The lifetime challenge of funding your own retirement
It is changes in society and government policy rather than economics or financial asset performance that make lifetime financial planning different for today’s generation of young people. For those with expectations of near-average lifetime earnings, the challenges are not necessarily so different. But for young people with higher expectations, the changes are pretty dramatic. Almost regardless of how entrepreneurial are their expected income sources, they are likely to have to plan their earnings and retirement income as if they were running a business.
In this item, I look at the drivers for lifetime cash flow planning and put some numbers to the task for an expected retirement income, by way of example, of £35,000 pa after tax in real terms. To the extent space permits, my answers place the accumulation of assets for retirement in the context of personal beliefs and ambition about earnings opportunity and also of the typical dynamics of ‘life stages’ which are partly predictable and partly dictated by fate but never run smoothly like a financial plan.
Personal responsibility and pensions
In the past, anybody with expectations of an above-average level of lifetime earnings could either depend on their employer to fund an acceptable level of retirement income or, being a professional or entrepreneur, they tended to see self-funding as something that went with the territory. Provided earnings lived up to expectations, self-funding has confronted no widespread difficulties in spite of the extravagant costs of personal pension products. In terms of economics, inflation has played havoc with planning but (in the UK, at least) a traditional bias to real assets (partly to absorb high costs) has ensured that asset returns in real terms have got the job done. The volatility of financial assets and changing correlations between different assets has meant that some generations did much better than others but the job still got done. The only threat to those already retired is if they did not plan on inflation protection for their pension and inflation subsequently cuts real pension incomes more than they can bear. That remains to be seen.
If those with entrepreneurial income face no fundamental change in the nature of their task, the same may not be said for employees in the private sector, for whom personal responsibility in the shape of self-funding is now a new imperative in their lives. For reasons spelt out here in several other items, the final salary pension scheme is not likely to be an option for many. It is easy to see why when I answer the questions about what self-funding requires, because individuals still have the scope for making tradeoffs between targets, required certainty and investment risk that are no longer open to occupational schemes. If individuals collectively cannot afford the contributions needed in order to meet high targets with near total certainty, neither can the nation’s employers.
Some people with high earnings expectations will choose a career in public service. If they are current with trends in pension provision, they may even see the possibility of taxpayers meeting their high targets for retirement income with total certainty as an important attraction. However, this is self-defeating as it is not possible for public and private funding of retirement income to diverge indefinitely. If shareholders cannot let management continue to make generous pension promises, taxpayers cannot let governments do it either. Ultimately, all the money comes from individuals and we are all in the same boat.
Realistically, then, all aspiring young people, whatever their expected sources of earnings, stand to gain from making action plans for the financial dimension of their lives and doing so in the context, amongst other aids, of a financial model that is realistic as well as intuitive, both about the dynamics of personal lives and households and about the uncertainties of financial markets.
There are challenges for professional advisers here too. They must be as good as mentors of the principles of goal-setting as they are as providers of the numbers that inform planning. The mathematical rigour of planning does not always go with the passion of creating opportunities. The simplification assumed in financial models typically cuts across the complexity of the uncertainty in the real world. And advisers need to grapple with new academic theories about lifecycle goal priorities which have some counter-intuitive implications for approaches to risk-taking for different ages and stages.
Goal-based planning has a very different complexion for young people than for people with accumulated savings whose current scale is a known and whose targets and time horizons are both much clearer. More of future spending power for young people is explained by future earnings than by capital or past earnings. Getting this across is vital if people in their twenties are not to despair as soon as they see a route map for lifetime cash flows laid out with numbers. Financial markets are hard places with bounded scope to carry investors to their goals. Income generation is where people with the high lifetime spending expectations can make the difference. That said, many young people will have formed a narrow view of their control over their destiny, such is their propensity to make early experiences mean more than they need. Financial planning has a horrible habit of focusing on the distribution of limited means when it should be about helping to lift the perceived limitations on means.
My advice, then, is to leave pensions out of early-stage life planning and focus on a priority which is easier to relate to and a more powerful incentive to income generation. This is building up a stock of financial assets to empower yourself. Give it a name, maybe some defined targets for its use and give it a target level and time horizon, like £50,000 by age 35. The ‘money is power’ approach to building capital resonates because it provides flexibility to change your life today. You do not get this flexibility in a pension plan. You may also decide you do not get it owning a property with largely borrowed money.
