Don’t Count On (Retail) Statistics

Related entries: Business Issues, Consumer Market, Economy, General, Research

Another day, another set of statistical gymnastics. Last time it was foreign investment figures (with or without the financial services sector), before that it was the trade surplus (or capital in-flows disguised as trade), the we had the GDP restatement and miscounting of bad debt by Ernst and Young. Now it is retail sales – with thanks to Access Asia’s endless research (“China’s Retail Sales…What Exactly Are They?”), which is reproduced from their latest newsletter, below:

    “At the end of December last year, as reported previously in this newsletter, the Chinese government, through its National Bureau of Statistics (NBS), announced that it had re-estimated the value of the Chinese economy, in terms of Gross Domestic Product (GDP), upwards by 16.8%. This re-estimate, according to the government, was due to having previously underestimated the private service and retail economy with the overall economy. Little in the way of further detail has been given since this, with no indication of which parts of the service and retail economy, and what this means for the total value (or sectoral value) of retail sales to consumers in China. However, as Carsten Holtz (associate professor of Social Science, Hong Kong University of Science & Technology) points out in the latest edition of the China Economic Quarterly (CEQ), even the latest calculations seem to be a fudge, and the reworking of historical figures for GDP on the expenditure side seems to have been based on previously published real annual growth rates for sectors within the economy, rather than a proper re-evaluation of each sector.

    This is significant, because the Chinese government has given an increased value for the total retail market in China. However, there remains a long-standing problem with the value given for retail sales in China. According to latest NBS press releases, the value of “retail” sales in China in 2005 was RMB6,717.66bn. Of this, RMB170.06bn was achieved by the “others” category; RMB888.68bn in sales was derived from the catering sector; and RMB5,658.92bn came from the combined “retail & wholesale” sector.

    The reason that China observers have long held suspicions about the real size of China’s consumer economy, is due to some inconsistencies in how the data is gathered and presented. For instance, national statistics for retail sales still continue to quote figures for “large- and medium-sized” enterprises only. Furthermore, retail sales (officially) includes retail (under which are included sales of vehicles and fuels – which are not classed as retail sales in other economies), as well as wholesaling, catering and “other” – which includes sales by manufacturers and other agencies. Also, the “retail” market does not seem to tie in with what is reported as “household consumption” within the China Statistical Yearbook (CSY), which is published by the NBS.

    Access Asia investigated what is included under the “retail & wholesale” sector, and if there is a value split between what is wholesale, and what is retail. Although total values for each are not expressed explicitly in the CSY, the NBS does provide value splits by sectors within the total, for both retail and wholesale. By adding each of these sectors for each year, we find that in 2004, total wholesale value was RMB4,293.39bn, while actual “retail” sales amounted to only RMB1,245.58bn!

    Based on the historical figures from the CSY, and the latest 2005 figures provided by the NBS, Access Asia has estimated that actual retail sales, taking out wholesale, catering and “other” sales, amounted to RMB1,273.26bn, whereas the Chinese government talk of retail sales being RMB6,717.66bn – a difference of RMB5,444.40bn!

    This significance of this is great. Foreign companies are investing large amounts of money to develop a consumer market for their products and services within the domestic market in China, based on the assumption that total retail sales (as expressed by official Chinese government data) reached a value of RMB6,717.66bn in 2005. This gives them a per capita retail sales figure of about RMB5,096 (US$631.48) for every man, woman and child in China. If the average household, as claimed by the NBS, has about 2.97 people, then the average household retail spend each year would be about RMB15,136 (US$1,875). Based on these assumptions, you would be able to work out the likely per capita sales on a certain product per annum, combined with unit values, and thus plan a likely market demand for a product in China.

    However, if the total value of the actual retail market is RMB1,273.26bn, not RMB6,717.66bn, then per capita retail sales only amount to about RMB966 (US$120), so for the average household, annual retail spending is RMB2,869 (US$355). Therefore, if you base your marked demand projections on the government’s figure of RMB6,717.66bn, you might think there is a strong market for a certain product, but if looked at from the perspective of retail sales only being RMB1,273.26bn, then the market potential does not look so great.

    This confusion of broader consumption being termed as “retail sales” runs throughout data provided on markets in China, both at a macro and a micro level. Data coming from trade associations and companies in China also reflect this confusion in definitions. Therefore, when retail sales for certain product sectors are given, this is often overall trade, including wholesale trade to retailers and direct B2B sales to institutions.

    What is clear, at this stage, is that retail sales figures, for most products, have been inflated due to including non-retail trade. Access Asia has been attempting to investigate this across all consumer goods sectors in China, in order to try to establish the true split between retail and other trade in goods in each sector. Our aim in assessing what is the true size of the retail market in China is to be able to provide reliable figures for our forthcoming report Retailing in China 2006, due at the end of this year. Yet, we are also looking to publish a guide to the problems with these figures, and the significance of this, in conjunction with the CEQ, in December.”

What is also clear is that these guys know their stuff, and that they are doing an important job in questioning the reams of inter-related statistics that are churned out every month by the NBS and others. There is always a risk in taking statistics at face value so, if relying on them to make critical business decisions, it is well worth taking some time (and budget) to:

    • Confirm data definitions (they may differ from your assumptions, and could include such tricks as double counting of wholesale and retail sales, or using registered residents rather than census data as a sample base).

    • Go to the original source for the original figures (are they credible sources? Have the figures been correctly reproduced? Especially if translated from Chinese. Have they been revised since publication?).

    • Cross-reference multiple sources (data collection, analysis, and results, can vary widely between different organisations – with different agendas, abilities and resources).

    • Do some customised research to test findings in target markets (very low national “per capita” figures may be irrelevant if the target market is limited to a small niche in a few wealthy areas such as Beijing, Shanghai and Guangzhou, and it can be very useful to drill down into key issues that are not covered by external sources).

    • Do a reality check (does it all make sense? If not, find out why – don’t just walk on into the unknown and hope for the best…)

Calculator, anyone?

Leave a Reply

You must be logged in to post a comment.