4% safe withdrawal rate?

[gkp]

Looking for some ARF investment thoughts. Is it reasonable to assume I can withdraw 4% from an ARF (age 56 so need to last 30+ years, with approx €1m fund)?
Would this require 5% fund growth to cover charges?
What equities/bonds ratio would be needed to get this return?

 

[Zebedee]

Isn’t there an imputed distribution of at least 4%pa? Ie you pay tax on it regardless of whether you take the income on it or not.

Beyond this I think it is best to talk to a financial adviser (I’m not one) to take a wholistic view of your circumstances, finances, risk appetite etc.

 

[Conan]

You must draw down a minimum of 4%pa from the ARF (5% pa after age 70). So in investing the ARF funds you do need to consider the issue of risk v return. However I don’t necessarily see that the objective should be to target a gross return of 5% (allowing for charges). This assumes you are trying to keep the €1m fund intact.
Had you bought an Annuity, you get a guaranteed income for life but the fund is given away. There is no fund available on death (other than perhaps a spouse’s Pension). So targeting a 5% return on the basis that you also want to keep the €1m fund intact on death is seeking to do more than an Annuity. That will involve taking a higher investment risk is order to possibly achieve such a return. That might work out OR it might not work out. Taking a lower risk investment strategy will probably result in a lower return but might also deliver less volatility. So if you take out 4% pa but only earn say 3% pa net of charges, then the fund will gradually reduce over time. But the fund will still see you out. Yes you might have a lower fund passing on to your estate, but so what.
In my opinion, I would focus more on managing the retirement income (and the investment risk) than focusing on leaving the €1m to my estate. You cannot take it with you on death (at least so I am reliably informed).
A good financial advisor should be able to give you a profile of the numbers under various assumptions . So look at 4% drawdown with 5% , 4%, 3% growth rates.

 

[Cervelo]

Couple of things to think about
Can you live comfortably now on a 4% drawdown after tax
Have you factored in years of no growth and what this will do to the longevity of your fund and lifestyle
Have you factored in inflation to your future drawdowns

 

[RETIRED2017]

Lots have other savings /pension income along with ARF drawdown you have to take out 4% min you may not be spending all of it how you invest any spare cash along with pushing out date of starting ARF if it is possible,

 

[Steven Barrett]

Looking for some ARF investment thoughts. Is it reasonable to assume I can withdraw 4% from an ARF (age 56 so need to last 30+ years, with approx €1m fund)?
Would this require 5% fund growth to cover charges?
What equities/bonds ratio would be needed to get this return?

The million dollar question. If we knew how long we would live for, what inflation would be and what the markets would return, we would know the answer.

Assuming a 60/40 mix, taking €40,000 out a year that is index linked for 40 years. Looking at the returns and inflation going back to 1900, the success rate over a 40 year period is 44%. Now, there's 2 World Wars and a Great Depression in there to skew the figures. For example, if you retired in 1914, you could safely withdraw €16,130 a year (inflation proofed) until 1953. Some period horrific investment markets in there.

If you retired just 7 years later with the same pot of money, you could have taken out €71,625 a year for the same period of time!

Then you have to consider that if you retire at 56, you are going to spend more than the 86 year old you. For a good part of your retirement, your spending may increase over time and for the later part your spending will decrease over time.

So really, there's no magic blend or number. You need to look at what you want to do and what you need. Unfortunately, the unknowns of health, mortality, market performance and unplanned for expenditure gets in the way of providing that certainty that people look for.

Steven
http://www.bluewaterfp.ie (www.bluewaterfp.ie)

 

[Sarenco]

Personally, I think it would be unwise to assume that an appropriately diversified, balanced portfolio of publicly traded stocks and bonds will do anything more than simply keep up with inflation over the course of your retirement, after all costs and taxes are taken into account.

Of course, returns on the portfolio might turn out to be better (or, less likely, worse) than that but I think that's a reasonable assumption for planning purposes.

So, if you are projecting a 35-year retirement (entirely reasonable for somebody planning to retire in their mid-50's), you would need a savings pot equivalent to 35 years' worth of your projected expenses. A 60-year old planning on a 30-year retirement would need 30 years' worth of projected expenses, etc.

I certainly agree with Steven that annual expenses in retirement are unlikely to follow a uniform pattern (do they ever!). However, I think it's unwise to assume that expenses will necessarily fall as we age and become less active - long-term/end of life care can be extremely expensive.

Finally, I would personally ignore any State contributory pension or other benefits (including the Fair Deal scheme) for planning purposes. Hopefully those benefits will still exist 30 years from now but really who knows?