The size of this capital stock and how much eventually can get carried over to pensions depends on many circumstances, including lots outside our control. But at least the new pensions regime allows us the flexibility to move non-pension capital into pensions in a brief period of time (as a function of 100% of earnings per year).
Decisions about how to manage the capital depend partly on the possible claims on its use, some of which may be relatively short term. But the empowerment function of early capital accumulation may in any case stand conventional thinking about investment risk on its head, if individuals place a high value on loss avoidance until a sufficient core stock has been build up rather than maximising the time over which an average expected rate of return (including some risk premium) compounds.
There will also be circumstances in which pension contributions do start at 25, either because an employer is paying them (perhaps replacing a final salary arrangement) or because an individual embraces the lack of flexibility of personal pension contributions as a savings discipline.
In this case, what sustainable rate of contributions is required to meet the target? Realistically, the target should today be set as an income starting at 65. Earlier retirement is then one option for benefiting from all the better outcomes in the range of uncertain outcomes for the plan, a direct alternative to retiring on a higher income. This gives at least a 40-year time horizon to the contributions. A financial model based on historical real equity returns is likely to point to equity-based investment strategies initially because all the expected band of real outcomes at this horizon lies above the near-certain real outcome when investing exclusively in index linked gilts. Such a model should also reject non-inflation proofed bonds as having at very long horizons similar real return uncertainty to equities but with no risk premium.
To deal with the wide range of outcomes resulting from normal market volatility and its dependence on the investment time horizon, it is helpful to think in terms of a minimum required outcome. This acceptable ‘floor’ fixes the combination of contributions and investment risk whose resulting outcome range exceeds it with (logically) confidence of 99% or thereabouts. In your 20s it may not be easy to focus on either a minimum income target in real terms or a desired outcome but it is certainly not impossible and will in any event be refined over the course of the journey. For this example, I suggest setting a low floor at 60% of the desired target of £35,000. This is equivalent to a quite high level of risk tolerance, even for young people. The target should always be expressed in ‘real terms’ so having the same purchasing power whatever happens to inflation.
Apart from targets, time horizons and risk tolerance, setting contributions requires some other practical assumptions:
Expected returns should reflect initial market conditions but for regular contributions over a long period the returns will be close to ‘normal’. The outputs in this example reflect expected returns of between 5.3% pa (UK) and 6.2% pa (Japan).
Exposure to equity markets will be highly diversified and ‘efficient’ in terms of risk and return
The expected returns above assume low implementation costs, relying mainly on trackers
Risk is controlled by mixing risky assets with index linked gilts, the effect of which is to narrow the range of outcomes
Risk and return projections for the plan assume that the exposure to risky assets will be replaced by risk free assets as the time horizon shortens
Though the expected equity returns assume ‘mean reversion’, no extra return is assumed from buying low and selling high when rebalancing portfolios over the course of the plan
The capital output from the model will be used at age 65 to purchase an annuity with spouse benefit and 3% pa inflation allowance. The model outputs assume an annuity yield of 4.6%. A return to evidenced real yields might support a higher rate for converting capital to income but cannot be taken for granted when for all sorts of reasons there is growing demand for assets that match inflation-sensitive liabilities without risk. Illustrations are also offered assuming a rate of 6% which could be a function either of higher real yields or of buying less inflation protection
The contributions required from age 25 to meet these targets with 99% confidence are £643 per month. The capital sum required at 65 is £951,000. At the higher annuity rate the capital required is nearly £730,000 and the contributions reduced proportionally.
I have deliberately not expressed these as a percentage of current pay. Though this is traditionally the way pension contributions are expressed, it produces incorrect information about adequacy when real rates of pay are expected to increase significantly over a career (as is likely in this example).
Starting at 35
If we assume the same targets and risk tolerance and the low annuity rate but no prior accumulation of assets that are available for retirement planning, the contribution rate on a conservative annuity assumption is nearly double at about £1,191 per month.
However, if a capital sum of £50,000 has been accumulated outside pensions by whatever means and is now available to be added (actually or notionally) to the pension fund, the contribution rate should be just over £960 per month.
