I'd expect there are lots of factors with glass. Making a new plastic bottle every time might be cheaper on a resource-use than making a glass bottle and washin

I'd expect there are lots of factors with glass. Making a new plastic bottle every time might be cheaper on a resource-use than making a glass bottle and washing it for re-use (which uses energy, and water) some average number of times before it gets broken (5? 20? 50? No clue). But it might not. However, other factors include: labor; transportation & waste costs (glass is heavier and breaks easier); and labor-geographic efficiency enabled by lower transport costs—for example, you might choose fewer bottling plants farther apart for cost efficiency, so you can reduce overhead and concentrate your labor costs in cheap locations, rather than every town having a small bottling plant, or stores having fill-n-cap stations directly in them, or whatever. [EDIT] that is, with lower transportation costs of plastic you might be able to concentrate production in places with cheaper labor.