in Canada has picked up in recent years. But while advisors don’t lack options – with products such as tontines, guaranteed minimum withdrawal benefits (GMWBs) and advanced life deferred annuities (ALDAs) available – we lack frameworks for decision-making and product allocation to optimize client outcomes.
The inability to enter the variables of assets, age, gender, market return and volatility, risk tolerance, desired income and other factors into financial planning software and receive an answer that gives clients their best possible combination of products is a major hindrance.
To date, the available products’ novelty, the lack of solid allocation frameworks, the behavioural factors and compensation differentials – combined with a general lack of appreciation of the challenges and mathematics of decumulation – have contributed to low adoption.
Given these limitations, how can an advisor utilize what’s available to help their clients?
The first step is to educate themselves on mortality risk, the probability of living to different ages, and the mechanics of the products available.
The second step is to identify clients who require decumulation products. That’s best handled by way of a financial plan that utilizes Monte Carlo analysis and market-based assumptions for both return and volatility. (There are resources for this available published annually by FP Canada, BlackRock, Inc., Vanguard Group Inc., J.P. Morgan Chase & Co. and other financial institutions.)
By creating and stress-testing financial plans, we can assess the probability of ruin (running out of money) at advanced ages. For those unlikely to run out of money, longevity products may still be useful of enhancing estate values potentially through access to mortality credits; for people at greater risk, the products could be essential.
How to model decumulation solutions
Annuities: These are the easiest products to test. Simply obtain quotes for annuities of different premium amounts and test them in the financial planning software by entering the annuity payment, reducing the investment value by the premium, and re-running the Monte Carlo analysis. The numbers should be clear as day. The optimal allocation will likely require testing multiple annuities and comparing the impact.
ALDAs: The same can be repeated for ALDAs, only this time, testing of various start ages may be required. Even if the financial planning software doesn’t support ALDAs yet – and most don’t – just reduce the RRSP amount by the premium and enter a new fully taxable income beginning at the start period of the annuity payment age.
GMWBs: If you’re using a new or legacy GMWB that hasn’t started making distributions, your best bet is to model it as an investment account with a higher fee/lower return and then convert it to an annuity at the desired income start date. This fails to account for the guarantee resets for higher market value, but it should be close enough on a linear basis. The limiting factor is that Monte Carlo will be useless as there’s no way to adjust the annuity payment for the market value of the account at the time of purchase.
If the GMWB is already making income distributions, then your best bet is to model it as an annuity. In this case, the Monte Carlo analysis can be used and the impact is effectively measured, but the asset and estate values will no longer be accurate as they won’t include the value of the GMWB account.
Tontines: For Purpose Investment Inc.’s Purpose Longevity Fund, the best solution is to use its projected distributions as an indexed annuity. Of course, that fails to reflect the client’s asset base, estate value and, if purchased outside of a registered account, taxable nature accurately.
For Guardian Capital’s Modern Tontine Trust, create a separate account for which the return expectation is that the account value will be the expected value of the tontine (based on Guardian’s calculator) at the end of the 20 years.
In either case, Monte Carlo analysis can’t be used effectively for tontines because volatility impacts the pools. Despite the inability to test for volatility, the adoption of either product should lead to higher probabilities of successful outcomes based on their provision of mortality credits.
What works for clients
In a few years, we’ve gone from one option for mortality risk hedging when decumulation planning to four or five (given the different types of tontines). They all have some value proposition worth considering for clients who face the prospect of running out of money in retirement.
That said, the end solution for each client will depend on a combination of the facts of their case and financial plan, their personal preferences and, in some cases, their willingness to accept their estate being smaller if they pass away early. While few like that idea, the prospect of ruin in advanced age should scare them – and their advisors – even more.
Jason Pereira is a senior partner and financial planner at Woodgate Financial Inc. in Toronto.