When a company announces that it has found cheaper labour, what has it actually found?

A better way to produce? A more capable workforce? An infrastructure that allows people to accomplish more with less? Or people who will accept less because refusing would cost them too much?

All of these possibilities can appear under the same heading in an investment presentation: cost reduction. Yet they describe profoundly different relationships between a business and the people whose work sustains it.

A factory can become more efficient by wasting less material, preventing machine failures, improving its products, or organising production intelligently. It can also improve its accounts by paying people less for substantially the same work. The first changes how production happens. The second changes how its benefits are distributed. Both may increase profit. That does not make them the same achievement.

Before celebrating the savings, we should ask where they came from—and where the costs went.

The phrase “cheap labour” encourages us to begin with the employer’s expense. From that perspective, a wage is a number to be contained. From the worker’s perspective, the same number determines housing, food, transport, care, rest, and the possibility of planning beyond the next payment. What appears as a reduction in one column may become an additional shift, a postponed treatment, or a child’s unmet need somewhere else.

The cost has not necessarily disappeared. It may have changed its address.

This is the central concern: a business can turn another person’s restricted choices into a source of financial advantage. It may not have created the poverty it encounters. It may even improve some of the opportunities available. But does either fact settle how responsibly it should use the bargaining power that poverty gives it?

If we did not create someone’s vulnerability, does that mean we have no responsibility for how we benefit from it?

To answer seriously, we need to distinguish low wages from exploitation without allowing the distinction to become an excuse for ignoring either.

A wage that is lower in one country than another is not, by that fact alone, proof of exploitation. Prices, exchange rates, infrastructure, technology, and productivity differ. Paying the same nominal wage everywhere is not a complete theory of fairness. Nor does the same job title guarantee identical working conditions or output.

But the opposite shortcut is equally inadequate. A wage does not become fair simply because it matches the local market. The local market may itself reflect widespread deprivation, weak bargaining power, exclusion, or ineffective protection.

“Everyone pays this much” describes a practice. It does not justify it.

The more revealing questions concern what the wage makes possible and what the employment relationship allows. Can a person meet ordinary needs through ordinary working hours? Can they refuse a dangerous task? Can they organise with colleagues? Can they question an unpaid hour without risking their livelihood? Can they leave without immediately losing the means to survive?

The International Labour Organization’s 2024 agreement on living wages places workers’ and families’ capacity to maintain a decent standard of living at the centre of the definition, taking national circumstances into account and calculating the wage for normal working hours. That final condition matters: a life made barely affordable through permanent overtime is not the same as an adequate wage. (International Labour Organization, “Living Wages,” 2024 agreement.)

A worker who needs exhaustion to make the wage sufficient is partly financing the apparent affordability of the job with their own recovery time.

There is a familiar answer to this criticism: the worker agreed.

Agreement matters. Workers in poorer countries are not passive figures waiting for outsiders to explain their interests. They compare opportunities, support families, take risks, learn, negotiate, and make decisions under conditions they often understand better than distant observers. Accepting a difficult job can be an intelligent response to the available alternatives.

But an intelligent response to an unfair situation does not make the situation fair.

Consider a hypothetical worker choosing between an inadequately paid factory job and irregular work that pays even less. Taking the factory job may improve that person’s life. The improvement is real. It still leaves open whether the employer could offer better conditions, whether the worker has meaningful bargaining power, and whether the gains from production are reasonably shared.

Two questions must therefore remain separate: Is this better than the worker’s existing alternative? And is this a fair use of the power the employer possesses?

An arrangement can benefit both parties while allowing one party to take unfair advantage of the other’s limited options. Mutual benefit does not, by itself, establish a fair distribution of that benefit.

If the only standard is “better than nothing,” how far above nothing must a worker ever be allowed to rise?

The argument that a business creates jobs is important, but incomplete. A job is not a uniform unit of human progress. Its value depends on its pay, safety, stability, demands, opportunities, and the voice it gives the person performing it. Counting positions without examining those conditions can make an employment policy look successful while leaving the quality of working lives outside the measurement.

A government can announce ten thousand jobs. The announcement does not tell us whether those jobs allow ten thousand people to stop living in permanent insecurity.

Nor is a wage simply a transparent measurement of the value an individual creates. It is also shaped by the structure of bargaining.

