What definition of pseudoscience would capture economics without capturing medicine, ecology, or meteorology?

Everyone's just using models here, and the way we incorporate statistical observations to define the limits of the models' scope, and refine the models over time, or reject the models entirely, applies to economists, meteorologists, seismologists, and many branches of actual human medicine.

Popper would define pseudoscience as predictions that can't be falsified, but surely that can't apply to the idea of the weatherman predicting rain and being wrong, right?

Kuhn came along and argued that science is about solving problems within paradigms, and sometimes rejecting paradigms in scientific revolutions (geocentrism vs heliocentrism, Newtonian physics versus Einstein's relativity), but it wasn't a particularly robust test for separating out pseudoscience.

Lakatos categorized things further at explaining how model-breaking observations could be handled within the structure of how science performs its work (limiting the scope of the model, expanding the complexity of the model to fit the new observations, proposing specific exception handlers), but also observed the difference between the hard core of a discipline, in which attempts at refutation were not tolerated, and auxiliary hypotheses where the scientists were free to test their ideas for falsifiability.

But when you use these ideas to try to understand how science works, I don't think economics really stands out as less scientific than cancer research or climatology or other statistically driven scientific disciplines.