Gold, like all other assets, do not go up / down in a straight line. Even if you beleive that it’s heading towards $2000 an ounce, it will go up, stop, come back down, consolidate in a range, then rise again.

As frustrating as it may be, gold reached its latest peak in May and it’s only been 3 months. Three months is not unusually long for gold to consolidate before it resumes its uptrend (if it does). This essay looks at the 6 up,down,consolidate cycles in the past 5 years: http://www.zealllc.com/2006/gamehui.htm

I don’t think the government would purposely want to start a speculation in gold, since it wouldn’t impress the people (voters). Governments hate gold anyways, since they can’t tax and monitor the people if everyone avoids using his bank account and starts using gold. Also, they want to pay their government bills and salaries, and by buying gold, they would have to print even more money for themselves to offset the money they already printed for gold. And last, the US doesn’t produce as much gold as South-America/Africa/Australia, so a high price would benefit those countries. Even in good faith, gold doesn’t really create any productive capacity or jobs, at least in the stock and RE bubble the beneficiaries were in the US.
There could be, however, inadvertent buying of gold by the government if things get really bad, and the dollar collapses. To restore faith, or later to be able to sell bonds, the government might slowly accumulate gold to have something real after the crash. This is in my opinion the reason why they don’t sell any now.

Everything is possible, so diversifyyyy!
Regarding the liquidity there are two schools of thought:
1) Those who believe that creation of dollars just devalue the dollar, and prices of everything will go up. They adhere strictly to the definition of inflation as the “supply of money and credit”.
2) Those who believe that money and credit creation props up certain assets for which they were created for.

I believe that it is probably to some extent both. If say the government prints money for a war, military companies (their shares and employees’ salaries) will go up most. While they in turn spend the money on raw materials and other goods, some of that (but not all) money will slowly flow through the economy and to some extent drive up other prices.
If say the government creates credit by lowering interest rates, they also influence a little what credit is created for (by all these little regulations, 401k rules, margin rules, GSEs, etc.). People since 1995 took credit mostly for houses or cashed out for consumer items, and few people took a loan to buy gold.
Creation of money always leads to driving up the price of gold, since money rarely gets removed from the system. Creation of credit has hardly helped buying of gold as we saw where it went in the 90s. If credit contracts, it is therefore unlikely that gold gets sold either. In a severe depression it could do worse than cash, but way better than the other assets. And in that case people might still buy gold as flight to safety. So, yes, if the government would insist on contracting credit, gold could go down. But how likely is that?