4% safe withdrawal rate?

The dividend payment causes a company's share price to drop by precisely the same amount as the dividend. So, for example, if a company makes a dividend payment equal to 5% of the stock price, shareholders will see a resulting fall of 5% in the price of their shares. If the shareholder immediately reinvests the dividend payment back into company stock, the shareholder reverts to the position he was in immediately prior to the dividend payment (market movements and attritional costs aside).
Sarenco I haven't fully plumbed the theological differences between yourself and Colm, but I guess it is simply that you are at cross purposes. However the above quote is not quite accurate. On a look through basis, the day before the dividend payment the investor owns, say, 100 of productive assets and 5 of cash. (I am presuming the company generates the dividend from its cashflow and not from disposing of productive capacity.) The payment of a dividend crystalises the position to being the investor owning 100 of productive assets (on a look through to the company) and separately 5 in cash. If she uses the 5 to buy more of the same shares she finishes up, again on a look through basis, with 105 of productive assets. This is not then a reversion to the previous position.
 
Whitecoat,

Whatever about the musculoskeletal impact on the players, it's the brain cell destruction and the associated hangover of poor supporters like me that's the real concern.

We agreed that the guidance for Irish financial advisers on this matter is virtually non existent compared to, say, the UK or USA.

On a serious note, as Marc has correctly confirmed, there has been such little published research in Ireland regarding this - this is very unsatisfactory.
 
as Marc has correctly confirmed, there has been such little published research in Ireland regarding this - this is very unsatisfactory.

I agree, and would go further. There is a complete dearth of information on how the funds of individual DC/ ARF investors have performed. We see lots of figures on the performance of unit-linked funds, but they're not what the end consumers see. It's the end result that matters to them, after all charges by insurance companies, advisers, etc.

It's almost impossible to get even more basic information, such as where individual investors are putting their money. One of the few statistics I've uncovered serves as a damning indictment of the financial advice industry: a 2015 survey by a working party of the Society of Actuaries in Ireland found that 44% of insured ARF's/AMRF's were invested entirely in cash or cash-like funds (i.e. including capital protected), another 44% in managed funds (which include a cash element) and 12% in single asset funds. Does anyone have any more up-to-date information? Are any of the advisers who contribute to this forum prepared to tell how their clients' funds are invested, or how they've performed (net of all charges) over (say) the last 3 or 5 years?
 
Last edited:
Colm,

All fair comments (fair as in good - not fair as in fair!!)

One observation. An "unsophisticated" adviser could honestly claim that his clients achieved perfectly satisfactory returns over the last number of years given how markets have performed - i.e. pot luck. [I still believe in the role pot luck (randomness) can play in things in spite of my trusty crusty socks letting me down big time yesterday].

What concerns me most is the lack of developed debate on the subject. The OP's original question is just so sensible and obvious - we really should have been able to answer it with: here is a link to a very useful template/guide on the subject produced by the XXX Professional Body.

My belief is that the lack of data and debate and guidelines leads to poor quality advice to the public. I have seen new business ARF reports from reputed consultancy firms - frankly, the quality of commentary and analysis was very low grade. God alone knows what some of the other advisers out there are up to.
 
@WhiteCoat makes an interesting observation about the use of defensive assets in a European context.

My own analysis in this area has slightly improved (over the last 3 years or so) on a "traditional" Short-term, global Government and Investment Grade Euro Hedged Bond approach for the "defensive" part of the portfolio

But one has to recognise that you can't generate returns without exposure to risk and not all risks are worth taking and even those that are worth taking aren't guaranteed to pay off over your time frame (ask a value investor!!)


upload_2019-2-4_11-14-49.png

Source Morningstar to end of Jan 2019
 
@Marc

Fair dues, and thanks for sharing with us. I’m glad I'm not the only one prepared to share my innermost secrets.

For what it's worth, here's my approach to estimating how much of an "income" I can draw from my portfolio, and how much I think could be drawn from the sample portfolio Marc shared with us.

I learned my trade in the old world, before the era of stochastic simulations. Being stone age therefore, I look at financial questions firstly in deterministic terms, and then try to allow - more intuitively than scientifically - for the possibility (certainty!) that things won't work out as planned.

I have a spreadsheet, which projects expected investment return and pension outgo each month, starting from the current smoothed value of the fund (using smoothed values rather than market values is another story, too complicated to go into here; basically, I don’t think the market is always right). I also include a modest provision for contingencies.

