still the magic number?

This communication is for financial adviser use only. Structured Products 4% still the magic number? Sustainable spending for UK retirement December 2016

Contents By Leigh Fisher, Intermediary Distribution at Investec Structured Products Leigh Fisher is the London and Home Counties Sales Manager for Investec Structured Products. Having started her career at Schroders in 2002, she then spent 8 years at Neptune Investment Management latterly as Sales Manager in the Third Party Distribution team. Prior to joining Investec, she was Head of IFA Sales for Natixis Global Asset Management. Call Leigh on +44 (0)7525 822 183 or email her at leigh.fisher@investec.co.uk Introduction: Revisiting the safe rate 3 A safe foundation? 4 Far horizons 5 Spending power 6 Factoring in expenses 6 Risk and diversification 7 A question of timing 8 Leaving a legacy 8 The quest for guarantees 9 Structured income and SWR 10 Conclusions: The age of complexity 11

Revisiting the safe rate It s more than 20 years since William Bengen s ground-breaking research into safe withdrawal rates for pension pots. 1 The US financial planner was the first to make his calculations using actual historical market returns, rather than averages. Famously, he came up with a safe level of income that could be withdrawn over 30 years to ensure investors continued to receive an income in line with inflation, without running out of capital. The figure was 4%. Since then, the Safe Withdrawal Rate (SWR) has been refined by Bengen himself and countless others. It has been challenged and criticised; new layers of research have added nuances and caveats; but the 4% figure has stuck as a universal rule of thumb. How useful is it today? As life expectancy rises sharply, and market volatility becomes almost the norm, is an SWR of 4% still a relevant basis for calculations? And, what other strategies could UK advisers be using to protect the income and capital of retirees for today and tomorrow? 3

A safe foundation? Looking at US portfolio performances and the effect of income withdrawal over 30-year periods between 1900 and 2008, Bengen concluded that 4% was the Safe Withdrawal Rate. Bengen s breakthrough William Bengen called it SAFEMAX. It s more commonly known as SWR the maximum safe level of income that can be withdrawn from a portfolio over a 30-year period to achieve an income that keeps pace with inflation, while still preserving capital. Bengen s 1994 research 1 was based on portfolios split equally between bonds and equities. Critically, he used actual results from past points in history, rather than the historical averages which had formed the basis of studies to that point. Looking at US portfolio performances and the effect of income withdrawal over 30-year periods between 1900 and 2008, Bengen concluded that 4% was the Safe Withdrawal Rate. In 96% of cases, following the SWR rule for 30 years leaves a portfolio equal to the amount originally invested. In the remaining cases, the worst scenario is that income payments leave zero capital. A complex picture Since then, further research has explored the impact of various factors on SWR, from portfolio diversification to risk tolerance. This has created a complete, yet complex picture. In general, the 4% rate has proved resilient, though it has not been immune to challenge and criticism. One concern about the Bengen formula is that it relied on US data. When Pfau updated the research in 2010 for 17 developed countries, 2 he found that an SWR of 4% would have worked in only three other countries besides the US. For the UK, Pfau found that even with perfect foresight of the best portfolio combination, the SWR was 3.77%, improving to 4.17% where a 10% probability of failure was acceptable. Whatever the starting point, calibrating the right SWR invariably needs to be with the individual client s needs and circumstances in mind, taking into account factors such as: their retirement time horizon likely changes in spending habits over the drawdown period any desire to leave a legacy the client s attitude to risk the preferred asset allocation mix, and openness to diversification the impact of fees and taxes consideration of annuities and other products that could offer guarantees on income. 4

