How should you manage your money when you retire? Should your portfolio change when you finally sign your F.U. letter to the boss?
Is the famous 4% rule really safe or is there a better way?
While the passive path to accumulating your pension pot is well lit by blogs, books, and preachers of the gospel, the more difficult question of how to safely ration your retirement savings has no simple answer.
Attempts to supply a silver bullet to retirement spending often flounder. Proposed solutions may be unrealistic, mistranslated, too narrow, or grossly oversimplified on their journey from academic journal to custom and practice.
Michael McClung’s achievement is to survey that landscape with the rigorous eye of an engineer who wants to build a house that won’t fall down.
He’s poured his findings into Living Off Your Money. It’s a practical, safety-conscious, and evidence-based manual that DIY investors can use to avoid the retirement quicksands.
Hazards ahead
One big thing lifelong savers need to grasp as they contemplate retirement is that we become more vulnerable as we rundown our stockpile.
An unfortunate sequence of returns can put us on a crash course early on. Inflation and even the blessing of a long life can put us on prison rations in our twilight years.
The situation is worsened because traditional retirement rules-of-thumb like the ‘4% rule’ are about as reliable as ‘red sky at night’.
The 4% rule is prone to failure, numerous caveats that don’t fit into 140-characters or fewer (or even 280), and it’s barely applicable outside the US. And where the 4% rule can leave some retirees on the brink of poverty, it can leave others departing the stage with most of their hard-earned loot unspent.
The system offered by Michael McClung takes a data-forged sword to those twin-headed terrors. His design relies upon two important techniques that many retirees may struggle with:
- Dynamic asset allocation
- Dynamic withdrawal rate
Dynamic asset allocation means that your yin and yang of equities and bonds is no longer fixed by some permanent cosmic ratio. Instead, your percentages can pitch up and down depending on the motions of the market.
A 50:50 portfolio could, with McClung’s system, average between 30%-70% equities over the course of a retirement.
In extreme conditions you could end up with 100% of your portfolio in equities. Conversely when equities are storming ahead you’ll convert them into high-quality bonds, ensuring there’s fodder in the barn for when winter comes. And when equities are blown away like dandelions in a category five hurricane you’ll live on bonds until they’re gone. There’s no automatic annual rebalancing here.
With a dynamic withdrawal rate, your income rate can also vary every year.
A tempestuous retirement could see withdrawal rates swing between 2.5% and 6%. Benign conditions might bless you with an average withdrawal rate of 7.7%. When your portfolio swells, a dynamic withdrawal rate lets you spend more. When conditions worsen you batten down the hatches.
All this may make the system sound random, but it’s rather that the plan flexes in response to market feedback. It gives you a brake and an accelerator to apply rather than putting you on rails until your retirement train terminates.
If that sounds like market-timing, it isn’t.
Trial by data
Living Off Your Money builds on the work of other retirement researchers. (These guys have lower profiles than North Korean late-night comedians, and are probably only familiar to you if you’re into obscure financial planning journals.)
All have sought to improve upon the cult of 4% inflation-adjusted withdrawals plus annual rebalancing.
For his part McClung reverse-engineers their systems, tests them to within an inch of their algorithms, and then bolts together the best parts to come up with his recommendations.
The major difference between McClung and most other retirement researchers is that McClung has subjected these formulas to more tests than a talking ape.
Standard practice is to pit your proposals against the historical performance of US equities and bonds and leave it at that.
The danger is that a system that worked well when US assets outgrew those of most other nations may not look so clever when planted in poorer home soils. Even US investors may not enjoy such sunny days again. Non-US residents have no reason to expect to.
McClung guards against this by testing his contestants against the UK and Japanese datasets. Neither has enjoyed the same hot-hand as the US.
No retirement strategy trumps all others, everywhere, every time. Optimisers are missing the point – you might as well try to optimise a baby. What works in one situation won’t always work in another. McClung acknowledges this and recommends a plan that:
- Works well during historically difficult retirement periods
- Is robust across geographies
- Maximises withdrawals
- Avoids catastrophic failure like a zombie plague
- Leaves a large margin for error
He doesn’t stop there. McClung also checks his system versus the chilling effects of a low-growth world. His recommendations assume a globally diversified portfolio and performance. McClung’s mindset is world-first, not America-first, which makes his work directly applicable to UK investors in a way that most retirement research isn’t.
The Living Off Your Money strategy can also be calibrated for shorter and longer retirements. That is especially handy if financial independence is on your ‘to do’ list.
Don’t misunderstand me – McClung isn’t claiming his method is fail-safe. Very few retirement strategies would look good after a dose of German-style hyperinflation and being on the wrong side of two World Wars.
There are no guarantees, only probabilities.
The downside
There’s always a downside in investing and the trade-off demanded of you by the Living Off Your Money approach to retirement spending is that you can tolerate a volatile income and asset allocation.
Yes, you’ll probably be able to spend more over the course of your retirement. But there will be times when you’ll need to spend less. (The reality is that many retirees do naturally vary their income anyway outside of the confines of the retirement researcher’s lab.)
Sticking to the plan may also mean going all-in on equities in extreme conditions. Many retirees couldn’t cope with those strains.
To help alleviate some of these issues, McClung explains ways to take the edge off his purest prescriptions.
Floors and ceilings can be used to contain your equity allocation. There’s also an extensive section on creating guaranteed income to cover the bills when your withdrawal rate dips alarmingly low.
You may need to work longer to be able to afford such optionality. That’s the price of sleeping well at night.
Easy doesn’t do it
While McClung is a master of retirement theory, he doesn’t wallow in it. He never loses sight of his goal of creating a book that can genuinely help people.
The explanations are clear, and McClung carefully ropes off step-by-step practical sections that can be chewed on separately if you’d rather skip the methodology hors d’oeuvres.
Yet his work is steeped in integrity. McClung goes to great pains to explain his guiding principles and assumptions and – unlike some financial writers – all of his recommendations can be fulfilled in real life. There’s even a spreadsheet on his website to support anyone who wants to implement his strategy.
None of this changes the fact that reading this book and managing your portfolio by its light requires a fair degree of investment literacy.
The truth is, nobody should manage their retirement investments without a strong financial education and Living Off Your Money can help school anyone, regardless of whether you ultimately apply its teachings.
Long-term Monevator readers will be in their element. But if you just want to get by with a couple of blog posts and a few simple rules that could be printed on a tea towel then this isn’t the plan for you. Your best bet would be to accumulate so many assets that you are left with plenty of room for error.
On the other hand if you have a strong risk tolerance, genuinely enjoy engaging with investing, and want to do more with less then McClung might just change the course of your retirement.
If you’re not sure which camp you fall into then McClung has made three sample chapters available for free.
Alternatively, check Living Off Your Money out on Amazon and let us know in the comments below what you think of it.
Take it steady,
The Accumulator
Comments on this entry are closed.
Great to see your review from a UK investing perspective!
The sample chapters are thought-provoking stuff. Even if you don’t want to implement complex rules and would rather stick to a static asset allocation and withdrawal rules, the book provides a comparison against the fancy approaches that’s worth looking at.
My takeaway was that a static 60/40 allocation is good enough, but I hope to read the book in full soon to dig deeper into McClung’s research.
See also the discussion that ensued on Bogleheads about this book:
https://www.bogleheads.org/forum/viewtopic.php?t=192105
Interesting review. 30 years experience dealing with real life clients in real life market conditions tells me that it will be an individual’s emotional tolerance to risk which is the biggest factor in their investment success. In my experience, and as you rightly point out, very, very few people can handle a higher level of volatility than their psychometric risk profile indicates they can cope with.
I’ll have to read the book, but an alternative approach which has been proven to work is to set out, in writing, what you will do if ‘x’ happens – x being some kind of event that’s not run of the mill. So, for example (and assuming risk tolerance et al are in line), “If the value of our portfolio drops by 20% we will reduce withdrawals by 25% for up to 6 years”.
By considering courses of action before an event happens, and having that written strategy to fall back on a) enables you to plan in advance (e.g. by holding higher cash /bond reserve) and b) means you know exactly what to do if the event happens. You can model various scenarios in a lifetime cash-flow forecast and see what levers would need to be pulled and select the option which works best. It might mean doing something in the short term (e.g. working longer) you hadn’t envisaged, but at least you have the time to make the adjustment, and are contingency planning with a clear head not fogged by the panic of suddenly having to take corrective action.
Ultimately it’s all a trade off, as you say. And no amount of testing algorithms can tell you how real human beings will react in a real life situation. Everyone is different, there’s no one size fits all.
If you are chopping and changing asset allocations according to market conditions,this is ‘active investing’.
This sounds to me very sensible. Pre retirement life is all about dynamic allocation – our earnings and outgoings are never the same from one year to the next unless you never change jobs or get promoted or get married or buy a house – so why should post retirement be different? The equivalent variables are portfolio size, and health & life expectancy.
