Abstract
Purpose
This paper aims to investigate the cyclical associations among energy prices and key macroeconomic variables for the USA.
Design/methodology/approach
To this end, the recently developed Hamilton filter (HF) and the oft-used Hodrick–Prescott filter (HPF) are used. The two methods produce starkly different results regarding the relationships between energy prices on the one hand and output and employment on the other.
Findings
While the HF suggests that energy prices are acyclical, the HPF suggests they are procyclical. However, the associations between energy prices and inflation are robust across the two methods, indicating that energy prices are strongly correlated with – and lead – the consumer price index (CPI). Furthermore, unlike the results produced by the HPF, those produced by the HF are robust across seasonally adjusted and unadjusted data.
Research limitations/implications
Given the inherent seasonality in energy prices and the differences in the underlying processes that generate macroeconomic and energy prices, the results obtained from the HPF filter should be interpreted with caution.
Originality/value
To the best of the authors’ knowledge, this is the first study that uses the recently developed HF to examine the associations between the cyclical behaviors of three key macroeconomic variables in the USA – the industrial production index, the CPI, and total nonfarm employment – and the prices of natural gas, crude oil, gasoline, diesel, and heating oil. Second, this study presents a comparison of the results produced by the two filtering techniques. Third, recognizing that energy prices are characterized by seasonality, this study tests the robustness of the results produced by the two filters across seasonally adjusted and unadjusted data.