Far horizons Women aged 65 today have an 11.1% chance of reaching the age of 100. Men of the same age, the chance is 7.2%. The longevity factor Most SWR research assumes a 30-year time horizon. But with life expectancy rising sharply, this assumption may be out of date for those retiring today and in the future. The latest Office for National Statistics (ONS) data 3 shows that women aged 65 today have an 11.1% chance of reaching the age of 100; for men of the same age, the chance is 7.2%. There s a 25% chance that a 65-year-old woman will reach the age of 96. Couples dilemma This has clear implications for anyone facing retirement. Couples face particularly tough choices. If both are aged 65, there is a 25% chance one of them will reach the age of 97, and a 17% chance one of them will reach 100. This puts the onus on advisers to carefully discuss the options, and to adjust the SWR and asset allocation to account for a longer time horizon or indeed a shorter one. Studies by Bengen (1996) 4 and Blanchett (2007) 5 both suggested an increase in SWR by 1% for a 20-year return period, and a reduction of 0.5% for longer periods. In other words, SWR rises to 5% for 20 years, and drops to 3.5% for 45 years. Asset allocation mix For optimal asset allocation, it s wise to consider an increase in equity exposure for longer time horizons, and lower equity exposure for shorter periods. For a couple both aged 65, there is a 25% chance that one of them will reach the age of 97, and a 17% chance that one of them will hit the centenary. At the same time, shorter time periods need a lower equity exposure of around 30% within the asset allocation, as most portfolios are less able to recover quickly from periods of greater stress. This rises to 65 to 75% equity exposure for a 45-year time horizon, due to the longer period of inflationary increases to income, taxes and expenses. Pfau (2012) 6 suggests a slightly lower equity exposure to achieve a 95% confidence level in maintaining the SWR. For cautious clients who are keen to achieve 99% confidence levels, a sound approach is to reduce the equity exposure to 30% or lower for any time horizon, as well as reducing the SWR. 5

Spending power A retiree might adjust spending after age 75 to inflation minus 4%, and then after age 85 to inflation minus 2%. By the end of the period, this client would be spending around 45% less. The ageing effect Spending patterns naturally change with age. Many clients will see their expenses declining in later years, as they become less active and the need for discretionary spending gradually reduces. How should this be reflected in planning? In a 2001 paper 7, Bengen considered how a retiree might adjust spending after age 75 to inflation minus 4%, and then after age 85 to inflation minus 2%. By the end of the period, this client would be spending around 45% less than if they had simply planned to withdraw income in line with inflation every year. As a result, the overall SWR rose from 4.1% to nearly 4.8%. Spending and performance Some clients also find it prudent to adjust their spending in line with portfolio performance. This can be achieved by drawing up a set of decision rules, as suggested in research by Guyton and Klinger in 2006. 8 Their proposed rules would cap inflation increases on withdrawals at 6%, while if the return on the previous year s portfolio was negative, there would be no inflation increase. The same researchers proposed using capital preservation or prosperity rules. These would adjust the income by 10% if it deviated by more than 20% from the original spending path in the first 15 years. Guyton and Klinger found that clients who were prepared to adjust their income in this way reducing spending in difficult markets, and passing up inflation increases when markets were flat could achieve the same overall lifetime spending, but at a starting level that was 20% higher. Factoring in expenses Fee deduction Any fees due to advisers, providers or platforms must naturally be accounted for when calculating a client s income projections. But it isn t just a simple case of deducting the fee from the SWR. According to research by Pye (2001) 9 and Kitces (2010) 10, the effect of expenses on income levels is just 35% of the overall costs. The effect of expenses on income levels is just 35% of the overall costs. That s because over time, the portfolio will be eroded as more and more withdrawals are made. So the percentage fee will steadily reduce, assuming the portfolio is not bucking normal long-term economic cyclical trends. For example, in the first year a 1% fee is 10,000 from a 1m portfolio. But, after 30 years with 100,000 left, the fee is only 1,000. Tax impact Again, the effect of tax is more complicated than simply applying a straightforward reduction of the tax rate, whether 20%, 40% or 45%. Research by Bengen suggested a 20% tax rate reduces SWR by 0.3%. 11 However, Kitces, assuming a higher dividend yield and greater portfolio turnover, came up with a slightly higher rate of 0.66%. 12 Since most UK retirees take income via drawdown arrangements, little tax is paid within the portfolio itself most is post-income withdrawal. That means the client needs to increase their income. So a slightly higher drawdown level is required, which crucially erodes the assets slightly faster. 6

