Turning Weak Signals into Strong Predictions
Why some machine learning models unlock economic forecasting potential.
Turning Weak Signals into Strong PredictionsOfficially, the 2010 Dodd-Frank Act was passed into law as the Dodd-Frank Wall Street Reform and Consumer Protection Act. But portions of the legislation don’t relate directly to reforming Wall Street nor to protecting consumers, at least not in the sense the US Securities and Exchange Commission has traditionally sought to protect them. These sections deal instead with the disclosure of nonfinancial information—specifically, the disclosure of socially undesirable practices such as the use of conflict minerals or a failure to ensure worker safety. If the aim of these sections is to bring about socially beneficial change, the evidence suggests that approach is working.
Chicago Booth’s Hans B. Christensen and Mark G. Maffett, Booth PhD candidate Lisa Yao Liu, and Rice University’s Eric Floyd examined how Dodd-Frank has affected the safety of workers at US mining companies. Among its regulatory imperatives, Dodd-Frank requires all mine-owning public companies to report mining-related injuries and safety citations as part of their annual 10-K and quarterly 10-Q disclosure forms. In addition, these companies must issue an SEC 8-K filing whenever they receive an “imminent danger order,” an event that signifies a particularly serious breach of safety. Comparing mines owned by companies subject to Dodd-Frank with mines owned by companies that are not, the researchers find that the additional reports required under Dodd-Frank led to an approximately 11 percent decrease in mine-safety citations and a 13 percent drop in mining injuries.
Although the SEC filings merely duplicate safety reports already available in public records—mine operators are required to report safety information to the Mine Safety and Health Administration (MSHA), which makes the documents public via its website—they increase investor awareness of safety issues in ways that affect the company’s appeal as an institutional investment and its stock price, according to the study.
For example, the researchers find that when mining companies disclose imminent danger orders through 8-K forms, overall mutual-fund ownership of the disclosing companies falls more than twice as much as it does if the order is disclosed through the MSHA website alone. The disclosures have a particularly strong effect on managers of mutual funds dedicated to socially responsible investing.
The study also finds that the median stock-price reaction is 140 basis points more negative when a safety citation is reported in both an 8-K filing and on the MSHA’s website, compared to when citations are disclosed only on the website.
The heightened awareness generated by the SEC filings may be due to the fact that such filings are simply a familiar vehicle for information among investors. “SEC-required disclosures on forms 8-K, 10-Q, and 10-K are effectively the billboards of the financial community,” the authors write. By contrast, investors may be unaware of the safety data posted on the MSHA website, or they may find it difficult to parse.
Given the relevance of mine-safety disclosures to investor sentiment, mining companies have a clear incentive to avoid them by improving their safety records. But those improvements come at a cost: the researchers document a 0.9 percent drop in productivity relative to privately held mines that are not required to make safety disclosures through the SEC.
Nonetheless, by regulating transparency, the SEC appears to have cleared the way for the market to, in some sense, help regulate safety. It’s a strategy that has applications for other industries as well. “The idea is that, once [regulators] remove the frictions that prevent information discovery, market forces can be an effective means of reducing socially undesirable behaviors,” the authors write.
Why some machine learning models unlock economic forecasting potential.
Turning Weak Signals into Strong PredictionsAn expert panel discusses the yet-unknown and steps policy makers should take.
How to Design Policy in the Age of AIBig companies try to impose discipline on spending by setting budgets for everything, but the rules lead to lots of waste.
How the Budgeting Tail Wags the Corporate DogYour Privacy
We want to demonstrate our commitment to your privacy. Please review Chicago Booth's privacy notice, which provides information explaining how and why we collect particular information when you visit our website.