In How Much is Enough? I explained why a portfolio of 25x your annual spending is (probably) enough never to have to work again.
If you spend £25,000 per year, this implies a portfolio of £625,000 is enough. This is just a 4% Safe Withdrawal Rate (SWR), expressed another way.
There are different views on this. Plenty of gloomy people out there will tell you that the SWR is less than 4%. But The Escape Artist is back to tell you why those SWR pessimists are probably wrong.
The SWR is not some magical number plucked out of the air by academic gurus like a rabbit from a magicians hat. The SWR is based on the long term returns that can be generated by a portfolio. The higher the long term returns, the higher the SWR. My key point is this – the SWR is determined in part by the starting level of valuation of a portfolio.
Imagine you have a diversified portfolio of 25+ high quality stocks; highly cash generative blue-chip companies which have survived a century or more. Lets focus on the valuation of this portfolio as measured by the free cash flow yield and assume this is 6%. This means a £0.5m portfolio is generating Free Cash Flow available to Equity shareholders (FCFE) of £30,000 a year (£30,000 is 6% of £500,000).
The return on equities is delivered by a mix of dividend income and share price growth. The FCFE not paid out as dividends is held within the company. It can be used for paying down debt, making acquisitions, buying new factories or buying back shares. If you have any faith in capitalism, you will accept that FCFE (whether distributed as dividend or retained in the company) will benefit you as a shareholder. All other things being equal, every £1 retained in the company will increase the market value of the equity by £1.
We can use the FCFE yield plus a growth assumption as a way of estimating future returns from a portfolio. If the FCFE yield is 6% and you expect the underlying cashflows attributable to those stocks to grow at 3% per annum, then the expected return on the portfolio is 9% per year.
Its easy for me to imagine this scenario. That’s because the current FCFE yield on my active share portfolio is about 6%. The portfolio includes some of the best companies on the planet: resilient businesses which have done very well thank you for the last 150 years or so, surviving wars, depressions, booms and busts.
Obviously the actual returns will be different from the expected returns. Actual returns exhibit random volatility as they rise and fall with changes in valuation. If share prices rise, the portfolio FCFE yield falls and the actual returns will be higher (and vice versa). The reason that investing can feel scary is that, in the short term, the “noise” of share price volatility outweighs the “signal” of expected returns based on fundamentals. Over the long term though, the actual returns converge towards the expected returns.
So a fair guess for a portfolio 100% invested in these stocks, would be that it could support an annual withdrawal rate of ~9%. This will sound ridiculously optimistic to many but my active equity portfolio has delivered 12+% annualised returns for 18 years now over a period which included 2 whopping bear markets. Yes, its possible that The Escape Artist is a lucky monkey throwing darts at the Financial Times but it does highlight that 9% per year is not impossible.
But is it also possible that my portfolio companies cash flows could fall – say, in a recession? Of course, which is why I would never rely on withdrawing 9% per year. By assuming a 4% SWR, I am factoring in a margin of safety*.
So perhaps the withdrawal rate on an all-equity portfolio could be as high as 9% a year without eating into the capital. What about portfolios that include a fixed income element?
Lets assume a 75% equities : 25% government bonds portfolio (my current asset allocation). If UK longer dated gilts have a yield to redemption of about 2% per year, the expected annual portfolio return now becomes:
Expected portfolio return = (75% x 9%) + (25% x 2%) = ~7%.
So on this basis, in the absence of major market falls, I might expect that the portfolio could support a withdrawal rate of perhaps 7% per year. This still offers a margin of safety over a 4% SWR assumption*.
At this point we need to think about the impact of market risk (sometimes referred to as sequence of returns risk for retirees). We can’t wish away volatility, it is a fact of life. But its important as an investor to distinguish between 2 different forms of risk:
1. Market risk
2. Business risk
Market risk is the risk that asset prices fall. If you have to sell shares / units to meet your living expenses, then a stock market crash entails the risk you will be forced to sell more shares at low valuations to raise the cash you need to live off.
