We don’t know when we will die, but we do / should know how much money we need each year to sustain a given lifestyle, and we should be able to project future investment returns based on long term historical performance. Leading up to - and now in - retirement I have been documenting my lifestyle costs rigorously, month by month. I have this data for 3 years now, so I can be fairly confident what my target annual pension income needs to be. For me this turns out to be quite a lot more than the $72,000 for a couple to retire ‘comfortably’, as proposed by ASFA. 

Let’s say for the sake of argument I have accumulated $1 million in super, and I need $100,000 to live comfortably each year (and I have no other sources of income). If I assume 5% inflation and 5% annual investment returns (which you might be fairly sure of with an income / fixed interest focus in your portfolio), I would run out of money in 10 years. 

If I increase the assumed annual investment returns to 7%, that only gives me another year before I run out of money. (Over the past 15 years the median Growth fund has returned 7.2% per year).

But we all expect superior returns than those offered by a super fund, don’t we? (Otherwise why be a self-directed investor?). If I increase the assumed annual investment returns to 9%, that still only buys me another year before I run out of money. However if I increase assumed annual returns to 15% then I run out of money after 20 years. 

Conclusion: I would either need to reduce my lifestyle costs or increase my accumulation balance before I retire, since (for me at any rate) it is too aggressive to assume an annual 15% return over a retirement lifetime, which could be as long as 30 years or more, health permitting.

Looking at another scenario: the maximum amount you can currently accumulate in super is $1.9 million. With this balance transferred to pension mode, with a 5% annual investment return I would run out of money in 19 years. Not so bad. With a 7% annual investment return I would run out of money in 22 years. If I just manage slightly superior returns to an average super fund (in a Growth option, mind you) at 9%, then my money would last for 30 years.

Food for thought. I’m not trying to suggest to anyone how much super you need to retire on, but with some simple (albeit crude) modelling you can begin to see what kind of returns you need to achieve given a starting balance and a realistic annual lifestyle cost, and this in turn might influence your asset allocation post retirement.

As I'm rapidly approaching "that time" I've recently been doing a LOT of reading and research in this space and even attended two whole day "retirement events"

All looking at both financial and the many non financial aspects of retiring

Some of the books I've recently read include (the first three are Australian so have that local context):

• The Golden Years: How to plan a happy and financially secure retirement
• How to Have an Epic Retirement
• Retirement Made Simple
• Retirement Planning Guidebook: Navigating the Important Decisions for Retirement Success - by Wade Pfau (who's a very well known Retirement Researcher/expert in the US - it's obviously US based (and was quite technical) but some of the content is general enough to apply anywhere)

Morningstar also recently ran a four part Morningstar Retirement Bootcamp series (the presentations are available for download and the videos are available on Youtube) - Note: I know Mark is a big fan of the bucketing method and income investing so that's the asset allocation approach you'll see highlighted in the last part of the series here

There's also quite a few blogs with extensive content in this space - however most of them are US based - some of which applies and much of which doesn't in our domestic context (and that goes way beyond just our tax, pension, medical, aged care and other systems differences - even things like the so called 4% rule doesn't actually work (stand up to historical calculations) in a historical context in Australia (or in fact almost every other country other than the US) - however the FIRE crowd seems to eat it up) - anyway some I found interesting include Early Retirement Now's - Safe Withdrawal Rate Series and The Retirement Manifesto

There appear to be two broad (opposing) camps of retirement philosophies: those who are probability-based and and those who are safety-first - e.g. if a Monte Carlo simulation (the standard tooling used by most modern retirement calculators - e.g. the TelstraSuper Retirement Lifestyle Planner or the Supercalcs Mercer Retirement Income Simulator) indicates a 90% probability of success:

• the probability-based people say 90% is reasonable (plan likely to work), can adjust as plan progresses if needed (e.g. reduce spending if required);
• however the safety-first people say 10% failure rate is too high, must split into essential expenses which need 100%, and discretionary which can be adjusted

and this favouring of one or the other then leads into different strategies for investing in retirement - e.g.

