If you’ve ever read academic papers or journals about the finance and investing industry, thinking you’ve finally understood all the exotic variables that govern stock market returns, chances are you’d be wrong. That’s because, as Duke University Finance Professor Campbell R. Harvey argues, most of the research findings on the topic published in financial economics “are likely false.” Shaking the foundation A recent research paper Harvey co-authored with professors Zhu Heqing and Liu Yan titled “… And The Cross Section of Expected Returns” pointed out a significant statistical mistake which many finance researches have committed. This includes even…
If you’ve ever read academic papers or journals about the finance and investing industry, thinking you’ve finally understood all the exotic variables that govern stock market returns, chances are you’d be wrong.
That’s because, as Duke University Finance Professor Campbell R. Harvey argues, most of the research findings on the topic published in financial economics “are likely false.”
Shaking the foundation
A recent research paper Harvey co-authored with professors Zhu Heqing and Liu Yan titled “… And The Cross Section of Expected Returns” pointed out a significant statistical mistake which many finance researches have committed. This includes even Harvey, who said, “Everybody’s in the same boat, including me. My previous research suffers from the same problem.”
The issue with most financial economics research, which Harvey, Zhu, and Liu found, is this. Say you’re now trying to find the cause for a particular phenomenon. Typically, most researchers would be satisfied if a variable they’re testing survives a 95% confidence level test, meaning that there’s only a 5% chance that a positive finding has happened because of a random stroke of luck. That means there’s a very slim chance that your statistically significant result is actually a false positive.
But a problem arises when researchers start testing for hundreds, or even thousands of variables (something which finance researchers are likely to do). With a 5% chance of having a false positive with each test, the more theories that get thrown onto the wall, the likelier it is that a researcher would end up with a “statistically significant” result that is actually a false positive.
In other words, if enough data is mined, even ridiculous examples like, say, a decline in the number of pirates worldwide, might also be found to have a relationship with stock market growth.
The practical implications
What can his findings mean for us investors though? This is what Harvey thinks:
“The implications are provocative. Our data mainly focuses on academic research. However, our paper applies to any financial product that is sold to investors. A financial product is, for example, an investment fund that purports to beat some benchmark such as the S&P 500. Often a new product is proposed and there are claims that it outperformed when it is run on historical data (this is commonly called “backtesting” in the industry).
The claim of outperformance is challenged in our paper. You can imagine researchers on Wall Street trying hundreds if not thousands of variables. When you try so many variables, you are bound to find something that looks good. But is it really good – or just luck?”
Delving into the truth
In light of all the above, it’s perhaps fair to backtrack and think about one of, what seems to be, the most powerful drivers of stock market returns which is succinctly summarized by Warren Buffett as follows:
“If a business does well, the stock eventually follows.”
Does this relationship still hold water? I’d argue yes. One of Harvey’s thoughts about financial research is that there has to be an “economic foundation” which goes behind the thinking of what variables might affect future market returns.
On that note, it’s hard to argue against a company becoming more valuable as the amount of profits and cash flows it can generate increases over time. And since stocks are simply partial ownership stakes in companies, it thus stands to reason that a company which does well would have a stock price that tags along eventually. On the reverse, a company which does poorly, would also see its share price performance suffer.
I once did a study of Singapore’s best- and worst-performing shares over a period of more than 10 years stretching from the start of 2004 to the start of June 2014. Amongst the top 10 best-performers were Raffles Medical Group Ltd (SGX: R01) and MTQ Corporation Limited (SGX: M05). Meanwhile, the worst two happened to be Surface Mount Technology (Holdings) Ltd (SGX: Q7Q) and Metech International Ltd (SGX: QG1). This is how their business and share price changes look like:
|Share||Total return||Earnings per share growth|
|Surface Mount Technology||-99.9%||-99.6%|
Source: S&P Capital IQ (data covers the period 1 January 2004 till 1 June 2014)
I trust it’s obvious to see that it’s the businesses that do well which have the good share price returns and vice versa.
Foolish Bottom Line
There would never be a shortage of spurious variables which purport to be able to predict future market returns. But what Harvey has found ought to make us think twice about deviating from the most basic but reliable driver of stock market returns – the share’s underlying business performance.
To learn more about investing and to keep up to date with what's exactly happening in today's market, click here now for your FREE subscription to Take Stock Singapore, The Motley Fool's free investing newsletter. Written by David Kuo, Take Stock Singapore can also show how you can GROW your wealth in the years ahead.
The Motley Fool's purpose is to help the world invest, better. Like us on Facebook to keep up-to-date with our latest news and articles.
The information provided is for general information purposes only and is not intended to be personalised investment or financial advice. Motley Fool Singapore writer Chong Ser Jing owns shares in Raffles Medical Group.