I had the opportunity to attend the Fletcher School of Law and Diplomacy’s Managing Political Risk Conference in Boston, MA, on Saturday, and wanted to share a few of my insights this week. The conference provided an interesting window into the myths that the industry perpetuates- some of which I’ve covered in past posts (just scroll down). Those who are interested in reading my live-tweets of the event can search my Twitter feed, @MilenaRodban, using the hashtag #MPR2015.
The day-long conference consisted of panels on (R)evolutions in Political Risk Assessment; Investment and Operations in Complex Environments, and The Shifting Geopolitics of Oil and Gas. In each of the panels, there was some measure of sweeping and misleading generalizations, such as assertions that we now have “ubiquity of data” (wrong- we lack a great deal of information about some of the most intriguing places that we analyze, such as North Korea). This is one myth that the sector has a vested interest in perpetuating, given clients’ increasing desire to incorporate “big data analytics” into the products they order. Of course, you can only analyze data if there is data to analyze, and many firms claim to have proprietary methodologies for doing so. But not only do we lack a good deal of data, but we also lack the means to apply meaningful quantitative analytical techniques that can produce any useful results, if there are even any results to be found. Still others advocated for the need to institute certification procedures, because anyone can claim to do political risk now. The problem with that is that certification usually breeds standardization of thought, which we certainly don’t need in the field. What we need to methodological rigor, to adapt better analytical techniques to help our clients understand how we arrive at our conclusions.
There were also some great insights, such as a spin on an old quote by Ezra Solomon: “the only function of political risk is to make astrology look respectable.” Indeed, as those of you who follow my Twitter feed know, I often call out bad analysis based on absolutely outlandish attempts at “gaming” the future. I was surprised and happy to hear multiple admissions that a great deal of bad work is being done in the political risk sector, with clients having no way to truly judge the quality of work since it often goes only to a single client, and is never judged in public. The emphasis on risk, without equal emphasis on opportunity, was also criticized, as was the fact that most clients who pay for political risk analysis are not educated enough or capable of acting on the information they receive, and there hasn’t been much effort expended on correcting this.
Generally, the panels were optimistic about the field, and the opportunities and challenges available to those who choose to enter it. But in private conversations, many lamented the fact that though the field has done a good job of identifying the issues that need to be fixed, there are few ideas to be found for how to fix them. My biggest issue with the conference was that it was the equivalent of preaching to the choir- panels where the panelists told the attendees things we already knew. As I told a couple of the organizers, an idea for next year would be to have an interactive workshop with some consumers of political risk analysis, to get a better idea of how they view the field, what problems they see and how we could do a better job, as an industry, to address their needs.
Here’s the thing we all need to understand- the number of political risk firms really exploded in the wake of the global financial crisis of 2007-2008. Before the crisis, when money was plentiful and cheap, businesses could gamble on risky bets and write off losses without a problem. But after the crisis, when money was suddenly scarce, businesses became much more risk averse, and unwilling to take big gambles. They turned to a multitude of political risk firms for help in deciding which gambles weren’t too risky to pursue. Back then, it was easier to say “don’t go there” and be taken at one’s word, and make a good deal of money without trying too hard. But now, as money is becoming plentiful and cheap once again, the appetite for risk increasing, and the most attractive investments are in dangerous places- avoiding them isn’t an option. Yet political risk firms haven’t used their profits to invest in more advanced capabilities, and they can’t do much to meet the more complex current needs of many clients. That’s why so many political risk firms are now jumping on the big data bandwagon- it’s easier to write code to analyze “data” than build extensive local networks of contacts or open a host of global offices. Which is why, after a couple of sessions on the need to add methodological rigor, using a combination of qualitative and quantitative techniques, the mood of the conference suddenly changed during the keynote.
The keynote address featured Dr. Nassim Nicholas Taleb, author of the INCERTO masterwork, which includes The Black Swan, The Bed of Procrustes, Fooled by Randomness and Antifragile. Mr. Taleb’s lecture was without a doubt the liveliest of all, which should not surprise anyone who has ever heard him speak. Mr. Taleb is well known for his skepticism regarding the very idea that risk, and specifically black swan events, can be predicted, and he was dismissive of the idea that big data analytics could predict much of anything. After all, he argued, there were trillions of data points available before the global financial crisis, but no one was able to predict that. His dismissal of statistical analysis was not shared by some in the audience, who were eager to extol the merits of the work being done with data, but Mr. Taleb effectively ended that discussion with his explanation of the turkey problem. Relying on past data is similar to having a turkey, fed for 1,000 days straight, with each day confirming to the turkey that it is well-loved by its owner. That is, of course, until a few days before Thanksgiving, when the turkey’s life suddenly ends. All the past data could have never predicted that final day for the turkey. Could the turkey come back to life, he'd wonder how he'd missed it, but of course, he's dead, and no amount of better monitoring or forecasting would have done him any good. It's natural, when something goes wrong, to wonder if we didn't do enough of something, but what we should really wonder is if we were doing the right thing to begin with. Mr. Taleb was similarly dismissive of using models for forecasting. As he said, when we have a model, we look only for the data that the model requires to work, and don’t really look closely at a situation, often missing important clues. If you’re lost in a city, he said, an economist you encounter along your wanderings will say, I don’t have a map for this city, but I have one for another (similar to applying a model for one type of situation to another, which may not have the same characteristics). You’re better off having no map, and looking around for clues to help you find your way, than having a useless map. The keynote ended to thunderous applause by the younger attendees, and pursed lips among those practitioners of political risk who strongly disagreed with Mr. Taleb’s assertions.
After the keynote, I spoke to several attendees from a variety of fields. Many of them were obviously unhappy with the points Mr. Taleb made, accusing him of being an entertainer, whose ideas barely hold up upon close review. The problem is, of course, that many current methodologies in political risk similarly don’t hold up under close examination, and by dismissing Mr. Taleb’s arguments, we’re not doing the field of political risk any favors.
This is, of course, why I’m not at all interested in predicting the future- check back next week for my take on where political risk can go from here, if we’re brave enough to try something different.