• probability-based people look at safe withdrawal rates and use a total returns portfolio approach using MPT and are generally not looking to things like annuities (as the generally lower returns do not offset the marginal increase in safety over a SWR (Safe Withdrawal Rate)) - just use an account based pension in retirement
• safety-first folks live in the lifecycle finance theory world where they prioritise spending goals (essential basic needs, contingency fund, discretionary expenses), risky assets are "risky" and history is no guide, so need to do asset-liability matching (assets matched to goals/needs) - time segmentation and bucket strategies and including cash, annuities, bond ladders are some of the fundamental tools for essential spending; an account based pension is just for the discretionary bucket)

Of course it's not just "black" and "white" and often people will fall across the spectrum and choose aspects of each to incorporate

Personally after my research I think I'm so far leaning towards having a partial bucket strategy (the emergency fund part at least) and using a brief falling (into retirement) and then a rising (reverse) equity glidepath in the 5-15 years after retirement (essentially to offset some of the risks of sequence of return around retirement) but moving back to a 75% equity weighting over those years for long term growth (this last part is not typical of a traditional lifecycle asset allocation/glidepath of a traditional super fund (where in the 10-20 years leading into retirement you're shifted from a growth portfolio to a balanced then conservative portfolio by retirement and then stay there (e.g. 30:70 defensive:growth) for the rest of retirement). Also to set an expectation/guide of the sort of reliable ongoing spending in retirement I'm also looking at calculating via a 3.25% SWR (which appears to be historically reasonable in Australia - Note: I'm calculating for 40+ years)

An additional problem in this space is there are so many withdrawal strategies (variants and in addition to the standard SWR) out there it's hard to choose - most "intuitively sound reasonable" but it's hard to know in practice and some are way complicated (it's harder to see some of these working when you're old and senile trying to redo the calcs each year or if you pass away expecting your better half to just take up where you left off) - including: Constant Dollar (sort of SWR), Vanguard Dynamic Spending Strategy, Endowment Strategy, CAPE-based, Hebeler Autopilot II, Guyton-Klinger, 95% Rule, Sensible Withdrawals, Variable Percentage Withdrawal (Boggleheads) & Dynamic SWR to name a few

I stumbled across this Livewire article discussing the Growth At A Reasonable Yield (GARY) portfolio, which sounds like it focuses on that elusive sweet spot of 'quality' companies which exhibit both capital growth and pay a worthwhile dividend. 

The recent best performers in this model portfolio are cited as CDA, which I already hold, EVN, which I have been mulling over for a few months, but am not too keen on because of its purely domestic focus (hence limited growth potential), and TLS, which has been a long term capital destroyer. 

The GARY portfolio seems to be the creation of Deep Data Analytics, whose website is not at all helpful in explaining the basis for stock inclusion in the portfolio.

I recently listened to the Money Puzzle podcast (22/8 episode) in which James Kirby interviewed Sam Sicilia, CIO Hostplus for the past 16 years. Hostplus has been the #1 performing Australian Super fund over the past decade (8.3% pa). Their average fund member is 37 years old, so unsurprisingly the talk starts with the value of compounding and long-term investing for those working toward retirement with runways of 30 or so years.

But the discussion then pivots to the issue that (quoting Sicilia): "Super was designed at a time when people retired at 65 and conveniently died at 68. Today they live to 98. How do you intend to live for 30 years after retirement without generating returns in retirement."

Sicilia's thesis is that you invest for growth in the accumulation phase (pre-retirement), then move to the decumulation phase post-retirement, but now that decumulation phase investing must have an accumulation (meaning growth) component, or else retirees will run out of funds. "All I'm saying is that if you intend to live 30 years post-retirement, then you are essentially a 30-year-old that retires at 60, even though you're a 60-year-old that retires at 90, right?"

I've recently heard several other fund managers who work with SMSF investors echo this thinking - that there needs to be a mind-set change and retirees need to be investing for growth as well as income, and that investing for income alone (and holding capital-preserving assets like bonds and cash) will see their capital erode too quickly. Couple that with recent surveys showing that the primary reason so many retirees are dying with large Super balances is because they are frightened of running out of cash and not because they are hoarding it for future generations.

Personal experience has taught me that the need for growth as well as income in retirees' investments is one many financial advisors haven't yet absorbed. Their learned wisdom seems to be that if you're retired and in pension phase, then you must move to preserving capital (Defensive Assets = Domestic Cash and Fixed Interest) and otherwise invest in equities for income only (regardless of any possible capital growth). The impact of inflation on capital is barely if at all acknowledged.

When it comes to the current reality of pension-phase investing, there seems to be a fundamental disconnect between, on the one hand, fund managers (and savvy self-managing SMSF investors) and, on the other hand, the body of financial advisors.

Just my 2 cents.