• QA 137
    Why are there differences between the disappearance and final inventory figures in section 02.01.03 of the Guide, and the ICO data base?
    How were the disappearance and final inventory figures in section 02.01.03 of the Coffee Guide calculated? I assume ICO data was used, but for example the inventory on ICO's data base does not match and the disappearance figure is neither the disappearance data alone (member countries onl;y) nor added to net imports of non-member countries of the ICO.
    Asked by:
    Agribusiness Banker - Brazil

    Differences arise because Coffee Guide data are updated once a year whereas ICO data are constantly revised as new information comes to hand. This why section 01.01.02 clearly states that the Guide is not a statistics source as such and only provides a small number of the most important trend statistics. Instead, the reader is referred to a number of databases, of course including that of the ICO.

    Your observation is important however and warrants more detailed comment…

    ICO data are derived from Certificates of Origin and returns submitted by producing member countries, as well as information from consuming countries. Data from producing member countries is not always submitted promptly and sometimes not at all or very late and, as a result the statistics will always be incomplete to a degree.  Information from other sources such as government institutions often contain estimates that are subject to revision and, there are also problems with some of the import data where, for example, quantities of soluble/roasted coffee are not converted but simply reported as green bean, there is double reporting and so on.

    As a result ICO data, both current and historical, is constantly being revised, a large and in fact never-ending task. But an important one because, in the end, the ICO's database whilst not perfect either certainly is the most detailed of all. The Coffee Guide on the other hand does not pretend to be a database and therefore only updates its data once a year, based on what is then available and including any changes that may have been made to data for previous years.

    To answer your question differently though, the data as given on the Coffee Guide website is correct but, the figures are based on the ICO's Coffee Statistics No. 41, and on their Green and Processed Coffee Trade Statistics 2000 - 2005 (Corr) published July 2006.  By now the ICO is already on Coffee Statistics No. 43 * and yes, the figures are different, not only for 2005 but for other years as well. However, the figures for 2005 are not significantly different in that disappearance is now put at 84.8 million bags against our figure of 85 million bags, and inventories are put at 20.1 million bags against the figure of 19.8 million quoted by the Coffee Guide. In the context of the Coffee Guide and its target audience these are not major differences and certainly do not change the trends that we aim to show.

    As and when we come to update the 2006 statistics, any changes that may have been published for earlier years will then automatically be incorporated.

    * To point out that whilst the figures being queried are easily accessible in ICO statistics 43, others in the table are not because ICO statistics are published on a quarterly basis. ICO Statistics 41 deals with calendar years whereas Statistics 43 deals with the twelve-month period July/June.  Furthermore, the breakdown into imports and re-exports by type is only published once a year. Basically the quest for correct and all-inclusive data is never ending…

    Posted 16 February 2007

    Related chapter(s):
    Related Q & A:
    Q&A 007, 057, 076, 091