Risk and diversification Some clients are willing and able to accept a greater allocation of equities and probability of failure, in return for a higher initial SWR. Risk: the trade-off Attitude to risk has an important bearing on income projection. Lower risk clients will prefer a lower starting SWR, but some clients are willing and able to accept a greater allocation of equities and probability of failure, in return for a higher initial SWR. As Kitces noted in a 2012 paper, 13 the probability of failure doesn t typically represent the likelihood of a total failure of the plan but does mean the client will have to make spending adjustments at some point. The greater the possibility of failure, the more likely the need for an adjustment, and the more severe the potential adjustment could be. Two separate studies from 2007 considered the trade-off for those willing to accept a 25% probability of failure or adjustment. One suggested an SWR above 5% for clients with a 75% exposure to equities; the second posited withdrawal rates of 5.5 to 6%. 14 Diversification: a cloudy picture Evidence about the effects of extra diversification in a portfolio on SWR remains unclear. A 2003 study found only minimal benefits to global diversification. Reseachers in 2005 found that extreme exposure to global stocks actually reduced SWR. 15 Kitces suggests increasing SWR by 0.5% to 1% for significant multi-asset class diversification. Other studies have considered the use of international equities, international bonds, REITS, commodities and so on. It s currently difficult to provide a definitive list, since data is not widely available for all asset classes to be able to test the full range of outcomes over a 10-year period. Kitces suggests increasing SWR by 0.5% to 1% for significant multi-asset class diversification, with the caveat that limited evidence means the exact value remains largely theoretical. 16 7

A question of timing The main criticism of much SWR research is that it fails to account for the timing of the first withdrawal, which has a big effect on the starting balance. Consider the example of Mr A and Mrs B. They start out with the same portfolio of 1m, but Mr A retires now and takes an initial income of 45,000, while Mrs B decides to work a further year before taking any income. The first year is a gloomy one in the asset markets, causing portfolios to fall by 15%. SWR research fails to account for the timing of the first withdrawal, which has a big effect on the starting balance. Mrs B s first year of income therefore starts at 38,250 that is, 4.5% of what is now a portfolio of 850,000. Mr A, by contrast, can increase his income by inflation. Assuming inflation is 3%, he takes 46,350 in the second year. However, Mr A is enjoying a higher income, even though his portfolio is now the smallest ( 811,750 after his year 1 withdrawal and the 15% market decline). Kitces solution to this is to make clients who retire after a bear market eligible for a higher SWR, albeit on a lower balance, than those who retire before market decline. Leaving a legacy Recent legislation has made it far more attractive to pass on assets free from inheritance tax, using pension wrappers. Recent legislation has made it far more attractive to pass on assets free from inheritance tax, using pension wrappers. To achieve this, it s as important to maintain the value of the asset itself as to save the inheritance tax charge. Structured products as an investment vehicle can play a useful role here, since many defensive payoff profiles incorporate capital protection features. Bengen (2006) 17 has showed that the SWR only needs to be reduced by 0.2% to leave the principal sum the investor began with 30 years earlier, whether this is for legacy purposes or simply to provide a safety net where the client outlives the investment period. 8

