LMCA - Liberty Media

[Wilson-TPC]

I think I figured it out...

 

Take a look at the section where it says example stake.

 

Please correct me if I'm wrong.

Liberty_Media.xlsx

[merkhet]

The spreadsheet should have live data now.

 

My Mobile Stream stake was buggy -- couldn't pick up the right feed from Google Finance, so it's been updated now.

 

The share counts are estimates.

[loganc]

Great interview with Maffei.  My favorite quote (paraphrased) in response to the question as to what is still interesting about SIRI:

 

They are going to grow EBITDA by a huge amount ... 25 to 30 percent ... grow free cash flow by a massive amount.  The thing is a free cash flow generator of ungodly proportions and it is still growing sort of double digit top line.  I mean ... what more do you want?

 

All the while, SIRI now approaching 52 week lows.  Hard to say what the YTD buyback looks like, but I would guess they have repurchased 8% of the company through 3Q14.  The recent weakness of SIRI has been such a blessing and I hope it continues. 

 

Also, I noticed the comment Maffei made about Yahoo and the BABA ownership.  I guess I had forgotten about or didn't notice the sale of 20% in 2012 for ~7B pretax.  Talk about snatching defeat from the jaws of victory to the tune of ~40B.

[dowfin1]

The spreadsheet should have live data now.

 

My Mobile Stream stake was buggy -- couldn't pick up the right feed from Google Finance, so it's been updated now.

 

The share counts are estimates.

 

I am confused re deferred taxes in the sum of parts valuation and corresponding discount.  On SIRI and CHTR stakes, Liberty Media has taxable gains (if liquidated) of over $9B.  While deferred taxes can be viewed as interest free loans and have economic value if assets are not sold, should there not be a discount from the sum of parts valuation for these taxes as the market has already done?

[Wilson-TPC]

Good question.

 

If any case study has been done on Malone and his Liberty stake. Any prudent investor will understand that all transactions should be done in a tax efficient manner. With the sirius stake, it would be preposterous to think that Liberty will even contemplate the idea of liquidating. It would be tax inefficient and unjustly retarded. They now have a commanding stake of 57% and will grow to 60+% by the end of this year. Why sell something good?

 

Charter already has an upcoming spinoff.

 

The market does apply a discount, but not because of the tax. 

[ItsAValueTrap]

The spreadsheet should have live data now.

 

My Mobile Stream stake was buggy -- couldn't pick up the right feed from Google Finance, so it's been updated now.

 

The share counts are estimates.

 

I am confused re deferred taxes in the sum of parts valuation and corresponding discount.  On SIRI and CHTR stakes, Liberty Media has taxable gains (if liquidated) of over $9B.  While deferred taxes can be viewed as interest free loans and have economic value if assets are not sold, should there not be a discount from the sum of parts valuation for these taxes as the market has already done?

 

1- With SIRI, Liberty did sell some shares and pay a small amount of tax on that.  However, the benefits of that outweighed the downsides. 

 

In the future, it is likely that Liberty and Sirius will some day merge.  That would be done in a way that won't trigger the payment of taxes.  So you might as well assume that the present value of future tax liabilities on the Sirius stake is zero.

 

I also believe that the tax basis on some of the Sirius shares is fairly high.  One of the investor presentations goes over that.

 

2- With Barnes and Noble, Liberty did pax taxes on selling most of its shares.  So the deferred tax liabilities for that stake I believe is meaningful.

 

3- With Charter it is likely that the taxes will get deferred for a very long time.  I don't think that Malone will be selling down that stake anytime soon.

The same probably applies for Live Nation.

 

Anyways, the present value of the DTLs should be fairly low.  It might change the discount in Merkhet's spreadsheet by a few percent... which is not that meaningful IMO.

[dowfin1]

thanx for the well reasoned responses. 

[Picasso]

Is this stock really that cheap?

 

60% of the value in sum of the parts is SIRI which is currently trading at a 6% FCF yield.  To say that there should be no discount applied to this stake because they do not plan to sell seems odd to me.  If you want to extract the value out of the holding you either need to apply the FCF appropriately in capital allocation which is not an extraordinary high FCF given the risks SIRI is facing.  So the returns on their stake in SIRI will compound as they reinvest the cash-flow in buybacks but so will the deferred tax liability.  You cannot ignore the tax liability forever even if you think you can ignore it today.

 

I am not sure I want to pay 1x or even 0.9x the total value of the stake for SIRI within LMCA.  I can go buy SIRI stock today at $3.17 and if the stock does well and it trades for $5 in a year, I will only lose about $0.45 in taxes if I sell.  On the other hand, LMCA probably owns SIRI at a net price of $1.00 so if they realize a gain at $5 there will be a tax drag of about $1.  So I can buy SIRI directly and turn $3.17 into $4.55 or I can buy LMCA post spin of LBRDA and own SIRI at $2.82 and turn that into $4 if LMCA stayed at a 10% discount.  Then I would need to pay tax on my sale of LMCA to net out much less than if I just owned SIRI directly.

