This wiki article is inspired by this post by @TeaGhost. This is very much a work in progress and should grow over time through discussion. As always, legal disclaimer: this is not investment advice.
Intro
So here you are – you paid off your debts, you set aside an emergency fund (and maybe even f-you money) and you’re investing your money into a global stock index fund like VT. Now you ask yourself: where do I go from here?
This wiki article aims to show why and how to diversify a portfolio for an investor with Swiss domicile and CHF as home currency. It takes a bit of a different approach, in that it doesn’t list the various asset classes and tells you what they’re good for (e.g. “what are REITs and when should you buy them”). Instead, it lists various goals you might have, how you could achieve these goals and what to expect from these adjustments.
The three main goals are:
- Improve risk-adjusted return: you feel like the risk of 100% stocks is not worth the returns, so you want to reduce risks while increasing the return relative to the risk you take.
- Optimize returns: you want to increase returns slightly while keeping risk the same, or maybe even slightly reducing volatility and maximum drawdown.
- Increase returns: you want to increase returns significantly and are willing to accept even higher risks.
Improve risk-adjusted return
Less pain, less gain
Reduce concentration of mega-caps
Show/hide details
What is this about?
Titles in indices are usually weighted by market capitalization, so situations may arise where only a few companies make up a large portion of an index. For example, currently (summer ’24), Nestlé, Novartis and Roche make up nearly 50% of the SMI, and the so-called Magnificent Seven (Apple, Amazon, Alphabet, Meta, Microsoft, Nvidia and Tesla) make up about 30% of the S&P 500.
By reducing the weight of mega-caps, the goal is to reduce cluster risks.
How would I do that?
Use funds that either put a cap on how much weight a title can have within an index (like the SLI, which caps titles at 9%) or that weighs each title equally and not according to their market capitalization.
Did it work in the past?
Not really, at least not for the goal of increasing risk-rated returns. According to this article, equal-weighted indices tend to have slightly higher returns, but also higher volatility and higher maximum drawdowns. And this study found:
The study found that on a risk-adjusted basis the mean weekly Sharpe ratios were not significantly different for the S&P 500 market weight index as compared to the S&P 500 equal weighted index for four of the five periods tested. However, for the period from April 2009 to March 2020 the S&P 500 index Sharpe ratio was statistically significant which indicated on a risk adjusted basis the capitalization weighted index outperformed the S&P 500 equal weighted index.
Reduce exposure to U.S. stocks
Show/hide details
What is this about?
US stocks currently takes up more than half of the global market capitalization. By reducing the weight of US equities, the goal is to reduce cluster risks.
How would I do that?
Use separate funds for US and for non-US regions.
Examples
- Split VT into VTI and VXUS
- Split MSCI World into Xtrackers MSCI World ex USA UCITS ETF 1C and Xtrackers MSCI USA UCITS ETF 1C
Did it work in the past?
It did when US underperformed and didn’t when it outperformed (no shit Sherlock). Looking back 50 years, US stocks have massively outperformed the rest of the world, but all of those gains came from the last 10 years (see #1 here). There will most likely be periods when US will underperform again. Perhaps US underperformance starts today and people who underweighted US will have beaten the market when looking back in 20 years. Or it starts in 5 years and those people will underperform the market because they lost out on those 5 years on US gains.
Increase allocation to bonds
TBD
.
Optimize returns
Same pain, maybe a bit more gain
Base weighing of countries and/or emerging markets on GDP instead of market capitalization
Show/hide details
What is this about?
Usually, multi-country ETFs base the weight of included countries based on the market capitalization of their publicly-traded stocks. Alternatively, countries can also be weighed according to their GDP. This way, the weight of countries that have relatively more companies that are not publicly traded would be increased and the weight of those that have less would be decreased.
How would I do that?
The easiest way to use this strategy is to leave the weights within developed markets and within emerging markets at market cap (so use “normal” ETFs for DM and EM) and only use GDP-weights to balance DM compared to EM. So instead of the usual 90% DM and 10% EM, the weights would change to 70% and 30% respectively.
Alternatively, special ETFs that weigh the countries based on GDP. The availability of such ETFs is limited or even nonexistent.
Did it work in the past?
In the past 50 years, developed countries, when weighed based on GDP, performed more or less the same as when weighed based on market cap. If one were to exclude the US, GDP-weighted performed slightly better with about the same volatily and maximum drawdown.
Add asset classes with low correlation to stocks
TBD
.
Increase return
More pain, more gain
Private equity
TBD
Leveraged ETFs
TBD
High-risk bonds
TBD
