Irish Times opinion piece by Colm Fagan: "Auto-enrolment plan seriously flawed"

[nest egg]

Hi @Nermal
An assumption on future returns plays a minor role in the smoothing process, particularly at the start, but it really is minor, and becomes more minor over time. The smoothed graph can be viewed as a line cutting through market values, reducing the magnitude of the humps and hollows of market value movements. I've tried to append the graph of smoothed and unsmoothed indices if contributions were invested in the UK index between 1990 and 2019 to show how it looks pictorially. I hope my stupidity in all things technical hasn't stymied my efforts.
In case that hasn't worked, I attach the paper I submitted to the UK Institute and Faculties. That shows the same graph on the bottom of page 10. I suggest you also run through the numbers in Table 1 on page 9, calculating smoothed returns for the first six months of 2020, assuming a scheme start date of 1 Jan 2020 (just before Covid struck the stock markets). The calculations should answer your question. They show monthly unsmoothed returns spanning a 20% range (from -15.1% to +4.9%) while smoothed returns span a 0.2% range (0.13% to 0.33%).
I'm sorry if the answer to your question is a bit long-winded, but if it were child's play, it would have been done years ago. I think you'll get the hang of it quickly though if you just take 15 minutes or so to put the numbers of table 1 into a spreadsheet. It could actually be a bit of fun!

I'm going to ask a very naïve question here... isn't there a fund someone has already come up with, which has an optimal mix of stocks/bonds/other investments, which would provide a similar smoothing effect, and avoid the need to re-write the rule book? (eg: Vanguard Global Balanced Fund)

 

[Duke of Marmalade]

I'm going to ask a very naïve question here... isn't there a fund someone has already come up with, which has an optimal mix of stocks/bonds/other investments, which would provide a similar smoothing effect, and avoid the need to re-write the rule book.

Yes, it's called "lifestyling" and involves giving up all or most of the Equity Risk Premium with a resultant major negative impact on long term pension savings. It is the default approach for many and was the basis of the original "strawman" proposal put forward by the Department of Social Protection which has not changed.

 

[nest egg]

Yes, it's called "lifestyling" and involves giving up all or most of the Equity Risk Premium with a resultant major negative impact on long term pension savings. It is the default approach for many and was the basis of the original "strawman" proposal put forward by the Department of Social Protection which has not changed.

Lifestyling, at least for my pension provider, is a process which gradually moves my funds into less and less risky assets, as I get closer to retirement. Therefore I could miss out on all those (potentially juicy) returns close to my retirement age. Which isn't what I'm saying at all - but maybe I've misunderstood the lifestyling concept.

In my case, I've opted out - and instead I'm invested in a passive equity fund, until I retire unless I make a further change. Ideally though, and taking inspiration from Colm's concept, I'd like to have the highs and lows smoothed out. both investing in, and withdrawing from the fund, along the way.

If I had access in my pension to invest in say Vanguard's Global Balanced Fund, would I get more or less the same benefit as Colm's approach? A smoothed return.

Ideally I would stay in this fund, with my current pension provider, after I retired, but even if I had to "cash out" and setup an ARF with the same fund at that point in time, my approach would stay the same (albeit with likely higher charges for the reasons other posters have already mentioned).

Is there a distinct advantage to Colm's approach vs investing in the long term in something like that Vanguard fund?

 

[Duke of Marmalade]

Lifestyling, at least for my pension provider, is a process which gradually moves my funds into less and less risky assets, as I get closer to retirement. Therefore I could miss out on all those (potentially juicy) returns close to my retirement age. Which isn't what I'm saying at all - but maybe I've misunderstood the lifestyling concept.

In my case, I've opted out - and instead I'm invested in a passive equity fund, until I retire unless I make a further change. Ideally though, and taking inspiration from Colm's concept, I'd like to have the highs and lows smoothed out. both investing in, and withdrawing from the fund, along the way.

If I had access in my pension to invest in say Vanguard's Global Balanced Fund, would I get more or less the same benefit as Colm's approach? A smoothed return.

Ideally I would stay in this fund, with my current pension provider, after I retired, but even if I had to "cash out" and setup an ARF with the same fund at that point in time, the approach would stay the same (albeit with likely higher charges for the reasons other posters have already mentioned).

Is there a distinct advantage to Colm's approach vs investing in the long term in something like that Vanguard fund?