 

25% of other value is CHTR which is in the process of a turnaround that a lot of event driven guys are aware of.  Investors call CHTR cheap because of the potential, but the stock trades at one of the highest multiples among cable co.'s.  While I have a lot of respect for the fellows running Charter, it does not mean that implies an easy go of things.  The $95 basis in CHTR also implies some deferred tax liability which will grow if CHTR is successful.  Again, does not seem like something I want to pay 1x for.

 

I'll ignore the other 15% for now (or more depending on the value of some lesser known holdings) since most of the success in LMCA will come from their CHTR and SIRI strategy. 

 

I am ignoring some of the clever financial engineering these guys can do such as using LBRDA as an acquisition currency or wholly acquiring SIRI.  It seems to me that the discount on shares of LMCA should be significantly more than 10% given 1) complexity 2) tax liabilities 3) SIRI headwinds 4) a choppy CHTR turnaround.

 

A big discount on LMCA is not a bad thing since they buy up a lot of their own shares.  But perhaps it would be logical to wait for a larger discount first?

[Wilson-TPC]

Is this stock really that cheap?

 

60% of the value in sum of the parts is SIRI which is currently trading at a 6% FCF yield.  To say that there should be no discount applied to this stake because they do not plan to sell seems odd to me.  If you want to extract the value out of the holding you either need to apply the FCF appropriately in capital allocation which is not an extraordinary high FCF given the risks SIRI is facing.  So the returns on their stake in SIRI will compound as they reinvest the cash-flow in buybacks but so will the deferred tax liability.  You cannot ignore the tax liability forever even if you think you can ignore it today.

 

I am not sure I want to pay 1x or even 0.9x the total value of the stake for SIRI within LMCA.  I can go buy SIRI stock today at $3.17 and if the stock does well and it trades for $5 in a year, I will only lose about $0.45 in taxes if I sell.  On the other hand, LMCA probably owns SIRI at a net price of $1.00 so if they realize a gain at $5 there will be a tax drag of about $1.  So I can buy SIRI directly and turn $3.17 into $4.55 or I can buy LMCA post spin of LBRDA and own SIRI at $2.82 and turn that into $4 if LMCA stayed at a 10% discount.  Then I would need to pay tax on my sale of LMCA to net out much less than if I just owned SIRI directly.

 

25% of other value is CHTR which is in the process of a turnaround that a lot of event driven guys are aware of.  Investors call CHTR cheap because of the potential, but the stock trades at one of the highest multiples among cable co.'s.  While I have a lot of respect for the fellows running Charter, it does not mean that implies an easy go of things.  The $95 basis in CHTR also implies some deferred tax liability which will grow if CHTR is successful.  Again, does not seem like something I want to pay 1x for.

 

I'll ignore the other 15% for now (or more depending on the value of some lesser known holdings) since most of the success in LMCA will come from their CHTR and SIRI strategy. 

 

I am ignoring some of the clever financial engineering these guys can do such as using LBRDA as an acquisition currency or wholly acquiring SIRI.  It seems to me that the discount on shares of LMCA should be significantly more than 10% given 1) complexity 2) tax liabilities 3) SIRI headwinds 4) a choppy CHTR turnaround.

 

A big discount on LMCA is not a bad thing since they buy up a lot of their own shares.  But perhaps it would be logical to wait for a larger discount first?

 

Can you please elaborate on some of the "headwinds" you see with Sirius XM?

[Picasso]

I think Maffei was talking at the FBR conference about how EBITDA was growing at 30% plus some other nice metrics.  So clearly they are doing well right now.  But even with 50% growth over the next few years in EBITDA, you're only looking at maybe a 9-10% 2017 FCF yield.  So if things keep trucking along for another few years the stock will produce a lot of cash but the question is the long term nature of the business on any DCF model.

 

I think AT&T announced the other week that they are doubling their data cap usage without raising subscription costs.  It seems highly, highly likely that we are going to see the availability of wireless bandwidth increase over the next several years.  Many major car makers are going to roll out with 3G/4G or whatnot directly into the vehicles.  What is the margin propensity to pay an extra $20 dollars a month when you have direct connection to internet radios or music sharing services like Spotify?  That is going to be a very tough environment for SIRI.

[giofranchi]

So if things keep trucking along for another few years the stock will produce a lot of cash but the question is the long term nature of the business on any DCF model.

 

Liberty Media is not Sirius. Though it might seem it is, because of the very large stake in Sirius and all the talk about a merger. But it surely is not. Liberty Media is an ever evolving entity that focuses on two or three businesses at a time, maximizes their earnings potential, and then moves on and shifts its attention onto something else. Liberty Media is how Malone keeps doing what he does best. It is not a matter of discount. Not at all. Anyone who waits for a larger discounts has not understood what Liberty Media is all about (imo).

 

Three or four more years of increased profitability for Sirius (and Maffei evidently thinks it very likely) is all that matters for Liberty Media shareholders. Because in three or four years everything can change.

 

I already have a very large position, but on Monday I will buy more. If Liberty Media keeps going down, it will most probably be because of some kind of market correction. Of course, I cannot predict that, but if it occurs, I will gladly go on buying.

