Does the increase in transactions perfectly motivate price increase? Maybe not. Or at least, it's an open question (see gbianchi's model, for example)
Does the no. of transactions rise very obviously in an exponential fashion? You'd have to be pretty biased to deny that.
And there it is, one of the "fundamental reasons" for price growth you keep looking for.
I think you have to be pretty blind to deny that ;)
Well, for fairness sake, you can find growth functions that model the price increase that aren't exponential (cue Jorge: exponential "spurts", purely motivated by outside news), and you can always question whether the (EDIT fundamental) growth function so far continues to hold.
I'm just saying, looking at this chart (or many similar ones graphing the network fundamentals) and concluding we're not undergoing exponential growth requires some serious mental acrobatics... :D
And that's your keyword, Jorge :P