Using Data for Good Is Good for Business

We hear a lot about businesses looking for ways to tap into data to make better decisions. For Bloomberg, financial data is not just about the numbers that make up corporate profits. It’s also about the metrics of a company’s sustainability in terms of its social and environmental impact.

A recent Bloomberg article, “Sustainable Investing Is Booming. Is It Smart?” details the rise in “SRI (socially responsible investing) and ESG (environment, social and corporate governance).” Citing the Forum for Sustainable & Responsible Investment, the article points to a jump of 28 percent for “U.S. investment funds incorporating ESG criteria” whose assets have grown more than fourfold to $4.3 trillion in the two-year period ending in 2014.

Bloomberg data contributes to that trend, as it points out in customers using ESG data increased 76% in 2014. That increase is measured over a span of five years, from 2009—when only 2,415 customers used such data—to 2014, when the number rose to 17,010. Bloomberg delivers ESG data on more than 11,000 companies as it integrates ESG data into its Equity Screener (EQS), Portfolio Analytics (PORT), Relative Value (RV) and its analytical models that include an ESG Scorecard.

“Doing good is good business,” declares Gideon Mann, head of Data Science at Bloomberg. He explains that tracking companies’ ESG can “make a difference on a global scale.” As a result of “the way global capital is allocated,” investors can “agitate for change.”

Mann explained the factors that go into the ESG Scorecard. The E is about concerns for the planet that include what the company is using as its source of power and how it is managing waste. For manufacturing companies, a concern would be the pollution a firm produces and how it disposes of it.

Social concerns include how a company treats its workers, its commitment to diversity, its attitude toward labor unions, etc.

For governance, they look at the structure around the board and the answers to questions such as these: Are there any conflicts of interests at play? Is the board too closely tied to the government? Does the chairman wield too much control?

Creating Analytics to Understand the Market

Data for good at Bloomberg started by creating analytics to understand how markets works and introducing more access to information. That type of transparency made it “less about who you knew and more about what you knew in financial markets,” according to Mann.

When asked about examples of the effect of corporate scoring, Mann offered Walmart as “an interesting example.” As Bloomberg’s report on the retailer’s initiatives shows, there are different views on it.

“Some investors see Walmart as an innovator because of its efforts to green its supply chain and invest in renewables, as well as offer organic products,” he explains. However, others maintain that the brand that is so closely linked with low prices is still “too dependent on low cost, low wage, low quality, and provoked regulatory attention and censure as a result of union busting, for instance.”

According to that view, Walmart’s social component is not up to par. Consequently, its “customers are the same as their employees, too squeezed and poorly paid to even afford Walmart.”

In addition to trying to direct socially conscious investors toward companies that share their values—effectively rewarding companies committed to sustainable goals—Bloomberg contributes to the exchange of information that can show how data can be used toward the same end. That’s the idea behind the Bloomberg Data for Good Exchange.

It held a workshop in New York City in September 2015. There, it brought together data science experts “with the NGO [non-governmental organization], public sector and nonprofit partners that can benefit the most from applied data science.”

Greater pools of data, such as those offered by cities, provide even more learning opportunities. As an example, Mann points out that with around 8 million people in New York City proper, every hour generates huge quantities of data in relation to where people are living, working, in transit, etc. When that kind of data is brought to bear on solving problems, it can prove instructive—not just for one city, but for all cities.

That was the case when New Orleans applied data-driven methods to deal with the backlog of damaged buildings in the wake of Hurricane Katrina. Using data models, Mann explains, the city was able to more quickly arrive at decisions of whether to sell or knock down each building. New Orleans also learned from New York’s example of targeting buildings to prioritize delivery of smoke alarms.

Sharing information and solutions is what makes it possible for data science to improve the delivery of services within the city.

Whether it is a question of city services or businesses investments, “good” choices now have measurable metrics for environment, social and governance issues, as well as for the bottom line.