Wrangling Uncertainty
Announcing the Two Lanterns Platform: Part 2
In the first part of this series analyzing the political risk industry (and why the new Two Lanterns Platform may help) I spoke about the troubles with the Long Tail.
In this newsletter, I’m going to talk about a bigger problem - perhaps the central dilemma that political risk faces: uncertainty.
Predicting the Future is Hard
Most professional services are based around knowing the answer to a client’s questions. If you go to an accountant, you have the question of how much in taxes do I owe. You don’t want the accountant to use the word “probably” in his answer. That’s a recipe for an audit.
But accountant don’t vacillate, because they are professionals and know the tax code. Although it takes time and skill to become an accountant, there is a body of knowledge out there that can be learned. We aren’t going to accountants and asking them to predict what our taxes will be fifteen years from now. No one would expect them to know the answer to that.
But political risk analysts are in the job of predicting what might happen in the future. Even when we are experts on the topic at hand, our predictions are not 100% certain. Those are the questions we’re faced with.
When it comes to making predictions, there’s no easy way to do it.
Yet it’s the core of the job.
How to manage amid uncertainty
There is no silver bullet that can help us predict the future. There are, however, some best practices. In my mind, there are three prerequisites to make a good forecast:
You need to have knowledge of the subject.
You need to include multiple perspectives in a forecast.
You need to blend Steps 1 and 2 together.
Let’s look at a simple example. What is Joe Biden going to do when he becomes the president?
In order to answer this, I need to know what Biden’s agenda is (Step 1: Knowledge). I should then read what different people think Biden might do to ensure that I’m not missing anything (Step 2: Gather perspectives). I then think it through and come to an opinion of my own (Step 3: Synthesize).
Doing this, often for much harder questions than Biden’s plans, and doing so affordably, is the crux of the political risk industry.
In some cases, Step 1 is the big hurdle. What’s the agenda of Bhutan’s finance minister? There aren’t a lot of people with that knowledge. Even the basics go far in answering that question.
In others, Step 2 is the challenge. Maybe you know only one expert on Bhutanese politics and have to hope they aren’t biased or you scramble to find other information.
For all, how to perform Step 3 defines their firm and approach.
Mechanisms for synthesizing
If you look at the marketing materials for political risk firms, many focus on why they are best at Steps 1 and 2. They will speak about “deep regional knowledge” or “extensive expertise in the sector.”
That is important, but it doesn’t tell us anything about how they perform Step 3, other than the assumption that their expertise helps them synthesize all that information in their own heads.
Some firms leave this as the implicit method of synthesis without paying much more attention to it. Others are more explicit that there is an explicit process for synthesis. Some new firms use innovative approaches like prediction markets, forecasting tournaments, or data services.
The pros and cons of each approach can be debated elsewhere. But I stress this to show that even though political analysis can be simplified into a basic three step process, there is no agreement in the industry about the precise best method for each step. There are, however, some best practices.
The Big Hole in the Process
Most everyone in political risk would agree on the following principles for each step.
More knowledge is better than less knowledge.
A variety of higher quality perspectives are better than a few lower quality perspectives.
There are some techniques better suited to the problem at hand than others.
This means that, all else being equal, to forecast Biden’s agenda, we would want to know as much as possible about his plans. We’d want to hear from a number of knowledgeable people on the topic and not a single cable news pundit. And we’d probably want a lengthy and nuanced report rather than to create a prediction market.
These principles highlight one of the political risk world’s blind spots. By not including a client’s knowledge and perspectives on an issue, we are missing out on additional resources for Steps 1 and 2.
This is completely understandable as a consulting model, I should stress.
The consulting model is based on a client paying someone else to tell them what they don’t know. If the consultant only repeated back what the client told them, they wouldn’t be doing their job. And for anyone selling a subscription, the logistics of getting client feedback on a topic can be so difficult as to be prohibitive.
From a broader perspective, there’s even more knowledge left untapped. If you’re writing a report on the Biden agenda, perhaps the client you’re writing for doesn’t know much. But all of your clients collectively probably know a lot. Being unable to access that knowledge and perspective easily would be a huge help to the process.
This means that we’re currently trying to forecast the future without access to a lot of knowledge and perspective that could make our forecasts better.
A toolbox, not a hammer
So there is something we’re missing on Steps 1 and 2. But what about Step 3?
Approaches vary from firm to firm. Some are specialized in writing reports. Others have built a subscription product. Others build prediction markets. It is necessary in a competitive market to have a comparative advantage in some product. But that can lead to using that technique in situations where it’s not the most applicable.
Pre-existing relationships, brand reputation, and topic expertise can lead a client to choose a firm/expert to work with, despite a synthesis technique that is not the best for this situation. There’s nothing inherently wrong with that - it can still lead to useful products - and many companies have multiple service offerings to meet this challenge.
The downside, however, is that clients might be using firms that don’t have the best approach for the question. Consultants are missing out on clients that they would be the best fit for. As with any market that contains frictions (which I’ll speak more about in the next newsletter), there are missed opportunities.
How Two Lanterns addresses this
The Two Lanterns Platform seeks to improve all three steps.
We tap into client knowledge by letting your team contribute to the process.
We allow you to see the perspectives of the whole community.
We let you choose the technique that’s best suited for your question.
Admittedly, it is still difficult to predict what will happen in politics. This platform should not be seen as a panacea for all political risk questions.
Our hope is that the platform will break down logistical barriers that had previously hindered the political risk process and result in the aim of all of our work: a better understanding of what’s going to happen.
Please be in touch if you’d like to join the platform as one of its early users and let me know if you have thoughts on how to improve the product.