Show Notes
Why can’t economists forecast better? The short answer is they don’t try.
The four-day weather forecasts of today are as good as one-day forecasts 30 years ago. Economic forecasts have been consistently lousy throughout the period with no sign of improvement. And yet there's evidence they could improve, though probably not by as much. How could they do that? By taking a leaf out of the weather forecasters' book.
We discuss Ben Bernanke's review of the Bank of England's forecasting and ask why Philip Tetlock's work on superforecasting has received so little attention. The answer is "no reason", it's just that he's not an economist. And the profession of economics puts its store in the cleverness and technical prowess of its forecasters, rather than in their ability to consistently outperform other forecasters.
If you'd like to access the video, it's here.
00:00 Trailer
00:58 Start
05:15 Introduction to Forecasting
06:27 Philip Tetlock and His Legacy
07:58 The 2008 Economic Downturn: A Case Study
11:38 The Essence of Super Forecasting
12:38 Psychology in Forecasting: Kahneman's Perspective
17:17 The Weather Analogy: Forecasting's Gold Standard
20:39 The Future of Economic Forecasting
23:09 Concluding Thoughts and Reflections
More from Uncomfortable Collisions with Reality
Welcome to another edition of Uncomfortable Collisions with Reality, where Gene Tunny and I react to a UN report on AI that invested me as the intellectual godfather of the phrase "digital public goods". I have been mulling on the quWelcome to another edition of Uncomfortable Collisions with Reality, where Gene Tunny and I react to a UN report on AI that invested me as the intellectual godfather of the phrase "digital public goods". I have been mulling on the question of public goods since well before chairing the Government 2.0 task force, realising that many of the public goods of the twenty-first century are digital. Most people haven't noticed yet, but the Ostroms' quadrant in which a public good is both non-rival and non-excluded is no longer good enough as a definition. It denies public goods as a problem solved. But Wikipedia is a public good and it's excludable. It's just that its founders wanted to open it and could raise the funds to do so as a charity. Likewise, just as free-to-air TV figured as a public good in some 1950s and 60s textbooks, so Google, Facebook and Twitter(X) are public goods even though they're funded by advertising. (And of course you're free to think that X or any of the other social networks are public BADs, but that means they're just public goods that generate negative value. We then go on to explore ways in which social media could contribute to social mobility - something I wrote about hereAnd Gene asks me some questions about AI. I think it's obviously intelligence. Chatting with Claude and ChatGPT reveals a mysterious, shape-shifting intelligence that can easily beguile us without being truly conscious. By launching careful pilots and managing the risks, we can ensure this alien technology serves its ultimate purpose of improving our real lives.The UN report mentioned is here.
Are we suffering from an austerity of the imagination? Why does modern political and policy reform feel completely stuck in a rut?In this episode of Uncomfortable Collisions with Reality, I sit down with Luca Tiratelli to diagnose the exhaustion of the modern policy landscape. We explore why radical solutions seem impossible today, distinguishing between the Overton window of vested interests and the Overton window of imagination. We unpack the creeping careerism infecting academia and the public sector, where producing content and appeasing stakeholders has replaced frank, fearless advice.Luca shares his thoughts on the necessity of utopian thinking and having a clear vision to drive reform, whilst I argue that mapping out an unknown future often distracts from making difficult trade-offs in the present. Finally, we discuss why our current electoral democracy encourages risk aversion and competitive misrepresentation, and how alternative governance models like sortition might break the gridlock and breathe new life into our institutions.
Greg Smith has a well thought out, deeply compelling and scary take on the world we suddenly find ourselves in. I thought you should hear from him and so have just recorded this conversation. I strongly recommend you check it out. (Note: this was recorded in March but only hoisted on YouTube until now - early October. But it's still very much current.
What if technology stopped keeping us apart and started bringing us back together?
Why, in an age of constant connection, are so many of us lonelier than ever?