Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
A vector-autoregressive model of actual output and expected output obtained from surveys is used to test for information rigidities and to provide a characterisation of output dynamics that accommodates these information structures. News on actual and expected outputs is decomposed to identify innovations understood to have short-lived effects and these are used with the model to derive a ‘news-adjusted output gap׳ measure. The approach is applied to US data over 1970q1–2014q2 and the new gap measure is shown to provide a good leading indicator of inflation.