Yes depends on 2nd pillar. I know a friend where interest rate is fixed at 3.5% and then another one where it’s fixed at „BVG min + 1%“
OK, I did that calculation, and it seems so obviously in favor of pillar 2 (compared to free investment) that I want to know if I’m missing something:
| Initial investment | Free investment | Pillar 2 (+ tax savings free) |
|---|---|---|
| CHF 10,000 | CHF 18,681 | CHF 22,467 |
after 30 year with the following assumptions:
- Pillar 2 is an 1e fund with 75 % equities (Swiss focus), some bonds, gold etc.
- mean real return (assumed): 4 %
- mean volatility (assumed): 15 %
- cost: 1 % p.a.
- Free investment is my regular allocation with 85 % equities (global), some bonds, gold etc.
- mean real return (assumed): 4.5 %, of which dividends 2 % (assumed)
- mean volatility (assumed): 17 %
- cost: 0.1 % p.a.
- Correlation between the two (e.g., for Monte Carlo sim): 0.80 (assumed)
- Further assumptions:
- Years to retirement: 30
- Marginal income tax rate: 35 %
- Marginal wealth tax rate: 0.15 %
- Pillar 2 exit tax rate: 8 %
Now, even with a simple calculation of compound geometric growth, pillar 2 pays off.
Let’s say we have CHF 10,000 to invest.
Case 1: invest in free investment (3b)
- arithmetic mean: 4.50 %
- less volatility drag (1/2 * vola^2): -1.45 %
- geometric mean: 3.05 %
- less TER/cost: -0.10 %
- less wealth tax on amount invested: -0.15 %
- less dividend tax drag: -0.70 % (2 % taxed at 35 %)
- net CAGR: 2.10 %
So, compound CHF 10,000 over 30 years:
CHF 10,000 × (1 + 2.10%)^30 = CHF 18,681
Case 2: invest CHF 10,000 in Pillar 2, invest tax savings in free investments
- arithmetic mean: 4.00 %
- less volatility drag (1/2 * vola^2): -1.12 %
- geometric mean: 2.88 %
- less TER/cost: -1.00 %
- net CAGR (pre-exit): 1.88 %
- less exit tax drag: -0.28 % (this depends on the holding period, I broke it down)
- net CAGR (post-exit): 1.59 %
So, compound CHF 10,000 over 30 years:
CHF 10,000 × (1 + 1.88%)^30 = CHF 17,459
× (1 − 8%) exit tax = CHF 16,063
This is obviously worse than the free investment, but we also get to invest 35 % x CHF 10,000 into free investments (I assume we get the tax savings later, so invest one year less):
CHF 3,500 × (1 + 2.10%)^29* = CHF 6,404
So, together Case 2 has:
CHF 16,063 + CHF 6,404 = CHF 22,467
If your marginal tax is 35% and exit tax is 8% and your 2nd pillar is 1E with 75% equities, I think you will be fine with 2nd pillar. This is similar to why 3a with investment option is quite lucrative
just to keep in mind
1e is not standard so you might be forced to sell at wrong time if you change employers. There is some extra time being proposed for transition so that might become less of an issue. But still the 30 year assumption only holds if you continue to be in 1e system for 30 years. If next employer doesn’t have 1e, your returns would be lower but more stable.
4 Likes
Huh. This is the biggest boost, not available to mere mortals. However 1e also comes with risks of changing rules, including when changing jobs.
By the way, I model the difference between tax-advantaged (2, 3a) and taxed investments by assuming starting amounts of X and X*(1 - marginal tax rate). Don’t think it matters.
1 Like
I does, makes the difference even bigger. Plus, seems more intuitive like that.
But as you say, 1e vs. normal pillar 2 are different things to compare.
2 Likes
Even with a withdrawal tax? 
For normal 2nd pillar, I think 10 years is a good rule of thumb limit: longer investment horizon - stocks in taxable, shorter - 2nd pillar.
Yeah (marginal - withdrawal) is a better approximation.
Tho withdrawal can very a lot by Canton so it’s going to be a best guess on future situation.
