Political uncertainty and portfolio managers in emerging economies

ABSTRACT

This paper claims that investors dislike political uncertainty about future policies and value stability in the political environment. We test the validity of this hypothesis by first studying variations in equity flows from global investment funds to emerging markets in periods surrounding elections. Second, we interpret changes in democracy score as another event potentially creating political uncertainty and assess their effects on equity flows. Our results corroborate our hypothesis. They indicate that the period following an election is generally characterized by a fall in equity flows, and that this occurs only where the incumbent is not re-elected, suggesting continuity is valued by investors. We also find that this effect is confined to presidential regimes, a result we interpret as further evidence that potentially radical swings in policy affect investors’ choices. Finally, a decrease in the democracy score implies lower equity flows, while democracy in itself does not affect equity flows, which is consistent with our view that equity funds are vigilant when potentially adverse changes in the political environment arise.

KEYWORDS:

• Equity flows
• elections
• democracy
• emerging markets

ACKNOWLEDGEMENTS

The authors would like to thank two anonymous referees whose remarks helped to significantly improve this work.

Notes

1. Frot and Santiso (2010) offer other insights regarding additional results from the literature on the links between finance and politics.

2. See economic historians’ narratives on the financial origins of major political events such as the French Revolution, the 1848 Revolution or the outbreak of the First World War (Ferguson, 2006; 2008).

3. On the relations between political democracy and financial globalization between 1870 and 2000, see Eichengreen and Leblang (2006). See also Campos and Coricelli (2009) on the subtle relation between financial liberalization and democratic regime. See also Haber et al. (2008) on the relations between political institutions and financial development.

4. See Jensen (2008) for a presentation of these arguments applied to foreign direct investment.

5. Our dataset also contains information about bond flows, but for a shorter time span, fewer countries and fewer funds. For these reasons, we only study equity flows in this article. Readers may refer to Frot and Santiso (2010) for results about bond flows.

6. Frot and Santiso (2010) describe others (Chile and Ecuador) that also fit our story.

7. Although we believe it makes sense to have a specification with a lag, all the results are virtually identical if we drop this lag from the list of independent variables.

8. One might be concerned that there is not enough variation within countries for the fixed-effect estimators to precisely estimate the democracy coefficient. We ran the same regression with random effects, and a simple ordinary least-squares without country fixed effects. The coefficient on the democracy score is not significant in any of them.

9. Note, however, that Li and Resnick (2003) reach the opposite conclusion.