My proposal for one fund with a smoothed return

Colm has published a new paper on this topic which you can download here

And he will be presenting it in a webinar as follows:

Date Wednesday, 20th January, 2021
Time Start time: 12.30 pm End time: 2.00 pm

Description
This event will be run as a live webinar using Zoom. Click here for information on using Zoom.

The paper can now be accessed here

The paper proposes a new approach to auto-enrolment (AE), designed to deliver higher pensions than under the previous government’s AE proposals, at half the cost: total contribution 7% (3% employee and employer, 1% state) compared with 14% (6% employee and employer, 2% state) under the previous government’s proposals. The higher benefits come from investing in equity-type assets with smoothing of investment returns, combined with a seamless and low-cost transition from pre-retirement to post-retirement.

To register for this event, please proceed with 'Make a reservation' below and we will send you details of your ZOOM invite shortly.
If you are unavailable to attend this webinar, it will be recorded and then uploaded to actuview afterwards. You will find information on using actuview on our website here and this includes a Frequently Asked Questions section, which you may find useful.


Colm Fagan
Biographical details
Colm is a past President of the Society of Actuaries in Ireland. He presented his first actuarial paper 43 years ago. He promises that this will be his last. His first was acclaimed as “remarkable” 24 years later, in a history of financial economics published in the British Actuarial Journal (Volume 8, Part I, p38). If history repeats itself, Colm is unlikely to be around to read the profession’s judgement on this one!
 
Hi Colm

So far, I have read the first 30 pages.

I fully agree that people should remain in equities all their life and that lifestyling is a terrible idea.
In particular people should have their DC fund in equities during retirement and not in cash as many (most?) people have at present.
I fully agree that there should be some form of smoothing of returns
I really like the idea of a mutual managing our pension funds

I strongly recommend that you focus on getting those principles accepted and focus less on the complicated smoothing formula.

I have a lot of respect for the Minister for Social Welfare, Heather Humphries. She is far better than her predecessor and probably would engage with the above ideas. If you could convince her to adopt them, then you could move onto the next stage.
 
While I support the idea of smoothing, I don't like your approach.

It's hard to understand.

Most people would be suspicious of it. That it's some form of clever financial engineering which we won't know if it will work until we have had about 50 years experience of it. It might be too late then to discover the flaw.

It's terribly inflexible. To prevent people gaming the system, they have to make scheduled contributions and scheduled withdrawals. Life doesn't work out according to a schedule. I might want to take a break from contributing. I will want my tax-free 25% on retirement, but I might also want to take it early, take it late or take more than 25% from my ARF. You are limiting me to 8% a year.

I think that a much simpler form of smoothing would be easier to understand and much more flexible. This is how I would explain it to potential members of the scheme

"The stock market is very volatile but gives a good return over the long term.
Our target is to give you a return of 3% a year over the long term.
In the years when the returns are in excess of 3%, we will hold them back to bring the returns in the bad years up to 3%.
The big advantage for you will be in retirement. Firstly, you should have a much higher fund with which to retire. And secondly, and more importantly, because of our smoothing, you will be able to stay in the fund during your retirement which should combine low risk and high return."

If the fund "declares" 3% a year in the early years but there is a sustained stock market fall, the fund would be in deficit but it should recover in time. It might be necessary to lower the initial figure from 3%.

There could be a rolling lock-in of returns every 10 years. For example, whatever I contribute in year 1 would have an interim return of 3% a year. After 10 years, the return could be "capitalised" so I would keep it, assuming that the fund allowed it.

We have a lot of experience in Ireland of building up reserves for which the original contributors benefit very little
The Credit Unions have huge reserves built up from not paying very high rates of interest to shareholders.
The Building Societies also built up reserves.
The mutual life companies did so as well and then became Plcs and distributed the reserves to the members.

So if my investment of €5,000 in year one gets a market return of 100% over ten years, but you lock in only 40%, I won't mind.

Brendan
 
Last edited:
Brendan,
I think you are describing what I used to recognise as a With Profit fund.
I think we all agree on the overall objective - to allow members invest in equities for life whilst pooling the investment timing risk.
There are pros and cons to both your and Colm's approaches.
You are right to point out the history of discretionary pay out systems - they tended to hold a lot back over time. Colm's approach takes away discretion in pay-outs and over time distributes full market value.
I agree that the formula is complicated but I do not think Colm envisages punters getting into the weeds, in fact he points out the simplicity of the approach at the point of delivery - like a bank account but with interest payments backed by equity returns.
 
