Summary: The paper analyses the method adopted by the index of employment protection systems developed in 1999 by the OECD, and discusses its reliability and shortcomings, in particular with regards to Italy. The paper highlights a series of methodological problems arising from the aggregation of the respective indicators, and casts doubt on the possibility of developing an aggregate index generated as sum of partial indexes. The paper identifies several errors and imprecision in the development of the index, in the light of an accurate reading of legal data. The greater inaccuracies regard the index of rigidity attributed to legal provisions applicable to "employment termination indemnities", "individual dismissal", "redundancy lay off" and "temporary redundancy benefits". Re-examination of legal data has led to a substantial downgrading of Italy's labour market rigidity index. Yet, the most significant finding is the extreme volatility of the index, according to diverse interpretations of the legal data. This observation is fraught with consequences regarding proper application of information provided by the rigidity index.