Historic safe withdrawal rate (SWR) for Ireland

[SPC100]

Investment Management & Financial Planning News - Timeline

News and updates for Financial Advisers on Timeline's Planning and Portfolios as well as our research an updates on financial markets.

www.timelineapp.co

Rare to see Irish case being studied like this!

It tries to answer the question: How much can you sustainably take from your pension and not see it reach zero within 30 years?

It would appear that 4% of your initial portfolio is a bit agressive and 3p.c would be better rule of thumb. This is based on historical return data.

This framework is taking a percent of your initial portfolio as your first year withdrawal, and then adjusts that number by inflation for every subsequent year.

I don't understand why Ireland has such enforced high minimum drawdowns. It nearly guarantees portfolio will reach zero before 30 years are up.

edit: as per my further posts below, the state is using a different model. so it is not directly comparable

 

[jpd]

It is designed so that pension funds are not used to pass wealth on to the next generation. By enforcing a high (?) minimum drawdown, the wealth will circulate in the economy now and not be saved and passed on to the next generation.

That is the theory, at least - although I am not 100% convinced that it has the desired effect

 

[SPC100]

The eforced taxation of minimum drawdown will 'catch' some tax from high net worth individuals.

FWIW, I think the taxation is enforced but drawdown is optional.

The design doesn't appear to make sense for most of the population, who would want to be fairly sure their pension pot would last 30 years. I guess most of them will at least drawdown the minimum percent - "Hey. it's the minimum." - as this is effectively a behavioural nudge (see Thaler).

IMO, The government is reckless in effectively encouraging that level of drawdown for everyone.

Edit: as per later posts. I no longer consider this reckless. The states withdrawal model is different.

 

[AAAContributor]

Thanks for the article. On the drawdown issue, the State has to whet it's beak eventually! It'd be interesting to see figures for higher-rate taxpayers to see the amount of tax saved on contributions during the accumulation stage versus tax paid on drawdown and/or inheritance tax paid on inherited ARFs. There's probably a nice subsidy there from the State in the aggregate.

 

[NoRegretsCoyote]

Don't forget the contributory state pension which most private pension holders will also receive.

This serves as free insurance against your fund going to zero and should allow for a higher risk appetite.

All the evidence suggests that you spend a lot less when you make it to your eighties or nineties too.

 

[LDFerguson]

Bear in mind that the ARF was designed to offer a choice to retirees who, prior to its introduction, had only an annuity to provide their pension. If I run an annuity quote for a couple, both aged 65, with 100% pension continuing for the life of the spouse (given that 100% of an ARF would also pass to the spouse) the annuity rate is marginally over 3%. So, if someone doesn't want to go down the ARF route, they can stick to the older alternative - the annuity. They'll get a 3% withdrawal, but the entire capital is gone from Day One. But they're guaranteed to get the income for life. You pays your money, you takes your choice.

 

[SPC100]

It would still benefit, if 3% was the minimum drawdown, and it would not be encouraging citizens to risk running their fund to 0.

From an irish life doc - here is the minimum withdrawals

From the year you turn 61, tax is payable on a minimum withdrawal on the 30 November each year of 4% of the value of the fund at that date
From the year you turn 71 the minimum withdrawal is increased to 5%
Where the fund value is greater than €2 million the minimum withdrawal will be 6%


Remember that most will have taken 25% tax free, so 4% is effectively 5% of the original pension.

Note this is taking the percent of the current value, as opposed to what is detailed in the OP, i.e. percent of the original value and increasing by inflation.

I accept 4% of each years value will never actually reach 0, but that doesn't change the thrust of my point.

re: state pension, if anything, in my mind, this should reduce the minimum withdrawal the state encourages.


Maybe I am being unfair, and maybe they have run their models using their minimums and maybe it gives reasonable incomes in most cases over a thirty year period.


edit: added solme amrf details


Approved Minimum Retirement Fund (AMRF) holders can make only one withdrawal up to 4% of the value of the AMRF asset value at the date of withdrawal
An AMRF becomes an ARF when you reach 75 or on earlier death or should your circumstances change and you are are in receipt of a guaranteed income of €12,700.

 

[SPC100]

Liam, thanks for indicative quote, does that annuity payout change e.g. can it go up or down due to inflation? or is fixed?

 

[RedOnion]

I don't understand why Ireland has such enforced high minimum drawdowns. It nearly guarantees portfolio will reach zero before 30 years are up.

It's 4% of the balance each year. Not 4% of the starting value. If you withdraw 4% of the balance, it will never go to zero. Or am I missing something?

 

[SPC100]

Yes, i pointed that out above, it is a different model. I have also edited my op to make that clearer.

I guess the question becomes which model is better.

