Tech

Tim Culpan is a technology columnist for Bloomberg Gadfly. He previously covered technology for Bloomberg News.

Christopher Langner is a markets columnist for Bloomberg Gadfly. He previously covered corporate finance for Bloomberg News, and has written for Reuters/IFR, Forbes, the Wall Street Journal and Mergermarket.

As far as financial metrics go, analysts and investors have no shortage of data points to slice and dice in their eternal quest to understand the companies they buy, or short sell.

Yet among this cacophony, few are more misleading or useless than gross merchandise value.

Investopedia defines GMV as "the total value of merchandise sold over a given period of time through a customer-to-customer exchange site."

Back during the dot-com boom, similar cool-sounding but ultimately useless metrics were being bandied about by executives keen to talk up their companies. These were businesses that attracted a lot of eyeballs, giving them impressive page views and clicks. But they weren't necessarily generating a lot of well, revenue or profit.

Revenue was an old-economy metric, and investors needed to adjust to the new way of valuing companies, explained the founder of one internet company in 2000. 

China's JD.com and Alibaba are among the e-commerce firms that regularly refer to GMV in their earnings reports, invariably citing it ahead of metrics like revenue and income. Analysts are inclined to follow that lead in their own reports.

Investopedia summarizes executives' and analysts' love for GMV as such:

"Gross merchandise value is one element of an e-commerce site's performance, since the revenue of the business will be a function of gross merchandise sold and fees charged."

The theory is that pumping more transactions through your system will bring economies of scale, widening margins and boosting profit. Unfortunately, that's not the case. GMV's biggest problem is that it doesn't always translate to revenue or profit.

Myth of Scale
While gross merchandise value keeps rising, JD.com is garnering fewer sales per unit of transaction value, an indication that it's not leveraging economies of scale to drive revenue
Source: Bloomberg, company filings

At JD.com, actual revenue per unit of trade has been falling, not rising. At the same time, continued growth in GMV has led to wider and wider operating losses, rather than any improvement.

Alibaba's numbers show a similar disconnect. More than once in the past two years, GMV and revenue growth figures moved in the opposite direction, casting doubt over GMV's actual impact on expanding the top line.

Metric Mismatch
Growth figures for gross merchandise value at Alibaba and JD.com rarely coincide with revenue growth
Source: Bloomberg

Another problem with GMV is the method of calculation. In the prospectus for its May 2014 U.S. public offering, JD.com used a similar definition as Investopedia, but added an important caveat, describing GMV as "the total value of all orders placed on our website...regardless of whether the goods are sold or delivered or whether the goods are returned."

Such a definition raises concerns about juicing the numbers. Sid Choraria of APS Asset Management reckons that returns and cancellations could be as high as 25 percent while "brushing" exacerbates the problem further.   (JD.com claims its direct-sales model -- one part of its e-commerce offering-- makes it less prone to brushing than rivals.) Choraria recommends discounting JD.com's stated GMV figures by 35 to 50 percent. Admittedly, that's pretty harsh, but if any discount needs to be applied, then the quality of the data is by definition questionable.

Profit Motive
JD.com has failed to post net income since its 2014 U.S. listing despite sales and GMV growth
Source: Bloomberg

JD.com says its GMV calculation excludes consumer orders "exceeding 2,000 yuan ($304) that are not ultimately sold or delivered." That sounds transparent, except when you consider this threshold is six times the company's average GMV-per-order of 378 yuan.

Alibaba, on the other hand, said in its 2014 IPO prospectus that its calculation "does not take into account how, or whether, the buyer and seller settle the transaction." The company excludes from GMV "certain product categories over certain amounts and transactions by buyers in certain product categories over a certain amount per day," it said in its latest annual report, without specifying that figure. 

A footnote in JD.com's most recent results stated that if it calculated GMV the way its largest competitor, Alibaba, does, the figure would be approximately 38 percent higher. Alibaba meanwhile plans to hold investor briefings this week to discuss GMV, among other issues.

The great upside to having generally accepted accounting principles, or GAAP, is that investors can work off the same set of definitions across both different companies and time periods. By continuing to report non-standard, nebulous numbers such as GMV, companies add little value to their financial reports and instead confuse investors. But, perhaps that's the point.

This column does not necessarily reflect the opinion of Bloomberg LP and its owners.

(Updates to give more information regarding Alibaba's GMV calculation in 14th paragraph.)

  1. In JD.com's most recent earnings call, there were 30 mentions of GMV versus 18 of revenue.

  2. Brushing is the practice of placing fake orders on e-commerce sites either by canceling later, or sending empty boxes out, in order to artificially inflate transaction-volume data.

  3. APS doesn't disclose its position in various stocks, but has a disclaimer that states the fund "may have a position in this security at the time of posting and may trade in and out of this position without informing the community."

  4. Alibaba's China retail marketplaces revenue was 18.3 billion yuan, according to its March quarter results, a 41 percent increase year-on-year. GMV transacted on its China retail marketplaces came in at 742 billion yuan, a rise of 24 percent.

To contact the authors of this story:
Tim Culpan in Taipei at tculpan1@bloomberg.net
Christopher Langner in Singapore at clangner@bloomberg.net

To contact the editor responsible for this story:
Katrina Nicholas at knicholas2@bloomberg.net