CHTR - Charter Communications

[Spekulatius]

Fiber to the home used to cost $3000 when Verizon did their buildout, but this has dropped to below $1000/ home (LBTYA mentioned £650 for example). Some of cost is cost/ home fixed, some of it is discretionary when customer actually orders it.

 

I would imagine that getting $80/month of high margin business with a $1000 upfront investment is reasonable  ROI, because based on what I have seen, the uptake of reasonably priced FTTH is quite high.

 

I had FTTH first (50MPS first) then switched to cable (200MBPS) and then switched back to FIOS(1GPS). At that point, Played an online game and when I switched  from the slow 50MPS to cable, I noticed increased lag. I investigated with a ping tracer they hater were ping spikes and packet losses, which I had not seen before even with slower nominal  data speed. I was extremely happen when gigabit FIOS was offered and didn’t look back since. The nominal data speeds mean very little when you game or even stream and have ping spikes and packet losses, which depend on you local network, what your neighbors are doing etc because bandwidth is always shared to some extend with cable.

 

FTTH get rid of these issues and that it why I is superior.  This will become very obvious when people start to stream video music on circus devices at the same time.

[Liberty]

utledge said that his whole footprint can pretty inexpensively get to a gig, and has a path to 10 gigs of bandwidth. Fiber is competitive but much much more expensive, and so the ROI for overbuilds doesn't make much sense.

 

I have got. FIOS Gigbit and cable can’t touch it. I actually tried it and switched it toi cable for a while l because they gave me a good offer with 250Mbit, but the issue is not speed, it is ping and packet losses. I ditched cable fairly quickly.

 

Fiber, where it is available  has high market shares and houses when sold listing the availability as a plus. I think it is also getting cheaper to get fiber to the house.

 

Fiber is great. Cable has tons of fiber up to local nodes too. The point is, building new fiber where it doesn't already exist, especially in areas that aren't super densely populated, is VERY expensive. Anything where you have to dig trenches and deal with permits and such, and get new wires to houses will be expensive. If you had to re-build a cable network from the ground up today it would be astronomically expensive too. Over most of Charter footprint, overbuilding fiber doesn't make much sense, especially since most people don't even know what latency or packet losses are (guessing you're a gamer). Even Google has found Google Fiber harder than they expected...

[Liberty]

Fiber to the home used to cost $3000 when Verizon did their buildout, but this has dropped to below $1000/ home (LBTYA mentioned £650 for example). Some of cost is cost/ home fixed, some of it is discretionary when customer actually orders

 

These costs cited don't mean much because they'll vary a lot depending on where you are (population density, regulations, labor costs, etc). Fiber in rural alabama won't cost the same per passing as in Hong Kong.

 

Extending fiber to somewhere that already has it is one thing, bringing it to somewhere that doesn't is another, etc.

[Shooter MacGavin]

Fiber to the home used to cost $3000 when Verizon did their buildout, but this has dropped to below $1000/ home (LBTYA mentioned £650 for example). Some of cost is cost/ home fixed, some of it is discretionary when customer actually orders

 

These costs cited don't mean much because they'll vary a lot depending on where you are (population density, regulations, labor costs, etc). Fiber in rural alabama won't cost the same per passing as in Hong Kong.

 

And also GBP 650 is the incremental cost to connect to Virgin Media's already existing Fiber backbone in the UK across more dense areas.

 

Here's a nice breakdown of DSL vs. Fiber vs. Cable..

 

https://broadbandnow.com/report/dsl-vs-cable-vs-fiber/

[Jurgis]

I had FTTH first (50MPS first) then switched to cable (200MBPS) and then switched back to FIOS(1GPS). At that point, Played an online game and when I switched  from the slow 50MPS to cable, I noticed increased lag. I investigated with a ping tracer they hater were ping spikes and packet losses, which I had not seen before even with slower nominal  data speed. I was extremely happen when gigabit FIOS was offered and didn’t look back since. The nominal data speeds mean very little when you game or even stream and have ping spikes and packet losses, which depend on you local network, what your neighbors are doing etc because bandwidth is always shared to some extend with cable.

