From the Arab Spring to the recent uprising against Victor Yanukovic's government in Ukraine, corruption and favouritism have motivated people to pour into the streets to protest against the economic and political arrangements benefiting connected individuals and firms. Such protests have sometimes been successful in unseating unpopular rulers, as illustrated by the recent events in Tunisia, Egypt, Libya, and Ukraine. But are they effective at limiting the extent of corruption and favouritism that set them off in the first place?
We shed light on this question by studying Egypt's Arab Spring. On 11 February 2011, Hosni Mubarak, Egypt's president and de facto dictator since his accession to power in 1981, was forced to resign in the face of large protests in the main square of Cairo, Tahrir Square (Berman 2013). Mubarak's regime was a perfect specimen of economic favouritism. Rampant corruption and repression, which excluded vast segments of the population from political participation, was a major trigger of the protests. Mubarak's fall was followed by a phase of military rule until June 2012, when Mohammed Mursi, an Islamist, was elected president. Mursi's presidency, in turn, was followed by a second phase of military rule starting in July 2013. Throughout these four phases of Egypt's Arab Spring, politically connected firms (in particular, those connected to the Mubarak’s National Democratic Party (NDP), the military, and the Islamists) have seen their fortunes ebb and flow, offering a window to study the real-time effects of episodic street protests against a changing cast of ruling political elites.
The effect of street protests on the stock market valuation of connected firms
Using information from Egyptian and international print and online media, we construct a daily estimate of the number of protesters in Tahrir Square. We then analyse the effect of these protests on the returns of firms connected to the group then in power. Our main specifications estimate the differential changes in the stock market values of different types of connected firms relative to non-connected firms as a function of the size of the protests. The blue bar in Figure 1 shows the effect of 500,000 protesters demonstrating against the current regime in Tahrir Square on the daily stock return of firms connected to the incumbent regime.
- It shows that on days with such protests, firms connected to the incumbent regime tend to lose 0.8% of their value relative to non-connected firms.
The green bar shows the effect of the same protest on firms that are politically connected to one of the two rival groups. Interestingly, this effect is also negative, albeit statistically insignificant. So while protests against the regime appear to disrupt the ‘pipeline’ of privileges and unearned income that flows to politically favoured companies, their political rivals do not appear to benefit from this disruption. In this sense, investors appear to believe that protesters may effectively achieve their goal of reducing future corruption and favouritism in the economy.
Figure 1. Effect of 500,000 protesters in Tahrir Square on the daily stock returns of firms connected to the incumbent regime (blue bar) and rival connected firms (green bar), relative to non-connected firms.
Notes: The red lines show 95% confidence intervals. The specification on the left hand side uses data from 1 Jan, 2011 through 31 July, 2013 and controls for time- and sector fixed effects as well as the interaction of time effects with various firm characteristics, such as its leverage, market beta, etc. The specification on the right hand side drops all trading days surrounding changes in regime, constitutional changes, elections, and other major political events.
Interestingly, the pattern shown in Figure 1 holds regardless of who is the target of the protests. Under Mubarak, we find a significant effect of protests on the relative valuation of NDP-connected firms, but no effect on military and Islamic-connected firms; under military rule, protesters harm the valuation of military-connected firms, but not the valuation of their rivals. When Islamists are the target of the protests, bigger protests harm Islamic firms but have no effect on NDP- and military-connected firms.
The right panel of Figure 1 shows that this association between street protests and the stock market valuations of firms connected to the incumbent power group is not just a reflection of spikes in protests during some of the key events, such as the fall of a president, a change in a constitution, elections, etc. When we drop all days surrounding such major political events from our sample, we find almost the same results as before. In other words, even during periods when protests did not lead to changes in governments or institutions, and even when such changes appear unlikely, protest activity is strongly associated with swings in the relative stock market valuations of (incumbent) connected firms.
Figure 2 shows that the timing of the effect of street protests on the value of firms connected to the incumbent power group is consistent with our interpretation that street protests, and not some other omitted variable, drive the relative fall in the valuation of firms connected to the incumbent government:
- The impact on stock returns is on the day of the protest and the following day, but does not occur before the protest itself.
Figure 2. Effect of 500,000 protesters in Tahrir Square on the daily stock returns of firms connected to the incumbent regime before and after the day of the protest.
We further use data from Twitter to shed light on several interrelated questions. First, we document that activity on Twitter predicts protests in Tahrir Square, suggesting that social media has helped coordinate street mobilisations. Second, we also find that this social media activity has no direct effect on stock market valuations with or without simultaneously conditioning on street protests.
We draw two main conclusions from these facts.
- First, since the Tunisian and Egyptian Revolutions, a widespread view in political science has been that social media has fundamentally altered the nature of successful mobilisation against a repressive regime. Our results are consistent with the view that it has indeed facilitated coordination among potential demonstrators.
However, the fundamental challenge does not seem to have changed: discontent voiced on social media has no effect on corruption and favouritism unless it results in mobilisation in the street.
- Second, large protests and street mobilisation appear to be effective means of changing the future distribution of power and act as a constraint on the ability of connected firms to siphon off returns to corruption and favouritism.
This interpretation is consistent with the results in the previous literature that protests during various critical periods have an autonomous impact both on future protests and on certain economic and political outcomes (Aidt and Franck 2014, Madestam et al. 2013), and by our finding on the interplay between social media activity and protests.
Acemoglu, D, T A Hassan, and A Tahoun (2014), “The power of the street: evidence from Egypt’s Arab Spring”, NBER Working Paper w20665
Aidt, T S and R Franck (2014), “Democratization under the threat of revolution: evidence from the great reform act of 1832”, Working Paper .
Berman, S (2013), “The Continuing Promise of the Arab Spring”, foreignaffairs.com, 17 July.
Madestam, A, D Shoag, S Veuger, and D Yanagizawa-Drott (2013), “Do political protests matter? Evidence from the tea party movement”, The Quarterly Journal of Economics 128 (4), 1633–1685.