Benchmark

“A benchmark tells us compared with what; it does not tell us automatically what good should be. ”

Benchmarks give numbers context. A wage is compared with a living-wage benchmark. Farm productivity is compared with regional yield. An emissions pathway is compared with a sector trajectory. Without a reference, performance can be difficult to interpret. The value of a benchmark depends on comparability. The population, geography, period, method and boundary should match the use.

Comparing small rain-fed farms with irrigated commercial estates may reveal a difference, but it does not establish underperformance. The comparator may reflect different resources, objectives and risks. Living-income and living-wage benchmarks illustrate both value and limitation. Methods developed by Anker and Anker estimate the cost of a decent standard of living in a particular place and household context.

They provide a reference for the gap between actual income or wage and a decent level.

They do not by themselves determine who should pay, which intervention will close the gap or whether household income data are accurate. A benchmark can be historical, peer-based, normative or performance-based. Historical benchmarks compare with past performance. Peer benchmarks compare organisations or groups. Normative benchmarks reflect a standard or acceptable condition.

Performance benchmarks may define eligibility or crediting thresholds. Each supports a different judgement. Peer benchmarking can encourage improvement, but the average is not necessarily sustainable. If an entire sector underpays workers or exceeds ecological limits, performing better than peers can remain inadequate. A race to the top requires a reference grounded in outcomes, not only relative position.

Benchmarks can also become targets.

A programme may aim to reach the living-income benchmark or perform above a carbon-intensity benchmark. The shift should be explicit. A benchmark describes a reference; a target states an intended result. Treating the benchmark as automatically achievable or sufficient can oversimplify structural constraints. Methodological differences are common.

Two water-stewardship tools can use different basin boundaries, stress data and time periods. Two biodiversity benchmarks can use different reference ecosystems. Two income benchmarks can assume different household sizes or public-service access. Publishing the number without the method invites false comparison. Benchmarks can exclude context.

A global threshold may be useful for consistency but insensitive to local rights, climate or infrastructure. Local references can be more relevant but reduce comparability. Organisations may need a hierarchy: global principles, sector methods and place-specific values. There is also a danger of strategic comparator choice.

A company can select a weak peer group, an old baseline or an easy jurisdiction and claim leadership. The comparator should be chosen before results are known or justified through independent criteria. Performance standards in carbon markets show the stakes. A benchmark can help assess whether an activity goes beyond common practice, but its stringency affects whether credited reductions are genuinely additional.

If the benchmark is too weak, ordinary performance earns exceptional claims. Benchmarks can be descriptive or normative, and the distinction should be explicit.

A peer median describes what comparable organisations currently do; it does not establish what adequate performance requires. A living-income benchmark estimates a decent standard of living; it is not the average income in a producing region. A science-based threshold may remain demanding even when every peer misses it. Benchmarks also age.

Inflation changes livelihood costs, climate science changes pathways and common practice changes as regulation and technology advance. Periodic review is necessary, but historical comparisons should not be rewritten silently. Users need to know which benchmark applied at the time, which one applies now and whether an apparent improvement reflects performance or a weaker reference.

The discipline is to ask what judgement the benchmark is meant to support. Is it describing position, setting adequacy, establishing additionality or guiding a target?

A credible benchmark makes its comparator, method and limitations visible.

Practical application

Choose benchmarks aligned with the population, geography, period, boundary and intended decision. Document the method and test whether the comparator reflects adequate performance or only current practice. Use multiple references where relative and normative performance differ. Review benchmarks as prices, science and context change, and disclose when a benchmark becomes a target or eligibility threshold.

Document why each comparator is appropriate and who approved it. Where several benchmarks are plausible, show the conclusion under each rather than selecting only the favourable one. For livelihood, climate or ecological thresholds, separate the benchmark of adequacy from peer performance. The organisation can then see whether it is improving, leading and sufficient - three different judgements.

Where a benchmark is used for incentives or public claims, obtain independent review of its fitness for purpose and disclose whether performance is above the comparator yet still below an adequacy threshold.

Why it matters

Benchmarks influence claims of leadership, adequacy and progress. Poor comparators can normalise sector-wide failure or make ordinary performance appear exceptional.

Common misconception

A benchmark is often treated as best practice or the level an organisation should achieve. It is a reference point; its normative value depends on how and why it was established.

Connections

Baseline is an internal or scenario reference, Target is an intended future result and Metric enables comparison. Living Income and Living Wage depend on place-specific benchmarks. Additionality may use performance benchmarks to test whether action exceeds expected practice.

A question worth asking

Does your benchmark represent what is common, what is achievable or what is actually required for people and ecosystems to thrive?

Selected references

Anker, R. and Anker, M. 2017. Living Wages Around the World: Manual for Measurement. Living Income Community of Practice. 2020. The Living Income Benchmarking Methodology. OECD. 2024. Effective Results Frameworks for Sustainable Development. Greenhouse Gas Protocol. 2004. The GHG Protocol for Project Accounting.

International Organization for Standardization. ISO 14031:2021. Environmental Management - Environmental Performance Evaluation.