In August 2018 we resumed publishing our monthly journal. From 1997 to 2011, we provided independent analyses and insights each month on macro scenarios, asset class valuation, portfolio construction, decision making in the face of uncertainty, and downside risk management to investment managers, financial advisors, family offices, and sophisticated individual investors from around the world.
They used The Index Investor for three purposes:
To improve their investment process and portfolio construction;
To improve their risk management process, and obtain early warning of dangerous situations, such as the ones we provided ahead of the 2008 and 2000 crises;
And to improve their client communications on matters related to macro uncertainties and the risks and opportunities they create.
In 2011, we suspended operations (but left our website and all our briefings and back issues up) when our key writers moved on, either to other publications, or, in the case of contributing editor Tom Coyne, to spend four years on the Good Judgment Project team, which won the Intelligence Advanced Research Projects Activity’s forecasting tournament, with forecast accuracy that was more than 50% better than the tournament's control groups. The team's experience is described in Professor Philip Tetlock's book, “Superforecasting.”
Tom later co-founded Britten Coyne Partners, a firm that provides education courses and consulting services on strategic risk management and governance. It helps clients to better anticipate, more accurately assess, and adapt in time to emerging strategic threats to their survival.
Our new Research Library is also free. Investors can browse our curated content on a wide range of issues affecting medium and long-term asset class valuations, including technological, economic, environmental, national security, social, demographic, and political trends and uncertainties, as well as potential "grey swan" wildcards like environmental, infectious disease, cyber, and large-scale electromagnetic events.
So why are we publishing again?
As former Bank of England Governor Mervyn King has so aptly described it, we now live in a world of radical uncertainty, that has become a much more dangerous place for investors. Former US Treasury Secretary and Goldman Sachs Co-Chairman Henry Paulson shares this view, noting that "we are living in an age of unprecedented risks."
In recent years information overload, anxiety, and fear have all grown exponentially worse, and synthesizing meaning and insight from everything that is happening around us has become harder than ever.
Perhaps more importantly, the exponential increase in connectivity the world has seen over the past 20 years has made global economic, political, and financial systems far more complex, and therefore uncertain. Research has demonstrated that in such circumstances, evolution has primed human beings to become more prone to conformity and to place more emphasis on imitating what others are doing (so-called "social learning").
Paradoxically then, as uncertainty increases, people — including many sophisticated investors — are more likely to become attached to a smaller (not larger) number of dominant narratives and expectations about the future.
This makes the highly interconnected systems very vulnerable to small changes in information, feelings (fear is transmitted between human beings far more quickly than greed), and behavior that are all rapidly transmitted to many people, amplified by algorithms of various kinds, and whose impact is therefore increasingly both rapid in speed and exponential in size.
In the complex, uncertain world we live in today, accurate prediction depends on our ability to deliberately integrate forecasts from a wide range of sources that are ideally based on different underlying methodologies and information.
Pilots and air traffic controllers use a term — "having the bubble" — to describe having a sense of how a complex dynamic situation is evolving in real time, and maintaining confident control over your decisions and actions within it. Another term for this is "situation awareness."
We're publishing again to help our subscribers maintain their situation awareness in our rapidly changing, uncertain, and non-linear world, where threats to corporate and investment goals and strategies abound.
Here's what one of them recently wrote to us: "I am delighted to be able to get your analysis again. We get everything from Wall Street, and they all seem to be saying the same thing. Your take is greatly appreciated."
Our mission is to help our subscribers develop an edge in anticipating downside risks, accurately assessing them, and adapting to them in time to avoid large losses.
Our forecasting methodology focuses on how endogenous system dynamics can cause regime changes, even in the absence of exogenous shocks.
Specifically we focus on key system stocks (e.g., debt levels), and how ongoing system flows (e.g., government deficits) can eventually cause them to exceed critical thresholds and produce non-linear effects.
Our forecasting methodology analyzes stocks and critical thresholds in five areas whose effects tend to follow a rough chronological sequence, from technological change to economic, national security, social, and political changes and effects. A graphical representation of this causal framework is shown below. What is not shown are the many feedback loops between these issues, that create what Eric Beinhocker has termed a "complex reflective system", and underlie Rudiger Dornbusch's famous quote: "Things take longer that you think they will, then happen faster than you thought they could."
Our forecasting process is based on tools we've developed and learned from many sources over the years, including the Good Judgment Project and Peter Pirolli and Stuart Card's information foraging and sensemaking model (see their classic article, "The Sensemaking Process and Leverage Points for Analyst Technology”):
Here's what subscribers get in each monthly issue of the new Index Investor:
(1) Estimated asset class over/under valuations, and updated market stress indicators, using the same methodologies we've used in the past.
(2) Narrative forecasts and quantitative probability estimates for macro system and financial market regime changes over the next 12 months. In our forecasting methodology, possible regimes include the Normal Regime, where equities perform well; a High Uncertainty Regime, where negative asset class valuation changes of 20% or more can quickly occur; a High Inflation Regime, and a Persistent Deflation Regime. Our forecasts take the form of a contingent probability tree, an example of which is shown below. Our goal is to help subscribers have a good sense of what could happen not just one, but two or more steps ahead.
