If Starlink, OneWeb, etc were cancelled tomorrow, it's likely GTO launch demand would pick back up significantly.
Starlink is both the cause of and solution to this problem.
1. May 2014: OneWeb (formerly 'WorldVu') receives its initial $500 million in funding [1]
2. January 2015: SpaceX's satellite constellation is announced. [2]
3. December 2016: OneWeb receives an extra $1.2 billion from SoftBank and existing investors [3]
4. March 2017: Blue Origin signs OneWeb as a customer [4]
If SpaceX was a neutral transport provider, there's little reason that OneWeb would have chosen Blue Origin over SpaceX. Blue Origin is probably offering very low prices, but SpaceX arguably could match them.
[1] https://spacenews.com/40736google-backed-global-broadband-ve...
[2] http://old.seattletimes.com/html/businesstechnology/20254807...
[3] http://fortune.com/2016/12/19/softbank-oneweb/
[4] https://spacenews.com/blue-origin-gets-oneweb-as-second-new-...
As to why do it themselves as opposed to just be a provider to OneWeb, I think this is a sneaky way to do some price discrimination - SpaceX, like everyone else, may have some spare launch capacity, so having an internal (hopefully-profitable) project to soak up that demand rather than cratering launch prices may be worth it.
What does it mean? It more or less officially confirms that Starlink is the only way they can finance the plans for Mars and BFR. There isn't going to be a huge uptick in launches in the coming years, such that that will be enough to dent their capital requirements. Musk has been attempting to finance some of SpaceX's capital demands with debt recently to avoid further diluting his stake, because it risks pushing him below a controlling interest. It's very questionable whether he'll be able to hold the line on that if the launch rate is going to remain fairly stagnant versus 2018.