Page 3 of 4

Newer firms like London-based Generation Investment Management, founded a decade ago by Gore and former Goldman Sachs Asset Management CEO David Blood, have experimented with additional techniques for assessing sustainability information in selecting investments — and have chalked up superior returns. Some managers are experimenting with algorithms that can integrate and correlate financial and sustainability data in picking long-term winners. Arabesque Asset Management is billing itself as an ESG quant shop on the claim that sustainability information is now of adequate quality and quantity to extract patterns and generate actionable information for investment decisions. Omar Selim founded the London-headquartered firm in 2013 based on an asset management project he had developed at Barclays Bank and tapped Harvard professor Eccles as chairman and former Global Compact director Kell as vice chairman of its board.

“ESG is a new set of information that people would have integrated long ago if they had had it,” says Andreas Feiner, a founding partner and head of ESG research and advisory at Arabesque. “It really follows in the tradition of Benjamin Graham’s concept of value investing, but now you have the capacity to get qualitative information in a quantitative way — about all the different forms of capital — that provides a portrait of the future of a company’s ability to generate cash flow and profits.”

A final driver of the shift to a much deeper understanding of the relationship between investing and value creation is coming from the immense and still rapidly growing world of pension funds.

“In many ways it goes back to John Maynard Keynes in the 1930s,” says Keith Ambachtsheer, an adjunct finance professor and director emeritus of the Rotman International Center for Pension Management, and author of The Future of Pension Management, released this month. “He always wanted to move away from the beauty contest of short-term trading, to focus on the creation of wealth. With an integrated approach to thinking and investing, it is not about reporting on social issues versus financial issues but about arriving at a complete story. That’s really the key: How do you get a holistic view?”

While day-to-day skirmishes are taking place over the value of ESG information, a much larger metamorphosis is overtaking the capital markets that will reshape not only how investments are made but the face of capitalism itself. This transformation will happen when all the forces of change, each gathering strength separately, finally converge: as investors demand more-­comprehensive sustainability information, as stock exchanges put pressure on companies to release it, as standard-setters render that information more reliable, as companies begin to generate billions of new data points, as these data points are collected and released through Bloomberg terminals and other mechanisms, and as they are eventually correlated with other data from government, social media, user logs, geographic information system maps and market transactions. Already, the world creates an estimated 2.5 exabytes — or 2.5 billion gigabytes — of data every day, meaning that in the past two years human beings have produced 90 percent of the data ever created.

There are plenty of voices in the marketplace that harken back to a simpler time and argue that this is all a waste, and that the only solution is to move back to tighter definitions of materiality, to smaller pools of information and to narrower definitions of corporate purpose. Even the greatest proponents of big data understand that the raw information piling up inside acres of hard drives is useless in itself. As Steve Jurvetson, a venture capitalist and an early investor in Tesla Motors, said during a conference on deep learning at Stanford Graduate School of Business in 2014: “There is nothing really exciting about it until we have a methodology to make use of it, because we can’t understand big data. None of us wants to read through a phone book.”

To comprehend the data, analysts will increasingly turn to advanced artificial intelligence. These programs are constantly learning from new information and can offer analyses in a rapid manner. The better the program, the more valuable the information and connections that can be derived and given to decision makers. A current limitation in the data science community is the processing speed required for some of the artificial intelligence programs being employed. Currently, artificial intelligence is not advanced or efficient enough to fully gather all the information possible from these massive sets of data, and large investments must be made either to implement proper computing systems or to rent cloud computing services from a company like

To address such concerns, enormous resources must be pooled and focused. In 2015, JPMorgan Chase & Co., Goldman Sachs and Morgan Stanley decided to form a data company so that they could all save money instead of cleaning and formatting their own data. Bloomberg has created a data science unit that is already facing crushing internal demand for its services. “We are at this weird point,” says Curtis Ravenel, Bloomberg’s head of sustainability. “Some people think that there is too much being disclosed. Others worry that the amount of information has exceeded our ability to do anything with it. So this means that we need better tools, because big data is going to change everything.”

Whether we like it or not, and whether we are ready or not, we have entered a universe in which quadrillions of data points are gathered on almost every aspect of our individual and institutional lives. Right now tens of thousands of people are laboring to discover what kinds of insights such big data might offer. Despite the mighty challenges in addressing data quality, availability and connectivity, the central questions that we now face as a society are philosophical: What do we want to know, and why? What are the signals that we are looking for in the middle of this ocean of noise?

We could extract intelligence for many reasons: to advance our own interests, to understand markets better, to create improved products, to predict the future, to expand our control and to make money. All those incentives will drive the system of big data forward. But here is the surprising reality: In addition to all of that, we also could employ the ever-improving field of data science to understand the complexities of business and capital markets in breathtaking new ways.

There already is a strong drive from several quarters to place corporate and investor performance in their larger regional or global contexts. Many of the disclosure instruments invented over the past two decades, including the GRI, have focused on the release of aggregated information at the institutional level, which is most relevant to investors. In the past the push was for a company like Ford Motor Co. to gather all the information from all of its far-flung units and report on its total energy use. But now, with the creation of global goals on environmental and social concerns ranging from the national greenhouse-gas targets of the COP 21 Paris meeting to the comprehensive Sustainable Development Goals for 2030 released by the U.N., the focus will expand to include the institutional contribution to the sustainability context — the impact within specific geographic limits and targets. Through this lens one asks less about the performance of a company as a whole and more on how its products — and research, marketing and capital expenditures — affect the overall greenhouse-gas emission targets of a particular place, such as France.

An equally promising project has been launched by Steven Lydenberg, a longtime pioneer in the integration of ESG information and founding director of Harvard’s Initiative for Responsible Investment. Lydenberg has proposed the bold Investment Integration Project to look at the collective action problem — how the combined decisions of portfolio managers often end up negatively affecting the stability and sustainability of whole industries and ecosystems.

As part of the evidence for this expansion of perspective, Lydenberg points to Bank of England governor Mark Carney, who in September surprised financial markets by highlighting climate change as a major threat to global economic security. Carney, who chairs the powerful Financial Stability Board, has launched the Task Force on Climate-­Related Financial Disclosures, chaired by Michael Bloomberg.

One of the most alluring opportunities offered by the new approach to sustainability data is to put real flesh on the conceptual bones set forth by the IIRC’s multicapital model. Today the idea of correlating the relative inputs and outputs of six forms of capital would seem conceptually valuable but computationally impossible. The creation of new data driven by new algorithms will allow us to explore previously hidden correlations that could have a major impact on value.

For example, though every CEO ritually announces that “our employees are our most important asset,” the real assessment of the value of human and social capital within companies has been skimpy. To tackle the task, one must ask new questions about the relationship between the complex needs and skills of employees and other forms of performance. Under what circumstances would training one’s own workforce provide a better stock price or a lower cost of capital? How could we tell which forms of work flow or decision making would best enhance all the capitals required by a business model to succeed? Could broader and more democratic participation and higher degrees of diversity increase a company’s ability to respond to a complex and rapidly changing technical and competitive environment?

Single Page    1 | 2 | 3 | 4