Where Did all the Optimism Go?
By John M. Burkhardt
Risks & Rewards, February 2023
In December of 2021, I wrote in Risks and Rewards about the dynamics of the historically unprecedented bull market that emerged subsequent to the COVID-fueled market crash of February/March 2020. The analysis at that time was that the unprecedented bull market was fueled by an intrinsic optimism bias among retail investors. In short, the widespread availability of cash resulted in enormous additional investments from retail investors making decisions largely on the basis of behavioral economic heuristics. An irrational belief in their own capacity for successful investing–i.e., optimism bias–drove a critical mass of investors to continuously double down even in the face of transient dips, buoying market stability and supporting long-term market growth.
Importantly, I asserted that this optimism would tend to fuel itself. As dips proved transient and market growth continued, retail investors would continue to invest in a sanguine fashion, sustaining or accelerating market growth well beyond the indications of a fundamentals-based model.
“Absent some genuine infrastructure-altering event, such optimism bias can persist indefinitely,” was the prediction made.
When Optimism Runs Out of Gas
As of this writing (December 2022), the “historically unprecedented bull market” is clearly no longer anything of the sort. The recovery and rebound disappeared with considerable alacrity, beginning in January of 2022. Retail investors have retreated from most active positions and institutional investors are largely in holding patterns to limit loss. We therefore need to ask ourselves, what happened? Was the original thesis around investor optimism wrong?
As it turns out, no; the December 2021 prediction included a significant caveat that has since come to pass. The full analysis given at the time was that optimism bias would persist so long as liquidity remained among retail investors. Once liquidity evaporated, loss aversion would dominate retail behavior.
Throughout 2022, inflation has been the major story in the United States. Different sources provide different estimates, but the general consensus is that inflation is approximately 7 percent, representing a roughly 50 percent increase over 2021’s elevated rate, and is the highest level of inflation seen in the country since the early 1980’s. This has substantially reduced available liquidity among investors, and has driven the retail market into the predicted loss-averse behavior.
While it is generally accepted that inflation is bad for investors and markets, it is not immediately obvious why this is the case. Superficially inflation appears to simply be an exercise in scaling, and as such, little more than a multiplicative adjustment of pricing models. Individuals’ dollars are worth less during periods of inflation, so all that should be required is a recalibration of predictive models for whatever the rate-adjusted value is, with a small fudge factor based on historical observations. It is a simple algebraic transformation; why should this have such a profound impact upon overall market behavior?
What is often missed are the human elements triggered by inflation, most notably its impact on household finances and the resulting additional constraints. Yes, there is a rescaling of dollar values that is immediately relevant to pricing and predictive models, but there are downstream consequences of this that go well beyond model recalibration. When dollars lose value, many financial decisions faced by individuals invert their architecture–i.e., the switch from the gain domain to the loss domain. Allocation of household resources moves from finding opportunities for wealth accumulation to minimizing drains of savings accounts. Household forecasting becomes extremely conservative as salary inflation lags cost-of-living inflation. Recalibration of individual budgets requires contraction across many categories of expenditures. The rescaling of value ultimately results in a fundamental shift in the types of decisions that individuals–i.e., retail investors–must make on a daily basis.
Inflated Expectations
Consequently, dealing with inflation and making predictions during inflationary periods does not just mean recalibrating the models, but updating them to a completely different regime. It is known that actual human decision-making–not optimized calculations, but observable quantified behavior–displays different dynamics in gain outcomes versus loss outcomes.
As mentioned previously, investment behavior in the retail sector exists in a dynamic balance of optimism and loss aversion. Optimism bias predominates during growth periods; indeed, one can make a compelling case that optimism bias is a prerequisite for any sort of active management of assets. When markets start to consistently decline, however, loss aversion becomes the primary driving factor. In broad strokes, losses are felt more strongly than gains of equal value. Losing money hurts more than gaining money feels good. This asymmetry in perceived impact drives investors to focus on avoiding losses more than pursuing gains.
What is often missed, however, is how loss aversion inverts when losses become certain. Individuals readily accept greater overall potential losses if there is a possibility of avoiding loss altogether. Essentially, people become gamblers. There is an evolutionary imperative underlying this behavior: In a physical landscape where any loss could mean injury or death, even the slightest chance of avoiding loss is preferable, even if it means accepting a potentially greater overall impact. Projecting this behavior forward 200,000 years onto the investing landscape, investors are willing to gamble and take risky positions, creating even greater exposure, to try and avoid otherwise certain losses.
Consequently, one expects to observe an acceleration in overall loss once inflation sets in, liquidity evaporates, and markets decline. Probabilistically, most risky positions will not end well, so the net impact on market conditions will be negative. Importantly, however, this phase cannot occur indefinitely. Investors can only gamble and attempt to dodge the pain so long as they have resources–liquidity–available to gamble with.
