The Index Investor
April 2019

Key Takeaways

Many asset classes remain significantly overvalued.

Market stress indicators remain weak. Asset class return indicators imply that aggregate expectations favor either a return to the Normal Regime or a transition to the High Inflation Regime. We believe this conventional wisdom is wrong.

We have not changed our estimated probability that 12 months from now the macro system will still be in the High Uncertainty Regime (60%) or the Persistent Deflation Regime (30%).

We are closely watching housing market developments, given that sector’s 33% weighting in the US Consumer Price Index. Given the weight of additional evidence received this month, further signs of accelerating housing market weakening will cause us to increase the estimated probability that within the next 12 months we will transition to the Persistent Deflation Regime.

This month’s feature article presents Conviction Narrative as a complement to Bayesian forecasting methods. We present four different narratives that argue why, in 2022, the macro system will be in each of our four possible regimes.

Asset Class Valuation and Momentum Indicators (@29March19)

Note: The language we use to describe our estimated likelihood of asset class over or undervaluation is based on US Intelligence Community Directive 203 on Analytic Standards, which includes the following table:

Market Stress Indicators (@29Mar19)

Market Stress Indicators: Forecast Discussion

We view financial markets as a complex adaptive system. The size of changes generated by such a system follows a power law rather than a normal (Gaussian) distribution. The critical point is that large changes are much more common in complex adaptive systems than most people’s intuition leads them to believe.

While predicting the behavior of complex adaptive systems remains far more art than a science, various researchers have found that large changes in such systems are often preceded by subtle warning signs, as stress accumulates within them. While this research is not definitive, we believe that five warning signs are worth monitoring as potential indicators of growing stress within financial markets that could suddenly give rise to large changes in asset class valuations.

Our first indicator is the month-to-month autocorrelation of broad asset class returns (i.e., the relationship of this month’s returns to last month’s). A system under increasing stress loses resiliency, causing it to take longer to recover from perturbations; hence, autocorrelation increases as it approaches a critical transition (see, “Early Warning Signals for Critical Transitions” by Scheffer, et al).

The one-month autocorrelation of returns for the broad asset classes we monitor was about unchanged from last month. This indicates that financial markets remained less ordered in March, and thus potentially further away from a critical transition point (which would most likely be accompanied by sudden and substantial changes in asset class values) than they were last month.

The second market stress indicator we monitor is the Economic Policy Uncertainty Index published by the Federal Reserve Bank of St. Louis (via its FRED economic database), which is based on research by Baker, Bloom, and Davis (see their paper, “Measuring Economic Policy Uncertainty”). The index is based on automated text analysis of leading newspapers and magazine publications, to identify the frequency with which words and phrases are used that indicate uncertainty.

In humans’ evolutionary past, when uncertainty increased the probability of survival was enhanced by staying close to a group. All of us still have that instinct. Research has found that as uncertainty increases, we have an unconscious bias towards higher conformity of our own views with those of a larger group (i.e., reduction in cognitive diversity). Behaviorally, heightened uncertainty induces more “social copying” of others, likely due to both conformity bias and the rational belief that others may be acting on the basis of superior information. This increase in conformity and copying makes a social system more ordered as uncertainty increases, and also reduces its responsiveness to perturbations (i.e., increases autocorrelation) because of delays in the social copying process.

The key point is that increasing uncertainty induces more, not less order in social systems, and in so doing primes them for sudden non-linear change.

Based on our continued research into the insights this index can provide, in 2019 we are focusing on the number of days, in the previous 30 days, that this index was in the top quartile of all values since the index series begins at the start of 1985. We then compare this statistic to the full set of rolling 30-day periods, and calculate its percentile at the end of the most recent month. At the end of March, our rolling 30 day count of top quartile values was in the 48thth percentile – a slight increase from last month, which indicates the macro system is still not strongly primed for sudden, non-linear change.

Our third market stress indicator is the spread between the yield on AAA rated bonds and the 10-year US Treasury. This is a proxy for the level of investor concern about financial system funding liquidity. At the end of March 2019, this spread stood at 1.21%, (the 48th percentile since the series began in 1983), up only slightly from last month’s 1.13% spread. This indicates essentially no change in market stress.

Our fourth market stress indicator is the yield spread between speculative BB rated bonds and the ten-year US Treasury. Throughout history, excessive credit growth has been a root cause of many financial crises. An indicator of such growth is falling credit spreads, particularly in the case of riskier borrowers. In contrast, rising BB spreads indicate growing investor concern about the consequences of such growth, and the financial distress lower rated companies could experience in an economic downturn. At the end of March 2019, this spread was 2.39% (21st percentile since the series began in 1996), unchanged from last month. This is a very low level for this late in what is already an exceptionally long period without a serious economic downturn. As such, it likely indicates the further buildup of hidden stresses in credit markets.

Our fifth market stress indicator is what we term the “political risk premium” that is implicit in the price of gold. Our starting point for estimating this premium is the three different roles that gold plays. First, gold is a store of value in a world of fiat currencies. When the rate of money supply growth exceeds the growth of nominal GDP, gold’s price should increase to maintain its purchasing power. Between 2007 and 2017, the US money supply (M2) grew by about 86%, while nominal US GDP grew by 35%. The stock of gold grew by 18%, based on mine production over this period. We therefore infer that 33% of the increase in the price of gold represented the maximum potential gold price change that could be attributed to a desire to hedge inflation risk (86% less 35% less 18%).

Second, gold is a unit of account. We take this to mean that the annual change in GDP expressed in terms of physical gold (i.e., nominal GDP divided by the price of gold) should equal the change in real GDP calculated using the GDP price deflator to account for actual inflation over the period. A key challenge is the point at which to start this calculation.

We chose the price of gold in 1995/1996. In that period, the change in real global GDP measured using the IMF’s price deflator just about equaled the change in GDP measured in terms of physical gold. We interpret that coincidence as indicating that at that point in time, concerns about future inflation and political risk were minimal, and the change in the price of gold was mostly driven by its role as a unit of account. We calculated a subsequent series of gold prices that would produce the same change in “gold GDP” as the actual real GDP as calculated by the IMF. Between 2007 and 2017, “gold as a unit of account” warranted a 21% increase in its price.

