Most studies I read contain “data available on request”, with the rare exception putting it directly on Github… I might publish a google doc with a subset, but so far, I don’t see what for.
CAGRs Sept '22 - July '23:
Portfolio 1 (no inflows): 7.71%
Portfolio 2 (no inflows): 14.5%
Portfolio 3 (~100% inflows stripped out TWR-style): 15.15%
Easy to see that others will be in the same ballpark; Sept '22 was a local minimum after all.
I still stand by this, btw: the return decomposition into price delta (which would apply if no rebalancing nor inflows took place) and into fund quantity delta (which applies both due to dividends, CHF 1.00 transfers lol, and churning between not-fully-correlated funds) is as theoretically solid as it’ll get.
We’re not asking for a Google sheet, what we are asking for are (others may complete):
the funds you are using and their allocation in each of the portfolios you are tracking.
the starting value of each portfolio and the date at which it was.
the amount and date of all new inputs made since then (“1 CHF on the first/last business day of each quarter” or somesuch suffices as an answer, no need to spell out every single date since you have a standing order).
Those are the numbers specific to your own experiment that we can’t find in previous academic studies. You’ve already provided clues as to the allocations you’re using for some of the portfolios but having the funds spelled out comprehensively would help (no need to go overboard here either, VIAC has only a narrow set of available funds. “CSIF SMI”, for example, would be sufficient to identify the fund (I think we would only need the fund provider (CSIF/Swisscanto/UBS/other) and the index/commodity tracked, including the mention that it is ESG when/if it applies).
Both with starting values on 18.09.2022 around 14k (stocks+re: 14360 / stocks ch: 13700). Saw weekly inflows in Q4/22, see monthly inflows from Q1/23, give or take on the 7th or the next business day.
shannon’s demon is, as mentioned, VIAC Global 100. Starting @ 3625, saw 4868 weekly inflows ~135CHF until late August - which I have, obviously, removed from performance metrics.
TWRs are bullshit across the board and not to be used except for marketing; you won’t be able to retire off anything but absolute returns in Swiss francs
The 90% Shares - 9% Gold Portfolio is something we can work with; particularly as the fund count is relatively low. Can you as well give us the TWR Performance Figure that VIAC provides you on this figure, from 27 September 2022 to 27 August 2023? Best is to have a look at VIAC’s TWR Performance chart and simply give us both the Performance Figures that VIAC reports on both these dates. Meaning that maybe Sept '22 was somewhere at 5.XY% and end of Aug was at 17.XY%. We can re-base what this means. Important that you don’t change VIAC’s Reporting Timeframe when capturing these data points.
If you want to go the extra mile, appreciate if you can as well give us some interesting Quarter/Year End Performance Figures as well, so that we see where the performance was.
The only additional thing we would need is a dump of all dividend transactions you had in this portfolio. This to both ensure we don’t lose any dividend dates/transactions when re-calculating the “non-rebalanced” return… and to ensure you didn’t receive any WHT credits for positions you held years ago.
And ultimately, can you give us the overperformance and at what precise date you measured these? For such date, would as well appreciate the VIAC TWR figures. AND we need to know whether you receive any referal credits in VIAC or not (aka reduced Management Fees).
Now it only needs someone to crunch this data, which is probably a 2-3 hours job or so… :). But with the above data, it can be calculated.
Over the past year I’ve experimented with adding CHF 1 weekly (in Q4/22) and later CHF 1 monthly (from Q1/23) to the first two portfolios, and ~1/52th of the full 3a contribution limit to the third portfolio.
You’ll be able to find the quarterly updates further up this thread, and I’ve long since noticed that the outperformance attributable to rebalancing (see table above, last row) is highly beneficial. The numbers are in line with research on the topic (also in this thread).
I conclude that if you’re using VIAC and only contribute to one portfolio at present, it’s a good idea to set up a monthly transfer of CHF 1,00 to the others.
Firstly, I remain doubtful that sub-monthly re-balancing should outperform monthly balancing except by coincidence. But even if we accept this:
Triggering a 150CHF cost for a marginal (and uncertain) gain is a best ‘banditry’: pushing costs of your tactic onto other users of VIAC (though you might want to check whether this is the case, I’ve seen that some funds allocate the cost of such transactions to the ones that created the transactions).
Given that there is already regular rebalancing, I think attention and efforts are best put elsewhere rather than try for these micro-optimizations that have negative costs for the community and very limited upside.