Apologies. It is difficult to know how to pitch posts on AAM. Clearly you are quite clued in and yes you have described "lifestyling" accurately. The basic point though is that any mix involving bonds/low yielding assets/cash is expected (but not certain) to produce lower returns than 100% equities in the long term. I think Colm has described his own strategy as 100% equity but obviously without smoothing. Depending on your circumstances you may be comfortable with that and that does seem to be your approach - you seem to be ok with short term volatility. "Lifestyling" essentially recognises that many folk and especially those at whom AE is targeted would not be comfortable with that amount of volatility.
Charges are another but also very significant aspect which Colm's proposal also addresses.

 

[nest egg]

Apologies. It is difficult to know how to pitch posts on AAM. Clearly you are quite clued in and yes you have described "lifestyling" accurately. The basic point though is that any mix involving bonds/low yielding assets/cash is expected (but not certain) to produce lower returns than 100% equities in the long term. I think Colm has described his own strategy as 100% equity but obviously without smoothing. Depending on your circumstances you may be comfortable with that and that does seem to be your approach - you seem to be ok with short term volatility. "Lifestyling" essentially recognises that many folk and especially those at whom AE is targeted would not be comfortable with that amount of volatility.
Charges are another but also very significant aspect which Colm's proposal also addresses.

No worries, the only reason I'm even half-way 'clued-in' is because of the regular contributors to this forum, like your good self.

I certainly see the merit in Colm's approach (I'm following it as closely as I can) From all of my professional experience though, change is difficult. Given where we as a country are with pensions/retirement provision, I'm just wondering if we could give this future auto-enrolled population 80% of the benefits of Colm's approach, with a lower hurdle to jump?

Wouldn't it be a good start if the default fund for everyone auto-enrolled was something like the Vanguard fund? No need for lifestyling, no high charges, and ideally no hard-stop either at retirement age. That pitch is one which may be an easier sell to those in power, in the short term at least.

 

[Duke of Marmalade]

@nest egg It would take us a little off topic to critique the Vanguard Global Balanced Fund and I am not familiar with it. It's name suggests something like a 80/20 equity/bond split. Let's say in the long term equities outperform bonds by 4% p.a. (ERP) then long term this will expect to underperform a 100% equity fund by 20% x 4% = 0.8% p.a. Not disastrous for sure but still compared to the 0.5% cap on charges which was originally proposed it is significant.
Global sounds good. A well diversified global equity approach is what I think Colm is proposing.
What about volatility? Well let's say we would not be at all happy if our retirement fund fell by 20% in a year. Let's say the 100% equity fund fell by 20%, then other things equal the 80/20 fund would fall 16%. Is that much different?
Talk of an "optimal mix" sounds a bit like hubris to me, though I have seen analysis which suggests 60/40 is "optimal".. "Optimal" will be relative to a personal risk/reward appetite.

 

[Conan]

The “lifestyle “ approach was a reaction to two separate issues:
- those who wanted to reduce volatility as they approach ”drawdown” at retirement in order to ensure their retirement lump sum
- the balance of the fund being used to buy an Annuity (the price of which is determined - to a significant extent- on bond/interest rates)
But with the introduction of ARFs very few retirees from D.C. schemes have bought Annuities, probably because in a decade or more of low interest rates, Annuity rates have looked poor value. Most retirees have gone the ARF route and reinvested the residual fund (typically 75%). But in doing so, many (but not all) have tended to adopt a relatively conservative investment strategy (“I am reluctant to take much investment risk”, “ I am not sure I can live with Equity volatility for the entire ARF fund”). In my experience, very few have adopted Colm’s strategy of 100% Equities (unless they we not reliant on the ARF drawdown for income in retirement).
So for the typical retiree, they require advice on whether ARF or Annuity (and Annuity rates are now looking better as interest rates are on the rise), and if going the ARF route they require advice on what ARF, what investment strategy, what level of drawdown etc. So invariably all of this comes at a cost. Even if all of this was set at a charge of say 1% pa, it means the fund has to earn 1% pa before the client gets any return.
So apart from Colm’s “smoothing proposal”, the idea of leaving the residual 75% in the AE fund and drawing income from that fund, is likely to result in an expense saving for the client.
As someone once said “I works fine in practice, but does it work in theory”.

