Reading The Little Book of Common Sense Investing by John C. Bogle was an eye-opener for me. It made me realize that constructing a personal ETF portfolio isn’t as complex as it seems, and the results I achieve might not differ much from those of highly-paid professionals. Over the past few months, I’ve become increasingly convinced that I need to take control of my retirement savings, especially considering my skepticism about the future stability of nations over the next 30 years.
Additionally, after delving into Bitcoin and the Austrian School of Economics, I’ve come to understand that in our Keynesian economy, money cannot simply be saved passively—it requires active management. With this in mind, I’m embarking on a new chapter in my financial journey, preparing to manage my savings effectively. Let’s dive into the details!
Before we explore my ETF choices, it’s important to first consider the currencies in which I’ll be investing:
All the ETFs I’ve chosen are accumulation-based, meaning they reinvest dividends, which is beneficial for tax efficiency and maximizing compounding. When selecting ETFs, I focused on two main criteria: Total Expense Ratio (TER) and fund capitalization.
If you’re interested in exploring these ETFs further, you can easily check them out on JustETF by clicking on the names in the list above.
I structured my portfolio based on my preferred currency exposure:
This led to the following allocation:
Fees were a key factor in my ETF selection. Over a long time horizon, high fees can significantly diminish the benefits of compounding returns. While I sometimes chose ETFs with slightly higher spreads, I aimed to strike a balance by selecting well-known ETFs with lower overall costs.
ETF | Management fee | Transaction fee | Total fee |
---|---|---|---|
S&P 500 | 0.07 % | 0.02 % | 0.09 % |
STOXX 50 | 0.09 % | 0.01 % | 0.10 % |
MSCI Pacific ex Japan | 0.20 % | 0.02 % | 0.22 % |
MSCI Switzerland 20/35 | 0.20 % | 0.00 % | 0.20 % |
MSCI Emerging Markets | 0.18 % | 0.03 % | 0.21 % |
Gold | 0.12 % | 0.00 % | 0.12 % |
My goal with this portfolio is to create a balanced, long-term investment strategy. Historical performance is not always a reliable indicator of future returns, especially given the potential rise of BRICS and the shifting global economic landscape. Therefore, I’ve diversified across regions and sectors to mitigate risks.
Sector Exposure:
I categorized sectors into three groups:
Country Exposure:
I intentionally reduced my reliance on the U.S., with only 20% of my portfolio in U.S. equities. Europe and Switzerland each also have 20%, with the remaining 20% diversified across emerging markets, including China, Australia, India, Taiwan, Singapore, and Hong Kong.
The ETFs in my portfolio exhibit correlations ranging from 60% to 80%, with gold being nearly independent. This indicates that my portfolio is reasonably diversified, reducing the risk of significant losses during market downturns.
As someone who enjoys math and simulations, I couldn’t resist diving into the historical data. However, it’s important to emphasize that the results presented here are not necessarily indicative of future performance. Markets are unpredictable, and past performance does not guarantee future results. With that caveat in mind, let’s explore the simulation.
All computations can also be found in the GitHub repository!
The goal was to estimate the portfolio’s annualized average returns and volatility, along with the 95% credible interval for these metrics.
To start, I used actual historical data to compute the portfolio’s average annual return and volatility. This was achieved by applying the covariance matrix and the ETF weights.
To assess whether the distribution of returns for each ETF is normally distributed, I conducted three statistical tests: Shapiro-Wilk, Jarque-Bera, and Anderson-Darling. The results from all three tests indicated that none of the ETFs follows a normal distribution. Given this, I chose to proceed with the analysis using a Monte Carlo simulation approach, which doesn’t rely on the assumption of normality.
I ran a Monte Carlo simulation by generating 10,000 series of returns, each obtained by randomly sampling with replacement from the original return series. Using these simulated datasets, I calculated the average returns and volatility. The results, presented with a 95% credible interval, are as follows:
These results suggest that while the portfolio is expected to have relatively low volatility compared to the S&P 500, it remains well-balanced.
Results:
The lump-sum investment strategy significantly outperformed the DCA strategy over the long term. However, DCA remains the only viable option for investors who don’t have access to a large amount of capital upfront. In my case, the choice is to gradually invest through DCA, accepting that this approach might lead to lower potential gains compared to a lump-sum investment.
While the historical performance of this portfolio may not match that of the S&P 500, there are important factors to consider:
In summary, while my portfolio may not outperform the S&P 500 in the short term, it’s designed to be resilient against geopolitical shifts and the potential rise of emerging markets. I’m satisfied with the balance between risk and reward and will begin dollar-cost averaging (DCA) into this portfolio using a fee-free broker like Scalable Capital.
I understand that this portfolio isn’t perfect and that adjustments may be necessary over time. However, I believe that taking action now, even if imperfect, is better than waiting indefinitely for the “perfect” solution that may never come.
Leave a Reply