Turn on Javascript in your browser settings to better experience this site.

Don't show this message again

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Find out more

Human decision-making significantly impacts the markets in ways that quantitative models can’t address.

Avoiding the pitfalls of quantitative investing

  • 02janv. 18
  • Arne Staal Head of Multi-Asset Quant Strategies

The rise of the investment algorithm has coincided with a nine-year bull run. What happens if 2018 is the year of the bear? The one thing we know is that it will be humans, not computers, who will have to navigate through the next crisis.

Like mirages, above-market returns promised by new strategies can disappear as soon as you invest.

Technology helps… and hinders

Increased use of data and technology can provide real investment insights but also increases potential for discovery of false signals. Next year will see growing use of algorithmic investment strategies, which will be increasingly based on big datasets and machine learning. This can be a force for good: Quantitative approaches can help investors understand the true drivers of their portfolios and bring hope of better investment decisions and new sources of return. But the easier it is to access “big” data, the easier it is to be misled. Like mirages, above-market returns promised by new strategies can disappear as soon as you invest. Statistical and human biases conspire to produce fake discoveries in historical data, and quantitative approaches have no appreciation of a changing world.

Keynes’ description of economics applies equally to investment: “a science of thinking in terms of models joined to the art of choosing which models are relevant to the contemporary world.” The global financial crisis reminded investors of the limits of models. Economists failed to predict the crisis. And political decisions mattered more than economics in bringing an end to the downturn.

Models miss political interventions

Politicians change the rules in ways models cannot capture. The Asian financial crisis of 1997–1998 led to the International Monetary Fund providing financial help to Thailand, Indonesia and South Korea. Malaysia took a different route, imposing capital controls that pegged the currency but locked investors in.

If we look further back, post-war investors in most UK public utilities were forced to hand their investments over to the state in return for government bonds.

Occasionally, markets cease to function at all. World War I caught the whole financial system by surprise. London was the center of the financial world. UK banks had lent significant sums to German companies. The outbreak of war proved to be the Lehman moment, but without the safety net of the current web of central banks, governments and supranational bodies. The financial system froze, with 50 stock markets around the world closing, some for many months.

These risks remain relevant. In 2013, Cyprus imposed capital controls to limit outflows from its banking system. The UK Labour Party’s current policies include nationalization of railways and water utilities.

Don’t discount the human factor

But surely market closures are a thing of the past? In An Engine, Not a Camera: How Financial Models Shape Markets, Professor Donald MacKenzie tells one story from the crash of October 1987, when the S&P 500 Index fell 22% and the S&P 500 futures contract fell 29% in trading on the Mercantile Exchange. This left several securities firms close to failure. There was a risk the exchange would be unable to open the following day. Continental Illinois Bank was the exchange’s clearing agent and owed $400 million. With minutes to spare, the chairman of the bank stepped in to release the funds needed to allow the exchange to open. As MacKenzie puts it, “The resolution of the crisis shows something of the little-understood network of personal interconnections that often underpins even the most global and apparently impersonal of markets.”

When things go wrong, humans step in. A similar network kept the massive failure of hedge fund Long-Term Capital Management from bringing the financial system to a halt. Human intervention kept the doors open at the Royal Bank of Scotland and other financial institutions during the global financial crisis.

Tools, not solutions

Quantitative investing is rapidly replacing traditional active management. Algorithms help reduce the bias and noise that hamper human decision-making. Information technology empowers us to back our judgments with empirical evidence. But the science must make sense and be used in context. In a world still driven by humans’ fear and greed, investors need to understand the limits of quantitative finance.

Into 2018, big data is set to become bigger. Quantitative techniques have become vital tools for today’s investors. But they are tools, not solutions. Successful investing requires an understanding of both the numbers and the context in which they are used. This requires judgement. Data scientists have a role to play, but investment still needs investors.

Important Information

Foreign securities are more volatile, harder to price and less liquid than U.S. securities. They are subject to different accounting and regulatory standards, and political and economic risks. These risks are enhanced in emerging markets countries.

Companies mentioned are for illustrative purposes only and are not intended to be a recommendation to buy or sell any security.

Indexes are unmanaged and are included for illustrative purposes only. You cannot invest directly in an index.

ID: US-281217-54742-1

This Content Component encountered an error