 

FTTH get rid of these issues and that it why I is superior.  This will become very obvious when people start to stream video music on circus devices at the same time.

 

I doubt that your packet losses and ping spikes are due to neighbors sharing the network and/or inherent cable (vs FIOS) issues. I'm pretty sure it's due to either the infrastructure the company has locally or losses on physical cable or possibly even connection to backbone (I've even had one game where the issue was backbone to another backbone connection that was being crapped by whatever the backbone interconnect politics was playing at the time). IMO and from what I've heard all of these can occur with FIOS too, so it's somewhat a crapshoot what you gonna get in specific location/specific companies/etc. Best bet is to have a house that has multiple providers and test all. And even then the quality may change in couple months (for who-knows-what reasons).

 

However, I can't prove any of the above, so FWIW. 8)

[Liberty]

I had FTTH first (50MPS first) then switched to cable (200MBPS) and then switched back to FIOS(1GPS). At that point, Played an online game and when I switched  from the slow 50MPS to cable, I noticed increased lag. I investigated with a ping tracer they hater were ping spikes and packet losses, which I had not seen before even with slower nominal  data speed. I was extremely happen when gigabit FIOS was offered and didn’t look back since. The nominal data speeds mean very little when you game or even stream and have ping spikes and packet losses, which depend on you local network, what your neighbors are doing etc because bandwidth is always shared to some extend with cable.

 

FTTH get rid of these issues and that it why I is superior.  This will become very obvious when people start to stream video music on circus devices at the same time.

 

I doubt that your packet losses and ping spikes are due to neighbors sharing the network and/or inherent cable (vs FIOS) issues. I'm pretty sure it's due to either the infrastructure the company has locally or losses on physical cable or possibly even connection to backbone (I've even had one game where the issue was backbone to another backbone connection that was being crapped by whatever the backbone interconnect politics was playing at the time). IMO and from what I've heard all of these can occur with FIOS too, so it's somewhat a crapshoot what you gonna get in specific location/specific companies/etc. Best bet is to have a house that has multiple providers and test all. And even then the quality may change in couple months (for who-knows-what reasons).

 

However, I can't prove any of the above, so FWIW. 8)

 

First question I'd ask if he was on wifi or on a wired connection...

[Munger_Disciple]

First question I'd ask if he was on wifi or on a wired connection...

 

Good question.

 

Also, why on earth would anyone need 1 Gbps connection to home?

 

 

[Spekulatius]

Because I can get 1 Gig and it’s not that pricey

 

Jurgis is correct that some of the issues appeared to be backbone related ( some local nodes in Long Island) but again, it’s the providers job to get this right.

 

It was also played on wireless on a mobile device. But then again, I had the Verizon router and the cable companies router, but then again, it is the providers job to get the hardware right. My pingtracet indicated that wireless wasn’t the issue though.

 

When on cable internet, I even had issues with Netflix or Amauon video streams jamming up. Again YMMV but in my comparison test, Fios performed better. Small dataset, I know.

[Liberty]

Because I can get 1 Gig and it’s not that pricy.

 

Jurgis is correct that some of the issues appeared to be backbone related ( some local nodes in Long Island) but again, it’s the providers job to get this right.

 

It was also played on wireless on a mobile device. But then again, I had the Verizon router and the cable companies router, but then again, it is the providers job to get the hardware right. My pingtracet indicated that wireless wasn’t the issue though.

 

When on cable internet, I even had issues with Netflix or Amauon video streams jamming up. Again YMMV but in my comparison test, Fios performed better. Small dataset, I know.

 

Ok. Glad it fixed your problem. Whatever fixes it for you was the right thing to do, I suppose, but with this context, it pretty much has nothing to do about cable vs fiber as technologies.

