New light on the 'impressionistic view' of the balancing item in Australia's balance of payments accounts

Recently, Fausten and Brooks offered (what they refer to as) an 'impressionistic view' of the temporal evolution of Australia's balancing item, which is a measure of the accuracy of the balance of payments accounts. They claim that the balancing item 'has been increasing in magnitude and volatility, violating with increasing frequency internationally agreed acceptability criteria for smallness'. In the present paper it is shown that Fausten and Brooks' results derive from data that incorporates excessively a dynamically asymmetric concentration of revisions and is therefore unsuitable for statistical analysis. This paper develops, and empirically evaluates, a model of the process of revisions of balance of payments data. This model illustrates that dynamically inconsistent time series of the balancing item, such as that employed by Fausten and Brooks, are bound to generate an artificial impression that it follows an 'explosive' time trend. Subsequently, it is illustrated that when alternative, dynamically consistent editions of the balancing item data for the same period as that examined by Fausten and Brooks are employed, their results are reversed. Indeed, the findings here contradict diametrically the conclusions of these authors by suggesting that the decline in the frequency of balancing item 'violations' observed in the latter portion of the relevant time period is unparalleled in the history of the balance of payments accounts.