Building a very diversified portfolio as first timer investor

Thanks for the advice… indeed my first doubt was am I making it more complex than I should? I will think on what to trim down. Regarding the specificity of high dividend yield and sector fund, my thoughts were:

  • USDV tracks the “Aristocrats” index, which means the top 100 companies that have increased dividends every year for at least 20 consecutive years. This personally gives me a feeling of high reliability/solidity of the companies tracked by this index.
  • Nasdaq, yes probably I have already exposure on this with the other indexes, but from an economy growth perspective feels also the only sector worth investing also on its own for the coming decades, but yes very much personal taste driven choice :slight_smile:
  • Investing in SRI is kind of a personal/ethical thing, not really driven by the numbers, as I feel I would like to invest in Socially Responsible company. But yes, number wise the answer would be “drop it”

Why would it make a difference for future returns? (and if it matters, you’d also want to track the companies that have increased share buybacks, since they’re equivalent)

Aren’t those just companies that mostly stopped growing (returning money to shareholders since they don’t know what to do with it).

Overall, I’d suggest testing if doing this extra complexity is actually better than just VT. Adding QQQ/USDV/UIMR it’s just changing the allocation of a few companies a bit, but all of those are still highly correlated equities and will move in the same way.

Real estate is different as it might not be as correlated (but then 5% if pretty small, when I was playing with backtesting tools a 5% allocation rarely made much difference in the outcome).

2 Likes

That’s a very conservative choice, though not necessarily “ideal”. Full replication means higher trading and other running costs - and inherently not perfect tracking of an index. Index composition will continuously change, while a fund will only replicate that at certain intervals.

No physically replicating ETF will be able to "always full"y do so.

While TER is highly correlated with returns, it’s not necessarily aa measure of actual performance or tracking error. It’s not hard finding higher-TER ETFs outperforming lower-TER ETFs for (at least) years. On the same index. But it should be a pretty good rule of thumb, to try minimising TER.

I think you got confused here.

Especially since you’re speaking of “in general without exposure to US…”. If anything though, the WHT numbers 15% and 30% numbers you stated apply to US dividends, so aren’t applicable.

For dividends received from other countries, the WHT tax will depend on each and every source country’s withholding tax rates on dividends provided by domestic tax laws - and then, possibly, any relief on that provided by double tax treaties to Ireland or Luxembourg respectively.

As an example (just to drive home that point): Albania has, to my knowledge, a withholding tax rate of 8%. The withholding tax burden on your Albania ETFs wouldn’t be 15% or 30% or in-between, they couldn’t be more than 8%. And then (now I’m making things up), Luxembourg could possibly have a more favourable DTA with Albania, thus making the Luxembourg ETF come out ahead of the Irish. It all depends.

But back to reality: not every country even has a withholding tax on dividends themselves. Case in point: The United Kingdom usually does not charge withholding tax on dividends distributed by resident companies. So the Irish and Luxembourg ETF holds in UK stocks would be on par here - and neither ETF take a hit on dividend distributions.

1 Like

Even though it might get marketed or classified as such (a “technology” index), Nasdaq-100 isn’t really an economic “sector” index.

It’s just a stock exchange index that happens to overweight certain sectors (technology, due to concentration of actual exchange listings) and underweight others (financials due to arbitrarily excluding them of the index).

Technology stocks make up for less than 60% of it. Soft drinks and candy bars, children’s’ toys and trucking services or cosmetics and construction supplies are hardly “technology”.

Multiple studies have shown that dividends or increase in stock price is the same. Moreover, with dividends you will pay more taxes in CH. I think it’s a bad idea.

