Feature

posted 13 Dec 2003 in Volume 9 Issue 1

ECA course: Part three

The basics of financial planning

“Assumptions” are what we base much of our thoughts on. Useful and false ones abound. To quote Rudyard Kipling in The Old Men: “This is our lot if we live so long and labour unto the end – That we outlive the impatient years and the much too patient friend: And because we know we have breath in our mouth and think we have thoughts in our head, We shall assume that we are alive whereas we are really dead.” Michael Hague continues with part three of the ECA course: The basics of financial planning, with assumptions in mind. In the quest for better understanding, you will probably be surprised to meet “the man in black” and to engage in the “Monte Carlo simulation”. Who said lawyers and doctors have all the good jargon?

In addition to demonstrating a prescribed standard of professional education, the granting of the licence to become a certified financial planner requires a number of years experience, plus, in most cases, the submission of a written financial plan for scrutiny by the Institute. The standards for that financial plan include the setting of assumptions. This is one of the most crucial elements of preparing a financial plan. Make incorrect assumptions and, no matter how accurate your financial maths, you will get incorrect information and may end up taking inappropriate actions, that is, rubbish in rubbish out. The standard for financial plans says that assumptions should be:

Stated;
Reasoned;
Reasonable.

The first two are quantitative points, that is, were the assumptions stated and were the reasons behind them given?

The third one is much more subjective and, dare I add, beloved of lawyers probably for this very reason. Were they reasonable assumptions to make? If the client agrees with the assumptions, and if the maths have been correctly performed, then they cannot disagree with the conclusions. However, our financial-planning colleagues in the US frequently refer to the “man-in-black test”. This is a little sinister but is not to be confused with “the men-in-black test”, which involves Hollywood aliens and has nothing to do with either reasonableness or financial planning. Put simply they mean when you are stood in front of the man in black, being called to account for your financial planning, will he also judge that your assumptions were reasonable?

Who sets the assumptions?

A number of financial planners argue that assumptions should be set by the client. They ask the client questions such as: “Mr Client, what do you think inflation will be in the future?” Or they will ask: “Mr Client, what rate of investment return shall I assume in your financial plan?” I believe it is wrong to ask the client for this information.

Imagine you are stood in front of the man in black. The client’s solicitor says: “My client knew so little about these matters that he realised he needed to take professional advice. Given that lack of knowledge, why did you think it was right to ask him what he thought future inflation would be?” How will you be able to defend yourself?

Planners need to have a process for deriving assumptions. They can then discuss the methodology with the client. If the client brings in factors that would have affected the way in which the assumption was arrived at, then the assumption can to be varied to meet his particular circumstances. This is totally different to asking what they think the assumptions should be.

What assumptions do we need to make?

General assumptions

We make some very general assumptions, which, while they are common to all clients’ situations, it is worth stating. They are usually lifestyle based and do not generally result in numerical answers:

That the client’s current position is the one that we are planning for, unless they have come specifically for a plan in preparation for a change. An example of this would be that, if the client is married and has children, we assume that he is going to remain married and be financially responsible for his or her children. If they are coming to us for a financial plan in anticipation of divorce then that is a totally different matter;
That the client is ethical. We do not do financial planning that involves tax evasion. Nor do we get involved in financial planning where, for example, one member of a partnership or marriage wants to achieve something at the expense of the other. For this reason, if a couple are getting divorced, and we have acted for both of them in the past, we will only act for them jointly during the divorce, not for the husband against the wife or vice versa. We believe that to do so would be a conflict of interest;
That the client will want to at least maintain their current standard of living. If the client wants to plan for an expanded standard of living then we will work with that but, in all scenarios, such as planning for retirement or protecting income in the event of ill health or a death, we will assume that the current standard of living is that which we are seeking to maintain rather than aiming, for example, to create a very wealthy widow;
That all information has been disclosed. Financial plans cannot be prepared unless the client is willing to give full information. I know that a lot of independent financial advisers are aware that their clients have investments that they do not disclose and may even have income that is not disclosed. Comprehensive financial planning requires comprehensive information. If it subsequently transpires that the client has not disclosed something to his planner, then the planner must not be put in a position where they are responsible for any errors that arise as a result.

