Power Has an Address

A label tells you where a garment was made. It tells you much less about where its conditions of production were decided.

The factory has managers. The buyer has deadlines. A purchasing team negotiates a price. A lender expects payment. A government sets rules and decides how seriously to enforce them. A worker arrives at the beginning of a shift with obligations that do not appear anywhere on the label.

When something goes wrong, each institution can point to another part of the chain. The buyer did not employ the worker. The factory did not set the retail price. The lender did not decide the schedule. The government did not sign the commercial contract.

These distinctions have legal and operational meaning. They can also leave a person facing a system in which everybody contributed to the conditions and nobody appears to own the consequence.

The phrase global economy can make this arrangement sound too large to examine. It is built from decisions made in particular places by institutions with names, accounts, contracts and people authorized to act.

Power may travel. It does not become supernatural when it crosses a border.

Follow the order back

Imagine a factory that has agreed to deliver an order on a tight schedule. The buyer changes a specification late, but the delivery deadline remains. The factory must absorb the extra work, negotiate a change or risk losing a customer it depends on.

What happens next depends on its capacity, contract and management. It might improve coordination, add properly compensated capacity, seek an extension or accept a smaller margin. It might also push the pressure down onto workers through excessive hours, unstable schedules or unsafe shortcuts.

The buyer may have a code of conduct forbidding those practices. That code matters. Its credibility depends partly on whether the buyer’s purchasing behavior makes compliance feasible and whether workers can report violations safely.

A responsible sourcing policy should therefore inspect both the factory floor and the commercial decisions that shape it. The question is not only whether a supplier signed a standard. It is whether price, lead time, payment and changes are consistent with the standard, and what happens when they are not.

This does not make the buyer responsible for every decision a supplier makes. A supplier retains obligations and agency. It does mean that a powerful buyer should not treat its own contribution to a foreseeable pressure as outside the moral scope of the relationship.

The same reasoning applies beyond clothing. A lead firm can set conditions that smaller businesses pass along. A public purchaser can demand a low price while overlooking the labor practices required to deliver it. An institution can celebrate its own standards while rewarding another institution for making those standards difficult to meet.

Follow the order, the money and the timing. A chain of responsibility is easier to understand when it includes the decisions that create pressure, rather than beginning only where the pressure becomes visible.

The committee and the factory

There is evidence that institutional design can change conditions, although the effects are neither automatic nor uniform.

Laura Boudreau’s research, published in Econometrica in 2024, studied a randomized effort to enforce Bangladesh’s safety-committee mandate in eighty-four supplier factories during 2017–2018. The programme involved the Alliance for Bangladesh Worker Safety and multinational buyers. Enforcement increased compliance, and stronger committees produced small improvements in objective safety measures without detected losses in wages, employment or productivity. Benefits were concentrated in better-managed factories; weakly managed factories did not improve and job satisfaction declined.¹

This was a particular intervention, not a verdict on all factories or all private oversight. The Alliance was a different organization from the Accord on Fire and Building Safety. The study’s variation is especially instructive: a formal committee and outside pressure can help, but organizational capacity influences what the mechanism can accomplish.

For a worker, representation must connect to the ability to solve a problem. A meeting that records concerns without resources or authority may disappoint people more deeply than no meeting at all. Raising expectations creates an obligation to build the means of answering them.

The broader lesson is practical. Voice, enforcement and capability need to meet in the same place.

A worker should not be the cheapest adjustment

Global production can bring employment, skills, investment and income to places that need them. A working-class politics should not dismiss those gains from the comfortable position of someone whose own job is secure.

It should ask why the person with the least reserve so often becomes the easiest place to put uncertainty.

If demand falls, temporary workers may lose hours. If a buyer pays late, a supplier may delay wages. If a currency moves, management may seek savings where resistance is weakest. None of these responses is mechanically inevitable. They arise from a distribution of contracts, bargaining strength and protections.

