New Sunday Times Feature - Diary of a Private Investor

My evidence is my own research for the paper mentioned above. .

Colm,

Your proposal involves a pooling approach which has many merits which I have already acknowledged elsewhere previously. However, my question was a specific reference to the high-lighted quote.

The statement I was questioning was about individual investment allocations. The very reason pooling has merits (apart from improved charges) is that it minimises investment and longevity risks. On an individual basis which is what we are talking about here and implied by your statement (the one that I was questioning), all-in equity is simply not the most effective strategy as certain individuals will, sooner or later, get clobbered by a sequence of return risk.

In my opinion, the evidence produced by the gentlemen referenced is both extensive and conclusive - hence the reason for stating that I would be amazed if contrary credible evidence is available. I have looked at this deeply and have never seen any credible contrary material.


Duke,

There are piles of analysis on averaging-in investing (DCA or ECA, if you will). Asset allocation and withdrawal strategy are the key as per material referenced earlier. I'd love to see credible Euro based comparative studies for many reasons!!
 
Last edited:
In my opinion, the evidence produced by the gentlemen referenced is both extensive and conclusive - hence the reason for stating that I would be amazed if contrary credible evidence is available.
As I said, I haven't studied their work. I am satisfied however that the approach I proposed (which involves smoothing across time and across cohorts) allows 100% investment in suitably diversified real assets. I'm out of commission for a few days from tomorrow.
 
Colm,

In the context of this discussion, your comments which I high-lighted in post 200 have/had a definite implication/inference at individual investor level. I believe these comments to be simply incorrect and provided extremely reputable supporting sources. Larry Swedroe is another who has much useful material in this context also. For many reasons, I'd love to be shown contrary evidence. If credible contrary evidence does exist, I'll be beyond gracious in my acknowledgements!

This has nothing to do with your pooling proposals which as I have said have many merits!
 
Last edited:
When you say a mix between bonds and equities is recommended, can cash substitute for bonds?

I agree it seems risky to have 100% allocation of equities.

It seems obvious that's risky.
 
A quick read of an interview with Pfau demonstrates the value in having a percentage in an annuity
In the event that a person lives a long time the annuity continues to pay. It's a pooled bond that had advantage when one lives a long time.

So back to question asked by Duke
How about 25% into annuity
And then drip feed the 75% into equities over say 5 years

At that rate year 1 25% annuity 15% equity
Year 2 30% equity
Year 3 45% equity
Year 4 60% equity
Year 5 75% equity


That doesn't account for withdrawals
 
Last edited:
Duke,

I've been thinking about your question.

If I was going to an adviser, what I'd like to now is:

1. What is the asset allocation that has best chance of success?
2. Do I immediately place funds on this basis or do I average in?
3. What is the withdrawal strategy - i.e. from which asset classes?
4. How is the portfolio balancing done?

My contention is that for the individual or individual couple (for the avoidance of doubt I needed to mention this ;)) that these are the right questions! There's a whole branch of finance that examines this question in the U.S. generally referred to as safe withdrawal* rates - I am unaware of any publicly available comparative evidence-based studies in Euroland. It will be interesting to get the comments on this from the financial planners within the AAM community.

The bottom line is that a couple, both aged 60, seeking an annuity with 100% reversion and a modest enough 2% escalation rate, would receive an initial annual income of just over 2% according to the Irish Life calculator just now. That is - for a million Scooby-doos, the annual pension is €20,280 - which is highly unattractive in my opinion.

There must be a way to improve on this - whilst restricting "bomb-out" risk to the absolute minimum. When minimising bomb-out risk is a concern, a 100% or a 100% plus! return seeking asset allocation is just not a good idea.

*not to be confused with family planning, Irish-Catholic style!!
 
Last edited:
Duke,

I've been thinking about your question.