Gold’s third role is as a hedge against inflation and what we term “political disaster” risk. We subtract the 21% estimated compensation for actual inflation from the 33% “gross” inflation risk hedge to derive an apparent 12% increase in the gold price that reflected the true risk premium to hedge against possible future inflation. However, between 2007 and 2017 the price of gold actually increased by 81%. This implies that 48% of this (81% less 21% less 12%) represented a premium for some other type of uncertainty at the end of 2017. The interesting question is the nature of the uncertainty for which gold is believed by some investors to be a superior hedge than traditional ports in a storm like short-term US government securities, or similar securities issued by other developed countries.

The logical inference is that the uncertainty in question must reflect a situation in which short term US Treasuries would be a less effective hedge than gold. This could be a world of widespread hyperinflation, capital controls, and/or radical changes in nations’ governments (of course, this would also imply a preference for investing in gold coins rather than bullion, as while the latter may be a store of value, it is far less convenient as a means of paying for transactions).

To put this in further perspective, this gold price “disaster risk” premium sharply increased from 2008 to 2012, then declined before sharply increasing again after 2016. Arguably, a significant part of the former increase reflects concerns about the potential inflationary consequences of dramatic quantitative easing by central banks. But this is not likely to be the case after 2016.

Over the last month, the price of gold fell by (2.59%). Since the end of 2017, it is now down by (0.41%). Therefore, on a rough approximation, the political uncertainty premium at the end of February 2019 stood at about 48% (48% - .41%) compared to a low of about 39% at the end of August 2018. On balance, this is another indication of falling market stress.

To conclude, four of our indicators – asset class return autocorrelation, top quartile values for the Economic Uncertainty Index, the AAA spread over Treasuries, and the price of gold – point to a relatively low level of underlying market stress this month. Our metric for the BB spread is also consistent with this conclusion, but seems exceptionally low for this late stage of what has bee an extended, if relatively weak, recovery.

Macro Regime Forecast and Implications for Asset Class Values

Perhaps the wisest insight I’ve come across in 40 years of forecasting is this quote by the late economist Rudi Dornbusch: “Crises take a much longer time coming than you think, then happen much faster than you would have thought.”

While there were certainly new pieces of high value information this month (as you can see later in this report), their cumulative weight was not sufficient to cause a change in our forecast regime probabilities this month. To be sure, there are more indications that the global economy’s three demand motors (the US, EU, and China) are simultaneously weakening, which is a distinctly negative sign. Moreover, new articles by Larry Summers and Claudio Borio of the Bank for International Settlements made the case, respectively, that “secular stagnation” is well-underway in the private sector, and that it’s negative impact will likely be amplified by financial market conditions, including high leverage and rates that are near the zero lower bound.

Finally, and perhaps most importantly when it comes to the persistent deflation regime, Daniel Alpert published a new paper arguing that house prices are once again above levels that have historically proven to be sustainable. Together, these information inputs would seem to argue for an increase in the probability that we will be in a persistent deflation regime 12 months hence, and a decrease in the probability of remaining in the current high uncertainty regime, which was once again supported by many new pieces of high value information. On balance, we decided to wait for another month of new evidence before changing our current regime probabilities.

Forecast Methodology

The focus of our monthly macro forecast is twofold. First, the probability of a change in financial market regime that causes changes of 20% or more in asset class valuations over the next year. Second, contingent on such a change taking place, the probability of a subsequent transition to other regimes.

Our analysis focuses on four possible macro regimes: (1) Normal Times, where equity asset classes perform well; (2) a High Uncertainty regime that is usually short and transitory, where asset classes like short-term government bonds perform best and equities suffer significant declines; (3) High Inflation (which we deem 5% or more, year-on-year), where commercial property, real return bonds and other traditional hedges are favored; and (4) Persistent Deflation (a year-on-year decline in the US CPI), which up to now has only been seen in Japan, and in which the relative performance of different asset classes remains uncertain, but will likely favor high quality bonds and the consumer staples equity sector.

Our forecasting methodology is derived from our experience on the Good Judgment Project, as described in the book, “Superforecasting” by Gardner and Tetlock, as well as a range of other sources, from the intelligence community to systems dynamics and complex adaptive systems to statistics and political economy.

We start with base rate/reference case data about the historical probability of large changes in equity and bond valuations. We then analyze the current situation from both a quantitative and qualitative perspective. In the latter, we focus on the key endogenous drivers of macro regime change, including technological, economic, national security, social, and political trends and uncertainties. We also focus on three potential sources of exogenous shocks that could also produce a macro regime change, caused by environmental, disease, and cyber related events.

While most of our attention typically focuses on various flows (e.g., economic growth, change in the price level, sales, earnings, job creation, etc.), endogenously caused regime changes result when those flows push key stocks beyond a critical threshold or tipping point, often setting off non-linear reactions across multiple areas. As noted by Hyman Minsky and others, a classic example is the steady accumulation of outstanding debt until it reaches the point where it can no longer be serviced and triggers a crisis.

Base Rate Data

Since the end of World War Two, there have been fifteen months where a downturn in the US equity market began that eventually reduced asset class value by 20% of more. That is a hazard rate of about 1.75% per month. Put differently, in any given month there is a 98.25% probability that a 20%+ downturn won’t occur, or, in a given year, an 81% probability.

However, as the time without a 20%+ downturn extends, the compound probability that one will not occur shrinks. At the end of August 2018, it is more than nine years since the last equity market decline of 20% or more. The probability of that happening is only 15%.

To estimate the base rate for a 20% fall in bond prices (which historically has been caused by a sharp increase in inflation, as we saw in the late 1970s and early 1980s), we analyzed monthly historical AAA bond yields since 1919. For consistency, we used them to calculate the price of a ten-year zero coupon bond. We then calculated the probability of a price decline of 20% or more over three different holding periods: 12, 18, and 24 months. In any month, the annualized probability of a decline of 20% or more over the subsequent 12 months is 12%; over 18 months, 20%, and over 24 months, 25%.

The Current State of Quantitative Regime Predictors

Our quantitative methodology focuses on the level and change in three-month returns, over the most recent and previous three-month periods, for those asset classes which should perform best under different regimes.

