Is Investing a Science?

To address this question Andrew Berkin (head of research at Bridgeway) and I wrote an article that first described the scientific method and then discussed its application to investing. The article was published in the Fall 2015 issue of the Journal of Investing. The following is an abbreviated version. I hope you find it helpful.

You’re probably familiar with the phrase, “Past performance does not guarantee future results.” This statement, or one like it, is found on almost all mutual fund materials given out to individual investors. And for good reason: Returns of asset classes can vary dramatically from quarter to quarter and year to year. Yet, investors are tempted to read too much into past performance because they’re looking for greater certainty about the results of their investing strategy. Wouldn’t it be great if investing could be distilled into a set of laws that govern results in a predictable fashion, similar to what happens in the hard sciences such as chemistry or physics?

Unfortunately, we do not think investing is a science—at least not in the sense of the natural or physical sciences. Scientific phenomena are governed by the rules of nature (think Newton’s Universal Law of Gravity), which don’t change. Investing is contaminated by the complicated interactions of investors. This is why science has laws, while investing has hypotheses and models.

However, even if investing isn’t a science, we can address some of its uncertainty by applying the scientific method to our investment approach. The scientific method is a system for answering questions about the world through careful observations and rigorous experiments. The basic steps of the scientific method include asking a question, doing background research, constructing a hypothesis, testing that hypothesis, analyzing the data and then sharing the results.

Scientists conduct research in this fashion in order to reach objective conclusions based on the evidence. While scientists may follow their intuition when first developing a hypothesis, the framework for testing those hypotheses is entirely objective. In a similar fashion, going by your gut instinct isn’t enough when analyzing investments; you need objective testing and analysis to gain confidence that your hypothesis is correct.

For instance, we might start our scientific investigation by asking which investment strategy has the highest odds of successfully achieving our goals with the least amount of risk. Our next step would be to research the current thinking around the question and develop a hypothesis about what helps drive investment results. That hypothesis then needs to be tested, which we can do through careful analysis of existing financial data. Ultimately, this research will generate a conclusion that we can communicate to our peers for review and criticism. If the result holds up to this scrutiny, we now have a model that can give us greater confidence in our investing decisions.

Value investing is a great example of how the scientific method influenced investment strategy. Investing legend Benjamin Graham touted the benefits of value investing as far back as the 1930s, but there was no evidence to explain the value effect. Even the development of the Capital Asset Pricing Model (CAPM) in the 1960s could not explain this effect. But in the early 1990s, economists Eugene Fama and Kenneth French used a systematic approach to show that the value effect existed in U.S. market data dating back to the 1960s. Further, they proposed that the higher returns offered by value stocks were a reward for bearing risk.

Their work received much interest—and much criticism. So the authors, together with Jim Davis, analyzed data going back even further and found that the same effect persisted all the way back to 1927. What’s more, the value effect held true even when they looked internationally across different stock markets. This rigorous and continued application of the scientific method has shown value investing to be both persistent and pervasive, and the Fama-French model has continued to hold up more than 20 years after its initial publication.

Of course, there are no guarantees in investing. But applying the scientific method has helped us move investing away from the anecdotal and into the realm of evidence-based analysis. We may never be able to predict the future with financial markets, but we can continue to conduct research that increases our understanding of likely results and their potential variability. That, in turn, can give us the greatest possible confidence in the outcome of our investments.