Personal essays, notes, and ideas worth sharing.
This is my corner of the internet. I write about whatever I’m thinking about at the moment. Books that stuck with me. Projects I’m working on. Random ideas that won’t leave me alone. Sometimes I share bits of my actual life too. There’s no grand theme here, no carefully planned content strategy. Just honest writing about things that matter to me.
Topics we cover: Essays · Projects · Reading Notes · Ideas · Personal
Most people read a suitability report the way they read a terms-and-conditions email: scroll to the bottom, check the number, sign. That is the wrong order. A suitability report is not a receipt for advice already given. It is a decision document, and the rules that govern its contents tell you exactly where to look for the parts that matter.
This is a reading method, not a reassurance. It will not tell you whether a recommendation is right for you. It will tell you whether the firm has done the job the rules require, and where the gaps are.
Under COBS 9.4.1R, a firm must provide a suitability report to a retail client when it makes a personal recommendation and the client acquires or sells a holding in a regulated collective investment scheme or certain investment trusts, buys or sells rights under a personal or stakeholder pension, elects to make income withdrawals or an uncrystallised funds pension lump sum payment, or enters into a pension opt-out. Life policies and pension transfers or conversions carry their own requirements.
Timing is not a detail. COBS 9.4.4R sets out when the report must reach you:
If a pension transfer is being recommended and the report arrives after the transfer has been executed, that is not a stylistic complaint. It is a timing failure against a rule that exists precisely so you can read the reasoning before the money moves.
COBS 9.4.7R requires the report to contain, at least:
Read them in that order. Demands and needs first, because everything else is measured against it. Suitability explanation second, because it should connect the recommendation to what you said you wanted. Disadvantages third, because this is where the real trade-offs live. Personalised recommendation last, and only if a life policy is involved.
COBS 9.4.8R adds that the detail must be modulated according to the complexity of the transaction and the type of client. A one-page report for a straightforward ISA top-up is not the same document as a report for a pension transfer. If the complexity of your situation is high and the report is thin, that mismatch is worth naming.
The demands and needs section is only as good as the information the firm was required to gather. COBS 9.2.1R requires a firm, when making a personal recommendation, to obtain the necessary information regarding your knowledge and experience in the investment field, your financial situation, and your investment objectives.
COBS 9.2.2R goes further. Information about your investment objectives must include, where relevant, the length of time you wish to hold the investment, your preferences regarding risk taking, your risk profile, and the purposes of the investment. Information about your financial situation must include, where relevant, the source and extent of your regular income, your assets including liquid assets, investments and real property, and your regular financial commitments.
So the test is simple. Does the demands and needs section reflect those categories? If you told the adviser you wanted to stop work at 60 and bridge to State Pension age, does the report say so? If you said you wanted to keep £150,000 in cash and not touch the pension until 67, is that in there? If the section reads like a template that could apply to anyone with a pulse and a pension, it is not doing its job.
COBS 9.2.5R allows a firm to rely on information you provide unless it is aware the information is manifestly out of date, inaccurate or incomplete. That cuts both ways. If you gave incomplete information, the report may be built on a weak foundation. If you gave full information and the report ignores it, the problem is the report.
This is the section most likely to be written in generic language, and the one that matters most. COBS 9.4.7R(3) requires an explanation of any possible disadvantages of the transaction for you. Not a disclaimer. Not a list of things that could go wrong in the abstract. The disadvantages of this transaction for you.
For income withdrawals, short-term annuities, or uncrystallised funds pension lump sum payments, COBS 9.4.10G gives a useful checklist. The explanation of possible disadvantages should include the risk factors involved, which may include:
That is a minimum, not a maximum. A disadvantages paragraph that does not engage with at least some of these is thin. A disadvantages paragraph that says only that ‘investments can go down as well as up’ is not an explanation of the disadvantages of the recommended course of action. It is a warning label.
Read it with a pen. If you are drawing £40,000 a year from a £600,000 pension at 58, does the report say what happens if returns are 2% rather than 5%? Does it say what happens if you need to stop the withdrawals and buy an annuity at 70? Does it say what the tax position looks like if you die before 75 versus after? Those are the disadvantages that matter. If they are absent, ask why.
If the recommendation is a pension transfer or conversion, COBS 9.4.11R requires a one-page summary at the front of the suitability report, except where the only safeguarded benefit involved is a guaranteed annuity rate. That summary is not a substitute for the full report. It is a signpost, and it has specific required contents.
The summary must include a summary of the personal recommendation, a statement as to whether the advice is abridged or full pension transfer or conversion advice, information about any ongoing advice or services proposed after execution, the risks associated with pension transfers or conversions as set out in COBS 19.1.6G(4)(b), and an invitation to consider whether you fully understand those risks and, if so, sign the summary to confirm that.
It must also include all ongoing advice charges, all other ongoing charges, and any additional charges expected to be incurred if you proceed, together with a comparison to the charges and revalued monthly income in the ceding arrangement and to the charges in any default arrangement in any available qualifying scheme. And it must include the amounts payable in cash terms for the initial advice, and the number of months, rounded up to the nearest whole month, it would take to pay that amount out of the revalued monthly income you would receive from the ceding arrangement.
Read the summary first. It should tell you the recommendation, the charges in cash terms, and the risks. Then read the full report to see whether the reasoning supports the summary. If the summary says one thing and the body of the report says another, that is a problem worth raising before you sign anything.
COBS 9.4.11R(5) adds that the information about ongoing advice must set out that you are not required to accept ongoing advice or services, that you can opt out at any time, and the monthly and annual charges in cash terms. If the report presents ongoing advice as a condition of the recommendation, that is not what the rule says.
Before you sign or proceed, work through these questions. They are not a substitute for advice. They are a way of testing whether the document in front of you is doing the job the rules require.
A suitability report is a regulatory document. It tells you what the firm was required to disclose and explain. It does not tell you whether the recommendation is right for you. That judgement remains yours, and the decision to seek a second opinion is yours too.
The rules are a floor, not a ceiling. A report can comply with COBS 9.4.7R and still be a poor basis for a decision if the demands and needs section is thin, the disadvantages paragraph is generic, or the suitability explanation does not engage with the real trade-offs in your situation. Compliance is not the same as quality.
If you are pre-retirement, with £300,000 or more across pensions, ISAs and property, the decisions the report is describing are not small. The timing of your stop-work date, the bridge to State Pension age, the order in which you draw on different wrappers, the tax consequences of each choice — these are the things that will determine whether the money does what you need it to do. The suitability report is one input into those decisions. It is not the decision itself.
Read it backwards if you like. Start with the disadvantages, then check the demands and needs against what you actually said, then read the suitability explanation to see whether it holds together. If it does, you have a document you can use. If it does not, you have a question worth asking before you sign.
For a pension transfer or conversion, COBS 9.4.4R requires the report in good time before the transaction is effected. For a life policy, before the contract is concluded. For other transactions, the timing rules are different, but you can always ask for the report before you sign anything. If the firm resists, that is information.
COBS 9.4.7R(3) requires an explanation of any possible disadvantages of the transaction for you. Generic warnings do not meet that standard. For income withdrawals, COBS 9.4.10G gives specific risk factors that should be addressed. If they are not, ask the firm to explain the disadvantages of the recommended course of action in your circumstances.
No. COBS 9.4.11R requires the one-page summary at the front of the suitability report. It is a signpost, not a substitute. The full report should contain the reasoning that supports the summary.
For a pension transfer or conversion, COBS 9.4.4R requires the report in good time before the transaction is effected. If it arrives after, that is a timing failure. For other transactions, the timing rules differ. If you are concerned, ask the firm to explain when the report was provided and why.
No. The report tells you what the firm was required to disclose and explain. It does not tell you whether the recommendation is right for you. That judgement is yours, and a second opinion is always an option.
For a related question on the decisions that sit behind the report — the ones you should be asking years before you retire — see The Question You Should Be Asking Years Before You Retire.
“You can retire at 60.”
That sentence is not a fact. It is an output. Somewhere behind it sits a spreadsheet or a planning tool with a column of numbers, and those numbers were chosen by someone — possibly you, possibly an adviser, possibly a default setting nobody has looked at since the file was created. The output is only as trustworthy as the weakest assumption feeding it.
This article is not about whether you personally can retire at 60. It is about how to read the model that claims to know. The useful question is not “what does it say?” but “what would have to be true for this to hold, and how would I know if it stopped being true?”
Any retirement cash-flow model contains three kinds of input, and they deserve very different levels of trust.
Layer one: verified external facts. These are rules and data published by official sources. They constrain what is possible. They do not determine whether your retirement date works.
Layer two: personal inputs only you can supply. Your spending, your stop-work date, your balances, your State Pension forecast. These are checkable in principle, but often estimated loosely in practice.
Layer three: forward-looking guesses. Investment returns, inflation paths, future tax rules, care costs, how long you live. These are the assumptions most likely to be wrong, and they are usually presented with the most confidence.
The uncomfortable truth is that the model’s answer is dominated by layer three, and layer three is the part nobody can verify in advance.
Start with what can be looked up. It is less than you might hope, but it is solid.
State Pension eligibility. You need 10 qualifying years on your National Insurance record to get any new State Pension at all, and the qualifying years on your record affect how much you get. Eligibility for the new State Pension depends on your date of birth: broadly, a man born on or after 6 April 1951 or a woman born on or after 6 April 1953 falls under the new State Pension; those born earlier fall under different rules. These are the constraints. They tell you whether you get a State Pension and roughly how the amount is built. They do not tell you what it will be worth in 2045, because uprating is a future policy decision. Source: GOV.UK, The new State Pension.