The 4% rule looks like a useful rule of thumb which has been made to bear a lot more weight than it can stand. If we are going to use it at all, again, the rational thing to do would be to apply it dynamically by resetting twice a year and adjusting expenditure to 4% of adjusted portfolio value, and adjusting further for health to the extent that it affects our propensity to spend money, and our life expectancy.
Amazon link fails unfortunately
Thanks for bringing this book to my attention. I have skimmed the 3 chapters (mainly the recommendations and intro) and researched reviews of Prime Harvesting. For me that will be enough as I won’t read the 300+ pages of why-so.
I like the harvesting approaches of skimming off the excess equity to put into less risky bonds; I was thinking of doing the same in anticipation of a market drop sometime soonish. I downloaded the spreadsheet so I can test it out as I like the discipline it brings.
A 60/40 split seems a reasonable asset split, but with the current UK tax environment and me hoping for a circa 40 year retirement I think I need to keep more equities. My risk appetite is comfortable with that, I think.
I just need to make sure I have the right assets in those classes, and my costs are reasonable value for money.
Re: active investing – if you’re chopping and changing because you *think* the US is overvalued after you read a few news stories or emerging markets is hot right now cos it did well over the last year then I agree that’s active investing.
If however, you are using a mechanical set of rules with tightly defined triggers that have basis on evidence then I disagree – that’s not active investing. Even if you use a global portfolio then your asset allocation changes all the time as the market’s opinion changes. It also changes according to the decisions made by the S&P 500 committee or the guardians of the FTSE. Even if you lifestyle then you are making changes to asset allocation – but in accordance to a defined plan, not on a whim.
@ Anthonyh – the book isn’t 300 pages justifying prime harvesting. Prime harvesting is one aspect but variable withdrawal rate and how to set your withdrawal rate in the first place are key (along with guidance on floors and ceilings). The guaranteed income section is also strong. McClung’s asset allocation evidence also indicates that more than 60% in equities is pretty risky. (At the outset).
I would thoroughly recommend having a look through ERN’s analysis of safe withdrawal rates if you haven’t already, where he also dives into some of McClung’s ideas: https://earlyretirementnow.com/2016/12/07/the-ultimate-guide-to-safe-withdrawal-rates-part-1-intro/
It’s extremely impressive and includes a lot interesting ideas backed up by lots of data. I have no affiliation, I just thought it would be of great interest to many Monevator readers 🙂
May have pressed a button prematurely there. This is all worthy stuff and yet I do feel increasingly unconvinced that we will ever see our retirements in the context of a universe where the past acted even in broad terms as a guide to the future. Not after Brexit and N.Korea have finished with us and spat out what is left. I don’t mean to sound like Eeyore (that donkey knew a thing or two) but it is all really pretty jolly depressing. I always try to look on the bright side of things, but that is rather difficult when the best presents you can think of for your children are a vegetable plot (for Brexit, so we can grow more food like the Government chap said we should) and a hazard suit (in case the one with the interesting haircut gets antsy).
Thanks for highlighting this book and the concept of prime harvesting. I’ve read the free extract and ordered a copy as I’m keen to understand the whole story, as I regularly get asked by friends and family about finance and pensions.
It’s good to see someone has taken the time and effort to test the various methods of extracting cash in retirement, and I’m surprised to see how poorly the age related method performed.
“Pick your target retirement length based on life expectancy, typically 30 to 40 years. ”
But 50% of people will outlive average life expectancy.
David Swensen in pioneering portfolio management writes with practical experience (at Yale) on managing withdrawals for a charitable educational endowment, a typical target spending rate is 4.5% pa. the dynamic withdrawal rate used has worked well for decades, albeit on a larger scale but is fairly simple.
Where Income is based on a 4.5% withdrawal rate applied on 20% of this year’s portfolio value, say 18% of lasts years value , 16% of the years before ( you can tweak the % figures to suit) but the principle is to base the rate on previous years values with a weighting for the more recent years.
I have adapted a version for myself as a guidance and the concept is fairly simple and dynamic spending is a very useful concept for anyone moving on an “income pot of assets” I’ll take a look at the book sounds interesting.
The problem ultimately is that this type of modelling requires a level of knowledge and interest in the topic, that is probably found in around one person in a thousand …
I worry that McClung is over complicating things, and am sure the boggleheads are. Doing 50 year forecasts with only 100 years of past data for testing the algorithms is scientifically weak, and shows that economics isn’t really a science. In one sense it’s good that the author avoids tax, as that would give a us bias, but it’s key to uk plans, and going 40 years back shows that the rules could change so much 40 years hence. Similarly market fluctuations pre computers based on human frailty may not apply in our future.
All the schemes sound practical for 50somethings, but not 80somethings as our faculties fade.
I don’t think anyone should obsess about raising a 90% change of success to 95%, and just focus on a backup plan if the headwinds gather. So a variable spending plan with a floor covered by the state seems the simplest approach.
@dearieme – “50% of people will outlive average life expectancy”
Not quite that simple, e.g. 5 x 70 plus 1 x 76 = 71 average but only 1 above average.
I too have ordered the book. I don’t understand the criticism of Prime Harvesting being too complicated. The principle is quite simple: sell stocks when they are doing well and squirrel away the profit in bonds from which you draw an income, don’t sell stocks when they are not doing well and continue to draw your income from bonds until, potentially, they run out at which point you draw from stocks, replenish your bonds when stocks are doing well again.
I can see the limitation of the data sets for testing the idea but against that you have to take into account that the strategy “makes sense” and isn’t a result of arbitrary data mining.
“Pick your target retirement length based on life expectancy, typically 30 to 40 years. ”
But 50% of people will outlive average life expectancy…….
It doesn’t say based on “average” life expectance, it says “life expectancy”; 40 years is 105 for a 65 year old.
“Not quite that simple”: indeed
“e.g. 5 x 70 plus 1 x 76 = 71 average but only 1 above average” but not quite that foolish either.
“It doesn’t say based on “average” life expectance ..”: ‘life expectancy’ is a well-defined term. Making up his own meaning – or your own meaning – is rather silly.
The problem with these sorts of schemes is that what a portfolio looks like at any point in time (equity to bond weight) will depend on when you started to drawdown. For example, if I started drawing down 2 years ago the equity/bond ratio I would have now could be quite different to how I would choose it to be if I started drawing down now. That is illogical.
I keep things very simple. Choose a number of years N, that I want my portfolio to last. Divide portfolio value P by N and that is my budget drawdown for the next year. The budget is not a rigid amount, but if I go overbudget I do so in full knowledge that my current drawdown rate may be unsustainable if I want the pot to last N years. To work the other way, if I spend X over the year, then my estimate for how long the portfolio might last would be P/X.
“Making up his own meaning – or your own meaning – is rather silly….. “
We can all have our views about who is being silly.
Here is the full quotation:
“Pick your target retirement length based on life expectancy, typically 30 to 40 years. Websites such as https://personal.vanguard.com/us/insights/retirement/plan-for-a-long-retirement-tool can identify the age where the life-expectancy probability is between the recommended amounts of 1% to 5%.”
Where it is clear the author is talking about life-expectancy probability between 1% and 5%, not average life expectancy.
@dearieme
No matter what parameters it derives from, nor what name is attributed to it, a data set will only give, with certainty, the precise percentage result you claim in the wholly exceptional case where the distribution of the data in the set is perfectly symmetrical about the mean.
@dearieme. Yes, but you can use the interim life tables available on the net to do your own percentile calcs. I have spent many an otherwise unoccupied hour at work enumerating the boundaries own mortality. I must be fun to know.
At 65, life expectancy for a man is about until 83. 35 years takes you to 10% territory. Of course, with a younger spouse to provide for, need to cover that. So in my case after 28 years there is still a 50% chance one or both of us will be blogging but even in this case, 37 years gives 10% chance.
Only in pension speak is an early death a winning ticket 🙁
@ John B – simplicity when it comes to retirement often equates to paying a higher price – in terms of years worked / pot accumulated. I think McClung has made it as complicated as it needs to be to find out how to maximise your income from a given stash. Jim Otar (author of Unveiling The Retirement Myth) came up with a traffic light system to categorise near-retirees who fell into one of three zones:
Green = loaded = nothing to worry about (these guys can keep it as simple as they like)
Grey = just about managing = strong returns will see them alright, poor returns could put them in trouble (McClung’s book is of particular value to this category).
Red = Nowhere near enough savings = in trouble, urgent action required.
My dash for FI will put me firmly in the grey zone and I’m more than happy to learn withdrawal rate kung-fu in exchange for more year not in the office.
