The Way Businesses’ Social Performance Gets Measured Isn’t Working

VISIT | The Valuing Respect website 

Companies’ social performance matters!

Thirty years ago that might have been viewed as a radical campaign slogan. Today it is a pretty mainstream observation. The reality of how business activities can negatively affect the lives of people across the world has become inescapable. 

Business conversations at major gatherings such as the World Economic Forum are increasingly focused on the purpose of businesses within society. We’re a long way from the days when most companies assumed they should externalize as many social and environmental costs as the law allows in order to maximize their profits. Today, business leaders recognize how self-defeating that proposition is. As Unilever’s Paul Polman has put it, “There is no business case for enduring poverty or runaway climate change.” 

In fact the Business and Sustainable Development Commission produced a 120 page report last year emphasizing the imperative (and opportunity) for business to get behind the global Sustainable Development Goals. A key pillar of their message was the need to strengthen the “social contract,” including by contributing to decent jobs, the eradication of forced labor and child labor, tackling discrimination and income inequalities, advancing access to clean water and healthcare and empowering women and girls.

Also Read

November 2016 | Publications

Business, Human Rights and the Sustainable Development Goals: Forging a Coherent Vision and Strategy

Read

Socially responsible investors have long promoted the need for companies to take ownership of reducing their impacts on people as well as planet. Now mainstream investors are joining the conversation as well, with a rapid growth in “ESG” (Environmental, Social and Governance) investment funds now available — not least in response to demand from millennial and women clients.

All good news, yes? Well, that depends.

Current measurement of companies’ social performance: at best superficial, at worst misleading

The growing attention to companies’ “social performance” – the impact they have on people – naturally raises the question of how to measure which ones are doing well and which are not. But most of the metrics being used to make these assessments are at best superficial and at worst misleading. Moreover, a proliferation of ratings, rankings and ESG products is embedding many of the same weak data into their methodologies.

As a result, companies are allocating critical resources to gathering and publishing information that gives us little insight into how their business actually affects the basic dignity and welfare of people. And markets are left rewarding often poor or inadequate behaviors, while leading practice can go unrecognized and under-supported.

That’s a major setback for champions – inside and outside business – of good corporate social performance. And it betrays the hundreds of millions of people who are most vulnerable to the negative impacts that can result from business activities.

What should we conclude from this? Perhaps it’s that knowing that companies’ social performance matters is only a starting point; it’s how we measure that performance that will be critical to progress.

From audits to training to big data: why these measurement approaches don’t tell the whole story

Right now, we’re falling short on the challenge of measurement – and we are doing so collectively. This is a shared problem across business, investors, regulators and civil society. There are no easy answers. If there were we would have them by now. But the need for better approaches is now urgent.

The UN Guiding Principles on Business and Human Rights, endorsed by the United National Human Rights Council in 2011, set the global standard for what is expected of companies (and governments) in advancing business practices that respect people’s human rights. So why, nearly seven years later, are we not doing a better job of defining what success looks like?

Why are so many analysts giving such weight to how companies spend their profits through philanthropic initiatives or employees’ community volunteering, as against how those same companies make their profits – and whether they respect people’s human rights in the process?

And even when the focus is on how profits are made, why are the metrics dominated by low value input/output data such as the number of hours of human rights training provided to staff, of human rights clauses included in contracts, or of impact assessments or social audits carried out at operational sites and suppliers?

Let’s take just the data produced from social audits of factories and farms in corporate supply chains. These are still viewed by thousands of companies and their investors (including many “social impact investors”) as the answer to addressing labor rights risks. Yet research has long since shown the inadequacy of relying on such data or seeing it as the means to progress.

Most companies will audit just their first tier suppliers, but most of the 40 million workers subjected to forced labor, and the 80 million children in dangerous forms of child labor, work at more remote levels of large companies’ supply chains.

Company managers can report audit data that show dramatic improvements at factories after “corrective action plans” are instituted. But those same managers will admit privately that they expect many of the same problems to be back within a year or so, since underlying causes are often not being addressed and their companies often move on from suppliers that invest seriously in improving their standards when that leads to rise in the price they are charged.

Also Read

August 2013 | Collaborations

From Audit to Innovation: Advancing Human Rights in Global Supply Chains

Read

Indeed, pretty much any leading company has frustrated managers eager to share stories of how superficial this information is as a means of measuring or driving positive social impact, or of comparing companies’ performance across an industry. Yet the appeal of social audit data continues: it provides easy, clear numbers that corporate leadership and investors can take as the basis for important business and investment decisions, ignoring the reality that they paint not a picture of true social performance but a mirage that masks a more unsettling reality.