If I was going to an adviser, what I'd like to now is:

1. What is the asset allocation that has best chance of success?
Thanks for the comprehensive reply elac. I too have been playing with the Irish Life annuity calculator and your example is very close to that of my friend. Yes annuity rates are awful. But the problem with this question is what does it mean? Investing in emerging markets may have the best chance of success but it also has the "best" chance of failure. My initial thoughts are to "advise" some balanced passive fund, possibly with global diversification. Having said that I think any investment in govies at these levels is very bad value.
2. Do I immediately place funds on this basis or do I average in?
That's the question I am currently grappling with. I think averaging in reduces timing risk for her and for me as her "advisor". Referring to an earlier comment I think "sequencing of return risk" is mainly concentrated in the early years. If you get off to a good start you are reasonably protected against future bad runs in the market.
3. What is the withdrawal strategy - i.e. from which asset classes?
4. How is the portfolio balancing done?
I envisage only one asset class so these questions become redundant.

On annuitisation I feel fairly confident in saying not now. But maybe at age 75 some annuitisation should be considered. At that age the mortality kicker is greater than the additional return you might expect from investing in equities. The point about insuring longevity risk is it can be deferred unlike other risks such as mortality or health where you might defer it till it becomes an uninsurable risk.
 
Excellent observations, Duke.

I am hosting lunch today - seemed like a good idea at the time! - so I better get the shrimp on the Barbie prompto. There's a lot in it - arguably deserving of its own thread so we don't bring Colm's thread off on a tangent - albeit an important one.

So a few comments for now:

1. Kick the tyres on the material produced by the three guys I referenced last night - you'll enjoy it - there's some smart thinking therein!

2. Your comments on "emerging markets" (and by extension return seeking assets generally) are spot on - generally will provide the best outcome but in this case, generally ain't good enough (until Colm's "pooled" solution gains traction.......this stuff is highly personal!)

3. Similarly, Euro-govies are really problematic in translating the US research to the Irish ARFer / income drawdowner.

4. Agreed that "sequencing of risk" is significantly skewed in its impact on the early years (tús maith leath na h-oibre and all that).

5. I need to consider the bit about one asset class. [Presumably, if your pal is in a balanced fund - there's more than one asset class? The research I referred to is very interesting and it involves having different pots of return seeking and defensive assets and examines the optimal basis for withdrawing funds. I honestly never thought about how this applies in a "balanced" fund environment which is pure dumb of me. I need to think about the implications of this.]

6. Regarding annuitisation - fair comments. Just to add - I remember reading (some time ago so I'm a little hazy on the detail) a pretty convincing analysis by Larry Swedroe as to why deferred annuities are a very valuable tool for the "healthy" retiree (i.e. one buys an annuity at say age 60 that only starts paying out from age 75 or something.) The idea being that a % of one's retirement fund is allocated to such a product. [Interestingly, a bit of an overlap here with Colm's pooled solution!]

It goes without saying that US annuities are way higher than ours anyway due to the difference in sovereign yields. There is no such product available in these shores - unless something has changed very recently.
 
Upon mature reflection drip feeding is illogical. After the end of the drip feeding period you are right back where you started. Transaction costs aside, you still have the choice of being fully invested, fully in cash or somewhere in between and if drip feed was the right choice in the beginning then cash in and drip feed all over again is the right choice at the end of the period. Of course you might think current markets are artificially high, as I do, but that inexorably says don't go in yet, wait for the correction. However, it would be hubris on my part to give that advice to my friend. This is difficult:cool:
 
Last edited:
Duke,

Where's the money coming from? - as in is it from a DC arrangement or what? [Where money is coming from a DC arrangement, in the "normal" scenario, averaging-in would not be a particular issue because the money has already been invested. Make sense?]
 
Duke,

Where's the money coming from? - as in is it from a DC arrangement or what? [Where money is coming from a DC arrangement, in the "normal" scenario, averaging-in would not be a particular issue because the money has already been invested. Make sense?]
No it's from a DB arrangement.
 