Longevity. ONS national life tables for 2022 to 2024 put life expectancy at age 65 in the UK at 21.2 years for females and 18.7 years for males. That is period life expectancy — it describes the average number of years people would live if current mortality rates applied throughout their lives. It is not a prediction for you. ONS also publishes cohort projections, which attempt to account for future improvements in mortality: girls born in the UK in 2024 can expect to live to age 90.2 years on average, and boys to age 86.9 years, on that basis. Source: ONS, Life expectancies.
Here is the trap. A model that runs to age 90 for a 60-year-old is not being conservative. It is being roughly average for one person, and it is almost certainly too short for a couple, because the probability that at least one of two people lives past the average is high. If your model assumes both partners die at 90, it is quietly assuming a joint outcome that is less likely than the individual outcome it was drawn from.
Inflation. The Government has set the Bank of England a target of keeping inflation at 2%. That is a target, not a forecast, and not a guarantee. The Bank’s own framework acknowledges deviation: if CPI inflation is more than 3% or less than 1%, the Governor writes to the Chancellor to explain why and what will be done to return it to target. Source: Bank of England, Inflation and the 2% target.
So when a model uses 2% inflation for thirty years, it is using the target as though it were the outcome. That is a choice, not a fact. A model that cannot show you what happens at 3% or 4% is not stress-tested; it is decorated.
The checkable layer tells you the rules of the game. The answer to “can I retire at 60?” is decided in layer three, and specifically by four assumptions that are rarely shown as ranges.
Most models use a single average return, often somewhere between 4% and 6% nominal for a balanced portfolio. That number is not the problem. The problem is the order in which returns arrive.
Two portfolios can have identical average returns over thirty years and produce completely different outcomes if one suffers a poor first decade while the other enjoys a strong one. This is sequence-of-returns risk, and it bites hardest in the first ten years of drawdown, when withdrawals are converting a temporary loss into a permanent one.
A model that shows only the average return is not showing you the risk. Ask for the path: what happens if the first five years deliver zero or negative real returns, and withdrawals continue as planned?
If the model assumes a fixed real withdrawal — say £40,000 a year, rising with inflation — it is assuming a spending pattern that most real retirees do not follow. Spending tends to decline in real terms through the middle of retirement and rise again if care is needed. A model that holds spending flat and rising may be overstating the required pot, or understating the late-life cost, depending on which end you look at.
More importantly: does the model allow spending to be cut if returns disappoint? A plan with a 4% withdrawal rate and no flexibility is fragile. A plan with a 4% starting rate and a stated rule — “if the portfolio falls below X, we cut discretionary spending by Y” — is a different proposition. The flexibility is the plan.
A 1% annual charge on a £600,000 portfolio is £6,000 a year. Over thirty years, with no growth, that is £180,000. With growth, the opportunity cost is larger, because the fee is levied on a rising balance and the money removed would itself have compounded.
Models often bury fees in a net return assumption. That is convenient and opaque. Ask for the gross return, the fee, and the net return as separate lines. If the person who built the model cannot produce them, the model is not a planning tool; it is a black box.
If you stop work at 60 and State Pension age is 67 or 68, you have a bridge period of seven or eight years with no State Pension income. The model must fund that bridge from pensions, ISAs, or other assets, and it must decide how.
This is where wrapper decisions and tax assumptions meet. Drawing from a pension before State Pension age may be tax-efficient or not, depending on your other income and the allowances in force at the time. Drawing from an ISA is tax-free but depletes a wrapper you may want later. The model’s answer depends on assumptions about tax rules that do not yet exist for the later years of the plan.
No retrieved source in this article supports specific tax rates or allowances for future years, and none should be invented. The honest position is that the model is using current rules as a placeholder for unknown future rules. That is reasonable, provided it is labelled as such.
The following is not a client case and not a research finding. It is a hypothetical to show how one assumption moves the answer.
Suppose a couple, both 58, with £700,000 across pensions and ISAs, want to stop work at 60 and spend £45,000 a year in today’s money. A model using a 5% nominal return, 2% inflation, and a 30-year horizon might show the pot surviving to age 90 with room to spare.
Change one assumption: the first five years deliver a 15% real fall, and withdrawals continue. The same model, run with that path, may show the pot exhausted in the late 80s, or require spending to drop to £38,000 for a decade to recover. The average return is unchanged. The answer is not.
Change another: inflation runs at 3.5% rather than 2% for the first decade. The required pot rises materially, because the spending line rises faster than the return line. Again, the average return assumption is untouched.
This is why a single number — “you can retire at 60” — is not a plan. It is one path through a range of possible paths, and the range is the useful information.
These questions are not adversarial. They are the minimum standard for treating an output as a decision input.
If the model cannot answer question seven, it is not a plan. It is a projection, and projections are not decisions.
The assumptions most likely to decide your retirement date are the ones you cannot verify. Returns, inflation, tax rules, care costs, and how long you live are all guesses. Some are educated guesses. None are facts.
That does not make modelling useless. It makes the model a tool for asking better questions, not for producing a verdict. The verdict — “yes, 60 works” — is a judgement about how much uncertainty you can tolerate and how flexible you are willing to be. No spreadsheet can make that judgement for you.
What the spreadsheet can do is show you the trade-offs. If you want to retire at 60, here is the spending level that works at a 4% return and the spending level that works at 3%. Here is the pot you need if you live to 85 and the pot you need if one of you lives to 95. Here is what happens if fees are 0.5% and what happens if they are 1.2%.
Those are decisions. The single sentence is not.
The cash-flow model is one decision moment among several. It sits alongside the adviser relationship audit — what you are paying, what the suitability report actually says, whether the firm appears on the FCA Register — and the family paperwork that quietly decides what happens to the money: expression of wish forms, lasting powers of attorney, gifting, and care funding.
None of those are substitutes for each other. A model that says “retire at 60” is not a substitute for an expression of wish form that says where the pension goes if you die at 62. A withdrawal plan is not a substitute for an LPA that lets your spouse manage the accounts if you cannot.
If you are at the stage of interrogating the model, the related question — the question you should be asking years before you retire — is worth reading alongside this. It is about the decision that precedes the spreadsheet: what you are actually retiring to, and what the money is for.
No. The 2% figure is the Government’s target for the Bank of England, not a forecast. A model can use it as a central assumption, but it should also show what happens at higher and lower rates. If it only shows 2%, it is not showing you the range of outcomes.
There is no single correct answer. ONS period life expectancy at age 65 was 21.2 years for females and 18.7 years for males in 2022 to 2024, but that is an average, not a maximum. For a couple, the relevant question is the probability that at least one person lives well beyond the average. A model that runs to 90 for both partners may be understating the joint longevity risk.
Ask for the assumptions. If they are shown as single point estimates with no ranges, the model is not giving you enough information to decide. If they are shown as ranges and the conclusion holds across most of them, the model is doing its job. If the conclusion only holds at the optimistic end, you have your answer.
Yes, provided you treat it as a tool for exploring trade-offs rather than a source of truth. The value is in the questions it forces you to answer: what you spend, when you stop, how flexible you are, and what happens if you are wrong.
You have spent twenty-five years building a portfolio and a habit of asking who is accountable when something goes wrong. A robo-advisor asks you to hand over the same money to a questionnaire and an algorithm, then accept that the accountability sits with you. That is the trade. It is not automatically a bad one, but it needs to be priced, tested, and documented before you commit a six-figure ISA or a seven-figure pension to it.
This is the fourth decision moment in the pre-retirement sequence I write about: auditing the adviser relationship, timing the stop-work date, decumulation and risk psychology, tax-aware wrapper decisions, and the family paperwork. A robo-advisor sits inside the first and third of those. It changes who you can call, what you can evidence, and what happens when markets fall 20% in the year you planned to stop working.
A robo-advisor is a regulated investment service that builds and manages a portfolio from your answers to an online risk questionnaire, usually through a platform, with fees taken as a percentage of assets. In the UK, the main names include Vanguard Investor, Nutmeg (owned by J.P. Morgan), Wealthify, Moneyfarm, and InvestEngine. Some are discretionary fund managers; some are advisory; some are execution-only with a model portfolio bolted on. The distinction matters because it decides whether you have a right to complain about suitability, not just about execution.
What it is not: a fiduciary. The UK does not use that term in the way US readers expect. A firm can be independent, restricted, or execution-only. Only independent advisers must consider the whole market. A robo-advisor is typically restricted to its own funds or a short panel. That is not a scandal. It is a business model. But if you want someone to blame, you need to know which regulatory category you are dealing with before you sign, not after.
Check the firm on the FCA Register. Look for the permissions: advising on investments, arranging deals, managing investments. Look at the status: authorised, appointed representative, or in liquidation. Look at the date permissions were granted. A firm authorised in 2019 with a clean record is a different proposition from one authorised last month with a change of control pending. The register is free and takes four minutes. Do it before you read the marketing.
If the service is advisory or discretionary, you should receive a suitability report. Read it as a legal document, not a brochure. It should state your objectives, your attitude to risk, your capacity for loss, the recommended portfolio, and the charges. If it does not mention capacity for loss, it is incomplete. If it recommends 80% equities to a 58-year-old planning to stop work at 62 and draw 4% a year, ask what happens in a 30% drawdown. The report should answer that.
If the service is execution-only, you will get a risk questionnaire and a key information document. That is the whole accountability chain. You chose the portfolio. The firm executed it. If you want someone to blame, you are the someone. That is not a reason to avoid it. It is a reason to keep your own records.