[vince]

anyone hear anything lately that may have caused the price to drop this much? not complaining, i'm buying more.  i do wish tho that they would be able to buy more stock back at these levels

[atbed]

BAML came out with a note on VZ this week. They were very bullish on 5G wireless broadband which may be the cause of weakness on CHTR stock this week. Their general opinion is that wireless broadband will be a viable #2, but they don't expect an impact on cable/fiber until 2020.

 

anyone hear anything lately that may have caused the price to drop this much? not complaining, i'm buying more.  i do wish tho that they would be able to buy more stock back at these levels

[vince]

Thanks atbed

[vince]

so charter trading for 84 billion.  buy it thru lbrda and its 78 billion.  buy it thru gliba for 70 billion.  ebitda for 2018 should be 16 billion.  normal capex at 15% on 42 billion revenue equals 6.5 billion.  interest about 3.5 billion.  no cash taxes, fcf 6 billion.  pretty good multiple whichever vehicle u choose.  now lets see 4 years out.  with 4 years of 5% ebitda growth (mgmt thinks high single or low double digit growth but trying to see value if operating environment is tougher) we get 20 billion ebitda, 13 billion ebit, 8.5 billion ebt (we borrowed another 25 billion to keep leverage ratio constant).  for sure they will have found a way to push cash taxes back further.  so fcf for next 4 years (4.5, 5.5, 7.5, 8.5) equals 26 billion (i still used elevated capex).  so we have 50 billion with debt and fcf to allocate.  lets assume all on buybacks (which is conservative cause an acquisition will probably add more value).  at an average of 450 dollars a share we can buy back 110 million shares.  not good at figuring out options but lets assume we lose 20 million shares back to mgmt, we will have 185 million left outstanding and about 45 dollars of fcf per share.  a 15 multiple on fcf seems rational, brings us to almost 700 a share.  even a multiple of 10, using pretty conservative numbers thru-out gives us 450, very good in this low interest rate , high priced environment.  full disclosure, i have 50 percent of my net worth in this idea.

[BG2008]

so charter trading for 84 billion.  buy it thru lbrda and its 78 billion.  buy it thru gliba for 70 billion.  ebitda for 2018 should be 16 billion.  normal capex at 15% on 42 billion revenue equals 6.5 billion.  interest about 3.5 billion.  no cash taxes, fcf 6 billion.  pretty good multiple whichever vehicle u choose.  now lets see 4 years out.  with 4 years of 5% ebitda growth (mgmt thinks high single or low double digit growth but trying to see value if operating environment is tougher) we get 20 billion ebitda, 13 billion ebit, 8.5 billion ebt (we borrowed another 25 billion to keep leverage ratio constant).  for sure they will have found a way to push cash taxes back further.  so fcf for next 4 years (4.5, 5.5, 7.5, 8.5) equals 26 billion (i still used elevated capex).  so we have 50 billion with debt and fcf to allocate.  lets assume all on buybacks (which is conservative cause an acquisition will probably add more value).  at an average of 450 dollars a share we can buy back 110 million shares.  not good at figuring out options but lets assume we lose 20 million shares back to mgmt, we will have 185 million left outstanding and about 45 dollars of fcf per share.  a 15 multiple on fcf seems rational, brings us to almost 700 a share.  even a multiple of 10, using pretty conservative numbers thru-out gives us 450, very good in this low interest rate , high priced environment.  full disclosure, i have 50 percent of my net worth in this idea.

 

What vehicles is in that 50%? CHTR, GLIBA, etc

[vince]

all 3

[BG2008]

Care to disclose further allocations and the logic behind it? Thanks.

[vince]

when you ask about further allocations, you mean to other companies? and logic behind the heavy concentration?