Colm,

In terms of income drawdown in retirement, if you have any evidence to support this statement, I'd love (and be amazed :D) to see it.

In the U.S., others like Wade Pfau and M.McClung have produced very detailed research to the exact contrary.

I’ve now had a chance to look at Wade Pfau’s conclusions.

First of all, we are comparing apples and oranges: my proposed approach differs in a number of key respects from the approach described by Wade Pfau so his conclusions cannot be used to draw any conclusions from the work I’ve done. I think I’ve explained my approach in some detail, and I stand by my statement, as quoted by you in post #200 on this thread. If you disagree, let me know which of my conclusions/ assumptions you don’t accept, and we can take it from there.

In this response, I would like to focus exclusively on Wade Pfau’s work and explain why I have serious reservations about his methodology and conclusions. In a nutshell, he takes a highly theoretical approach, which ignores the reality of stock market investing.

To keep things simple, I’ll just focus on the scenario where he assumes 100% investment in equities (he also looked at bonds and bills, in varying percentages for all three).

He assumes 100% of the money is invested in the index at the outset. Then 4% (or whatever) is withdrawn each year by cashing the required number of units. Taking this approach, he found that the money ran out in a significant proportion of past time periods.

Let’s look at what this means for a real-life portfolio invested in equities, not his artificial construct.

1. He assumes that every single cent is invested in equities at the start, with no cash being retained in the fund. This is wrong (but it’s only a minor mistake). Every equity fund has a small liquidity element, probably less than 5%. This cash element will in practice be used if needed to meet withdrawals in the first few months. As we shall see later, however, it may not even be needed. If the liquidity in the fund is used to fund “income” payments, it will have to be replenished to ensure the liquidity level never falls below (say) 1%.

2. He assumes that all dividends are reinvested immediately, even if the equities thus purchased are sold again within minutes to fund “income” requirements. That’s a ludicrous assumption. If someone is taking a constant “income” from the fund, they will hoard whatever dividends they’ve received until the next “income” payment is due, thus reducing the number of shares that need to be sold to fund the regular outgo.

3. Much the same is true for stock turnover within the fund. The mechanics of this depend on whether the fund is being managed actively or passively. Let’s assume it’s being managed passively (the argument is even stronger for an actively managed fund). Some people think that a passively managed fund has little or no stock turnover. That’s what’s implied by the term passive. It’s far from the case. A passive fund must track the index, so if a stock is removed from the index, the fund must sell the shares in that particular stock and buy ones in the stock that replaces it in the index. Similarly, a share buyback, which reduces the number of shares in issue, requires a passively managed fund to sell some of its holding of that share. The opposite is true for rights issues, or IPO’s. As an aside, managers of passive funds objected to Mifid II’s decision to include the costs of these purchases and sales in the costs of such funds. Here again, Mr Pfau in his wisdom assumes that, if a company is removed from the index, the fund will sell shares in that company, then buy the shares in the company that replaces it in the index and immediately sell those shares again to pay the “income” to the beneficiary. Very intelligent behaviour, I don’t think. In real life of course, when the fund sells the shares in the company that has been removed from the index or that has completed a buyback, it will not reinvest if it needs the money to pay an “income” to beneficiaries (assuming it’s already used up dividend receipts and hasn’t enough cash for the “income”). In fact, rebalancing is much easier in a fund where there is a continuing flow of new money or a regular outflow than where there is no net change in the money invested in the fund.

4. He assumes that the investment strategy for a fund devoted exclusively to paying a continuing regular income to beneficiaries is exactly the same as for a fund with no requirement to deliver a regular “income”. Once again, that is completely unrealistic and artificial. I’ve looked at my two biggest portfolios to explore the truth or otherwise of this assumption. One of the portfolios is my ARF, from which I am required by law to take an “income” of 6% every year. The other is the home for the proceeds from selling my shares in the business where I once was a major shareholder. I don’t have to take an “income” from that portfolio. Both portfolios are invested virtually 100% in equities. I decide the equities to hold in both portfolios. The average dividend yield on the portfolio from which I must take 6% per annum is more than three times the dividend yield on the portfolio from which I don’t have to take a regular income. The average dividend yield on the two portfolios combined is probably not too far off the average dividend yield for the index. I would think my experience is not untypical of the market.