ESG index outperformed standard indexes, so it’s a good choice to go SRI/ESG. However, only 5% of ESG funds won’t make the difference, so invest more in SRI funds

The SRI is a choice on its own, meaning that I wanted to invest a portion on Socially Responsible ETFs regardless of the “balance”. However, it turns out that Top10 position of World and US SRI index I was checking, are anyways companies already over represented on my portfolio like Microsoft, Pepsi, P&G, Intel, etc., while the EMU one exposes me to a different set of companies. Moreover, TER is comparable and performance wise has actually been even better in the past 5y (I know, past performance is not sign of future ones). Here the tre index/ETF and related factsheets I considered:

Thanks, I will have a second look at it then :slight_smile:

Definitely. Any advice for a good backtesting site/tool? the one I’ve found lack many ETFs (or at least those I’m interested into) in their database :frowning:

If I may, I would disagree if my understanding is correct. TL;DR Acc vs Dist does not make any difference in terms of taxes in CH, considering that both ETF are registered at ICTax.
The Swiss tax authorities will indeed tax distributed dividends, but also the value appreciation part that is caused by accumulated dividends. That’s why if it is not registered and they don’t know which part of the value appreciation is caused by accumulated dividends and which by increase in value price (capital gain, which is not taxed), they will tax the whole appreciation.

Thanks, I hadn’t noticed you had just replied to the SRI thread, quite informative. I will take a second look. My rationale for the choice of UIMR is this other reply above.

Acc vs Dist does not make any difference in terms of taxes in CH -> this is correct.

What I meant is that companies which don’t distribute big dividends keep the money. Thus the stock price increases as much as companies which would have distributed dividends. As a Swiss investor, the dividend will be taxed but not the capital gain. That’s why choosing an ETF like High dividend yield is a bag idea.

  • The taxes will be higher
  • The total return will be the same

https://www.portfoliovisualizer.com/backtest-portfolio but you might want to substitute ETFs with the underlying index (or a proxy).

We’re talking about the underlying stock here, not the fund. It is true that for a Swiss (and also true for many countries) investor stock buybacks are better from a tax point of view instead of dividends. While for the company both are a different way to redistribute capital to shareholders.

1 Like

You’ve had some good feedback here already and have clearly put a lot of time and thought into constructing this. My thoughts for your consideration:

Avoid dividends as much as possible in Switzerland.
There are two ways you make a return in an investment:

  1. Increases in price of the fund
  2. Dividends paid by the fund

For all intents and purposes these returns are equal. A company can decide how to return its net profit/investments to its shareholders. It can either do so through company investments or share buybacks which should increase the price of the fund, or by paying out dividends with cash on its balance sheet which will lower the price of the fund.

In Switzerland you have two tax drags which you should try and minimise as a smart investor.

  1. Wealth tax
  2. Dividends taxed as income

Wealth tax is unavoidable (unless you are investing through a 2p or 3p which is a reason to try max these before anything).

Dividends taxes are really incredibly unfavourable in Switzerland.

  • You get taxed on them as though they are extra income so at your marginal income tax rate (maybe 40% at CHF100k+ income).
  • They are also taxed through wealth tax (maybe 0.5% more).
  • They reduce the value of your fund as the companies have less cash on their balance sheet. This means you get less tax-free capital gain returns.
  • Finally, in some cases you may have un-refundable WHT so could lose 0%-30% more in worst cases.

For the reasons above, in Switzerland it is beneficial to invest in funds with lower dividends and more capital appreciation.

As REITs and Bonds are generally more highly (or completely) taxed as dividends you may want to reconsider allocations of those entirely.

Too many ETF funds

Diversification is great. It is the one ‘free lunch’ of investing whereby you can increase return/decrease risk without sacrificing the other.

However, to achieve ample diversification you don’t need to invest in 9 ETF funds.

This will make things significantly more confusing for you, your taxes, and any rebalancing of your portfolio. It will also likely lead to lower returns due to extra fee drag.

You have picked your 9 ETFs to get exposure to:

  1. 25% USA Mid-Large Cap
  2. 10% USA Small Cap

Unless you have picked your 25% / 10% weightings for a specific reason, why not just go for Vanguard’s VTI which captures the entire US market (small/mid/large)?

