The Japanese Food Sanitation Law sets a limit on the migration level of caprolactam for food-contacting nylon products. Here, we carried out an interlaboratory study in twenty laboratories to evaluate the performance of the official GC-FID test method and a GC-MS method as an alternative test method to the official method. Each laboratory quantified caprolactam in three test solutions in 20% ethanol as blind duplicates using GC-FID or GC-MS. The official method (GC-FID with absolute calibration) gave trueness, repeatability (RSDr) and reproducibility (RSDr) values of 96-97%, 3.3-5.4% and 4.0-6.7%, respectively. These values met the target criteria (trueness: 80-110%, RSDr: 10%, RSDr: 25%). The performance of the method was further improved by the introduction of heptalactam as an internal standard. As for GC-MS method, some values of the RSDr exceeded 10% when absolute calibration was used. However, when an internal standard was introduced, the trueness, RSDr and RSDr of GC-MS method were all acceptable at 94-96%, 2.0-4.4% and 7.0-9.4%, respectively. Therefore, GC-MS with an internal standard is available as an alternative test method to the official method.