Data and governance requirements of sovereign sustainability-linked bonds

Sustainability-linked Bonds (SLBs) for sovereigns are a powerful mechanism to operationalise sustainability commitments and render them more credible, but the challenges to scaling up Sustainability-linked Bonds are formidable.  
‍

As outlined in our blog on ‘A sober assessment of the sovereign SLB Value Proposition’, the reporting requirements of SLBs are generally more onerous than disclosures covering conventional bonds which mainly draw on established macroeconomic and financial statistics.  
‍

By contrast, the datasets underlying climate and nature KPIs tend to be large, unstructured, configured in non-standard formats, and dispersed across multiple sources within the public sector or among third parties. Some of these datasets may also be sensitive and subject to data protection or localisation laws, preventing their transfer to offshore cloud data centers for processing.  
‍

Extracting, transforming and loading of data into pipelines that deliver KPI consistently and reliably over the life of the bond can entail complex workflows and advanced data management systems, which in turn requires trained staff, state-of-the-art technology infrastructure and sound data governance frameworks.
‍

For illustration, KPIs that measure deforestation using satellite imagery and remote sensing techniques can require specialised geographic information systems (GIS) and third-party providers that are costly and complicated to onboard.  
‍

For this reason, Uruguay’s 2034 SLB incorporated a native forest cover KPI that was based on geospatial data from two space agencies (images from the European Space Agency’s Sentinel satellite, topographic elevation from NASA’s Shuttle Radar Topography Mission). A machine learning algorithm classified the images into different forest types with the aim of isolating native tree clusters. The analysis was carried out by a team of six data scientists and engineers coordinated by General Forestry Directorate (DGF) using the Google Earth Engine cloud computing platform to perform the calculations and GIS software (QGIS and ArcGIS) to process the results.
‍

The centrality of data in the SLB structure also puts a premium on data transparency and integrity. Investors must have confidence that the KPIs truly reflect underlying performance and that there is zero scope for manipulation or human errors. This means each data point can be interrogated and traced back to source. Each step of data journey should be clearly mapped and automated to the extent possible using end-to-end processing tools such as application programming interfaces (APIs).  
‍

Some of this work can be outsourced to the growing cottage industry of measurement, verification, and reporting (MRV) providers in the climate and nature data space. However, providers need to be properly vetted and locked into service level agreements that provide quality assurance over the long run and specify contingency plans to avoid any disruption of data flow.  
‍

Second party opinion providers (SPOs) add an extra layer of quality control; in the case of Uruguay, this is provided by Sustainalytics for the sustainability framework and the United Nations Development Program (UNDP) for the external verification of KPIs.  
‍

However, it doesn’t absolve the sovereign of ultimate responsibility over the integrity of the data.
‍
Pulling data from other agencies within government can throw up major coordination problems. Data sharing commitments need to be enshrined into durable agreements. These, in turn, need to be underpinned by a robust governance structure that can survive successive political cycles.  
‍

Coordination mechanisms such as inter-ministerial committees are often prerequisites as they clarify protocols for sharing data, define roles and responsibilities for ensuring data flows are unobstructed, and resolve internal disputes.  
‍

In the case of Uruguay, agreements were set out in memorandum of understandings (MOUs) and implemented by a special inter-ministerial task force (the SSLB Working Group) composed of the Ministries of Economy and Finance; Environment; Industry, Energy and Mining; Livestock, Agriculture and Fisheries; Ministry of Foreign Relations (see Figure 1).

Figure 1: Uruguay’s governance architecture
‍

Source: Uruguay MEF

For coordination mechanisms to be effective, the incentives of the participating agencies need to be aligned. Commitment is unlikely to materialise organically if data sharing requires organisational changes or additional investments in technology and talent that detract from other policy priorities.  
‍

Institutional inertia and bureaucratic politics often come into play in such situations, especially in contexts where political polarisation and weak state capacity makes it hard to achieve consensus on targets or implement KPIs. A simple directive from the executive or other high-level authority such as the minister of finance, while necessary, may not be sufficient to guarantee swift buy-in and sustained compliance from participating agencies.

In practice, aligning incentives often means understanding the wants and needs of all stakeholders involved and finding ways to match them. This may require some amount of “horse trading”, whereby providers of data receive something of value in return — budgetary resources, other data, control over usage, participation in the selection of KPIs and underlying projects, etc. The pooling of data itself can be an inducement since it creates a richer data set on which to run models.  
‍

As a hypothetical example of such a data-for-insights trade, say the ministry of environment supplies geospatial imagery of forest cover and the ministry of mines geological surveys of extractive resources, then overlaying the two can help to locate ecological areas at high risk of illegals exploitation. Adding hazard maps to the stack can also help to identify natural assets that protect against climate shocks.  
‍

The sharing and commingling of data across agencies may also unlock efficiency savings if it cuts duplication of work or systems (e.g. only one agency licenses GIS software rather than every other). Still, arranging these value exchanges can be time consuming and contingent on favorable political conditions. The costs in terms of budget, time, and political capital are significant and increases the need for the benefits of SLBs to be clearly understood – as outlined in our blogs on ‘The understated benefits of sovereign SLBs’ and ‘Accelerating sovereign SLB transactions: building the ecosystem’.

‍

‍

‍

‍

Read Next:

The understated benefits of sovereign sustainability-linked bonds

‍

Previous blog series:

Go with the flow, not just the stock: Sustainability-linked bonds and public debt dynamics  

Sustainability-linked bonds and public debt dynamics: a detour into sovereign debt maths

How sustainability-linked bonds can improve sovereign credit ratings: an illustrative example

‍

‍

‍