Specific assumptions

Specific assumptions arrive at numbers to use when preparing plans. Others may be needed for specific clients, but the ones I will cover are:

That prices will rise, so we need to set an assumption for future inflation;
That earnings will rise, so we need to set an assumption for increase in earnings;
Planning will be for the client’s lifetime. This means that we have to make an assumption about how long the client is going to live;
That money will be invested. This is going to involve a range of assumptions for different types of investment;
Taxation, like eventual death, is one of the few certainties in any form of financial planning. We need to make an assumption about future tax rates;
Education fees will continue to rise, so a rate needs to be assumed;
Long-term care costs will rise faster than inflation, so a rate needs to be assumed.

These should be examined in more detail but, before I do so, I need to make sure readers are clear about the difference between escalation and revaluation.

Escalation and revaluation

One of the things that is most often misunderstood in areas of financial planning is the difference between escalation and revaluation.

Escalation: Is an increase to a movement in money during the time it is occurring. For example, a pension in payment may escalate each year.

Revaluation: Revaluation is the increasing of a figure between the time that it is set and the time it starts to be paid. An example may be that, if someone leaves a pension scheme they will have a retained pension. This may be revalued between leaving and the time it starts to be paid. Thereafter, it may escalate.

This is important as revaluation and escalation may apply to the same figures, but be at different rates. The most common example of this is when planning for retirement. If we analyse the client’s current expenditure, we can arrive at a target income that will maintain their current standard of living in retirement. Between now and when they retire that income will rise.

If we have assumed that it will rise in line with national average earnings then we will revalue the target by national average earnings, making the assumption that, as their pay rises, their expenditure will also rise.

However, once we arrive at a point when a client is living on their retirement income, we will assume that that needs to escalate in order to allow for increasing prices. This escalation, logically, will be in line with the retail prices index, which, as we will see, is usually assumed to be different to the increase in the national average earnings.

Specific assumptions generally do enable us to arrive at a number to use when performing calculations.

Relationship to inflation

The actual figures arrived at are less important than is the relationship between them. As an example, assuming an investment return of six per cent per annum might be reasonable during a time when inflation is four per cent. However, it might be an impossible target if we were experiencing inflation of less than one per cent or be a ridiculously easy target if inflation returned to ten per cent per annum. For this reason, we recognise the interdependency of all the assumptions that we make and, as a base assumption, refer most things back to inflation.

For this reason you will see that earnings rise by a factor above inflation. Similarly, the cost of school fees, long-term care costs and investment returns are all related back to inflation. By doing so, we should arrive at a position where the financial plan works, even if inflation is higher or lower than expected.

There is a way of testing this. If your financial plans have been constructed in a suitable way, it is possible to test them by applying various rates of inflation, having the model automatically adjust all the other assumptions. The technique for doing this is known as a “Monte Carlo simulation”, reflecting the fact that you are effectively rolling dice in order to randomly arrive at a figure for inflation and then measure the key results from the financial plan for each time you have rolled the dice.

Having carried out tests of this type, we find that there are certain extreme situations where plans won’t work. These tend to be only those where the relationship with inflation breaks. If it were possible to find a scenario where inflation was high but investment returns were low, or vice versa, then most financial plans would fail. Given that in such a scenario it would be possible to invest in index-linked gilts and obtain a return greater then the high inflation, I cannot see how, other than by sheer bad management of money, we could have a protracted period of high inflation and low investment returns. For this reason, we assume that the relationships will continue but that they will not be flat margins.

Investment returns will not be a flat percentage above inflation. School fees will not rise by a flat percentage above inflation. They are far more likely to rise at a multiple of inflation.

Prices will rise

I will not go into the economic theory behind why we have inflation. We have not yet come across a set of personal circumstances that would mean that we shouldn’t assume any form of inflation for the future.

Simple reference to past inflation figures is not a reasonable way to make assumptions about future inflation. Inflation is a function of various economic factors and, unless you believe that the future will closely resemble the past, it is wrong to assume that future inflation can be derived by studying the past.

The government currently has set an inflation target and has given the Bank of England autonomy over the way in which it seeks to achieve that target. The question has to be: do you believe that it will be achieved? If you don’t, then using the government’s target is inappropriate. Similarly, it is inappropriate to assume that the government will miss its target by, for example, 0.5 per cent and thus assume that inflation will be 0.5 per cent higher or lower.

So how might you do it? Remember that the important thing is that your assumption should be stated, reasoned and reasonable. Let us take those in reverse order. In doing so I will go into more detail about inflation than I will about other assumptions, but I am sure the reader can apply the same principals to all the figures we need to derive.

