We increasingly see claims low volatility in the markets may be structural. Even as we agree that some of the analyses we see make good points, we are concerned we may be setting ourselves up for a major shock. Let me explain.

Before getting into the details about the current environment, let me make some general observations. In my experience, complacency, with its cousin low volatility, is the best bubble indicator I am aware of. Perceived safety gets investors to pile into investments that they later regret. When it happens on a massive scale, major market distortions may be created that can lead to financial crises. And as the tech bubble that burst in 2000 shows, even if there is no systemic risk, the unwinding can be most painful to investors. In 2008, however, the perception that home prices always had to rise had become engrained in highly levered, yet illiquid, financial instruments, causing the unwinding to bring the global financial system to its knees. In our assessment, however, make no mistake about it: assuming for a moment the next crash won’t take down a major bank, doesn’t imply it cannot wipe out much more of your wealth than you ever anticipated.

Indeed, lower bank leverage is given as an argument as to why volatility is lower these days. Except that the run-up to the financial crisis of 2008 – a period in which banks were extra-ordinarily levered – also showed low volatility in a variety of markets. It was the perceived safety, embodied by quasi-sophisticated value-at-risk (VAR) models that got risk managers at financial institutions to gear up. What could possibly go wrong, right?

A more convincing argument I hear as to why low volatility is structural may be that information nowadays gets absorbed more quickly. On the one hand, we have computers scan the news in milliseconds, often trading without human intervention. And we have more computing power, allowing for a more efficient implementation of any investment process. Market makers in exchange traded funds also help in the execution efficiency of markets, possibly exerting downward pressure on volatility. However, let’s not forget that volatility lowered in this fashion may have the same implication as low volatility in the building up of any bubble: it is the perceived risk that is lower, not actual risk. Machines are fantastic at certain aspects, be that keeping spreads tight in an exchange traded fund, or scanning Twitter for keywords. Trades initiated in this fashion provide liquidity to the markets, but that liquidity can evaporate rather quickly when the machines go off-line. Let there be a glitch in the markets for whatever reason (say, someone dumps a large number of derivatives in off hours), and today’s incarnation of automated traders tend to wait it out. In the meantime, stop loss orders of other market participants may be triggered, possibly causing flash crashes.

One argument is that low volatility is due to a lack of economic surprises. We get emotional about elections, but when the economy is stuck in a low growth environment with few deviations, it only makes sense for markets to be less volatile as well. That may be the case, but in our assessment won’t stop from investors pushing asset prices to unsustainable levels on that backdrop.