Green IT Is Not About the Planet. It Is About Institutional Pressure.

I was reading a corporate sustainability report recently, the kind with the matte cover and the wind turbines on page three, and I noticed that the section on data center energy efficiency was longer than the section on actual emissions reductions. The company had a net zero target, a scope one and two accounting methodology, and a paragraph about renewable energy certificates. It did not have measurable progress on any of those things. The report was thirty pages of signal. The operations were unchanged. DiMaggio and Powell (1983) have a name for what I was looking at, though they were writing about organizations in general, not about companies with green cloud strategies. They called it isomorphism, the process by which organizations in the same field become structurally similar. Not because they independently converged on the optimal form. Because institutional pressure made similarity the safest option.

The three pressures they identified map onto green IT adoption with uncomfortable precision. Coercive pressure comes from outside. The EU's Corporate Sustainability Reporting Directive, which started phasing in during 2024, requires large companies and listed SMEs to disclose sustainability information according to European Sustainability Reporting Standards. That is external regulation. Companies are not choosing to report their emissions because they care about the climate. They are reporting because the EU told them to, and the penalties for noncompliance are real. This is coercive isomorphism by the book, and the book is DiMaggio and Powell, who specified that coercive pressure must be external. An internal executive mandate to be green is not coercive isomorphism. The EU regulation is. As I wrote when I discussed institutional isomorphism in AI adoption, the difference between external regulation and internal championing matters because internal mandates can be reversed by the same executive who issued them. External regulatory pressure persists regardless of who sits in the corner office.

Mimetic pressure is the one I find hardest to unsee once I started looking. Every major cloud provider now has a sustainability page. Google has been reporting its carbon footprint publicly since 2009. Microsoft published a carbon negative pledge. Amazon has the Climate Pledge. The pages look similar: the targets, the language, the framing of digital infrastructure as an enabler of green outcomes for other industries. I wrote before about the environmental cost of computing and the gap between reported and actual carbon accounting. The mimetic pattern here is not about whether these commitments are sincere. Some of them might be. The pattern is that once one major provider published a sustainability framework, every competitor had to publish one too, because the absence of such a page became a liability. Under uncertainty about what genuine sustainability looks like in cloud computing, organizations copy the visible artifacts of what looks like success. Google published a carbon footprint dashboard. AWS published one. Azure published one. The dashboards are the mimetic signal. Whether they change operational decisions is a separate question that institutional theory does not promise to answer.

Normative pressure is quieter and slower. Green IT certifications, sustainability reporting standards, ESG rating methodologies, industry frameworks for measuring IT carbon intensity. These are being built by professional associations, standard-setting bodies, and consulting firms. When every chief sustainability officer gets trained through the same certification programs, reads the same GRI standards, and attends the same conferences, they will produce sustainability reports that look alike. As I noted in my post on AI policy as governance theater, normative convergence makes practices feel like best practice rather than what they are: shared professional habit. The reports converge because the people writing them share a professional frame, not because the reports reflect operational reality.

The concept that locks this together for me is institutional decoupling. Meyer and Rowan (1977) argued that organizations adopt the formal structures that their institutional environment demands, myths and ceremonies that signal legitimacy, while their actual work practices continue along a different track. The formal structure and the actual operation are loosely coupled. Decoupled. The sustainability report is the formal structure. The data center running on whatever energy mix is available is the actual operation. I need to be careful here. I do not have a local copy of Meyer and Rowan in English, and my understanding of their specific argument about institutional myths and ceremonial conformity is based on how the concept has been summarized and applied in later literature I have read, including the Robey and Boudreau (1999) paper in my local files, which references Meyer and Rowan for the claim that organizations conform to institutional models even when their efficiency is threatened. The decoupling argument, as I understand it from this secondary reference, is that the organization maintains a gap between what it says it does and what it actually does, and that maintaining this gap is rational from the perspective of institutional legitimacy, even if it is irrational from the perspective of technical efficiency.

I think the decoupling argument is the sharpest lens for reading ESG reports in the IT sector. The report signals commitment. The operations reflect cost and convenience. The gap between the two is not a failure of implementation. It is the predicted outcome of institutional pressure that demands the appearance of conformity without demanding the substance. The organizations that produce the most polished sustainability reports are not necessarily the ones with the lowest emissions. They are the ones that face the strongest institutional pressure to appear green, which means the ones most exposed to regulation, public scrutiny, and peer comparison. The pressure produces the report. It does not necessarily produce the reduction.

This connects to something I tried to articulate in my earlier post on whether IS research has an environmental cost problem it mostly ignores. The IS field studies adoption, value, and governance. It rarely studies the gap between what organizations say about sustainability and what their IT operations actually deliver. That gap is where institutional theory earns its paycheck. Adoption of green IT policies is one thing. I wrote about why organizations adopt AI policies that copy their IT policies, and the mechanism is the same: under uncertainty, mimic the perceived legitimate template. Effective reduction of environmental impact is a different thing entirely. Institutional theory predicts the first. It does not guarantee the second.

The EU regulation is going to generate an enormous amount of disclosures in the next few years. The CSRD requirements apply to thousands of companies, and the disclosures will be detailed, auditable, and comparable across firms. This is coercive isomorphism at scale, and it will make the sustainability reporting population look similar because the rules demand similar reports. What coercion cannot do is make the underlying operations change at the same rate as the disclosures. The regulation forces the report to exist. Whether the report changes the power purchase agreement, the hardware lifecycle policy, or the data center cooling system depends on whether the organization treats the sustainability commitment as internal, not just external.

There is also a question I am not sure how to answer yet about whether some decoupling is functional. Meyer and Rowan, as I understand the argument through secondary sources, were not saying that decoupling is always pathological. In some cases, maintaining a gap between formal structure and actual operation is what allows the organization to keep functioning. If an organization were forced to close every gap between its stated policies and its actual practices, the cost of full compliance might exceed the benefit of the activity. Decoupling can be a rational organizational response to contradictory institutional demands. A company might face pressure from regulators to report carbon intensity, pressure from investors to deliver quarterly returns, and pressure from customers to keep cloud service prices low. Fully satisfying all three simultaneously may not be possible. Decoupling is how the organization lives with the contradiction.

The uncomfortable observation is that the more institutional pressure builds around sustainability reporting, the more organizations will produce reports that look similar, and the harder it will be to tell which reports reflect real operational change and which reports reflect institutional conformity. I wrote about digital colonialism and the way that infrastructure control concentrates in a small number of actors. Green IT has the same concentration problem. The companies that publish the most detailed sustainability metrics are the same companies that build, own, and operate the data centers whose environmental footprint is hardest to verify from the outside. The regulator can demand the report. The regulator cannot easily verify the operation when the operation is a server floor behind a badge reader in a building owned by the reporting company.

I keep coming back to the same distinction. Adoption of a green IT policy is an institutional outcome. Reduction of environmental impact is a technical and organizational outcome. Institutional theory explains the first. It does not explain the second, and it was never designed to. The question for anyone reading a sustainability report is whether the organization has moved from the first to the second, or whether it has stopped at the first because the institutional pressure only required the first. DiMaggio and Powell would not be surprised. They were describing how organizations respond to pressure from their environment. The pressure makes organizations look alike. It does not make them work alike.