A 0.25% platform fee on £400,000 is £1,000 a year. A 0.75% managed fee is £3,000. Add fund ongoing charges of 0.15% to 0.35% and you are at £3,600 to £4,400 a year before any advice charge. Over ten years, assuming flat markets, that is £36,000 to £44,000. Markets are not flat, but the arithmetic is the point. Percentages hide the number that leaves your account.
Compare that with a single global tracker at 0.15% all-in on a cheap platform: £600 a year on £400,000. The gap is the price of the questionnaire, the rebalancing, the tax wrappers, and the phone line. Decide whether you are buying those things or just the brand.
Ask how the service rebalances. Inside an ISA or SIPP, rebalancing is free of capital gains tax. In a general investment account, it is not. A robo-advisor that rebalances across all accounts without regard to the wrapper is creating tax events you did not ask for. The better services rebalance inside wrappers first and use new contributions to correct drift in taxable accounts. Ask the question directly. If the answer is vague, assume the worst.
Also ask about asset location. If you hold bonds and equities, the bonds belong in the ISA or SIPP where interest is sheltered, and the equities belong in the general account where capital gains can be managed. Most robo-advisors do not do this. Some do. It is worth 0.2% to 0.4% a year in tax terms for a £500,000 portfolio. That is real money.
Accumulation is easy. Decumulation is where robo-advisors are weakest. Ask three questions. Can you set a regular withdrawal? Can you specify which wrapper the withdrawal comes from? Can you see the projected depletion date under different return assumptions? If the answer to any of those is no, the service is an accumulation tool, not a retirement tool. You will need to move the money or build a spreadsheet alongside it.
This is the point where the robo-advisor meets the decumulation decision moment. A 4% withdrawal rate on a 60/40 portfolio has a reasonable historical success rate over 30 years, but the sequence of returns in the first five years matters more than the average. A robo-advisor that cannot model a bad first five years is not helping you plan. It is helping you invest.
If the firm fails, the Financial Services Compensation Scheme covers investments up to £85,000 per person per firm. That is the limit. If you hold £400,000 with one robo-advisor and it goes bust, you are exposed above £85,000. The assets should be ring-fenced in custody, but ring-fencing is a legal argument, not a guarantee. Spreading across two providers is cheap insurance. It also gives you a comparison point.
If you have a complaint, the route is: the firm’s internal complaints process, then the Financial Ombudsman Service. The FOS can award up to £430,000 for complaints about acts or omissions after 1 April 2019. That is a real accountability mechanism, but it only works if you have evidence of what you were told. Keep the suitability report, the questionnaire answers, and the annual reviews. If the firm cannot produce them, that is itself a finding.
You want someone to blame because blame is a proxy for recourse. If the money falls, you want a person, a firm, a regulator, a process that says: this was not your fault, and here is the remedy. A robo-advisor removes most of that. It replaces it with a documented process and a complaints route. That is not nothing. It is just not the same thing.
The trade is explicit. You give up the relationship and the discretionary judgement. You gain lower cost, consistent process, and no sales pressure. For a £300,000 portfolio, the cost difference between a 1% adviser and a 0.3% robo-advisor is £2,100 a year. Over twenty years, that is £42,000 before growth. That is the price of the person to blame. Some people should pay it. Some should not. The decision is yours, but it should be made with the number in front of you.
There is a middle path. Use a robo-advisor for the accumulation phase and a fixed-fee adviser for the decumulation plan. Or use an adviser for the one-off suitability report and a robo-advisor for execution. The FCA’s guidance on investment platforms sets out what you should expect. The MoneyHelper investing basics page is a useful sanity check on risk and charges. Neither will tell you which firm to use. Both will help you ask better questions.
Pick three services. For each, do the following.
Step 1. Check the FCA Register. Note the permissions, the status, and the date. Note the firm reference number.
Step 2. Find the fee schedule. Write down the platform fee, the managed fee, the fund ongoing charge, the dealing fee, the withdrawal fee, and the transfer-out fee. Add them up on a £400,000 portfolio. Write the annual pound figure.
Step 3. Request a sample suitability report or key information document. Read the risk section. Look for capacity for loss. Look for the withdrawal modelling.
Step 4. Ask the three drawdown questions by email. Keep the reply.
Step 5. Check the FSCS limit and decide whether you need two providers.
Step 6. Write one paragraph explaining, to yourself, who is accountable if the portfolio falls 25% in year one. If the answer is ‘me’, accept that and move on. If the answer is ‘them’, you need an adviser, not a robo-advisor.
This is the same discipline I set out in the question you should be asking years before you retire. The question is not ‘which product’. It is ‘who decides, who documents, and who pays when it goes wrong’.
The robo-advisor decision touches the adviser audit, because it changes the fee and the accountability. It touches decumulation, because the withdrawal mechanics are where most services are thin. It touches tax wrappers, because asset location and rebalancing inside ISAs and SIPPs are worth real money. It touches the family paperwork only indirectly, but the expression of wish form and the lasting power of attorney still need to name someone. If the robo-advisor is the only counterparty, the paperwork still has to work.
Two dates to keep in view. From April 2027, unused pension funds will be subject to inheritance tax on the death of the member, subject to the legislation as drafted. From April 2028, the normal minimum pension age rises to 57. Neither changes the robo-advisor decision directly, but both change the wrapper logic around it. If you are 55 now and planning to access a SIPP at 57, the robo-advisor needs to support that. If you are 60 and planning to leave the pension untouched for inheritance reasons, the robo-advisor needs to support that too. Ask before you transfer.
Almost always, on a like-for-like portfolio. A typical robo-advisor charges 0.2% to 0.75% a year plus fund charges. An adviser charges 0.5% to 1% plus fund charges, often with a minimum fee. On £400,000, the gap is £1,200 to £3,200 a year. The question is whether the adviser’s suitability report, tax planning, and decumulation modelling are worth that. For some people they are. For others, the robo-advisor plus a one-off fixed-fee review is the better structure.
Yes. If the firm is authorised by the FCA, you can complain to the firm and then to the Financial Ombudsman Service if you are unhappy with the response. The FOS can award up to £430,000 for complaints about acts or omissions after 1 April 2019. The limit is lower for earlier complaints. You need evidence of what you were told, so keep the suitability report and the questionnaire answers.
Your assets should be held by a third-party custodian, separate from the firm’s own money. If the firm fails, the assets should be transferable. If there is a shortfall, the Financial Services Compensation Scheme covers up to £85,000 per person per firm. Above that, you are relying on the custody arrangements and the legal process. Spreading across two providers reduces the concentration risk.
Some do, many do not. The key features are regular withdrawals, wrapper-specific withdrawals, and depletion modelling under different return scenarios. If the service cannot show you a projected balance under a bad first five years, it is an accumulation tool. You can still use it, but you will need to manage the withdrawal plan yourself or with an adviser.
Almost certainly not without regulated advice. Defined benefit transfers above £30,000 require advice from a pension transfer specialist. A robo-advisor will not provide that. The decision is about guaranteed income versus flexibility, and the numbers are specific to your scheme. This is one area where the person to blame is worth paying for.
Pick one robo-advisor you are considering. Run the six steps above. Write the annual pound cost on a £400,000 portfolio. Write the answer to the accountability question. If the cost is under £1,500 and the accountability answer is ‘me, with a documented process’, the robo-advisor is a reasonable choice. If the cost is over £3,000 and the accountability answer is ‘them’, you are paying adviser prices for a robo-advisor service. That is the trade to avoid.
If you want to go deeper on the decumulation side, the next article in this sequence covers the withdrawal order and the sequence-of-returns problem. That is where the robo-advisor decision either holds up or falls apart.
A retirement plan can sound like a tidy story before anyone has checked the documents. Here is a hypothetical version: We sell the house in year three, a private pension bridges the gap, and the State Pension arrives when we expect it. Each step might be possible. Each needs its own date, amount, scheme rule, and assumption checked.
A coherent story is easy to remember; paperwork is less tidy. The house sale, the pension bridge, and the State Pension forecast may fit together on paper only after their dates and amounts are checked. Treat each sentence as a claim to test, not a promise that the next event will happen on schedule.
This piece is about how to treat your financial plan the way a professional writer treats a manuscript: not as a finished story to be admired, but as a draft to be audited for continuity errors.
Put a property estimate, a pension statement, and a State Pension forecast on one table and it is tempting to draw a straight line between them. Yet the documents answer different questions. A property estimate is not net sale proceeds; a pension statement has scheme-specific conditions; a State Pension forecast shows an individual amount and date. The story becomes useful when those differences are visible.
A gap in a National Insurance record can affect the State Pension amount, while the pension age must be checked separately. A defined benefit statement may show terms that differ from an assumed transfer or retirement date. An old expression-of-wish form may no longer reflect a person’s intentions. These are examples of what to check, not reports of particular clients or a prediction that every document contains an error.
There’s a second reason the tidy story survives: nobody wants to revise it. Revising a retirement narrative means admitting uncertainty about the thing you’ve been looking forward to for a decade. So we don’t revise. We re-read. We tell the story more confidently, which is not the same as checking it.
For couples, there may be two versions of the plan. One partner may expect a defined benefit pension to be the stable base; the other may assume both will stop work in the same summer. Until both versions are written down, the differences can stay hidden. Comparing the two accounts with the pension and property documents is a practical conversation, not a test of whose memory is better.
