Japan's balancing item: do timing errors matter?

Abstract

This study aims to answer the question of whether Japan's balancing item patterns are a matter of timing errors or not. Using time series econometrics techniques, and the components of balance of payments accounts, this study provides an empirical answer that Japan's balancing item pattern is essentially due to timing errors.

Acknowledgement

This study is a minor research under the Monash University Malaysia Research Grant #B-10-05.

Notes

¹ Brooks and Fausten's (1998) study has been criticized by Tombazos (2003) in respect of the data of balancing item used, by showing that Brooks and Fausten's results derive from data that incorporate excessively a dynamically asymmetric concentration of revisions and are therefore unsuitable for statistical analysis.

² The study also examined the possible structural breaks of balancing item, and the endogenously determined break dates are 1973:1, 1973:2, 1987:1 and 1996:3.

³ This study does not discuss the issue of revisions of balancing item as done by Fausten and Pickett (2004) and Tombazos (2003), both for Australia, since the exact number of revisions cannot be determined due to the unavailability of information from the Bank of Japan and related agencies on-line. We, however, reserve this issue for further study.

⁴ Duffy and Renton (1971, p. 461) noted ‘… in a way that is meaningful from an economic point of view – for the “significant” and negative coefficient on the lagged first differences of the balancing item suggests that this item accounts for timing errors in the recording of transactions’.

⁵ As contained in the IMF Balance of Payments Manual, a balancing item is considered ‘too big’ if it exceeds 5% of the sum of gross merchandise imports and exports.

⁶ This study does not employ the new approach of cointegration for examining the possible long-run relationships between components of balance of payments accounts, and balancing item in Japan. Perhaps economic theory should be used to support the use of cointegration approach. As noted, the influences of the components of balance of payments accounts are an empirical issue rather than theoretical argument. Balancing item is fundamentally an accounting matter from the balance of payments accounts, and their economic relationships are unclear. The relationships between balancing item and the components of balance of payments accounts can be explained from the behavioural economics’ viewpoint.

⁷ dx for Windows version 3.00, EconData Pty Ltd (site licence: 2094A), Monash University Malaysia.

⁸ The general VAR estimates with 3 lag length (based on Hannan–Quinn criterion from 12 months lag length, and the estimator is OLS) are not reported here, but available from author upon request.