> What is the point of models, if they aren't true? Because they can make testable predictions or approximations for the truth. This is not unique to Economic

> What is the point of models, if they aren't true? Because they can make testable predictions or approximations for the truth. This is not unique to Economics, even the "hard" sciences have to deal with this problem. Before relativity was conceptualized, Newton's models for gravity and inertia were considered "good enough" even though they were observed to break down at the cosmic observational level. Even today, many measurements have to account for uncertainty in unit size (even at the macro scale!) and have to be accounted for when crunching numbers in serial calculations. (Wikipedia's got several acceptable pages on several types of uncertainty, especially measurement uncertainty.)