Portfolio

Bitcoin first. Roughly 40% of the portfolio is bitcoin, held for a decade at a low cost base. It wasn't bought at this size. It grew into it on a decade of outsized returns, compounding past the traditional holdings while they plodded along. That is more concentration than I would choose to keep, and it will come down, but not yet: the portfolio is still in its growth phase, and I see growth left in the bet on new technology. The crypto holdings will be trimmed once the next cycle top arrives, with the proceeds moving into the equity sleeve. Until then the position is held, not traded against its own volatility, and monthly contributions flow solely into the equity sleeve: the crypto positions neither shrink nor get added to. A further ~10% sits in other crypto assets on the same terms. Bitcoin also does the work of gold here: digital gold, so there is no physical metal on top of it.

The equity sleeve. The remaining ~50% runs on a written policy statement: mostly all world, with a few tilts backed by research. Long-horizon real capital growth from a globally diversified, all-equity portfolio, harvesting the equity premium plus documented factor premia (size, value, profitability), not security selection or market timing. Reference universe: MSCI ACWI IMI. Own the whole market first (every region, every cap tier) before any tilt is applied; tilts are overlays, never gaps. Costs stay near index level (blended ~0.17% p.a., ceiling 0.25%). Weights below are shares of the equity sleeve:

Less US than the index. The US has grown so large and so expensive that it now dominates global cap weights, and holding the full weight would tie too much of the future to one market's valuations. So within the equity sleeve, roughly 50% look-through US exposure versus ~64% in global cap weights, held within a fixed 50-55% band on a ten-year horizon. Reviewed against the US vs ex-US implied equity risk premium differential, rather than against relative performance, and not ratcheted with valuations.

Discipline. Every tilt carries a stated rationale and a stated vulnerability, and exits when its rationale breaks, not because of a drawdown. No rebalancing for its own sake. New contributions steer the equity sleeves back toward target and winners are left to run: momentum is real, and cutting winners to feed laggards is a poor trade. Only a sleeve that drifts more than 3 percentage points absolute or 20% relative from target is sold back, and the portfolio as a whole may stay skewed while part of it performs. Valuation signals (CAPE, implied ERP) are long-horizon inputs only; when they disagree, positions are sized to stay defensible under the less favourable reading.

Kept out. Market timing, leverage, single stocks, nominal-yield valuation frameworks, and abandoning a tilt because it underperformed.