The starting question changes
Many circular economy programmes begin with a familiar catalogue: recycling systems, industrial symbiosis, digital tracking, product redesign, or new regulation. Those tools can be useful. But selecting the tool before understanding the local system can turn circularity into a template exercise.
A 2026 Nature Sustainability analysis offers a useful correction. The authors reviewed 183 peer-reviewed articles covering circular economy activity in 28 low- and lower-middle-income countries. In that literature, circular solutions were often more bottom-up, more closely tied to socio-economic needs, and more dependent on community, informal, and adaptive practices than the policy-, industry-, and technology-led models that dominate much Global North circular economy discussion.
That does not make every local practice sustainable, fair, or ready to scale. It changes the first operator question. Do not begin with which circular solution should be installed. Begin with why resources are already being kept in use, who does that work, what value is retained, and what conditions would improve the result without removing the people or benefits that made the loop possible.
What the research actually covered
The analysis is substantial, but its boundary matters. The researchers searched Scopus and Web of Science at the beginning of January 2022, manually checked one additional journal, and included English-language peer-reviewed articles published from 2012 onward that explicitly engaged with the circular economy concept. Their final sample contained 183 studies across 28 countries classified in 2021 as low- or lower-middle-income.
The team coded the literature across five dimensions: motivations, circular activities, value-retention options, associated impacts, and enablers or barriers. The strongest recurring contrast was contextual. The reviewed Global South literature emphasized local and small-enterprise activity, resource scarcity, livelihoods, cultural practice, and socially or technically adaptive approaches. The comparison literature for the Global North emphasized formal industry, public policy, technology, and environmental goals more strongly.
This is a review of published research, not a census of every circular practice. The authors note important limits: English-only sources, a narrow requirement that studies explicitly use the circular economy concept, uneven regional coverage, and a literature search that ended in early 2022. The findings should guide better questions, not become a universal claim about all countries or communities.
Circularity can begin with necessity
In a resource-rich corporate strategy room, circularity may be framed as reducing excess material use or meeting a climate target. In a resource-constrained setting, the same loop may begin because replacement goods are unaffordable, formal services are limited, livelihoods depend on recovery work, or a community must stretch water, food, building material, or equipment further.
Those motivations affect project design. A repair network built around household affordability should not be evaluated only by tonnes diverted from disposal. A material-recovery activity that supports income should not be redesigned only around processing yield. A water-reuse practice may be linked to basic service reliability or climate resilience, not simply efficiency.
Environmental outcomes still matter. The point is that the route to them may pass through livelihood, access, resilience, or resource-security needs. A project that ignores those needs may improve a technical indicator while weakening the system that kept materials in use.
Do not erase the people already operating the loop
Informal workers, local repairers, small enterprises, community groups, and household practices can already perform collection, sorting, repair, resale, sharing, and reuse. Calling the system informal describes its relationship to institutions; it does not mean the activity has no knowledge, value, or organization.
Formal investment can bring safer working conditions, reliable infrastructure, finance, data, market access, and legal recognition. It can also displace existing workers, centralize value, impose reporting that small actors cannot carry, or optimize for a material stream while overlooking livelihoods and access.
The operator task is therefore connective. Identify who performs each activity today, who owns the material at each step, who is paid, who carries risk, and who would gain or lose under the proposed design. Then decide which formal support strengthens the loop and which change would merely replace it.
- Map existing actors before naming a new delivery partner.
- Separate unsafe or exploitative conditions from the useful knowledge embedded in the practice.
- Ask who owns recovered value before and after the intervention.
- Design data and compliance requirements that small operators can realistically meet.
- Record the distributional outcome alongside material and financial performance.
Use a five-question context map
The study's framework can be translated into a short project gate. Each question needs observable evidence. If the answer is still an assumption, record it as an open issue rather than letting it harden into the project model.
- Why: Which scarcity, livelihood, service, cultural, environmental, or business need motivates the loop?
- What: Which activities happen now, at household, community, enterprise, city, or supply-chain level?
- Value: Is the system avoiding demand, extending product life, recovering components, cycling biological resources, or recovering material value?
- Outcome: Which economic, social, environmental, and resilience results matter, and who experiences them?
- Conditions: Which skills, relationships, finance, infrastructure, policy, market access, and safeguards enable or constrain the system?
