Treat every promise a mining company makes as worthless until it carries a number, a date, and a consequence. That sounds harsh. It is the difference between a contract that protects your community for thirty years and a page of good intentions the company can set aside the moment it becomes inconvenient. A mine will offer you jobs, a clinic, a road, a share of the revenue. Whether any of it reaches your community depends on the drafting, not the warmth. It depends on how the promise is written, who signs it, how it is monitored, and what happens when it is broken.
Companies make promises easily and keep the ones that are watched. That is the pattern I have seen from inside these operations. This article is about learning to read what a company offers with a cold eye. You learn to tell a binding commitment from a vague gesture, to spot the red flags, and to get everything in writing before anyone signs. It draws on my handbook for communities, “When the Mine Arrives,” and on the details that decide whether a file gets acted on or filed away.
A promise you cannot enforce is not a promise
Your agreement with the company is a contract, sometimes called a Community Benefit Agreement, a Community Development Agreement, or an Impact Benefit Agreement. Its name matters less than its enforceability. An agreement that lists the company’s obligations but has no way to monitor them, no reporting requirement, and no consequence for breach is a statement of intention, not a contract. In the words of the handbook, an unenforceable agreement is merely a memorandum of understanding, and memoranda of understanding do not survive changes in company management, ownership, or finances.
So the first test you apply to any promise is simple. Can you prove whether it was kept, and can you do something if it was not. A commitment to “prioritize local employment” fails that test. There is no number to measure, no date to check, and no consequence attached. Now compare a clause that passes. It requires 35 percent of semi-skilled and skilled positions to go to community members, reported quarterly to a monitoring committee. A shortfall triggers a corrective plan within 30 days. Same subject, entirely different value.
Get your own independent legal counsel to check enforceability before you sign anything. The requirements vary by country. Your lawyer should advise whether the agreement must be witnessed, notarized, or registered with a government authority to bind the company. Ask, too, whether it can be enforced in your national courts. This is one area where spending a little early saves a great deal later. Do not rely on the company’s lawyers to tell you what protects you. Their job is to protect the company.
What turns a commitment into an enforceable one
Four elements separate a real commitment from an empty one, and you should look for all four in every clause. The first is specificity. A promise needs a number: a percentage, a monetary amount, a quantity, a quality standard. Vague words like “prioritize,” “support,” “where feasible,” and “endeavor to” are the language of escape. The second is a timeline. A commitment without a date is a commitment without a deadline, and a commitment without a deadline is never late. Every obligation should say when it happens and how often.
The third element is monitoring and reporting. A number you cannot verify is as weak as no number at all. The clause should require the company to report against the target on a fixed schedule, in a format your community can actually read, disaggregated enough to check the specific commitment. It should give a joint monitoring committee the right to inspect, to request more detail, and to bring in independent verification paid for by the company. My guidance on what belongs in a community benefit agreement returns to this point: reporting is what makes every other term real.
The fourth element is a consequence. A commitment with no penalty for breach relies on goodwill, and goodwill is exactly what erodes over a thirty-year mine life. Effective clauses attach real consequences to non-compliance: financial penalties, the right to suspend certain activities, or escalation to the project’s lender. Every agreement also needs a succession clause that binds any future owner. Mines are bought and sold, and a new owner may refuse to honor a deal it did not sign. Specificity, timeline, monitoring, consequence. If any one is missing, the promise has a hole in it.
How hard you can push on these terms depends on your leverage, not your uncertainty about what is reasonable. A community sitting atop a world-class deposit, with skilled advisers and a supportive lender, can insist on tight drafting and real penalties. A marginal deposit in a depressed market gives you less room. Know your position, and draw your lines where your leverage actually lets you hold them.
Read the structure, not just the sentiment
Once you can test a single clause, read the whole agreement for the promises that sound generous but deliver little. The mining sector is full of them, and they follow recognizable shapes.
Employment targets that exclude contractors are the most common. The target may cover only the company’s direct employees, not the hundreds or thousands of workers hired by contractors. If so, the real share of local jobs can be a fraction of what the agreement suggests. Insist that targets cover the entire project workforce. Social investment that serves the company is the next red flag. A road that connects the mine to the port but bypasses your village. A training center that produces exactly the workers the mine needs. A hall built mainly for the company’s own meetings. Each looks like a benefit and functions as an operating cost for the company. Then there is revenue sharing that sounds large and pays nothing. Consider a scenario drawn from patterns across mining agreements. A community signs for “five percent of profits.” For years the company reports little or no profit while shareholders are paid well. It gets there through inter-company loans, parent management fees, and transfer pricing to affiliates. Five percent of a minimized number is close to zero. Royalty-linked payments on gross revenue, independently audited, are far harder to manipulate.