I have read the 3 free chapters and I am very impressed. My current strategy, essentially annually rebalancing a 60/40 portfolio and varying withdrawal rate according to how the portfolio performs is clearly sub-optimal according to this book. The long term withdrawal rate my wife and I desire is quite a low percentage of the current value of our investment portfolio, so although our current strategy is likely to be fine, we would like the option of giving away more of our assets if the investments perform well enough, so if there are better ways that are evience based I am ok with that.
I have ordered a copy of the book. Thanks to those who posted the links to other sites on here. I will wade through these as well.
Hi Naeclue, yes those initial 3 chapters sold the book to me too, and McClung keeps up the rigour and craftsmanship all the way through.
Well this is especially interesting for me, as it seems to reflect what I do, and I have always felt rather out of step with the investing community (and especially the retirement investing subset) as a result.
I retired twelve years ago at 43, with a nice lump sum from the sale of a business, straight into the teeth of the last financial crisis.
Licking my substantial wounds, I set about enhancing my cash buffer so as not to be pushed ever again into raiding equities when they were struggling.
I use cash instead of bonds, keeping five or six years of outgoings in that form, and topping up that bufffer from investments only when they have grown well.
I intend to let that reserve fall as low as two years of outgoings when things take a downturn, and maintain them between one and two years until things pick up again and I can replenish them fully.
Most would decry keeping so much in cash, but it means I can ride out pretty much any storm.
Flexible (dynamic) asset allocation and withdrawals.
@ Naeclue – sounds like a sensible approach 🙂
@TA – are we due an update on your dash to retirement? I enjoyed hearing about the topic from a ‘normal’ person (I.e. not a tech millionaire). Please keep us updated, your posts on the topic to date are some of the most human I’ve read and were a refreshingly British antidote to all the Americans!
Hey thanks, L. That’s lovely to hear. Writing ‘Monevator the book’ has been taking up most of my blogging time but I’m now at the 70% mark. Partly thanks to McClung’s system enabling me to knock a sizeable chunk off my target figure. A wise person once told me that the first half of any goal is all slog but the second half whizzes by like you’re running down hill. I think that’s true and am feeling happier and stronger with every milestone that passes into the rear-view mirror.
Well I’ve bought the book. It’s bigger than I was expecting and crammed full of data…this will take sometime to read.
I have generally overthought the whole issue of how to invest for capital protection and decumulation. Thinking 3 years cash plus liquity buffer, say 15%, 30% IL, direct holding plus the l&g global fund, then the rest in GLS 80:20 and s*d it!
Not sure Lars K should read pages 13 and 14…..
It is very US centric which means a lot of the detail and finer calculations are not directly applicable. Where UK or global data is used, this is often by way of back testing a model generated on US data.
The ideas about asset allocation and drawdown strategies are interesting but the asset allocation advice is very intricate and again US weighted, rebalancing across 10 or 12 equity funds would be tricky!
So, could benefit from a thorough edit ( even the US don’t have a word negitive) and unduly prolix, but gives a framework as to how to engineer a portfolio for decumulation and is an interesting read anyhow.
McClung tests with UK data is to see if his assumptions stand up under UK historical conditions. They do. The system is then tested again using reasonable assumptions to account for a global approach. Naturally an American writer writing for the far larger US market is going to be writing for a US audience first, but it’s not hard to translate to UK circumstances. Do you know of any significant work on deaccumulation strategies that’s written primarily for a UK audience?
The asset allocation advice offers options to suit your taste from complex to simple. Again, it’s not hard to convert to the UK market and you don’t have to choose the most complex allocation option because he gives you alternatives. In fact, that’s one of the most brilliant parts of the book – McClung offers a clear menu of choices so you can customise his recommendation to suit your individual circumstances.
Yes, there are a couple of spelling mistakes and he writes like an engineer not a poet but that’s to overlook the real achievement here.
@TA. I did point out the decumulation strategy, actually he critiques quite a few and it was an eye opener for me. As you say, there isn’t much about this out there. The discussion about rebalancing was also valuable, especially as it feeds straight in to fund choices ( to the detriment of my intended approach of using multitasset/LS type funds, sadly).
In a few months I will have to do this for real so I am especially interested. Will read through again, more focussed, but translation to a non-US perspective I still believe to be not straightforward in terms of detail. Guess we will have to differ on that.
Nevertheless, I bought the book and am reading it again, it contains a hell of a lot of research. The main thing in dealing with a problem is to frame the question appropriately, and he certainly does that. Now all I have to do is apply it!
Re: fund choices, the main thing we can’t replicate here is the small value funds – unless you sign up to the cost of a Dimensional Fund Advisor.
I’ll probably replace the US / Int small value funds with one Global Multi-Factor ETF.
REIT US / Int becomes one Global REIT.
The emerging market small and value choices have equivalent funds / ETFs here.
I might well add a UK fund too just in case the £ has a golden decade against all the odds.
Great book! Thanks for the recommendation. Having FIREd a few years ago and having been decumulating since, it’s come at a good time. I was already following my own “sell equities when they’re up and don’t buy them again” plan so it’s great to have this backed up with some evidence based reassurance and refinement.
It is interesting how the author advocates strongly tilting away from the standard global market cap portfolio – which is challenging my previous conviction to stick to it.
IMO, it is more of a challenge finding appropriate UK alternatives for the bonds. In particular the 50% allocated to a TIPS fund which I believe in the US still offer a positive real return. Inflation linked bond funds here offer negative real return and are much too risky due to the duration. For now I’ve got a mix of short term bonds and cash but I do feel somewhat exposed to inflation. I guess we should just adjust our overall retirement expectations in the UK down a little to account for it unless anyone can suggest a suitable UK equivalent to TIPS?
Hi Humble Pie,
Yes, I agree. This is a real problem for UK investors. Here’s a piece I wrote about how big a problem and the list of (unsatisfactory) alternatives: http://monevator.com/why-uk-inflation-linked-funds-may-not-protect-you-against-inflation/
The simple fix would be for the HM Government to restore NS&I inflation-linked certificates as a fixture but failing that… the mix is short term bonds, cash and as many individual linkers as you can stomach. For many retirees, a small negative real return is a reasonable trade-off against the ravages of rampant inflation.
I intend to just grit my teeth and have a direct holiding of an IL gilt, maybe initially 12 year duration. Say 20% of retirement portfolio. L&G global IL fund has reasonable duration. Found that fund courtesy of this site ( thanks). Not pleasant buying into a probably over priced asset, but that is only probably, and didn’t someone say you are not diversified enough unless you have stuff you would rather not own.
With regard to the book, the idea of selling off the bonds first was an eye-opener for me. If I aim to drawdown at say 3% and that equates to £9k pa, having 90k in bonds gives roughly 10 years of cover ( inflation needs to be allowed for) for the equities to get back on song with the IL gilt standing in case if unexpected inflation. Still working on this!
I’ve read the book cover to err.. index, and reread a few chapters too. Thanks for highlighting this book as I’ve learned a lot. I was lucky to end up in the green zone some years ago and I thought the book wouldn’t affect my own position of being passively invested in the general global equity market. However the portfolio construction chapter has changed my mind. I think I’ll migrate towards the triad portfolio.
I’ve all also played with the spreadsheet and calculated the current Overall Valuation Level as 3.3 (4:2:4 US and 4 Dev CAPE and 2 Emerging CAPE). Ironically it calculated an initial withdrawal rate of 4% for me.
I agree that main problem in implementing the strategy is replicating TIPS.
Thanks for doing the valuation sum, I guessed 4. I think there is a more subtle issue in that the research is based on US economic conditions so I am not sure how valid a straight read across using investment alternatives is. Markets used to be much less coupled than they are now both through limited technology and through exchange controls, taxation on foreign holdings etc. So I am cautious about a literal transliteration (:)).
My overall valuation level was 3 – done about 6 weeks ago. I got a dev world CAPE of 3 otherwise same as Passive Pete. Initial withdrawal rate of 4.3%. By the time I subtracted costs and 40-year retirement penalty I was down to 4%. However, that was a big improvement on the 3% I had been planning for.
This reminds me, I wasn’t sure how dividends were to be treated. Are they reinvested, ie on what basis were the market returns based ? Must read it again.
Market returns are real total returns. Obvs you don’t reinvest your dividends if you’re living off them.
I also bought the book. Thanks for the recommendation. I am still trying to rationalise two views of the world. The comment from @Chester Dog above seems a reasonable approach to me. Keep a few years expenses in cash and the rest in equities. Tim Hale has some stats on the longest time to recover for his “whisky” fund and @CD’s 6 years should cover it. I have seen this scheme in other books too.
How do I reconcile that with the back tested evidence from McClung that we should have 50% bonds when starting retirement. The past performance data for US bonds up to 2010 used by McLung avoids the recent low interest rate era – although his book claims this does not matter.