I'm conscious that many will find the above to be an overly conservative approach. But in an uncertain world, I think a reasonable degree of caution is entirely appropriate.

 

[JoeRoberts]

). This assumes you are trying to keep the €1m fund intact.
.

v good point Conan.
I think a lot of advisors drive the idea of keeping as much of the fund as possible intact in order to ensure their maximum take from the fund in charges. If the retiree reduces the fund substantially then somewhere among the cloak and daggers of charges the "advisor" loses out. It is a very subtle idea planted and easily sold.

People need to be realistic about how much money they actually need and can spend once they reach 70/75 +, regardless of supposed increases in longevity which generally comes with limiting health factors. This assumes they own a house and don't need to pay rent or mortgage in retirement which is a point Brendan has correctly emphasised for many years that should be a priority over a pension.

 

[Colm Fagan]

I've discussed this topic on a number of occasions with other contributors. I disagree with the conventional wisdom.
See, for instance, post #292 (19 August 2018) of my "Diary of a Private Investor".
I stand by what I wrote then. For what it's worth, my ARF is now worth considerably more than when I started it in December 2010, despite making net withdrawals of more than 4% a year on average since then. I plan to increase the rate of withdrawal to more than 6% a year in future: I prefer to spend the money than leave it to the next generation!

 

[Sarenco]

For what it's worth, my ARF is now worth considerably more than when I started it in December 2010

With respect Colm, that period happens to coincide with the longest uninterrupted bull market in global stock market history. Your personal good fortune hardly represents a sound basis for any prudent retirement plan.

What would the value of your portfolio look like today if you had retired in 1928? Or 1964? Or 1999?

 

[Marc]

The following analysis relates to some of the issues when drawing income from an ARF.

The objectives at retirement can be boiled down to the following:


1. an overwhelming need for income and the sustainability of that income, and

2. a secondary (typically less important) need to leave assets for beneficiaries


The forces working against the achievement of these objectives can also be boiled down to the following:


1. financial market risk

a. this includes general market risk and sequencing risk

2. longevity risk

a. the time horizon in retirement is unknown

b. this speaks to the real risk of an investor outliving their capital base


We wanted to introduce real world constraints in a modelling framework to draw realistic conclusions and guidelines from our research.

The largest differentiator of our research is that we used a simulation process to produce portfolios that take any number of different paths possible given the risk and return characteristics of the portfolio.

Higher risk portfolios can withstand higher drawdown rates over long periods of time.

However, even with low drawdowns there is a small chance of failing over the period.

Low risk portfolios over long periods of time guarantee failure.

Risk assets are absolutely necessary to generate inflation beating returns and to
maintain living standards
Drawdowns above 6% become touch and go. 4% drawdown rates seem generally sustainable but not for conservative portfolios.

What about spending in retirement?


“Our findings suggest that typical consumption in retirement does not follow a U-shaped path –
consumption does not dramatically rise at the start of retirement or pick up towards the end of life to meet
long-term care related expenditures.”

Dr Brancati, International Longevity Center


Consumption in retirement starts relatively high and ends low. This pattern is common to both high and low
income groups, is robust to the inclusion of factors other than age and is not simply the result of the time period
in which the data was collected.


Of course many people will need care in later life, but this is not typical.

Consider the following:

• only 16% of people aged 85+ in the UK live in care homes
• the median period from admission to the care home to death is 462 days. (15 months)
• around 27% of people lived in care homes for more than three years, and
• people had a 55% chance of living for the first year after admission, which increased to nearly 70% for the second year before falling back over subsequent years.

Source AgeUK 2016

Some questions you need to ask yourself;


How much risk do you need to take in order to support a given income?

Is an ARF the right answer should you use an annuity?

What happens if the first few years of your retirement experience negative returns? What should you do?

How do you adapt to changing market conditions, changing interest rates changing inflation expectations?

Should you expect mean reversion in asset classes? If so how should you respond?

 

[Steven Barrett]

Finally, I would personally ignore any State contributory pension or other benefits (including the Fair Deal scheme) for planning purposes. Hopefully those benefits will still exist 30 years from now but really who knows?

Didn't you say on the mortgage lump sum or start a pension thread

Speculating about future taxes or levies is futile - we can only make reasonable decisions based on we know today.

Joking aside, from a planning point of view, potential home care screws everything up. You can be looking at about €60,000 a year for home care. And to be conservative, 5 years of care (people usually don't last longer than that in a home). Even without inflation, that's €300,000 net income. As the ARF is taxed, we'd be looking at about 50% of the ARF value to be kept just in case you find yourself in a care home. That's a lot of money not to spend when young an healthy for something that may or may not happen in 20/ 30 years time.

