Actually, Mr. Musk assumes a number of things are used up (expended) during the landing; both an Mars and on Earth. Unless he's planning on heat shield/retro rocket landings; no parachutes. Unlike personnel jumps, spacecraft parachutes are not reusable. The degradation in material strength due to exposure to the space environment (unless you're assuming they're kept heated and rad-shielded throughout the space voyage) means you have to discard them. The fabric is capable of multiple drops; it's the straps which are not. NASA did a few thousand tests on just that question; for several decades.
At 1/1500th, a Saturn V-class vehicle would run to $1,057,483.15 (hardware) + 9,523,209.41 (propellant) + 3,550,204.61 (Admin & Facilities) US / launch. The Saturn V could loft 118 tons to LEO or 47 tons on a TLI trip. (Lunar payloads had to slow themselves up to about 1,060 m/sec, further reducing the effective LLO payload to at or near 30 - 35 tons, maximum. Landing would take an additional load, minimum of 1,870 m/sec...nominally around 2,350 - 2,570 m/sec. Which puts your ability to throw to the lunar surface right around 25 - 28 tons.
So the Falcon Heavy lifts 53 tons plus launch fairing (around 2 tons from the Internet scuttlebut) to LEO. Then you have to lift a second payload up, to propel the first to lunar orbit. Part of the first payload including the lunar lander. That puts you at 106 tons to LEO and about 39.75 tons to the lunar surface. For which he's assuming a minimum charge of $1,336,667.18 US, adjusted for inflation -- assuming the 2 launches occur during 2014. Since the propellant costs almost that amount, I somehow doubt his assertion he can decrease launch costs by a factor of 100. I'd rely more on a factor of 33 1/3.
And the numbers I quoted were NASA numbers, not SpaceX. Elon Musk's SpaceX already launches for around 1/15th of NASA costs. So, the inference I'm making is: "NASA is overcharging the people of the United States by a factor of 1,500?" Sorry, I just don't buy it. I've been watching the space market for longer than Mr. Musk has been out of short pants. He may be a billionaire but he's had few setbacks, of late. Those can delay your entire launch schedule by YEARS, for a single incident. (Which goes a long way to explain a doubling of NASA costs...to 7.5x SpaceX rates.) Statistically, launch failures occur at a rate of 1:5 to 1:20, around the minimum.
Or are his rockets So Good they're supposed to last 100 launches? That's what the Engineers quoted NASA prior to the commencement of the STS Space Shuttle program. And they lasted around 30 - 35 launches, average. So you can take that 1/100th estimate and go ahead and reduce it to 33 1/3. Which puts a Saturn V-class, reusable) SLV at around $40.241 M-USD total cost/launch and SpaceX at $8.91 M-USD/2 launches. That actually sounds realistic, 4.5 - 5x cheaper than a NASA re-usable system.
My estimates put the target of 4 people (plus equipment) on the moon at $290.25 M-USD, 3 - 5 years from today...at the 1/100th price point...which, as I stated before, isn't a realistic target. More like $251.76 M-USD for a "fully reusable launch vehicle." Then again, my goals are different: colonization vs. simple transportation. And that includes the payload, not just launch costs.
