In our recent webinar, 87% of advisers said they were not fully confident in their ability to deliver the expected retirement outcome for clients approaching retirement.
More telling still: when the scenario shifts from a steady market environment to one involving a 30% fall, confidence drops sharply. A plan that looks workable at 7% growth can feel very different once you factor in the order in which those returns actually arrive.
This isn’t a question of competence – advisers understand the risks. The challenge is that understanding a risk is not the same as having a clear framework for managing it. That gap between awareness and a structured response is where the real problem lies.
What advisers are doing today
When things don’t go to plan, how do advisers actually respond? Our survey data reveals an instructive truth: there is no consensus.
We found that:
- 43% say they would reforecast the cashflow model.
- 27% would rely on the existing model to absorb the fall.
- 21% said it depends on the client, the circumstances, and the size of the drawdown.
- Only a small minority actively scenario plan for sequence risk before it arrives.
In practice, advisers solve the same problem in different ways, often without a consistent structure behind their decisions.
That inconsistency is understandable. Markets are uncertain, clients are different, and cashflow modelling tools have become increasingly sophisticated. But inconsistency also means that the quality of the response can vary, and in retirement planning, that variability carries real consequences.
What modelling is designed to do
Before examining where modelling runs into difficulty, it is worth being clear about what it does well – because it does a great deal.
Cashflow modelling gives advisers a structured way to test whether a retirement plan is likely to work. It allows for assumption testing across different growth rates, inflation levels, and spending patterns. It also creates a common language for adviser-client conversations, turning abstract numbers into projected outcomes that clients can engage with.
At its best, modelling answers a specific and important question:
Is this plan likely to be sustainable over time?
That is genuinely valuable to know, and for many clients, in many circumstances, it is the right tool for the job.
The issue is not that modelling fails, but that there are situations where modelling is being asked to do something it was never designed to do.
Where modelling starts to struggle
Cashflow models are built on assumptions…
- Growth rates
- Inflation
- Spending
- Longevity
All of these are inputs, and all of them are uncertain. In normal conditions, that uncertainty can be managed through scenario testing and regular reviews.
In practice, however, markets do not always cooperate as we have in recent times. The last few years have provided back-to-back stress tests: a global pandemic in 2020, followed by one of the worst years for multi-asset portfolios in recent memory in 2022. These are not freak events in the statistical sense; they are the kind of conditions that will occur again in some form.
The specific vulnerability here is sequencing.
A client who retires into a falling market and begins drawing income from a depleted portfolio faces a materially different situation from one who experiences the same average return but in a different order. The model may show identical long-run outcomes; however, the lived experience and the actual portfolio value can be very different.
Modelling can show what outcomes might look like, but it cannot control the path taken to get there. For clients close to or in retirement, that path matters enormously.
The shift in the question
This is where the nature of the problem starts to change.
In the accumulation phase, the primary question is whether a plan is likely to succeed over a long time horizon. Probability is a reasonable lens for that. A client with 20 years of saving ahead of them can absorb a significant fall – the sequence of returns matters less when there is time to recover.
When looking at the transition to retirement, the question shifts. It becomes less about predicting what markets will do and more about how a plan behaves when things go wrong. The horizon is shorter, the client is drawing income rather than contributing and tolerance for a poor sequence of returns is lower.
When the question changes, the tools used to answer it may need to change too.
The structural alternative
Structure, in this context, means something specific. It is not a different way of modelling outcomes; it is a different type of solution entirely.
Where modelling tests what might happen, structure shapes what will happen. It introduces constraints into the portfolio that reduce reliance on timing. So rather than asking whether a plan is likely to succeed, structure asks: what outcome can be defined and delivered, regardless of the market environment in which the client retires?
It’s important to recognise that this approach does not eliminate uncertainty. However, it does reframe the relationship with it. Instead of managing the probability of a good outcome, structure reduces the degree to which a good outcome depends on market conditions falling in the right order at the right time.
For clients approaching retirement – particularly those with defined income needs or lower tolerance for early losses – this distinction can be significant.
