Auto Enrolment should be to a State Sponsored Defined Benefit Scheme

[Duke of Marmalade]

Attached is a summary of the Congress of Trade Unions' response to the Strawman. It has the usual sweeteners for its members e.g. employers and the government should pay a bigger share of the overall contribution. But it has one big suggestion. The drawdown should be in the form of a state earnings related pension to supplement the state basic pension i.e. it should be DB.

The Department in their roll out of the Strawman made it clear that the decision had already been taken to go DC rather than DB. And all submissions slavishly followed that warning off any reference to DB, except Congress. Makes me suspect that the whole consultation process is bogus. Why was the biggest decision of all not subject to consultation?

The DC approach to retirement provision is so terribly flawed from both the individual and society perspective. Colm has made a brave and innovative effort to correct the deficiencies in the DC approach as described in the Strawman but just see below how a DB approach cuts immediately to the chase on all those deficiencies.
I am talking about a properly funded scheme with benefits and contributions determined on actuarial principles but, being state sponsored, not subject to the short term MFS constraints etc. Some of the advantages would be:
1) The assets backing the fund would be real assets, correct for society as a whole and not compromised by the individualisation of risk associated with DC. DC is sub optimal in the aggregate. The Norway Pension Fund is the obvious model with its 70% real asset allocation.
2) The drawdown phase is taken care of through community and generation pooling of investment and longevity risk. The drawdown phase is even more important than the accumulation phase. Woodenman continues to put this on the long finger. I suspect that it will never be resolved and we will settle for the politically safe UK NEST approach.
3) As a state sponsored single vehicle costs would be minimised.
4) There would be no spurious and costly chasing of new members by a range of commercial providers. (The state would outsource admin and investment to commercial agents but the end consumer would be remote from that interface.)
5) There would be no need for costly professional financial advice, which is especially problematic in the drawdown phase.
6) We could cancel our order of a carousel from Las Vegas. Joking aside, we will avoid the nonsense of individuals having dramatically different approaches to pension provision, decided by lottery.
7) It's been done before. Whilst Colm's suggestion goes some way to giving the above advantages in a DC approach, it does suffer politically from its "hasn't been done before" tag.

Oh, I am sure there are downsides, which I would welcome to hear of and try and address in this thread.

 

[Protocol]

Ok, interesting.

The ICTU proposal is sort of like having a pre-funded State Pension?

That document doesn't mention the benefit rates, so they don't seem to be defined in advance?

It would be like having a giant SI pension fund, but who decides the benefit formula?

 

[Protocol]

Sorry, I see now that the benefit is an annuity, a top-up to the SPC, ok.

The size of the annuity to depend on the value of conts, ok.

 

[Duke of Marmalade]

Sorry, I see now that the benefit is an annuity, a top-up to the SPC, ok.

The size of the annuity to depend on the value of conts, ok.

Yes, but crucially the annuity would be backed by real assets and there would be no need for stiff regulatory capital. In the current environment that could at least double the affordable annuity rate. Commercial interests might cry foul but they don't do that for the basic state pension and they don't really serve this constituency very well - too expensive.

 

[Brendan Burgess]

Hi Duke

Who would set the annuity rate? Would there not be huge pressure to pay too high a rate as the cost would be dumped on future generations.

What happens if the annuity rate turns out in retrospect to have been too high?

I am not sure that I would like to be contributing to a fund with an actuarial deficit in it.

 

[Peanuts20]

Isn't there a point of principle here that many (most?) DB schemes are in financial difficulty. What would happen in a state DB scheme if the funds were not there?. Would the taxpayer have to bail it out?

 

[Duke of Marmalade]

Hi Duke

Who would set the annuity rate? Would there not be huge pressure to pay too high a rate as the cost would be dumped on future generations.

What happens if the annuity rate turns out in retrospect to have been too high?

I am not sure that I would like to be contributing to a fund with an actuarial deficit in it.

Valid questions but I think you are seeing this through the lens of a typical commercial provider. A state provider doesn't have to worry over much about having its sums precisely right. Admittedly that gives rise to political moral hazard but as I say this is not a new proposal. The UK had a SERPS system. I don't know whether that is still the case and how it dovetails with NEST.

 

[Duke of Marmalade]

Isn't there a point of principle here that many (most?) DB schemes are in financial difficulty. What would happen in a state DB scheme if the funds were not there?. Would the taxpayer have to bail it out?

DB schemes are in financial difficulty because they value their liabilities at extremely low bond rates. If we believe in the ERP, much higher discount rates can be justified. This requires an investment policy consisting of real assets. We have found that the commercial sector for many reasons cannot operate this paradigm. But the demise of DB is a fundamental societal failure. The State should step in to salvage the concept. A world where the third of the population who are in retirement are living on their nerves following the Eurostoxx 50 index is surely a step backwards for society.

