The Santa Fe Institute hosted ACtioN’s Annual Risk Meeting with UBS to explore Beliefs, Narratives, and Market Structure on October 7th & 8th, 2020. Narratives and belief dynamics impact market structure, market behavior, and by extension market risk. For example, belief convergence drives crowding, contagion, and the growing prevalence of passive management. Belief instability drives market volatility. This meeting seeks to examine market risk through the lens of belief dynamics, changing narratives, and market structure.
Related Events & Projects:
The relationship between belief dynamics and market structure can be thought of as a special case of the relationship between belief dynamics and mechanism design. This latter, broader topic, is closely related to ACtioN’s ongoing discussion of Emergent Engineering.
The discussion of outcomes and belief dynamics is also closely related to the CouterBalance Applied Complexity seminar series. CounterBalance is an ongoing monthly seminar series aimed at sharing the latest research with, and facilitating a candid discussion among, scientists, practitioners, and policy makers working on issues of cyberhate, misinformation, social polarization, and radicalization.
Day 1: The Spread and Manipulation of Beliefs and Narratives | Wednesday, October 7th
Sense-Making at Scale with Simon DeDeo
Simon’s talk focused on understanding how explanations work, and exploring the systematic ways in which explanations go wrong. He puts belief in extreme conspiracy theories on a continuum with healthy epistemic practices. In so doing, Simon tries to identify the universal mechanisms that drive belief adoption, such as Unification, Concision, Validity, and Consilience. In the second half of the talk, he focuses on consilience. He first demonstrates the role of consilience in shaping the direction of scientific progress, but then shows how consilience fuels the growth of anti-social conspiracy-based internet communities such as QAnon and the Red Pill.
Computational History and the Stories Surrounding President Trump: Measurements of Timelines, Fame, Story Turbulence, and Collective Chronopathy with Peter Dodds
Peter’s talk explores how fame arises, how it drives stories, and how Twitter data allow us to observe the dynamics of these processes. Peter started by highlighting his group’s efforts to link their Twitter-based happiness measure (hedonometer.org) to the topics that are being most frequently discussed on Twitter (storywrangling.org). This motivated a discussion of the role, methods, and importance of stories and storytellers. Peter paid particular attention to the role communication network structure plays in enabling different stories to superspread. In the second half of the talk, he used these frameworks and analytic tools to compare the fame dynamics of Donald Trump and Joseph Biden in the build up to the 2020 presidential election. A full list of data sources Peter referenced can be found here.
This first practitioner panel focused on some of 2020’s most important issues related to market structure, including new data sources and how they impact the future of trading. Moderated by Vlad Khandros of UBS, this panel includes Hubert De Jesus (Global Head of Electronic Trading and Market Structure, BlackRock), Jonathan Kellner (CEO, Members Exchange), and Doug Cifu (CEO, Virtu Financial).
Day 2: Modeling and Predicting the Impact of Beliefs and Narratives in Markets | Thursday, October 8th
The Complexity Economics Revolution with J. Doyne Farmer
Treating distinct investment strategies as different species, and the wealth invested in each strategy as its population size, Doyne introduces an ecological and dynamic model of financial markets. Notably, he demonstrates that efficacy of any investment strategy is a function of relative “population sizes” of the other investment strategies. For most initial distributions of wealth across the different investment strategies, the system is drawn towards an equilibrium point. However, the path towards that equilibrium is highly non-linear. As a result, most systems do not reach the equilibrium in the first 200 years! This long disequilibrium trajectory predicts volatility spikes and mispricing, which are hallmarks of market malfunction.
Culture, Conformity, and Strategic Uncertainty with Willemien Kets
Willemien provides a mathematical formalization of the process by which culture enhances coordination at the expense of exploration. She examines which problems reward, and which problems punish, groups with different levels of cultural conformity. Specifically, she identifies the criteria under which the uncertainty produced by minimal cultural conformity helps low conformity groups avoid the inefficient lock-in often associated with coordinated decision-making. Willemien then applies this framework to explain a range of phenomena, including the failure of strong culture firms to adapt to severe exogenous shocks and the success of non-neurotypical analysts in helping their teams’ identify new solutions.
This second practitioner panel focused on how firms use data and diversity to navigate a world of self-destructive beliefs and manipulative narrative frames. Moderated by SFI Applied Complexity Fellow Joshua Garland, this panel included Mani Mahjouri (CEO & CIO, Blueshift Asset Management), Jeffrey Young (Founder & Chief Economist, DeepMacro) and Dario Villani (CEO & Co-Founder, Duality Group).