 

[nest egg]

@nest egg It would take us a little off topic to critique the Vanguard Global Balanced Fund and I am not familiar with it. It's name suggests something like a 80/20 equity/bond split. Let's say in the long term equities outperform bonds by 4% p.a. (ERP) then long term this will expect to underperform a 100% equity fund by 20% x 4% = 0.8% p.a. Not disastrous for sure but still compared to the 0.5% cap on charges which was originally proposed it is significant.
Global sounds good. A well diversified global equity approach is what I think Colm is proposing.
What about volatility? Well let's say we would not be at all happy if our retirement fund fell by 20% in a year. Let's say the 100% equity fund fell by 20%, then other things equal the 80/20 fund would fall 16%. Is that much different?
Talk of an "optimal mix" sounds a bit like hubris to me, though I have seen analysis which suggests 60/40 is "optimal".. "Optimal" will be relative to a personal risk/reward appetite.

Agree, no need to critique it - there are many well diversified funds on the market. It's one I recalled offhand and I mentioned it purely as an example. Your assumptions are helpful though in terms of the evaluating the right fund to select and there will be tradeoffs of course.

My counter proposal is a "hybrid" of Colm's, where everyone benefits from the following by default:

1. No cliff-edge at retirement, you stay in the fund forever (leveraging its economies of scale and simplicity)
2. No lifestyling (not necessary due to the default fund)
3. A level of smoothing still achieved, using a well diversified passive fund (this should also keep charges low)

Options can exist for the small minority who may want to do something else.

While very likely inferior to Colm's full approach, crucially it is better than what's being proposed by the Govt. From my standpoint, it also has the distinct advantage that it doesn't require any politician / civil servant to take a "brave" decision (in the Yes Minister sense of the word!) on something new.

 

[Duke of Marmalade]

Agree, no need to critique it - there are many well diversified funds on the market. It's one I recalled offhand and I mentioned it purely as an example. Your assumptions are helpful though in terms of the evaluating the right fund to select and there will be tradeoffs of course.

My counter proposal is a "hybrid" of Colm's, where everyone benefits from the following by default:

1. No cliff-edge at retirement, you stay in the fund forever (leveraging its economies of scale and simplicity)
2. No lifestyling (not necessary due to the default fund)
3. A level of smoothing still achieved, using a well diversified passive fund (this should also keep charges low)

Options can exist for the small minority who may want to do something else.

While very likely inferior to Colm's full approach, crucially it is better than what's being proposed by the Govt. From my standpoint, it also has the distinct advantage that it doesn't require any politician / civil servant to take a "brave" decision (in the Yes Minister sense of the word!) on something new.

You must have read my submission to the AE consultation It was headed "one fund for life".
I was aware of Colm's approach and said it required serious consideration, which is all Colm is asking for.

 

[nest egg]

You must have read my submission to the AE consultation It was headed "one fund for life".
I was aware of Colm's approach and said it required serious consideration, which is all Colm is asking for.

You've cheered me up Duke! The big question is, do you have anyone's ear?? I hope so.

 

[Duke of Marmalade]

The “lifestyle “ approach was a reaction to two separate issues:
- those who wanted to reduce volatility as they approach ”drawdown” at retirement in order to ensure their retirement lump sum
- the balance of the fund being used to buy an Annuity (the price of which is determined - to a significant extent- on bond/interest rates)
But with the introduction of ARFs very few retirees from D.C. schemes have bought Annuities, probably because in a decade or more of low interest rates, Annuity rates have looked poor value. Most retirees have gone the ARF route and reinvested the residual fund (typically 75%). But in doing so, many (but not all) have tended to adopt a relatively conservative investment strategy (“I am reluctant to take much investment risk”, “ I am not sure I can live with Equity volatility for the entire ARF fund”). In my experience, very few have adopted Colm’s strategy of 100% Equities (unless they we not reliant on the ARF drawdown for income in retirement).
So for the typical retiree, they require advice on whether ARF or Annuity (and Annuity rates are now looking better as interest rates are on the rise), and if going the ARF route they require advice on what ARF, what investment strategy, what level of drawdown etc. So invariably all of this comes at a cost. Even if all of this was set at a charge of say 1% pa, it means the fund has to earn 1% pa before the client gets any return.
So apart from Colm’s “smoothing proposal”, the idea of leaving the residual 75% in the AE fund and drawing income from that fund, is likely to result in an expense saving for the client.
As someone once said “I works fine in practice, but does it work in theory”.

I recently had a relative retiring with a pension target of €40k p.a. His fund was c. €1m which was about right for his target. 1% charges are €10k p.a. or 25% of target pension This is the sort of drag on income that young folk face in their mortgage. That ain't right.