The quest for guarantees To annuitise or not? Annuities are seeing a resurgence in popularity. After a striking move towards drawdown agreements when the new UK pension freedoms came into effect in April 2015, FCA statistics suggest a recent rise in clients buying or considering annuities. This is probably due to the volatility in the markets in the intervening period, triggering a desire by investors to seek out plans with guarantees or protection. The volatility in the markets has triggering a desire by investors to seek out plans with guarantees or protection. An income fund (bond, property or equity income) offers clients total flexibility, but with no protection for the income or capital. In fact, the evidence suggests there are few or no benefits on SWR from traditional annuities. Studies find annuitising part or all of the portfolio has little positive or negative effect on the ability to sustain income, compared to that of an invested portfolio. This is mainly due to the poor annuity rates now available, particularly for index-linked annuities. Guaranteed products: the pros and cons An income fund (bond, property or equity income) offers clients total flexibility, but with no protection for the income or capital. For clients seeking a degree of certainty in a volatile market, it s worthwhile weighing up the alternatives when planning a portfolio. With an annuity, the level of income is known from the start and is guaranteed over the annuity period. Capital is either lost to the annuity provider, or in the case of a temporary annuity, returned with a loss of capital. Variable annuities are more expensive: the cost of guarantees reaches 2% per year. This is high in relation to the negative effect that they have on any growth prospects of the underlying portfolio. The result can be the long-term stagnation of the portfolio and income. But, variable annuities do offer the flexibility to start or stop income payments, albeit within prescribed limits. Non-contingent structured income plans offer fixed income streams with protected capital, without having to give up access to the capital. As with an annuity, the income is payable regardless of the performance of an underlying index level. But the cost of these guarantees compare well to those of annuities, since the investment bank can hedge daily to mitigate costs. The disadvantage is that capital can be at risk if the index level falls below a certain threshold (usually 40 to 50%) or if the issuing bank defaults. 9

Structured income and SWR Their flexibility enables advisers to tailor them to fit each client s risk appetite and individual circumstances. Best of both worlds Structured income plans offer all the benefits of income guarantee, but within the framework of a drawdown portfolio. Their flexibility enables advisers to tailor them to fit each client s risk appetite and individual circumstances. They can also boost SWR. Contingent income plans are available offering income levels of up to 7%. Non-contingent plans, such as the Investec FTSE 100 Enhanced Income Plan, gives a fixed monthly income of 0.42% per month, or 5.04% per annum. Capital is at risk if the FTSE 100 falls by over 50%, or if Investec defaults (the bank is rated A2 by Moody s as at August 31, 2016). But even if the market falls by over 50% during the term, the client will still receive 30.24% in income. The capital returned will be reduced on a one-for-one basis. Defensive kick-out To add more certainty to the UK equity portion of a portfolio, a product such as the Investec FTSE 100 Defensive Kick Out Plan 34. The plan pays a positive return, even if the FTSE 100 Index drops by up to 10% from its initial level on any anniversary date. Provided the FTSE is above 90% of its initial level at anniversaries 3, 4, 5 or 6, the plan will return 8%. Recent history shows how this Plan can deliver. The six most recent plans to mature have delivered a positive return, against a falling FTSE index. Clients who had invested in equity over this period would have seen their capital reduced. The plan proves its worth in a rising market too: 11 of the 19 maturities to date have paid a higher return than the FTSE 100 Total Return Index. 10

