The old Wall Street seasonal trade looks good on paper. Strip out three exceptional years and the whole thing falls apart.
Every spring, the same advice does the rounds. Sell in May and go away. It has the ring of received wisdom, which is precisely why Deutsche Bank's strategists have now spent four consecutive years trying to pull it apart.
Their conclusion, published this week, has not changed: the strategy offers no more certainty than a coin toss.
The numbers look good until you look harder
On the surface, the case is not without merit. A strategy of selling the STOXX Europe 600 at the end of May and returning at the end of September would have delivered 9% annualised returns since 1987, compared to 7.4% for a simple buy-and-hold approach. That 1,314% cumulative gap sounds compelling.
The trouble is that it masks a deeply unreliable record. In 25 of the 39 years tested, the strategy underperformed a straightforward buy-and-hold. The median relative performance is slightly negative. A strategy that loses more often than it wins is not a strategy, the Deutsche Bank team argues. It is a story built on a handful of exceptional years.
Three summers did almost all the work
The years in question are 1998, 2001 and 2002, when equity markets fell hard over the summer months. Remove those three years and the Sell in May trade would have underperformed buy-and-hold over the long run. Over the past decade, it has underperformed in eight of 10 years. In 2025, following the strategy would have cost investors 2.7 percentage points against the market.
That is not seasonality. That is the residual echo of the dot-com bust.
Bonds help, but not enough
Deutsche Bank tested a modified version of the trade, replacing cash on the sidelines with European government bonds during the summer period. The numbers improve: the bond version delivered 11% annualised from 1998 against 6.5% for buy-and-hold. Yet even this enhanced approach only beat the market in 13 of 28 years. Again, strip out 1998, 2001 and 2002 and the outperformance largely evaporates.
It fares even worse in the US
The strategy has a poorer record still against the S&P 500. Since 1973, the straight cash version delivered 9.3% annualised against 10.4% for buy-and-hold. The bond version did better at 12%, but its hit ratio tells the same story: it beat the market in only 22 of 53 years. Last summer, when US equities gained 14% between May and September while Treasuries returned just 3%, following the strategy would have meant missing most of that run.
Deutsche Bank's advice is direct. Ignore the calendar and focus on fundamentals. For the past three years at least, that approach would have served investors better than any seasonal rule of thumb.