In periods of market decline such as that currently being experienced, retail investors are expected to exhibit an initial phase of suboptimal behavior which eliminates most of the sector’s available capital. And as discussed earlier, inflationary periods change the decision architecture of individuals and their available set of choices, so periodic infusions of cash largely disappear. With intrinsic optimism having yielded to loss aversion, and with little available capital and thus no ability to gamble further to avoid losses, retail investors are not doing anything at all.
When Prediction Fails, Markets Stagnate
Savvy readers will note that for most retail investors, there is an easy way to re-enter the market and fuel loss aversion induced gambling: Sell off assets to generate the desired liquidity. So why is that not being witnessed to any material extent?
At present, investment markets and the economy as a whole have extremely high levels of uncertainty attached to them across all sectors and asset classes. It is unclear what will happen with interest rates in 2023; whether inflation will decline in the near future is unknown; the direction of the housing market is an enormous wildcard. Asset prices are in decline across bonds, securities, commodities, and alternative classes. In short, the future economic landscape is remarkably difficult to predict, and no asset class appears especially attractive at the moment.
Lack of clarity as to the best path forward is the most potent precipitating factor for “Status Quo Bias.” This is often described as a tendency for people to persist in whatever they are already doing. However, it is more precise to describe it as an unwillingness to make a decision with a high degree of uncertainty around it; this has the effect of pushing off any sort of action to an indefinite future. With no clarity on the economy, and a lack of good investment options, retail investment activity is expected to stagnate.
On top of this, it is important to point out that loss aversion ceases being meaningful beyond a certain point. Examination of Kahneman & Tversky’s Prospect Theory curve show a tapering impact of perceived gains and losses at the margins, but there is a practical aspect to this as well. Put simply, one cannot lose more than one has. While potential gains theoretically have no cap, there is a fixed limit on the amount of loss an individual can experience. Loss aversion ceases to be a meaningful influence on behavior beyond a sufficient level and/or duration of loss, and efforts to arrest declines in portfolio value will taper off.
Taken together, retail investor activity can reliably be expected to drop off during periods of prolonged inflation and market decline, as status quo bias takes hold and additional losses lose motivating efficacy. Similar to optimism bias during market growth, this stagnation can last indefinitely without major alterations to the underlying decision architecture of investors.
Hunting for Hope
While the picture presented here is rather grim, history shows that no decline lasts forever. There will be an eventual turnaround in markets, and optimism will once again drive retail behavior. The question then becomes, what will the signs of recovery be, and what will drive retail investors to behave with optimism?
Although the rise in inflation was the leading indicator for the current decline, a reduction in inflation will not necessarily be the leading indicator for recovery. Restoration of optimistic investor behavior depends not on lower inflation, but higher liquidity. Once there is a surfeit of capital on the individual level, investors can resume gambling against losses, doubling down on market dips, and providing the underlying stability that was observed during 2020 and 2021.
Any sort of government policy that genuinely reduces short-term costs and/or provides infusions of cash to individuals will precipitate at least a transient recovery (the long-term impacts of such an intervention would be considerably more complex). For a sustainable recovery, the most predictive element will be a re-normalization of salary/wage indices to cost-of-living indices. At present growth in cost-of-living indices is outpacing growth in salary/wage indices; once wages catch up to costs in a lasting fashion, retail investors will have the capacity to resume their pre-decline behaviors. Note that employment levels are not specifically predictive of a market recovery, as employment does not directly imply anything about salaries and available capital.
Lessons Learned
It seems obvious to state that if investors have a reliable source of liquidity, markets will recover. However, the behavioral analysis as to why this is true and how such a recovery will unfold is quite nuanced. This requires a considerable shift in investor thinking and behavioral inertia, moving from decision avoidance to risky gambles to optimism-driven growth.
While investors can be expected to resume some activity with any sort of cash infusion, reversing market direction requires a large-scale move across the larger retail population. Therefore, investors need to have enough recurring capital to sustain their loss aversion until the market shifts. The sustainability and recurrence of liquidity is crucial, and perhaps the most important aspect in ascertaining whether a recovery will be transient or lasting. Optimism bias and its associated investment behaviors are feed-forward functions, but they require a lasting fuel to overcome the deep losses currently being experienced.
Statements of fact and opinions expressed herein are those of the individual authors and are not necessarily those of the Society of Actuaries, the newsletter editors, or the respective authors’ employers.
John M, Burkhardt, Ph.D., is the CEO of Capita Neuro Solutions and Adjunct Professor in the Enterprise Risk Management program at Columbia University. He can be reached at jburkhardt@capitaneuro.com.