VIAC should just enable this for everyone. It’s proven to be beneficial above and beyond the rebalance bands VIAC uses. I’ve linked the research above, feel free to ignore it. And yes, I’ve asked them to implement that.
I clicked through the links and scanned through papers (admittedly very quickly) but couldn’t find anything of the order of 4% p.a. mentioned in research
InvestResolve uses “dynamic” asset weighting methods. Trust me, in hindsight I can always come down with a “formula” that based on volatility or whatever alters portfolio allocations the way I generate an over-performance. This however has nothing to do with rebalancing but this is simply a matter of over-fitting and backtesting.
As Portfolio 3 was a plain-vanilla VIAC 100% Shares strategy. Why don’t you post your YTD TWR Chart from your VIAC app, and show us the return on 30 September? This will allow straight and direct comparison with the Performance Figures stated by VIAC?
VIAC is set up (at least on paper) so that re-balancing and trading cost are borne by the respective investor. Same applies for CSIF / Swisscanto Index Funds; there the cost is embedded into the Buy/Sell Uplifts. In practice, there is still indirect cost as an over-active investor both increases infrastructure load for VIAC and reduces netting/pooling opportunities for other investors.
In my view, its just a stupid idea for the individual investor as the performance of your portfolio would likely not benefit (vs. a monthly re-balancing). I believe in Monthly or Quarterly Re-Balancing; but weekly just burns money…
I didn’t find any research matching that description in post 4 or 21.
Do you mean this article in post 22 ? I scanned it and I did not see anything suggesting a 4% p.a. premium for daily rebalancing in comparison to weekly rebalancing which I understood was your experiment :
In the ReSolve AM Post, see e.g. Figure 5. MAX-DIV 3.2%, ERC 2.6%, HRSS MAX-DIV 3.3%, ENSEMBLE 2.5% – fully in line with the annual numbers posted above
If I’m not mistaken that paper also convinced me to substitute weekly with monthly rebalancing on two portfolios.
Not so; it’s all wrapped in the admin fee. From the FAQ:
What does the VIAC administration fee include?
The VIAC administration fee is 0.52% per annum (at most 0.40% p.a. due to the fee cap) on the invested pension assets. There is no administration fee on the cash portion. The administration fee covers the trading costs and the fees paid by the custodian bank for cash account and custody account management. This fee also includes all foundation administration fees, technology deployment and support. There are no transaction costs, retrocessions or other commissions.
The index funds definitely have a subscription spread and some also a redemption spread. This can be seen clearly in the factsheets. However, as I understand it, this is simply to approximately cover the costs the fund incurs as the money remains within the fund. VIAC’s pooling should reduce the effective spread.
The custodian bank’s transaction fees are covered by the admin fee.
VIAC‘s Admin Fee covers Trading Fees. It neither covers FX Fees nor does it cover the Index funds equivalent to Bid/Ask Spreads. Both are not trading fees as such.
To mitigate this, Viac applies pooling and hedging but most of the time, this means you can sell without fees but you pay upon purchase.
Remember that particularely the EM funds have hefty uplifts upon the sale of fund shares (to cover India‘s Capital Gain Tax).
Edit: Jay just wrote the same, but expressed in a clearer way - so just refer the above post.
These are first of all Risk-Parity Like Portfolios where the majority if benefita comes from the Asset Allocation that builds on the low correlation among Gold, Long Term Treasuries and Shares. You could re-balance such Portfolio once per year and you already had some correlation benefit.
At the same time, they use an active weighting scheme that is prone for overfitting. Just because 29.75% Gold, 50% Treasuries and 30.25% Shares was the best performing Allocation in the last year, there is zero rationale why this should continue to be the same way. Even worse as they use logic to change the percentages on a regular basis and using an obscure volatility formula.
Sad article hence mainly demonstrates the permanent Portfolio aka investing in Non-Correlated Assets plus overfitting of strategies. The paper is not about re-balancing intervalls as such and no (retrospective) value is given the actual re-balancing interval.
By reading and partipating to this forum, you confirm you have read and agree with the disclaimer presented on http://www.mustachianpost.com/
En lisant et participant à ce forum, tu confirmes avoir lu et être d'accord avec l'avis de dégagement de responsabilité présenté sur http://www.mustachianpost.com/fr/
Durch das Lesen und die Teilnahme an diesem Forum bestätigst du, dass du den auf http://www.mustachianpost.com/de/ dargestellten Haftungsausschluss gelesen hast und damit einverstanden bist.
