May 15, 2012 12:00 PM
By Geoffrey Dennis, Citi Global Emerging Markets Strategist
To "Sell in May" is a fabled investment thesis. But does it actually work? Over the past 20 years, EM equities have tended to do better in the three months up to May (+2.8%, on average) than the three months after (-0.4%). We also show that May-October has been the worst rolling six months of the year for EM since 1990 (average monthly fall of 0.2%). However, while, since 1991, EM equities have risen in the three months to mid-May in 15 out of 22 years (68% of the time) they have still risen in the three months after mid-May in only 11 of 21 years (52%). So, "Sell in May" does work, but barely. Asia and EMEA have clear "Sell in May" effects; for Latin America, they are less clear.
With this in mind, investors currently fear a repeat of 2011. While YTD performance is better this year, the short-term patterns seem similar; the focus is so intense this year as "sell in May" worked so well in 2011. However, "sell in May" has failed badly in six of the past ten years in EM and has only clearly worked in three of the past 10 years (including 2011). In many recent years (2004-7, 2010) there were 'spring' corrections - remember those - which should have been bought. Average full year returns were 27% in those five years.