Modelling vs Structure: Two different answers
These two approaches are not in competition; they simply answer different questions.
| Modelling answers: | Structure answers: |
| Can the plan work? | What happens if markets fall at the wrong time? |
| What is the probability of success? | How much uncertainty can be removed from the equation? |
| How sensitive is the plan to different assumptions? | What outcome can be relied upon, regardless of market timing? |
The choice between them is not about which is better. It is about which problem you are actually trying to solve.
For a client with flexibility in timing, a long horizon, and comfort with volatility, modelling may provide all the information needed. Where a client has a fixed retirement date, defined spending requirements, and limited capacity to absorb early losses, probability alone may not be enough.
The mistake is not using modelling but assuming it answers both questions.
The certainty question
There is a tension in this debate that advisers recognise: certainty has a cost and not every client is willing or able to pay it.
That cost is specific…
- A growth cap
- Reduced upside participation
- A defined term that clients are committed to
Some clients will look at those trade-offs and decide they are not worth making. Feedback from our webinar reflects this split:
- A third of advisers say that offering greater certainty would involve trade-offs their clients would not accept
- Nearly two-thirds say that certainty would change client behaviour – reducing anxiety, improving decision-making, and supporting more consistent planning conversations
The interesting question is whether advisers are making that decision on the client’s behalf before the conversation has properly happened.
Advisers who say clients will not accept the trade-offs may be right, or they may be projecting.
The reason is important because the two-thirds figure suggests that when the conversation does happen, certainty changes things. Anxiety goes down, decisions improve, and planning conversations become easier.
That is not necessarily a client rejecting a growth cap – it may be a client who was never properly shown what they were getting in exchange for it.
When modelling is enough, and when it isn’t
While the underlying tools may use the latest technology, the broader practical framework for this decision does not need to be complex.
Modelling works well when clients have a long time horizon, genuine flexibility in when they access income, and the capacity to absorb short-term falls without a lasting impact on their plan. In these circumstances, probability is a reasonable guide when a well-constructed, regularly reviewed model provides sufficient confidence.
Structure becomes more relevant when timing is defined, when income needs are specific and non-negotiable, or when a client is in the transition window. Close enough to retirement that a significant fall in the early years would cause lasting damage to the plan. In these circumstances, relying on probability alone starts to feel less comfortable.
The trigger is not a number or a date, but the degree to which a client’s plan can tolerate a poor sequence of returns, and whether the tools being used actually address that.
Applying structure in practice
In practice, this often means structuring part of the portfolio around a specific outcome over a set period.
AQ Protected Portfolios (AQP) follow this approach. They combine market exposure with a capital protection profile at maturity, supported by a volatility management overlay.
The aim is not to maximise returns, but to reduce reliance on market timing at the point where it matters most.
This is not a guarantee, but it is a different type of answer to a different type of question.
For some clients, the answer is structure. For others, it is the process.
If structure is one answer to the sequence problem, AQ Global is another: a globally focused, evidence-based portfolio range built around AQ’s disciplined fund selection process, now at 12bps across the global range, reduced from 15bps.
More details on the process, portfolios and updated pricing can be found on the AQ Global page.
The real decision
In summary, the challenge for advisers is not choosing between modelling and structure, because both have a role to play. The challenge is being clear about which problem is on the table and reaching for the right tool to address it.
For clients with time and flexibility, modelling supports sound planning. For clients approaching retirement with defined needs and limited tolerance for a bad run of returns, the question shifts, and the approach may need to shift with it.
The advisers who navigate this well are not those who abandon modelling. They are those who recognise when modelling alone is no longer sufficient and who have a clear, consistent approach for what comes next.
This content is intended for financial professionals only. These are the author’s views at the time of writing and may be subject to change. This content is not intended to provide the basis for any investment advice or recommendation. Any forecasts, figures, opinions, tools, strategies, data, or investment techniques are included for information purposes only.
The information presented is considered to be accurate at the time of production and has been obtained from or based upon sources believed by the author to be reliable and accurate, but no warranty of accuracy is given and no liability in respect of any error or omission is accepted. Please visit our Regulatory Information and Terms of Use pages for more information.
Production