As you can see in the following table, based on three month returns to the end of last month, this analysis indicates that the balance of expectations across all four regimes was in rough balance, which we consider to be another indication of high underlying uncertainty.

That said, comparing the change between the two three month periods, positive momentum is strongest for both the Normal and the High Inflation Regimes.

Like Professors Andrew Lo, Doyne Farmer and others, we regard financial markets as a complex adaptive system (CAS), that exist as part of a larger macro system comprised of other CAS between which there are multiple feedback loops. These other systems include those that produce technology innovations, and economic, environmental, national security (including cyber), social, demographic, and political outcomes.

We also find that these systems tend to operate and generate effects in a rough chronological sequence, albeit with many feedback loops between them. The following chart highlights that the changes we observe in different areas at any point in time are actually part of a much more complex evolutionary process.

While most media coverage of these systems focused on flows (e.g., the size of the government deficit), rapid non-linear change in complex adaptive systems is often caused by a key stock (e.g., the amount of outstanding government debt) exceeding a critical threshold.

The next table highlights the key macro system stocks that we monitor.

In the next section, we will discuss information received over the past month that is related to these stocks, and which we believe is significant to our assessment of the probabilities that a critical threshold will be reached and a regime change will occur. We will conclude with our estimate, at the end of this month, of how close the macro system is to these critical thresholds, and the implications for financial market regime change probabilities.

In our methodology, we classify new information as significant and highly valuable if either it (1) is an “indicator”, which reduces or increases our uncertainty about the value of a parameter in our mental model for making sense of the dynamic macro system, or (2) it is a “surprise” which increases our uncertainty about, and causes us to revaluate the structure of our mental model.

How Close is the Macro System to One or More Critical Thresholds?

As we have noted, the macro drivers of financial market regime changes typically follow a rough chronological sequence, from technology to economic, security, social, and political causes and effects. Yet there are many feedbacks loops between them, creating complex root causes for many of the critical thresholds we have identified.

Understanding the time dynamics in this complex system is critical to avoiding substantial downside investment risk.

We use the UK Met Office Warning Model to communicate our assessment of these time dynamics. We estimate the time remaining before a critical macro system threshold is reached that could trigger a regime change, which is usually accompanied by substantial changes in asset class valuations.

The model uses three increasingly serious levels of warning, from “Be Aware” (condition yellow), to “Be Prepared” (condition orange), to “Take Action” (condition red).

For our purposes, we denote as “Be Aware” (yellow) critical thresholds that we assess to be three or more years away. We estimate that “Be Prepared” (orange) thresholds could be reached within 1 to 3 years. “Take Action” thresholds are very likely to be reached within one year.

Given their nature, we also note that in our three “wildcard” areas (Environment and Energy related; Disease and Human Caused Bioevents; and Cyber and Electromagnetic Events), our forecasts have higher levels of uncertainty.

The following charts summarize our current estimate of the time remaining before different critical thresholds will be reached.

At the highest level, we believe the global macro system can be in one of four states, based on its degree of order versus disorder, and degree of social cooperation versus conflict. A very coarse-grained reading of history suggests that these states evolve in a predictable cycle, from ordered/cooperative, to disordered/cooperative, to disordered/conflicted, to ordered/conflicted.

We believe that the system is currently in its most uncertain state, characterized by high degrees of underlying disorder and social conflict, both domestically and internationally.

Our current forecast question is this: what is the probability we will either remain in the High Uncertainty Regime or transition to another regime over the next twelve months?

We currently estimate there is a 60% chance of remaining in the High Uncertainty Regime over the next 12 months, which will see the commencement of what promises to be a tumultuous US presidential campaign, and, one hopes, resolution of what has become an extended struggle over what form Brexit will take, if it happens at all.

We also conclude that over the next 12 months, the probability of returning to the Normal Regime is slight, at 5%, as is the probability of entering the High Inflation regime over the next 12 months, at only 5%.

We estimate that the probability of entering the Persistent Deflation Regime over the next 12 months is 30%.

While we expect the High Uncertainty Regime will produce declines of 20% or more in equity asset classes, it is unlikely that any other asset class will experience a gain of 20% or more. We estimate there is a roughly even chance that gold could be the exception, with that increase heavily tied to continued global confidence in the US government and economy. While the apparent inflation and political uncertainty premia in the gold price today are high relative to the last 25 years, they are still below the peak reached in 2012, and it is possible that herding in the face of increasing uncertainty could produce 20% price gains.

Pre-Mortem Analysis

One of the most important forecasting disciplines is to ask yourself why your forecast could be wrong. Dr. Gary Klein’s research has shown that a very powerful and insightful way to do this is via a “pre-mortem analysis.” This method asks you to assume that it is a point in the future, and your forecast has been proven wrong (or your strategy or company has failed). You are then asked to look backward from this imagined point in the future, to explain why you failed, what you missed, and what you could have done differently to avoid your fate.

The pre-mortem method takes advantage of the fact that humans reason much more concretely and in more detail when explaining the past than they do when trying to forecast the future.

So let us assume that it is one year from now, and our current forecast has turned out to be wrong.

How did this happen? What developments did we fail to anticipate?

• At the end of March 2019, the leaders of the world’s three major powers – Xi Jinping, Donald Trump, and Vladimir Putin are all facing weakening economies and declining political popularity. History teaches us that this can lead to increased “foreign adventurism” to distract the public from worsening domestic conditions, as a nation rallies around its leader in a period of heightened external conflict. Should such a conflict develop between China and the United States, or between Russia and one or more European countries, it would generate a sharp increase in uncertainty that would likely cause an equally sharp economic slowdown and, given high debt levels, speed the arrival of the Persistent Deflation Regime.

• As described in this month’s feature article, while remote, a supply side shock of some type could produce a sudden increase in inflation – the most likely scenario being a reduction in oil supplies due to a sudden intensification of the simmering conflict between Iran and Saudi Arabia.

