The process of assessing human rights risks for new operations is fairly straightforward: companies should assess the risks they pose to human rights and modify operations and project designs accordingly. But what about existing operations? Companies with small but high-impact portfolios (for example, mining companies) should carry out Human Rights Impact Assessments on each existing mine.

Companies with thousands of operations and facilities spanning dozens of countries across the globe cannot do the same, nor should they. Coca Cola, for example, is not expected to carry out detailed assessments of every warehouse it uses across the European Union to the same level they would be expected to assess their sugarcane suppliers in Colombia. Unilever is not expected to carry out equally detailed assessments of Skippy peanut butter’s corn syrup suppliers, which are large US-based, mechanized farmers, as their palm oil suppliers (which are high-risk operations predominantly in Indonesia and Malaysia).

Likewise, financial institutions that make thousands of loans and equity investments do not have the resources or incentive to scrutinize a loan to a worker diversity training program in Costa Rica to the same extent they scrutinize a loan to a greenfield mine in sub-Saharan Africa.

For business enterprises like these, a screening process is needed to identify the highest risk operations for prioritization in a human rights due diligence plan. Not all business enterprises will view risks the same way. For example, development banks will look for low income contexts to invest as a matter of priority, while consumer goods companies might avoid the risks inherent to such markets.

NomoGaia has created BankRight as a tool to identify the country-level risks to civil, political, economic, social, cultural and labor rights in 190 countries. The tool is based on highly-regarded, publicly-available indices produced by global institutions and refined over years and decades. Sourcing information is available directly within the tool.