Leaving it late
Many people with entrepreneurial income effectively leave pension funding till late in life and may even rely on the sale of a business to fund their retirement. For this illustration, it is more interesting to see what happens when the task has simply been postponed. At 45, with just £50,000 of capital accumulated, and at 50 with £100,000, the required contribution rates soar to £2,067 and £3,148 if the targets are unchanged. Realistically, the annuity purchased in such a case will be inferior but even so the contributions then need to be well over £2,000 per month from age 50.
It is sensible to think forward to what someone’s retirement expectations will be at around 50 when funding has been put off to this extent. If the capacity to meet these required contributions is there because of high disposable income, realistically the acceptable retirement income will itself have moved higher. Put differently, regret at falling short of the original target will be much greater and the original floor may now look unbearably low.
In this example the degree of funding risk is still very high and a conflict has emerged between the desire to make the risk management consistent with meeting a worst case outcome with near total certainty and the need to gamble to live comfortably. In terms of asset allocation, the difference is between about 30% exposure to risky assets (consistent with the risk tolerance assumed at the earlier ages and the now shorter horizon) and 100%. Something has to give if the contributions are not achievable. If the gamble takes the form of carrying much more investment risk over the 15 years remaining, an annuity will only be bought if the bet has paid off by 65. Otherwise it is likely to be extended by going into drawdown.
There is not necessarily anything wrong with extending the bet by using a drawdown plan and indeed most people with well-funded pension plans when they approach retirement (and experience by then of successfully planning a financial goal) are likely to elect for drawdown. The problem when not well funded is that risk aversion is typically increasing naturally as we age and the forced gamble may be experienced as a source of great stress and hence as a failure.
Presumably, many people who defer pension provision in this way have a plan B if the gamble does not pay off. Selling the family home and trading down is clearly an option and the diversion of savings capacity to meeting the high costs of home ownership may in any event explain the lower pension funding. A similar Plan B is inheritance but this too may depend in part on house prices. In this example, the magnitude of the equity required to be released or inherited is possibly greater than people realise. If only half the required contributions can be made, it is nearly £300,000.
The other clear risk to late funding is the security of employment income. The notion of contributing till 65, let alone maximising savings rates in the final 15 years, assumes that labour income can then be earned. For young people planning today, this is perhaps a reasonable assumption as the labour market will have had time to adapt. For people already caught in the late funding trap, longevity has increased but employability has not.
These different examples are summarised in the table below. Each can be reduced by about 25% on the assumption of higher annuity rates or else weaker inflation protection. This roughly corresponds to the difference in funding rates implied by setting the risk tolerance low instead of high (the two sets of numbers). You could rely on a reversion to high real interest rates to compensate for the lower contributions in the table but I do not recommend relying on them to bail you out if you opt for the lowest contributions and then take less investment risk. All that does is push the risk somewhere else.
Comparisons with occupational pensions
The funding principles are identical but the rules (and resulting risk preferences) are not.
The future employer liability must reflect a degree of inflation proofing for both salary and benefit whereas individuals have complete flexibility about the target income level and about the form of annuity or alternative income
The present value of the expected employer liability must be calculated using bond yields not expected returns for the assets invested in, leading to volatility in the difference between the market value of the assets and the present value of the liabilities that can only be minimised by investing in bonds
The direction international accounting conventions is headed in make the equivalent of the desired target for an individual the floor target for the employer
To minimise volatility in funding adequacy all the time without restricting investment to bonds employers will need to ‘over fund’, whereas individuals can fund freely at a rate that assumes a risk premium (as in my examples)
If you think that it looks hard to achieve the £35,000 target in this example when setting a floor at 99% confidence a full 40% below it, imagine how much more onerous it would be if £35,000 also had to be the floor. Even raising the floor in my individual example to £29,750, just 15% below the target, increases the contributions by about 50% – and that is without assuming a fully matched index linked gilt portfolio as the most risk-averse choice.
Even this risk-averse approach is an unreal comparison. My experience is that any individual who finds he or she can afford the luxury of meeting their purchasing power target without risk-taking will raise the acceptable outcome to a level assuming some payoff from taking risk. What is a rational risk preference for an employer, given the new rules, will probably never be rational for an individual.
The accountants are right to restrict firms’ ability to operate with unfunded liabilities and regulators are right to ensure fiduciaries have full backing for long term promises made. If this results in the loss of paternalistic funding, I say we might as well embrace personal responsibility and get on with the job of designing the structures to support and foster it.