The policy question is how to share adjustment more fairly. A buyer with greater financial capacity can accept more predictable payment obligations. Public institutions can enforce wage claims and provide income protection. Workers can negotiate arrangements that prevent every commercial shock from becoming an immediate household emergency.

Small suppliers need consideration too. A responsible standard that adds costs without adjusting commercial terms may simply drive a vulnerable business into concealment or closure. Assistance with capability, realistic timelines and consistent enforcement can matter alongside the rule itself.

This is not a request to lower standards whenever compliance is inconvenient. It is a demand to make the whole arrangement capable of meeting them.

The powerful should have to explain how their preferred price is compatible with the conditions they claim to support.

Borders can increase dependence

A worker who moves for employment may gain an important opportunity. Migration can improve income, support family members and expand a person’s future choices. It can also create additional dependencies that an employer or intermediary may exploit.

Consider the conditions rather than assuming the outcome. Does the worker understand the contract? Have recruitment debts reduced their ability to leave? Is legal permission to remain tied to a particular employer? Can they safely report mistreatment? Do they have access to their documents, wages and an independent source of advice?

Where several of these vulnerabilities combine, an ordinary employment dispute can become a threat to the person’s entire position in the country. The cost of speaking rises sharply.

A government that wants lawful, well-regulated migration should care about that vulnerability. Workers who cannot report abuse safely provide an opportunity for bad employers to undercut good ones. A system that makes people dependent and then blames them for being difficult to protect is badly designed.

The remedies require legal specificity, but the direction is clear: transparent recruitment, enforceable contracts, access to help, protection against retaliation and reasonable ways to change an abusive relationship. A worker’s nationality should not become a discount on the dignity owed to them.

Local workers have a stake in these protections. When one group can be paid less or frightened into silence, employers may use the difference to weaken standards for others. Turning workers against one another can protect the institution benefiting from the unequal rule.

There are legitimate debates about migration levels, public capacity and labor-market effects. They should be conducted without requiring the people already doing the work to accept mistreatment as the price of being present.

The world has improved and remains intolerable in places

A global argument needs room for actual progress. The International Labour Organization’s Global Wage Report 2024–25 found that wage inequality had declined in roughly two-thirds of the countries studied since the early 2000s, while substantial gaps remained. Wage distributions are not the same as wealth or all household income, and national experiences differ.²

That evidence rules out a lazy story in which every society is becoming worse in every respect. It also encourages a useful question: what institutions and changes contributed to improvements, and where have they failed to reach people?

The remaining hardship is immense. Under the revised indicators used in the 2025 WHO–World Bank monitoring report, 4.6 billion people were not fully covered by essential health services in 2023, and 2.1 billion faced financial hardship from out-of-pocket health spending in 2022. These are different measures and observation years. Incomplete coverage does not mean receiving no care at all.³

The report records progress over the longer term alongside a slowdown. A society can improve and still fall far short of what its people need.

Our moral imagination should be able to hold both facts. Celebrating gains does not require telling those left behind to wait quietly. Recognizing suffering does not require erasing the work that has already reduced it.

For people in low-income settings, the central problem may include an absolute shortage of clinics, trained staff, electricity or transport. Better bargaining cannot conjure resources into existence overnight. It can influence how resources are mobilized, allocated and protected from capture. Productive development and fair institutions belong in the same project.

We should want the world to become wealthier in ways that make more people secure, capable and difficult to discard.

Responsibility across distance

A distant consumer cannot personally inspect every workplace behind a purchase. Nor can a worker safely become the sole source of evidence against every institution on which their livelihood depends.

Responsibility has to be organized at levels able to carry it. Buyers can make contractual commitments and disclose relevant information. Governments can enforce labor and safety law. Worker organizations can preserve knowledge and act collectively. Independent bodies can examine whether promises are being kept.

The design must account for conflicts of interest. If the organization being inspected chooses and pays the inspector, what protects the inspection’s independence? If a buyer can leave immediately after a violation is found, who protects workers from losing their livelihood as a consequence of reporting? If a public authority lacks resources, what assistance strengthens it without allowing an outside institution to dictate its priorities?