Economists use the term monopsony power to describe an employer’s ability to influence wages and conditions because workers cannot easily move to other opportunities. There need not be only one employer. Transport difficulties, caring responsibilities, the expense of job searching, or the inability to survive between paydays can all restrict mobility. The OECD’s review of labour-market monopsony describes substantial evidence that employer power can suppress pay and employment relative to more competitive conditions. (OECD, “Employment Outlook 2022: Monopsony and Concentration in the Labour Market.”)

This changes the meaning of an apparently voluntary wage. The question is not merely what the employer offers, but what makes refusing the offer possible.

A company can sometimes wait for another applicant. A household may not be able to wait for another meal. Both parties possess a formal right to say no; the material consequences of saying it can be radically unequal.

What kind of freedom exists when one side can afford to delay the agreement and the other cannot?

The search for cheaper workers can exploit precisely this asymmetry. If the product, equipment, organisation, and expected output remain broadly comparable, but the business moves to a workforce with fewer alternatives and weaker protections, the improvement in profitability may come primarily from a shift in bargaining power.

To call that innovation without examining the mechanism would confuse the discovery of a better method with the discovery of people who can demand less.

This does not make every relocation unethical. Businesses may move closer to customers, suppliers, skills, or reliable infrastructure. Investment in a poorer country can create useful employment, training, and productive capacity. Rejecting such investment indiscriminately could remove opportunities from the people we claim to defend.

The ethical test must follow what the investment does, rather than treating its foreignness or its lower wage bill as a verdict.

Does it increase workers’ capabilities and options over time? Does it respect their ability to organise? Do better productivity and sustained commercial success produce improvements in their lives? Or does the business remain competitive only while their ability to demand more remains weak?

An investment that helps people move beyond poverty differs from one whose profitability depends on their remaining close to it.

Time is crucial here. The conditions under which an enterprise begins should not automatically become the conditions under which its employees must remain. A company may initially face limited resources and genuine uncertainty. Years later, it may have stronger sales, better machinery, and substantial accumulated gains.

If sacrifice was presented as temporary, what would count as the moment when it ends?

Workers are often asked to identify with the enterprise during difficulty. They may be encouraged to accept restraint because “we are all in this together.” The credibility of that language can be tested when conditions improve. Does shared sacrifice lead to shared gains, or does solidarity disappear once there is something to distribute?

This is where we must separate survival, profitability, growth, and continuously increasing profit. They are different objectives.

A business can, in principle, remain at a stable scale, cover its costs, renew its equipment, pay its employees adequately, and provide an income to its owners. Continuous expansion is not a logical requirement of every enterprise. Yet competition, debt commitments, rising costs, and investor expectations can make expansion appear necessary within a particular business structure.

“We cannot continue operating” and “we cannot deliver the return we promised” are not interchangeable statements.

Both can describe serious problems. But the first concerns the continuation of productive activity; the second may concern how much particular stakeholders expect to receive. Before workers are told that a concession is unavoidable, they should be able to understand which problem they are being asked to solve.

Imagine companies as balloons expanding inside a room.

At first, each appears to have space. Expansion looks like an individual achievement. Over time, the balloons approach one another. Their surfaces meet. Further growth begins to displace, compress, or constrain their neighbours.

If every balloon is instructed to keep expanding, while the room remains fixed, those instructions cannot all be fulfilled indefinitely. Something must change: the pace of expansion, the arrangement, the boundaries, or the balloons themselves.

The metaphor captures an important tension, but it needs limits. An economy is not literally a room of fixed volume. New products, better methods, and useful services can expand economic possibilities. More value does not always require proportionally more material. Companies can all increase their sales when their total market grows.

They cannot all increase their share of the same market at once.

Suppose, purely as an illustration, that total demand in a market remains unchanged while every company plans to sell ten per cent more. The plans cannot all succeed under those assumptions. Some companies must take business from others, fail to meet their targets, change their products, or leave the market. What looks achievable in each company’s forecast may become impossible when all the forecasts are added together.

This is a fallacy of composition: assuming that what works for one participant must also work for everyone simultaneously.

The same tension appears when businesses view wages only as costs. A wage is an expense to the employer and purchasing power to the household. If many firms simultaneously restrain labour income while expecting stronger consumer demand, we should ask what will support the additional spending. Investment, public expenditure, exports, or changes in other incomes can affect the outcome. There is no automatic collapse implied. But there is a collective question that an individual cost sheet cannot answer.

