Value of political links, a la Turca

It was only natural that the stock prices of DMG and other media companies of the Doğan Group dove on the news. After all, even though DMG is likely to fight the levy in a lengthy legal battle, there is a possibility of a huge cost item entering DMG’s balance sheet, which needs to be priced in, not only for DMG but also for other Doğan companies in the media sector. However, the sharp falls (more than 10 percent for each Doğan company and nearing 20 percent for DMG over the course of three days) are much more than simple accounting can account for. More interestingly, the mere fact that Petrol Ofisi, a non-media company in the Doğan group, also had a fall in excess of 10 percent hints that something deeper and more sinister might be going on.

That something may be the value of political links. The notion that firms with political ties might be affected from the fate of their party would be a no-brainer to almost any Turk. But actually measuring political connections is no easy task. The trick is to find unexpected political events that have no obvious financial bearing on the firm so that any abnormal return (the difference between the expected and actual return) could solely be attributed to the political event and thus to the value of the political tie. The Doğan levy unfortunately does not pass this litmus test, but Turkey has seen its share of political events in the past two years, providing fertile testing ground.

In an exercise borne out of curiosity a few months ago, I tried to measure the value of political connections by following up on the fate of companies perceived to be close to AKP by the finance community (from a small survey I conducted) during the major political events since 2007: The e-warning at the end of April 2007, Tarhan Erdem’s July 19 survey showing AKP further ahead than thought, the AKP election victory, the filing of charges with the Constitutional Court against AKP in March 2008 and the Court’s decision at the end of July.

In AKP-positive events, the average cumulative abnormal return (CAR) turned out to be around 3 percent, with the effect almost doubling in AKP-negative events.

Destruction of market value
While the results above show the value of political connections, they do not demonstrate the destruction in value of being labeled by the AKP. Therefore, urged by the events of last week, I repeated the same exercise for Doğan companies. There does appear to be an AKP effect in the reverse direction, but only for the Constitutional Court events, and much smaller in magnitude- less than 1 percent CAR. But then again, the hatchet has just been unburied, so we might have to wait for the next political shock.

I should say that this small exercise is by no means original: The first academic paper to demonstrate such effects showed that firms close to Indonesia’s Suharto were affected disproportionately by changes in the dictator’s health. Another looked at the Nazi seizure of power after the Reichstag fire, snap election and constitutional change by March 1933, when the Nazis decided to get off the democracy tram (no pun intended), and found that stock prices of firms that had tied their fortunes to the Nazis in the process surged.

It is comforting to know that AKP finds itself in such good company.



Emre Deliveli is an independent consultant and part-time lecturer at Izmir Economics University. His daily economics blog is at http://emredeliveli.blogspot.com/