A good thing about McLung’s book is that he does use the data to work out an answer on asset allocation – rather than fall back on the usual comment that it’s a personal preference or it depends on your risk tolerance. I find these comments rather duck the issue.
I broadly agree although much depends on the question you’re asking. To take two extremes, Buffet publicly stated the portfolio he was leaving his wife is 90% in equities. Presumably because it’s so big that even 10% in bonds will see her through any storm. On the other hand, there’s no point the truly risk adverse being 50% in equities if that’s liable to panic them. But yes, the reality is that McClung has drilled into this question in such depth that he can come up with a ready recommendation if you accept the rest of his assumptions. He also has the nous to show how you might customise his recommendation. Is 50% much higher than you were thinking?
Incidentally, it took the author 5 years to research and write the book.
“Market returns are real total returns. Obvs you don’t reinvest your dividends if you’re living off them.”
I’m not sure that I understand this statement in the context of McLung’s book. If you follow Prime Harvesting, you will not be living off dividends by definition.
Yes 50% is higher than I was thinking. I am trying to get feeling for the numbers and struggling to make sense of it.
Tim Hale says real return on equities is 5.5%. The stock market p/e is high at the moment, we are likely to see less than this return say 3% for the next 20 years. UK 10 year bonds have negative real returns – call it zero. So if I have 50% bonds then my average real return will be 1.5% (50% x 3 + 50% x 0). It has to last say 40 years. With a 3 or 4% withdrawal rate that does not really add up,
I agree it’s very much about the question you ask.
McClung’s drawdown strategy is total return. In other words, you sell however much of your portfolio you need to in order to meet your living expenses. So you’ll be living off some combination of accumulated interest, dividends, capital appreciation and, at times, principal. A total return strategy isn’t concerned with the individual components, you only need to select whether sales are coming from bonds / equities.
Possibly the confusion arises from the idea of a dividend drawdown strategy where you’d live off dividends only (and interest too, presumably) and leave principal alone, regardless of whether capital appreciation had taken place. This is much more expensive in terms of portfolio size required.
So in the context of the above statement, you are liable to be living off some portion of your dividends in a total return strategy, even though a total return strategy doesn’t distinguish between the sources of your income.
Of course, another potential source of confusion is that with McClung, you are drawing down from bonds first, if you possibly can, but I’m not drawing a distinction between bond interest and equity dividends in the above comments about the difference between total return and dividend drawdown strategies.
Yes, I understood “living off your dividends” to mean that your drawdown strategy was “living off your dividends”, not that your drawdown strategy would sometimes be funded partially or perhaps fully from dividends and interest.
Personally, due to the bond outlook I intend to be on the aggressive side of McClung’s recommendation. Probably 40% in bonds. No less. If equities underperform then the bond side of the equation will be liquidated and I’ll end up with a much higher allocation in equities as time passes. We’re most vulnerable to sequence-of-return problems in the first 10-years, so it is a good idea to have a generous bond buffer during that period. If equities outperform and cause me to buy more bonds, well hey, it’s not a problem because the portfolio growth is lifting me clear of emergency situations anyway. Interest rates may well have risen by then too.
The beauty of this system is that it has several self-correction mechanisms which Hale doesn’t. Hale’s prescription is more of an accumulation portfolio.
However, I think McClung won’t work for anyone who’s uncomfortable with a portfolio that could go to 100% equities in their dotage. Given today’s outlook, it’s easy to understand why anyone would quail right now at committing a large chunk to bonds, I think it’s the potential for bonds to completely disappear from your asset allocation that needs to be given real consideration. There’s also an inherent danger in projecting the current situation out to the next 40 years.
@ Lewis – yes, sorry, I think I was confusing matters with my rushed commenting 🙂
Thanks @TA, that’s a good summary of the issues.
For Mrs Buffet, we assume that she has so much money from Warren’s will, that she is not planning to spend it all in her lifetime. In which case she can leave the excess in equities. Hence his proposed asset allocation of 90: 10.
More practically, if you are convinced you have retirement covered and are thinking about legacy, then you could have more in equities (for 40 years time).
Yes, agreed. Conversely, if I was just scraping by then I’d want a heavy allocation in inflation-linked bonds, or inflation-linked annuities – as much guaranteed income as possible.
Actually, that’s something I forgot to mention earlier. Later in life, perhaps around age 70, if things look iffy then I’ll consider liquidating some of my portfolio (perhaps most) in exchange for an inflation-linked annuity. The State Pension will come into play and there’s always a reverse mortgage if things are really dire. So there are fallback positions if the market deals us a bad hand.
@TA. Thanks. To avoid doubt (by me), coupons and dividends collect in a cash bucket which is drawn first (exhausted each year) then supplemented by bond sales ? Must read it again with a better focus.
Ah, just figured it out! Yeah, pollution by a natural yield predudice is a source of confusion.
I too was surprised by the sell bonds first message, plus the other highlight for me was that the usual total market portfolio performed poorly in retirement. The concept of Harvesting Ratio got me thinking that I might adjust my current portfolio which is loosely based on Tim Hale’s book. I agree that the HR is largely based on US data and, as the book says, it needs more research. However I’m prepared to make some adjustments based on the current research.
Thanks also to @TA for confirming that I appear to be able to find the correct inputs to the spreadsheet. The recent discussions have helped me clarify my plan of action.
I wonder if he hasn’t been a bit too ambitious with his wanderings off into asset allocation. As I read it, and am now re-read information, the prime harvesting strategy is indifferent to the nuances of equity allocation providing you were diversified, and the bond allocation was a mix of short/ medium term treasuries. In the asset allocation piece, this has become a portfolio tilted towards small companies and value, with a wodge of reits, and the bond allocation has suddenly acquired TIPs.
I haven’t tried the spreadsheet yet but I would be hesitant in accepting an initial drawdown rate of 4%, let alone higher, if only because the 4% rule has been savaged recently as too optimistic.
Regardless of who you read, the most important asset allocation you can make is between equities and bonds. The majority of your returns will turn on that decision. However, you can attempt to optimise by diversifying across factors (e.g. small and value) and other assets (e.g. REITS) and sub-asset classes (e.g. inflation-linked bonds). There’s no guarantee this optimisations will improve your position but they have worked out well across long-term horizons. This is standard and not new.
The 4% rule as savaged is nothing like McClung’s solution and that’s the foundation of the book. The common-or-garden 4% rule is based on withdrawing an inflation-adjusted fixed amount every year. So if inflation goes up by 3% between years 1 and 2 and your portfolio doesn’t rise too then you’ll be withdrawing more than 4% in year 2. McClung’s methods and others like them have safety mechanisms that mitigate the blind step up prescribed by the standard 4% rule. Still, the lower a withdrawal rate you can accept the safer you’ll be so your caution is sensible, it will just cost you more.
Ok and thanks. Am I right to think that the harvesting logic is independent of the details of the equity allocation? Happy enough with 50:50 but not sure either about the tilts or about buying TIPs to then just sell them out for income from day 1. Seems odd to me when short duration bonds were the underpinning for the harvesting logic. Still reading!
I agree that the ‘traditional’ 4% harvesting ratio is currently likely to be too optimistic. However, the calculated initial drawdown rate is based on the Extended Mortality variable withdrawal strategy which can decrease if market conditions are unfavourable, as demonstrated in the example at the end of chapter 4. Therefore there is an adjusting mechanism built into the calculation of future years harvesting rates that would reduce the percentage below 4% should the portfolio performance be less than originally anticipated.
I think the book tries to build up marginal gains by starting with the basic harvesting strategy then introducing variable withdrawal strategies that improve overall income on average. It then introduces the harvesting ratio to help identify stronger portfolio types as you describe.
I agree with @TA that the important asset allocation is between bonds and stock, and this is demonstrated early on in the book in Figure 5.
I’ve pretty much finished adjusting my portfolio away from large US stocks and into small, EM and REITs.
The default assumptions for comparing the harvesting strategies are 60:40 equity bonds, 30 year retirement and portfolios of bonds in intermediate (not short) term treasuries and stock in 70% total market and 10% each in small company, small value and large value.
The ‘bonds first’ strategy was shown by Spitzer and Singh to perform better than other types of strategies such as age related or annual rebalancing.
Therefore I think you are correct in thinking the Prime Harvesting strategy is independent of equity allocation strategies noted later in the book. I think the tilts are just a marginal gain that you might choose to ignore or work around as we can’t simply replicate those suggested funds here in the UK easily. I’ve bent the rules a little to avoid too many changes to my original portfolio. However I think that in order to make any meaningful improvement you might need to accept the bonds first approach to harvesting. The waterfall mechanism of starting with 50:50 equity:bonds then selling bonds to fund retirement only works if you allow the original ratio to change such that equity can become a much higher percentage as bonds are sold but equity is retained during lower performing periods.
