Citing ChannelAdvisor, Reuters confirmed that sales tax collection in Texas and California had an impact both before and after the collection started. Before, it inflated sales considerably, so Q3 2012 might actually have been helped somewhat by the start of collection in California:
Before Amazon began collecting the tax in California, ChannelAdvisor client sales were 5 percent to 10 percent above other states. The week before the September 15 start of the tax, sales spiked as high as 70 percent compared to other states.
And afterwards, predictably, the sales tax led to lower sales:
After Amazon began collecting tax, its California sales leveled with other states. Then, in early November, they slipped as much as 10 percent below other states, ChannelAdvisor data showed.
During one of the busiest holiday periods, in late November and early December, sales dipped further in California vs other states. Toward the end of the holiday period, client sales in California recovered, the data showed.
"There was a sales impact of about 10 percent at the worst point of the dip," Wingo said. eBay, another Amazon rival, is an investor in ChannelAdvisor. Wingo also owned Amazon shares, but sold them in the fourth quarter for personal tax-related reasons.
Most importantly, higher priced items were affected the most, as was predictable:
Amazon's tax collection in California had the most impact on fourth-quarter sales of more expensive items priced at $200 to $250, Wingo said.
Also as per ChannelAdvisor, Amazon.com responded by lowering prices on the most affected items, so there might have been some (further!) earnings impact from trying to retain revenues.
These data points confirm what Best Buy (BBY) had already said: that in the states where Amazon.com started collecting taxes, Best Buy saw a positive sales effect. And obviously it wasn't just Best Buy benefiting, so Amazon.com must have lost sales to a great many retailers.
Also important, apparently Amazon.com is not collecting sales taxes in 3P sales where it can avoid collecting them, so ChannelAdvisor's observations will likely underestimate the impact of these changes. ChannelAdvisor's observations would reflect a lower likelihood of customers to even consider Amazon.com, since the impact would be on products where Amazon.com was not even collecting sales tax.
Trend towards 3P is negative for operating and free cash flow
This is another interesting development. Quite often the trend towards more 3P (Marketplace) revenues is shown by analysts as a reason to be positive on Amazon.com. The reasoning is that these are higher-margin sales and thus as they become larger over time it should favor Amazon.com's overall profitability. Putting aside the fact that's the only time anyone seems to care about profitability, there's another interesting consequence that no one is talking about. Quite simply, the increase of importance of 3P in the sales mix is a negative for Amazon.com's operating and free cash flow (the other often-cited reason for Amazon.com to trade so expensively).
This is simple to understand. When Amazon.com sells 1P merchandise, it gets to collect cash from its customers right away while it takes more than 2 months in average to pay its suppliers. This leads to cash generation from the cash cycle. However, when Amazon.com sells 3P merchandise, while it also collects from its customers right away, it also pays quickly to the merchant. It pays every 14 days, and may pay as fast as every 24 hours. What was a 75 day before having to pay turns into an average paying period of 7 days, or 1/10th of the time.
What this means is that overall, as 1P decelerates and 3P gains relevance, Amazon.com will generate less and less cash from the cash cycle.
On Amazon.com's "Free Cash Flow"
Lately, as Amazon.com's earnings have plunged, a great many explanations cropped up on why the stock continued levitating. Helping this search for reasons, was Jeff Bezos' comments that Amazon.com doesn't seek to optimize margins, it seeks to optimize free cash flow. And sure enough, Amazon.com reports a measure of "free cash flow" that's positive (if not growing).
There's just a problem with this. Free cash flow is also defined as "represent(ing) the cash that a company is able to generate after laying out the money required to maintain or expand its asset base." Now, Amazon.com is not distributing anything to its shareholders, yet it is issuing debt, which means that after "laying out the money required to maintain or expand its asset base" nothing is left and there's a deficit - meaning there is no free cash flow whatsoever.
Once again, we get clues as to the sales tax impact on Amazon.com. And it isn't good. It will probably lead to further revenue deceleration, and, barring that, further earnings impact. This on a $120 billion market capitalization that trades for 3,200 times TTM earnings and >150 times ever-plunging forward earnings, with deep threats to its business including losing its larger advantage (no collection of sales taxes). Amazon.com will end up losing most of its value at some point.
Disclosure: I am short AMZN.