Infrastructure promises hide a related trap: the white elephant. A school without teachers, a clinic without medicine or staff, a water system that breaks within a year because no one budgeted for maintenance. These come from a specific failure, where the company decides what to build without asking the community or planning how to keep it running. Two things fix it. The agreement should fund maintenance and operating costs, not just construction, and the community should keep genuine authority over how any development fund is spent. A fund the company controls is a company fund with a community label. Look for a board with a community majority, an annual independent audit shared with everyone, and clear rules against governors steering money to their own relatives.
Resettlement carries its own red flag: compensation set at market value rather than full replacement cost. In many rural areas, land has low market value because no real land market exists, yet enormous use value as the basis of food, water, and social life. The international benchmark, IFC Performance Standard 5, is clear: avoidance first, full replacement cost, livelihood restoration, and no family moves until the new site is genuinely ready. Hold the company to that standard whether or not a lender formally requires it. And remember the rule that runs through the whole handbook: every benefit must come with a protection. A new clinic does not pay for poisoned water. Sign nothing where the matching protection is missing.
The promise behind the promise: how it will be watched
A commitment is only as strong as the machinery that checks it, so read the monitoring and grievance provisions as carefully as the benefits. A joint monitoring committee is the standing body that reviews the company’s performance, and its value lies in its powers, not its meetings. It needs unrestricted access to data, provided before each meeting in a readable format. It needs the right to visit the site, including unannounced visits, because scheduled inspections show only what the company chooses to present. Watch for “health and safety” used as the reason to reschedule or narrow a site visit. In my experience it is the most reliable phrase for keeping a committee away from whatever the visit would surface. The committee also needs authority to commission independent testing at the company’s expense. And it needs a written escalation path to senior management, the regulator, and the lender.
Grievance mechanisms deserve the same scrutiny. Most company-run mechanisms are structurally biased, because the party being complained about is the same party that investigates and decides the complaint. The UN Guiding Principles on Business and Human Rights set the benchmark a good mechanism should meet: legitimate, accessible, predictable, equitable, transparent, rights-compatible, and a source of continuous learning. Insist on an independent review stage whose decision binds the company, complaints accepted orally and in the local language, protection from retaliation, and defined timeframes for each step. My analysis of why having a grievance mechanism is not the same as it working goes deeper on this gap. One detail from inside the process is worth carrying with you. The decision that most often determines whether a complaint is taken seriously is the categorization at intake, long before any investigation begins.
Test every clause before you sign
The Company Promise Evaluation Checklist turns this cold reading into one instrument you can run over any offer, clause by clause, before it becomes binding. It is a checklist of 16 checkpoints across five sections. The first three sections cover testing enforceability and legal status, checking the four elements of a real commitment, and screening for the classic red flags. The last two examine the monitoring and grievance machinery, and confirm that every benefit is matched by a protection. Each checkpoint is scored In place, Partial, or Absent, so a single glance shows which clauses are solid and which are goodwill dressed as obligation. Any clause scored Partial or Absent is a term to renegotiate or strike before signature, not after. Run it with your independent legal counsel, one provision at a time, and do not let a warm meeting substitute for a clause that holds. Download the Company Promise Evaluation Checklist, work through the draft agreement line by line, and keep it as your record of what still needs fixing.
Why a mediated, structured negotiation gets you enforceable promises
Reading promises well is defensive, and defense alone does not get you a better contract. The deeper problem is that a community negotiating alone, against a company with experienced lawyers and a standard template, tends to end up with the company’s language. Vague clauses survive because no one at the table has the leverage or the process to insist they be tightened. An adversarial fight does not fix that either; it hardens positions and rarely improves the drafting. What changes the drafting is a structured negotiation, facilitated by an independent third party. Both sides put their interests on the table, a neutral keeps the process fair, and every commitment is written to be tested. Independent facilitation is not a favor to the company. For the weaker party, it is how vague promises get turned into enforceable ones.
That is what the Social Accord Architecture (SAA) is designed to produce, and it protects the community’s ability to hold the company to its word. It treats the agreement as Modular Agreements rather than one undifferentiated page. Each unit carries its own evidence of success or failure, so you can prove one stage before the next locks in. It builds in a Feedback Loop, the real-time data that tells you whether a promise is actually being kept. A structured, mediated process is what gets those features into the contract in the first place, and what keeps the promises enforceable across the decades a mine runs. If you want help reading an offer or negotiating enforceable terms before you sign, reach me at [email protected].