The aim is that, if expected returns are achieved, there will be enough in the fund to last us until we're well into our 90’s. If either or both of us kick the bucket sooner, there'll be something left in the kitty. I haven't factored our house into the equation. That’s a form of "insurance policy", in case the fund fails to deliver the required return or we live even longer than expected. The default assumption is that our home will form part of our estate.

My fund is invested exclusively in equities, with a tiny amount in cash. I expect an average return of slightly over 6% a year before charges by external providers, 5.75% after charges. (I manage my own investments to keep charges to a minimum). Plugging the 5.75% assumed return into the spreadsheet, the affordable pension is more than is needed for “survival”. However, I plan to withdraw an average of the affordable amount from the fund each month anyway. If times get tough, we can cut back, but I don't plan on cutting back otherwise: I much prefer to spend the money now than leave it for someone else to spend when we’re dead.

Looking at the portfolio Marc put up, I calculate an expected blended return of around 4.4% a year before charges. The calculation assumes a zero return on cash, around 1% a year on government bonds, etc. I won't go into detail on the assumptions for each asset category; that would distract from the core issues, but the message is clear: I expect a low return from low risk assets and a high return from high risk assets. I haven't allowed at this stage for the greater volatility of the high-risk assets. The difference between the 4.4% for Marc's sample portfolio and my 6% plus (both before charges) is the higher average risk profile of my portfolio, particularly that I don't have any cash or low-risk bonds.

From my (limited) experience of the retail financial services market, I estimate an average charge of 2% per annum on a portfolio of the type Marc outlined. That includes asset management fees, additional charges by the unit trust/ unit-linked fund provider, and finally the financial adviser’s fee/ commission. The 2% pa estimate for charges could be wrong - how far out I don't know. Thus, the expected net return on what I'll call Marc's portfolio is 2.4%.

If I plug a projected 2.4% return into my spreadsheet and assume the same level of regular outgo, I run out of money eight-and-a-half years sooner than I've assumed for my portfolio. Alternatively, the affordable "income" is much lower. The risk that things will not work out as planned is not as great for Marc’s portfolio as for my pure equity portfolio, but it's still a risk, which will have to be allowed for.

I suppose the main message I want to convey is that volatility of future returns is an important consideration, but the expected return, before volatility but after charges, is far more important in determining what level of “income” can be taken from a pension fund.
 
Last edited:
Colm,

The expected return of the portfolio is included in the first graph and is the benchmark CPI + 2.32%pa (we assume 2% ECB inflation target) so the nominal expected return is 4.32%pa. So, 4.4% isn't a bad guess.

The total cost of the portfolio is 1.82%pa so again, not a million miles away.

The stochastic calculation looks like this

upload_2019-2-4_16-51-7.png

The identical analysis of a proxy for "your" portfolio looks like this

upload_2019-2-4_16-54-2.png
 
Last edited:
That looks like nifty software.

Is the x-axis projecting into the future? The shape of the graphs look like what I learned in my studies donkeys' years ago as an "expanding funnel of doubt". If it is projecting possible fund values into the future, I presume it's not allowing for future drawings from the fund? There's no way I'd allow my (completely theoretical) €1.5 million fund to grow to €6 million, if things went well. My ideal is not to leave anything - possibly other than the house - when my wife and I fall off our perches. The plan is to have spent/ dispensed any funds surplus to our requirements while we're alive.

For what it's worth, taking your hypothetical €1.5 million fund, I would allow for withdrawals of €92,000 a year (increasing with inflation), or 6.1% of its starting value, on my portfolio (after setting aside a small lump sum for contingencies), based on my investment return and expense assumptions. The corresponding figure for your asset configuration (and return and expense assumptions) is €63,000 a year, or 4.2% of the fund's starting value. The €63k income is less risky than my €92k income - although I couldn't see the increased risk on my portfolio coming through clearly in your graph.
 
Last edited:
Yes, the x axis is age.

So, let's zoom in, and look at the downside risk, since the key to managing other people's money (as distinct from one's own) is the need to manage against the rare (but perfectly possible) risks of catastrophe rather than assuming a benign investing environment. It's why for investment advisers, diversification is always the answer.