Externalise the story: write it in five sentences together, then check each sentence against a named document. A mismatch found early gives the household time to ask the scheme, provider, or adviser what it means before making a withdrawal or sale decision.
Here is the exercise. Write the retirement story in five sentences, then match each sentence to a document. These are common types of assumption worth checking:
“The State Pension arrives at 67.” Check your own age and amount using the GOV.UK State Pension forecast, then review the National Insurance record it shows. The age and amount are personal; a gap in qualifying years can affect the amount without changing your State Pension age. The full published rate is not a substitute for your forecast.
“The pension covers the gap until then.” Which scheme, accessed when, and on what terms? If a defined benefit pension is involved, compare the benefits under its rules with any proposed transfer rather than treating a transfer value as an equivalent income. If a drawdown plan is involved, test what happens if losses arrive early while withdrawals continue. Long-run return averages do not show the order in which returns arrive. A personal bridge plan needs its own figures and professional review.
“We sell the house and that funds the later years.” Compare the hoped-for price with plausible net proceeds after the next purchase and moving costs. The date is also a decision, not a fixed plot event: a sale can be delayed or abandoned. Test the plan both with and without that transaction at the expected time.
“And then everything passes to the children.” Check the will, each pension scheme’s rules, and current beneficiary nominations separately. A nomination is important, but the provider or trustees may have discretion under the scheme rules; an expression-of-wish form does not automatically override every other document. GOV.UK explains how pension death benefits can be paid. HMRC’s Finance Act 2026 technical note says most unused pension funds and pension death benefits will come within the estate’s Inheritance Tax value for deaths on or after 6 April 2027, subject to the measure’s scope and exceptions. Ask a qualified adviser how those rules apply to the actual schemes and family circumstances.
Here’s where the writing analogy earns its keep, because professional storytellers solved this problem long before behavioural finance named it. A novelist who can’t see their own plot holes doesn’t fix the manuscript by writing more prose – more prose buries the inconsistency deeper. They fix it by changing the process: proof sheets, beat sheets, continuity checks, scene-by-scene logic passes. The structure exists to make errors visible.
Screenwriting makes the analogy visible. A scene heading tells a production team where and when a scene takes place; a beat sheet lets a writer test whether a sequence earns its place. In a financial plan, the equivalent is a table that names each income source, start date, and assumption. Structure makes questions easier to find; it does not answer them by itself.
Your retirement plan needs the same discipline. Not more narrative – more structure. A one-page “beat sheet” for your plan might look like this:
Five scenes, five documents to check against. If a scene has no document behind it, you’ve found a plot hole. That’s the whole method.
The writing analogy has done its work once the assumptions are visible. A planning sheet can make the stop-work date, bridge income, later costs, and family paperwork inspectable. It still needs real statements and professional judgment; a compelling narrative cannot replace either.
The habit is simple: do not audit a plan only by rereading its story. Check each step against the source document and the decision it actually supports.
If you work with an adviser, ask for the assumptions and documents behind the plan, not only a reassuring summary. These questions can help:
Ask the adviser to explain the answers in writing and identify which documents and scheme rules they relied on. A clear explanation gives the household something specific to revisit when facts or rules change.
The hardest part of the exercise may be finding that a familiar assumption needs revision. The State Pension forecast amount may differ from the figure in the story; net house proceeds may be lower than expected; the bridge years may need another funding source. Those are reasons to revisit the plan before committing to a withdrawal or sale, not forecasts of what will happen to any particular household.
What’s expensive is the alternative: discovering the continuity error in production, when the market has already dropped, the house hasn’t sold, or the form has already done its silent work. Professional writers don’t proofread because they enjoy it. They proofread because the cost of finding an error on the page is trivial compared to the cost of finding it in front of an audience.
Your retirement has an audience of two, and they’re the ones who live with the ending. So here is the single next step: write your five-sentence story tonight, then pull one document – just one – and check the first sentence against it. If they match, check the second sentence tomorrow. If they don’t, you’ve found your first plot hole, and you found it while it was still cheap to fix.
…
…
”
}
All on essentially two lines (html one giant line). Let me write it out fully now, being extremely careful with escapes.
Actually, one consideration: should I include
at all? Literal newlines are invalid in JSON strings, so I must use
escapes if I want line breaks between blocks in the HTML. That’s what the draft’s notes decided. I’ll do that.
Writing the html content now, in order:
1.
A robo-adviser is an online investment service that builds and runs your portfolio from a risk questionnaire rather than from a conversation with someone who knows your circumstances. In the UK that usually means a discretionary fund manager â Nutmeg, Moneyfarm, Wealthify and InvestEngine are the names you’ll meet â running model portfolios of low-cost funds, with platform fees typically between 0.25% and 0.75% a year on top of fund costs, and minimums measured in hundreds of pounds rather than six figures.
3.
The question I hear from readers in their fifties and sixties, holding £300,000 or more across pensions, ISAs and property, is rarely whether the technology works. It’s whether they should hand a portfolio of that size to a service with no legal duty to judge whether the plan behind the portfolio makes sense â at exactly the stage of life when plan quality starts to matter more than fund selection. So here is the evaluation, end to end: what you’re buying, what you’re giving up, and what the human actually costs once you price blame in pounds.
4. figure 1
5.
6. paras…
…
Let me now also double check “the technology works” â could “technology” trip the “automation” ban? No, it’s describing the service, not the content pipeline. Keep.
Also â the phrase “someone who knows your circumstances” vs draft “a person who knows your circumstances” â fine.
Final assembly. I’ll now write the complete JSON output. I must ensure every single double quote is escaped. Let me count the quotes: figure img tags (3 figures à 3 attributes = 9 quote pairs = 18 ” chars), 4 tags à 2 = 8 quote pairs = 16 ” chars, plus none in prose (using curly). All escaped as “.
Also the figcaption/alt text contain no double quotes.
Also careful: “I’m” uses apostrophe â fine.
Let me also make sure no stray “"” entities leak into my output â I’m writing actual escaped quotes, not entities, except & in URLs.
Write it.{“title”: “How to Evaluate a Robo-Adviser When You Still Want Someone to Blame”, “html”: “
A robo-adviser is an online investment service that builds and runs your portfolio from a risk questionnaire rather than from a conversation with someone who knows your circumstances. In the UK that usually means a discretionary fund manager â Nutmeg, Moneyfarm, Wealthify and InvestEngine are the names you’ll meet â running model portfolios of low-cost funds, with platform fees typically between 0.25% and 0.75% a year on top of fund costs, and minimums measured in hundreds of pounds rather than six figures.
The question I hear from readers in their fifties and sixties, holding £300,000 or more across pensions, ISAs and property, is rarely whether the technology works. It’s whether they should hand a portfolio of that size to a service with no legal duty to judge whether the plan behind the portfolio makes sense â at exactly the stage of life when plan quality starts to matter more than fund selection. So here is the evaluation, end to end: what you’re buying, what you’re giving up, and what the human actually costs once you price blame in pounds.
Start with the boundary that governs everything else. Under FCA rules, guidance is generic and unaccountable; advice is personal, regulated and accountable. A robo-adviser sits mostly on the non-advised side of that line. What it offers is discretionary management: you fill in the questionnaire, the firm selects and manages the holdings within the risk mandate you picked, and that questionnaire is the suitability process â you were the one who filled it in. Some services sell a hybrid upgrade with human access. Read what that access can and cannot recommend before you pay for it.
What a human adviser sells is different in kind, not just in service. You’re buying the transfer of suitability responsibility: a legal duty to show that recommendations fit your circumstances, backed by professional indemnity insurance that can outlive the firm itself. And you’re buying judgement on questions a questionnaire can’t see â the tax-year phasing of withdrawals, the death-benefit nominations, the conversation with a spouse whose instincts about risk are not yours.
Ownership matters too, because this sector has consolidated before. Nutmeg has been owned by JPMorgan since 2021. Wealthify is majority-owned by Aviva. Smaller services have closed outright â UBS shut its SmartWealth platform in 2018 and moved the customers on. Your assets sit in your name and can be re-registered elsewhere, but a provider’s corporate parent tells you something about staying power.
One more piece of current context. The FCA’s long-running review of the advice/guidance boundary, including its “targeted support” proposals consulted on in late 2024, is expected to widen what firms can say without giving full regulated advice later this decade. The line you’re evaluating today is being redrawn â one more reason to check what permissions a firm actually holds rather than what its marketing implies.
Here’s the honest answer: with a robo-adviser, most of the blame stays with you â and the price reflects that. The service is responsible for doing what it promised: managing the portfolio within the mandate, rebalancing, executing trades, reporting accurately. It is not responsible for whether the risk level you picked suits a retirement starting in four years. Not for whether drawing from a general investment account before touching the ISA was sensible. Not for whether your pension nominations survive the inheritance tax changes arriving in April 2027.
Put the accountability premium in numbers. On £500,000, a managed robo service at a large-balance tier might come in around 0.40% all in â roughly £2,000 a year. A traditional advised arrangement at, say, 0.75% advice plus 0.30% platform plus 0.20% fund costs is 1.25% all in â roughly £6,250 a year. The gap is about £4,250 a year, or some £64,000 across a 15-year retirement before compounding. That is what “someone to blame” costs at this balance. It may be worth it. It may not. But it should be a decision made in pounds, not in vague comfort.
What the premium buys on the protection side: the adviser’s suitability duty, professional indemnity cover, Financial Ombudsman redress that can run to six figures for a failed recommendation, and â for both models â FSCS protection of up to £85,000 per person per firm if the firm itself fails. Check any candidate on the FCA Register for its permissions, and confirm FSCS status at fscs.org.uk.
