I had literally never thought of this in any way. I've just finished reading two papers: one which shows that net-investment isn't really correlated to savings

I had literally never thought of this in any way. I've just finished reading two papers: one which shows that net-investment isn't really correlated to savings or interest rates which exactly supports your position (with data!); and, a second which shows that the ratio of old-to-young people is an incredibly good predictor of interest rates (with data!). You're both right? I suspect that means there must be a third option.