Capital Beyond the Portfolio

Capital at Risk — Paper 15

I. The Unit of Analysis

Financial services necessarily divide capital into structures. A SIPP, ISA and general investment account (GIA) have different tax treatments, access provisions and regulatory characteristics. The investments within them may also be subject to different mandates. Those distinctions are real and necessary.

The difficulty begins when the structures used to hold capital also determine how the client's affairs are considered.

An attitude-to-risk questionnaire provides useful information about a client's willingness to accept investment risk. At adviser level it sits alongside objectives, capacity for loss, time horizon, liquidity requirements and wider circumstances. Suitability requires considerably more than a risk score.

Further down the investment chain, much of that client-specific information falls away. A DFM operating an MPS may know the mandate and risk classification without knowing much about the individual whose capital sits within the model. An underlying fund manager is further removed again. They manage the capital they receive according to the mandate they have been given.

The adviser knows the client, while further down the chain the investment solution increasingly knows the mandate.

II. Capital Across Wrappers

Different wrappers can reasonably have different objectives. A pension inaccessible for many years presents a different planning problem from capital required in three. Time horizon, liquidity requirements and capacity for loss can justify different mandates across the same client's capital.

A firm might instead conclude that substantially the same MPS is suitable across a SIPP, ISA and GIA. That can also be perfectly reasonable. The difficulty is assuming that suitability within each wrapper tells us everything we need to know about the arrangement across them.

Tax makes this particularly visible. An additional-rate taxpayer may hold low-coupon gilts inside an ISA while assets with less favourable capital-gains treatment remain in the GIA. There need be nothing unsuitable about either portfolio for the asset location to make little sense across the client's affairs.

A case from my own practice made the point more clearly. The client wanted approximately £60,000 in cash. Their GIA was fully invested and contained a structured product showing an unrealised loss of around £100,000 at prevailing market prices. It had been intended to be held to term and we still wanted to own it. If its conditions were met, the target return was approximately 230%.

Within the client's wider SIPP portfolio, around £100,000 was held in cash alongside its other investments. The obvious source of the £60,000 was therefore the pension. The client had already used the tax-free cash available to them from the pension, so any further withdrawal would have been taxable pension income. They were already a higher-rate taxpayer, so that meant income tax at 40%. The pension therefore held the liquidity but could not release it efficiently, while the GIA held no cash but contained an asset we did not want to abandon.

Instead, the structured product was sold from the GIA through the broker at the prevailing secondary-market price, producing broadly the amount the client required. The SIPP trustees then used existing cash already held within the scheme, rather than a new contribution, to acquire the same structured product in the secondary market. These were two separate market transactions at prevailing secondary-market prices rather than a transfer between the wrappers.

The client still owned the same investment exposure, but it now sat inside the SIPP. The GIA held the cash. The economic loss on the structured product had already existed before either transaction and was crystallised on disposal. The tax consequences of the disposal and reacquisition formed part of the planning. Any subsequent gain on the reacquired holding would arise within the pension rather than the taxable account. The transactions did not create the £100,000 loss; they changed where the remaining exposure was held.

The structure subsequently produced a return of approximately 230%. That outcome did not make the original decision correct, but it explains why we had wanted to retain the exposure. Considered separately, the SIPP held cash that was expensive to release and the GIA held an investment we did not want to sell permanently. Neither wrapper, on its own, contained enough of the client's position to produce that answer.

III. Degrees of Freedom

Paper 03 considered the additional degrees of freedom that appear as capital grows and the problem moves beyond simply accumulating assets. Asset location is one place where those degrees of freedom can be used.

Applied across wrappers, they also change what bespoke investment management can actually do. This is rather more interesting than another argument about active and passive management, or whether one collection of funds is better than another.

Model portfolios have legitimate advantages. They provide consistency, scalability, supervision and cost efficiency. Their relative lack of client-specific discretion can itself be useful: fewer individual decisions can make oversight easier and reduce dependence on one adviser or investment manager getting every judgement right.

Vertical integration can reduce some of the same handoff risk by bringing advice, investment management, wrappers and administration together. It can also result in the investment proposition being shaped around what the integrated infrastructure can efficiently support.

For more complex clients, bespoke management offers something different. The attraction is the ability to use individual holdings and their location as part of the financial plan, rather than any assumption of superior manager skill.