5. I’m sure there are other artificial assumptions that I haven’t considered. For example, it’s likely that Mr. Pfau assumes that “income” is taken in one lump sum at the end of each year. In real life, income is taken gradually throughout the year. This assumption probably has a significant impact on the conclusions.

What do these differences mean in practice. I don’t know. I can only relay my own experience. My ARF has been in drawdown since 2010 and I don’t recall a single instance of having to cash an investment to meet the “6% income” requirement. Note that I’m NOT saying that the fund returned more than 6% each year. No. I’m only saying that the 6% “income” requirement each year was met by a combination of dividend receipts, or temporarily allowing the level of liquidity in the fund to fall, or holding back proceeds from selling stocks that I’d gone cold on instead of reinvesting the entire proceeds in stocks that I preferred at that time. There was also an element of timing when I took the “income”, i.e. I didn’t take it on the same date each year; instead, I took it when I thought the time was opportune. That safety valve wouldn’t be available to a professional manager who promised to deliver €X per month to a beneficiary, but it hasn’t played a significant role in the end result of not being forced to sell investments to meet the income requirement. The consequences for Mr Pfau’s conclusions don’t need to be spelt out.

Let me know if you still believe Mr Pfau.

I will reiterate what I said at the start, that my proposed approach is quite different.
 
Colm,

Is there a certain amount of times that I need to say the same thing?:D

Let's start by agreeing - I too think it's a mighty fine idea not to conflate apples and oranges. However, how we apply this belief in practice seems to differ somewhat. Here's my sense of the recent twist in this debate. I'll admit to getting very trigger happy with the quote function....

1. The venerable Duke asked a question about the asset allocation strategy of his friend (singular/solo/individual, etc,). It was a query at the individual level and not a question in relation to your pooling proposal. Let's call it a query in relation to apples.

2. You replied
I recognise that others won't share my risk appetite, yet their best hope by far of receiving an adequate income throughout their (hopefully) long years of retirement is to invest the absolute maximum possible in real assets - equities, property, and other assets with similar characteristics.

I took this, as mentioned previously, as a comment in relation to the Duke's query about the individual - a reply about apples

3. I questioned this comment because I have seen piles of research saying exactly the opposite to your assertion.

In terms of income drawdown in retirement, if you have any evidence to support this statement, I'd love (and be amazed :D) to see it.
In the U.S., others like Wade Pfau and M.McClung have produced very detailed research to the exact contrary.

I was talking about apples.

4. This was met by your response:

My evidence is my own research for the paper mentioned above. Importantly, my approach in its purest form involves pooling across generations and across market peaks and troughs (through smoothing), with a constant flow of new money and exits, although some limited analysis indicated that it also worked for single cohort entries.

In other words - for some reason in post 201, in response to my question about apples, you started to talk about your pooling proposals - let's enjoy ourselves and call this a reply about oranges.

5. I politely tried to get the debate back on track and in my next post (203), I again clarified in an attempt to avoid the confusion:

Your proposal involves a pooling approach which has many merits which I have already acknowledged elsewhere previously. However, my question was a specific reference to the highlighted quote.

........where the highlighted quote was in relation to the statement that you made as set out in point 2 above.

In other words, I was asking can we stay talking about apples not oranges please.

I thought that that should clarify things!

6. In your next post 205 - guess what?

I am satisfied however that the approach I proposed (which involves smoothing across time and across cohorts) allows 100% investment in suitably diversified real assets.

Hey - you're back on to oranges again?

