Benj....
Good questions, and unfortunately there are no definitive answers.
First, though, a general perspective. It would seem obvious that the amount of money it would take to move the markets the 10-30% you postulate would vary widely based on how and when it was done, with it requiring much more money to move a market that was already overbought (like moving it up from current levels) than it would take to move a market that was already severely oversold. The PPT has members (the Fed and Treasury) that have many years of experience with intervention in the currency markets so we are dealing with folks who are neither novices nor dummies, and they have a thorough understanding of technical analysis and timing.
Now think about how a rally like this generally unfolds. First, the PPT would have to come in and turn the markets, which by definition are already very oversold and susceptible to a ramp. To accomplish the turn, they would not buy the whole market, but instead they would buy index futures, ETF's and key index stocks such as INTC, MSFT, CSCO, and AMAT, among others. The initial buy-in would be expensive, but not hugely so when you think about the assets available - remember that almost every such sudden turn with the PPT fingerprints has been accompanied by mention on CNBC and the financial media by mention of an "unknown very large futures buyer that sparked the turn today."
Once the turn has been made, the shorts run to cover and take profits and the mo-mo boys jump on the bandwagon to make sure that as many shorts as possible continue to cover and this serves to give the rally a life of its own, and also lets the original buyer of the futures begin to lighten up his long positions by feeding them into the rally at a profit. From time to time, he will need to goose the rally a bit to prevent profit taking from setting in, but these are usually short term forays into the markets which then resume short covering and mo-mo buying to move higher. Finally, real investors and mutual funds begin to move in on the chance that this rally is the "real deal" - the bottom of the market, and while all of this is driving the markets upwards, the original buyer is still feeding his position slowly back into the markets at a profit even though from time to time he has to goose the markets a bit to keep the upward momentum going.
The net result is that the total cost for such a move is minimal, but only if you have enough money to work with to make a successful initial move, which is where the risk lies. Indeed, the overall outlay for a rally like we have had could end up as a net profit if it gets enough internal momentum to allow the buyer to exit his entire position. If carried too far though, I would expect that a point of diminishing returns is reached where it takes too much money to keep the rally going and little chance to pull money back out.
The most expensive part of the deal is propping up the dollar, but in that task other countries (Japan and the EU plus Britain) should be more than willing to help out since a strong dollar benefits them far more than it does us. Gold is a very small market (the total market cap of all the gold stocks put together is less than DELL) so it is easily manipulated, and there is no doubt plenty of folks willing to help hold the price of gold down for their own benefit (JPM, GS, BAC, and others), and some of them are already in the PPT.
So to answer your initial question, the amount of money required is great, but only because of the initial outlay and even that outlay would be recaptured as the market moved higher. The net outlay for the whole move would not be anywhere near what you would think, because of the pile on effects of short covering, mo-mo players, and some new money coming in.
Just my opinion, though.
mlsoft
AI
Market Data 
Markets 
