I wouldn't focus too much on just counting the number of variables. Oftentimes the more parsimonious explanation is the one that emerges from a simple principle

I wouldn't focus too much on just counting the number of variables. Oftentimes the more parsimonious explanation is the one that emerges from a simple principle being applied to a lot of moving parts. In this case, I would say that the perspective where it's the value of a dollar that varies is not particularly parsimonious, because it leads to a strange set of implications that could only be resolved by adding epicycles to the theory. For example, if you favor the perspective where it's the value of the dollar that varies between New York and Peoria, then you also need to account for why the dollar/euro exchange rate does not tend to vary between the two. It would be peculiar indeed if it were that the euro's value also varied by location, but just happened to vary in lockstep with the dollar. (I realize this is the scenario that's happening in TFA, but that's kind of exactly what the TFA is getting at: the rial is on the verge of forking into two different currencies.) You would also have to account for why the relative nominal prices of goods don't stay constant from location to location. Suppose that a quarter pounder costs $2 and a tube of toothpaste costs $4 in NYC. If it's the dollar value that changes, then one should be able to assume that a $1 quarter pounder in Peoria implies that toothpaste costs $2 in Peoria. My appetite for air travel right now doesn't permit me to test this myself, but I suspect that you will find that this is not actually the case.