@Brendan Burgess Thanks for your comments. As per my private email exchange with you, I'll leave replying for the time being, but I'll make sure to take your views into account when preparing my presentation for the webinar on the 20th.
I'll just make the simple point that what I'm proposing may look like alchemy to some people - delivering equity-type returns with volatility lower than a bond portfolio - but there is a cost, in the form of loss of flexibility. I would argue that the amount of flexibility lost is less than that lost by members of DB schemes. They pay contributions at a specified level for their entire membership, take a gratuity at retirement and have a fixed pension for life (or one increasing by inflation, if they're lucky). No-one complains about their loss of flexibility.
 
@Colm Fagan you make a very good point on the rigidity of DB schemes. DB is the ultimate in pooling of investment timing risk and longevity risk. The demise of DB is a big societal loss. Personally I would have the AE being DB but this is politically unacceptable because of a perceived state guarantee. Your proposal goes some way to replicating DB in a DC setting. It must therefore bear some of the DB rigidity.
 
Hi Colm

What I am trying to do is make suggestions as to how your proposal might gain political acceptance. I think it's a great idea and should be accepted, but I was at your last presentation to the Society of Actuaries and there was general scepticism. I don't know if that has reduced since then or not but if you can't convince your colleagues, then you have little chance with the Minister. I think you have a great opportunity with Heather Humphries at the moment. In due course, she will be replaced, and it's unlikely that the next Minister would be as open to new ideas.

A lot of people bemoan the loss of Defined Benefit schemes. If you can portray your scheme as a Defined Benefit scheme without the cost of the guarantee, you could well get acceptance.

I think you should address the lack of flexibility. On reflection, the fact that I can't make AVCs to this is not really a disadvantage. I can make AVCs to a separate pension scheme.

I do think that not being able to take more than the scheduled withdrawal rate is a disadvantage. Could you allow people take out more than the scheduled amount, but at the market value or the smoothed value whichever is lower?

Come to think of it, you could allow people to make AVCs at the market value or smoothed value, whichever is higher.

Brendan
 
No one wanted to hear about bogle's crazy tracking ideas back in the day....

Colm, if you can't get government acceptance to use it, what about setting it up a co-op entity and selling it as a product?

After 5 or 10 or 20 or 50 years of proven returns the idea might seem more palatable/main stream.

I and I suspect several others here and elsewhere would be willing to help/get involved to ensure better financial independence for many in our country.

Disclaimer: haven't read your recent paper, but i liked the idea since you first suggested it.
 
I am starting on the balance of the paper. This graph is striking.

5222


For about a third of the time, the smoothed value will be well above the market value.

So, in month 157, I will be paying €145.60 for €100 worth of market value.

For about 70% of the time, I will be paying less than the market value.

It would be very hard to attract new entrants during the periods when it's so far underwater.

I think you will need a reserve to smooth out the smoothed value even further.

Brendan
 
And that was a model based on favourable returns.

This is based on a repeat of the Japanese stock market.

5223


It would be hard to see the fund maintaining public support in this scenario.

Brendan
 
SPC100 said:
if you can't get government acceptance to use it, what about setting it up a co-op entity and selling it as a product?
Click to expand...

The problem is that the fund would be insolvent for a good part of its time and that is part of the design.

A Life Company could offer the product, but they would have to set aside huge reserves.

Brendan
 
@Brendan Burgess
I'm reminded of the saying: "To every complex problem there's a simple and neat solution. It's invariably wrong."

We are dealing with a very complex problem here. That's why I've written 65 pages.

Take for your example your comment above about month 157, where you'd be paying €145.60 for €100 worth of market value (leaving aside the question of what 'worth' means).

Let's actually see what I wrote (Sections 7.16 to 7.19) rather than picking and choosing what you want to quote:
  • Such concerns are overblown. Take for example the situation in month 157 of the simulation, when smoothed value equates to 145.6% of market value. At that point in the simulation, it is reasonable to ask: “Will members be prepared to keep contributing, knowing they are paying so much over the odds to buy into the fund?”
  • Counterintuitively, the answer is almost certainly “Yes”. A contributor at that date is looking at a smoothed return (before charges – to be discussed in Section 10) of 5.4% over the previous two years. At the same date, a different investor, one fully exposed to the vagaries of the market, is looking at a negative return of minus 39.2% (again before charges) over the same two-year period. The contributor to the smoothed fund is thanking their lucky stars that they did not opt for a market-based product and, while they may have some concerns that the smoothed return will be held back in future by the need to bring the 145.7% back to 100% at some future date, the strong likelihood is that they will be more than happy to stay the course, in the belief that the return to 100% will not prove too painful.
  • As it happens, in this scenario the 145.6% ratio of smoothed value to market value is back below 110% in less than a year and is below 100% in two years. The smoothed return is 0.9% in the next 12 months and 3.7% in the following 12 months (again, before charges), so the experience of getting the ratio back to 100% has not proven particularly painful.
  • The graph in Figure 16 shows a second peak in the ratio of smoothed value to market value at month 230 (a repeat of February 2009, which marked the nadir of the Global Financial Crisis, at least as it manifested itself in UK share prices). In this case, the smoothed return over the previous 12 months is +2.7% while the market return is a negative 33%, so once again the contributor whose return is smoothed is pleased with their decision. In this case, the ratio of smoothed value to market value is back below 100% in just 9 months (a repeat of November 2009).
I hope you agree that the additional quotes from what I actually wrote give a different perspective.
 
