Consider a Brazilian investor with R$1,000 to allocate each month from January 2000 through June 2026. Both strategies below use month-end closing prices and pay roughly 2.1% in conversion costs, a 1.1% financial transactions tax plus a 1.0% currency spread, each time capital moves offshore.
The first is dollar-cost averaging: convert and invest R$1,000 in the S&P 500 every month, whatever the exchange rate. No view, no discretion, no pause.
The second is impossible. Contributions accumulate at domestic risk-free rates and convert only at exchange rates identified in advance as exceptionally attractive, then buy the same index. Call it the oracle strategy, because it assumes foreknowledge no investor has.
The monthly buyer finished with roughly R$2.92 million. The oracle finished with roughly R$2.17 million. The gap is about R$748,000, or 34.5%, in favor of the investor who never tried to be right about the currency.
The result is worth stating precisely, because it is easy to overclaim. It does not show that currency levels are irrelevant, and it reflects one historical path under a specific set of assumptions. It shows something narrower and more useful: a perfect answer to the currency question did not cover the cost of the time spent waiting for it. If flawless timing loses, the realistic version, hesitant and only occasionally right, starts from a worse place.