No single certificate resolves these questions. A credible system needs evidence, independent worker participation, remediation and consequences that reach the institutions able to change the relevant conditions.

Remediation deserves special emphasis. Discovering abuse should lead to making people safer and repairing harm where possible. A policy that rewards firms mainly for avoiding association with a problem can encourage them to sever the relationship and leave the affected workers with the damage.

Sometimes ending a relationship is necessary. Sometimes a supplier refuses to improve or the harm cannot be managed safely. The decision should still account for the people whose work made the relationship valuable in the first place.

An ethical supply chain cannot be defined solely by how cleanly a powerful company can detach itself from a dirty fact.

Solidarity must survive the price tag

There is an uncomfortable question for consumers in wealthy places: what if better conditions cost more?

Sometimes improvements can be financed through reduced waste, better management or a different distribution of margins. Sometimes costs genuinely rise. A serious argument should not promise that every desirable reform pays for itself immediately.

The distribution of that cost matters. A low-income household should not be asked to solve global labor injustice through purchasing choices it cannot afford. Large buyers and highly profitable intermediaries may have greater capacity to absorb costs. Public policy can establish common standards so that responsible behavior does not depend entirely on a niche market for ethical products.

This is also why a boycott is not automatically the best response. It can create pressure, but it can also harm workers if demand disappears without a plan. The people directly affected should have a meaningful role in deciding what support would help them. A distant campaign that ignores their judgment can reproduce the hierarchy it opposes.

Solidarity begins with listening and continues through the less dramatic work of following a commitment. Did wages arrive? Was the hazard repaired? Were the workers who raised the issue protected? Did the buyer alter the purchasing practice contributing to the problem?

A photograph of concern is not the end of the relationship.

The map needs more than a villain

Suppose a factory has unsafe machinery. A useful map identifies the owner responsible for maintenance, the managers assigning work, the inspector responsible for enforcement, the buyer whose contract affects resources and the workers with direct knowledge of the risk.

It also identifies who can authorize a repair, what the repair requires and what protects people during the interval. It distinguishes influence from legal duty rather than pretending they are identical.

A campaign may then choose several connected demands: immediate protection, a funded repair plan, independent verification and purchasing terms that do not undermine the improvement. Different institutions will be responsible for different parts.

This is more complicated than finding one bad person. It is more likely to produce a result that lasts after that person is replaced.

The same method works for many global problems. Trace the decisions. Identify the dependencies. Locate the duty and the practical capacity. Ask where people can organize and which institutions can make cooperation binding.

Distance makes this work harder. It also makes it more necessary. An institution should not be able to expand the reach of its decisions while shrinking the range of consequences for which it must answer.

The worker at the far end of the chain belongs at the beginning of the moral argument.

An internationalism made of useful obligations

It is possible to speak beautifully about humanity while leaving every operative responsibility vague. A useful internationalism is more demanding.

It asks governments to cooperate on standards and enforcement where unilateral action is too easily undermined. It asks buyers to make purchasing behavior consistent with their public commitments. It asks worker organizations to build relationships across borders so that one group is not continually offered as the cheaper replacement for another.

It also respects local knowledge and political agency. People living with a problem are partners in solving it, not raw material for someone else’s campaign. Their priorities may differ from the story a distant audience finds most compelling.

This kind of cooperation will be uneven. Interests conflict. Institutions fail. A country may need investment while wanting stronger protections. A worker may prefer an imperfect job to no job. A supplier may need time and assistance to improve. These facts create the work of negotiation; they do not make dignity an optional luxury.

The transformative ambition is a world in which production can cross borders more easily without accountability being left behind. Prosperity should travel through the chain, and responsibility should travel back along it.

We cannot achieve that by wishing away national institutions or commercial constraints. We can build rules, organizations and relationships that make the next decision less one-sided than the last.

There is an address. Find who can answer there. Then make sure the people affected have a way to be heard together.