So, let's model your assumed investment return against historic equity volatility and start with an income of 5% of the €1.5M keeping this expenditure constant so that you don't have to suffer declining income in retirement.

upload_2019-2-4_18-48-12.png

Each bar represents a 5 year period



upload_2019-2-4_18-48-59.png

So, yes, there is a possibility of a good inheritance, or rising income in retirement. But, we could be 5% confident you would be broke 10 years in and on average, you should expect to run out of money at age 84!


Keeping everything else constant but changing the income to 5% gives the following results.

upload_2019-2-4_18-51-57.png


upload_2019-2-4_18-52-23.png

If anything, I am understating the actual risk in this analysis as I am using "market" volatility whereas the risk of a small undiversified stock portfolio is considerably higher than that of the market due to the risk of any one company going bust has significantly more impact.

I'll now model the €92,000 income from your post above

upload_2019-2-4_19-5-50.png

upload_2019-2-4_19-6-15.png

I other words there is only a 25% probability you’d make it past 90.

Given the combined life expectancies of you and your other half, you can’t afford to take that risk.
 
Last edited:
Hi Marc
Before discussing your latest posting, could I go back to your previous one (#49)? That shows the value of the portfolio falling in the middle outcome (the black line) of "your" portfolio, and rising only very slightly over the period for "my" portfolio. I had been under the impression that these graphs assumed no withdrawals, but that can't be true if the middle outcome shows market values falling over time. I've obviously missed something. Can you explain?

Now moving to your last posting, you'll have me on the breadline unless I change my ways! The projections are missing the key point that I have no intention of maintaining withdrawals at the current level (adjusting for inflation) if returns don't meet my expectations. As Charlie McCreevy said years ago: "If I have it, I spend it; if I don't, I don't" I'm in the lucky position that I won't be on the breadline if I have to cut back. I'm aware of that possibility; I'm not sure the same is true of my other half! (Thankfully, she doesn't read AAM!).
 
Colm,

They are both taking 5% of the value of the account.

I’ve also reflected your charge differential.

My post 21 includes the work we are doing on variable spending strategies but remember in Ireland you get nailed for the tax on at least 4 or 5% whatever you choose to do in terms of actual withdrawals.

I’d argue in an environment where you are effectively forced to take a minimum withdrawal then a 100% equity strategy is simply reckless.

I’d also suggest and indeed judicial opinion supports the view, that private investors can not afford the downside risk of holding a concentrated stock portfolio when a globally diversified portfolio can be attained at extremely low cost.
 
Last edited:
Or maybe even a higher % for our Sunday Times journalist?!

I'd suggest editing your post.....maybe put "at least" before 4 or 5%!! :confused::eek:
 
Last edited:
I’m revisiting this thread because I’ve extracted some statistics on past returns on my own retirement savings (ARF, AMRF and some non-exempt investments), which may shed light on a question that has generated much discussion on this forum.

The question is whether someone in drawdown (my status for the last few years) is best advised to invest in bonds or equities, or a combination of the two.

The conventional wisdom is that it’s wrong to invest close to 100% in equities because, while they produce a higher return than bonds on average, the occasional nasty market fall can cause havoc. Someone drawing a regular income must cash equities when they’re on the floor. If this happens shortly after retirement, the portfolio might never recover.

I believe that conventional wisdom and associated academic research are wrong. They start from the false premise that the amount withdrawn each month/year remains the same, irrespective of how markets perform. In practice, retirees withdraw more when markets are up and less when they’re down. This changes the conclusion.

My own experience over the last five years bears this out. My savings are entirely in equities. I’m lucky that the amount required to keep a roof over our heads and food on the table is less than each month’s planned withdrawal, where the “planned withdrawal” is calculated as the regular withdrawal that will exhaust our funds by the time we die (assuming we live to a ripe old age). It helps that I’m still working part-time, so I’m not completely dependent on my retirement savings, but I’m not unique in this. Very few go from working full-time to doing nothing.

After making sure we’ve enough to cover the necessities, I work on the principle of “If I have it I spend it; if I don’t, I don’t.”, i.e. when times are good, I splurge on a nice holiday, change the car sooner than I might otherwise, redecorate a room, give better presents to the grandchildren, whatever. When times are bad, I cut back.

