The effectiveness of the Fed's bond-buying policy may seem like a tool design issue, but it actually depends more on the economic cycle and market expectations.

Comparing 2008 and 2019 makes this clear. During the 2008 financial tsunami, the root cause was the financial system itself exploding—subprime mortgage defaults triggered a domino effect, and interbank financing froze directly. At that time, QE was like the last firefighter, with the Fed directly stepping in to buy problematic assets (like Bear Stearns, AIG), forcibly pulling the collapsing credit system back from the brink. How effective was it? PMI rebounded from below 40 to above 50, showing visible signs of recovery. But at what cost? The Fed's balance sheet expanded to 25% of GDP, leaving a bunch of aftereffects.

In 2019, another liquidity crisis occurred, but this time the nature was different. Trump's tax cuts blew up the fiscal deficit, and a bunch of government bonds drained bank reserves. Plus, the strict liquidity coverage ratio requirements from the Dodd-Frank Act caused financial institutions to hoard liquidity frantically. This problem couldn't be solved simply by adjusting interest rates; the Fed had to find another way—targeted asset purchases.

Interestingly, this approach didn't have as strong a pull on the real economy as in 2008. In early 2020, the US manufacturing PMI was still hovering below the 50 line, indicating that although liquidity increased, real economic vitality hadn't improved much.

Expectations management has also evolved. During the QE era, Bernanke and others created the "forward guidance" tool, which became the Fed's most core communication weapon…