The recent raid in Germany of Deutsche Bank and its ESG asset management division DWS over greenwashing claims has revealed a deeper mess, contends an article in the Financial Times about the matter. Though former DWS chief sustainability officer Desiree Fixler claimed that the billions under management at DWS were “ESG integrated,” in reality that label doesn’t actually mean much in terms of action taken by managers, and the terms under which a fund can claim they are ESG compliant are arbitrary and spotty.

In a sign of just how broad the term has become, more than 3,000 investment groups managing $103 trillion worth of assets claim that they are somehow integrated ESG into their management. And a new joint paper from analysts at MIT and the University of Zurich notes how difficult it is to create a structure for ESG investing that is objective and rigorous. The paper, which studied ESG ratings from KLD, Sustainalytics, Moody’s, S&P Global, Refinitiv and MSCI, found correlations them vary from 0.38 to 0.71. Compare that to the correlation from more established and quantitative credit agency ratings at 0.92—a divergence that becomes even wider in subcategories than in the aggregate, the article details.

That “makes it difficult to evaluate the ESG performance of companies, funds, and portfolios,” the paper says, as quoted by the Financial Times. Companies trying to find the best ESG funds to invest in get mixed messages from the rating agencies about which actions are valuable. Companies are then less likely to strengthen their ESG performance and are more likely to under-invest in ESG. The markets may not factor ESG performance into price, but that price could be affected by investor preferences. In either case, “the divergence of the ratings disperses the effect of ESG performance on asset prices.” CEOs might also reach for a specific rating but miss the mark on ESG performance improvement. And having the wider divergence of ESG ratings produces uncertainty into any ESG-focused actions that a manager might take.

That creates what the paper calls “aggregate confusion.” Measurement divergence (how various factors are measured) makes up 56% of the confusion with scope divergence (what is measured) accounting for 38%. That means that different rating agencies are weighting different factors such as lobbying activity, sustainability, or inclusivity. The paper calls for greater transparency, clarity and regulation, but as the Financial Times points out, “once the finance industry has a set of established rules, the gaming starts a nanosecond afterwards.”