Software for retirement planning

J

jpmackey

Registered User
Messages
39
Hi,

Just wondering if anyone can recommend a retirement planning software package suitable for Irish people?

Most I have seen promoted are targeting either US or UK people, with country specific taxes, ISAs, Roth IRAs etc etc

I am currently using self-developed excel models but they are really lacking in stress testing eg Monte Carlo sims, which I could develop but it would be a serious effort so I am willing to purchase a package instead.

Cheers
JP
 
Let's say you have a very good package designed for the Irish pensions and tax system.

What questions would it answer for you?
What decisions would you make differently today based on the numbers crunched out by the package?

My gut feeling is that the future is very uncertain and trying to put numbers on it would give it a false degree of certainty.

All you could get in broad terms is
1) You have loads of money so don't worry.
2) It will be about ok, so don't go too mad now and if you retire early you might get into real difficulty later.
3) It's going to be very tight to stop going on holidays and make your car last as long as possible.

I don't really know what other outcomes it could have.
 
Last edited:
I've used ChatGPT to figure out stuff like what (as an early retiree in my late 50s and unlikely to return to paid work) might be an optimal sustainable drawdown rate from my pensions given my known current annual expenditure, my desire to minimise the tax take (both now and when mandatory drawdown rates kick in at age 70-75), my desire to bridge the small gap between my current PRSI contributions/credits and 2,080 to qualify for the full state contributory pension, what I might need to do to qualify for the benefit payment for 65 year olds, the fact that I have other non-pension savings/investments and a small additional fluctuating income stream, and so on. It can even run analyses of the likely effect on pensions of a given drawdown rate, assumed growth rate, charges etc. including stochastic and Monte Carlo simulations.

As ever you need to double check that what it's giving is reasonable and based on current accurate information because it can give bogey information sometimes. It obviously helps to give it well thought out, clear and comprehensive questions.

But I've found it a very useful back and forth process and helpful for clarifying my thinking and understanding and arriving at a suitable strategy for my needs.
 
Last edited:
Here's an example of one such analysis that it did for me - it even generates the Python scripts for running such simulations should you want to reuse/adapt them if you have the necessary technical expertise.

 
Last edited:

Thanks, Conor, that site is a great resource that I haven't come across yet. Plenty to peruse. Will try out the template.


I think I agree with you mostly. What I am worried about is that I have glaring mistakes in my understanding of pensions & taxes that is giving currently giving me a false degree of certainty! If a package highlights to me that my plan only has say a 50% chance of success then that will raise a flag for me that I currently don't have waving.

I've used ChatGPT...
Click to expand...

Super stuff, Clubman, many thanks for the detailed info...I never thought to use ChatGPT for such a purpose. Will give it a go.
 
Brendan Burgess said:
My gut feeling is that the future is very uncertain and trying to put numbers on it would give it a false degree of uncertainty.
Click to expand...
Far from it, at least in my case. I’m 20 years out from retirement, less if I get my skates on AVC-wise and if markets are favourable. Sequence of Returns Risk is the big determinant of retirement success (where success equals running out of time before running out of money). History doesn’t repeat but it does rhyme, so having some knowledge of different historical retirement cohorts and how to play the market cards you’re dealt for the day you retire is powerful, particularly when you can choose your date of retirement.

To be brutally frank, I’m 46 now and should qualify for COAP by 60, so a lost decade (like the 70’s) as I maximise my AVC’s throughout my 50’s, only for the markets to go on an 80s-esque tear when I turn 60 would be my ideal scenario. If I was choosing an early retirement date in the near future though, I’d definitely be in the one-more-year camp with the returns of the past decade and the concerning outlook for the coming decade.

Studying those historical cohorts, as EarlyRetirementNow discusses at length, is hugely valuable in terms of maximising your retirement.
 