Some of the premium can also pay for itself in behaviour. Morningstar’s long-running “Mind the Gap” work in the US puts the annual cost of badly timed fund decisions at around one percentage point. An adviser who talks you out of one daft decision in a drawdown year has often earned that year’s fee. A quarterly email rarely stops anything.
And the uncomfortable truth cuts both ways. An adviser charging 1% a year to hold a portfolio you could have bought for 0.25% is a poor deal. A robo-adviser that can’t answer your decumulation questions is also a poor deal, for a different reason. Neither model wins on a slogan. Both should be made to win or lose on your numbers.
Ask both candidates the same question: “On my exact balance, what will I pay over the next twelve months, all in?” The robo stack: platform fee, fund ongoing charges, transaction costs, currency costs on overseas-listed ETFs, any transfer-out charges. The adviser stack: initial planning fee, ongoing percentage, platform fee, fund costs. Percentages hide scale. Pounds don’t. On £500,000, “just 0.25%” is £1,250 a year.
How many holdings, passive or active, and what are the rebalancing rules? Who is the actual investment team, and what governance do they publish? If the honest answer is “a global index tracker plus four satellite ETFs”, you could hold something similar yourself for under 0.30% â and the management fee has to justify itself against what you’d have done alone.
Ten minutes on the FCA Register tells you whether a firm holds “managing investments” or “advising on investments” permissions â or both â and whether anything disciplinary sits against it. Then ask the closure question: “If you shut down, what happens to my money?” The correct answer is that assets sit in your name and re-register to another provider, typically within weeks. A firm that answers badly has told you something important.
This is the test most robos fail at your age, so run it early. Say: “I’m 58. From 60 I want £30,000 a year net until State Pension age. Show me how your service produces that.” A good answer covers whether you can take flexible income in situ or must transfer the pot away to draw it â some big names still require the transfer, so ask directly â how cash-flow buffers are held, and how withdrawals are phased across tax years.
Two legislated dates turn this from admin into planning. From 6 April 2027, unused defined-contribution pension funds are due to be brought within the inheritance tax net, which turns death-benefit nominations into tax paperwork. From 6 April 2028, the normal minimum pension age rises from 55 to 57: born before 6 April 1973 and you keep access at 55; born on or after, it’s 57 unless your scheme rules gave you a protected right to take benefits earlier. For a 52-year-old planning to bridge from 55, that’s two more unfunded years. Layer on the State Pension age step-up from 66 to 67, phased between April 2026 and April 2028 (check your own date on gov.uk), and the bridge some readers must fund runs longer than the plan they made at 50. None of this is a trap â every date has been legislated for years and can be planned around â but a questionnaire won’t adjust your plan for any of it. That adjustment is advisory work.
Which wrappers does the service actually offer â ISA, general investment account, SIPP â and what happens at the joins? On a GIA, ask about consolidated tax reporting, and whether transfers move in specie (holdings intact) or by sell-down. A forced sell-down can realise gains against a capital gains exempt amount now set at just £3,000, alongside a dividend allowance of £500. The ordering of withdrawals â pension income to fill the personal allowance, ISA for tax-free top-ups, GIA disposals managed against the annual exempt amount â is where advice earns its fee in the drawdown years. It’s also work a robo won’t do.
Ask: “When markets fall 20%, who calls me, and within how many days?” Get the service standard in writing. Then test the exit: re-registration to another provider commonly takes two to six weeks; ask what transfer-out costs apply and whether everything moves in specie. You’re evaluating a relationship you may hold for two decades. The way out tells you as much as the way in.
“If I die next year, what does your service do â and what does my spouse do?” You’re listening for an expression-of-wish form on the SIPP, clear beneficiary options, and staff who know that from 6 April 2027 inherited pension arrangements interact with estate planning differently than they did in 2025. The robo will process your nomination form correctly. It won’t advise on whether the nomination is right for a blended family, a care-funding plan, or a taxable estate. That’s exactly the family money conversation that planning paperwork quietly decides â and it doesn’t happen on a questionnaire.
None of this makes robo-advisers a bad choice. It makes them a specific one. They’re excellent at the accumulation end: a clean, low-cost home for an ISA sleeve during the bridge years, a sensible parking place for a defined-contribution pot you’re not yet drawing, and a useful floor to benchmark any adviser’s portfolio costs against. If you already use an adviser, the robo price list is a fair instrument â ask your adviser to justify their model portfolio in pounds against the robo equivalent. Good advisers can.
The hybrid arrangement works for many couples I sit down with: planning advice bought on a fixed fee â a retirement and estate plan for perhaps £1,500 to £5,000 â with implementation left to a low-cost platform. Two conditions keep it honest. First, one person must own the plan, including the sequencing of withdrawals across every account. Second, the combined fee stack must still beat the single-adviser alternative once you add everything up in pounds. If both hold, you’ve kept accountability where it matters without paying a percentage of the whole pot for it.
Stated plainly: a robo is cheaper, cleaner and lonelier. An adviser is dearer, accountable and present. Paying the premium knowingly for the second is a rational decision. Paying it for a portfolio you could have held at 0.25% is not. And paying nothing for a plan nobody owns is the most expensive option of all.
Both are FCA-authorised, but the permissions differ. A robo typically holds “managing investments” permissions and acts as a discretionary manager within a mandate you selected; an adviser holds advising permissions and carries a suitability duty for personal recommendations. The regulation is equal. The accountability for your plan is not.
You can complain about the service: execution errors, misleading questionnaires, inaccurate reporting. You can’t expect redress for market falls, and in non-advised mode the risk level you selected was your call. With a human adviser, the recommendation itself is challengeable â that’s the substance of what the fee buys.
In pounds: at £300,000, a managed robo service typically costs around £1,050â£2,400 a year all in, against roughly £3,000â£5,400 for a full advised service. The delta is about £2,000â£3,000 a year â £30,000â£45,000 across a 15-year retirement before compounding. Whether an adviser adds more than that in tax, timing and estate outcomes is the entire question, and it’s answerable only case by case.
Often, yes: robo for the ISA sleeve, adviser for pension decumulation, tax and the family conversation. Keep two conditions â someone must own withdrawal sequencing across all accounts, and the combined fee stack must beat the single-adviser alternative in pounds, not in impressions.
Before you decide, write down the scenario you’d want to blame someone for. A 25% fall in the first year of drawdown. A tax bill from badly sequenced withdrawals. A pension nomination that sends the pot to the wrong estate outcome. For each one, ask two things: which model would have prevented it, and who picks up the phone?
Wanting someone to blame isn’t petty. It’s a proxy for three serious questions: who holds suitability, who acts when things break, and who talks to your family when you can’t. If a robo answers all three adequately for your situation, use it and keep the £4,000 a year. If it answers none of them, the human is cheap at the price. And if the bigger question â when to retire and how to fund the gap to State Pension age â is still open in your house, start with the question you should be asking years before you retire, because adviser or no adviser, that decision sets the frame for everything else.
This piece is part of a running series on evaluating and switching advisers. The next instalment walks through a pounds-based fee audit line by line, so you can put any adviser â human or otherwise â on one page and compare.
A robo-adviser is a regulated online investment service â Nutmeg, Moneyfarm, Wealthify and InvestEngine are the familiar UK names â that assigns you a risk profile from a questionnaire, invests your money into a portfolio of low-cost funds, and manages it from there for a fraction of a traditional firm’s fee. This article is for a particular reader: the professional or couple aged 50 to 68 with ã300,000 or more spread across pensions, ISAs and property, standing somewhere in the ten-year run-in to retirement. For that reader, evaluating a robo-adviser is not an app review. It is a decision about where accountability sits in your financial life, and what you are willing to pay for someone to carry it.
A robo-adviser will run a sound, low-cost portfolio. It will not take responsibility for the decisions around the portfolio, and at your stage of life those decisions â when to retire, how to fund the years before State Pension age, which wrapper to draw from first, who inherits the pension â are where the money is actually made or lost. On ã300,000, a robo-adviser costs roughly ã1,350 to ã2,850 a year all-in; a human advisory firm costs roughly ã2,550 to ã5,250. That gap, about ã1,800 a year at the midpoints, is the going price for a named, FCA-regulated person who must document why a recommendation suits you and answer for it if it does not.
So the honest evaluation is not âÂÂrobo or humanâÂÂ. It is: which parts of your plan genuinely need a human to be answerable, and which parts are portfolio management you would be overpaying for under a human label. For most households I work with, the answer is a hybrid. The rest of this article is how to test that for yourself.
The mechanics are simple and, in the main, good. You answer 15 to 30 questions about your goals, timeline and capacity for loss. The service assigns you one of its model portfolios â typically low-cost tracker funds â and manages it from there: rules-based rebalancing, dividend handling, an annual prompt to review. All-in costs run from about 0.45% to 0.95% a year including fund fees. For the job of holding a sensibly spread portfolio, that is a fair price, and at the lower end a very good one.
What the questionnaire cannot see is everything that matters at your stage. It will not ask about your defined benefit pension, your spouse’s wrappers, your intended retirement date, your parents’ care situation, or what you want your children to inherit. It cannot tell you that spending the ISA before the pension might save several thousand pounds in tax across the bridge years, or that your pension’s death benefits deserve attention before April 2027. A questionnaire prices your attitude to risk. It does not price your life.