Inflation – reasonable

One way of ensuring that an assumption is reasonable is to derive it from information either provided by or used by those best able to make an assessment. If every expert pointed you to one particular assumption and you, from your experience, saw no reason to deviate from that, then I would suggest that it is a reasonable assumption to use.

We arrive at an assumed future inflation rate by a reference to appropriately dated government gilts. The major purchasers of these are the large pension schemes, insurance companies and investment houses. They aim to get equal value from differently constructed gilts. In particular, they aim to get the same return from an index-linked gilt as they would from a fixed-rate gilt. This is not the place to go into great detail about the difference between the two types of gilts, but suffice to say that one gives an income during its lifetime that is fixed and known at the outset whereas the other gives an income that is a margin above inflation. If the institutions are to get equal value from each of these gilts then they need them to be priced in such a way that inflation plus the margin equals the rate available from a fixed-rate gilt.

Inflation – reasoned

The pricing of gilts is slightly more complex than it might at first seem, as it needs to reflect the difference between the anticipated rate obtainable now and that which was available when the gilt was first released. This means that you may pay more or less than the face value for a gilt now but will always receive its face value when it is eventually redeemed. By calculating the return you get on the price you pay, taking into account the income you are going to receive while you own the gilt, and the fact that you will eventually get its face value back, you can arrive at what is known as the gross redemption yield. Comparing the gross redemption yield between fixed interest and index-linked gilts will give you a margin. This margin represents what the market believes inflation will be.

Inflation – stated

This may be a different figure over different periods of time. At the time of writing, the Financial Times is giving a breakeven inflation rate for some index-linked bonds and, for example, give a figure for the 2.5 per cent Index Linked Treasury 2009 as being 2.9 per cent. To carry out fuller calculations, I recommend you visit the Debt Management Offices website, www.dmo.gov.uk, where full information on gilts in issue, plus their market prices, can be found.

I started this by saying why I felt the method was going to be reasonable. I have then given the reasons and, finally, I have stated the assumption. This meets the required standards.

Client’s earnings will rise

In most cases, it is wrong to assume that earnings will rise by inflation. Historically, this has never been the case and the difference is reasoned by reference to some basic economic theory.

In the same way as economic theory would lead us to conclude that inflation will be a factor in our client’s financial planning, we also anticipate that there will be growth in the overall economy. Earnings need to rise to keep pace with inflation, but they also need to reflect the sharing of that growth with those who have helped to create it. Because earnings rises include an element of economic growth, we assume a different rate of earnings escalation to that which we have for inflation.

We regularly review the way in which we arrive at an assumption for earnings growth. I believe that the often-used basis of earnings rising by a fixed percentage above inflation is incorrect. An increase of, say, two per cent above inflation was relatively modest when inflation was ten per cent per annum. However, an increase of two per cent above inflation now would see earnings rising close to twice as fast as inflation.

Analysis of the relationship between retail price index and national average earnings does not show an easy correlation. This may be partly due to the time lag between prices rising and pay increasing to keep up. We also see times when wages are held down by certain economic pressures but then are increased significantly as part of wage negotiations to catch up again.

We start our basic assumption with a revaluation of earnings equal to the assumed inflation rate plus the government’s target for economic growth. However, if we are close to the planning event, such as being within a couple of years away from retirement, we will return to assume that the revaluation is in line with inflation only.

Taxation

Tax, no doubt, will form part of our client’s financial futures. What’s equally certain is that the actual tax regime cannot be known in advance.

However, it is widely recognised that, unless there is a major change in government policy, tax changes move the tax burden around rather than make great differences to the overall amount of tax that is paid by any one individual. For this reason, we assume that the current tax system is the one that will apply in the future. If the future tax regime is different, then we believe that it is reasonable to assume that its overall effect will be about the same. Why use an unknown system for calculating tax if you can use a known one and expect to arrive at a similar outcome.

Life expectancy

In the first two parts of this course, I introduced the concept of a lifetime cash flow. If such a cash flow is to be prepared, then an assumption needs to be made about how long a lifetime actually is.

In the early days of life assurance companies, tables of life expectancy were constructed by visiting local cemeteries and recording the age at which people died. Unfortunately (or fortunately!), those tables proved to be somewhat inaccurate as they failed to take account of local factors, such as how many of those deaths were earlier than they might have been simply because you were recording deaths in a disease-ridden city.

Things have moved on since then and actuarial science is now well established and provides a plethora of information on which to base our assumptions.