(I’m currently discounting my tax money by >10% to account for this)
2 Likes
Huh, now I start to remember your comments that multiplication is commutative. Shame on me 
.
So instead of counting withdrawal tax at the end, one can apply different discounts to initial amounts: (1 - marginal tax rate) to taxable and (1 - withdrawal tax rate) to 2nd pilllar. After that, add usual compounding terms with different CAGR, et voilà, ready to compare.
2 Likes
Ignoring spreads, that shouldn’t be a problem as you can then put the money to a similar portfolio in the taxable account. Or you cannot do that?
1 Like
Yes. Formulas aside, they are comparing 13.5k (1e) vs. 10k (3b) to start with, instead of 10 vs. 6.5.
8% withdrawal tax on the 1e part included.
Seems fair. In ZH, married scale it’d currently be 10% at some 1.4 MCHF. Even 10% marginal (10.7%) only at some 572k.
Not too bad, and one could manage it with split accounts or WEF withdrawals in case it becomes an issue.
1 Like
New money - yes, but not those which are already in the second pillar.
If a new employer doesn’t offer 1e, all money go to a usual pension fund scheme OR you are allowed to move some to vested benefits - completely legal if your 2nd pillar balance larger than the new pension fund allows. So, another uncertainty.
Or you decide to stop working soon.
3 Likes
OK, some very good input here.
Regarding 1e risk: I was under the naive assumption that I could just keep my 1e if the new employer does not have a 1e. But that might not be true (have to read up, call them—or does anyone know?) and makes it less straightforward.
I ran the same calculations and simulation with a regular pillar 2 as well, and even then pillar 2 came out on top due to the high tax savings. Basically if you invest 35%+ more money, that is hard to catch up just through higher returns.
Normally you have to leave the scheme, but are you planning to leaving any time soon? I’d invest with optimism.
So many times I did ‘what if’ and it cost me. e.g. I didn’t subscribe to a generous employer scheme with a new job because they had some recovery penalty clause and I “didn’t know how long I’d stay”. It turned out I stayed for over a decade and the ‘free money’ I gave up probably would have been over a million with compounding growth.
3 Likes
It’s not. You have to transfer it out just like the rest.
It will, depending on assumed returns and taxes.
Typically after some 10 years.
3 Likes
Your assumed taxable portfolio is too conservative/pessimistic for my taste. 30 years investment horizon and bonds??? 17% volatility is 100% stocks, what you describe should be around 10%, I would say.
I calculate taxable growth with 5% CAGR, which is on a pessimistic side for 100% stocks.
1 Like
More good points. 
I don’t plan on leaving this job soon but you never know.
Will 1e have to go into regular pillar 2 or vested benefits then?
Also, taxable portfolio might be conservative but it’s the allocation I am happy with (it’s more like 85 equities, 5 treasuries, 10 gold btw; has been for years). 17 % vol is probably too high for that, true.
But I treat the 1e similarly conservative.
Anyway, I will do the calculations with regular pillar 2 and my specific tax rates just to be sure.
Two more things that are hard to price are:
- liquidity: I feel like I have enough in taxable to not worry about P2 illiquidity but liquid is always better than illiquid
- risk: P2 returns have less risk, how do I price this?
It’s possible if you have equivalent cash lying around in taxable account. I think it comes down to overall portfolio of the person
At time of job transfer 1e needs to go to the regular pension plan of new employer if they don’t offer 1e.
But currently it’s easy to “forget” it. In the future it’s likely you’ll have up to two years before transferring (and then you won’t be able to forget it).
OK, this was super valuable input.
I plan to stay with the current employer for quite a while, then potentially move abroad (EU) to my partner‘s home country. I assume moving abroad would also trigger a move to a „normal“ pillar 2 account? (Very specific question and not assuming someone here knows but if you do, please share.)
Anyway, I modeled the whole thing with switching to another, typical P2 (1 % real return with 5 % vol), regime after five years, and even then P2 comes out on top for my specific situation.
So if I haven’t made any grave mistake, this seems like a no-brainer.