• While we believe it is highly improbable, we can envision a scenario in which for a range of possible reasons, both Xi Jinping and Donald Trump leave their current roles, and are replaced by leaders who are more committed to lessening conflicts both between China and the United States and in the international system as a whole. This would likely provide a strong boost to confidence (and thus lead to an equally strong reduction in uncertainty). Whether this would also create an opening for a reduction in domestic political conflict in the United States, and thus progress on policy reforms to address weak growth and rising inequality isn’t clear.

Note: Combining this Forecast with Others and Extremizing the Result Should Increase Predictive Accuracy

Research has found that three steps can improve forecast accuracy. The first is seeking forecasts based on different forecasting methodologies, or prepared by forecasters with significantly different backgrounds (as a proxy for different mental models and information). The second is combining those forecasts (using a simple average if few are included, or the median if many are). The final step, which significantly improved the performance of the Good Judgment Project team in the IARPA forecasting tournament, is to “extremize” the average (mean) or median forecast by moving it closer to 0% or 100%.

Forecasts for binary events (e.g., the probability an event will or will not happen within a given time frame) are most useful to decision makers when they are closer to 0% or 100% than the uninformative “coin toss” 50%. As described by Baron et al in “Two Reasons to Make Aggregated Probability Forecasts More Extreme”, forecasters will often shrink their probability estimates towards 50% to take into account their subjective belief about the extent of potentially useful information that they are missing.

When you average multiple forecasters’ estimates, you are including more information, which should increase forecast confidence and push the mean estimate closer to 0% or 100%. However, this doesn’t happen when you use simple averaging. For this reason, forecast accuracy is increased when you employ a structured “extremizing” technique to move the mean estimate closer to 0% or 100%.

You can download an extremizing model from our website to use when combining the forecasts you use in your decision process.

The extremizing factors in our model are those that the Good Judgment Project found maximized the accuracy of combined forecasts. Note that the extremizing factor is lower when average forecaster expertise is higher. This is based on the assumption that a group of expert forecasters will incorporate more of the full amount of potentially useful information than will novice forecasters.

Human beings, whether individually, in small groups, or in larger organizations, inevitably face the challenge of how to overcome doubt and act in the face of uncertainty.

In point of fact, this is a so-called “dual search problem”, in that it requires the decision maker to simultaneously explore both the external “possibility space” of alternative future environmental trajectories, and the internal “policy space” of alternative courses of action to achieve their goals, with limited resources and usually in the face of constraints and opposition.

Here I’m going to focus on the first challenge – forecasting the different ways the future could evolve, and anticipating the opportunities and threats that could emerge along different paths.

In previous issues of The Index Investor, I’ve discussed one approach to meeting this challenge that is fundamentally based on the 18th century insights of Reverend Thomas Bayes. This is also the approach that underlay the Good Judgment Project Team’s methods. To simplify, you begin the process by establishing your “prior” forecasts (based on some combination of previous experience, deductive theory, intuition, and/or existing evidence), and then update it (to a “posterior” forecast) over time as you obtain new evidence.

The amount by which your prior probability is adjusted should be proportionate to the information value of the new evidence. A proxy for this is the “Likelihood Ratio”, which compares the probability of observing (or not observing) a piece of evidence if a hypothesis is true (e.g., this event will occur) relative to the probability of observing (or not observing) it if the hypothesis is false. The higher its Likelihood Ratio, the higher the information value of the piece of a new piece of evidence, and thus the greater the adjustment that should be made to the prior probability.

As we have noted in the past, we complement this “Bayesian” approach by also taking into account “surprising” evidence, which is not consistent with our existing mental model of a forecasting problem, and its range of possible outcome. Surprising evidence increases your uncertainty about the dynamics driving a system or situation, and should therefore cause you to reduce the probabilities you assign to the possible outcomes you have identified (e.g., by creating a catch-all option of “something else”, or by widening the confidence ranges for the probabilities you have attached to your current set of possible forecast outcomes).

While the Bayesian approach to forecasting is systematic and powerful, it has a major shortcoming: It conflicts with the way homo sapiens have operated for tens of thousands of years. Rather than acting like logical Bayesians, for most of our evolutionary past we have chosen to believe forecasts that were based on the most convincing stories.

Before there was the printed word, and before mathematics was widely understood, valuable information was organized, preserved, and communicated through the act of storytelling. Researchers have found that storytelling was likely a key contributor to the development of cooperative behavior that enabled our ancient ancestors to build larger and more successful groups. And given its importance, superior storytelling abilities probably conferred evolutionary advantage (e.g., “Cooperation and the Evolution of Hunter Gatherer Storytelling”, by Smith et al).

Increasing research into the neurobiology of storytelling is helping us better understand why this is the case. To begin with, we retain more information in memory when we receive it in the form of an effective story. At a more granular level, we are also learning what makes stories effective. Stories with a plot that develops tension hold listeners’ or readers’ attention much better than those that don’t (e.g., “The Emotional Arcs of Stories are Dominated by Six Basic Shapes” by Reagan et al). And character driven stories enhance that attention – and memory recall – still further (e.g., “Why Your Brain Loves Good Storytelling”, by Paul Zak, and “Storytelling is Intrinsically Mentalistic” by Yuan et al).

More broadly, retention of stories in memory has also been shown to be a function of both their emotional valence (positive or negative) and the degree of emotional arousal they trigger (high or low), which are also drivers of human beings’ instinctive “approach or avoid” reactions. As you would expect, stories negative valence with strong arousal have strong memory retention, because of their obvious evolutionary benefits for survival (avoid!). Similarly, stories with positive valence and low arousal also have strong retention. Interestingly, retention is weaker, as are approach/avoid reactions, for the other two types of story: positive valence/strong arousal, negative valence/weak arousal.

One consequences of this is that well crafted stories that contain misinformation are very hard to dislodge from people’s memory (e.g., “Misinformation and Its Correction: Continued Influence and Successful Debiasing”, by Lewandowsky et al). As Daniel Kahneman has found in his research (see his book, Thinking Fast and Slow), much of our cognitive activity takes place subconsciously, driven by what he calls “System 1”, which is based on association and automatically seeks to make new information cohere with existing belief structures (i.e., our internal stories). Only when this proves impossible do we become aware of discrepant information, usually via the feeling of surprise, which triggers much more effortful “System 2” processing to understand its significance and its relationship to existing belief structures.