Yep, I’ll hopefully have qualified with 2080 contributions. The valuable part of what I mentioned above has been figuring out how to bridge from retiring at e.g. 60 through to COAP at age 66. Tbh, the SWR workbook has even been really helpful testing out scenarios re deferring COAP versus, say, purchasing a small annuity from your pension pot to at least make sure significant other doesn’t run out of cash if my demise happens long before hers.
 
I had a typo in my response "a false degree of uncertainty" where I meant "a false degree of certainty" so I am not sure which people are agreeing with or disagreeing with.
 
Last edited:
conor_mc said:
Studying those historical cohorts, as EarlyRetirementNow discusses at length, is hugely valuable in terms of maximising your retirement.
Click to expand...

So Conor

Based on the ERSWK which you so strongly recommend, what different decisions are you making today?

conor_mc said:
maximise my AVC’s throughout my 50’s, only for the markets to go on an 80s-esque tear when I turn 60 would be my ideal scenario.
Click to expand...

The markets will rise and fall independently of what ERSWK says.

You are 46 now. You can't afford to retire now. That is probably clear enough.

You will have a bit more visibility when you are 56 but still won't know if you can go a year or two early.

By all means, count your money and play with this "what iffery" as it is fun, but I would be surprised if you can tell me any major decision that you would make today which would be different as a result.

 
Here is a question faced by some people in their 50s.

"Can I afford to gift my son €100k as a deposit on a house?"

There are probably 3 cohorts
1) I can do so comfortably, as I will be mortgage-free in retirement and will have a fat civil service pension.
2) I don't have that money now so I can't do it.
3) Not sure. Because the future is so uncertain. However, I can lend him the money now and he can repay it if I ever need it.

I just don't think tinkering around with Chat GPT or ERSKW will give you any clearer answer.

An engineering friend of mine told me of a principle in engineering "You cannot add a more precise number to a less precise number." and I see accountants and financial planners doing that all the time.

Your earnings over the next 20 years, the rate of return on your investments, your health, are all less precise numbers.
 
jpmackey said:
I am currently using self-developed excel models but they are really lacking in stress testing
Click to expand...

ClubMan said:
Here's an example of one such analysis that it did for me
Click to expand...
I'm not being funny but ClubMan's ChatGPT output is basically a visual version of what i get with my own Excel model. The performance for whatever your investment vehicle(s) is/are out there from which you can calculate averages, medians, standard deviation etc.
  • My version of stress testing is to basically assume the worst 20 year annualised return will continue for the rest my life.
So you can continue refining your own model would be my suggestion.
 
ClubMan said:
I've used ChatGPT to figure out stuff
Click to expand...
I was trying to use ChatGPT for some frivolous things in recent days and it couldn't get anything right, not even basic things. After apologising for getting the answer wrong, it would consistently spot out the wrong answer over and over again. I'm sick of seeing this message from it:

 
Last edited:
As I've said, I've seen it make mistakes and have had to ask it to correct things sometimes, but having double checked what it gave me back and further analysed it offline I'm happy that using it as I've described has been a useful exercise.
 
Last edited:
conor_mc said:
less if I get my skates on AVC-wise and if markets are favourable
Click to expand...
In my opinion above-average/median returns should be treated with extreme caution as they'll highly likely revert to the mean later.

My own model suggests a safe retirement age of 60 with annualised returns at 65% of average/median and a drawdown rate which would leave me comfortable in retirement.

It's currently still well ahead of profile despite recent market falls, but I'd need it to be probably double the profile value before I'd risk retiring at 55 (my theoretical earliest retirement date).
 
ClubMan said:
have had to ask it to correct things sometimes,
Click to expand...
This weekend, it had been entirely unable to correct itself. In the sources it listed for it's incorrect answers it actually had a source with the correct answer, but it didn't use it for some reason.
 
BIG-notorious said:
In my opinion above-average/median returns should be treated with extreme caution as they'll highly likely revert to the mean later
Click to expand...
Let's suppose you are 59 and considering retiring. Your returns have been 9%p.a., but you consider the average/median return to be 7%. What action do you take regarding this extra 2%?
 
You must log in or register to reply here.
Menu
Log in
Register