A new comparator for third pillars

[Baptiste Wicht]

Hi everybody

I have done one more comparator for the blog: A comparator for third pillars

Third Pillar Comparator - The Poor Swiss

This calculator assumes that you are contributing once a month to your third pillar. The world allocation defines how much of your portfolio is invested

thepoorswiss.com

I would love to get your feedback!

Currently, this works with a monthly investment, but I am thinking I should switch that to a yearly one because most people know how much they put per year, but not necessarily per month. Also, I am wondering whether I should include the number of years until retirement in the calculator to compute the total different over many years. What do you think?

 

[gaijin]

I would love to get your feedback!

Looks great! Three spontaneous thoughts:

• "Portfolio value" -> as with the broker comparison, I'm not sure a lay man/woman understands this term. The following suggestions are probably too long, but might give an idea how to rephrase it: total already existing money in all your third pillar accounts (cash and/or invested)
• "World Allocation" -> are you talking about bonds, equity, or other forms of investments (I guess technically even cash could be a world allocation if the custodian bank is abroad. But that's probably not a realistic scenario with a 3rd pillar provider)
• How should people with a 3rd pillar with these dreadful live insurance products use your comparison? Maybe you can add a field or information that will convince them to cancel their existing product immediately

 

[Baptiste Wicht]

Thanks for your feedback @gaijin, this is much appreciated!

I have amended the two terms and tried to make them simpler. And I have added a section in the doc about the life insurance product. I am not sure about adding a field. I think that this will complicate things and it's too difficult to estimate the fees of life insurance 3a. What do you think this field should do if added?

 

[gaijin]

What do you think this field should do if added?

Fully agree that an additional field can be confusing. I like your sentence about the insurance part, I think that's good enough.

 

[Michael]

I should include the number of years until retirement

I would like that. Recently convinced a friend to switch from a UBS Vitainvest 3rd pillar to a cheaper one, by showcasing that over x years nearly 100,000 CHF could be saved (calculation included 5% average return). If I understand your current comparator correctly, it doesnt include the fact that the costs of the 3a go up every year when you pay in more money (or due to stock market gains). Doesn't this heavily underestimate the effect of a good 3a in the long run?

Also, people with little background in this topic will look at it and think "oh wow these are expensive 3a's, why would I do that when my cozy UBS cash 3a is entirely free and I even get 0.6% interest?!". Personally I would love to see some of the broker comparator functionality in this 3a and include some average returns (maybe even adjustable with standard of 5%) which would show the huge difference between the UBS cash and most of the accounts you show here over a 30y window.

Maybe it would be overkill to add all the different bank cash 3a's in the comparison, but one could have one "cash account example" as contrast, so people realize what they miss out on.

 

[Baptiste Wicht]

Thanks for your feedback @Michael !

If I understand your current comparator correctly, it doesnt include the fact that the costs of the 3a go up every year when you pay in more money (or due to stock market gains). Doesn't this heavily underestimate the effect of a good 3a in the long run?

Do you mean the value goes up? The costs should not go up over time.

Maybe it would be overkill to add all the different bank cash 3a's in the comparison, but one could have one "cash account example" as contrast, so people realize what they miss out on.

I really like this idea. I would indeed not add all of them, but a reference 1% interest rate 3a seems reasonable to add.

I think I should probably split the comparator in two: Fees and Returns. Adding the number of years to Returns makes the most sense since for fees it does not matter much.

What do you think?

 

[Michael]

The costs should not go up over time.

It shouldn't? Finpension has a 0.39% yearly fee. So if I have 30k today on a 3a, I pay 117 CHF this year. Now when I pay in 7200 for 20 years, and every year I have 5% gains (simplification), I end up with a value of ~318,000 CHF in 20 years. And 0.39% of that 318,000 are then 1240 CHF which is >10x of today.

Am I getting this wrong?

I think I should probably split the comparator in two: Fees and Returns. Adding the number of years to Returns makes the most sense since for fees it does not matter much.

If my understanding above is not wrong, then both scale with years (& total amount). But returns will be much more significant than fees (5% vs. 0.39%).

 

[Baptiste Wicht]

Ok, I see what you mean. For me, they only scale with the amount. If the fees scaled with years, the fee itself in percentage would go higher with time. But you are right, hopefully the amount in your 3a grows over time and as a result you pay more fees. But for each 7200 amount, you pay the same amount of fees.

