We re-examine the empirical results on news shocks to total factor
productivity (TFP) by identifying technological diffusion news that
produces an S-shaped response of TFP. By revisiting Sims (2016), we
show that Barsky and Sims’ (2011) result that TFP rises immediately
following a favorable news shock, which is interpreted as a failure to
identify diffusion news, is sensitive; even within Barsky and Sims’
identification framework, the identified news shock is found to produce
an S-shaped response of TFP, as seen in Beaudry and Portier (2006).
Furthermore, we show that VAR information content and measurement
errors can play an important role in identifying technological diffusion
news. We document that stock prices convey more valuable
information about diffusion news than consumer confidence and that
when the identification of news shocks is less affected by measurement
error shocks to TFP, the identified news shock is likely to capture
diffusion news.