4 ms·
Their costs have reportedly increased by ~11x http://money.cnn.com/2011/07/08/technology/netflix_starz_contract/index.htm http://money.cnn.com/2011/07/08/techn
by quandrum 15y ago
Their costs have reportedly increased by ~11x
http://money.cnn.com/2011/07/08/technology/netflix_starz_contract/index.htm http://money.cnn.com/2011/07/08/technology/netflix_starz_con...
- encoderer 15y agoTheir licensing costs. And even then, that's a proforma FY2012 guess from an analyst that's not affiliated with the company. It's worth asking if Wedbush Securities has a short position on NFLX. This is a company that operates dozens of physical warehouses, employs thousands of people, and pays an astonishing amount in postage fees. Their licensing costs are surely just a fraction of their bottom line.
- seanalltogether 15y agoYou've got that the other way around. In 2010, Fullfillment Expenses cost them 9.4% of their revenue, Cost of Subscription cost them 54% of their revenue.
- cube13 15y agoAccording to Netflix's 2010 SEC filing, their operations expenses are around $400 million. Adding in development, marketing and general administrative costs, their total non-license costs were around $800 million. They paid $1.2 billion in licensing costs. It's their most expensive cost, by far.
- encoderer 15y agoThat includes their physical disc's, doesn't it? The post I replied to -- and my reply -- was addressing streaming licensing: Pachter predicts Netflix's streaming content licensing costs will rise from $180 million in 2010 to a whopping $1.98 billion in 2012. In other words, of that $1.2bn you quote from their 10-K, 1.02bn was physical discs.
- losvedir 15y ago> It's worth asking if Wedbush Securities has a short position on NFLX. It doesn't look like they take positions themselves. They seem to be your standard financial services broker/dealer, in which case they do research for clients who take positions. It's likely they have a "Sell", or at least "Hold", rating if they're writing negative notes. But that's as it should be; I think you have the causality backwards in your implication.