 

[Michael]

But for each 7200 amount, you pay the same amount of fees.

This is true. But if we only look at one timepoint (e.g. the 20000 start that is currently implemented), people might think "oh well, i pay 23.- more custody fees with Frankly than Finpension, that is not that much" (ignoring exchange fees for now). And they decide to stay with Frankly.

But the true story is, that after 30 year, even with the tiny difference of 0.39 vs 0.45, you end up having paid 1855.- CHF more with Frankly than with Finpension (13910.- vs 12055.- total custody fees over 30y).

Given that 3a is a long-term decision, I think this value should be highlighted more than the 23.-. Intuitively people would make the calculation "23.- more per year, 30 years, so i pay 690.- more with Frankly".

Of course all of these rates will not stay the same for the next 30 years, but assuming they do should still come closer to reality than ignoring it.

 

[Baptiste Wicht]

I have added a new field for the years until retirement. I agree that this shows better the scope of the difference.

In the future, I will also try to figure out the best way to integrate a comparison with a cash account.

 

[Michael]

One thing to correct is changing the "(per year)" in the results.
Also, I am heavily thinking about switching to True Wealth 3a, even though no one really trusts that they keep the fees so low for long...




Edit: Oh and the same issue seems to be with the Broker comparator, right?

 

[Baptiste Wicht]

Thanks!

One thing to correct is changing the "(per year)" in the results.

Good catch. Done.

Edit: Oh and the same issue seems to be with the Broker comparator, right?

Indeed for the two last scenarios. Done as well.

Also, I am heavily thinking about switching to True Wealth 3a, even though no one really trusts that they keep the fees so low for long...

Over a long period, this is indeed a significant difference.
I am also not convinced this will stay that cheap for ever, but it's definitely interesting currently. The only downside is losing the very customizable portfolios of FP.

 

[jimmack]

It would be nice to visualise the benefit of maxing out the 3a contribution in January, vs investing monthly over the course of the year. Another factor I was considering, was the stock allocation between 3a providers as well as end of year performance. Do we tend to see much difference in EOY performance between providers?

 

[Michael]

Good points but hard to implement in a calculator...

Do we tend to see much difference in EOY performance between providers?

That's highly dependent on the strategies you choose. There is no way knowing whether the assets in Global 100 from Finpension will outcompete a similar strategy at VIAC next year... And you have many different strategies to choose from also within each provider.

benefit of maxing out the 3a contribution in January, vs investing monthly over the course of the year.

Personally I do it monthly, but invest the rest in normal ETFs. So if I would invest all 3a at the beginning of the year, I would have invested into the other ETFs later. Now the question would be, what performs better; my 3a portfolio or my other ETFs; and that's hard to predict. Maybe 3a would be a bit beneficial because of no tax on the dividends, but then this money is also locked up earlier than in the other strategy.

It would be possible to show the benefit for people who do not invest their money otherwise, meaning they just have spare 7.2k on the bank account at the beginning of the year, but then only gradually pay into 3a. I assume this scenario is what you refer to regarding the benefit of maxing out 3a in January?

 

[jimmack]

I guess, that if there was a way of finding comparable strategies between providers and then using historical data to compare then this might reveal one provider outperforming the others...or maybe not.

The point you made about paying 3a monthly and using the rest of your cash to buy ETFs is a good one, and not one I had fully considered. In my case, I get paid a bonus in Jan and Feb each year, so I have this urge to just max out the 3a immediately and forget about it. But for sure, my ETF strategy is outperforming my 3a, so I might do things differently next year. But in any case, I am still investing 4 figures each month into my ETF (QQQ), so I'm doing well, but there is room to optimise as you have highlighted.

I'll take this moment to thank you for your excellent blog, which has been helping me to invest for years now. Keep up the great work!

 

[Baptiste Wicht]

As @Michael pointed out, this is going to be very difficult to implement. I would need to gather historical data for all of these providers with multiple portfolios. I simply do not have the resources to do that.

We expect most passive providers to have similar performance in the long-term. So, we mostly base decisions on fees. It's true that it's not entirely correct and there are some differences, but I do not think they will matter much in the very long term.