Thanks, being too fussy with equity allocations would also make the rebalancing more complicated and expensive. Didn’t the equity allocation avoid large cap altogether? I will recheck. If so it would mitigate against general global trackers which are capitalisation weighted, and the bonds first harvesting strategy also stands against multitasset funds, eg Vanguard LS. This, together with my predilection for a direct holding in IL means some thinking ( or a hybrid approach).
I will plough through the spreadsheet to see what the initial drawdown ratio is but I shall stick to no more than 3% in the early years and see how I get on.
Thanks for your patient replies.
I’ve just finished a first reading of McClung’s book and think it has a lot useful ideas, in line with many of the earlier commenters. The basis for both dynamic withdrawals and dynamic asset allocation are well grounded in the historical data, and these methods survive out of sample testing. McClung make the case for a portfolio with equal investments in each asset class rather than in proportion to market cap. I’ve wondered before if this is a better option, but have followed the global market cap plan along the lines advocated by Lar Kroijer and others. McClung recognizes that his equal balanced portfolios (e.g his Triad one) will be less comfortable for typical investors, and evaluates wider range, but all the best performing portfolios using his methods are equal balanced. What I’d like to be able to do is work out how my own portfolio measures up with his key measure, the Harvesting Ratio, to assess how far off it is compared to his 4 candidate portfolios. Broadly speaking I’m about 60%/40% equities/bonds with the bulk of the equities developed world but boosting EM to about 13% and keeping UK bias down a bit. The big difference with a typical global market cap portfolio and all McClung’s main candidates is their lack of large cap elements in both US and Intl equities, much more emphasis on small cap, and large and small cap value for US and Intl. He even splits EM into small and value elements for equal investment. So I’ve got no feel for whether it would be a radical re-working to rebalance to say a Triad portfolio or if I am in about the right ball-park and the changes would be good, but second-order effects to an already diversified portfolio, though based on market cap. Insights on this would be welcome. I note Accumulator has already thought through some of the potential for UK fund choices, so perhaps there is more to come there. Add the Triad to the example portfolios at some point?
McClung’s portfolio also contains REITs. Not sure about this either. I adopted the view that the large cap firms are all heavily into property anyway (Lars K again I think) so there is no point in having a separate REIT element in your portfolio. I read something about weird taxation rules on them too, which further put me off them. How crucial are REITs to the success of adopting a McClung-style approach I wonder?
Finally, McClung doesn’t say much about bonds at all, by design. I guess that’s as this is the ‘living expenses’ pool and you just stick to as simple as possible plan, almost a savings plan, and forget about it. In my case I’ve gone down a bit of a rabbit warren by adding non-UK sovereign bonds and corporate bonds to my UK gilt and IL bond funds. Plus a bit of cash for the next couple of years expenses and some P2P lending. So it could be I need to tidy this all up a bit, but I am loath to tinker further without good reason. Foreign debt and corporate bonds are a useful diversifier, according to Lars Kroijer, but I wonder if he has backed off from this position in the latest edition of his book – does anyone know if that is so?
My key questions then are: is the first-order benefit gained from applying McClung’s drawdown and portfolio allocation strategy rather than annual rebalancing to fixed asset proportions; and is modifying a globally diversified market cap portfolio to a Triad (or similar) portfolio necessary to benefit from McClung’s strategy or is the global cap portfolio likely to be adequate and the required changes only offer second-order benefits?
My interpretation was that the drawdown/withdrawal strategy is independent of the exact asset allocation. The triad and other recommended portfolios in the book are just those that (according to McClung’s testing/analysis) have the best chance of retirement success.
Whether you include small/value etc should really depend on your own view of how much these are likely to outperform the simple global market cap portfolio over the term of your retirement. McClung thinks they will but if you search around you will see that there are mixed views even from the industry heavyweights.
I’m also not sure about REITs. I include some as a diversifier but I thought in general they have a smaller expected return than general equities. It wouldn’t surprise me if some other sub asset classes (eg infrastructure funds) do just as well for this purpose.
FWIW I’ve settled on the following for the equities portion of my portfolio:
25% Developed market cap
12.5% Developed small cap
12.5% Developed value
20% Emerging markets
20% UK (plan to reduce this home bias over the next few years)
10% REIT
This is fairly similar to the portfolios in Tim Hale’s book. After reading McClung I’ve increased small cap and value but didn’t feel comfortable going all in.
@IH. Re LK 2nd edition , ‘….adding other government and corporate bonds….makes good sense’. So seems still ok. I have bought three copies of the book so hope the minor copyright infringement is tolerable 🙂
Good questions, but I don’t think there are simple clear answers as the book describes different metrics for measuring the different stages.
To my mind there are three stages of improvement from the default 60:40 total market portfolio and a fixed rate inflation adjusted withdrawal rate with annual rebalancing.
1. The Prime Harvesting strategy, where figure 26 shows its success rate as 94.4%, for a 5% withdrawal rate compared with a success rate of 77.8% for the default position shown in figure 5.
2. The Extended Mortality variable withdrawal rate can improve the efficiency of the strategy as measured in the HREFF-3 column from 68.7% in figure 48 to 99.7% in figure 59.
3. The composition of the portfolio can be made more robust as shown in figure 124 where the Total Global Market has a HR of 0.18 and MSWR of 4.8% compared with the HR-8×12 (similar to the Triad) that has a HR of over 0.35 and MSWR of 6.8%.
It’s therefore difficult to ascertain the levels of importance for each stage as we are effectively trying to compare apples with pears. We’ll all have different views and priorities, some might want to leave a legacy, in which case they might not want the most efficient harvesting mechanism.
@MrO – thanks for the info on LK’s 2nd ed. He noted in the press release for it that there were simplifications and I wondered if that was where they would be.
@Humble – I agree with you that including small cap and small value is taking a view on the market. On balance I’m still in the ‘take no view’ camp. It could be that as these sectors are I think higher variance they potentially increase the frequency that the 20% harvesting boundary is breached, so improve the efficiency of the method. I imagine this is a second-order effect.
@PassivePete Thanks for indexing those key relevant points back into the book. I’m wondering if the ordering of the points is coincidentally their ranking in their impact. Points 1 & 2 I have got a good grip on, but I will have to re-read on the point three material. This looks like a much bigger effect than I remembered, but I wonder how robust it is. My recollection is by this point in the book he’d had to substitute a proxy scaled market, and argue for the applicability of the measures over short time scales, so this suggests to me a major re-jig of my portfolio would be premature.
Bought the book. Fascinating. Trying to apply it to the UK financial landscape is a bit trickier. So I guessed at a withdrawal rate. ( I couldn’t even get the US links on his spreadsheet to work). I got the impression that the asset allocation of your portfolio isn’t that important, as long as it is diverse. I suppose some of the UK Vanguard Global tracker funds (for example) might do.
@John Pilkington I had a few problems with the spreadsheet links too but I managed to find the information eventually by digging around the federal reserve website and using google. I calculated the overall valuation level to be 3.3 whereas @TA calculated it as 3 (6 weeks previously).
My inputs are:
Market value of equities outstanding 25,306.4
Net worth 23,293.7
ICAGR20 4.87%
CAPE10 Value 31.30
Developed markets CAPE 10 24.3
EM CAPE 10 16.5
I agree that asset diversification is more important than where it’s allocated. Figure 124 was illuminating for me in that it shows that the MSWR can be improved significantly without increasing volatility by diversifying into REITs and EMs and by using equally weighted allocations. @IanH has highlighted that this section of the book is based on Simba’s spreadsheet that has data for only 27 and 40 years, so it’s less robust than other parts of the book. Still it was enough for me to make changes to my portfolio.
Many thanks for that information, Passive Pete, and your comments. This is such a helpful community.
@ John – I googled ‘Market value of equities outstanding Federal Reserve’ and got the pdf as the top result: z.1 Financial Accounts of the United States Flow of Funds, Balance Sheets, and Integrated Macroeconomic Accounts
I’ve exchanged a few emails with the book’s author – Michael McClung – and he mentioned that he sees equities as the main inflation hedge. As long as you have access to high quality government bonds then he’s not concerned about access to linkers.
@ Ian – Michael also made a wise and reassuring comment: Good strategies should have built-in resiliency to moderate variations.
Perhaps a reasonable analogy is that it’s like eating well. You could eat cauli instead of broccoli every now and then, it probably won’t make much difference. You can even have the occasional cream cake. But there comes a point where you’re eating cream cake so frequently and not broccoli that you’re on a different regime.
I think Passive Pete is right when he says that diversification across broad asset classes with historically sound returns is more important than the precise allocation. Still, Lars’ portfolio is really designed for accumulators whereas Michael has gone a step further by taking a range of lazy portfolios and testing them against his criteria for retirement.
