Recent complex financial products sold to households contradict the basic premise of canonical innovation theories: Financial innovation benefits its adopters. In my 2006–2015 sample of over 28,000 yield enhancement products (YEP), the securities offer attractive yields but negative returns. The products lose money both ex ante and ex post due to their embedded fees. On average, YEPs charge 6–7% in annual fees and subsequently lose 6–7% relative to risk-adjusted benchmarks. Simple and cheap combinations of listed options often statewise dominate YEPs. Competition, disclosure, or learning do not eliminate this inferior financial innovation over my sample period.

We find that 30-minute changes in bond yields around scheduled Federal Open Market Committee (FOMC) announcements are predictable with the pre-FOMC Blue Chip professionals’ revisions in GDP growth forecasts. A positive pre-FOMC GDP growth revision predicts a contractionary policy news shock (positive change in bond yields), a negative GDP growth revision predicts an expansionary policy news shock (negative change in bond yields). Failing to account for this predictability biases the estimates of monetary policy effects on the economy. First, the Fed’s information effect dissipates as the truly unpredictable policy news shock does not affect professionals’ beliefs about the economy. Second, net policy shock has a more negative impact on actual future GDP than the raw policy shock.