1. How many years/money did it save: historians have various estimates but among those that have looked at how the information was used by the Soviets (information that only came out in the early 2000s), the answer is in terms of time is: at most a year, maybe no time at all. Why? Because the Soviets did not use the information to save time or money. They used it, at most, as a guide to their program, and more often, as a "check" on the research their own scientists were doing. In general they did not use it to avoid any "blind alleys" because they were not sure how much to trust the information. As for cost, the most expensive part of any bomb program, by far, is the industrial production necessary to produce the fissile material. That would not be dramatically affected by espionage (even if you know how to do it, you still have to do the building, mining, operations, etc.). To put this into perspective, 84% of the Manhattan Project's cost was in building the facilities necessary to produce enriched uranium and plutonium. A mere 4% was spent on "research and development." And again, the Soviets did not save on R&D just because they had espionage information. They just informed their R&D with the information. But they still spent on it, and may have ultimately spent somewhat more (because they distrusted their own scientists, the people running the project frequently had the same work performed by different teams redundantly).

2. Manhattan Project security was not lax at all. It was, however, designed primarily to keep the project secret from American sources: from the press, from Congress, from leaks and incidental disclosures. It was also designed to detect and deter sabotage and rumors. It was not designed to find "moles" (scientists who happened to be working for the Soviets as well). Detecting "moles" is very difficult indeed — to totally avoid them, they would have had to prohibit all of the scientists from ever leaving the project sites. In the USA that is not really a viable option (it was, however, in the USSR), especially if your primary concern is getting the bomb built on time (in which morale plays a role). Remember also that there were, at most, maybe 10 "moles" in the entire Manhattan Project (and only a couple of these were truly important; the value of the others to the Soviets was mostly to corroborate their information) — a very small number to detect among all of the others, and any program that increased the "false positive" rate (e.g., flagging innocent people as suspicious) would also slow it down.

3. Yes, it was a factor in McCarthyism and the Red Scare. The first major spy was discovered in 1950 (Klaus Fuchs), and it had a dramatic effect on American politics. The Fuchs investigation also led to the discovery of the Rosenbergs, which was one of the core legal cases at the time. In general, it is hard to imagine McCarthyism having the overall effect it did without the revelation and fear of atomic spies — it added an existential horror that underwrote the entire anti-Communist endeavor.

A good overall book on these issues is Michael Gordin's Red Cloud at Dawn (FSG, 2009), which is an account of the Soviet bomb program that makes a lot of reference to the Manhattan Project and to the questions regarding espionage. On the failures of Manhattan Project security... the best book on this has not yet come out! My book has an entire chapter on what Manhattan Project security did and did not accomplish, and why it was particularly ill-suited for catching spies. But it won't be out for probably another year.