When organizations don't want to or won't contribute to making things better the policy response is frequently to put in place a protectionist model that works like this:
The model is not intended to foment excellence or encourage greatness, but to prevent an organization from screwing things up in some narrow part of what it does. Generally speaking, this isn't the thing that defines them as an organization, but something that once in place must be complied with in order to function as that organization. The model works by identifying a point in a metric as the difference between failing and not failing. If the organization fails, it is punished or sanctioned to promote compliance. If it does not fail, it is free to go about its larger work as an organization.
And because there exists a real advantage to not comply given the costs and effort often required, repeated verification for compliance is likely to be necessary. Successful cheating (think Volkswagen and their emission scam) creates a competitive advantage, and so trust by those doing the compliance check is going to be in short supply.
This model is reasonably effective when trying to promote the basis for water quality standards, or ensure we have minimally clean air to breathe, but that's it. The model draws its power by selecting a small piece of what an organization does, most generally in the form of an easily measurable output, and forcing compliance through the risk of sanctions or punishments, sometimes to the point of denying the organization the right to exist.
It is also important to note that full compliance says little about the organization's effectiveness. That determination is elsewhere.
But the greatest limitation of this model is that the line that gets drawn in the sand as the minimum will have a ceiling effect as well. If a factory is well above the threshold and needs to cut costs, it can limit some of its efforts to do that and still be in compliance. That sort of regression is to be expected, but in many cases it will be seen as a reasonable trade-off in creating an overall benefit for society. In situations where parts of an organization are in compliance and others are not, those that are in compliance will be ignored and those that are not will get the attention, which will contribute to the regression as well.
Thus this model would be a terrible model for an organizational accountability, or for building trust with stakeholders, or creating a continuous improvement mindset, since it doesn't care about such things.
What is so damned frustrating regarding school accountability is that this is the very model that was chosen, but the rhetoric of policy makers presumes it has all the capacity necessary to transform every school into a great organization, and so they hammer away on the model and wonder why schools aren't magically transforming themselves to match the rhetoric. But it's even worse than that because they picked a metric (predictive test scores) that doesn't do what they they think it does or mean what they think it means. So now they get mad when schools don't comply with the metric that doesn't mean what they think and wonder why schools aren't becoming great in a system that was never designed to make them great in the first place.
Makes you wonder.