Does the flap of a butterfly’s wings in Brazil set off a tornado in Texas?”Was the question Edward Lorenz posed as he dove into his exploration of Chaos Theory.
It was his way of stating that a tiny, inobservable change in input can lead to a massive, very observable change in output. This occurs in nonlinear (aka chaotic) systems where intricate interdependencies between components of the system take little wing-flaps and amplify them exponentially through a series of feedback loops.
The same thing happens in financial markets, where changes in one part of the world can lead to very drastic effects across global capital markets, Covid-19 being the case in point. These effects in turn force policy makers into responding, which creates new sets of consequences, etc. We are therefore living with a set of consequences which stretch back and forward through time, a reality which is simply too complex for the human mind to comprehend.
Our natural response is to reduce this complexity into smaller, more understandable parts, but doing so makes us blind to just how far reaching the consequences of our actions are. Ironically, by ignoring complexity, we become stuck inside a complex feedback loop of our own making, where each new action we take keeps us locked into situations with no good options.
We hope that as readers make their way through this Collection, they can get a sense of the nonlinear nature of markets, and an appreciation of the importance of thinking through the lens of complexity rather than simplicity.
We live in a world of complexity. Not complexity as in “it’s complicated”, even though being complicated is certainly part of it. Complexity as used by scientists to describe dynamic systems full of interdependency, feedback loops, multiple variables, and constantly evolving relationships.
In less scientific parlance, chaotic systems. In pop culture terms, chaos theory and the butterfly effect. In layman’s terms, complete confusion!
In such an environment, it is impossible to separate individual systems from the whole, as the whole is one huge interdependent mass.
For example, financial markets, themselves complex systems, are inextricably linked to myriad other complex systems. These include, but are not limited to, human emotion, domestic economies, large global economies, local politics, global geopolitics, demographics, and even the weather.
Needless to say, human minds cannot comprehend the result of so many variables which are constantly changing. And, computer models are difficult to create because even pinpointing which variables matter is a massive headache considering that a variable that matters today might not matter tomorrow.
The natural, and most common response that traders have to such an overwhelming task is to engage in reductive thinking.
Reductive thinking is useful in simple systems where variables are easily identifiable and change in predictable ways. This is because the systems can be reduced to smaller, simpler variables which are easier to understand, manipulated or changed in some way, then put back together again.
However, because markets are anything but simple, reductive thinking tends to take the form of market mantras like “buy low, sell high”, “buy when everyone is fearful”, “sell when everyone is bullish”, etc.
Unfortunately, these mantras, while pithy soundbites, don’t provide much in the way of useful advice.
How low is low, and how high is high? How do you tell when everyone is fearful? How do you tell when everyone is bullish?
To be continued…
Another, more pernicious example of reductive thinking is how people love to attribute single factors to explain price movements.
You’ve definitely heard it before, stuff like: “XYZ stock went up because it reported higher revenue”, or “Treasury yields are lower because of safe haven buying”.
These statements do more harm than good because they cause traders/investors, especially inexperienced ones, to 1) confuse correlation with causation, and 2) believe that markets are simple and easy to understand.
In the case of 1), a lot of times, markets will spike on a headline, such as “XYZ notches up record revenues”, which leads to people conflating the two events.
But correlation is not causation, and if one wants to attribute a fundamental reason for prices moving higher, there are so many other factors involved. Not least XYZ’s net income, margins, and costs.
It is far more likely that the buying came from a mixture of people. Those who truly believe that only revenues matter, those that assume higher revenues means good overall earnings, and those looking to profit from all the buying the people in the first two groups were doing.
Not a single, simple, factor after all.
In the case of 2), reading reductive statements put out by analysts and the financial media only reinforces people’s natural tendency to think in reductive, simple cause and effect terms.
Because of A , B will happen. The simplicity and straightforwardness of such causal statements are after all, quite beguiling, and the human psyche likes to latch on to them as concrete truths.
Unfortunately, complex systems do not work in linear causal ways.
For example, a statement like “Treasury yields are lower because of safe haven buying” encourages people to think of Treasuries only in terms of it being a “safe haven”. But the truth is, Treasuries only became the go to “safe haven” asset relative to stocks at about the start of the new millennium.
The point here being that market relationships are not constant.
