Ideas Made to Matter
Data processing boom in investing favors large firms, slights small companies
How do investors know whether they’re making a good investment in a public company? These days, the answer lies in data.
“Data collection and data processing have really exploded in the finance industry,” said MIT Sloan assistant professor of finance “There has been a huge technological improvement not only in producing data but also in data processing technology, and that raises an important question: What firms are benefitting from this trend?”
Farboodi is the co-author of new research that shows that investors are choosing to process data about large, high-growth companies most often, slighting smaller firms.
“Data gives investors an information advantage, [but] we find there’s a huge divergence,” Farboodi said. “It’s big, large-growth firms that have the most data being processed about them,” whereas the trend has really stagnated for small stocks.
The lack of data processing about small firms — which have an important role in job creation — puts them at a comparative disadvantage when it comes to raising the financing required to sustain growth.
In their published research — titled “Where Has All the Data Gone?” — the authors explain how they arrived at this finding: by developing a statistical measure of the amount of data.
Learning how to measure data in finance is complicated; there aren’t good ways of doing so, Farboodi said. Previous attempts have relied on measuring the volume of news stories, expenditures on information technology, or analyst coverage, with the thinking being that increases in these areas point to a higher amount of data available on a given company.
But these methods don’t directly measure how much information investors were able to learn about a particular stock, Farboodi said.
The model she developed with colleagues measures the amount of data that investors process about various companies and shows how the amount of data can be inferred from “price informativeness” — that is, how closely current stock prices reflect future firm outcomes. The authors’ model can be used across a wide range of scenarios and asset characteristics. For example, investors could take a group of assets on the S&P 500 or Nasdaq composite and see how much data investors are processing about a particular asset class.
The authors’ model shows that a company’s size and growth prospects are what make data about that company valuable to investors.
With this in mind, the authors looked at cross-sectional patterns in data in the U.S. equity market from 1962 to 2016, taking stock prices from the Center for Research in Security Prices. They measured stock prices at the end of March for each year and accounting variables at the end of the previous fiscal year.
Then they grouped companies by size, based on whether they belonged to the 500 largest firms in terms of market capitalization. Size is an important factor for investors because it determines the amount of value available to profit from, provided their information on the company is good.
“Investors can more easily trade large positions on the equity of larger firms,” the authors write. These companies “are more valuable to learn about because investors can make larger trades on such assets to exploit their informational advantage.”
Second, they classified companies into high-growth and low-growth categories based on their book-to-market ratios. Growth matters for data because it scales with the earnings-to-valuation ratio. “Firms with high [growth] have prices that are a high multiple of earnings and therefore have prices that are very sensitive to earnings news,” the authors write.
Investors are more interested in large, high-growth firms
The authors found that a company’s size and growth affect how investors perceive its value. Specifically, in the past five decades, investors have been disproportionately more interested in learning about large, high-growth firms.
Larger firms are more valuable to learn about, especially if they are expected to grow faster. “Both size and growth increase the value of information, which is also amplified by their interaction,” the authors write. “The combination of being large and growing quickly makes a firm a desirable target for data analysis.”
There are a number of companies in the S&P 500 index that wouldn’t benefit from this trend, such as Walmart. “Walmart is a huge firm, but it is a value firm, so it does not have a lot of huge growth prospects,” Farboodi said. “The value of data processing is not improved that much because there’s not much more to learn about them. Investors don’t have that much of an incentive.”
The situation is “more concerning” for small, publicly traded companies that have high-growth prospects but still fail to attract investors’ interest, she said.
Farboodi stressed that it’s important that investors channel funds toward small firms with high-growth prospects and not overlook them. Many would make promising investments and contribute overall to economic growth. “You don’t want funds to be taken away from them,” she said.
In that vein, one area bucking this trend is venture capital firms. These companies process an exorbitant amount of data about startups, Farboodi said, noting that those companies didn’t appear in her team’s research because they aren’t publicly traded.
In the future, Farboodi plans to conduct research that examines how different companies benefit from big data and digital technologies, such as how a company’s use of a certain technology affects profitability.
Farboodi’s co-authors are Adrien Matray of Princeton University, Laura Veldkamp of Columbia Business School, and Venky Venkateswaran of New York University’s Stern School of Business.
Read next: Big data helps large firms grow even bigger