The current situation is that the State will look after you if you need a care home. You pay for all of it as long as you can afford to, then the Fair Deal kicks in and you give up some of the value of your home.

v good point Conan.
I think a lot of advisors drive the idea of keeping as much of the fund as possible intact in order to ensure their maximum take from the fund in charges. If the retiree reduces the fund substantially then somewhere among the cloak and daggers of charges the "advisor" loses out. It is a very subtle idea planted and easily sold.

People need to be realistic about how much money they actually need and can spend once they reach 70/75 +, regardless of supposed increases in longevity which generally comes with limiting health factors. This assumes they own a house and don't need to pay rent or mortgage in retirement which is a point Brendan has correctly emphasised for many years that should be a priority over a pension.

If your advisor is taking that approach, you'd question the solvency of their business. You save for decades for retirement so you can spend the money and enjoy yourself. You are now entering the deccumulation stage of life when you start spending all that money you have saved for so long. People who have been prudent enough to save for their retirement are usually prudent enough in the spending of it too. They don't want to blow it all in the first few years, so you see a gradual reduction of the value of ARF funds over the years, not a sharp decline.

Those with small ARF pots, should be taking out as much as possible each year within the tax limits.


Steven
http://www.bluewaterfp.ie (www.bluewaterfp.ie)

 

[Sarenco]

4% drawdown rates seem generally sustainable

Over what timeframe? 25, 30, 40 years? I assume you mean 4% adjusted for inflation.

And how would you define what is "generally sustainable"? Would you consider a 10% chance of failure, based on your model, to be acceptable? 5%?

only 16% of people aged 85+ in the UK live in care homes

I'm actually surprised it's as high as 16%. An increasing number of seniors obviously choose to pay for long-term care to be provided within their own homes.

Of course many people will need care in later life, but this is not typical.

Is that not a contradiction in terms?

 

[Cervelo]

Drawdowns above 6% become touch and go. 4% drawdown rates seem generally sustainable but not for conservative portfolios.

Over what timeframe? 25, 30, 40 years? I assume you mean 4% adjusted for inflation.
And how would you define what is "generally sustainable"? Would you consider a 10% chance of failure, based on your model, to be acceptable? 5%?

Sorry for jumping in on your question Sarenco but I was wondering if Marc could clarify his interpretation of a "conservative portfolio" when responding

 

[Marc]

For example this portfolio is very conservative (expected volatility under 3%) and has a Gross expected return of CPI + 1.46%pa before costs. Take a 4% annual income and you are going to run out of money.


Higher risk portfolios can withstand higher drawdown rates over long periods of time. However, even with low drawdowns there is a small chance of failing over the period.

Low risk portfolios over long periods of time guarantee failure. Risk assets are absolutely necessary to generate inflation beating returns and to maintain living standards

Drawdowns above 6% become touch and go. 4% drawdown rates seem generally sustainable but not for conservative portfolios



IMPORTANT

Our analysis assumes that the income is 4% of the starting portfolio value.

Imputed distributions are 4% of the CURRENT value of the ARF

So, if the value of the ARF is €1 you are only required to impute an income of 4 cents so you will be left with 96cents.

An ARF cannot "bomb out" due to the imputed distributions since 96 cent is still a "positive value"

However, I'm more interested in declining living standards over retirement. If your income is down to 4 cent a year you have an issue

 

[WhiteCoat]

What guidance notes/best practice have the professional bodies issued around this question and how has the guidance changed over the years? Are the various representative bodies broadly in agreement? Are links available?

 

[Colm Fagan]

With respect Colm, that period happens to coincide with the longest uninterrupted bull market in global stock market history. Your personal good fortune hardly represents a sound basis for any prudent retirement plan.

Sarenco, you're right: I was lucky in my timing - but not that lucky. As recorded in my diary update of 7 January, my portfolio fell by 28% in 2018. The ARF fell 15%; other investments (including the AMRF) brought the average fall to 28%. The long-term returns quoted in my last post allowed for the 2018 result.

My general point is that, as @Marc noted above, long-term income, not short-term security of capital, is the prime requirement for an ARF holder. (And by "income" I don't mean dividends; I mean the income derived from capital gains and dividends). Cash delivers a zero income, bonds aren't much better (possibly worse in the long-term). Good quality equities and real estate are the best bet for achieving a reasonable long-term return, especially considering that the expected lifespan for a new retiree is probably more than two decades when future improvements in medical science are allowed for.

