Roughly 10%. http://en.wikipedia.org/wiki/Ski_Jumping
Ski jumpers wear different sorts of clothes than other skiers. Their suits are not form-fitting and are lined with a spongy material that makes them a little stiff - you can see this if you watch how the suit reacts as a jumper walks away at the bottom of the slope. The main purpose is to get the suit to catch a little air to help them get more hang-time - periodically there will be a row about somebody wearing a suit that’s a size or ten too big to increase the effect - but I imagine it also helps cut down on brushburns and wardrobe malfunctions in the event of a bad fall.
Some time after the V-style became predominant there were tendencies to use larger suits (like with the crotch at the knees) and experiment with the air flow properties of the material (allow air flow through the front, impede it in the back, instant balloon and improved air flow). Now there are strict rules about the size of the suit, and about the suits penetrability to air. Still, like in all sports, the athletes have to push as close to the limit as possible, so as not to give the other guy an edge, and sometimes there’s just a little too much stretch in your suit and you get to watch the competition from a spectator seat instead.
For those who aren’t familiar with jumpoing, this is the source of some big problems as jump performances get better and better. At some point the top performers with latest new equipment will be going so far they ‘jump out of the hill’ and start landing on the part where the slope is flattening out.
Then you need either to build a new jump/hill, change the rules (e.g. start lower on the hill) to shorten the flights or get ready for casualties 
I think this sort of thing happened when they introduced the V-style and new suits mentioned above, for instance.
It’s not really that much of a problem. Starting lower on the hill doesn’t require a rule change, it’s the standard procedure in all events. You want the longest safe jumps possible with the jumpers and conditions you have on the day, so with varying wind, and with varying snow on the inrun, you always have to regulate the starting speed.
The only reason for building new hills, or altering old ones, is hill record inflation and improved hill designs, not increased efficiency due to changes in jumping style and equipment.
In case it’s not clear from the previous post, the starting point (that little bench the jumpers sit on before heading down the hill) is moved up or down depending on how the hill is working that day. You want the best jumpers landing in the “landing zone” to avoid the problem of flattened Finns.
The size of the hill (70, 90, 120 meters) is the estimated distance a jumper will fly, not the height of the jump.
Finnish pancakes.
And usually even the Finns will “chicken out” and land before they run out of hill. As has been pointed out earlier, most of the flight is relatively close to the hill, and by “starting to land” they’ll lose a lot of lift and touch down. No point in jumping the hill down when they’ll just stop the event and move the start down a notch.
On the other hand, when someone’s in the top three (and thus the last three) in the final, and gets lucky with conditions and does everything right, and could set a new hill record, they do sometimes land so far down they can’t make the landing.
I bet it smarts when you bang your butt like that.
We’ve got a ski jump here in our flat little midwest town. They jump year-round on that thing. I have no idea where I’d find the ballage to do it. I watched one of the current Olympiads from a jumper’s eye view the other day and I nearly puked. This feat is so not on my list of things to do before I die.