“Mexico leftist Amlo vows no nationalisation, no expropriations”

Financial Times headline, March 9, 2018

One of the challenges of analysing the effect of political change on capital markets in emerging markets is the short history of the asset class. With at most thirty years of data, a country with a four-year electoral cycle will only have seven or eight electoral data points to consider, and these must also be taken in the context of economic and market drivers at both the national and global levels.
 
And this matters in 2018 with a fully-loaded electoral calendar (notwithstanding the unscheduled transitions in power we have seen this year in Peru and South Africa). In particular, the two large Latin American markets of Brazil and Mexico go to the polls in October and July respectively (Colombia also votes for its president, in May).
 
Since emerging market equity emerged in the late 1990s as an asset class, Brazil has seen seven presidential elections. The US dollar return of Brazilian equities in those calendar years (as measured by the MSCI Brazil index) has varied widely, from +63.8% in 1994 to -44.1% in 1998. Overall, the average USD return in an election year in Brazil is +8.7%, compared with an average of +9.4% in those same years for the MSCI Emerging Market index, suggesting a limited impact.

Where the Brazilian electoral experience gets more interesting is if we separate out the elections that returned generally centrist/conservative presidencies (1989, 1994 and 1998) from those that returned more left-wing/populist governments (2002, 2006, 2010 and 2014). The first set saw an average market return of +17.9% compared with +7.7% for the MSCI EM index in those same years, while the second set saw an average return of +1.8% compared with +10.7% for the MSCI EM index. It seems that the focus needs to be on the election result, rather than the simple fact of an election. 
 
A potentially similar pattern is also seen in the Mexican presidential elections (although these are held every six years, so there are even fewer data points). Mexico has elected presidents in 1988, 1994, 2000, 2006 and 2012. Despite a wide range of results (including the disastrous year of 1994 which saw a leading candidate assassinated in March and a sovereign debt default in December), the overall pattern is that the average market return in these years has been +15.2%, compared with +9.1% for the MSCI EM index. Significantly, in none of those five elections has the most left-wing candidate been victorious, suggesting a similar market dynamic to that seen in Brazil.
 
What of 2018, though? The Mexican election is likely to be won by a populist left-wing candidate, Andrés Manuel López Obrador (normally known as Amlo). This implies significant risk to Mexican equities in the run-up to the election, and, despite other attractive features to the market, we retain our neutral position because of political risk. We would note though, that as per the quote above, Amlo has sounded far more centrist in recent weeks, and significant weakness in Mexican assets around the election may offer an exciting buying opportunity, as was seen in Brazil in 2002 when Lula was first elected.
 
The Brazilian election is complicated. With Lula (the left-wing PT party’s preferred candidate) in jail for corruption, the leading candidate is abrasive far-right politician Jair Bolsanaro. If he is to be Brazil’s next president, the better comparison may be the present-day Philippines under President Rodrigo Duterte. Since President Duterte assumed office in June 2016, the MSCI Philippines index has returned -10.6% in USD terms, compared with a return of +46.5% from the MSCI EM index. We consequently remain underweight Brazilian equities.