One of the commonly used methods for determining fuel consumption in internal combustion engines is based on the carbon mass balance. It assumes a direct relationship between the mass of fuel burned in the engine and the mass of emitted carbon-containing substances. Therefore, the accuracy and repeatability of fuel consumption results depend on the quality of measurement method used for determining pollutant emission. This paper focuses on the results of motor vehicle pollutant emissions and fuel consumption obtained in interlaboratory tests. Two different chassis dynamometer laboratories were compared, both using two measurement methods: 1) analysis of the concentrations of diluted exhaust components collected in bags, 2) continuous analysis of the emission intensity of diluted exhaust components. The tests were conducted in the Worldwide harmonized Light vehicle Test Cycle using a passenger car with a spark-ignition engine. Significant differences were observed in the results obtained in two laboratories. The average specific distance emission of particle mass had the highest non-repeatability, whereas the specific distance emission of hydrocarbons and carbon dioxide had the lowest. The latter led to high repeatability in fuel consumption as determined by the carbon mass balance method, particularly in the case of the method that measured the concentrations of exhaust gases collected in bags.