To pick up a few points: the historical evidence for small and value factors is robust but there’s no guarantee they’ll work in the future. I suspect what’s at least as important here is that the factors aren’t strongly correlated with a large cap portfolio. Same with REITS. HumblePie is right that they have a lower expected return than equities but they have a reasonably low correlation too, so are a very useful diversifier. You’ll get some property exposure by investing in a global equity portfolio but a far greater degree of diversification by choosing a dedicated fund. There isn’t any tax funny business to worry about if you invest in a REIT index tracker.
hi,
Is this a website for professionals?…I was trying to read the comments and the language made me feel that. any idea for a newbie who wants to invest into a SIPP…nothing fancy, just funds with medium risk…looking at the likes of VLS60….
Hi Victor
Just ordinary self-investors here, not finance professionals. You have landed on a post that appeals to people interested in details of different portfolio spending strategies. If you use the ‘custom search’ box top right on the homepage you can find more relevant articles. For example, this is one http://monevator.com/using-vanguard-lifestrategy-funds-life/ is closer to what you want.
Hi Victor (and thanks ianh)
We’ve lined up some ‘start here’ posts here: http://monevator.com/category/investing/passive-investing-investing/
Hi, I don’t know if anyone is still looking at this, but I have been ‘practicing’ with the spreadsheet from the book. My fortunate position is that I have a pension that will cover my essential needs and a pot of savings invested. So, I don’t really need the 50 stocks / 50 bonds balance suggested by Michael as the starting point for most people. I’m wondering if the figures would work for an 80 stocks /20 bonds balance.
I’m thinking the valuation tilt would be 60 % (the highest given) for a 30 year outlook. (I’m referring to the second and third ‘sheets’ or tabs on his spreadsheet).
Also, with an 80% stock holding, I would expect the withdrawal amount to be more variable year on year depending on the past year’s inflation percentage (which is also on another sheet in the spreadsheet).
Thanks guys for a most informative blog; only got through the first four chapters, and I can already see some of your points mirrored in my queries. I feel sure I will be returning to these later as I work my way through.
John P: For my tuppence worth, McClung’s priorities are on minimising volatility / avoiding running out of money. If you’re not reliant on your investment pot for daily needs, you’re probably in a better position to sacrifice safety for better average returns. McClung does provide this data in his tables; you might find his MSWR 50% curves more appropriate for your needs (his focus is clearly on the MSWR 100%).
[MSWR = Maximum Sustainable Withdrawal Rate]
Just finishing my first read through and just went through these comments again. Most of it makes sense and feels robustly tested (within the limits of what’s possible). There are many concepts which are useful for someone newer to this area even if you don’t wish to fully embrace the system.
I’m struggling to get comfortable with one key part and need to do some more reading. His favoured variable withdrawal strategy is EM or ECM. The section on these and on MUFP from which these are derived is more opaque than the rest of the book. I read it as saying that MUFP is based on looking at what would have been an ideal withdrawal strategy using hindsight with the US data. The same approach using UK and Japanese data shows markedly different withdrawal rates.
It goes on to say that to make this practical to use EM is base on a “magic” (my description) table (figure 58) which combines the mortality and ideal withdrawal rate tables (for all portfoilos). There are two significant issues here for me. There is no explanation of how the magic table is created and no explanation of how it addresses the significant differences across other datasets.
This is curiously at odds with the amount of detail and somewhat stretched testing in other parts of the book. He does go on to show some backtesting of EM and ECM against “global” portfolios, but this has to use more synthesised data combining and extending real available data.
I suspect I’m missing something as the author is very careful about his approach thoughout the rest of the book, but the dufference between a 5% initial withdrawal rate (figure 56) for the US data and 3.8%, 3.8% and 4.1% for Japan, UK and his alternate US dataset is big enough that I feel more explanation of how figure 58 combines these is warranted. He demonstrates how important withdrawal rate is to overall success throughout the book and then omits explanantion at this pivotal point. I need to go and read up on the source references for that stuff to see if there are any answers, but would welcome others thoughts.
Like other commenters above, the translation to readily available funds etc in the UK will need some work. Equity funds should be quite easy, but I’ll admit to still finding bonds and bond funds a mystery and the use of US bond language makes this part a bit harder for me.
I had a first attempt with the spreadsheet and it does need a fair bit of work to get some of the source inputs. There’s potential for error here. It would be great to see the author maintain an up to date set of data (or at least checked links) on his website for the book and guarantee to do this for as long as the book is on sale at least.
I couldn’t figure out what exact mortality data he was using in order to check a UK equivalent. I tried to check it against the 2015 US figures rather than the 2007 source he cites but couldn’t find a direct match. I think the 10% chance of exceeding the age is the cause of the difference, but need to dig into that a bit more to find a UK source to use.
If I figure this stuff out I’ll post as this site could be very helpful to make application of the book easier for the UK.
The only other comment I’d have is that whilst he meets a much higher standard of scientific approach than the snake oil dressed as financial advice elsewhere, the core testing methods used are opaque and (as far as I can tell) not peer reviewed or made available for scrutiny. He describes backtesting but we don’t see the algorithms.
We don’t know much about the author other than he appears to have no professional financial background or qualifications, is well-qualified in computer science and claims experience in data science and simulation. I’m not trying to knock him. I believe his intentions and approach are excellent, but given how important this stuff is to anyone who uses it for their future prosperity, I believe it’s a valid observation.
At a mundane level, the proofreader appears to have gone to sleep on the later chapters and some of the graphics/dense tables have not reproduced well.
I wouldn’t want any of the above to put anyone off checking out the book. The first three chapters give a fair representation of the rest and can be downloaded for free. I plan to spend quite a bit more time getting comfortable with the methods he describes and some of the sources. At the very least I’ll be attempting to use this to see how well (or otherwise) I’m positioned to retire early.
Hi Stuart, I’m keen to come back to you on your comment but unfortunately don’t have time to go back over the sections you refer to in detail at the mo. I will reply when I can because this is an important conversation. For now, I’ll say I didn’t share your misgivings but I need to go back over those sections to explain why.
Personally I think McClung is more transparent in his methodology than the majority of authors I’ve read on the topic. If you’re looking for peer reviews (he’s not writing for a scientific journal!) then check out the exhaustive discussions here:
https://medium.com/@justusjp/prime-harvesting-vs-rebalancing-graphs-2687930a995b
https://www.bogleheads.org/forum/viewtopic.php?f=10&t=192105
I also think his non-financial industry background is an advantage. There is a cadre of astounding US retirement researchers who have been responsible for advancing the field. But in my view, none have written with such singular purpose as McClung in testing and articulating a complete system that can be used by DIY investors. They tend to publish out of academic interest or to advance the knowledge of wealth management professionals who handle large retirement accounts. McClung’s non-financial background means he’s writing for someone in my position and perhaps yours.
By all means post any bond questions you have here. There’s a fair chance I’ll be able to answer them quite quickly.
Thanks – I’m doing some further reading and looking at other related sources. I tried to reply but realise I need to re-read and check a number of things first so I don’t confuse matters further with misunderstanding.
With ref to your thoughts on McClung’s background etc – I don’t disagree, but it’s worth noting. There is quite a bit of discussion of his methods now as you say. I’d read EREVN’s various posts on medium.com and need to re-read to absorb the subtleties of his assumptions and approach and how they differ from McClung’s as they case a somewhat different light on Prime Harvesting.
re: bonds. I’ve been in equities and cash throughout the last few decades whilst working. Pretty much everything and everyone says that now I’m older I need to reduce risk and volatility by holding bonds (e.g. McClung receommended 50-60% equities). I understand that conceptually, but…
As far as I can tell, rising interest rates are likely to impact on QE fuelled equity overvaluations (as the small rise so far did), but rising rates also directly hit the value of bonds and bond funds – so they appear to be much more correlated than traditional wisdom suggests. Maybe that’s just in the “stuff” happens category rather than changing the underlying wisdom.
Gov bonds and equities are not negatively correlated. They generally have a low correlation and it changes over time. There have been instances in the past when bonds and equities sank simultaneously (e.g. 1973-74), not many instances but it does happen.
Even when bonds do decline, they are much less volatile than equities. This piece helps explain why: https://www.sellwoodconsulting.com/rise-and-shine-why-bond-investors-still-shouldnt-fear-rising-rates/
Be careful extrapolating from recent events. A full blown recession would be a different kind of storm from the recent squall – gov bonds could play their hoped-for counter-balancing role in the next one, they might not.
Even then, if equities are tanking 20% or more while bonds decline in single digits, you’re still better off living off your bonds and resolutely not selling equities when they’re down.
If you use a short-duration gilt fund then you’d experience much smaller drops than with intermediate. Longs bonds are only for the brave right now. Or you can buy individual bonds and create a ladder.
