A real vacuum (or very close to it) would be expensive, yes. That's what Musk designed the hyperloop to avoid; it works with a "low pressure" tube instead. His original hyperloop proposal paper outlines this in the intro:
http://www.spacex.com/sites/spacex/files/hyperloop_alpha.pdf
Another extreme is the approach, advocated by Rand and ET3, of drawing a hard or near hard vacuum in the tube and then using an electromagnetic suspension. The problem with this approach is that it is incredibly hard to maintain a near vacuum in a room, let alone 700 miles (round trip) of large tube with dozens of station gateways and thousands of pods entering and exiting every day. All it takes is one leaky seal or a small crack somewhere in the hundreds of miles of tube and the whole system stops working.
However, a low pressure (vs. almost no pressure) system set to a level where standard commercial pumps could easily overcome an air leak and the transport pods could handle variable air density would be inherently robust. Unfortunately, this means that there is a non-trivial amount of air in the tube and leads us straight into another problem.
The "another problem" being that if you just do low pressure, it means your pod has air in front if it and has to pushing that air around (or if that air has nowhere else to go in the forward tubing, trying to compress it, because not enough of the air can squeeze around the pods at the edge of the tube). That's what the hyperloop is getting around by sucking air in the front and blowing it out the bottom. Working as an air cushion is something of a bonus, and if everything works out right it also lets you avoid the expense of maglev.