[vince]

allison transmission, trading at 5.5 billion.  average fcf over last 4-5 years is 500 million, should be between 550-600 million this year.  so we are starting with a 10 percent yield and that 10 percent is free to distribute, not needed to grow along with nominal gdp.  my experience shows that when you start with a 10 percent yield (or will reach it within a couple years with a high probability, like chtr), in a reasonably good business with sustainable earning power you are going to do well.  obviously if u start with a 10% yield and the business earnings grow with gdp u have a 15% return (10% yield plus 5-6% nominal growth) with no multiple expansion. (this assumes mgmt doesnt like burning our retained earnings on dumb allocations and is potentially conservative cause at these multiples, if mgmt buys back lots of stock we will actually be getting more than a dollar of market value for every dollar retained and therefore a bump to that initial 10% yield).  and u are starting at the low end of price to earnings multiples so much better chance you will get multiple expansion, which means an excellent return.  for this to work, the business must have low-negative working capital needs (payables and accrued expenses greater than sum of inventory and receivables) and maintenance capex (call the company) equal to or less than depreciation. (or else the 5-6 percent growth will consume some of the 10 percent starting yield).  back to allison... they have a 60% worldwide share of teir endmarkets in fully automatic transmissions for med-large sized trucks (think fire trucks, refuse, school buses, motorhomes). penetration in north america is 80%, in China and other large emerging economies its 5%. they check the box on the 10% fcf yield.  doesnt look like an outlier cause its been 5 years of that same level of fcf.  will continue this in another post cause we have to still  assess their likelihood of 5-6 percent nominal growth and im out of space

[vince]

my apologies, forgot i was on chtr board.  i will copy and paste the alsn writeup and finish it under allisons board for those who want to see it

[Jurgis]

when you ask about further allocations, you mean to other companies? and logic behind the heavy concentration?

 

BG2008 might be asking in what percentages do you split your position between CHTR/LBRDA/GLIBA.

[vince]

Hi Jurgis,

 

Its a little complicated because of timing.  at first it was mostly lbrda because of the shares i was given in spinoff from liberty and the rights to buy more at 20 percent discount.  acquired about a third of chtr exposure thru lbrda at 45 bucks.  then bought about a third directly in chtr at 190, i wanted to own some directly just in case.  more recently, the last third, been buying lvnta (gliba) cause i want more chtr but i want it at cheapest price possible.  in addition i have bought and sold some lbrda profitably and also some long dated conservatively striked chtr calls.

 

Vince

[vince]

Jurgis and BG2008,

 

what do u guys think about chtr, the cable industry and my valuation?

[Jurgis]

I hold LBRDA and GLIBA shares. I don't really have much to say about valuation.

[walkie518]

Jurgis and BG2008,

 

what do u guys think about chtr, the cable industry and my valuation?

 

Charter is evolving.  Adding mobile is no small feat.

 

Charter's MVNO w/Verizon should prove fruitful, but we also find that on the 20% of Charter's new customers' mobile usage (according to the company 80% is over wifi/Charter's own pipes today), is beholden to Verizon's network. 

 

I bet Verizon squeezes if Charter is successful?

 

Does Charter build, buy, or lease towers to overcome reliance on its MVNO?  Does Charter make a bid for Sprint?  Does a bid for Sprint ruin its relationship w/Verizon? 

 

Studies have shown that quad-plays at scale can reduce costs by 40%.  It's more likely that Comcast sees this 40% before Charter and Charter should be able to learn a lot from Comcast's new offerings.  Charter should be better as a result?

 

It might be that customers don't sign up for quad-plays without significant financial incentives?  How close does the price for a charter phone need to be to charter's cost to get this done the right way?  Does Charter need to take a loss on mobile and video to gain scale?

 

Charter is about building value.  Maybe I'm wrong and I should be thinking about how Charter can reinvest to build a better tomorrow for shareholders? 

[vince]

Hey Walkie,

 

lots of good points on wireless but not a material part of my valuation.  Dont really know how verizon squeezes them, they have a mvno agreement.  my understanding is our mvno usage based costs will be minimal because most of bits on wifi anyway.  in addition, it is not going to be a material profit source when viewed as a sole business,  but will lower churn (which would be material).  in my mind, if it is successful, it will drive consolidation on terms favaorable to cable.  but i have to admit i havent spent lots of time on the wireless opportunity.  if it disappeared tomorrow it wouldnt change my valuation