Who is expected to buy the expanding output if the people producing it are continuously asked to accept less?

The room also has physical boundaries that financial accounts can obscure. Energy, land, material resources, ecological resilience, human attention, and time are not unlimited. Whether a particular form of economic growth increases or reduces its pressure on those limits must be examined, not assumed.

And sometimes what looks like enlarging the room means pushing a wall into someone else’s living space.

A business that prevents material waste may create a genuine efficiency gain. A business that leaves pollution untreated may instead transfer a cost to its neighbours. A business that organises work better may reduce pressure. One that achieves the same output through unpaid hours may transfer the pressure to workers.

When a balloon grows, we should ask not only how large it has become, but what it has compressed.

Nor are all balloons equally resilient. Companies differ in cash reserves, access to finance, market power, and the capacity to withstand losses. Competitive pressure does not guarantee that the most socially useful enterprise survives. A business can produce something valuable and still be unable to endure a prolonged financial shock.

Failure also reaches beyond an owner’s balance sheet. Employees can lose income, suppliers can lose payments, and a community can lose a major source of work. Recognising these consequences does not mean every company must be preserved forever. It means that transition should be judged by what happens to people, rather than treated as complete once an inefficient balance sheet disappears.

Capitalism helps explain the pressure, but naming the system does not finish the explanation.

Private ownership, competition, and the pursuit of profit can reward businesses for reducing costs. Where workers have weak bargaining power and social or environmental costs can be shifted elsewhere, lowering the price of labour may become an attractive strategy. An employer who respects stronger standards may then face competitors whose apparent efficiency depends on avoiding them.

This is why individual goodwill cannot be the entire solution. A system that penalises decent conduct creates a problem even for people who wish to behave decently.

Yet these outcomes are not natural laws. Property rights, bargaining rules, labour protections, taxation, public services, competition policy, and corporate governance are institutional choices. Different arrangements change what companies can profitably do and what workers can credibly refuse.

To explain a harmful incentive is not to excuse the person who acts on it. To hold that person responsible is not to pretend the incentive does not exist.

Interest and debt belong in this analysis, but they should not be made to explain everything.

A company with substantial debt faces payment obligations that may continue when sales weaken. Higher financing costs can narrow the room for investment and employment decisions. The IMF’s 2025 examination of corporate vulnerabilities discusses how financial distress and higher interest rates can affect firms’ capacity to invest. That establishes a relevant pressure, not a universal explanation of wage exploitation. (International Monetary Fund, “Corporate Sector Vulnerabilities and High Levels of Interest Rates,” 2025.)

A debt-free company can still underpay workers. An interest-free financing arrangement can still contain unequal ownership and aggressive return expectations. Inflation can also reduce the purchasing power of wages. There is no single financial switch whose removal guarantees fair employment.

The more useful question is how financial commitments are transmitted through the enterprise. When expected income fails to materialise, whose claim is treated as adjustable? Can dividends be reduced? Can debt be restructured? Can owners accept a lower return? Or is the worker’s wage assumed to be the most available buffer?

If a financial structure repeatedly requires employees to absorb shocks they did not choose, should we examine only their willingness to sacrifice, or also the structure that makes their sacrifice necessary?

Responsibility becomes more complicated—and more important—when we follow the supply chain.

A factory owner may exercise power over employees while facing strong pressure from a much larger buyer. A brand can demand lower prices, tighter delivery schedules, and improved labour standards at the same time. Whether those requirements are compatible depends on what the buyer is prepared to pay and how it behaves when production conditions change.

An ILO study of purchasing practices in the garment industry examined fourteen brands and retailers, asking how buyers’ commercial decisions enable or obstruct better wages and working conditions. Its relevance is straightforward: the terms of purchasing belong inside the labour discussion. (Mark Starmanns, “Purchasing Practices and Low Wages in Global Supply Chains: Empirical Cases from the Garment Industry,” International Labour Organization, 2017.)

A promise of ethical sourcing loses substance if the commercial contract makes decent production financially implausible.

We should therefore follow the chain far enough to see who has the power to change it. Blaming only the smallest supplier can allow the strongest buyer to purchase both low prices and moral distance. Blaming only the buyer can conceal abuses the supplier could have prevented. Responsibility can exist at several levels without becoming so diffuse that nobody must answer for anything.