There may be known health factors that lead us to assume a short life expectancy but such things need care. We have a client who, convinced he would not live to see age 75, made a pension contribution with the intention of creating an additional lump sum for his wife on his death. As he is a retired doctor, it seemed reasonable to accept his assumption. His 75th birthday came and went and he is now drawing additional income on which he is paying 40 per cent tax. Not a major problem but a less than 100 per cent successful bit of financial planning.

It would have been much worse if we had prepared a plan for someone to spend all their money by age 75 – but then they lived longer.

Today, life tables are a major part of actuarial sciences. The government actuaries department (GAD) publishes a range of tables on their website, www.gad.gov.uk, in a very usable Excel format.

Comparing the data for various parts of the British Isles reveals quite a difference.

However, before those “north of the border” start a mass exodus south, they should remember that:

Figures are changing all the time. These are the 1999–2001 interim life tables. Similar figures ten years ago would have suggested around two years less life expectancy. This is the result of improved health care and general standards of living;
Other factors give even greater differences. For example, annuity life tables, which give a measure of the life expectancy of those, generally more affluent people who buy annuities, might show an average of five or more years’ greater life expectancy.

The secret to long life seems to be “have money and outlive the last set of tables”. That way you will be chasing an ever increasing life-expectancy target.

However, care needs to be taken in using these figures. We have had a recent example of how different the figures can be in differing circumstances. We recently recruited an actuarial sciences graduate. He brought with him his copy of the actuarial tables used by current students. As is often the case, the student information is somewhat out of date as it was only used in order to facilitate training and practice calculations rather than to arrive at real-life conclusions.

Two things that are immediately obvious if you read his tables are that life expectancy is increasing (simply because his old tables have significantly shorter life expectancies than do the ones on the government actuary’s website) and that the differences in life expectancy of different socio-economic groups is much greater than that found as a result of geographical differences. The English may live longer than the Scots but those who are wealthy enough to buy annuities live substantially longer than those who are not.

If we were planning for someone who was already in their 90s, we might want to make a life expectancy assumption that is in line with the GAD tables. If their health were considered to be significantly below average then it might be reasonable to assume a shorter life expectancy.

If we are planning for anyone younger than their 90s then we need to allow for the fact that they are living during a time when life expectancy for such age groups is being extended. At the other extreme, if we are planning for someone at the start of their working life then how on earth do you arrive at an assumption for them?

Our starting point is to assume a life expectancy for everyone of age 110. Statistically, there is a negligible chance that someone will live beyond that at the moment. We believe that the majority of the medical advances that are going to have great impact on life expectancy have already been made so the rate at which life can be extended in the future will now slow down. If someone invents a miracle cure, which means that people are going to live twice as long or even life forever then financial planning as we know it is going to be wrong for just about every single person on earth. This is one of those scenarios where not only does it seem highly unlikely; no-one has an answer to the questions that it would pose if it were achieved.

The only true solution is to arrive at a point where you have sufficient financial resources to be able to take an index-linked income from it and still maintain a capital value that grows faster than inflation. If life expectancy has been doubled then chances are that our entire economic fabric will be so different that the tools that might be used to achieve such a situation have become worthless. You might as well try to plan for a scenario where each of your clients wins the lottery every week – it is so unrealistic as to be outside of the parameter of any financial planning.

Education costs

It is often assumed that education costs will rise in line with inflation. A simple look at the historic increases tells us that that is not the case. Deeper consideration of the issues tells us that it would be unreasonable to assume that that would be the case.

Education costs, particularly those for private schooling, reflect the increase in cost of delivering the service. One of the biggest costs is teachers pay. As we have already assumed that pay will rise faster than inflation, then that alone would make it impossible for education costs to only rise by inflation.

In addition, education costs reflect increasing capital expenditure needed in order to bring school buildings up to the standard that we now expect of them, that more up-to-date (and more expensive) resources such as computers, electronic music equipment, etc., are made available to pupils.

This combination of factors means that we need to assume that education costs will rise considerably faster than inflation.

In order to arrive at a useable figure, I have looked at the relationship between inflation, the national average earnings index and the increases in private education fees over many years. This includes data from various schools groups such the Independent Schools Association, the Headmasters’ Conference and the Girl’s Schools Association. Analysis shows that education costs rise about 2.67 times the rate of inflation.