 

[Conan]

I recently had a relative retiring with a pension target of €40k p.a. His fund was c. €1m which was about right for his target. 1% charges are €10k p.a. or 25% of target pension This is the sort of drag on income that young folk face in their mortgage. That ain't right.

There still will be some fund charge even if the retiree can remain in fund post retirement. But the AE proposal suggests a max annual fund charge of c0.5% . And if no advice is required, then that results in a potential further saving.
For the typical AE client (those not offered membership of an employer sponsored group scheme), the facility to remain in the fund post retirement (not have to seek/pay for advice) makes perfect sense.

 

[Duke of Marmalade]

The tax angle is off topic. I'm sorry I raised it.
@Brendan Burgess You seem to think that AE is a quasi State supplementary pension and therefore open season for the State to raid irrespective of whether it is using smoothing or otherwise. It is no more vulnerable than any other private sector arrangement (which we have seen of course have been raided) or indeed the UK NEST funds. I agree that Joe Duffy does wield powers to usurp our legal framework but I suspect Joe will have retired by the time he can have a go at AE

 

[Brendan Burgess]

I don't know what you mean by "emergency measures". I never mentioned anything of that nature in anything I ever wrote.

We had a financial crisis and the government implemented emergency measures. they taxed pension funds. the forced the mergers of ebs and aib.

You need to address how your proposal will be ring fenced from political interference and an attack from Joe Duffy.

Brendan

 

[Brendan Burgess]

You seem to think that AE is a quasi State supplementary pension and therefore open season for the State to raid irrespective of whether it is using smoothing or otherwise. It is no more vulnerable than any other private sector arrangement (which we have seen of course have been raided) or indeed the UK NEST funds.

Hi Duke

It's not really as separate as a normal defined contribution fund.

the state is fully involved in setting it up.

If we hit a black swan event, and the actual is below the smoothed value for a long time, the state will be the ultimate underwriter.

and if there is a long running surplus in the fund, a desperate government will eye it.

These are certainly concerns. Colm needs to face up to them , deal with them and show how they will not come to pass. and not just dismiss them as invalid criticisms. He will not be able to get people on board by attacking them.

Brendan

 

[NoRegretsCoyote]

There are quite a few things in @Colm Fagan 's ideas that I really agree with. Namely low fees due to bulk management good and lifestyling overrated.

Now for the problems:

The problem is that a "smoothed" value of a fund is fine in theory but impossible to implement in practice. Everyone knows that financial markets fluctuate and are prone to prolonged optimism and pessimism. The problem is that where we are in the cycle is only clear in a decade's time and you have to make a drawdown tomorrow. I've been in a job where making policy depended on estimates of where we were the financial cycle and trust me it was really hard. Endless discussion about models and an impossibility of explaining them to people who made decisions.

The second problem is that Colm's proposal seems to depend on pooling of current and future pensioners' funds. This is just a bad idea as current pensioners will be better organised and incentivised to call on the funds where needed than future pensioners. There is also a property rights argument. This is in fact the individual's money, and if it's their money they have to be given discretion over its use, even if they make bad decisions. As for fees, I think decisions in and around retirement can be pretty difficult depending on financial circumstances. There are a lot of decisions to take, including many potential wrong ones. I had a long discussion with relative last year in his 60s about how and when he plans to set up his ARF to convert what is a seven-figure fund. This relative is one of the smartest people I know but he is still taking professional advice and while I am sure there is some goldplating my feeling is that this is helping him to make the right decisions for his circumstances.

Getting back to a "smoothed return", the only way to implemented this is some kind of sovereign guarantee to make up the gap. But Ireland's fiscal capacity is not decoupled from the global financial cycle. And the whole point is to reduce the call on future taxes from pensions so a guarantee would just leave us where we are starting from. Sadly I think the problems as outlined above make the "smoothed return" idea a non-runner. Otherwise the other ideas are worth considering and everyone should advocate them.

 

[Colm Fagan]

