Loopy World
“Civilization advances by extending the number of important operations which we can perform without thinking about them” — A.N. Whitehead
Global markets are a complex dynamic system connecting large numbers of participants through a vast network. This giant network is rife with interrelated feedback loops of various complexities and time scales, from high-frequency trading to low-frequency asset allocation and policy decisions. It is akin to an atmosphere comprising giant structures, medium-size swirls, and tiny vortices that connect to each other, and brings to mind the chaos theorist’s tale of the butterfly wing flap somewhere causing a hurricane on a different continent.
The complexity of the whole comes from the fact that these loops interact in ways that are not always well understood, and, further, from the fact that, even when interactions are understood, they may result in intricate dynamics that are inherently chaotic. The phase-space diagram of a simple 3-D Lorenz model provides a feel for how such loops can interrelate. One wonders if this diagram inspired the butterfly story in any way.
Chart 1: Illustration of a Lorenz Model
Complexity also comes from the vast and growing amount of information generated by markets every second, causing an information overload, making the task of generating insights and meaning more and more challenging. Focusing on the low-frequency part of the spectrum, below are three examples of loops at play in markets:
The Low-Interest Rate Loop
This loop, also called the QE trap, is blamed for many ills, one of which is harming macro investors. Reporting on the closure of Louis Bacon’s hedge fund Moore Capital, the Financial Times wrote: “The post-crisis era of low interest rates, muted economic growth and subdued market volatility has proven difficult for these funds”. There is a loop implied in this statement, and it is as follows:
Central Banks impose low interest rates => Investors sell implied volatility => Implied volatility falls => Realised (actual) volatility follows => Stagnant markets => Low growth => Low inflation => Low rates etc.
The first arrow is due to a mechanical link, namely the options market, as investors sell volatility in the form of options to generate yield. The following arrows are plausible, although not always causal. The quote implies that low volatility hurts macro hedge funds. Yet stagnant markets stem from low interest rates, a powerful macro trend. Simply put, the narrative suggests that macro hedge funds are suffering because of one of the largest macro trends in modern times, i.e. low interest rates induced by QE. Much has been written about the other effects of QE, such as the unhealthy accumulation of debt in the economy and the incentive it gives corporations to buy back their own shares, creating another feedback loop with potentially cataclysmic consequences if left unchecked. As central bankers and economists know, breaking the QE loop is no small task. It will require more application, co-ordinated policy efforts, and probably more time. The chart below puts US interest rates and equity volatility in perspective since 1990. Today, short-term interest rates are low, but no longer plumbing the depths of the 2009–2016 period. Longer-term rates, however, are near historic lows, causing the infamous yield curve inversion.
Chart 2: US Interest Rates and the VIX Index
The next chart shows an interesting ratio, the 2-year US interest rate divided by the VIX index, which measures the relative cost of equity options to interest rates. When this ratio is low, investors are incentivized to sell volatility to generate yield. When it is high, investors are incentivized to use the proceeds of high interest rates to buy volatility, via protected notes, for example. More on this in another article. For now, suffice to say that this ratio can be thought of as a switcher from the low interest rate, low volatility mode to a high-interest rate, sustained volatility mode. Today, the ratio is up from all-time lows, but not yet at the level of the 90’s during which the latter mode prevailed. We are still stuck in the former mode: low interest rates, low volatility.
Chart 3: The Cost of Money / Cost of Volatility Ratio
The Risk-Based Asset Allocation Loop
The growing popularity of systematic investing means that a growing number of asset allocation models, funds and indices are run using risk-based engines such as risk parity, mean-variance optimization and the like. Consider the following thought experiment. Imagine a simplified single-currency world running on a risk parity model, where the amount of capital going to a particular asset class is inversely proportional to its volatility. Capital would flow towards safe assets such as treasuries, and away from volatile assets such as emerging markets and commodities. Within equities, stable stocks would get more capital than volatile ones. As time goes on, the model would self-reinforce, with safe assets such as treasuries getting more and more capital, pushing yields lower, while volatile assets would get less and less capital, making them even more volatile. The process would follow the loop below:
Capital flows to safe assets => Yield of safe assets falls => Volatile assets get less capital, become even more volatile => Etc.
This cycle would go on until an outside intervention, such as an IMF or World Bank bail-out, comes to the rescue, injecting capital in the cash-starved, volatile segments of the system, so the cycle can start again. Looks familiar? Of course, the real world is more complex than this, with multiple currencies, complex flows, trade wars and multiple ways in which capital is allocated. But this simple loop oddly resembles what is actually happening in the real world, suggesting that some risk-based logic seems, somehow, to apply at a global asset allocation level.
The Evidence-Based Investment Loop
The ongoing process required to perform asset allocation over time can also be described as a loop. At ALPIMA, we call it the OMDA loop, which stands for Observe, Meet, Decide and Apply. It looks like this:
Chart 4: The ALPIMA OMDA Loop — Evidence-Based Investment Loop
You may note the resemblance with the OODA loop devised by American military strategist and USAF commander John Boyd during the Cold War, and now recognized as having wide-ranging applications in business, politics and law enforcement. The key difference here is that the adversary is not led by a single entity. It is changing markets that are recursive in nature and whose level of unpredictability is random, thus creating the need to regularly re-assess the asset allocation. The OMDA loop is what engineers call a feedback control loop, a classic concept in control theory. Helping investors operate this loop effectively requires that the right data be available to them in order to determine the best course of action at any given time. At ALPIMA, we do this via comprehensive and interactive dashboards that update dynamically and show relevant data on risks, performance, contributions, sensitivities to markets, correlations, etc. The OMDA loop is an evidence-based investment loop, helping investors make better, more informed decisions and avoid the long and well-documented list of biases known to bend human judgment (anchoring, risk aversion and the availability heuristic, to name but a few).
Bringing It All Together
Multiple interconnected loops, increasing complexity, an explosion of information, the rise of quantitative investing and stubbornly persistent human biases — All of this calls for a new generation of tools with smart analytics and dynamic visualization, in order to help those making investment and allocation decisions see the forest from the trees.
It is precisely with this in mind that we have built the ALPIMA platform and evolve it over time thanks to the valuable feedback of our clients and partners.
Contact us at info@alpima.net if you would like to find out more.
