The Fed can directly manipulate bank reserves. It does so by buying securities from banks. Every other factor in the economy, the Fed only indirectly controls and the more remote the connection between bank reserves and the other factor, the less control the Fed has and the more control over the factor in the hands of others.
By increasing the supply of bank reserves, the Fed can push down the Federal Funds rate, which is the interest rate bank’s charge each other for over-night lending. By increasing bank reserves, the Fed gives incentives to banks to create more credit, which lowers interest rates in various credit markets. But that effect requires banks to expand their loan portfolios, which they do not have to do. Instead, they can build excess reserves, in which case the Fed expansionary policy is blunted. If banks do expand credit in response to expansionary Fed policy, then the lower interest rates and increased demand (by borrowers using the newly created credit) for assets drives up asset prices. But if people are reluctant to borrow, then interest rates will collapse without much asset price inflation. And so on, the effect of Fed policy generates effects on the real economy only through people’s reaction to it. Since their reaction can vary, the effectiveness of Fed policy can likewise vary.
The extreme positions that the Fed has no effect on the economy and that the Fed controls the economy are both certainly mistaken. Sometimes the Fed seems to lead events and sometimes it seems to follow them. While the pattern of a boom-bust cycle set in motion by monetary inflation and credit expansion repeats qualitatively, quantitatively the effects of Fed policy vary widely depending on the circumstances. For example, a much more modest increase in bank reserves during 2003-2007 ignited the housing-bubble boom, but a much more expansive increase in bank reserves during 2009-2014 has not generate a commensurately larger boom.
The greater effectiveness of monetary policy during the boom compared to the bust is well known insight, even in the mainstream: