Time-series analysis of how government regulation has affected the U.S. air pollution trends since 1970 has generated considerable interest in the area of environmental economics . Using recently developed structural break tests that are valid for difference-stationary (DS) and trendstationary (TS) data and efficient unit root tests that allow for breaks we examine the trend behavior of two air pollutants, Nitrogen Oxides (NO X ) and Volatile Organic Compounds (VOC).We concentrate on answering two questions.

Use of the time‐series econometric techniques to investigate issues about environmental regulation requires knowing whether air pollution emissions are trend stationary or difference stationary. It has been shown that results regarding trend stationarity of the pollution data are sensitive to the methods used. I conduct a Monte Carlo experiment to study the size and power of two unit root tests that allow for a structural change in the trend at a known time using the data‐generating process calibrated to the actual pollution series. I find that finite sample properties of the Perron test are better than the Park and Sung Phillips‐Perron (PP) type test. Severe size distortions in the Park and Sung PP type test can explain the rejection of a unit root in air pollution emissions reported in some environmental regulation analyses.