In recent years, the importance of stories has been recognized in a wider range of fields, including economics and intelligence analysis.

In economics, three of the leading researchers in this area are George Akerlof (e.g., “Bread and Bullets”), Robert Shiller (e.g., “Narrative Economics”), and David Tuckett, whose research team has produced a series of fascinating papers on “conviction narrative theory.” Arguably, John Maynard Keynes predated all of these, and his 1936 discussion in Chapter 12 of “The General Theory of Employment, Interest, and Money” of the importance of what he termed “conventions” to decision making in highly uncertain environments.

Tuckett’s theory of conviction narratives is particularly interesting for our purpose here. As he and his team define them, “conviction narratives contain a few fundamental components, notably a focus on the specific emotional elements of narratives that evoke attraction or approach to an object of investment (broadly conceived), versus emotions that evoke repulsion or avoidance of that object. This emphasis on approach and avoidance in conviction narrative theory focuses the idea of sentiment on its implications for action in uncertain decision-making, thus focusing the often-vague topic of positive/negative sentiment.

In more ordinary language we focus on excitement about the potential gains from an action relative to anxiety about the potential losses. If excitement comes to dominate relative to anxiety, investment will be undertaken. Thus, in the simplest case, the key variables of interest are the aggregate relative difference between excitement and anxiety and shifts in this difference over time…Specifically, we suggest that action in uncertain contexts is possible because the human capacities for emotion and narrative are allied with cognitive processes to create a feeling of conviction” (“News And Narratives In Financial Systems: Exploiting Big Data For Systemic Risk Assessment”, by Nyman et al).

Shiller makes the point that at any time, a larger or smaller number of narratives may be circulating (which Tuckett would note are held by different people with varying degrees of conviction). And as we have noted in our work over the years, researchers have also found that when uncertainty increases, so too does human beings’ desire to conform to the group (which, in evolutionary terms increases the chances of survival), which leads to higher rates of social copying. Paradoxically, as uncertainty increases, this leads to a narrowing of the range of narratives in circulation and likely a weakening of the conviction with which they are held, thus increasing “social fragility” and setting the stage for rapid, non-linear changes in beliefs and behavior.

Researchers in the field of intelligence analysis (like Gary Klein, Robert Hoffman, and Marvin Cohen) have also focused on the importance of narrative, particularly with regard to the construction of causal explanations of past events that are then used to generate predictions about possible futures.

Hoffman and Klein and their co-authors have found that our explanations of causal processes in complex adaptive systems (like economies or financial markets) typically take one of three simplified forms: (1) a simple list; (2) a logical sequence; or (3) a story that incorporates context and complex causal relationships (e.g., “Naturalistic Investigations And Models Of Reasoning About Complex Indeterminate Causation”, by Hoffman, Klein, and Miller, and Hoffman and Klein’s series of papers on “Explaining Explanation”).

Before moving on to a specific forecasting example, we should also discuss the confusing way that different researchers use the terms “story” and “narrative”. Some consider them synonymous – for example, Shiller defines “narrative” as “a simple story or easily expressed explanation of events that engages the emotions of others.”

Others draw inconsistent distinctions between them. For example, one definition says that a story is about people and situations: “a sequence of characters’ motivated actions (i.e., events) and their consequences.” In contrast, there can be multiple narratives for a single story, which vary in the characters and events they contain, and the way they present them (e.g., chronologically, by subplot, by character, etc.). Yet another definition considers narrative to be a “meta” concept: a system of stories and the connections between them.


A Practical Application of Conviction Narrative Theory to Regime Forecasting

Sitting around the conference room table, the organization’s investment committee got ready to hear four different consultants make their best pitch for why four different regimes were likely to exist three years hence, in 2022.

Given currently high levels of macro and market uncertainty, the committee would then weigh them before taking a decision on whether to tactically adjust its investment policy’s baseline asset class weights, and/or take other actions to reduce the risk of a substantial downside loss, while retaining as many options as possible for significant upside returns.

The first presenter was Susan Davis, who was accompanied by a team of younger associates carrying thick briefing books.

“You assigned our firm to make the most convincing case we could that three years from now the macro system will be experiencing persistent deflationary forces, like those that for the past thirty years have battered the Japanese economy. The essence of our firm’s argument is this: in essence, deflation represents a consistent shortfall in demand, relative to potential supply. The evidence I will present over the next hour supports the hypothesis that this is, in fact, the situation we face today, with little hope that it will change over the next three year. In fact, some of the deflationary forces that are at work seem likely to intensify.”

“I’ll begin with a summary of the key points in our case, which we will then review in more depth:

Due to many factors –including offshoring, automation, and the adoption of more efficient business models – potential supply has increased in many industries, from electronics to clothing to energy to food and many others. To be sure, this has not been the case across the board. For example, in many developed markets, increasingly tight regulatory conditions have limited the growth of housing supply, and major sectors of the economy like healthcare and education remain relatively inefficient, in some cases with negative productivity growth over time. But in the main, potential supply has increased.

However, this increase in potential supply has not been matched with an increase in aggregate demand; in fact, quite the reverse has occurred. Growth in demand has been limited by a number of headwinds, include the aging of the population, rising inequality, flat productivity (which, along with rising concentration and corporate power in some key industries has arguably limited labor compensation growth), the poor performance of education systems (which is causing too many well-paying “middle skill” jobs to go unfilled), and for too many young people, the growing burden of student debt. In the future, we can expect to see increased use of rapidly improving labor-substituting technologies like automation and artificial intelligence put further downward pressure on demand.

In the past, we might have been able to look to growth in investment spending or net exports to offset these downward pressures on consumer expenditure. But both of these have weakened too. To begin with, the increasing digitization of our economy has sharply reduced the need for capital investments in traditional plants and equipment. And where it has not – think of Apple’s supply chain – much more of it has taken place in other nations as supply chains have become globalized. And as I noted above, residential fixed investment has been limited by stricter regulation, rising inequality, and potential homebuyers who, in aggregate, are more burdened with student debt than ever before.