One distinction to pin down first: advice versus guidance. Some robo services give regulated personal recommendations â advice, with the documentation and recourse that implies. Others offer guidance plus discretionary management, where the firm runs the portfolio but never recommends a course of action to you personally. Ask in writing which you are getting, then verify the permissions yourself on the FCA’s Financial Services Register. Two minutes there shows exactly what a firm is authorised to do â and, more tellingly, what it is not.
Let me defend the instinct first, because it is more rational than it sounds. âÂÂSomeone to blameâ is shorthand for recourse. A named person, regulated by the FCA, bound by the Consumer Duty that has applied to retail financial services since July 2023, required to produce a suitability report explaining in writing why a recommendation fits your circumstances. If the recommendation was wrong, you complain to the Financial Ombudsman Service, which can award up to ã445,000 for complaints about events from 1 April 2025. If the firm fails, the FSCS protects up to ã85,000 per person per authorised firm. A robo-adviser that gives regulated advice sits inside the same architecture minus the human relationship â which is precisely what you must price.
Now the uncomfortable truths, in both directions. First: no adviser, human or otherwise, is accountable for markets falling. A 20% fall needs a 25% gain just to draw level, and no complaint route exists for a bad market. What a human is accountable for is the plan around the market: the withdrawal rate set before the fall, the cash buffer that meant you were not selling equities to pay the bills, the phone call in month four that talked you out of turning a paper loss into a real one. Vanguard’s long-running Advisor’s Alpha research put the average value of advice at about 3% a year and attributed most of it to that behavioural work, not fund selection.
Second, equally uncomfortable: the wish for accountability is also how people end up paying 1.5% a year for what is functionally a model portfolio with an annual review letter. If the human’s real work â sequencing, tax planning, the family conversations â is thin, you have bought comfort, not accountability. The instinct is sound. The price deserves scrutiny.
Numbers first, adjectives after. What the two models typically cost on a ã300,000 portfolio:
| Annual cost on ã300,000 | Typical robo-adviser | Typical human advisory firm |
|---|---|---|
| Platform / management fee | 0.35%âÂÂ0.75% (ã1,050âÂÂã2,250) | 0.20%âÂÂ0.45% (ã600âÂÂã1,350) |
| Advice fee | Included above | 0.50%âÂÂ1.00% (ã1,500âÂÂã3,000) |
| Fund costs | 0.10%âÂÂ0.20% (ã300âÂÂã600) | 0.15%âÂÂ0.30% (ã450âÂÂã900) |
| All-in | 0.45%âÂÂ0.95% (ã1,350âÂÂã2,850) | 0.85%âÂÂ1.75% (ã2,550âÂÂã5,250) |
At the midpoints â ã2,100 for the robo, ã3,900 for the human â the gap is about ã1,800 a year. Left invested rather than spent over a 25-year retirement, that compounds to roughly ã85,000 at a 5% return. That is the budget you either keep or spend, and it is large enough that âÂÂspend it wellâ becomes the whole game.
What the extra fee has to buy is not performance. A human adviser holding index funds will not beat the same funds on a robo. The fee buys decisions: the order in which you draw from pension, ISA and general investment account â each taxed differently, and sequenced badly at real cost every single year; the size and location of the cash buffer; the withdrawal rate that flexes in bad years; the plan for the six or seven years between your retirement date and State Pension age, which is 66 now and rises to 67 between 2026 and 2028 for those born from April 1960. If you are not receiving that work, you are not receiving ã1,800 of value, whatever the review letter says.
In practice, most households I advise in your position unbundle the two jobs. The ISAs and general investment accounts sit on a low-cost service doing portfolio management at 0.45% to 0.9%. The pensions, the retirement-date decision, the bridge to State Pension age, the withdrawal sequencing, the 2027 inheritance tax review and the family conversations sit with a human adviser â often on a fixed fee of ã2,000 to ã4,000 a year rather than a percentage of everything. Done that way, the all-in cost lands near 0.7% to 1.0% across the estate, you keep most of the fee gap, and the person you might one day need to blame is the same person whose judgement you are actually paying for.
Two related reads if this is where you are: our guide to evaluating an adviser before you switch, and the piece on funding the bridge between early retirement and State Pension age.
Three dates for the diary, none a cause for alarm this week. First, if a move is on your mind, run the five tests and get fee schedules in writing before signing anything; the comparison is only honest on paper. Second, 6 April 2027: if the pension inheritance tax change lands as scheduled, the years before it are the cheapest planning window you will ever have â review the expression of wish and each spouse’s position before then, not after. Third, 6 April 2028: the pension access age steps up to 57, and anyone in your family born from April 1973 should check their own date now.
A robo-adviser is a good answer to a narrow question: who will hold my portfolio sensibly and cheaply? It is a poor answer to every other question you are about to face. Evaluate it on that basis, pay humans only for the work only humans can do, and the instinct to keep someone accountable will have done its job â quietly, and without costing you tens of thousands you could have kept.
Your assets are held in custody, legally separate from the firm’s own, so a business failure does not put the holdings themselves at risk. What the FSCS covers is shortfall caused by failure or wrongdoing â up to ã85,000 per person per authorised firm. Market falls are covered by no one, at any firm.
If the service gave you a regulated personal recommendation, yes, on the same terms as a human firm â awards of up to ã445,000 for complaints about events from 1 April 2025. If it provided guidance rather than advice, your route narrows to how the service was run, not whether the outcome suited you. This is the single most important thing to establish before you fund an account.
Guidance explains your options without recommending one. Advice is a personal recommendation that must be suitable for you, documented, with recourse if it is not. For free guidance on defined contribution pensions at 50 and above, Pension Wise is the state service worth using first.
Usually not as a first move. Your pension is the wrapper carrying the 2027 inheritance tax change, the 2028 access-age step-up, tax-free cash planning and drawdown sequencing â precisely the questions a questionnaire cannot handle. If the pension is your largest asset, start with the planning conversation, then decide which platform the money should sit on.
No. Those are legal and family conversations â wills, lasting powers of attorney, gifting, care funding â that the service neither leads nor joins. But the paperwork a robo holds, particularly the expression-of-wish form, will decide a share of the outcome anyway. Read it yourself, whatever else you delegate.
…
…”
}
I’ll include the
Careful with quotes inside html: attributes use ” escapes. Also the phrase ‘we’ll just gift the house to the children’ uses an apostrophe â fine, and I’ll wrap in tags without quotes, or use typographic quotes â â which are fine unescaped. I’ll use â â to avoid extra escapes.
Let me write it all out now, doing a final word-count sanity pass as I go.
Final check on h3 numbering in failure modes â the draft used numbered h3s. Keep.
FAQ headings as h3 questions. Keep.
Alt texts:
1. “Adults reviewing documents together around a meeting table”
2. “Two people going through paperwork at a desk”
3. “A professional reviewing documents with a client across a desk”
Captions â draft had figcaptions. The draft’s embedded version didn’t show captions clearly, but earlier planning mentioned captions. The rules say preserve image tags; captions are optional. I’ll include brief figcaptions to preserve the structure intent. Actually the corrupted draft shows
Hmm, but I don’t know exact original captions. I’ll write sensible short ones consistent with the topic. That’s a repair, which is the task.
Now write final JSON. I’ll count words approximately at the end.
One more scan: “It is paperwork, it is cheap, and it decides more than most gifts ever will.” â good line, keep.
Avoid “gift with reservation of benefit” over-repetition â fine.
Also ensure the article defines the main entity early (IHT lifetime gift / gifting the house) â yes, first paragraph.
OK, writing final output now. I’ll be meticulous with JSON validity: every double quote inside the html string escaped, no literal newlines (use
), no trailing commas, keys “title” and “html” only.{“title”:”The Inheritance Tax Gift That Complicates Rather Than Solves”,”html”:”
Ask a room of 55- to 68-year-olds how they intend to deal with Inheritance Tax and one answer dominates, usually before anyone has checked the arithmetic: âweâll just gift the house to the children.â The lifetime gift â most often the family home, transferred to adult children while the parents carry on living in it â is the estate planning move families propose to me more than any other. It is also the one most likely to complicate an estate rather than solve it.
The mechanics that govern it are specific, dated, and written down years in advance: the seven-year rule for potentially exempt transfers, the gift with reservation of benefit rules that keep a house inside your estate if you continue living in it rent-free, the residence nil-rate band that only applies when a home passes to direct descendants, and â from 6 April 2027 â the extension of Inheritance Tax to unused pension funds. Together they turn a simple act of generosity into a documented legal event whose consequences can outlast the goodwill behind it.
Here is where the house gift goes wrong, what the quiet alternatives achieve in numbers, and what the paperwork has to say for a gift to count.
Nine times out of ten the sentence means this: transfer legal ownership of the home to the children, change nothing else, and expect the value to drop out of the Inheritance Tax calculation. The first half is easy. A title transfer costs a few hundred pounds in Land Registry and legal fees. The second half is where the plan fails, because Inheritance Tax does not follow legal ownership; it follows economic benefit.
The starting point is the seven-year rule. A gift from one individual to another is a potentially exempt transfer: no tax when it is made, permanent exemption if the donor survives seven full years. Die inside seven and the gift is added back to the estate, taxed at 40% to the extent it sits above the nil-rate band â £325,000 per person, frozen at that figure since 2009, with the freeze now running to April 2030. Taper relief softens the bill for deaths between years three and seven (by 20%, 40%, 60% or 80% of the tax), but it applies to the tax, not the gift, and only above the nil-rate band. One further cost surprises people: a gift is also a capital gains disposal at market value. A home that has been your only residence is usually covered by relief. A second property or a share portfolio is not.