The most accurate data related to human rights impacts often exists where abuses have already occurred: the numbers of workers killed in the collapse of Rana Plaza in Bangladesh, or in demonstrations at the Marikana mine in South Africa; the devastation of homes and livelihoods from the Samarco disaster in Brazil or the BP Deepwater Horizon blow-out; or the burgeoning number of alleged and proven cases of sexual abuse in workplaces across multiple industries.

These are all highly relevant data but they are lagging indicators: that is, they measure, after the fact, instances where respect for people’s human rights has broken down in practice. They don’t help us identify which company is the next one with a similar human tragedy on its way, nor which companies are doing the best job of making sure it isn’t them.

Moreover, many of these same problems are baked into the ways in which “big data” is used to gain insight into corporate performance. Cutting edge software is used to scrape the Internet for allegations of abuse and to train computers to use these inputs to gauge human rights performance. Again, this information is not without value, if used as an input to careful research. But when plugged into algorithms that draw direct and simplistic conclusions on the relative performance of companies, it carries (and conceals) enormous risk.  

This may seem a grim diagnosis of the situation, but it’s not a new one. At Shift we consistently hear from those we work with across business, the investment community and civil society that they feel deeply frustrated by the insufficient grasp of how to measure companies’ human rights performance. And various other organizations have highlighted the many symptoms of the problem.

This shared challenge is also a shared opportunity for progress

This is a collective problem that we must all own up to it – and then take shared ownership in addressing.

Getting it right will be truly challenging. Years of sincere effort have demonstrated that there is no silver bullet, no easy answer. As a result, too many human rights abuses in business value chains continue unchecked or even unidentified. Moving forward will require the wisdom of the crowd, multidisciplinary thinking and co-creation.

It is in that spirit that Shift is launching this month a major new collaborative initiative – a crowd-based conversation if you will – to help work out how we can all do better. A conversation to break down the problem of measurement, talk about the barriers to improvement, and see which can be overcome and how. A conversation aimed at learning from and contributing to the very best of practice around the world and looking at how it can be transferred and scaled up across businesses and industries. A conversation that builds a common understanding of the kinds of information that offer a true perspective on which companies are respecting human rights and which are not.

Shift is joined in this initiative by three regional, non-profit partners: the Centre for Human Rights at the University of Pretoria in South Africa, the ASEAN CSR Network based in Singapore, and the Polish Institute for Human Rights and Business. We are grateful for the seed funding for this project from Humanity United and a three-year grant from Norges Bank Investment Management, the sovereign wealth fund of Norway. Together, we will work with a wide range of experts and organizations across all regions and consult with all stakeholder groups throughout the process.

The aim is not to end up with some single “answer” to the problem, nor a proprietary model to promote. The aim is to come up collectively with better approaches to this shared challenge of measurement – approaches that carry wide support, and which everyone can draw from as befits their own work and objectives. The process of research, consultation and the gradual crystallization of ideas, tools, principles, examples, models and conclusions will be open to all, with regular updates posted online.

This is an ambitious initiative, to be sure. But just as we see every day in our work how poorly companies’ social performance is currently being measured, we equally see a burgeoning number of innovators lighting the path to better solutions. New approaches to measuring how business decisions impact people are being quietly explored within companies and piloted through collaborations with trade unions and civil society. New methods to hear and learn from the people who are themselves impacted by business are being developed and tested. New areas of research are bringing to light data about impacts that was not previously accessible; and long-standing areas of research and practice offer us pertinent lessons – from the development, health and safety and change management fields, among others.

Furthermore, there is no shortage of highly-motivated and experienced people across civil society, business, the investor community and government who have shared with us their enthusiasm to help change the state of practice in this field. So we are excited and optimistic about what we can achieve together, and we invite everyone with an idea, a challenge, a question or an interest in this issue to contact us and join the conversation. A conversation about how to make markets more sustainable by equipping investors to more accurately assess the risks in their portfolios.

The groundswell of recognition that every company’s social performance matters opens up huge opportunities for positive change. To seize these opportunities, we need to be able to judge what’s working and what isn’t, which companies are really improving lives through responsible business conduct, and which are churning out numbers and words that build a Potemkin village aimed more at winning awards than having a positive impact. In this age of crowd-sourcing models, we can surely crowd-source – and crowd-craft – the tools we all need to tell the difference.