The "sequence of return risk" has often been cited as an argument against high equity levels in a retirement portfolio. The risk is overblown. Any theoretical studies I've seen on this topic start from the false assumptions that (i) the pensioner withdraws a constant amount each year, and (ii) they cash investments equal to the amount withdrawn. Both assumptions are wrong.

If I do well in a year, I treat myself - bigger presents for the grandchildren, a nice holiday, whatever - but if I've done badly, I cut back on expenditure. That doesn't fit with the model, but it is the reality.

Theoretical models also assume that the only way to derive an "income" from an ARF is by cashing investments. Speaking once again from experience, that's just not true. At the end of last year, when market values were on the floor, I had to raise some cash, but I managed to do it without selling investments (or by selling very little). There is always a small cash balance in the portfolio. At the end of 2018 I ran it down almost to zero. That too is a significant departure from the theoretical model, which would have me cashing investments to meet income needs, even if I had cash in the portfolio.

If the model allows for withdrawals to be flexed (even slightly) in response to market returns, and also allows for "income" to be taken from the cash balance in the fund when markets are depressed (with the cash being replenished when market returns are good), I'm sure that the matrix @Marc kindly shared with us would look quite different. As it stands, it should carry a health warning: "This matrix is based on artificial assumptions, which might bear no relationship to what happens in practice".

 

[Gordon Gekko]

I agree that the risks seem to be overblown.

For someone with €1.5m in his ARF, why not just keep, say, €300k in cash and €1.2m in equities?

And, to pick up on Colm’s point, isn’t a decent chunk of the 4% covered by the dividend yield on a typical portfolio?

 

[elacsaplau]

Over what timeframe? 25, 30, 40 years? I assume you mean 4% adjusted for inflation.

Excellent question. I, too, await clarity.

What guidance notes/best practice have the professional bodies issued around this question and how has the guidance changed over the years? Are the various representative bodies broadly in agreement? Are links available?

I don't believe any professional body in Ireland has produced meaningful guidance in terms of asset allocation, safe withdrawal rates, etc.

The "sequence of return risk" has often been cited as an argument against high equity levels in a retirement portfolio. The risk is overblown. Any theoretical studies I've seen on this topic start from the false assumptions that (i) the pensioner withdraws a constant amount each year, and (ii) they cash investments equal to the amount withdrawn. Both assumptions are wrong.

These are not false assumptions. The purpose of many excellent studies is to determine the "safe" withdrawal rate. These studies are designed to give investors a guide so that they better understand the trade-off between maximising returns and maximising income security. They also, importantly, model the impact of various rebalancing strategies. These reports provide extremely useful information and show, for example, that a high equity content is the optimum strategy compared with the old adage of age in bonds. "High" in this context can differ between reports but typically is in the 60% to 80% range. I have seen no credible report that has advanced "age in bonds" as the optimum play to best match the competing needs of the retiree.

I can recall nothing in these reports which suggest that people must slavishly follow an X% rule. For example, if the experience of a given retiree has been positive by a given point in time in his retirement, there is nothing to prevent him from taking a one-off lump sum or increasing his withdrawal rate, etc. Equally, these reports help to illustrate that the retiree could keep the withdrawals at a constant level to de-risk or have the comfort of knowing that they are likely to be leaving something behind. Favourable experience can also clearly serve as a protection against severe healthcare costs, etc.

In relation to the second point, the reports are designed with the typical individual investor in mind who invests his retirement account in unit linked funds. It is unfair the say that the approach adopted is false when the standard approach to withdraw funds in unit linked contracts is to encash units. What alternative assumption would you employ?

 

[Cervelo]

I gather from your post Marc that you would consider a 30/70 portfolio as conservative, presumably a 50/50 in neither here nor there and a minimum 70/30 would what be required to sustain a 4% withdrawal
But what is the best way or current thinking about how to build the portfolio, my current portfolio is a 50/50 with 6 ETFs and I'm thinking that it needs to move to a more aggressive stance and that 6 ETFs is to many
My current thinking is I need to reduce these to 3 which I know will narrow the diversity, increase the risk but I feel its the better option for my goals. but I'm also questioning whether EFTs is the right option
I note Colm Fagan's route (if I have this right) is to build a portfolio of individual stock and shares rather then ETFs and am thinking this might be the better option but wondering about diversity
as in do I hold 10 stocks of 100k or 100 stocks of 10k, I have been dabbling in the stocks since the summer and am enjoying the ride but not sure I would be confident enough to build a good portfolio
I know I'm asking a question that's the same as asking "how long is a piece of string" but I after the thinking behind each type of portfolio??