Analysis · Finance
How an Expectations Assumption Can Reverse an Inflation Signal
By AWEI · AI-compiled · Published · Analysis prepared · 1 source · doi.org
An inflation model can make stronger output look associated with weaker price pressure because of how it represents expectations. In Klaus Adam and Mario Padula’s 2010 study, using surveyed forecasts helped recover a positive output coefficient where estimates assuming rational expectations had struggled.
The supplied June 2009 working-paper version examines quarterly US data from late 1968 to early 2003. The researchers combined GDP-deflator inflation, using the GNP deflator before 1992, with average forecasts for the following quarter from the Survey of Professional Forecasters. These forecasts served as proxies for firms’ inflation expectations.
They estimated a New Keynesian Phillips curve, which relates inflation to expected inflation and marginal production costs under intermittent price adjustment. They compared two cost proxies: real unit labor costs and an output gap defined as log real GDP’s deviation from a fitted linear trend. With survey expectations, both had positive, statistically significant coefficients over the full sample.
The revealing difference was timing. Subsequent inflation was more strongly correlated with current output than surveyed expectations were. Requiring forecast errors to be unrelated to output therefore attributed more output information to expectations than the surveys supported. The authors showed how this discrepancy could push the estimated output coefficient below zero. The corresponding correlation pattern for unit labor costs did not produce the same reversal.
This was an observational model comparison, not a causal experiment. Professional forecasters may not represent firms’ beliefs, and both cost measures are approximations. Tests rejected stable coefficients across the three historical subperiods. Past inflation also retained explanatory power after survey expectations were included, exposing a limitation of the basic forward-looking model.
For interpreting inflation models, the useful check is what “expected inflation” actually means: a reported forecast or an expectation inferred through assumptions. Here, that choice changed the estimated output relationship. It did not establish a dependable forecasting or market-timing rule.
Sources used for this article (1)
Direct links to the publisher reports used to prepare this article.
- Source 1
- INFLATION DYNAMICS AND SUBJECTIVE EXPECTATIONS IN THE UNITED STATES Klaus Adam, Mario Padula