Conclusions While the 4% rule remains a useful baseline, it offers no more than a starting-point for a detailed retirement planning exercise. The age of complexity SWR depends on a multitude of variables. While the 4% rule remains a useful baseline, it offers no more than a starting-point for a detailed retirement planning exercise. The state of the market at the time of first withdrawal, the impact of fees and taxes, the optimum portfolio mix will all affect retirement planning; above all, the client s individual circumstances, ambitions and attitude to risk will weigh on the solutions chosen. Bengen saw the process of integrating all these factors as a layer cake, where the base withdrawal rate would be adjusted up or down according to each factor. Kitces, however, has warned that even this approach may be too simplistic, pointing out that many of these factors were evaluated in separate research studies, and their interdependencies have not been properly explored. The evidence serves to underline that in the age of so-called robo-advice, the role of human advisers is, if anything, more critical than ever. Investors need the adviser s expertise to help them plot a path through a complex web of factors to achieve the most sustainable income. And client and investor alike need to monitor these variables continuously to ensure the portfolio stays on track to deliver a comfortable retirement and more importantly, a sustainable one. 1 Bengen, Determining Withdrawal Rates Using Historical Data, http://www.retailinvestor.org/pdf/bengen1.pdf 2 Pfau, An International Perspective on Safe Withdrawal Rates: The Demise of the 4 Percent Rule? http://www3.grips.ac.jp/~pinc/data/10-12.pdf 3 Office for National Statistics, Life Expectancies https://www.ons.gov.uk/peoplepopulationandcommunity/ birthsdeathsandmarriages/lifeexpectancies 4 Bengen, Asset Allocation for a Lifetime http://connection.ebscohost.com/c/articles/5560373/asset-allocation-lifetime 5 Blanchett, Dynamic Allocation Strategies for Distribution Portfolios: Determining the Optimal Distribution Glide Path original not available online 6 Pfau, Capital Market Expectations, Asset Allocation and Safe Withdrawal Rates https://www.onefpa.org/journal/pages/capital%20market%20 Expectations%20Asset%20Allocation%20and%20Safe%20Withdrawal%20 Rates.aspx 7 Bengen, Conserving Client Portfolios During Retirement, Part IV http://connection.ebscohost.com/c/articles/4460514/conserving-clientportfolios-during-retirement-part-iv 8 Guyton & Klinger, Decision Rules and Maximum Initial Withdrawal Rates http://cornerstonewealthadvisors.com/decision-rules-and-maximum-initialwithdrawal-rates/ 9 Pye, Adjusting Withdrawal Rates for Taxes and Expenses, http://connection.ebscohost.com/c/articles/4349405/adjusting-withdrawalrates-taxes-expenses 10 Kitces, Investment Costs, Taxes and the Safe Withdrawal Rate, behind paywall at kitces.com 11 Bengen, Asset Allocation for a Lifetime http://connection.ebscohost.com/c/articles/5560373/asset-allocation-lifetime 12 Kitces, Investment Costs, Taxes and the Safe Withdrawal Rate, behind paywall at kitces.com 13 Kitces, The Next Generation of Monte Carlo Analysis, behind paywall at kitces.com 14 Spritzer, Strieter, Singh, Guidelines for Withdrawal Rates and Portfolio Safety During Retirement http://www.cfapubs.org/doi/full/10.2469/dig.v38.n2.34; Cooley, Hubbard, & Walz, Retirement Savings: Choosing a Withdrawal Rate That Is Sustainable https://incomeclub.co/wp-content/uploads/2015/04/retirement-savingschoosing-a-withdrawal-rate-that-is-sustainable.pdf 15 Cooley, Hubbard & Walz, Does International Diversification Increase the Sustainable Withdrawal Rates from Retirement Portfolios https://www.researchgate.net/publication/240621321_does_international_ Diversification_Increase_the_Sustainable_Withdrawal_Rates_from_Retirement_ Portfolios; Ervin, Filer & Smolira, International Diversification and Retirement Withdrawals http://www.emeraldinsight.com/doi/abs/10.1108/19355181200500006 16 Kitces, 20 Years of Safe Withdrawal Rate Research, behind paywall at kitces.com 17 Bengen, Asset Allocation for a Lifetime http://connection.ebscohost.com/c/articles/5560373/asset-allocation-lifetime 11

Structured Products For latest plan information visit www.investecstructuredproducts.com Call our team on 020 7526 9216 Important information The Plans referred to herein are not sponsored, endorsed, or promoted by the index providers, and the index providers bear no liability with respect to any such Plans or any index on which such Plans are based. The prospectus and relevant Plan brochure contains a more detailed description of the limited relationship the index providers have with the license and any related products. Investec Structured Products is a trading name of Investec Bank plc. Investec Bank plc (Reg. no. 489604) is authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority and the Prudential Regulation Authority and a member of the London Stock Exchange. Registered at 2 Gresham Street, London EC2V 7QP.