Now, you could substitute cash for bonds and use that to fulfil the same functions. You won’t see a rise in the value of your holdings with cash during a recession and if you’re keeping it in fixed term accounts then it will be adversely affected by rate rises, same as bonds.
You could even use a blend of cash and bonds – as long as you have plenty buffer to avoid selling equities when they’re down.
Btw, when I use the term bonds, we’re talking high quality domestic gov bonds i.e. gilts. Not corporate bonds or owt like that.
Thanks for the bond tips. I’ll read up on all that. I’ve been trying to get to the bottom of withdrawal rates before posting again, but going down a bit of a rabbit hole at the minute. Will post once I can make some sense of it. Thanks
Thanks for the informative site and introducing me to McClungs’ book. I am looking at whether to keep a DB pension scheme with approx. 60% part with max 2.5% index linking or take out the CETV and invest it. Anything thoughts about that would be interesting but specifically re the book a very simple question, on page 97 he gives some key withdrawal rate definitions but not the definition of inflation adjusted withdrawal rate. Is it the money taken in any specific year / inflation adjusted initial portfolio value?
Hi John B,
I haven’t checked the page number, so assuming we’re talking about a standard constant pound/dollar withdrawal then:
You start with your withdrawal amount in year 1 e.g. £10,000.
Your year 2 inflation-adjusted withdrawal = £10,000 x inflation. For example, if inflation is 3%, then:
£10,000 x 1.03 = £10,3o0 i.e. your year 2 inflation adjusted withdrawal.
Year 3 withdrawal = £10,300 x inflation at end of year 2. And so on. In a constant dollar / pound scenario (which is what people are talking about when referring to the 4% rule) it doesn’t matter what value your portfolio is. You take the same amount in real-terms every year until you or your portfolio run out of rope.
Also helpful: https://www.bogleheads.org/wiki/Withdrawal_methods
Thank you for your feedback. I am clearer on that now. Unfortunately it given me another question I hope you can help me with: In the book and in various places I see comments along the lines of “ stocks are a good inflation hedge” . Does that mean that stocks generally have growth rates > inflation or that stocks will roughly also increase by an additional amount due to inflation (roughly):
Eg Assuming average 5% future return from my portfolio, with 50% stocks at say 7% and 50% bonds at say 3% then if there is 3% inflation does that mean that future real return would arithmetically be:
I) Stocks: 7%-3% = 4%
Bonds: 3% – 3% = 0%
Or
2) Stocks: 7% +3% -3% = 7%
Bonds: 3% – 3% = 0%
I am about to dig into various books to try and understand that better but if you could let me know that might save me a day or two….. When you are beginner every little thing can lead to a question:)
Hi John, equities have historically beaten inflation over the long term. They can’t be relied upon to beat inflation annually or to spike when inflation spikes. Equities do badly when inflation is unexpectedly high e.g. 1970s.
Your scenario 1 is precisely what you’d do to work out the real return of equities if their results were reported in nominal terms. Which they usually are.
Has anyone read ‘Beyond The 4% Rule: The science of retirement portfolios that last a lifetime’ by Abraham Okusanya or have any views on how it compares with Living Off Your Money?
I have not, but I will now. Thanks for the tip off! Abraham is a rare voice in UK retirement research, so will be interesting to see what he’s come up with. Strangely he doesn’t even mention his book on his blog.
Hi people, an excellent thread. I have just finished reading the book and have ordered one for my brother who lives in the USA and will hopefully find it useful.
I found the book to be very interesting. Loved the harvesting section and the variable withdrawal section. Shared the misgivings about the portfolio allocation stated in this thread and have been assured by others comments . I noted with interest the Guaranteed Income part of the book but with the heavy USA leaning see that for us in the UK only annuities are really available – and as I have dual nationality as a Kiwi we don’t even have those back in New Zealand.
But the part of the book I found most difficult to understand was the talk about tilts and using percentages such as a x% tilt. Could somebody give me a pocket version of what that’s about please.
I would also be interested to hear people’s views on Beyond 4% and compare it with Living Off Your Money. Thanks.
Hi, read Beyond 4% this week. Review coming soon.
Re: tilts. If you choose a ‘safe withdrawal rate’ then that’s meant to protect you against the worst that history could throw at you. Most of the time history was kinder. If you stuck to the safest rate over the course of a non-worst-case-in-history scenario then you left money unspent. A tilt to your initial withdrawal rate allows you to use a higher withdrawal rate, if the initial market conditions look promising. There is some evidence that ‘valuation levels’ e.g. Shiller’s CAPE 10 ratio, have some predictive power when it comes equity performance a decade or so down the line. When Shiller’s metric is high, equity performance over the next 10-years tends towards the disappointing and vice versa. So McClung has used this relationship to enable you to increase initial withdrawal rate if the coast looks relatively clear. The caveat is, that if things go wrong e.g. big bad bear market and high inflation, then you’ll need to put the brakes on later. Hope that helps, a bit!
Many Thanks @TA, That’s just the hint I needed. A part of my confusion was based upon thinking the tilt reduced the withdrawal rate and I emotionally couldn’t engage with that thought – lol. Clearer now.
Thanks for the tip on Beyond the 4% Rule. Just bought it: hopes it helps in my quixotic quest for an easy answer,!
Beyond The 4% Rule review here:
http://monevator.com/beyond-the-4-rule-abraham-okusanya/
I quoted your comment, Mr Optimistic. I thought it nicely summed up the Holy Grail we’re all after.
Looks like I’d better register on this thread!
Has anybody looked into the implications of using Prime Harvesting in terms of combining the use of SIPP drawdown and ISAs?
Prime Harvesting relies on the initial equity value being set. As McClung documented, the only equity purchases should be when rebalancing between equity asset classes and dividend reinvestment.
This is fine for my SIPP drawdown, since I won’t be adding additional capital.
I’m happy carrying out Prime Harvesting and setting a SippSWR to drawdown out of the SIPP.
Assuming returns are reasonable (ho ho), the amount I intend to annually withdraw from the SIPP is a lot more (on average) than my outgoings (I live frugally). I am withdrawing more than I need in order to get the money out of my pension (before I croak), avoid LTA tax and maximise my annual basic tax allowance.
I intended putting excess withdrawals into ISA funds (for future capital spends) and then use Prime Harvesting to manage these. This will of course mean capital being added to the ISA funds over time. Therefore an ‘initial equity capital’ sum cannot be set, and so, following McClung’s documented rules, the Prime Harvesting process cannot be used for harvesting my ISA funds.
I’ve been thinking along the lines of having some sort of ‘Dynamic Target Value Of Stock’.
At any point in time, this would be the sum of all inflation adjusted ISAs that have been added to my ISA fund. This would then allow a ‘Dynamic 120% of Target Stock Value’ to be set and used as the harvesting threshold.
So a Target Value Of Stock at the end of 3 years might look something like…
Dynamic Target Value Of Stock = ISAyr1 (inflation adjusted for 3 years) + ISAyr2 (inflation adjusted for 2 years) + ISAyr3 (inflation adjusted for 1 year)
…and…
Dynamic 120% Of Target Stock Value = Dynamic Target Value Of Stock x 1.2
Does that seem reasonable?
Is anybody else using Prime Harvesting to harvest their ISAs or are you just spending your excess on beer?
Hi Nun, I don’t understand why changing the location of your assets makes any difference if you view your investments at the portfolio level. For example: you have £100,000 in equities across all accounts. If you sell £10,000 worth of equities in your SIPP account and use those proceeds to buy equities in your ISA account then what have you changed?
Hi @TA, When I read McClung’s book I initially thought that a virtual portfolio would be the obvious way forward, but the more I thought about it, the less comfortable I became.
My concerns with a virtual portfolio are down to:
• my requirement of a different risk profile for SIPP and ISA investments
• regulatory limitations on transfers between an ISA and SIPP
I’m wanting the risk profile of my ISA fund to be more aggressive than my SIPP fund and so there would not necessarily be a one-for-one swap. I appreciate that there is nothing stopping me changing the fund completely when I ‘transfer’ money from the SIPP to the ISA. My issue is how the risk profile of my ISA could potentially be altered later on by a trigger of a Virtual Portfolio 120% of Target Stock Value.
Consider the following scenario that muddies the wat…. highlights my virtual portfolio concerns. I’ve taken an extreme example to stress the point. Reality would be less extreme, but in my view have the same underlying problem.
Let’s say my virtual portfolio triggers the Virtual Portfolio 120% of Target Stock Value. Prime Harvesting says we should move 20% of the virtual portfolio equities to bonds.