But what about exports? The problem here is that we aren’t the only economy that is facing demand headwinds. I would argue that Europe’s are even stronger, while those in China are clearly becoming stronger as the population rapidly ages and overcapacity in many industries reduces investment spending.

Now economic history teaches us that not all periods of deflation are necessarily bad. Falling prices for consumer goods can make stagnant incomes stretch farther and actually produce increases in some people’s effective standard of living – assuming they are still working and earning income.

But history also shows us that some deflations can be very destructive. These are the ones that occur when an economy is highly leveraged. In this case, falling prices mean falling revenues and increasing difficulty in servicing debt, which forces borrowers into bankruptcy, causing more people to lose their jobs and incomes. The bad news is that we live in a world today that has very high debt levels.

Even that might not be so bad if interest rates were in their historical normal range, which would allow monetary policy to be used to reduce interest rates and make all that debt cheaper to service. But today’s interest rates are already at record lows – we’re close to or at the “zero lower bound.” So we may well be facing the bad kind of deflation, where the negative impact of falling prices is amplified by increasing financial distress and rising corporate failures and cutbacks.

To be sure, if we find ourselves in that position, it seems certain that the government will try to ramp up fiscal policy, and expect the Fed to monetize a lot of the new debt that will be issued to pay for it (because the increase in taxes on high income earners that I expect won’t come anywhere close to covering the fiscal stimulus that we’re going to need to keep debt deflation at bay). Whether the current political gridlock and policy arguments in Washington will enable that huge fiscal stimulus to happen is an open question. I’m not optimistic, given how polarized our politics have become.

And even if it passes, let’s not forget what happened to all those “shovel-ready” infrastructure projects that President Obama thought he was funding back in 2009. Regulatory red tape and litigation by activist groups made sure that a lot of them never got off the ground. And even if you get a lot of actual fiscal stimulus, there is also the question – raised recently by researchers – of how much impact it will have. The problem is that they have found that the fiscal multiplier – the demand bang you get for your buck – tends to shrink as the population ages and inequality worsens. So any fiscal stimulus will have to take that into account if it is going to succeed in boosting demand and keeping deflation at bay.

Another question is whether any fiscal stimulus will just be a one time fix – like a tax cut – whose effects will fade away, or whether it will come with structural changes that address many of the root causes of the fundamental deflation problem we face. Maybe it will. I’m sure some of the other presenters today will have a more rosy view of the future that we do.

The next presentation came from professor Nigel Ensor-Gregg from Portsmouth University, who was charged with making the case for why the macro system would be in a period of high inflation by 2022. His remarks were, as always, both colorful and thought provoking.

“Let me start by telling you that many of the arguments I’ve heard for why we’re going to find ourselves in a period of high inflation three years from now are, to be polite, complete bollocks – rubbish to you Yanks. But, as I’ll get to, not all of them.”

“Here are the traditional arguments for how you find yourself in a period of high inflation, and why I think they’re wrongheaded:

While some people think MV=PQ is of the same order of brilliance as E=MC2, it is not. If V and Q are constant, then an increase in the Money supply as big as we’ve seen since 2008 should have produced a huge increase in inflation. But it didn’t. Why? Because V (for “velocity”) has fallen to an all-time low, while Q (real output) has also increased. So we haven’t seen the hyperinflation that some predicted would result from central banks’ quantitative easing (“QE”) policies.

That begs the question, of course, as to why Velocity has fallen by so much. To unravel this mystery a bit more, let’s look at some data. At the end of Q4 2018, the NY Federal Reserve Bank reported that total household debt (e.g., credit cards, mortgages, auto loans, student loans, etc.) stood at $13.5 trillion. That is higher than the previous peak of $12.7 trillion in Q3 2008. Households haven’t been hoarding money.

What about businesses? According to the St. Louis Fed, at the end of Q3 in 2010, there was $460 billion in loans outstanding to non-financial corporate businesses in the US, and $3.9 trillion in debt securities. By Q4 2018, loans outstanding had grown to $1.2 trillion, while debt securities had grown to $6.2 trillion. However, this borrowing has, to some extent, been offset by a large increase in holdings of liquid assets, at least at some companies. Apple, to use an extreme case, has a cash hoard of nearly $250 billion.

That leaves banks – which now hold near record amounts of excess reserves. This is the main reason that velocity has declined so much. But that raises the interesting question of why banks haven’t increased their lending – rather like Sherlock Holmes’ dog that didn’t bark. Is it because regulations about lending or the banks’ credit policies have become stricter? Or is it because there just isn’t that much demand for loans? Whatever the reason, the fact remains that in recent years we’ve learned that velocity can change more than we’d previously thought, so a big increase in the money supply hasn’t led to the high inflation people thought QE would produce.

Another route to high inflation that people have put forth is a loss of global confidence in the US dollar, which would cause it to depreciate, forcing a rise in import prices that would drive other prices across the economy and thus an increase in the Consumer Price Index. A fair argument, as far as it goes. But it rests on a crucial assumption: that all those dollars being sold would have another place to go. And where might that be?

Not China – it surely wouldn’t want to see a sharp appreciation in its currency, which would make its exports more expensive. And how much trust do you have in the Chinese financial markets? What about the Euro? While that market might have more capacity than China, how much confidence do you have in what you’d be buying? It’s not as though there aren’t big variations in risk – and that’s just in the case of sovereign debt. What does that leave? Japan? Where the debt/GDP ratio is the highest among major nations? Gold? Its supply is limited. Would you really pay $10,000 an ounce or more? You get my point. The argument that a crisis of confidence in the US Dollar is going to cause high inflation rests on a very questionable set of underlying assumptions about where those funds would go.

What about good old-fashioned demand-pull inflation? Remember that, from way back when? When aggregate demand was grew faster than supply, and forced prices up? Any takers for that argument in today’s global economy, which is automating its way to ever higher levels of efficiency, and thus lower marginal costs of supply? I didn’t think so.

Now at this point, you are no doubt asking yourselves, ‘I thought we paid this professor to make the best case possible for how we could end up in a high inflation regime in 2022?” Hopefully you won’t be disappointed by what comes next.