Before and after: an imported template versus a context-first project
Consider a city seeking to improve recovery of a high-volume waste stream. A template-first project might begin by procuring equipment, defining a centralized collection model, and measuring the tonnes processed. Local collectors and small processors appear later as a stakeholder-management issue.
A context-first version begins with the existing flow. It maps who collects, sorts, transports, buys, repairs, resells, or disposes of the material; how income is earned; where contamination or injury occurs; which buyers accept the output; and why households participate. The team then selects infrastructure and governance that address the documented bottlenecks.
The resulting project may still use centralized equipment. It may instead combine neighborhood aggregation, safer handling, small-enterprise support, buyer standards, and shared data. The difference is not the size of the technology. The difference is that evidence defines the intervention, and the project measures whether material retention, working conditions, access, and local value improve together.
Measure progress without flattening the context
A context-first project still needs disciplined measurement. The difference is that measurement begins with the mechanism the project is trying to improve, not with a generic dashboard imported from another programme. A reuse system needs evidence about return, condition, repeated use, loss, and displacement of new purchases. A repair network needs evidence about product life, repair success, cost, access, and what happens to items that cannot be repaired.
Material indicators should be explicit about boundaries. Tonnes collected are not the same as tonnes returned to useful service. Recycling output is not automatically a reduction in virgin material demand. A reported job count does not show income quality, safety, continuity, or who lost work when a system changed. Keep these measures separate until the project can explain the causal link between them.
Qualitative evidence also belongs in the operating record. Interviews, worker observations, community feedback, grievance records, buyer acceptance, and maintenance logs can reveal why a technically sound flow is not being used. These sources need dates, owners, and documented methods just like quantitative data. They should not be treated as decoration around the material balance.
Set a small number of decision thresholds before the pilot begins. Define what would support expansion, what would require redesign, and what would stop the project. This makes evidence useful for control rather than only for a report written after the decision has already been made.
Finance and procurement should test the local operating model
A lender, buyer, donor, or corporate sponsor can unintentionally push a project back toward a template by asking only for familiar technology, volume, certification, or growth metrics. Those questions matter, but they do not establish whether the local operating model is credible.
Due diligence should ask who supplies the material, how participation is obtained, how prices and income are set, which actor controls records, who handles contamination or rejected material, and whether the project depends on unpaid or underpaid work. It should also test whether equipment can be maintained locally, whether buyers accept the recovered output, and whether policy support is active rather than merely announced.
Procurement can then describe the required outcome without prescribing every delivery detail too early. For example, a buyer may require traceable recovered material, defined quality, safe working conditions, and a correction process while allowing qualified local partnerships to determine collection and preparation. This protects the evidence boundary while leaving room for a model that fits the place.
Finance terms should distinguish construction, commissioning, stable operation, and verified outcome. A project can purchase equipment without establishing a durable circular flow. Staged evidence makes that difference visible and gives the operator a chance to correct the model before expansion locks in the wrong incentives.
A practical operating checklist
Use this checklist before approving a circular project concept, financing note, supplier programme, or public claim. It is deliberately short enough to use in an early project meeting.
- Write one sentence naming the local problem and who experiences it.
- Map the current material, product, money, information, and risk flows.
- Identify existing formal and informal actors, including who may be displaced.
- Name the value-retention strategy before selecting technology.
- Define at least one material, one economic, one social, and one operational indicator.
- Separate observed evidence from assumptions and imported benchmarks.
- Test whether compliance, data, and finance requirements are usable by the smallest critical actor.
- Create a grievance, correction, or redesign path for affected participants.
- Pilot at a scale that can reveal distributional and operational effects.
- Decide in advance which evidence would stop, narrow, or change the project.
Keep the evidence boundary visible
The reviewed study supports a context-first approach, but it does not prove that every bottom-up practice is environmentally preferable or that every formal system is unsuitable. It does not provide one performance benchmark that can be applied across countries, materials, or sectors.
Project teams still need local baseline data, technical assessment, worker and community participation, financial analysis, legal review, and outcome monitoring. Report material, social, and economic results separately before combining them into a broader conclusion.
The useful conclusion is modest and operational: circular economy design improves when the evidence includes the context that produced the loop. A project should not treat livelihoods, culture, scarcity, and informal organization as background details after the technical solution has already been chosen.