5224


So with half the level of contributions, a person can get the same amount of pension over their life.

You should use this to get government buy-in. It will cost the tax payer less to get the same benefits.

That will be a big motivator for the government in the coming years. The cost of everything.

I would go further and say that the government's contribution could be earmarked for the Reserve Fund to make sure that the smoothed value always exceed the market value.

Brendan
 
Colm Fagan said:
I hope you agree that the additional quotes from what I actually wrote give a different perspective.
Click to expand...

Colm, I certainly was not picking and choosing.

I think you have a great idea which has little chance of succeeding the way you are presenting it.


Brendan
 
Brendan Burgess said:
I would go further and say that the government's contribution could be earmarked for the Reserve Fund to make sure that the smoothed value always exceed the market value.
Click to expand...
Brendan, you're sanctifying market values. There's no need to. We can forget about them for much of the time. That's part of the reason I headed one of the sections "Market values must serve investors, not be their masters" You're letting them be your masters.
 
SPC100 said:
Colm, if you can't get government acceptance to use it, what about setting it up a co-op entity and selling it as a product?

After 5 or 10 or 20 or 50 years of proven returns the idea might seem more palatable/main stream.
Click to expand...
One of the problems is that I don't have that sort of time to see this through. I may even be gone in ten years, not to mind 20 or 50! A more significant challenge however is the requirement to get Central Bank and EIOPA (European Insurance of Occupational Pensions Authority) clearance. I'm confident that can be done.
In relation to getting approval from these two bodies, Brendan isn't helping matters by making misleading comments, which I hope are unintentional:
Brendan Burgess said:
The problem is that the fund would be insolvent for a good part of its time and that is part of the design.
Click to expand...
Insolvency is being unable to meet your liabilities as they fall due. I have spent ages - running Monte Carlo simulations until I'm blue in the face (or more precisely asking Brian Woods to run such simulations) - to show that the scheme will be able to meet its liabilities as they fall due in all reasonably plausible future scenarios. Brendan, please be careful how you word what you're writing. In particular, be careful about words like "insolvent". I put a lot of effort into thinking through all the complications. Please do the same.
 
Brendan Burgess said:
The problem is that the fund would be insolvent for a good part of its time and that is part of the design.
Click to expand...
No! No! Define insolvent. Insolvent yes if members could withdraw all their funds at any time and smoothed values exceeded market value, that is a tautology. But that is exactly why members would not be allowed to withdraw all their funds at notice.
The paper actually addresses the solvency aspect at some length. It shows that in 2000 simulations covering 60 years, some producing highly implausible and hitherto unseen prolonged market collapses, it does not become insolvent in the crude sense of running out of cash. The paper points to 2 simulated scenarios where at the end of 60 years the possibility of future insolvency does look material. But by then the fund will have accumulated capital. The paper claims that the proposed formula would be safer than the Solvency II standard for insurance companies.
 
Last edited:
I've spent some time over the last couple of days digesting Colm's paper, which is certainly very comprehensive and thought-provoking.

However, I'm still a sceptic.

In Colm's own words -

"...if smoothed values are above market values for an extended period when cash flows are negative, the obligation to pay more than can be realised in the market to members leaving exacerbates the shortfall and could eventually draw the scheme into an insolvency vortex."

To me, that makes the entire scheme a non-starter.

Sure, a cash reserve can be carved out of fees (which implies to me that the fees are too high to begin with) but how do you figure out when it is appropriate to release (or partially release) that reserve?
 
Duke of Marmalade said:
No! No! Define insolvent.
Click to expand...

Hi Duke

Maybe "insolvent" is the wrong word. But I consider something to be insolvent when its liabilities exceed its assets. You would probably use the word "underfunded" for this. For about 30% of the time, the liabilities exceed the assets.

Brendan
 
You must log in or register to reply here.
Back
Top