James Montier, of the investment firm GMO, has repeatedly made the case that historically, the main cause of periods of high inflation has been supply shocks – that is, a sudden drop in supply relative to demand. In many cases, these were made worse by large amounts of government debt denominated in foreign currencies (which inflation and the resulting exchange rate depreciation made more expensive to service), and political conditions that led governments to introduce indexing as a means of protecting people’s purchasing power. These typically led to exploding government budget deficits that were financed by printing money, which, via the indexing ratchet, caused inflation to increase non-linearly, sometimes to the point of hyperinflation – that is, inflation that is increasing at an increasing rate.

Given this, we need to ask two questions: What types of supply shock could have a powerful impact on the United States? And how likely is one of them to happen between now and 2022?

Let me offer you some possibilities to consider. Another oil shock is perhaps the most obvious candidate – for example, if increased Sunni-Shia conflict in the Middle East, or an Iran-Israel conflict caused Iran to mine the Strait of Hormuz or undertake direct military action against Saudi Arabia. Never forget that most of Saudi Arabia’s Shia population lives close to its main oil fields.

Infectious disease is always a threat – some have said that another global influenza pandemic on the order of the 1918 Spanish flu is just a mutation and a plane ride away. And we’ve also seen how Ebola is making a comeback in Africa, though so far it kills so quickly that outbreaks have burned themselves out. But again, mutations are a fact of viral life. And, like it or not, with increasingly easy access to genetic engineering tools and knowledge, we sadly can’t rule out an intentionally caused pandemic that could close down key supply chains and produce an inflationary shock.

Less likely, but far from impossible, are four other supply shock scenarios – a major crop failure caused by a weather related event, a cyber event or solar storm that causes widespread damage to critical infrastructure – e.g., the electrical grid, pipeline, or communications network, or the eruption of “kinetic hostilities” (probably by accident) between the US and severely disrupts supply chains.

So we have six scenarios that could produce a severe and prolonged supply shock that would likely trigger a sharp rise in prices. How do we estimate the likelihood that any of them will happen? That’s hard, because in many of these cases, the crisis in question would be relatively, and historical frequencies are either unavailable or unreliable guides in a system that is constantly adapting and evolving. We can, however, turn the question around, and ask what is the annual probability that over the next three years, none of these supply shocks will occur? Let’s say you think that probability is 95%. Then over three years, the probability that a significant supply shock will occur is 100% less 95% cubed, or about 14%. If you think the annual probability of non-occurrence is 99%, then the probability of occurrence over three years falls to just 3%. On the other hand, if you think the annual probability that none of these shocks (or one we haven’t anticipated) will occur is actually 90%, then the probability of an inflationary supply shock over the next three years jumps to 27% - or about a 1 in 4 chance.

Peter Fisher, a leading sell-side research analyst at a large investment bank, gave the third presentation. His charge was to make the best case that by 2022 the macro system will have returned to the normal regime, with equity asset classes expected to deliver annual returns in their historical range.

Fisher wasted no time in launching into his argument:

“Are equities today a little pricey? Sure, especially in the US and emerging markets. But we also think there’s a very good chance that events over the next three years are going to catch up to these valuations and make them look much more reasonable, and maybe even cheap. Here’s why:”

We think there’s been an overreaction on the downside. Let me give you some examples. Brexit is either not going to happen, or its impact isn’t going to be a negative as the scaremongers would have you believe. Think, for example, about the possible benefits of a comprehensive trade deal between the US and the UK – that would instantly add 15% to the United States’ GDP. That would be like adding another Canada – and another Mexico.

Now let’s look at China. How happy do you think people are about the growth of the surveillance state, the repression of private companies, a weakening financial system, a slowing economy, and picking a fight with the United States? And how happy do you think all those other leadership factions are in China, whose members have been targeted by Xi Jinping’s relentless purges, sorry, I mean “anti-corruption” drives? We think there is a credible case that sometime in the next three years that Xi is going to be replaced by a new leader who is going to tone down the conflict with the United States, go back to Deng Xiaoping and his successors’ encouragement of the private sector, and get Chinese growth back on track, including cutting some type of managed trade deal with the United States that will benefit both countries’ economies.

We also don’t believe that Donald Trump will be our president after the 2020 election. Maybe it will be a health issue, or maybe he’ll just get tired of their constant attacks. Or maybe he’ll get taken out in a primary by somebody like Nikki Haley, or maybe center right and center left members of Congress will finally band together to pass pragmatic legislation that is opposed by both parties’ radical wings. We just don’t see how the United States can continue down the political path we are on without something giving; there’s just too much stress accumulating in the system.

So think about what happens if we actually get health reform passed that people support, while tax rates get more progressive and rather than an increase in the minimum wage we boost the earned income tax credit that gives people a stronger incentive to work and a decent income when they do, and there’s reforms of the college loan system and maybe some type of national apprenticeship system that starts to get more people into good paying jobs without having to rack up tens of thousands in debt chasing a worthless college degree. And what happens if this center group finally passes immigration reform that people can support – maybe one based primarily on skill-based immigrants, like they have in Canada and Australia? And if the increasing capabilities of automation and AI technologies lead to more reshoring that creates more jobs right here in America?

You know what this adds up to? Faster growth. Reduced income inequality. Restoration of the American Dream. Less social and political conflict. And rising stock prices that will make today’s valuations look reasonable, or maybe even cheap. That’s where we think global macro is going to be in 2022.

The last presenter of the day was Jayne Coombs, from a Canadian risk consulting firm. She was tasked with making the argument for why in 2022 the global macro system would still be in the high uncertainty regime it is in today.

“I guess it’s fitting that I’m the last presenter you’re going to hear from today”, she began, “as I’m here to argue that three years from now we’re still going to be in a regime of high uncertainty, though by then global equity markets will have lost 20% or more in value compared to where they stand today. Let me tell you why the high uncertainty/high anxiety period we’re in today isn’t going to end anytime soon:”

To begin with, the most recent data show that the world economy’s three “growth motors” are all slowing down – fastest in Europe, a bit more slowly in China (the accuracy of whose data is always questionable), and now in the US.