Complexity does not increase neatly with AUM. A relatively wealthy client can have simple affairs, while a smaller pool of capital can become complicated once it is spread between pensions, taxable accounts, companies, trusts or different family members.

Bespoke management is more complicated and more expensive, and for many clients there will be no economic case for paying for flexibility they are unlikely to use. An AUM threshold is therefore a poor proxy. The same amount of capital can present a very different planning problem depending on how and where it is held.

IV. Cashflow

A client may have a twenty-year investment horizon and still require £200,000 in year three. Both facts matter, but the second one changes what can be done in the period before the withdrawal.

It does not follow that £200,000 should be sold immediately and left sitting in cash. There may already be cash within the portfolio. Bonds mature and positions are sold as part of normal investment management. If the investment manager is selling something anyway and a withdrawal is approaching, some of those proceeds can go into the pit lane rather than straight back into the market.

By pit lane, I mean capital deliberately taken out of normal portfolio rotation because it is approaching the point at which the client will need it.

There is a cost to doing this. Cash held early can create drag if markets continue to rise, while leaving everything invested for longer preserves expected return but increases the possibility that assets have to be sold into an unfavourable market when the liability arrives. Cashflow planning gives time to manage that trade-off.

An MPS will continue to rebalance according to its model. A fund manager will continue managing the fund according to its mandate. That is what each has been employed to do. The client's requirement next March will only affect the investment process if the information reaches somebody able to do something with it.

This is why the industry's repetition of "time in the market" can become too crude when applied to financial planning. None of this requires a view on market direction. Yield to redemption depends on the price and date at which a bond is acquired. Maturities occur on specific dates. Disposal timing can alter tax consequences. The client's liabilities arrive on actual dates regardless of what markets happen to be doing.

V. What Survives the Mandate

Paper 07 considered what happens when different professional advisers each see part of a family's affairs. Investment management adds another dimension because information about the client has to be translated into something the investment process can use.

A risk classification, investment objective or time horizon survives that translation fairly easily. Specific information about future expenditure, assets held elsewhere or tax consequences is harder to reduce to the same sort of instruction.

The planner may know about expenditure, tax, pensions and assets elsewhere. The investment manager knows the portfolio, its liquidity, forthcoming maturities and positions likely to be sold. Trustees, accountants and lawyers may each be acting properly within their own remit while holding information another part of the process needs. Their separate mandates do not automatically transmit it.

The case in Section II worked because the planning information and the investment information met before the decision was made. Had either remained confined to its own part of the process, the opportunity could easily have disappeared. No misconduct or obvious mistake is required for that to happen. The structures do not all contain the same information.

Paper 07 dealt with fragmented professional knowledge. Here there is an additional step because the client's circumstances have to be translated into mandates and instructions that can still influence decisions further down the chain. Some context will inevitably be lost. Whether the remainder is sufficient depends partly on what the capital is being asked to do.

VI. Beyond the Portfolio

A portfolio remains a sensible unit for investment management. It becomes less useful as the sole unit of analysis when the client's capital is spread across structures with different tax treatments, access provisions, ownership and purposes.

Portfolio-level suitability remains necessary, but it says little about whether the pieces have been arranged well together.

The structured-product case shows what becomes possible when the relevant information crosses those boundaries. Standardisation deliberately reduces reliance on that kind of individual judgement, and there are good reasons for doing so. Some decisions are unnecessary, expensive or difficult to supervise. For a sufficiently complex client, however, some of the decisions removed by standardisation have economic value, provided the information needed to use those additional degrees of freedom can cross the structures around the capital.

The structured-product case happened to produce an opportunity. The same boundaries can produce a constraint.

Case note

The structured-product example is drawn from the author's professional experience. Figures have been rounded and identifying client information omitted. The sale proceeds were broadly the amount required by the client; the SIPP cash balance is included only to show that sufficient existing scheme liquidity was available to reacquire the exposure.

The GIA disposal and SIPP acquisition were separate market transactions at prevailing secondary-market prices. The SIPP acquisition used cash already held within the scheme rather than a new pension contribution.

The client's liquidity requirement created the decision point, and the tax consequences of disposing of the GIA holding and reacquiring the exposure within the SIPP were considered as part of the planning.

The precise capital-gains treatment depends on the facts and on the matching, connected-person and other rules applying at the relevant time. This example should not be read as a generic tax strategy.