Is there a democratic premium? Elections and financial markets in emerging economies

Javier Santiso, Emmanuel Frot 27 November 2010

a

A

In past decades, nearly all the major financial crises in developing countries occurred in sync with electoral cycles. In the 2000s, this situation began to change, with many emerging countries experiencing a “decoupling” of their financial and electoral cycles. Brazil’s recent election, held in October 2010, conforms to the new trend. While Dilma Rousseff, the newly elected president of Brazil, might have been too left-wing for many investors around the globe, her steady march to victory went hand in hand with rising Brazilian stocks (up by more than 20% since mid-2010) and currency (up by more than 10% against the dollar).

The relationship between finance and politics is obviously not new, nor should it be seen as exclusive to developing economies. The past is full of examples of extreme, critical junctures linking finance and politics. Some economic historians such as Niall Fergusson (2008) argue that the origins of major political events such as the French Revolution can be traced back to a stock market bubble caused by a convicted Scots murderer. Between 1927 and the early 2000s, the excess return in the US stock market was consistently higher under Democratic presidents than under Republicans, according to Santa-Clara and Valkanov (2003).

The year 2010 thus far has also underlined how financial markets continue to be sensitive to political issues, in particular around election times. The sudden death of Néstor Kirchner, the former President of Argentina, in October 2010 is a recent reminder. Financial markets surged with bonds and stocks rallying massively after the news. On the day of the news, the MSCI Argentina index rose as much as 13% and shares in Galicia, the country’s biggest consumer lender, increased by more than 26% in New York.

The latest Greek crisis also underlines how much elections drive markets. In October 2009, the Greek Socialists won the general election, defeating the Conservatives who had been in power since March 2004. Some weeks later, the new government revised the 2009 deficit projection sharply upwards and in doing so they initiated a major financial meltdown that continues to send waves across Europe. The Greek turmoil prompted comparisons with Argentina who defaulted in the early 2000s during an election period. 

Democracies and emerging markets

The relationship between democracy and finance is particularly close in emerging markets, where we have been witnessing a major trend towards democratisation over the past decades, concurrent with increased activity in the domestic and global financial markets. Mexico for example, suffered major financial collapses in 1976, 1982, and 1994, all occurring at the same time as presidential elections. Yet more recently, the democratic transition in 2000 and more generally across Latin America with several presidential elections in 2006 have demonstrated that the electoral curse is not inevitable. Despite 2006 being a very busy year for elections, no country in the region experienced a financial crisis.

In several recent research studies (Nieto Parra and Santiso 2008, Nieto Parra and Santiso 2009), we show that, while analysts’ recommendations between 1997 and 2008 tended to be consistently negative as elections approached, in 2006 they were no longer systematically negative. In particular, for Brazil, Chile and Colombia, Uruguay or Peru, analysts no longer anticipate the risks of a credibility gap in economic policies. Exchange rates and risk premiums have not been eroded on a large scale as they have been on previous occasions. In 2006, most Latin American countries therefore experienced major elections without suffering any major crises. In 2010, Chile, Costa Rica and Colombia also held elections and also avoided financial turbulences in spite of a highly risk averse international environment.

New research on the democratic premium

In Frot and Santiso (2010) we use new datasets to study this topic. We study the influence of political events on portfolio equity flows, focusing primarily on data on portfolio flows before, during and after elections. We concentrate on the analysis of portfolio flows and fund managers themselves, rather than on stock market, foreign exchange, or interest rate variables. In doing so we ask the following simple questions: Is democracy good for portfolio flows? Is an improvement or deterioration in democracy indicators good or bad for portfolio flows?

Our results indicate that elections do affect portfolio equity flows, confirming previous research. Financial markets, in general, and financial markets in emerging countries, in particular, are highly sensitive to electoral outcomes as well as to regular democratic events such as elections. The period following an election is generally characterised by a fall in equity flows. However this occurs only if the incumbent is not re-elected (or was not a candidate). We interpret this result as evidence that uncertainty about future policy plays a big role in explaining the effect of elections. We also find that the fall in portfolio equity flows is restricted to presidential regimes. Figure 1 shows the effect of an election, taking place in month 0, on equity flows, using an 18-month window. The vertical axis is in millions of constant dollars, and measures the deviation from the counterfactual that no election occurs. There is little happening before the election, but equity flows are below the counterfactual in the months immediately following the election.

Figure 1. Effect of an election, occurring in month 0, on equity flows (million $)

Turning to ideology, we find no significant evidence in support of the hypothesis that a post-election change from left to right (or vice versa) leads to a decrease in portfolio equity flows. In contrast, the change in ideology affects bond flows, and the effect is more pronounced for a change from a right to a left wing government than the opposite. This set of results suggests that investors value continuity and stability in the political environment. A change of leader or of political ideology usually results in lower than average portfolio flows. This result also confirms previous work and findings, but relies on a broader number of observations than in past studies. Finally, democracy in itself is not found to significantly influence portfolio equity flows, but (negative) changes in democracy do. A decrease in the democracy score implies lower portfolio flows. Yet, an improvement in such score has no effect. Again, investors do not particularly value a given political environment, but dislike changes in this environment.

Conclusions

The findings of our new research support our hypothesis that elections have an effect on portfolio flows only when they create political uncertainty. Such findings give policymakers some room of manoeuvre.

Political uncertainty can be quickly resolved and portfolio flows return to previous levels within an average of 8 months after an election. While this is a relatively short period, the economic costs may still be substantial, the worsening around election times being frequently severe, at least in the 1990s.

Politicians can take steps to reduce political uncertainty by making their intentions clear. They can reduce the incongruous information foreign investors face by publicising their political agenda to the financial actors in the country as well as to their citizens. Another option is for politicians to send a message of credibility and policy options by tying the hands of the electoral candidates ex ante through pre-electoral commitments binding them to implement "market friendly" reforms. This strategy was successfully adopted by the former Brazilian President, Luiz Inácio Lula da Silva, in 2002.

At same time, and in the other direction, politics in OECD countries is increasingly priced now by financial markets. A government changes in Greece and a massive crisis follows; a weak coalition is anticipated in the UK and the exchange rate against the euro moves; a referendum is lost in Iceland and risk premiums go straight to the sky. It looks as though in this area of finance and politics, developed and developing countries are converging.

References

Ferguson, Niall (2008), The Ascent of Money: A Financial History of the World, New York: Penguin Press.

Ferguson, Niall (2006),“Political risk and the international bond market between the 1848 Revolution and the outbreak of the First World War”, Economic History Review, 59(1):70-112.

Frot, Emmanuel and Javier Santiso (2010), “Portfolio managers and elections in emerging economies: how investors dislike policy uncertainty”, Stockholm Institute of Transition Economics and ESADE Business School, Working Paper.

Martínez, J and J Santiso (July 2003), “Financial markets and politics: the confidence game in Latin American emerging economies”, International Political Science Review, 24(3):363-397.

Nieto Parra, Sebastián, and Javier Santiso (2008), “Wall Street and Elections in Latin American Emerging Democracies”, OECD Development Centre Working Paper No. 272.

Nieto-Parra, Sebastian and Javier Santiso (2009), “Revisiting Political Budget Cycles in Latin America”, OECD Development Centre Working Paper No.281.

Santa-Clara, Pedro and Rossen Valkanov (2003), “The presidential puzzle: Political cycles and the stock market
”, Journal of Finance, 58:1841-1872.

a

A

Topics:  Politics and economics

Tags:  elections, democracy, financial crises, emerging markets

Professor of Economics at ESADE Business School

Senior Econometrician, Microeconomix

Events