We don’t think this will end up in deflation in the US because the healthcare and education sectors of the economy together have a 10% weight in the Consumer Price Index. Both have growing demand and are highly inefficient with negative productivity growth, so we expect continued strong price rises in both. On the other hand, food and other goods (e.g., food, furnishings, etc.) together make up 34% of the CPI, and both have been under downward pressure because of changing supply and demand dynamics in those sectors. Energy has an 8% weight in the CPI. In the absence of a supply shock, a weakening economy should cause energy prices to decline.

The key variable in the argument whether we will enter a period of deflation is therefore housing (i.e., home ownership and rental costs), which has a 33% weight in the CPI. As the growth of housing supply is increasingly constrained by regulations, the question is whether even as the economy weakens, a combination of population growth and new household formation, along with low interest rates and more quantitative monetary easing will result in continued increases in housing costs. We think they will, and that all these factors, plus a heavy dose of fiscal stimulus will keep deflation at bay. Having seen what has happened in Japan over the last 30 years, we know that most policymakers are deathly afraid of deflation, and will go to extraordinary lengths to avoid falling into the same trap.

Beyond avoiding deflation, however, we see economic uncertainty continuing to increase as a number of forces continue to play out, including the increasing deployment of more capable automation and artificial intelligence technologies and their likely negative impact on employment, the potential for job losses and financial system problem due to high leverage levels as the demand continues to weaken, possible disruptions in Europe due to Brexit, worsening of Chinese-US relations and their potential impact on supply chains and trade, and what are likely to be a fierce debate over the form that any fiscal stimulus will take when economies fall into recession territory. In the US, this could take the form of fights between supporters of progressive initiatives like the “Green New Deal” and supporters of tax cuts (and at which groups they should be targeted). In Europe, a downturn could easily trigger another sovereign debt crisis in the Eurozone (this time with Italy at its center), as well as fights over how fiscal stimulus should be apportioned across nations – specifically, will Germany keep looking out for itself, or stimulate more than its politicians would like in order to support the wider Eurozone?

The next reason we’re forecasting that the current high uncertainty regime will continue for the next three years is that we don’t see the political situation getting any better, as long as the twin economic problems of insufficient demand and worsening inequality remain unresolved. We expect that in the United States and European countries, centrist political parties will keep losing ground to more extreme parties, particularly if worsening global economic conditions trigger much larger migration flows. As a result of all these drivers, political conflict, policy gridlock, and uncertainty will increase.

In the United States, we now think it is at least an even bet that Donald Trump will be re-elected, shocking as that may have seemed six months ago. But in what looks like a repeat of 1972, the Democratic Party seems poised to tear itself apart in a fight between its left and center wings, much as the Republicans did a decade ago when the far right Tea Party arrived on the scene. We doubt that Trump’s re-election will reduce uncertainty.

On the international front, we expect that Vladimir Putin will continue his “grey zone” initiatives against Western Europe, seeking to generate further domestic conflict between factions both within countries and across the continent. A critical question for us is how these efforts, if they succeed, would interact with Putin’s declining domestic popularity. Would he attempt to seize more territory – say in the Baltics this time, under the pretense of protecting threatened Russian minorities in one or more of those countries? Clearly, this dynamic is potentially a major source of increasing uncertainty. And the same is true, unfortunately, for China’s increasingly aggressive activities in the South China Sea, particularly at a time when, because of growing economic problems, Xi Jinping may be feeling less secure in his position. The third big international source of uncertainty that we see is Iran, which is unlikely to slow down either its attempts to foment conflict and expand its influence in the Middle East (e.g., in Iraq, Syria, Lebanon, and Yemen), or its ongoing development of weapons and delivery system that could pose an existential threat to Israel – where Bibi Netanyahu, who has a very keen appreciation of these threats and is a combat veteran of Sayeret Maktal, will still be in charge.

Last but not least, a key reason we expect the high uncertainty regime to continue through 2022 is the recursive nature of uncertainty itself, which is deeply rooted in human nature. As we become more uncertain, we rely more heavily for direction on what we observe others to be doing. Various researchers have different terms for this, like social copying, social learning, or increased conformity. While the definitions of these terms differ, the underlying phenomenon is the same – and in our hyperconnected world, it has now become supercharged. In the 1930s, Keynes noted how, in the face of high uncertainty – and indeed, as he noted, our ignorance about what the future may hold – we adopt what he termed “conventions”, which provide the conventional wisdom that gives us the confidence to act. But he also noted that these conventions ultimately rest on public confidence in their accuracy. The flip side of confidence is uncertainty – and hyperconnectivity now rapidly transmits changes in it across the population, which can cause “conventions” or the conventional wisdom to change more rapidly than ever before – which recursively increases feelings of uncertainty.

Having heard all four presentations, the committee prepared discuss and weigh the relative likelihood of the narratives they had heard, before deciding if any of them were sufficiently convincing to motivate a significant change in their portfolio asset class exposures…


Closing Thoughts

As you can see, the conviction narrative approach to forecasting is quite different from more systematic Bayesian methods. Yet on a very deep (indeed, evolutionary) level, most people are likely to find it more intuitively appealing, as through the use of story it simultaneously engages us on both a cognitive and emotional level.

For that very reason, stories can also lead us astray. One way to protect against that is to systematically critique any conviction narrative before you act on it. In our strategic risk consulting work over the years at Britten Coyne Partners, we have found two methods to be very useful.

The first is Gary Klein’s “Pre-Mortem” technique. Assume that you are at some point in the future, and your strategy or action plan has failed to achieve your goal. Looking backward, write down why this happened, including warning signs you missed, and what you could have done differently to increase the chances for success.

The second is Marvin Cohen’s “Story Critique” approach. Begin by evaluating the assumptions made in the story about “known unknowns”, and the quality and weight of evidence that supports each of them. As the number of assumptions with weak support increases, so too does the likelihood that important “unknown unknowns” are missing from your story and remain to be discovered. Cohen recommends two approaches to help you “forage for surprise.” Begin by teasing out and examining the implicit assumptions it contains. After that, move on to questioning the story’s assumed “known knowns”.

On balance, we believe that both the Bayesian and Conviction Narrative approaches to forecasting are equally valuable. The former forces us to pay close attention to the value of new information that we receive over time, while the latter makes explicit the evolving causal stories that drive our decisions and actions when uncertainty is high.