The Reliability of Quarterly National Accounts in Seven Major Countries: A User's Perspective

Abstract

National accounts data provide the most comprehensive overview available of developments in national economies. They are of great interest to a wide range of users of economic information. These users, which include governments formulating budgetary policies, central banks making monetary policy decisions, businesses considering investment decisions, and financial institutions making judgments concerning portfolio allocation have needs which may differ in various respects. However, since their interest generally stems from the likelihood that they will make better decisions if they are well informed about economic developments, they all have a strong interest in the accuracy of national accounts statistics. This paper examines the reliability of preliminary quarterly national accounts statistics. In particular, it considers the longer-term behavior of the provisional estimates to GDP growth and its main expenditure components through an examination of the revisions to those estimates. It covers the seven largest OECD countries and, as such, updates and extends upon previous OECD analysis on the topic. Overall, the results are broadly similar to the earlier work; that preliminary estimates for output growth have not been statistically biased, although the average size of revisions has been large but smaller than those exhibited by the demand components of GDP.