Also, thinking of Treasuries as safe haven assets views the asset class only from the perspective of portfolio management, whereas Treasuries serve a much larger role in the financial system.
They are also used as collateral in the massive global repo market that international banks and corporations use to quickly gain access to USDs. This effectively makes USTs a de facto global currency, and the reserve asset of choice for global Central Banks.
Furthermore, their role at the nexus of global repo and USD funding markets means Treasuries are affected by forces far more global, diverse, and complex than portfolio managers’ need for a “safe” asset.
Forces like global growth prospects, offshore USD liquidity, and international Central Banks’ reserve accumulation, to name a few. These in turn are part of two-way feedback loops with other factors, such as global trade volumes, commodity prices, and inflationary/deflationary forces in foreign economies that are too dependent on offshore USD funding.
Again, not a single, simple, factor.
To be concluded…
Reductive thinking is also widely used in financial forecasting. Examples abound daily in the financial media where statements that follow the same seductive Because of A , B will happen formula.
“Sell the USD because of QE!” Not true, see our QE Series for a more in depth look.
“Inflation is coming because of QE!!”. Also not true.
And finally, “Fiscal stimulus will save the economy!”, but… will it? Is an economy so simple as to just require government spending to be saved?
Unfortunately for these forecasters, it isn’t.
Firstly, fiscal stimulus is an injection of government cash into corporate America that will help businesses stay afloat for a while longer as the pandemic rolls on. The cash will also help businesses not fire their employees, which is the government’s way of propping the labor market up.
All this is well and good, but what happens when the money runs out?
Of course, the hope that the government has is that economic growth can rebound enough by then to keep the workers on payroll and make businesses profitable again. Essentially, they are buying time. Will economic conditions improve enough by the time stimulus money runs out?
No one can say for sure, but looking at the devastation wreaked in the labor market, it is highly doubtful. The massive amount of jobs lost makes it difficult for companies to take them all back into the labor market quickly enough.
This is due to companies needing time to regain confidence before expanding their operations again. Also, the lost jobs have already led to sharply reduced disposable income and spending, which directly reduces companies sales and thus their ability to rehire workers.
That’s just the labor market side of things.
There is also the credit aspect, where companies, especially those who do not have access to capital markets, struggle to get banks to loan them money. Because, just like with household stimulus checks, one off government payments are not future cashflow.
This means the stimulus does not really help them to get bank loans. No bank loans means no new capital investment, no new hires, and no increase in the money supply in the economy.
What started out as one factor has now morphed into multiple, interlinked factors
It’s not all bad for government stimulus though. It can be made to work, but for this to happen, the stimulus needs to be constant, direct spending in the economy, just like in WW2, where the government boosted the economy with its massive military spending.
The problem here is of course, how long can the government keep this up, and does the political appetite for this exist in both the legislature and electorate?
If this road is taken and the government decides to stop at some point in the future, the problems listed above will hit home again, except now the private sector is much smaller in size having been crowded out of the economy by the government!
So, will fiscal stimulus save the economy? It’s complicated (or should we say complex)!
After giving you some sense, and hopefully, an appreciation for how market outcomes are the result of complex phenomena, it is time to introduce the Iceberg model.
This model is an excellent tool to help you think through developments in complex systems, and react accordingly to them. Of course, for us, this would center on financial markets.
As with any iceberg, only a small percentage of the structure is visible above the waterline, with the rest of its massive bulk floating beneath. The same thing applies to Events in complex systems.
Events are the observable outcomes of multiple interactions within the complex system, hence their place at the top of the Iceberg model.
This would be, for example, the price of a stock moving sharply higher or lower on a headline. The price move is immediately obvious to all who are watching, which draws in a lot of traders who are keen to act, thinking the Event is all there is to the price move.
Of course, the Iceberg model tells us otherwise, and we must look at the submerged levels of the Iceberg to figure out what really is going on.
The second level of the Iceberg details Patterns of behavior in the system.
In financial markets, this would be how a market trends over time. Trends are the context behind price movements, and crystallize seemingly random short term gyrations in price over longer periods of time.
This is extremely useful in two ways, the first being it shows a trader whether to go long or short over a relevant time frame, with the trend being your friend and all. The second being that the long term trend hints at broader factors at play in the market; a clue, so to speak, asking the trader to conduct further investigation.