@Brendan Burgess
A number of different arguments are in danger of being conflated, and possibly confused. Neither of us want that.
I'll try to take them one by one. May not have the time to deal with all of them now, but I'll do my best.
So, what you mean by "emergency measures" is the government stealing people's money. Stealing is stealing, no matter who does it and where they take it from. Are you saying that it's different to steal it from an AE account than, say, from someone's post office savings account? A variation on that is that you're saying it might be easier to steal it from an AE account where the returns are smoothed, particularly if the market value exceeds the smoothed value, presumably on the argument that people won't know the money is missing. What I'm trying to make clear is that everyone will know the smoothed value of their account at any time. The market value can also be easily obtained. If money is taken from the account, both the market value and the smoothed value will fall.
I think you're confused by memories of what happened with the NPRF. There, the government just took the money (I'm sure they described it differently). However, that money wasn't allocated to individuals. A senior Civil Servant who was closely involved with the NPRF told me once that the government would never have dared taking the money if it had been allocated to individuals. Then it would be clear that it was theft.
In future, if you want to mention "emergency measures" as a threat to an AE scheme (whether returns are smoothed or not), it might be best to make your meaning clear by calling it theft. Don't mince your words.
As to the risk of government being on the hook if smoothed value exceeds market value, there is no risk. The smoothing formula is constantly pulling smoothed value towards market value (not vice versa, as some people claim that I'm saying). Therefore, the two will always return to balance. That balance will be restored quickly when cash flows are strongly positive. The example in the paper shows them returning to balance within two or three months even after a massive 15% fall in March 2020, assuming a scheme start date of 1 January 2020. The return to balance will be slower when cash flows are not so strongly positive, or are negative (expected to happen sometime after year 30), but balance will always be restored. Therefore, there is no risk of government being on the hook.
There is the problem, of course, that circumstances would arise where smoothed returns will be lower than market returns for a prolonged period, possibly extending for more than a year or two. One key objection, which I recognise as valid, is that this could cause members to lose faith in the scheme. I don't think it will, partly for the obvious reason that members of the smoothed scheme will be doing far better in such circumstances than members in schemes where returns are not smoothed, and the iron law that returns from risky assets will trump returns from "safe" assets in the long-term will deliver salvation eventually. The paper even shows that to be true for Japan post 1990. After 30 years, the Japanese stock market (with dividends reinvested) was 18% below its level 30 years previously, yet an AE contributor would have got an average positive return of around 2% a year over the period. That's far more than they would have got from Japanese bank accounts. Why? Because their funds were being invested in the market all the way through, when markets fell to the depths and then when they started to recover. The Japanese experience post 1990 is about as bas as it could ever get.
I'll take another look at all your arguments and come back if there are more that I haven't answered.

 

[NoRegretsCoyote]

The smoothing formula is constantly pulling smoothed value towards market value (not vice versa, as some people claim that I'm saying)

Of course, that's how smoothing formulae are designed to work! But these formulae are totally dependent on parameters estimated from long-term performance. These parameters are updated over time and if they are not regularly then something is wrong!

So you are introducing three concepts to someone about to retire: market value of the fund, "smoothed" value of the fund, and risk that smoothing parameters will be adjusted (possibly against them) in the future. This is conceptually very heavy to understand never mind to make policy on the basis of.

You could get ten different Nobel Prize-winning economists to estimate smoothing parameters and they will give you ten different models. Which one would you use? None of this is to deny that markets are irrational in both ways, for prolonged periods, and this is often to the detriment of people trying to plan a retirement.

I think your arguments would be much more acceptable (and indeed less complex) if you ditched all this stuff about smoothing and just said that people should be much more reliant on equities over fixed income products. People have decades-long investment horizons (even in retirement) and equities almost always return more over that kind of time horizon.

 

[Duke of Marmalade]

These are certainly concerns. Colm needs to face up to them , deal with them and show how they will not come to pass. and not just dismiss them as invalid criticisms. He will not be able to get people on board by attacking them.

Brendan

Boss that is an unwarranted swipe at Colm IMHO. He could have covered his proposal with one chapter. Instead he produced a 60 page paper for the Society of Actuaries, mostly devoted to implicitly or explicitly addressing the twin (valid) concerns arising if and when smoothed values exceed market values - (1) the fair sustainability of future contributions and (2) the possibility of an implicit State guarantee.
It is true that he did not substantially address the opposite scenario of where market values exceed smoothed values and the possibility that Joe Duffy will whip up a baying mob demanding their "surplus" nor that government will help themselves by stealing the surplus. Nor did he address the possibility of a Madoff type gal getting the key post of head of the CPA.
Reminds me of the presenter of a fear of flying course I attended, screaming at his audience -"if you ask enough questions, make no mistake you will crash this plane".

 

[Towger]

The government stole from individual's pension funds, during the recession. Not just the NPRF. It is no difference from stealing from current accounts, except it did not result in the masses demonstrating in the streets.

There are other issues in the thread, such as PRSAs having PRSI relief. The only pension type with any type of PRSI relief on contributions are ASCs. I will probably get attacked for calling ASC a pension.