Does the price paid for the product include the conditions the buyer publicly demands?

Governments in poorer countries are part of this relationship. They can offer infrastructure, education, reliable institutions, and growing capabilities as reasons to invest. They can also offer weak enforcement, suppressed bargaining, or the expectation that workers will remain inexpensive.

Those are different development strategies, even when both are advertised as job creation.

A government may face real constraints: unemployment, scarce public revenue, limited financing, and competition from other locations. These pressures deserve analysis. They do not make every concession necessary or every employment announcement sufficient.

When public money supports a private investment, what does the public receive in return? How are wage commitments measured? Are safety requirements enforced? Do workers have accessible remedies? Are incentives reviewed if promised benefits fail to appear?

Most importantly, is there an exit from the condition used to attract the investor?

If a country promotes itself through the affordability of its workers, what is the plan for those workers to become better paid? If every improvement is met with the warning that investment will leave, the development strategy risks making its promised destination incompatible with its method.

The country is told to become prosperous by remaining cheap.

A credible strategy must therefore be judged over time. It should ask whether investment builds transferable skills, productive capacity, local enterprise, public revenue, and stronger worker options. An opening ceremony is a beginning, not an outcome.

The threat of relocation also raises a problem that individual countries may struggle to solve alone. If investors can move between jurisdictions to avoid improving conditions, governments can be pushed into competing over what workers will be denied. Coordination on enforceable labour standards can help change the terms of that competition, although agreements require effective implementation and room for different economic circumstances.

The objective should not be to prevent poorer countries from developing. It should be to prevent their development from depending indefinitely on the weakness of their workers.

Workers across borders should not be turned into each other’s moral enemies. The person who loses a job when production moves and the person who accepts the relocated job may both be acting under circumstances they did not design. Condemning the second person for accepting less mistakes a constrained decision for control over the system.

What would solidarity mean if it defended both the worker threatened with displacement and the worker being recruited through poverty?

It would require better standards alongside real opportunities, and support for transitions rather than the simple removal of livelihoods. A campaign that closes a dangerous workplace without considering what happens to its workers may expose a genuine abuse while leaving those workers with another emergency.

Being on the worker’s side requires attention to consequences as well as intentions.

There are alternatives to treating cheaper labour as the main path to competitiveness.

One is to examine the organisation of production before treating the wage as the problem. In a randomised experiment involving Indian textile firms, Nicholas Bloom and colleagues provided management consulting to selected plants. Their study reported a seventeen per cent increase in productivity in the first year through improvements in quality and efficiency and reductions in inventory. It was a specific intervention in a particular setting, not a guarantee that every company can achieve the same result. It nevertheless demonstrated that significant productive gains can come from changing management practices. (Nicholas Bloom, Benn Eifert, Aprajit Mahajan, David McKenzie, and John Roberts, “Does Management Matter? Evidence from India,” 2013.)

Before declaring labour too expensive, has the business examined the cost of its own disorganisation?

Another alternative is to invest in the quality of supervision and working conditions. The independent Tufts University assessment summarised in Better Work’s “Progress and Potential” reported improvements in labour conditions alongside gains in business performance. Research on supervisory training found substantial productivity improvements, including an estimated gain of around twenty-two per cent in the studied training context. These findings do not show that every wage increase automatically pays for itself. They challenge the claim that respect for workers and commercial performance must always conflict. (Better Work, “Progress and Potential,” 2016; Tufts University evaluation of Supervisory Skills Training.)

Such evidence matters, but it should not become a new condition that workers must satisfy before their dignity is recognised.

What happens when the ethical choice genuinely costs money?

Some improvements may pay for themselves through lower waste, fewer errors, better retention, or stronger products. Others may require owners, buyers, or consumers to accept part of the cost. A serious argument for decent work must be able to defend that possibility too.

If safety is protected only when it raises profit, profit remains the final judge of whether safety deserves protection.

The purpose of a right is partly to establish what cannot simply be traded away whenever doing so becomes advantageous.

At the company level, improvement should combine adequate basic pay with safe work, predictable conditions, accessible grievance procedures, and meaningful employee representation. Profit-sharing can supplement an adequate wage; it should not replace it with uncertainty. Employee ownership can create opportunities for participation, but it is not automatically fair if workers take financial risks without corresponding control.