 

[LDFerguson]

Liam, thanks for indicative quote, does that annuity payout change e.g. can it go up or down due to inflation? or is fixed?

The sample quote I ran was for a level payment. It's possible to build in annual escalation on annuity payments, but this comes at a cost of a lower starting rate.

(As an aside, although it can sometimes seems counter-intuitive, I often advise against availing of escalation on annuities. If I run a level quote and an escalating quote and model the two side-by-side on a year-by-year spreadsheet, usually it takes over 25 years before the cumulative amounts paid by an escalating annuity will match those paid by a similar level one. Taking into account NoRegretsCoyot's point above about spending habits, I believe level annuities to be better than escalating ones in most circumstances.)

 

[NoRegretsCoyote]

The state wouldn't need to tax drawdowns on a nimplicit 4% basis if it taxed inheritance to children properly.

If someone dies young it's not uncommon that tax is not paid on pension contributions, returns, or when the balance is bequeathed to a child.

 

[NoRegretsCoyote]

(As an aside, although it can sometimes seems counter-intuitive, I often advise against availing of escalation on annuities. If I run a level quote and an escalating quote and model the two side-by-side on a year-by-year spreadsheet, usually it takes over 25 years before the cumulative amounts paid by an escalating annuity will match those paid by a similar level one.

Again it depends on your sample period. Take two ten-year periods and the HICP, my preferred measure of inflation:

96-06: +34% over the whole period
08-18: +2% over the whole period

In the first period inflation rose by a full third. This would really, really reduce the living standards of a pensioner on a non-indexed annuity. In the second period it rose by an amount you would barely notice, and there would have been little benefit to indexation.

Both of these periods when Ireland's monetary policy was set in Frankfurt too.

The lesson is that inflation in Ireland is historically volatile, and if you are risk averse then index-linking (at least partially) makes sense.

 

[Sarenco]

I accept 4% of each years value will never actually reach 0, but that doesn't change the thrust of my point.

Doesn't it?

There is nothing remotely reckless about a variable percentage withdrawal that starts at 4% and rises to 5% for the over 70s.

If anything, it's probably overly conservative.

 

[SPC100]

Fair point, I might be wrong. Reckless seems like an unfair claim.

We'd need to run the models with equivalent assumptions and see which gives a better outcome. In terms of providing the desired income, total income generated, and remaining wealth.

 

[SPC100]

I wish there were more papers studying the Irish case! I will read a bit more about the variable percentage withdrawal methodology as that is what we have here.

 

[NoRegretsCoyote]

I wish there were more papers studying the Irish case! I will read a bit more about the variable percentage withdrawal methodology as that is what we have here.

Is the Irish case all that relevant? Ireland doesn't have its own currency and my guess (correct me if I am wrong) is that there is very little home bias anymore in most pension funds under management.

A lot of listed Irish firms generate the bulk of revenues outside Ireland anyway.

 

[Sarenco]

We'd need to run the models with equivalent assumptions and see which gives a better outcome.

Well, it depends what you mean by a "better outcome".

A fixed withdrawal amount carries the risk that you will either run out of money during your lifetime or leave a very large estate.

A variable percentage withdrawal simply acknowledges the fact that the returns on the underlying portfolio will be, well, variable!

Bengen's work was simply intended to demonstrate that the sustainable amount that could historically be withdrawn from a portfolio was substantially less than the average return on that portfolio (because of the possibility of an adverse sequence of returns).

If you want (or need) the certainty of a fixed amount every year, you are better off buying an annuity.

 

[elacsaplau]

It's 4% of the balance each year. Not 4% of the starting value. If you withdraw 4% of the balance, it will never go to zero. Or am I missing something?

Hi RedOnion,

My understanding is that the initial seminal study looked at what the safe withdrawal rate was as a % of the starting lump sum that could be withdrawn each year and for this initial amount to be increased each year by inflation.

 

[LDFerguson]

Again it depends on your sample period. Take two ten-year periods and the HICP, my preferred measure of inflation:

96-06: +34% over the whole period
08-18: +2% over the whole period

In the first period inflation rose by a full third. This would really, really reduce the living standards of a pensioner on a non-indexed annuity. In the second period it rose by an amount you would barely notice, and there would have been little benefit to indexation.

Both of these periods when Ireland's monetary policy was set in Frankfurt too.

The lesson is that inflation in Ireland is historically volatile, and if you are risk averse then index-linking (at least partially) makes sense.

Sorry if I didn't make this clearer in my post: the examples I've looked at were comparing a level annuity with an annuity that escalates at a fixed percentage per year. In those examples, the level annuity always came out looking like better value unless the individual was wildly confident of living to a ripe old age.