I discovered that the money-weighted return on my retirement savings over the last five years (i.e. the “real” return, based on actual amounts withdrawn each month) is a full 1% a year more than the time-weighted return, i.e. the return assuming that there were no withdrawals and no deposits in the period. I believe that this 1% difference between the real return and the notional return is due in large part to my practice of withdrawing more in good times and less in bad times.

I believe that incorporating a simple rule on these lines into academic research, in place of the artificial assumption of a constant withdrawal amount irrespective of market conditions, would change the conclusion in favour of increasing the equity content of a retiree’s portfolio.
 
Last edited:
Colm,

We are currently developing an adaptive drawdown strategy

3844

However, you also need to factor in the imputed distribution requirement in Ireland which is an external constraint on the model forcing a higher liquidity requirement on the portfolio.

This means that the only strategy that "works" is a combination of an equity bias as you set out combined with a liquidity pool capable of smoothing the imputed distributions in down markets thereby avoiding the sequence of return risk making one a forced seller of equities in a depressed market.

However, since the impact of sequence of return risk is more dramatic in the early years of retirement, one could make a theoretical argument for ramping up equity exposure as an investor ages which is of course contrary to the natural inherent conservatism that goes with older investors and whilst theoretically sound I doubt the courts would agree with the argument of increasing the equity exposure for an 80 year old in order to maximize the inheritance for the next generation.
 
Hi Marc
I assure you that the requirement to take at least 4% or 6% a year from an ARF is not a constraint for me. I've no intention of leaving a big inheritance to the next generation(s). I would much prefer to give it to them now. The average return on the portfolio is well north of 6%, so cashing at least 6% a year isn't a problem. In any event, I also have (some of) the proceeds from selling my shares in the business that I once owned/ part-owned. Those funds don't have the constraint of a minimum withdrawal amount.
Liquidity is rarely an issue for someone managing their own portfolio. There's always a cash balance, either from dividend receipts (my largest shareholding has a dividend yield of close to 7%) or from natural turnover: selling down one holding and ramping up another.
Your model only allows for yearly transactions. That's completely unrealistic. You need monthly modelling to get anywhere close to the real market. Looking at my own portfolio (which I recognise is more volatile than most), the monthly market movements over the last 12 months alone have varied from highs of +12.0% and +11.6% to lows of -15.8% and -12.1%. I assure you that I withdrew a lot more in months with positive returns than in months with negative returns. Your model doesn't capture those intra-year changes in withdrawal amounts.
I cannot (yet) speak for 80-year olds. From this remove, I feel much the same as you. I'll probably be OK when/if I get to that age, especially if I'd had another decade of the equity risk premium under my belt by then, but I recognise that others won't have the same insouciance to market fluctuations. You've probably seen my smoothing proposals for Group ARF's and auto-enrolment. One of the aims of smoothing is to ease the concerns of such people while still enabling them to capture the Equity Risk Premium. I hope that someone will have implemented the proposals before the bulk of DC pensioners with substantial funds get to 80.
 
I’d argue in an environment where you are effectively forced to take a minimum withdrawal then a 100% equity strategy is simply reckless.
What would your thoughts be on taking a 5% withdrawal, living on say 3%, and investing the balance of the withdrawal in the same strategy as the ARF (but outside that structure)? Would that reduce the recklessness of a 100% equity strategy?
 
McGaggs

If you take the 5% from the ARF it is taxed.

So you have the net proceeds to invest.

Those proceeds are also subject to tax (income tax and CGT or exit tax) so you would be potentially subject to double tax

Therefore keeping money in the ARF is the “best strategy” which is why revenue force an imputed distribution in the first place.

A better approach is to delay retiring some benefits and leaving these to grow in a pre-retirement structure free from imputed distributions and only ARF such benefits as are necessary to meet required expenditure.

However this strategy has also come under attack by Revenue who will now force income in the form of imputed distributions from age 75 even if you don’t retire benefits.
 
Last edited:
Interesting post Colm.

Is this where capacity for loss comes into play? While retirees do indeed take out less when the value decreases, there is a minimum income required too. When you have an ARF of €2m+, personal investments and are still generating a regular income, you can afford to take the risk of an equity portfolio and so enjoy greater returns over the long term.

For a lot of people, loss aversion is a bigger driver and minimising their losses is more important than maximising the gains.

Academic research tends to look at the numbers only but not the impact of human nature. This is something that we have to work with everyday when advising people on their money.


Steven
www.bluewaterfp.ie
 
Back
Top