Endogeneity can be overcome by using some clever Instrument Variable, for one thing. But, causality is a different issue, it really has to do with model specification.

The simplest case, assuming there are NO other variables involved, if you run the following regression and find a to be significant:

Y = aX + u

Does that mean X "causes" Y?

No, the model could also be specified as

X = 1/a * Y + u'

it could also mean Y "causes" X.

Econometrics does not solve everything, one would still need some intuition. Causality is in a sense in the eyes of the beholder.

If you tell me, spending more money on other people makes one happy, I simply don't buy it, I think they got the causality messed up. Happy people spend more money on other people is more likely the case. And why are they happy in the first place? Probably because they already spent a great deal on themselves!