7. At this stage, I'm beginning to wonder what form of words, if any, I can use to politely keep everyone on the same page - so I go with:

In the context of this discussion, your comments.........
This has nothing to do with your pooling proposals which as I have said have many merits!

A desperate imploration for you to stay with apples

*******************************

So when I saw your post just now - with a "first of all, we are comparing apples and oranges".......I just laughed and cried.........This is getting old! It's just too hard work! ;)


My energy to debate this further has pretty much all but evaporated. At the outset, I was simply asking you whether you had any credible evidence to support your original assertion (i.e. the one that I challenged as per point 2 above).

I continue to wish you good luck with your pooling proposals, your beloved oranges. Also, I think there may be merit to some of your criticisms of Wade's methodology in his study of apples but my sense is that his central thesis remains valid. As a by the by, have the Society of Actuaries in the Ireland or the UK done anything on this particular question? If yes - what does it say? If not - that would be somewhat curious.

I know I'm giving you a little hard time - but I must admit to finding the oranges/apple thing a little irritating. Also, you are a thought leader and many people will be influenced by you and some will just follow you. Personally, I remain of the view that your recommended asset allocation is certainly far from being "by far" the best solution to minimising bomb out risk! Frankly, your personal experience over the last eight years does not provide me with sufficient comfort - not the most robust stress test!

The point of all the US studies that I have seen is that all-in equites is akin to the old adage of generally working very well but not working for "all of the people all of the time". Also, withdrawal rates are a very important part of the equation.

Just to be clear - what I'm talking about is a Japan style melt-down or worse. It may never happen but it could might and then, as Mr. Buffett reminds us - it's only when the tide goes out....

Oh - ar eagla le h-eagla - in the last few paragraphs I've been talking - almost exclusively about granny smiths again!! Nobody can blame me for mentioning this, right?!
 
Last edited:
What did I do to deserve this?
I now gather that you're taking issue with my statement "...their best hope by far of receiving an adequate income throughout their ... retirement is to invest the absolute maximum possible in real assets ..". That's not an opinion; it's a fact. If we define "an adequate income" as around 4% per annum (real), then an investor has zero chance of achieving that income in retirement if they invest in cash (currently yielding around zero) or bonds (highest Eurozone bond yield less than 1.5% per annum). Their only chance is to invest in real assets (property, equities, etc.), which have an expected real return north of 4% per annum. The chances of receiving that income throughout retirement are less than 100%, due to the sequence of return risk and the risk that the expectation of greater than 4% will not be realised. I never said it was 100%; I just said it was their "best" chance. Given the opposition from bonds and cash, "best" means anything above zero. While it's not 100%, I'm sure you'll agree it's above zero.
Have I now given the answer you've been asking for?
You made a big issue about work completed by the various authors quoted.
others like Wade Pfau and M.McClung have produced very detailed research to the exact contrary
the evidence produced by the gentlemen referenced is both extensive and conclusive
I believe these comments to be simply incorrect and provided extremely reputable supporting sources.
I put in some effort to understand what they were saying and to set out my disagreements. Given the reaction, I'm sorry I made the effort.
As a final comment, I enjoy engaging with contributors on technical questions. I have no time for contributions that do nothing to advance our understanding of the issues.
 
Two final observations - then I will need to park this for a few days to do other stuff.

1. ARFs were introduced in Ireland in 1999. I haven't done the back-testing since then but I'd be amazed if someone with an 80/20 (equity/bond) mix hasn't done better in financial terms than someone with a 100% equity allocation - if using sensible withdrawal strategies, rebalancing methodologies, etc. as described by say McClung in his excellent book "Living off your Money". [As a by the by, their emotional well-being is likely to be substantially better - all other things being equal - but hey, lets not go or introduce bananas at this stage! Associated with this, they would also be likely to have a substantially reduced risk of making behavioural finance errors like bailing when the waters get choppy!]

2. There is a very practical side to my original question. A couple, at the point of retirement with fund of €1m invested on a 80/20 equity/bond split, go to see their financial adviser.