Next, we have the third level of the Iceberg, representing Systems structure.
This level calls for market participants to map out each component in the complex system in order to understand the relationships between each, and how these relationships influence systemic outcomes.
For example, on a broad level, the components of the US economy would be the labor market, Corporate America, the banking system, government fiscal policy, the Fed’s monetary policy, and to some extent, capital markets.
It is important to remember that the relationships between these components will change over time.
Government fiscal policy is very different depending on which party has control over the legislative agenda, and it’s policies can have far reaching effects on the rest of the economy’s components.
Whether or not the labor market can keep its skills up to date with what Corporate America demands is another example of a changing relationship, with direct effects on corporations’ costs and productivity.
Relationships between components can run both ways, adding another layer of complexity. Keeping with the US labor market, it is quasi dependent on the US government and what education policies it adopts. Hence Corporate America is also affected by what the US government chooses to do with education.
It is crucial to note that in this globalized world, no economy stands as an isolated system.
This means that the US economy is itself a component in the global system. Which makes the parts of the US economy sub-components of a much larger whole whose interdependency, interrelationships, scale, and complexity is truly mind boggling.
The final layer of the Iceberg is Paradigm, and it is key to understanding and changing all complex systems.
So important in fact, that it needs its own article to explain fully and properly.
To be continued…
Finally, we have come to the deepest part of the Iceberg – Paradigm.
Paradigm is very special because it exists purely on an ideological level, but manifests itself physically throughout all the layers of the Iceberg above it.
Capitalism, for example, is an idea of how economies should be structured.
Societies that begin from this starting point then create institutions and entities that are required for a capitalistic economy to function. These are the components in the system, the Systems structure.
Once the components are up and running, their interactions over time coalesce into Patterns of behavior. These patterns then become the context behind observable outcomes, or Events.
As you can see, everything starts from an idea.
To see how this transpires in markets, replace “Capitalism” with “the Fed” in the above paragraph.
Markets believe the Fed is in control of all things economic and financial. Their entire worldview is structured around the Fed sitting in the center, that’s their current paradigm.
The result of markets holding such a paradigm is that everything is explained away using the Fed. For each level of the Iceberg:
Events: “Markets are up today because the Fed cut its target Fed Funds Rate by 25 bps”. This is a headline that should be familiar to all who read the financial media.
A reductive statement expressed in linear cause-and-effect terms used to explain real time happenings in the markets – this is exactly what the first, and only, visible layer of the Iceberg represents.
Patterns of behavior (market trends): “Treasury yields continue their downward trend as the Fed resumes QE” Another headline that should be familiar.
US yields have been moving lower for slightly longer than three decades now, with the Fed’s QE programs taking long term yields closer and closer to zero, and short end yields bouncing around zero/negative.
For whatever reason(s), USTs have been well bid for a very long time. But, because of the Fed-centric paradigm, the downtrend is, and has been attributed to Fed policy (and will continue to be until the paradigm changes).
System structure: What are the components of a Fed-centric financial system? A macro view of the system would include the Fed, banking system, capital markets, and the economy.
Interestingly, the diagram immediately shows that the Fed-centric view isn’t actually Fed-centric, simply because the Fed can only affect the economy and capital markets* through the banking system.
For instance, the Fed-centric view holds to the idea of Fractional Reserve Banking, which works in theory but not in practice. Instead, money is created from new loans made by banks, which is congruent with what we see in the diagram, where the banking system sits in the middle, not the Fed.
This precisely illustrates the power of Paradigm – it doesn’t matter if a theory works or not, it matters that people think it does.
It doesn’t matter if the Fed sits in the middle of the system or not, it matters that people believe it does. Every thought, line of reasoning, and inquiry into what happens in markets is thereafter erroneously built upon that belief.
Given how powerful beliefs are, and how deeply they influence our thought processes and behavior, it isn’t surprising that they are extremely difficult to change.
Similarly, Paradigm, at the deepest point of the iceberg, illustrates that it is the part of complex systems that are most difficult to change. Which also means that when it does change, the effects are profound, explosive, and more often than not unforgettable.
Which begs the question – What happens when the market’s paradigm changes?
*This is not always the case. The Fed can choose to purchase whatever securities it wants from the capital markets, although most of the time they choose to do so through the Primary Dealer network , whose members are all big Wall Street banks.