(3) A comprehensive, chronological "Evidence File", that contains two kinds of high value information that we have used to update our monthly forecasts. The first are "indicators" that cause us to either increase or decrease our uncertainty about the values of different parameters in our mental model of the complex macro system. The second are "surprises" that increase our uncertainty about the structure of that model. Evidence is categorized by month, with surprises highlighted, and divided into separate sections covering developments in technology, the economy, national security, society, politics, financial markets and investor behavior, and three potential "wildcards", which cover developments in energy and the environment, health and infectious disease, and cyber and electromagnetic events.
The Evidence File helps subscribers to better understand the chronological trajectory and dynamics of developments in each of these critical areas.
(4) In between monthly publications, we will publish flash updates — on our blog, via email, and via our Twitter @indexllc — if and when we obtain high value information that results in a substantial change to a forecast probability.
(5) A feature article providing an in-depth analysis of either a key macro-uncertainty (e.g., how close the system is to one or more critical thresholds) or an aspect of making good investment decisions in the face of complexity and uncertainty. These articles typically synthesize a broad range of academic research and practitioner experience to provide thought provoking insights about critical issues facing investors and their advisors. Put differently, in a complex system that is constantly adapting and evolving, the accuracy of statistical or machine learning based forecasting methods declines exponentially as the time horizon lengthens, as the historical data set on which they were trained bears less and less resemblance to the distribution of outcomes the system is likely to produce in the future. Under these circumstances, accurate forecasts beyond the short term must be based on causal and counterfactual, and not just statistical thinking. We know that our subscribers' attention is their scarcest resource, so our goal is to maximize the return on the time you invest each month reading The Index Investor.
In a complex system that is constantly adapting and evolving, the accuracy of statistical or machine learning based forecasting methods declines exponentially as the time horizon lengthens, as the historical data set on which they were trained bears less and less resemblance to the distribution of outcomes the system is likely to produce in the future.
Under these circumstances, accurate forecasts beyond the short-term must be based on causal and counterfactual, and not just statistical reasoning.
In order to increase accuracy, our forecasts can (and should) be combined with forecasts subscribers obtain from other sources. Ideally, these forecasts will be based on substantially different underlying information and methodologies.
We provide subscribers with tools that enable them to combine forecasts to improve predictive accuracy, including the “extremizing” methdology used by the Good Judgment Project.
In sum, our goal is to provide you with an ongoing understanding of the complex dynamics driving asset class valuations and returns, as well as forecasted downside risk probabilities for broad asset classes, that are based on an explicit methodology. which facilitates their combination with forecasts from other sources. We also provide custom research as well as speaker services on how to increase forecast accuracy, understanding the differences between active, passive, and index investing, and how to overcome the individual, group, and organizational obstacles to making good decisions in the face of uncertainty. Our speaker offerings include seminars for advisors' clients and speeches for larger groups. Click here to learn more.
You can download a free copy of a recent issue to get a better feel for what we provide subscribers each month.
You can also send us feedback about how to improve The Index Investor to better meet your needs. Or you can subscribe (our first issue for subscribers will be published in early September).
We believe that in addition to diversification across broad asset classes, avoiding large downside losses is also critical to achieving long term investing goals.
Hence, our focus is on identifying asset classes that are dangerously overvalued today, and, at a longer-term horizon, identifying emerging threats that could cause substantial changes in uncertainty and asset class valuations.
Our research is based on the application of complex adaptive systems theory and advanced forecasting methods to macro factors such as technological, economic, environmental, military, social, demographic, and political trends and uncertainties.
We believe that financial markets are filled with positive feedback loops that produce nonlinear effects through the interaction of competing strategies (for example, value, momentum, and passive approaches) and underlying decisions made by people with imperfect information and limited cognitive capacities who are often pressed for time, affected by emotions, and subject to the influence of other people.
While attracted to equilibrium, we believe that financial markets are best described as adaptive systems that never reach it. When they are operating far from equilibrium, substantial over and undervaluations are the usual result.
In contrast, traditional mean-variance optimization is based on an underlying assumption that markets generally operate in or close to equilibrium. This is why this approach often produces disappointing portfolio results.
Our benchmark model portfolio is equally allocated between broad asset classes in order to capture the underlying system and social drivers of financial market returns.
To achieve long-term portfolio goals, avoiding the large losses that follow substantial overvaluations is critical.
This can be achieved through a combination of systematic and episodic portfolio rebalancing that is driven by the extent of asset class over and undervaluation.
In a nutshell, the active investor believes that he or she can regularly generate (or choose fund managers who can generate) returns that are above the returns generated by a passive benchmark portfolio, due to a mix of information and/or forecasting skill (i.e., an "edge") that is superior to that possessed by other active investors.
In contrast, the passive investor wants only to match those benchmark returns at the lowest possible cost.