What actually makes a rule stick

This is article 2 of 3 in The Enforcer series. Here is article 1.

The test

Sustainability collapsed faster than financial auditing or manufacturing certification because it never had an enforcer, a party with its own money or name at stake, who demanded compliance regardless of conviction. Six institutions could plausibly play that role now. Regulators. Insurers. Lenders. Voluntary standards bodies that certify companies against a set of principles. Courts. And the large, dominant customers at the top of a supply chain.

Here's the test I'd apply to each. Does it change what a company would otherwise choose to do, or does it just look like it does? An enforcer that doesn't survive a change in leadership, a bad quarter, or a shift in political weather isn't an enforcer. It's the same conviction problem wearing a different suit.

Regulation: real teeth, filed down fast

The European Union built the closest thing to a hard mandate the sustainability world has seen. The Corporate Sustainability Reporting Directive and the Corporate Sustainability Due Diligence Directive required large companies to disclose and manage their environmental and social impact, with real penalties attached.

In December 2025, the EU finalised a deal to simplify both. The CSRD's reporting threshold rose from 500 to 1,000 employees, listed small and mid-size companies were exempted entirely, and a new revenue floor of 450 million euros was added. The CSDDD's threshold jumped from 500 employees to 5,000, and 1.5 billion euros in turnover. Mandatory climate transition plans were dropped, the EU-wide liability regime was removed, and penalties were capped at 3 percent of worldwide turnover. This happened because of sustained business lobbying. The one enforcement mechanism with actual legal weight bent the moment political and commercial pressure was applied. Regulation can build an enforcer. It can also unbuild one just as fast, because a regulator answers to politics.

Insurance: real, but narrower than it looks

Insurers are already repricing climate exposure. Property insurers are pulling back from wildfire- and flood-prone markets and raising premiums where they stay. An insurer's own solvency depends on getting the risk right, and it doesn't care whether a client believes in climate change. But look closely at what it enforces. It prices physical risk, whether a building will survive a fire or a flood, and says very little about whether a company cuts emissions or treats its supply chain fairly. Insurance is a real lever on adaptation. It isn't yet a lever on the broader sustainability agenda.

Lending: looks like an enforcer, mostly isn't

Sustainability-linked loans tie a company's interest rate to its performance against sustainability targets. On paper, this is exactly the mechanism this series is arguing for. In practice, the research is consistent. The targets are usually chosen by the borrower. So they're rarely ambitious, and rate adjustments are often too small to matter. It's the shape of enforcement without the pressure, a press release with a spreadsheet attached.

Voluntary standards: an enforcer with the incentive pointed the wrong way

There's a category worth separating from the rest because it fails the test for an entirely different reason. The UN Global Compact can, on paper, expel a company for violating its Ten Principles. But the policy sets the bar at "egregious or systematic abuse" that is "admitted by an authorised company representative or" where "there is a finding of guilt in a court of law." In practice, more than 20,800 companies have been delisted since the programme began, and the overwhelming majority were removed for failing to file their annual progress report, not for violating anything substantive.

B Corp shows the same pattern from a different angle. In 2022, B Lab certified Nespresso as a B Corp despite reporting that children as young as eight were picking coffee on farms in its supply chain. Existing B Corps pushed back hard. B Lab's response wasn't to tighten the standard. It defended the decision and lifted its own revenue cap, opening the door to more large companies. A bank gets safer by enforcing strictly. A membership organisation grows and stays relevant by keeping members. Voluntary certification isn't a weaker version of a real enforcer. Its incentive runs in the opposite direction.

Courts and permits: the sharpest teeth of all

The clearest working enforcer in this entire series isn't a company or a regulator. It's a court from the Netherlands.

In 2019, the Dutch Council of State ruled the country's nitrogen permitting system illegal under EU law. The consequence wasn't a fine or a disclosure requirement. It was permit suspension. Thousands of construction and farming projects froze overnight because municipalities could no longer legally issue the licences those projects needed to proceed. In January 2025, a Hague court went further, ordering the government to cut nitrogen emissions by 2030 or pay 10 million euros to the environmental group that brought the case. The government has since spent close to 1.6 billion euros buying out farms in the worst-affected areas, not out of conviction, because farms without a valid permit cannot legally operate.

That's a different category of enforcer than anything else here. It doesn't ask a company to disclose, or price a risk, or choose to comply. It removes the legal ability to operate at all until compliance is proven. Nothing else on this list has that kind of leverage.

Anchor customers: the same company, two different results

This is where the clearest corporate evidence sits, and it comes from one company doing both versions.

In 1989, Walmart launched a green shelf-tag programme, marking products it called environmentally friendly. At its peak, it covered more than 300 items. Walmart never verified the claims behind the label, and when customers discovered that a "recycled" paper towel had only a recycled cardboard tube, the credibility collapsed. The programme faded out by the mid-1990s.

In 2006, Walmart tried something different. It introduced a Packaging Scorecard, rating every supplier on packaging efficiency and feeding that score directly into buying decisions, with a public goal of cutting packaging across its supply chain by 5 percent by 2013. I was one of those suppliers. It was an unwelcome exercise. Nobody on our side wanted to do it, and we said so loudly throughout most of the process. We did it anyway because Walmart was the customer, and the scorecard determined whether we kept the account. Once we'd made the changes Walmart required, we found further savings on our own that nobody had asked for.

Same company, seventeen years apart. The 1989 version ran on branding and collapsed. The 2006 version ran on a scorecard tied to purchasing, and it held. The difference wasn't conviction. Walmart tied its supplier relationships to a measurable requirement because cutting packaging cut Walmart's own shipping and handling costs.

The pattern across all six

Line the six up and the pattern holds, in both directions. Regulation had teeth and lost them the moment political pressure landed, because a regulator's incentive is political, not financial. Insurance has real teeth, but only on the narrow slice of risk that threatens an insurer's own solvency. Lending looks like an enforcer and mostly isn't, because the lender's own return rarely depends on the borrower hitting the target. Voluntary standards bodies look like enforcers and mostly aren't, because their own growth depends on keeping members rather than losing them. The two that actually held — permits in the Netherlands and the 2006 scorecard — held because the enforcer's own position, its legal authority to operate at all, or its own operating costs, improved by enforcing.

That's not six separate findings. It's one finding, six times. An enforcer only holds when its own position improves by enforcing. Everything else is conviction with better branding, and conviction is exactly what just collapsed.

What this leaves unanswered

Most companies reading this are never going to be Walmart, and most readers don't have a court behind them the way Dutch nature reserves did. They can't wait for a regulator whose commitments dissolve the moment the political weather shifts, and they can't build a plan around a membership body that has every reason to look away. So what does a leader without any of that leverage actually do? That's where part three goes.

illuminem Voices is a democratic space presenting the opinions of leading Sustainability Thought Leaders, their views do not necessarily represent those of illuminem.

The world needs sustainability knowledge. At illuminem, no interest group or shareholder can influence our work. Thank you for supporting our mission to make high-quality and independent sustainability information free for all. Every contribution helps. Thank you for donating today.