 

[Brendan Burgess]

Admittedly that gives rise to political moral hazard but as I say this is not a new proposal.

But we Irish could not do that.

In the General Election campaign one party would propose to lower the contribution rate for the lower paid and raise it for the higher paid.

The DC system is the only one which protects against this.

Could you have DC up until retirement and then some sort of guarantee on the annuity bit afterwards? Or a with-profits annuity?

Brendan

 

[Duke of Marmalade]

Could you have DC up until retirement and then some sort of guarantee on the annuity bit afterwards? Or a with-profits annuity?

Brendan

Yes that is an interesting possibility. The fact is that the annuity market is to all intents dysfunctional. One could envisage the State providing annuities priced using the ERP and without having to service capital and of course at low cost.
Raises interesting issues around whether lifestyling would then be appropriate.
Indeed, on thinking about this, I can't see them solving the decumulation phase in a DC context and they might be coaxed into providing state annuities.

 

[Brendan Burgess]

But a straightforward DB scheme would not be possible in Ireland?

There would be so much messing that it would just become another huge government guaranteed deficit.

Brendan

 

[NoRegretsCoyote]

But we Irish could not do that.

In the General Election campaign one party would propose to lower the contribution rate for the lower paid and raise it for the higher paid.

The DC system is the only one which protects against this.

It's not as simple as that.

The current proposals are designed to insulate the state from any financial risk whatsoever. Let's be clear about this. There is also the property rights argument hovering in the background. If you are compelling people to part with their income then you have to give them some discretion over how it is invested.

@Brendan Burgess is probably right that a badly-run state fund would be vulnerable to bad management the way it is in somewhere like Italy. Politicians could meddle with various paramaters to turn it into something like a PAYG scheme with promises to boost payouts before elections.

I think this could be fixed with the right governance arrangements. You could have a paid committee of experts with long terms who would be impossible to fire. They would make public recommendations to government about the actuarial factors to use. It could be set up so that only active disagreement by the Government could stop their recommendations being implemented.

This wouldn't stop interference, but it would raise the political cost considerably.

 

[Steven Barrett]

But we Irish could not do that.

In the General Election campaign one party would propose to lower the contribution rate for the lower paid and raise it for the higher paid.

The DC system is the only one which protects against this.

Could you have DC up until retirement and then some sort of guarantee on the annuity bit afterwards? Or a with-profits annuity?

Brendan

Agree 100%. It would be politicised. Ireland is full of people who want services but don't want to pay for them. People will want to have a great DB pension but will want to pay as little as possible for it. And if politicians think it will win them votes, they'll tell people this is nothing but a tax and they are going to reduce it.

Also, the government have done nothing to solve the funding crises that is coming down the track for public servants pensions. Do you think they should be given the responsibility of managing the private sectors pension fund too?

The auto enrolment is going to be a disaster for a number of years. Jim is going to retire 3 years after starting in the scheme. As he's close to retirement, he stuck his money in bonds as they are "safe". The €1,800 he put in is only worth €1,500. He's now on Joe Duffy complaining about broken promises about having a pension to retire on and it is never being pointed out to him that only €1,800 has been paid in, half of which came from his employer.

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

 

[Sunny]

We don't have to follow Italy though. Look at what Norway have done with NBIM who manage their Pension Fund. Had a few dealings with them in the past and they were probably one of the most professional organisations I have met. The NTMA to their credit also has a very good reputation. I admit the challenge is getting the right governance and legislation in place at the beginning but that is hardly insurmountable.

 

[Peanuts20]

But regardless of who runs a DB scheme, fundamentally it needs to make a return on its investment to be able to pay out. That investment will either be in the stock market, the bond market or physical (real) assets which pay a return. And, as the ads always say, the value of your investment may rise or fall. Hence, regardless of whether it is DC or DB, there is always going to be an element of risk here and anything else is smoke and mirrors. Show me a bank manager who never made a loss and it'll be a manager who never lent a penny.

 

[PMU]

@Brendan Burgess is probably right that a badly-run state fund would be vulnerable to bad management the way it is in somewhere like Italy. Politicians could meddle with various paramaters to turn it into something like a PAYG scheme with promises to boost payouts before elections.

That's very unfair on Italy. It's not that the pension fund would be badly managed but that politicians would or most likely will raid it for short term electoral gain. It will almost certainly happen. For example, the National Pensions Reserve Fund ended up with two portfolios: (a) the discretionary portfolio, i.e. invested in multiple assets and controlled by the NPRF Commission and the directed portfolio, i.e. funds allocated under the direction of the Minister of Finance. There were always calls from TDs that NPRF funds be allocated to to their pet projects. It's difficult to see something similar not happening again.

 

[NoRegretsCoyote]

Also, the government have done nothing to solve the funding crises that is coming down the track for public servants pensions.

That's simply not true. The 2013 Single Scheme for new entrants will have benefits in the hundreds of millions per year by mid-century compared to the previous policy.