  1. 5% USA Tech Focus

Doubling down on tech is appealing looking at the past decade and the promise of tech going forwards. If that is your choice thats fine, just remember a lot of growth value is priced into these stocks already.

  1. 20% Europe Mid-Large Cap
  2. 5% Europe Small Cap

You have given Europe a weighting close to that of the USA when in practice its market cap is closer to 1/3. If you are choosing to overweight Europe as a long term strategy thats fine, if not I would lower it.

  1. 10% Pacific excl. Japan (mostly Australia & Hong Kong)

This is not really an Asia ETF, it is mostly Australia and Hong Kong. Again, putting 10% of your portfolio into these is overweighting them significantly compared to their market caps on a global stage.

  1. 5% EMs

5% in EMs is fine. No comments.

  1. 10% US Mid-Large Cap High Dividend Companies

See my points above about the Swiss costs of dividend stocks + you are overweighting these by investing in them twice.

  1. 5% Europe Socially Responsible

Personally, I think there are better ways to have a social impact than investing in equities a bank has deemed to be socially responsible. There is simply too big of a grey area/subjective view on what or what isn’t socially responsible. Sacrificing returns/less diversification/over allocation/higher fees makes most of these unappealing. Better just to invest normally and donate some returns. ¯\(ツ)

Alternate ETF Diversification

Why not just pick the ETFs frequently advocated on here? Either VT for total world small/mid/large caps. Or VTI + VXUS if you have specific weightings of US and ex-US you want to stick to (VT will just take their market cap weightings over time).

If you wanted to over allocate to tech stocks or EMs after that then add some QQQ or EIMI into the mix, just make sure you have properly considered these allocations in advance and that they are long term allocations you intend to stick to.

To add to this, if you do decide on weightings then make sure you rebalance to them in the long run. This will allow you to “buy high and sell low” rather than allowing your QQQ to skyrocket to 30% of your portfolio with extreme returns before crashing back down for example.

Finally, right now your mix of ETFs misses out on several markets (e.g. Canada) and over/under weights most of the others.

Keeping portfolios simpler is preferable for many reasons as an investor. Complicating them leads to worse returns/risk in 90%+ of cases.

REITs / Other diversification
Right now you have diversification across equities but not across asset classes.

As retail investors unfortunately we don’t have access to the great asset class diversifiers which maintain equity level returns (venture capital, leveraged buyouts, absolute return, timber, etc.)

Probably the best we have access to are:

  • US equities
  • ex-US equities
  • REITs
  • Bonds

We have already discussed equities enough, really just decide whether you want to stick to market cap weightings of countries/sectors/market caps or whether you think that the market has priced them wrong and want to overweight tech for example.

On REITs, if we were in the US I would say absolutely add them to a portfolio. Because we are in Switzerland and the returns are very tax inefficient I have personally decided to skip on them. Perhaps I could be persuaded if I found a great fund (good returns, good tax efficiency, low/no dividends) but I haven’t seen one yet.

On Bonds, treasury bonds are a decent diversifier (other than as a hedge against interest rates) but have too low returns. I wouldn’t bother unless you are creating a bond-tent as you near retirement or a big purchase (i.e. a house you really want in the near term) and need lower volatility.

Corporate bonds are very correlated to stocks and have lower returns (incentive misalignment vs shareholder structure). Again, I would avoid them unless lower volatility in the near term is an absolute requirement for you.

Diversification with 5% positions

As a rule of thumb, any diversification of less than 5% is fairly pointless in terms of volatility reduction. So I guess 5% is on the edge of pointlessness :P.

Focusing on REITs as your diversifier, I would either up the % or remove it entirely. As a 5% position it will have negligible effects on your volatility. I’d suggest either 0% or 15%-20%.

Domestic equities (small/mid/large caps) are heavily correlated so pretty irrelevant % wise. Here you are just over/under allocating vs market cap.

