The interesting number in this week's research is not the one about advisers. It is what that number implies about everybody else. Research commissioned by AJ Bell and conducted by Ad Lucem, surveying 350 UK financial advisers and 150 paraplanners online in May 2026, found that 88% of the advisers said tax wrapper decisions had become more complex over the past three years. Among paraplanners — the people who do this analysis all day, every day — 86% said the same. And 79% of the advisers said they would use a formal wrapper decision framework if one existed.
Read that as an accountant rather than as an adviser and it lands differently. These are specialists, choosing between pensions, ISAs, trusts and onshore and offshore bonds several times a week, with technical teams and provider support behind them, and nearly nine in ten of them say the ground has shifted under a decision they had long since settled. Now think about the client whose file is open on your desk. They are making a version of the same decision — where the surplus goes, how the extraction is structured, what happens to the pension — once, unaided, at the end of a long year, on the basis of something they half-remember reading. The lang cat's State of Advice Report 2025 put the proportion of UK adults who had paid for regulated advice in the previous two years at 9%. Most of those decisions are being made with nobody in the room who does this professionally.
What actually changed
The research attributes the added complexity to a specific accumulation, not a general sense of turbulence. Reductions in several tax-free allowances. Increases in capital gains and dividend tax rates. And unused pension funds due to be brought within the inheritance tax regime from April 2027 — the change that most directly rewires how owner-managers have been taught to think about the order in which they draw on their wealth.
Charlene Young, head of technical at AJ Bell, described the effect as advisers having to reconsider established approaches to how clients draw income from and pass on their wealth, and said that the level of agreement showed “this reset is something the entire industry must recognise”. Tony Wickenden of Technical Connection, who worked on the accompanying material, framed it as running across the whole wealth journey — accumulation, decumulation, preservation and transfer.
That framing matters for accountants, because each of those four stages has an accountancy footprint. Accumulation shows up as retained profit and extraction policy. Decumulation shows up as the owner who suddenly wants to know what the company can pay them. Preservation and transfer show up as shareholdings, trusts and the succession conversation nobody has scheduled.
The read-across for the accountant
There is a temptation to treat a story about advisers as a story for advisers. It is not, for one structural reason: the accountant sees the trigger, and the adviser does not.
Consider a fairly ordinary file. A trading company with £180,000 sitting in the current account, of which the accountant knows perhaps £60,000 is genuinely committed to working capital and tax. The owner is 54. There are two old workplace pensions from before they went out on their own, untouched since. Extraction has run on the same salary-and-dividends pattern for six years because it worked, and nobody has revisited it. The company has a director's loan that gets cleared and re-drawn each year.
Nothing on that file is a problem. It is a good client having a decent year. But every one of the pressures the AJ Bell research identifies runs straight through it — the dividend rate on the extraction, the capital gains position if the business is ever sold, and from April 2027 an inheritance tax treatment of those two dormant pensions that is not the one the owner assumed when they left them alone. The owner will not raise any of it, because none of it announces itself. The accountant is the only professional who can see all four facts at once.
The client does not know the decision has changed. The accountant is the only person holding the facts that show it has.
Where your ground ends
Being the one who sees it does not make you the one who decides it, and the distinction is worth stating precisely because it is where well-meaning firms get uncomfortable.
The tax analysis is squarely accountancy work. Establishing how much of that £180,000 is genuinely surplus, modelling what a given extraction route costs, explaining the corporation tax treatment of an employer pension contribution, setting out what a disposal would do to a client's marginal position — all of that is what clients already pay you for, and none of it requires anything you do not already hold.
What sits on the other side of the line is the selection. Recommending that this client's money belongs in a pension rather than an ISA, or in a bond rather than a general investment account, is regulated advice and requires FCA authorisation. The line is not about competence. Plenty of accountants understand wrappers perfectly well. It is about permissions, and about who carries the liability when the recommendation is tested years later.
The practical consequence is a sequence rather than a hand-off: the tax facts established first by the person who holds them, the wrapper decision made second by someone authorised to make it. Where that sequence breaks down is not usually at the perimeter. It breaks down earlier, when nobody mentions the issue at all — the accountant assuming the client has someone, the client assuming the accountant would say something if it mattered. Our piece on the cash question covers how that silence usually forms.
Two things worth doing this week
Run one filter across the client list. Not a review of every file — a single query: which corporate clients hold a materially larger cash balance than they did two years ago, and which owner-managers are within ten years of the age they have vaguely named as their exit? Those two lists will overlap more than expected, and the overlap is the population for whom the changes the research describes are live rather than theoretical.