It's not that the pension fund would be badly managed but that politicians would or most likely will raid it for short term electoral gain. It will almost certainly happen.

This can be mitigated at design stage with the right governance set-up to keep it as far away from politicians as possible.

 

[Steven Barrett]

That's simply not true. The 2013 Single Scheme for new entrants will have benefits in the hundreds of millions per year by mid-century compared to the previous policy.

I should have stated that they haven't started saving for the funding crises that is coming down the track

 

[Colm Fagan]

Some thoughts on the Duke's suggestion. I am naturally comparing it with my "smoothed DC fund" proposal.

The smoothed fund as proposed has all but one of the advantages outlined by the Duke at the start:

1) The assets backing the fund would be real assets, correct for society as a whole and not compromised by the individualisation of risk associated with DC. DC is sub optimal in the aggregate. The Norway Pension Fund is the obvious model with its 70% real asset allocation.
2) The drawdown phase is taken care of through community and generation pooling of investment and longevity risk. The drawdown phase is even more important than the accumulation phase. Woodenman continues to put this on the long finger. I suspect that it will never be resolved and we will settle for the politically safe UK NEST approach.
3) As a state sponsored single vehicle costs would be minimised.
4) There would be no spurious and costly chasing of new members by a range of commercial providers. (The state would outsource admin and investment to commercial agents but the end consumer would be remote from that interface.)
5) There would be no need for costly professional financial advice, which is especially problematic in the drawdown phase.
6) We could cancel our order of a carousel from Las Vegas. Joking aside, we will avoid the nonsense of individuals having dramatically different approaches to pension provision, decided by lottery.
7) It's been done before. Whilst Colm's suggestion goes some way to giving the above advantages in a DC approach, it does suffer politically from its "hasn't been done before" tag.

1. The fund is invested completely in real assets.
2. There is pooling of investment risk and (optional) pooling of longevity risk in the drawdown phase. I'll talk more about longevity risk below.
3. Low costs are achieved by the simplicity of the design: a deposit account with the same 'interest rate' for everyone, a fund governed by trustees (which can be considered as state sponsored).
4. No spurious and costly chasing of new members. Commercial operators can continue to offer their wares, their main sales pitch being tax relief at the marginal rate (v an effective 25% for the AE scheme), wide range of investment options pre- and post-retirement, and more flexible drawdown options than AE.
5. No need for costly advice, which is especially problematic in the drawdown phase. Investment choice at various life stages is the main reason for people to seek advice. There is no investment choice with the proposed smoothed fund.
6. No Las Vegas carousel here either!
7. You have me on that one!! Yes, SERPS has been done before in the UK, but can't Ireland be first for something worldwide sometime????

There is far greater transparency with my proposed approach. @NoRegretsCoyote makes a very important point:

There is also the property rights argument hovering in the background. If you are compelling people to part with their income then you have to give them some discretion over how it is invested.

My proposal doesn't allow any discretion over how a person's money is invested, but I see it like a deposit in a credit union. I don't know what the credit union does with my money and I don't care. All I really care about is the interest rate. I can always see how much principal I have and how much interest has accrued to date. It's exactly the same with my proposal. My understanding is that this transparency is missing with Duke's proposal. I am probably told that I have accrued €X of annual pension when I have contributed €Y, but this raises all sorts of questions over what happens if I die after just getting a pension for two years, what happens if I retire early, etc. I just can't put a value on what I've earned with my contributions to date without an actuary at my elbow, and even then its value depends on how long I live, etc.
Pooling of longevity risk implies a transfer of wealth from the poorer to the richer. Rich people live longer, so pooling of longevity risk gives the rich a bigger slice of the pie.
My proposed approach effectively eliminates this as a problem. People can see their account build up while they're working. They can also see it running down when they retire. They know that, if they die at any time, the balance in the account will be paid to their estate. I'm sure that people will see this as eminently fair.
I have proposed some (optional) sharing of longevity risk, but it's limited and much less likely to discriminate against the less well-off than Duke's annuity proposal. There is no pooling of longevity risk before age 75. After 75, it takes the form of a small reduction in the annual interest rate, which is used to fund annual payments to people who live beyond age 90.

 

[Duke of Marmalade]

I have proposed some (optional) sharing of longevity risk, but it's limited and much less likely to discriminate against the less well-off than Duke's annuity proposal.

Those are good points Colm but I am fascinated by this leftie twist to the argument. I am warming to the idea of state annuities to address the drawdown phase. I had in mind a single or, at worst, age related annuity rates. So in reality this would involve a range of cross subsidies:
Poor to Rich
Male to Female
Unhealthy to healthy
Smokers to non-smokers (this would be popular - aside: bashing smokers is de rigeur these days, all other minorities enjoy PC protection:mad
Single life to joint life (I haven't teased this aspect through)
etc.
But isn't the same true of the basic state pension?