Let’s say I look at the returns of the individual funds in my virtual portfolio and find that all the significant, extra equity returns are in my ISA fund. If it was a truly virtual portfolio I’d be able to sell 20% of my ISA equities and buy 20% of SIPP bonds (since I notice my bonds are going low in my SIPP and that’s where I could really do with them to fund my large/constant SIPP withdrawals and reducing my insomnia at having 95% equities in my SIPP).
Obviously the government won’t let me do that, so the other alternative is to sell 20% of the virtual portfolio equities in my SIPP to buy the SIPP bonds. But it’s the ISA returns that triggered the harvesting. Selling SIPP equities at this point may not be advantageous since the returns aren’t as good as those in the ISA. So I’m left with selling 20% of the ISA equities and buying 20% of ISA bonds. I don’t really want to do that either since I really want to have an aggressive stock/bond ratio on my ISAs.
I think if it was possible to have a free flow of transfers between ISAs and SIPPs, I’d be comfortable with a virtual portfolio.
Since that is never going to be possible, I’m on a quest for an alternative.
I’ve also a requirement to introduce additional capital to the ISA fund (not sourced from the SIPP) – a virtual portfolio would not resolve that issue.
Cheers, Nun
On p71 McClung says:”…additional tests were run using Prime Harvesting to manage individual asset subclasses (e.g. small stocks, large value stocks) as opposed to stocks as a whole. Performance only improved a little….Further study (with more data) may show the extra complexity of finer tracking is worthwhile for some retirees; however with the data in hand it’s easier and reasonable to avoid the additional complexity”.
I find this rather vague. My reading of the above is that his testing dataset(s) did not have the required granularity of performance data to provide a definitive conclusion, and it was easier to do his tests at the portfolio stock level.
I find the 120% Target Stock Value (for the portfolio) trigger convenient. But, my gut feeling is that it’s using a chainsaw to slice a blancmange.
We are all good investors and try to invest in asset classes that have less correlation. Therefore, there is more chance that some may be performing better, when others are performing poorly. The poorly performing asset classes may be preventing that 120% Target Stock Value (for the portfolio) being hit. Potentially, roles might reverse and the poor asset classes of yesterday, may be tomorrow’s stars, and yesterday’s stars, tomorrow’s fallen heroes. Worst case is that as the asset classes move up and down, the 120% Target Stock Value (for the portfolio) will get hit less, and potential gains not harvested.
Why not harvest at the asset class level? It guarantees that you harvest those gains for each asset class…rather than have the potential of losing them. McClung says that complexity is an issue. As I see it, all you are doing is tracking whether each individual fund (tracking your asset class), grows beyond inflation growth + 20%. I don’t really see that as being complex or much of an administration overhead.
I can’t see that it would stop you using his useful spreadsheet to derive the Initial and Maximum Withdrawal Amounts…you just wouldn’t trigger the ‘Rebalance=Yes’ on his ‘Prime Harvesting Worksheet’.
Great book. If only all investment books were written with such rigour.
Hi, I appreciate I may be missing something about your situation but possible approaches are…
You have multiple ISA accounts. How about allowing some to maintain your more aggressive risk profile but then allow some ISA accounts to work in tandem with your SIPP to maintain an overall retirement risk profile? Then bonds in your ISA, not a problem. That particular ISA or set of ISAs are viewed as part of your retirement portfolio. You still maintain your super-aggressive ISAs as a separate entity.
If that won’t do, then you just make a couple more trades:
1. Sell £10,000 worth of outperforming equities in your ISA.
2. Sell £10,000 worth of ‘getting by’ equities in your SIPP.
3. Buy £10,000 worth of bonds in your SIPP.
4. Buy the ‘getting by’ equities back in your ISA.
The bonds are now in the SIPP, and you’ve sold your outperforming ISA equities.
It’s worth adding, a scenario whereby you’ve got a stack of outperformers in your ISAs vs your SIPP doesn’t sound likely unless you have some highly eclectic positions that all happen to turn up trumps in the ISA.
You could also avoid this by having your ‘retirement’ ISAs hold the same equity positions as your SIPP.
Re: prime harvesting by sub-asset classes. McClung seems to be saying you can do this and he tested to see if it is worth doing. By the light of his evidence and in his judgement, it’s not, but you may well find it is worth it for you.
Thanks for the input.
Before I read McClung’s book I was effectively managing the SIPP and ISAs separately in what he calls Alternate-Prime (although I would harvest above inflation growth equities before heading off to bonds for withdrawals). I hit on that by luck rather than design. I do have asset allocation targets that fit across both the SIPP and ISAs; in that respect I do manage it as a virtual portfolio. There is a risk difference in that my SIPP is 60:40 and ISAs 80:20.
McClung’s book brought in the swerve ball of Prime Harvesting and the drifting stock/bond ratio. This did get me thinking about continuing with Alternate-Prime for my SIPP and changing to Prime for my ISAs (since I’d feel ok for the stock/bond ratio to drift).
Although I see a virtual portfolio essential in maintaining asset class targets, I’m not really convinced of the practicalities of using one Prime or Alternate-Prime strategy across the virtual portfolio.
I think, at the moment, I’m happy to continue with the Alternate-Prime strategy for my SIPP and ponder further about Prime for my ISAs.
I bought the book and had an initial read (I have a feeling I will go over it many times!) so through I would add my two pence.
First, it is great to see someone approach the subject with such depth and evidence based rigour. I read that ‘beyond the 4% rule’ book and this is a magnitude more detailed and robust.
In terms of the recommendations, I buy into the income harvesting strategy and the variable withdrawal strategy (and the valuation for the initial WR bit) but have trouble with the portfolio construction.
As someone who has drunk the passive kool-aid and agrees with Lars K that the market cap weighted index is the aggregate of all market players’ best guesses of best value, I find it difficult to shift to one of the recommended portfolios which are more equal weighted.
I understand the thinking behind factors and am not against a bit of a tilt but the Triad in particular seems to go too much the other way. For example, US large cap growth has an allocation of 6.5% in the global total market portfolio (VTWSX) but only 0.4% in the Triad. I don’t think it is balanced to strip out so much large and growth. Having said that, McClung is open in saying that these portfolios represent a departure to traditional accumulation ones.
@IanH and @PassivePete have hit the nail on the head for me: are these portfolios a first order or second order issue? If I used a market cap based allocation (assuming it was diversified with a sensible bond-equity split) would that seriously undermine the income harvesting and variable withdrawal elements of this ? Fig 124 shows that traditional market cap has a harvesting ratio of 0.18 and the 8×12 has 0.35. It’s difficult for me to get a feel for what that means exactly.
So, I’ll keep on mulling it over and will probably end up with a halfway house. Let’s not let the perfect be the enemy of the good: a well diversified portfolio with some tilts will probably do the job given the soundness of the overall framework and the other levers (cutting spending / get more income) outside of the decumulation machine.
Thanks to everyone for their comments – it is good to have other people on the same journey.
p.s. In the Bogleheads thread there were a few people complaining about having to pay $40 for the book. As a result of reading it I am upping my initial WR from 3.3 to 3.8%, saving me a six figure sum!
Anyone tried to use the portfolio recommendations? They’re notable for avoiding large cap US companies in general and instead specifying large-value instead (presumably to reduce volatility from growth stocks?). It doesn’t seem to be a class that is easy to replicate with a fund or ETF in the UK (and forget about GBP hedging). Just curious if people are ignoring this part of the book or have come up with cunning alternatives.
@Raheem – “Fig 124 shows that traditional market cap has a harvesting ratio of 0.18 and the 8×12 has 0.35. It’s difficult for me to get a feel for what that means exactly.”
One answer is about 1% withdrawal rate (90% success) according to McClung. Based on a 50/50 portfolio tested against 1972-2010 (fig 89, p. 163).
But when he gets into the actual portfolios chapter, the Harvesting Ratio (HR) figures are all lower (<.30) so I don't know how much store to place in the above.
Been thinking about this a lot as well having been back to re-read. I boiled the argument down to…. diversification comes more to the fore in retirement because we need more insulation from periods of poor performance and it turns trading off better portfolio returns to get reduced volatility results in a higher probability of successful retirement.
So the additional volatility of Large and Growth (which McClung dumps in favour of Large-Value) is not worth the extra return… in decumulation that is.
As you say, however, it can be tough to switch approach so significantly , but I suppose it depends on how good you think the analysis/testing is to offset how it "seems".
The reality seems to be that it's difficult/impossible to replicate the portfolios anyway as we do not seem to have access to the graularity of factors in funds for the US/internationally. At least I haven't found them and certainly not in hedged versions.
So it seems more like taking on board the diversification message and maybe playing with the Bogleheads' Simba sheet a bit to get a feel for what mixes can increase HR a bit perhaps.
At least that's what I'm resigned to as the complexity of trying to do any better than that really was resulting in the analysis paralysis sagely pointed out in the article.