Do Political Blogs Matter? Corruption in State-Controlled Companies, Blog Postings, and DDoS Attacks
This working paper probes to what extent net-activism can influence corruption behavior in non-democracies. The authors analyse the impact of anti-corruption blog posts on the stock prices of state-controlled firms in Russia. Their research uncovers a negative correlation between stock prices and blog posts both in the short and medium term. Moreover, they provide evidence that the relationship between stock prices and blog posts is causal, using an instrumental variable and placebo tests.
This research paper is part of a groundbreaking series of research propelled by the use of “big data” (internet-sourced & real-time data) to evaluate the role of net-activism and the blogosphere as a substitute for traditional media. This is particularly relevant for restricted democracies and authoritarian regimes where the media fails to provide the necessary accountability mechanism for public policies and politicians.
The blog activities monitored in this paper are the ones of Alexei Navalny, Russian anti-corruption activist and one of the opposition leaders who uses his blog (LiveJournal) to raise public awareness about corruption. The research question investigated by the authors is: within constrained democracies do anti-corruption blogs matter?
To answer this question the authors employed the following methodology: First, they tested the impact of blog posts on deviations in daily returns on stock prices of the flagged company (i.e., gap between the evaluation of the stock and the market price), controlling for mentions of the firm in other media and the effect of actual events not posted on this blog.
Then, aiming to test the existence of a causal relationship, the authors looked for an exogenous source of variation which may be able to affect blog posting activities but not the stocks’ prices of the companies mentioned in the LiveJournal. The chosen variable is the occurrence of Denial-of-Service (DoS) attacks (i.e. coordinated attacks that overload a server making the website inaccessible). This kind of attack prevents Navalny from posting or readers to have access to his blog but there is no reason why the stocks of the firms mentioned in the blog could be directly affected by the DoS.
The placebo test consisted of testing further the results on companies LiveJournal did not mention, which confirms the results and suggests the existence of a causal effect.
Finally, to reinforce the hypothesis that blog posts provide an alternative to traditional media and makes a difference in holding the corrupt accountable, the authors decided to study the medium-term impact (one month vs. daily impact) of these posts. They observe that the most prominent posts (the ones that mention a company name more than 5 times) have a significant effect on the stock volatility and volume of transactions.
Citation: R. Enikolopov et al., "Do Political Blogs Matter? Corruption in State-Controlled Companies, Blog Postings, and DDoS Attacks", Available at SSRN: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2153378, September 2012