I think we can all agree that asset allocation is the primary determinant of investment performance - so the couple ask the adviser - what is the optimal asset allocation for us in retirement where our objective is to minimise bomb-out risk but still have a substantial exposure to equities? [This is turn could lead into discussions about effective withdrawal strategies, re-balancing, etc.]

And so the real question becomes........how would an adviser in Ireland answer this question and what evidence exists to support such advice? I'd really love to know how the adviser community addresses this question in practice now - if indeed it is addressed at all in such terms!!
 
Colm,

Our posts crossed.

You seem to continue to misunderstand the point I was making. I give up. The points about 100% bonds/cash in your latest post are silly. None of the research is suggesting this and this is just another example of you writing about something I haven't said. The bulk of the research is that high equity allocations is very good. There is a difference between high and 100%! Why do you think Buffett gave a 90/10 instruction for his legacy assets?

We had similar nonsense in another thread which I previously highlighted to you.

Let's not get so defensive - in essence I asked you to produce evidence to support a statement that you made. I have seen none!!:D
 
Last edited:
I too give up. I said "the maximum possible". That doesn't necessarily mean 100%. The calculation of "the maximum possible" should take whatever constraints there are into account.
In relation to constraints and the debate about percentages to invest in various assets, I have a real problem with people using the term "equities", as if all "equities" were homogeneous. As you know, I'm a stock-picker, and every individual stock in my portfolio has its personal attributes, which make it different from other "equities". For example, one of my holdings is a company that owns a big shopping centre and a range of other office and retail properties. It's trading at about 75% of net asset value. That can be considered a property investment. Another of my holdings gets a regular stream of income from a very secure asset and pays a dividend of around 7% per annum. That is my "bond" proxy. Needless to say, Renishaw is a different kettle of fish again. They are very disparate businesses, with very different prospects in different economic conditions. I have very little time for theoretical claptrap that considers them as a single entity.
 
Last edited:
ARFs were introduced in Ireland in 1999. I haven't done the back-testing since then but I'd be amazed if someone with an 80/20 (equity/bond) mix hasn't done better in financial terms than someone with a 100% equity allocation -
Assertions like this frighten me. The main reason for the good performance of bonds since 1999 is because yields fell dramatically. Part of the return came from eating the seed-corn, which is all the more reason why someone investing in bonds now is condemned to a lousy return. In other words, good past returns on bonds is precisely why they're such bad prospects for the future.
 
AAM readers will know my solution: invest 100% (or more!) in equities. I recognise that others won't share my risk appetite, yet their best hope by far of receiving an adequate income throughout their (hopefully) long years of retirement is to invest the absolute maximum possible in real assets - equities, property, and other assets with similar characteristics.

Colm,

This is the relevant post. Sure the maximum possible does not mean a 100% equities. It's kind of implied, though, at least to a considerable extent by the previous line - as bolded above. Nonetheless, in case there was doubt in my understanding of what you were saying, I mentioned a few times "all-in" equities!! To me - "all-in" equities has a very definite meaning. Anyway, I can't allocate any more time to this now and do not wish to engage in any further I said/you said. That would just be silly.

I agree with you that the substantive issue is what is important and regret the turn this thread has taken. I just asked a simple question at the outset. Re-reading the thread, I am happy that my question and attempts at clarification were clear. Let's just leave it at that please!

I need to leave this debate to others for a few days now - else my own retirement will arrive sooner than anticipated! Hopefully, the adviser community in Ireland will comment on how they address the optimal asset allocation for retirees and what data informs such advice. For the avoidance of doubt, I understand different retirees will have different risk tolerances so I get that it's not a one size fits all. :(

I have just seen your comment on ARFs - I am not going to comment further other than to say that I have mentioned earlier how tricky the optimal asset allocation is in the Irish context given current levels of Euro sovereign yields. I really do need to go now and will not reply further......at least for a few days!!:)
 
Back
Top