This is the follow up article to "Amazon Earnings Preview: The 5 Metrics to Watch." I have developed five key metrics that are leading indicators of whether or not Amazon (AMZN) will fulfill its promise of delivering the type of net margin, cash flow and long term growth both analysts and investors expect. For the past eight quarters all of these five metrics have trended in the wrong direction leading me to write a prior article entitled "Average Selling Price is the Smoking Gun that Sinks the Amazon Business Model." Those trends continued, and in most cases accelerated in the just completed 4th quarter. Analysts are dead wrong to point to gross margin increases as a sign of a healthy business model, as that metric is a false indicator of potential with Amazon. Amazon is simply transferring costs below the line so that while gross margin goes up, net margin will continue to drop (case in point: in the 4th quarter net margin was cut in half as gross margin improved 330 basis points). Let's review the five metrics and the updated new data.
Metric # 1: Average Selling Price (ASP) of E-Commerce Revenue
What I wrote on 1/27/2013: Over the last eight quarters e-commerce revenue (Total Revenue less "Other") has increased an average of 36% year-on-year, while paid units shipped has averaged a 46% growth (Figure 1). That equates to a 15% decline in the ASP of Amazon's e-commerce products sold over the past two years. This not only pressures gross margin in this category, it also places tremendous pressure on both fulfillment and shipping costs as they become a larger percentage of the overall price of a product. The size of the delta between Paid Unit Growth and E-commerce revenue lets us know the magnitude of efficiency gains required in shipping and fulfillment costs just to maintain current margin levels.
Update 1/29/2013: In the 4th quarter E-commerce Revenue growth was 21% while paid unit growth was 32%. This 11% delta is an acceleration in the decline of ASP, and means that ASP dropped 8.3% Y/Y. Let's see if Fulfillment and Shipping Cost were able to maintain that efficiency level to counter this trend.
Figure 1: Amazon E-commerce Revenue Growth vs. Paid Unit Growth
Metric # 2: Shipping Cost as a Percentage of E-Commerce Revenue (SCPER)
What I wrote on 1/27/2013: Amazon's shipping cost per unit needs to decline by the same amount as the ASP decline for SCPER to remain flat. That means Amazon needs to drive 7.8% efficiency gains in their shipping cost annually just to keep up with ASP trends. Instead, SCPER had steadily increased until 2011 Q3 and has remained fairly flat since then. The figure points out that gasoline costs have largely driven (pardon the pun) the change in shipping costs, and hence why Amazon is adding significant fixed cost by building and leasing distributed fulfillment centers. In spite of the investments, SCPER was higher last quarter than when gasoline peaked in 2011 Q2, so there has been scant sign of return. Amazon needs this investment to pays off for its model to work, so if SCPER doesn't remain flat or decline from here, Amazon could be in trouble long term. If SCPER does continue to increase, Amazon will need increase shipping revenues which would potentially slow revenue growth.
Update on 1/29/2013: Even with declining average gasoline prices (Figure 2) and the roll out of additional fulfillment centers in Q4, Shipping Cost Percentage of E-Commerce Revenue (SCPER) increased Q/Q and Y/Y. The fulfillment center investment is not paying off. But yet on the call, analysts are excited about the efficiencies (that is because they mistakenly look at shipping cost as a percentage of total revenue - they miss the boat here completely).
Figure 2: Amazon Shipping Expenses (SCPER) vs. Average Gasoline Price
Metric # 3: Fulfillment Cost as Percentage of E-Commerce Revenue (FCPER)
What I wrote on 1/27/2013: Amazon's fulfillment cost per unit needs to decline by the same amount as the ASP decline for FCPER to remain flat. That means Amazon needs to drive 7.8% efficiency gains in their fulfillment costs annually just to keep up with ASP trends. FCPER has steadily increased since before the investments in new fulfillment centers even began (Figure 3). Due to the fixed cost mix of fulfillment, FCPER is seasonal and if it is above 2011 Q4's 9.8%, this will provide a bearish indicator on future net margin performance as there will still be no sign of investment return.
Update on 1/29/2013: FCPER came in 1,200 basis points above last year's level and is a very troubling sign. My model predicted it perfectly as fulfillment expense largely scales with unit growth, not revenue (in Q4 E-Commerce revenue was up 21%, units were up 32%, and fulfillment expense was up 36%). This is the primary reason (along with Technology and Content costs) that as gross margin increases, net margin decreases.
Figure 3: Amazon Fulfillment Expense
Metric # 4: North America Direct E-commerce (non 3P) Growth relative to Total US Growth
What I wrote on 1/27/2013: The law of large numbers is catching up with Amazon in the North American E-commerce market. If you assume Amazon 3P sales are at the same commission rate as eBay (11%, or sales through the site are 9 times larger than reported 3P revenue) and have a similar North American mix to direct sales, then Amazon North American E-commerce market share for the past four quarters is 32.4% relative to Comscore data (which does not include Canadian market data). Because of the expansion of 3P sales and raw size of Amazon, growth in the North American market is on a steady decline (from 50% last year to 30% last quarter), and is trending toward market growth. We also know US e-commerce sales were below expectations in Q4 (+13.8% vs. +16%) and Amazon 3P sales were up only 40% (vs. an average growth rate of 72% over the past two years). Add to this the additional headwinds of sales tax in CA and TX and market share gains by eBay (EBAY) and Best Buy (BBY), and Amazon revenue potential was squeezed. A good number for Amazon would be to maintain or beat 2x market growth to stop the decline in Figure 4. That would place North America Direct E-commerce sales at 27.6%. Any number below this would continue the downward trend of slower growth.
Update on 1/29/2013: Amazon completely failed as predicted in my prior articles on this metric growing North American Direct E-Commerce Revenue only 20.8% in Q4. Amazon growth as compared to market (Figure 4) has continued its significant decline over the past 10 quarters with no sign of improving.
Figure 4: Amazon North America Direct E-Commerce growth / Market Growth
Metric # 5: International E-commerce Growth
What I wrote on 1/27/2013: Many of the upgrades over the past few weeks have focused on international growth being the growth driver for Amazon going forward. This is required as the aforementioned North American revenue growth is slowing significantly and AWS will be too small for the foreseeable future to make a meaningful impact on overall Amazon growth. Unfortunately, international e-commerce revenue has been slowing as well and is now pegged at overall international e-commerce growth rates of about 20%. For Amazon to grow at 25% per year as investors have demanded, international performance needs to turnaround and grow between 25-30% over the next few years. A number in the area of 20% would be a bearish indicator, and at or above 25% would be a bullish indicator.
Update on 1/29/2013: Another growth disappointment by Amazon as International sales only grew 20.9% Y/Y. There is no sign of Amazon proving they can grow e-commerce sales above market internationally.
Figure 5: Amazon International E-commerce Growth
Amazon missed expectations on all of the five key metrics that are required for them to achieve the predicted growth and net margins analysts are forecasting. The following table summarizes Amazon's actual Q4 performance vs. the 5 key metrics. Analysts are misreading the financial performance of the company by focusing on gross margin rather than key segment revenue growth, and how ASP and fulfillment expenses are decreasing net margin while gross margin increases.