How to Build a Global ETF Portfolio from Scratch

Asset allocation — how you divide your portfolio between equities, bonds, and other asset classes — determines approximately 90% of your long-term return and risk, according to landmark research by Brinson, Hood and Beebower (1986, updated 1991). ETF selection within each category matters far less than getting the allocation right.

The primary driver of your target allocation is your investment horizon. As a rough guide: if your horizon is 20+ years, 80–100% equity allocation is historically well-supported; 10–20 years: 60–80% equity; 5–10 years: 40–60% equity; under 5 years: below 40% equity and consider whether you should be in markets at all for funds you may need shortly.

Within equities, the secondary question is geographic split. A market-cap-weighted global equity allocation (as in MSCI World) puts approximately 65% in US equities. GDP-weighting would allocate closer to 25% to the US. Most long-term European investors land somewhere between these two, with a deliberate tilt toward their home market or emerging markets depending on their views.

For developed market global equity, the three most efficient choices are: MSCI World (available from iShares, Xtrackers, Amundi, and others at TERs of 0.10–0.20%), FTSE Developed World (same coverage, slightly different methodology, available from Vanguard at 0.12%), and MSCI ACWI or FTSE All-World (add emerging markets at a 10–12% weight, TERs 0.20–0.25%).

For additional geographic tilts: MSCI Europe ETFs (TER 0.10–0.15%), MSCI Emerging Markets ETFs (TER 0.15–0.25%), and specific market ETFs (MSCI Japan, MSCI Pacific ex-Japan) if you want granular control. Small-cap exposure can be added via MSCI World Small Cap ETFs (TER 0.35%).

Key selection criteria beyond TER: fund size (prefer funds above €500M in AUM for liquidity), replication method (physical replication preferred over synthetic for most investors), domicile (Ireland-domiciled ETFs are most tax-efficient for EU investors), and share class (accumulating share classes reinvest dividends automatically and are more tax-efficient in most EU jurisdictions).

Bonds serve three functions in a portfolio: they reduce volatility, provide income, and — historically — have shown negative correlation to equities during equity bear markets (though this broke down in 2022). Whether and how much to include depends on your horizon and risk tolerance.

For long horizons (20+ years), bonds historically reduce return more than they reduce risk relative to equities. A 100% equity portfolio has higher expected return and, over 20-year rolling windows, has historically never produced a loss in any developed market. For shorter horizons or investors who genuinely cannot tolerate large drawdowns, adding 20–40% bonds is rational.

If including bonds: global aggregate bond ETFs (such as BGBL, AGGH, or the iShares Core Global Aggregate Bond ETF) provide diversified exposure at low cost (TER 0.07–0.10%). For European investors, hedging currency risk in the bond allocation is important — unhedged global bonds introduce significant currency volatility without commensurate return improvement. Look for EUR-hedged share classes (usually identified by "(EUR Hedged)" or "H EUR" in the fund name).

The mechanical setup of a global ETF portfolio involves four steps. First, select your broker: for EU investors, Scalable Capital, DEGIRO, Trade Republic, and Interactive Brokers all offer access to major ETFs with low transaction costs. Consider the combination of custody fees, transaction fees, and FX conversion costs for your expected trading frequency.

Second, set up a regular investment plan (piano di accumulo / Sparplan / plan d'épargne) if your broker offers it. Automatic monthly contributions at a fixed amount automate the savings habit and provide natural cost averaging across market cycles. Most EU brokers now offer free ETF savings plans.

Third, decide on your rebalancing rule: either calendar-based (rebalance once a year to target weights) or threshold-based (rebalance when any position drifts more than ±5% from target). Calendar-based is simpler; threshold-based is more precise. Both work well for long-term investors.

Fourth, track your portfolio in a tool that calculates time-weighted return and shows you your actual allocation versus target. This is where DonkyCapital adds value — you can see, in real time, whether your portfolio has drifted and what your actual performance has been relative to your benchmark.

The most frequent mistake is over-complicating. A three-ETF portfolio (global equity + European equity tilt + bonds) tracking your target allocation precisely outperforms most complex multi-ETF strategies on a net-of-cost basis over time. Each additional ETF adds complexity, cost, and rebalancing friction without adding proportional diversification benefit.

The second mistake is mixing accumulating and distributing share classes haphazardly. Distributing ETFs pay out dividends, which are taxable events in most jurisdictions. Accumulating ETFs reinvest dividends internally, deferring the tax event. For EU investors outside of specific tax wrappers, accumulating classes are almost always preferable.

The third mistake is failing to account for home-country bias. European investors often over-weight their domestic market (Italian investors buying Italian companies, German investors buying DAX components) beyond what a rational global allocation would suggest. Home-country bias reduces diversification without providing return compensation.

The fourth mistake is chasing recent performance — buying the ETF that did best last year. Sector ETFs on "megatrends" (AI, clean energy, cybersecurity) typically launch near performance peaks, carry higher costs, and revert to mean after their initial marketing momentum fades.