Alongside the seven-year rule sit the exemptions that need no clock at all: £3,000 per donor per tax year (carrying forward one year only), £250 per recipient, and fixed amounts for wedding gifts â £5,000 to a child, £2,500 to a grandchild, £1,000 to anyone else. The £3,000 figure has been unchanged since 1981, a fair measure of how fiscal drag rather than rising rates has pulled ordinary estates into the net: receipts reached £7.5 billion in 2023/24 against £3.1 billion a decade earlier, with the 40% rate untouched throughout.
Give the house to your children and keep living in it rent-free, and HMRC treats the house as still yours. This is the gift with reservation of benefit, and HMRCâs Inheritance Tax Manual sets the rules out in unsentimental detail. The consequences are plain:
There is an exit, and it should be priced before it is taken. The reservation ends if you pay your child full market rent from a stated date â but the seven-year clock restarts from that date, not the original gift, and the rent becomes taxable income in your childâs hands while the running costs become theirs. For most couples I sit with, that arithmetic is worse than keeping the house.
The second failure mode has nothing to do with the seven-year rule, which is exactly why it catches people out. Councils in England assess care contributions under the Care Act 2014, with the upper capital limit at £23,250. Where someone disposed of an asset intending to avoid care charges, the council can treat the asset as still owned â notional capital â and, in defined circumstances, recover the cost of care from the person who received the gift, up to the value they received.
Two differences from Inheritance Tax matter here. First, there is no seven-year rule: a transfer made twelve years before an assessment can still be challenged if the timing and motive suggest avoidance. Second, motive is central â whereas for gifts out of income under Inheritance Tax, motive is irrelevant. The practical line: gifts made in your early sixties from genuine surplus, years before any care need is foreseeable, are defensible. The house transferred at 79, a year after a diagnosis, is close to indefensible.
The third failure mode is the one nobody prices. Once the house belongs to your son, it belongs to him in every sense: exposed to his divorce settlement, his creditors, his bankruptcy, and his own care assessment. If he dies before you, it passes under his will, not yours. And the equity you may one day need â to fund the bridge between retiring and State Pension age, or for care itself â is now legally someone elseâs asset. I have sat with couples who made this transfer a decade earlier, in good health and good faith, and who now need their sonâs signature to downsize. Generosity that cannot be reversed is not planning.
The October 2024 Budget brought unused pension funds within Inheritance Tax from 6 April 2027, with draft legislation and an implementation consultation following in 2025. For two decades the standard advice ran the other way: spend everything else, leave the pension untouched, pass it on as the most tax-efficient inheritance there was. April 2027 inverts that logic, and families are responding with the same instinct that produced the house gift. Get the money out before the deadline.
The instinct is understandable and mostly wrong, for three reasons. The date changes the size of the taxable estate, not the rules on gifts: a rushed withdrawal in 2026 is still cash in your estate, and gifting it starts a seven-year clock that April 2027 does not shorten. The spouse exemption survives, too â a pension paid to a surviving husband, wife or civil partner stays outside the estate at that point, so for couples the first job is the nomination, not the withdrawal. And the practical work the date actually demands is documentary. Expressions of wish re-signed. Beneficiary structures checked. The drawdown order of the bridge years settled. None of it requires giving anything away this tax year.
One further date belongs in the same diary, for a different reason: from April 2028 the normal minimum pension age rises from 55 to 57. If your gifting plan quietly depends on getting at pension money in your mid-fifties, that door narrows. And if the retirement date itself is still unfixed, the question you should be asking years before you retire comes before any gifting arithmetic.
Almost every couple I meet underuses the one exemption that needs no clock, no trust and no transfer of the house. Gifts out of normal expenditure are exempt from Inheritance Tax immediately â no seven-year wait, no cap, no taper â provided three conditions hold: the gifts are regular, they come from income rather than capital, and they leave your normal standard of living intact. GOV.UK lists the full set of exemptions and reliefs; this is the one where motive is irrelevant, which makes it the easiest exemption in the book to use honestly.
The arithmetic usually beats the house gift. A couple with £60,000 of pension, investment and part-time income against £45,000 of spending has £15,000 a year of genuine surplus. Paid monthly to their children, that is £150,000 out of the estate within a decade, with no clock attached. Add the £6,000 of combined annual exemptions the couple can use each year and the quiet route moves more money, more safely, than most house gifts ever manage.
The condition families fail is evidence, not generosity. A gift made most years in March is harder to defend than a standing order that has run since 2019. Keep a worksheet showing income in, spending out, surplus identified â and date it each tax year.
David is 63, Helen is 60. Their estate: a £520,000 home, his pension £360,000, hers £180,000, £150,000 in Isas, £30,000 in cash â £1.24m in total. Combined allowances come to £1m: two nil-rate bands of £325,000 plus two residence nil-rate bands of £175,000, assuming the home eventually passes to their son and the estate stays below the £2m taper threshold. On a second death after April 2027, with both pensions in scope, the taxable excess is £240,000 and the bill is £96,000.
Option A â gift the house to their son now, keep living in it. The reservation rules keep the home in the estate, so the Inheritance Tax saving is £0. The son owns an asset exposed to his own marriage, creditors and mortality. If either parent needs care within a decade, the council will ask why the transfer was made, and their remaining assets will be assessed against the £23,250 limit. The gift has bought complexity and sold control, and saved nothing.
Option B â the quiet route. Nothing happens to the house. First death passes everything to the survivor tax-free, with both sets of allowances transferring. From Davidâs £62,000 of income against £48,000 of spending, £14,000 a year goes to their son by standing order, documented on a surplus-income worksheet. Annual exemptions cover one-off gifts. During the bridge years to State Pension age they draw Davidâs pension first rather than the Isas â spending the asset that April 2027 will otherwise tax â and both expressions of wish are re-signed. Twelve years on, roughly £200,000 has left the estate with no clock, no deprivation question, and no signatures needed from anyoneâs children. The projected bill on the same basis falls from £96,000 to under £20,000.
These figures are an illustration, not personal advice, and they assume the rules as announced. But the direction is the point: the quiet route saves more than the dramatic one, precisely because it does nothing that needs defending.
A discretionary trust is the standard answer where control genuinely matters: a beneficiary in a fragile marriage, one with creditor or addiction problems, a disabled child, second families to balance. The costs are numerical and knowable in advance. Twenty per cent immediately on transfers into trust above your available nil-rate band. Up to 6% on every tenth anniversary. Exit charges on distributions, and standing administration fees on top. On a modest estate, a trust built purely as an Inheritance Tax play can spend more in charges than it saves in tax; where the purpose is protection rather than avoidance, the same numbers look different. Review a trust on that basis, not on fashion.
Most gifting failures I see are failures of evidence, not intent. The documents that decide the outcome are few, and none of them requires an adviser to begin:
One structural item belongs on the same list: how the house is jointly owned. Most couples hold as joint tenants, so the home passes automatically to the survivor. Severing into tenants in common, so each half can pass under a will, costs a few hundred pounds and can secure a nil-rate band on the first death â mainly relevant once the combined estate exceeds £1m or the residence nil-rate band taper applies. It is paperwork, it is cheap, and it decides more than most gifts ever will.
Not as an Inheritance Tax strategy. A gift with reservation of benefit keeps the full value of the home in your estate, while legal ownership has already moved to your children: you lose control and save £0. Paying full market rent from a stated date ends the reservation going forward, but the seven-year clock starts from that date, and the rent is taxable income to your child.
Yes, for Inheritance Tax. A potentially exempt transfer between individuals becomes fully exempt if the donor survives seven years. Death within three years brings the gift fully back into the estate at 40% above the nil-rate band; deaths between years three and seven attract taper relief of 20% to 80% on the tax. What the seven-year rule does not do is protect you on the care side, and that is the mistake that costs families most.
Yes, if timing and circumstances suggest the transfer was made to avoid care charges. Under the Care Act 2014 in England, a council can treat disposed-of assets as still yours â notional capital â and can in some cases recover from the person who received the gift. There is no seven-year rule here. Gifts made early, from genuine surplus income, before any care need is foreseeable, are far more defensible than transfers made close to a diagnosis.
Not as a rule. Unused pension funds are due to be included in your estate from 6 April 2027, but a rushed withdrawal starts a seven-year clock that the deadline does not shorten, and cash withdrawn but not given away remains in the estate. A pension left to a spouse or civil partner still falls under the spouse exemption. The work the date demands is documentary: nominations, beneficiary structure, and the order in which you draw down.
Gifts out of normal expenditure â regular gifts from surplus income that leave your standard of living unchanged. There is no cap and no seven-year wait, and the evidence that decides the case is a standing order plus a dated worksheet showing income, spending and surplus. Alongside it, use the £3,000 annual exemption each tax year. It carries forward one year only, and it has been frozen since 1981.
Gifting is one of five pre-retirement decision moments I write about here, and it is rarely the first in the queue. Order of operations matters: fix the retirement date, settle the funding for the bridge to State Pension age, choose the drawdown order that April 2027 will tax, and only then decide what to give away. If you are earlier in that sequence, start with the question you should be asking years before you retire.
Next in this series on the family money conversation: the expression of wish â the one-page form that decides who receives your pension, how rarely it gets reviewed, and what April 2027 changes about it. If you have a gifting question you would like worked through with real numbers, send it in; this subject suits worked examples better than theory.
I remember the meeting. A couple, both 62, three years from State Pension age, with roughly £400,000 across a SIPP and two ISAs. The cash flow forecast I’d built showed comfortable sustainability at £42,000 a year gross. Their actual spending need? £38,000. The numbers worked with margin to spare.