Add one question to the year-end meeting. Something factual, not advisory: “Is anyone looking at your pensions alongside this?” It is a question about the client's affairs, which is exactly the kind of question an accountant is expected to ask, and it does not commit you to an answer. In a fair number of cases the reply will be that there is an old adviser nobody has spoken to in years, or nobody at all. That answer is the useful output. Our note on the signals already sitting in the accounts sets out what else the same conversation tends to surface.
What is not yet settled
Two things are genuinely open, and it is worth being clear which.
The April 2027 inheritance tax treatment of unused pension funds is the pivot for a lot of owner-manager planning, and the detail of how it will operate in practice — particularly the administration around it — is still bedding in. Anyone telling a client precisely how their estate will be assessed under it is ahead of the evidence. The right posture is to flag the direction, not to pre-compute the outcome.
The second is whether a decision framework of the kind 79% of advisers said they wanted actually changes behaviour, or simply documents it. AJ Bell has published one alongside the research, and is running a ten-location tour between 7 and 22 September to take advisers through it. Whether that shifts how wrapper decisions get made across the market is not something a survey conducted in May can answer. It is worth watching over the next year rather than concluding anything now.
Neither uncertainty changes the point for an accountant, though. The uncertainty is about how the specialists will handle a harder decision. The gap is about the clients who are not putting the decision in front of a specialist at all — and on that, the accountant is not a bystander. They are the only person with the facts in front of them. More on the shape of that gap sits in our overview of the advice gap.
Common questions
Is wrapper choice something an accountant can advise on?
The tax analysis is accountancy ground. Explaining how a company distribution is taxed, what a disposal does to a client's marginal position, or how a pension contribution interacts with corporation tax is ordinary tax work. What sits outside it is the recommendation itself — telling a client that a particular pension, ISA, bond or trust is the right home for their money is regulated advice under the Financial Services and Markets Act regime, and it needs FCA authorisation. The practical line most firms work to is that an accountant describes the tax consequences of options a client is already considering, and a regulated adviser decides which option suits them. Crossing that line is a permissions problem, not a competence one.
Why should an adviser's difficulty matter to a client with no adviser at all?
Because it calibrates the difficulty. The AJ Bell research surveyed people who choose wrappers professionally, several times a week, with technical teams behind them — and 88% of the advisers and 86% of the paraplanners said it had become harder over three years. A company owner making the same decision once, from a standing start, at the end of a busy year, is doing it without any of that. The lang cat's State of Advice Report 2025 found only 9% of UK adults had paid for regulated advice in the previous two years, so most of those decisions are being made unaided. The reasonable inference is not that owners are getting it wrong every time. It is that nobody should assume they are getting it right by default.
What changes in April 2027, and is it too early to raise it?
Unused pension funds are due to be brought within the inheritance tax regime from April 2027, and that change is one of the specific drivers the AJ Bell research cites for wrapper decisions becoming harder. It is not too early. A pension that was accumulated on the understanding it would pass outside an estate is now a different asset in planning terms, and the owners most affected are the ones who deliberately left pensions untouched while drawing dividends. Raising it in the 2026 year-end meeting gives a client time to take advice and act before the date, rather than discovering the position afterwards. Altro has a fuller piece on that specific change.
A client asked whether spare company cash should go into a pension or an ISA. What can we say?
You can say a great deal about the facts. How much of the balance is genuinely surplus once working capital, tax reserves and known commitments are set aside is your analysis, and it is the number the whole question turns on. So is the corporation tax treatment of an employer pension contribution, and the effect of extracting the money as salary or dividend first. What you would be moving into is the comparison itself — which wrapper suits this client's age, risk tolerance, retirement date and existing holdings. That comparison is regulated advice. The joined-up version is that you supply the verified surplus figure and the tax picture, and a regulated adviser does the selection.
What does a genuinely joined-up conversation on this look like in practice?
It starts from one set of facts rather than two. The accountant brings what only they can see — the surplus cash figure, the director's loan, the extraction pattern, the shareholdings, the fact the owner turns 55 next spring. The planner brings the regulated analysis and the recommendation. The client has one conversation instead of two disconnected ones, and neither professional is guessing at the other's half. The mechanics vary: some firms make a warm introduction and step back, others stay in the meeting so the tax reasoning is in the room. What matters more than the mechanism is sequence — the tax facts established first, the wrapper decision made second, by someone authorised to make it.