Productivity gains should also become a subject of negotiation. If the same output can be produced in less time, the possible benefits include better wages, shorter hours, reinvestment, lower prices, or increased returns. None of those distributions follows automatically from the machine or the method.

Technology creates possibilities. Institutions and decisions determine who receives them.

At the level of the supply chain, responsible purchasing means contracts that can support the conditions buyers require: realistic lead times, predictable payment, credible commitments, and a willingness to address the commercial causes of repeated violations. Monitoring conditions while continually undermining the means to improve them is not a complete responsibility model.

At the government level, minimum standards need enforcement, and workers need institutions through which they can negotiate above those standards. Wage policy should consider workers’ needs alongside economic conditions, with attention to effects on employment, prices, and informality. Implementation matters; a rule that exists only on paper does not provide a worker with an effective remedy.

Social protection, accessible care, transport, and opportunities to retrain can also change the bargaining relationship. They matter not only as benefits after employment has failed, but as conditions that make refusal less catastrophic.

A worker with another viable option enters the conversation differently.

This suggests a practical way to evaluate reform: does it increase a worker’s ability to say no without losing the conditions of a minimally secure life?

That ability should not depend entirely on exceptional courage. A person supporting dependants should not have to become a hero to report a dangerous machine or ask for wages already earned.

Financing can be reconsidered as well. Longer horizons, manageable debt, and realistic return expectations can create room for investment in people and production. None guarantees ethical conduct. But a business designed around obligations it can meet only by repeatedly squeezing its weakest participants deserves scrutiny before those participants are blamed for resisting.

We must also acknowledge limits. Some enterprises may be unable to survive while meeting acceptable standards. Pretending otherwise would weaken the argument. The answer cannot be that every business model has an unconditional right to continue at workers’ expense. Nor should closure be celebrated without a plan for those who depend on it.

The goal is to protect people through change, not to preserve every existing arrangement or to treat the destruction of jobs as a moral achievement.

Consumers have a role, but they cannot carry the entire responsibility. Labels may be incomplete, supply chains difficult to inspect, and household budgets constrained. A low-paid worker should not be expected to solve another worker’s exploitation by consistently purchasing goods they cannot afford. Responsibility should follow power, information, and the capacity to act throughout the system.

The same applies to the language of personal improvement. Training can expand a worker’s opportunities. It cannot, by itself, guarantee fair treatment in every occupation that still needs to be performed.

If the only advice offered to an underpaid person is to escape their job, who is expected to do that job next—and under what conditions?

A society cannot rely on essential work while treating the people who perform it as having failed to become someone else.

The balloon image returns here with a different emphasis. The task is not simply to decide which businesses may expand. It is to ask what supports that expansion, what it displaces, and who participates in deciding its limits.

Some balloons grow because production becomes more useful and less wasteful. Some grow because they capture a larger share of existing value. Some remain inflated because costs are carried outside their accounts. A meaningful economic judgement must distinguish these processes instead of treating size as proof of achievement.

A company can become larger while the lives supporting it become smaller: less time to rest, less freedom to refuse, less security between payments, less room to imagine a different future.

That outcome should not be hidden inside the word “growth.”

The position taken here is on the worker’s side. That does not require denying business constraints, romanticising poverty, or assuming that every employer has the same power. It requires refusing to make workers’ restricted alternatives the unquestioned foundation of somebody else’s success.

Employment is valuable. Investment is valuable. Productive enterprise can improve lives. Their value should be assessed through the lives they make possible, including those of the people whose work creates the product.

Before a government celebrates an investor, it should ask what happens when its citizens become able to demand more. Before a company celebrates a saving, it should ask who now carries the cost. Before an owner calls a return insufficient, it should be possible to ask what level of insecurity is being treated as sufficient for the worker.

And before we describe labour as cheap, perhaps we should ask what has made the person supplying it unable to charge more.

If the business succeeds as those constraints weaken, it may be helping to create development. If its success depends on those constraints remaining, then we have a different relationship to explain.

When a company searches the world for someone who will do the same work for less, is it discovering a better way to produce—or discovering how little choice another human being has?

And if the answer is the second, what would it take for that discovery to become a responsibility rather than a business advantage?