Long-term care costs

This is a real political hot potato at the moment. The care costs are changing significantly, not because the cost of care is changing but because the responsibility for care costs and the resources available to meet that responsibility is being passed from central government to local government and from one budget to another. This means that the proportion of the cost that is being left as the responsibility of the individual is changing more significantly than is the actual cost itself.

The factors that affected education costs are not dissimilar from those that affect the basic cost of providing for long-term care. Therefore, it is not unreasonable to assume that long-term care costs will increase by a similar multiple of inflation. What it is difficult to legislate for is whether or not those costs will ultimately fall to the individual or be met by local or central government. Here is one example of an assumption where we discuss the principle with the client and mutually agree whether to plan for all costs falling on them or for some assistance being available.

The actual calculations regarding care costs then need two assumptions. The first is the cost of care of the type and in the location that the client would be most likely to need it, and the second is the rate of escalation.

Finding figures for the cost of care in your area can be greatly helped by visiting www.bettercaring.co.uk, offered by the Stationary Office, where it is possible to find out how much it will cost for specific nursing homes or nursing homes in a geographic location. We then assume that that cost will be revalued and escalate at the same rate, which we set at being 2.67 times the rate of inflation.

The decision as to whether you are going to assume that support is available is very much client specific, as it depends on their financial resources, and may well depend on the view you take about the future for state funding towards this particular need.

Investment return

When I prepared the first article in this course, I almost immediately received a request for information about the investment return I had assumed. I did explain to the sender of the e-mail that this was not a single rate but a composite return. So let’s look at this in more detail now.

In our financial plans, we identify how much money needs to be set aside to meet three distinct spending requirements. We refer to these as the clients now, soon and later pots.

The now pot has to provide the money the client expects to spend in the next two years, plus a reserve fund to cover the spending he hopes not to have to make – but just might need to;
The soon pot has to provide the money that the client thinks he will be spending in years three, four, five, six and seven. If the client wants to be particularly safe, they may even want the soon pot to have money for years eight, nine and ten;
The later pot has everything else.

Our assumptions are based on now money needing to be instantly available, so only invested in very low-risk, instant-access holdings, such as bank or building society accounts. The soon money can be invested, but only if there is certainty of outcome. It would be appropriate, for example, to buy government gilts with soon money and then hold them through to maturity. It would equally be appropriate to buy National Savings bonds or other forms of guaranteed investment.

Investing later money depends on the client’s attitude to investment risk. Just as you would tailor an investment portfolio to be in line with that attitude to risk, you must also allow for different assumed rates of return. It would be unreasonable to assume a low-risk investment portfolio would obtain the same return as one that took a higher risk. If that were the case, then why would anyone take an investment risk?

Returning to some basic economic theory, there has to be a hierarchy of investment returns. If putting money on deposit meant that it consistently lost value against inflation then no-one would bother saving up. Therefore, we assume that deposit accounts will give gross interest that is slightly above inflation.

If guaranteed investment bonds only matched deposit bonds then, once again, there would be no reason to invest in one rather than the other. Consequently, we assume that soon money would get a return of 1.1 times the return on now money. Later money requires a matrix depending on the client’s attitude to investment risk. While it is outside the scope of this particular part of the financial-planning course, I will simply say that we have a risk tolerance questionnaire, which results in seven different risk categories.

The rate of return for each of these categories, expressed as a multiple of deposit rates, is as follows:

Risk adjusted portfolios – return assumed as a multiple of deposit rates

Risk group 1 - 1.2
Risk group 2 - 1.34
Risk group 3 - 1.47
Risk group 4 - 1.61
Risk group 5 - 1.76
Risk group 6 - 1.87
Risk group 7 - 1.96

It can be seen from this that the assumed rate of return for a client’s investment portfolio is first dependent on their particular circumstances and, second, is somewhat fluid as the client moves money from later to soon and from soon to now in order to spend it.

Perhaps a more detailed review of this aspect of planning will need to be covered in a future article as I am sure that assumptions made in this area will raise as many questions as have been answered.

Powerful modelling tools are the only way to see the results of such complex interrelationships of assumptions and the complexities of managing a client’s plan through the steps of implementation, monitoring and reviewing in order to help them reach the financial goals they have set.

Michael Hague is a partner in The Consultancy Group based in Bawtry. He is a fellow of the Institute of Financial Planning and is a certified financial planner licensee. He was Money Management’s overall “Financial Planner of the Year” in 2001, having been their “Investment Planner of the Year” in 2000. He can be contacted at: michael.hague@theconsultancygroup.com