Foreign equities are slightly less correlated so a good diversifier to add. Again, note the comments on over/under allocating vs global market caps.

This reply got rather long-winded but always fun to theory-craft portfolios! Happy to discuss any of the above in more detail :stuck_out_tongue:

4 Likes

Only in some cities, if you choose a bad place to live.

Citation needed.

Hear Hear!

It’s worth remembering to take into account expense ratios, even if replicating VT, VXUS+VTI can be cheaper than VT (that’s only taking expense ratios and not actual performance into account).

But OP did seem to want some small-cap, so you’d actually still want some more funds, which leads to VXUS+VTI+something for US small-cap + something for rest-of-world-small-cap - which is 4 funds.

Also remember: REIT’s can be tax-efficient in a 3A.

On running the numbers maybe I used a number too high here. I see for CHF100k taxable as single person 30% marginal Zurich, 37% Vaud, 38% Geneva, 33% Bern.

Anyway, paying tax at marginal rate on returns is never great compared to paying 0% for capital gains.

Was a rough approximation because will vary a lot on a case by case basis. But if we assume long run target of CHF2m net worth, take the mid at CHF1m we can see that in 2016 tax rates ranged from 0.16% to 0.65% depending on canton.

VTI includes US small cap.
VXUS includes ex-US small cap.
Adding any additional small-cap specific funds would result in portfolio overweight in small-caps.

This could be a really great solution to effectively diversifying across asset classes as a Swiss investor.

Do you know of any good solutions here? i.e. 3a offering REITs, low TER.

Currently all my 3a is in VIAC equities as everything else seemed pretty abysmal.

Looking at this in a bit more detail - I see with VIAC you can create a custom investment strategy allowing up to 30% real estate exposure. I assume regulations don’t allow higher.

A max of CHF2k investment into a REIT a year won’t be sufficient for most as a portfolio volatility reducer.

I think that’s very good tax advice. Yet very questionable investment advice (in the meaning of the word).

Because yes, from purely a tax perspective, you should avoid dividend distributions. And while I would agree you should take tax considerations into account when investing, that shouldn’t be the only one.

Much of your advices hinges on the assumption of “all returns being equal” - which I think it isn’t. For instance, dividend paying companies are, much more often than not, profitable. Especially if they manage to grow their dividends. Non-distributing companies are… well sometimes, sometimes not.

I’m not getting it!? That table shows wealth tax on wealth tax.
How does that pertain to your original statement regarding dividend taxation (emphasis mine):

Dividends are taxed through income tax.
How are they specifically taxed by wealth tax in a way that capital appreciation is not?

I disagree and it has been demonstrated multiple times that the dividend yield has no impact 8or even negative) on the total return.
One short exemple:
MSCI World high yield has a lower annualized total return that the MSCI world
https://www.msci.com/documents/10199/74fe7e16-759e-405c-96aa-8350623fae65

Looking at that chart: Just exclude financials (primarily banks), and I’d bet the dividend-paying stock would be outperforming.

EDIT: On second thought… maybe, maybe not. Maybe it’s just me, but from the looks, it suggests the HDY index does outperform for in upward trending markets - and underperforms at the tail end of them, and in downward trending markets.

Sure - I’m not advocating avoiding dividend paying companies altogether else you miss out on large sectors of the market. My feedback here was based on investorn00b’s portfolio which looks to overweight dividend paying companies.

All returns are equal in the sense that a company chooses how to return value to its shareholders, through dividends or alternatives.

I haven’t seen any full analysis on returns of dividend companies vs non dividend paying companies but I’d imagine this fluctuates by temporal trends more than anything and in the long run is irrelevant.

I would be very surprised if dividend paying companies have had higher absolute returns than non-dividend paying companies over the last decade. The large tech stocks which have led the market (Facebook, Amazon, Google) do not pay dividends and have contributed the most to recent stock market growth.