The best example of stockmarket risk is the 1987 crash when the US stockmarket lost 30% of its value over a few days…for no good reason other than there were more sellers than buyers. If you were Frankie the daytrader and you were long and leveraged, this would have been an extinction event for you. But, as the market quickly recovered, this would have been no problem for a FI early retiree that kept their head. You can help protect yourself from stock market risk by controlling your inner chimp and ensuring you do not panic-sell during a market crash.
No bear market lasts forever, so if you have enough cash to be able to fund say 3 years living expenses then you are largely protected from market risk.
If your investment income matches or exceeds your spending then you never have to sell units / shares and so the level of the stock market is of little interest to you other than a depressed market offers you the option to pick more cheap stocks and boost your income.
The concept of flexibility is important here – if you can reduce your spending when necessary, you have much more safety margin as you wont need to sell units in a market crash. This is why MMM emphasises a flexible and resourceful mindset and the ability to cut spending.
Real business risk is the risk that the companies that we own encounter adverse shocks that reduce their cash flows and dividends. But this risk can be minimised by diversification and / or by the careful selection of shares in businesses that have high levels of resilience and longevity.
What about investors in index funds? Its harder to get data on the FCFE yield on an index. But the concept still applies. Investors in funds that track over-priced indexes should expect lower future returns and a lower SWR. Investors in cheaper indexes can expect higher future returns and a higher SWR. We can use other metrics such as price : earnings multiples or dividend yield to assess the level of valuation of an index.
The cyclically adjusted price earnings (CAPE) multiple on the S&P500 is currently about 27x or about 70% higher than its long run average of 16x times. That’s why I don’t currently own a S&P 500 tracker.
At these prices, it makes sense for US focussed investors to at least consider:
1) lowering their previous estimate of the SWR (maybe),
2) adjusting away from expensive, low yielding bonds (probably) or
3) getting international equity diversification (almost certainly).
Flexibility in investing (as well as spending) matters. Risk can be mitigated by periodically rebalancing a portfolio away from more expensive indices and towards cheaper ones. The SWR studies I’ve read don’t seem to reflect that FI investors can invest flexibly; avoiding over-priced markets and finding value elsewhere in the world using an asset allocation that responds to relative valuations. FI-seeking investors don’t need to make pessimistic assumptions about the SWR if they are internationally diversified and have an appropriate asset allocation.
The VHYL ETF gives UK investors an easy way to gain diversified global equity exposure. The dividend yield alone is almost 4% on this ETF. Given that the dividend on this grew 5% between H2 2013 and H2 2014, this implies an underlying return (before changes in valuation) of almost 9%. That fact alone makes me question whether people pushing a 3% SWR are being too pessimistic.
To summarise: the Safe Withdrawal Rate is not a fixed magic number that is right in all circumstances.
Firstly, the cheaper your portfolio compared to its underlying sustainable cash flows, the higher your expected future investment returns and your SWR will be.
Secondly, the SWR also depends on whether your living expenses are covered by the income from the portfolio. If they are you (or you have the flexibility to adjust spending and avoid being a forced seller) then you can effectively ignore stock market risk.
Thirdly, international diversification and a value based asset allocation should allow a higher SWR.
So be careful when you read stuff that says a 4% SWR is too optimistic. Yes, its important not to be reckless, but there are also risks in being too conservative. If you keep working a job you don’t like, you risk wasting your life and damaging health and happiness.
As I’ve said before, debates around the SWR make a bunch of implicit assumptions. They assume that once you escape, you will never earn any other paid income ever again, never get a state pension or other benefits and never reduce your spending as you get older. These are incredibly conservative assumptions.
So remember to cultivate rational optimism. If you have a sensible portfolio, assuming a SWR less than 4% may be unnecessarily prolonging your stay in the Prison Camp. Why hang around longer than you need to when the gate is open and the guards are looking the other way?
* If we want to compare a real SWR with real expected returns then we can adjust for annual inflation, which in the UK is currently about 1% (RPI = 1.1%, CPI = 0.3%). So a nominal return of 7% becomes a real return of 6% which still leaves a margin of safety.