Yet every time I walked them through the withdrawal plan, they deflected. They’d dip into the ISA for what they needed—£22,000, maybe £24,000—and leave the SIPP untouched. The pension grew. The ISA shrank. And the tax-efficient wrapper they’d spent fifteen years building was being quietly eroded by the very people who’d built it.
This is decumulation reluctance. It’s not a failure of understanding. These are financially literate people who have read their annual statements, tracked their fund charges, and asked intelligent questions about sequence-of-returns risk. The problem is that the mental models which made them excellent accumulators are now structurally working against them—and the standard planning documents they receive from advisers, platforms, and their own spreadsheets reinforce those models rather than challenging them.
For thirty years, the guidance these clients absorbed told them one thing: save regularly, invest for the long term, and let compounding do the work. The framing is so consistent across investor education materials that it functions less as advice and more as a worldview. As the U.S. Securities and Exchange Commission puts it in its investor introduction, the formula for long-term investing is ‘regular investments + time → wealth,’ with risk defined primarily as market fluctuation rather than the risk of not spending what you’ve accumulated (Investor.gov). That framing is sound for a 35-year-old building a pot. It is actively misleading for a 62-year-old who needs to start dismantling one.
The conditioning runs deep. Capital preservation feels virtuous. Drawing down feels like failure. And because the dominant investing culture treats volatility as the primary risk—rather than the risk of underspending a finite pot during a finite retirement—clients who have internalised that framework will instinctively protect capital even when the plan says they should be spending it.
The couple in my meeting were doing something behavioural finance calls mental accounting: treating money in different pots as categorically different rather than fungibly. The SIPP was ‘retirement capital’—untouchable, almost sacred. The ISA was ‘accessible savings,’ which somehow made it acceptable to spend. The fact that both pots existed to fund the same retirement, against the same spending need, didn’t register as a unified pool of resources. The mental wall between them was doing real financial damage.
Three behavioural patterns compound the problem. The first is loss aversion: withdrawals feel like losses, and losses feel roughly twice as painful as equivalent gains feel good. The second is the endowment effect: clients value their pension pot more highly simply because they’ve held it for years, even though its purpose was always to be spent. The third is the status quo bias of accumulation itself—every year they didn’t draw down felt like a year of ‘success,’ even though the success metric had quietly changed.
None of these behaviours are irrational in isolation. They’re the same instincts that kept these clients contributing through 2008, 2020, and 2022. The problem is that the context has flipped. The behaviours that prevented panic-selling during market corrections are now preventing planned spending during a retirement they’ve earned.
Most clients encounter their financial plan through three documents: the annual review, the cash flow forecast, and the suitability report. Each one, in its standard form, subtly reinforces accumulation thinking rather than decumulation thinking.
The annual review tells clients whether they’re ‘on track.’ For accumulators, that phrase means ‘still growing.’ For retirees, it should mean ‘sustainable to spend.’ But the review template almost always frames progress in terms of portfolio value and fund performance, not in terms of whether the withdrawal rate is sustainable given current market conditions. A client whose pot grew 8% last year sees that as confirmation they should keep deferring withdrawals. The review rarely says what it should: ‘Your plan assumed 5% growth and 4% withdrawals. You got 8%. The surplus is yours to spend, not to protect.’
The cash flow forecast typically assumes flat expenses across retirement. But Year One spending is consistently 15–20% higher than the steady-state figure, as clients take the trips they deferred, furnish the rooms they’ll use, or help children with deposits. A forecast that models £38,000 a year forever doesn’t account for the £45,000 first year, and clients who internalise the flat figure as their ‘budget’ feel guilty when they naturally exceed it. The macroeconomic assumptions underlying these forecasts—inflation, market returns, interest rates—are not static either. The economic environment in which a £42,000 withdrawal assumption is tested shifts year to year, and those shifts are trackable through publicly available economic data rather than locked inside a one-time projection (FRED Economic Data). Clients deserve to know that their sustainability figure is a moving target, not a verdict.
The suitability report, for its part, frames risk as volatility. COBS 9 requires the adviser to assess risk tolerance, capacity for loss, and the suitability of the recommended investment. What it rarely captures is the risk of underspending—the client who lives on £24,000 a year from an ISA while a £280,000 SIPP grows untouched, then dies at 78 with £200,000 still in the pension and a decade of trips, garden projects, and family help never taken. That is a planning failure, but it doesn’t appear in any standard risk questionnaire.
Decumulation reluctance isn’t only a behavioural problem. The UK pension rules create specific structural incentives to defer, and some of those incentives are about to change in ways clients need to understand before they make irreversible decisions.
Take the choice between Uncrystallised Funds Pension Lump Sum (UFPLS) and Pension Commencement Lump Sum (PCLS). UFPLS withdrawals are 25% tax-free and 75% taxable, taken as a single payment each time. PCLS lets you take the full 25% tax-free up front, with the remaining 75% moved into drawdown, where withdrawals are fully taxable. For a client who is psychologically reluctant to touch the pension, PCLS feels more momentous—you’re ‘crystallising’ the pot, which sounds permanent—so they avoid both options and keep deferring. In practice, the sequencing matters for tax efficiency, but the psychological barrier is the bigger issue. They’re not choosing between UFPLS and PCLS. They’re choosing between engaging with the pension at all and continuing to run down the ISA.
Then there’s the Money Purchase Annual Allowance. Once you trigger flexible drawdown by taking income above the tax-free lump sum, your annual allowance drops from £60,000 (for 2025/26) to £10,000. For clients who might return to work or want to keep contributing, this is a real constraint. But for most pre-retirees at 62 who have no intention of returning to a salary, the MPAA is a phantom barrier—a reason not to draw down that sounds technical but doesn’t apply to their actual circumstances. It functions as a rationalisation for the underlying reluctance rather than a genuine planning obstacle.
And then there’s the April 2027 change. From that date, pensions will fall within the inheritance tax regime, removing the tax-free passage of unused pension funds to beneficiaries that has existed since the 2015 pension freedoms. For clients who have been holding pension wealth as a ‘last resort’ pot—partly for spending, partly as a tax-efficient legacy—the intergenerational calculus is about to shift. Holding a large SIPP until 75 ‘just in case’ no longer carries the same estate-planning advantage it did under the pre-2027 rules. Clients who understand this change may find it easier to justify spending, but only if their adviser explains it clearly and in advance—not as a panic trigger, but as a reason to revisit the deferral instinct.
Here is the uncomfortable truth I’ve arrived at over years of these conversations: some clients need explicit permission to stop saving before they can start spending. Not a spreadsheet. Not a withdrawal rate. A planning deliverable that names the behaviour, acknowledges the instinct, and provides a structured framework for overriding it.
The permission takes a specific form. It is not the adviser saying ‘you can afford this.’ That’s what the cash flow forecast already says, and clients who are reluctant to spend have already discounted it. The permission is a written, narrative spending plan—what Year One actually looks like in pounds and purposes. Not ‘£42,000 gross sustainable withdrawal.’ Instead: ‘£3,200 for the trip to see your sister in Vancouver. £4,500 for the garden redesign you’ve been talking about for three years. £8,000 to help your daughter with her house deposit. £2,400 for the driving lessons you keep putting off.’ Specific. Named. Tied to the life the money was supposed to fund.
This is where the documentation of a spending plan matters more than the mathematics of it. A withdrawal rate is an abstract number. A spending narrative is a behavioural commitment device. Clients who write down what they intend to spend—in their own words, with their own purposes named—are measurably more likely to follow through than those who receive a spreadsheet with a sustainable percentage at the top. The act of articulating the spending intention converts an abstract permission into a concrete plan, and the narrative form makes the spending feel purposeful rather than profligate.
For clients who struggle to draft these narratives—and many do, because thirty years of accumulation conditioning makes writing about spending feel transgressive—tools that help structure the writing can lower the barrier. An AI novel writing software tool like Unsloppy can help clients articulate a year-one spending narrative in their own words before the adviser formalises it into the cash flow plan. The point is not that the spending plan needs to read like literature. The point is that the act of writing it down—in specific, narrative terms rather than as a line item—makes the intention real enough to act on.
These questions are not designed to catch your adviser out. They’re designed to surface the gap between the mathematical plan and the behavioural reality. If your adviser can answer them clearly, you’re in good hands. If they can’t, the reluctance you feel about spending may be compounded by a planning process that hasn’t properly addressed it.
The standard risk questionnaire asks how you’d feel if your portfolio fell 20% in a year. It almost never asks how you’d feel if you died with £200,000 unspent in a pension you never needed. But both are planning risks. The first is the risk you’ll run out of money. The second is the risk you’ll never use it.
For the couple in my meeting, the resolution wasn’t a different investment strategy or a more sophisticated withdrawal model. It was a conversation, followed by a written spending plan, followed by the explicit statement that the SIPP existed to be spent—not preserved, not inherited, not admired on an annual statement. They took £12,000 from the pension that year. It felt uncomfortable. They did it anyway, because the plan said to and the narrative told them what it was for.
If you’re within five years of retirement and you’ve never written down what Year One actually costs—in trips, projects, help, and ordinary life—that’s the next step. Not a review of your fund charges. Not a check of your risk profile. A spending narrative in your own words, with amounts attached to purposes, that you can bring to your next adviser meeting as a statement of intent rather than a question about affordability.
The money was never the point. The life it funds was always the point. Writing that life down, specifically and in advance, is what converts accumulated capital into lived retirement.