 

Gio

 

[Wilson-TPC]

I think Maffei was talking at the FBR conference about how EBITDA was growing at 30% plus some other nice metrics.  So clearly they are doing well right now.  But even with 50% growth over the next few years in EBITDA, you're only looking at maybe a 9-10% 2017 FCF yield.  So if things keep trucking along for another few years the stock will produce a lot of cash but the question is the long term nature of the business on any DCF model.

 

I think AT&T announced the other week that they are doubling their data cap usage without raising subscription costs.  It seems highly, highly likely that we are going to see the availability of wireless bandwidth increase over the next several years.  Many major car makers are going to roll out with 3G/4G or whatnot directly into the vehicles.  What is the margin propensity to pay an extra $20 dollars a month when you have direct connection to internet radios or music sharing services like Spotify?  That is going to be a very tough environment for SIRI.

 

Would love to see your model on your EBITDA assumptions and FCF assumptions to arrive at a 10% yield because using your math, FcF would hit approximately 3.7125 billion in 2017 and I don't think that's a 10% yield.

 

By 2017, if I'm assuming that's your time frame for cars capable of data usage. That's 3 years away with roughly 48 million in new car sales. I don't know what your exact car penetration ratio is for cars enabled with data usage, but my assumptions are something close to 5-10% penetration ratio. Even with a smart car or whatever you call it, OEM relationship would still be on the side of Sirius XM as spotify would not have the financial power to help pay for subsidies in the car. The only player in this market that has the financial power to do so is SIRIUS XM. I think it's prudent to think about risks and ask for opposing views, but I think it's much better to project out the exact assumptions and weigh them with the current risks implied.

 

Please do share your model if you have. I'm very curious to see what year you have projected out to that assumes a penetration ratio in new auto sales that would start being detrimental to SIRIUS XM's current penetration and conversion ratios.

 

[Picasso]

I was not saying 50% per year, but 50% total EBITDA growth over a few years.  I think that is aggressive enough given how they have performed the past three years from what has been a major bounce back in new vehicle sales.  That appears to be slowing again but we have no way of knowing that for sure.

 

What subsidies will need to be paid for in the cars?  If you have car makers not implementing wireless internet in their vehicles, they will lose sales to their competitors.  There will absolutely be an arms race to turn your vehicle into another access point to consume content.  GM is already running ads on this and we're a couple years away still.  Tesla vehicles have 3G/4G already implemented with easy access to internet radio.  For a Tesla owner to now go back to a vehicle without internet would be like swapping your cell phone for a landline.

 

I do not have a model since this is all back of the envelope.  I think irregardless there should be a larger discount for this kind of stock.  I can point out other large cap, well covered stocks with 50-60% discounts to NAV.  For example Softbank is at a massive discount to NAV($76 billion versus $120 billion SOTP).

 

Anyone know the average discount to SOTP for LMCA pre-STRZA split?

[Guest JoelS]

Re the complexity discount, tough to have it both ways if by design:

 

"If you set out to optimize something like long-term return on equity, it leads you to complicated structures, primarily out of tax policies and postures around the world - and leverage. So the combination of leverage and tax. People will say "why don't you simplify, and I'd say, well, because look at what this does to returns - and they say, yeah but the stock would go up if you were simpler, and I say yeah, but I’m not concerned about today's stock price - I'm concerned about stock prices five and ten years down the road." - Malone

 

 

[Picasso]

I actually like the complexity discount in stocks like these.  The company can enjoy buying back shares more cheaply.

 

But a 10% discount on LMCA today is accounting for everything fuzzy about the stock, including the complexity.  A complex capital structure can be a 10% discount on its own.

 

Which is why I was wondering if anyone knew the average discount pre-STRZA split before I go plugging in the numbers myself.

[giofranchi]

I think irregardless there should be a larger discount for this kind of stock.  I can point out other large cap, well covered stocks with 50-60% discounts to NAV.  For example Softbank is at a massive discount to NAV($76 billion versus $120 billion SOTP).

 

Personally, I don’t know enough about Softbank, therefore I cannot say why its discount to NAV is so massive.

What I know is that I always think about the business, and never compare something I think I know, like Liberty Media, with something I don’t know, like Softbank, on some kind of financial metrics, like their discounts to NAV.

Without thinking hard about the business and understanding it well, the comparison of financial metrics doesn’t make much sense to me.

 

Gio

 

[Wilson-TPC]

I have discount at 17%.

 

That spread sheet didn't value a piece of the business.

[Wilson-TPC]

Okay. What made you use 50% overal EBITDA growth? Back of the napkin must have a reason.

 

Did you look at the total data cost vs current SIRIUS XM subscription cost? What about spectrum capacity for the carriers to offer data at an affordable monthly price? If cellphone usages have already clogged up most of the current spectrum usages, and cars require even lower frequency spectrums, what makes you think that Internet radio would be the affordable way to listen to music or exclusive contents?

 

Because cars will travel in areas where it will require low frequency spectrums, it would be very very expensive for carriers to buy these spectrums. The current carriers that have these low frequency spectrums are AT&T and Verizon and from current data plans, they are not cheap. The last remaining low frequency spectrums are in for auction early 2015. The last remaining ones and I suspect sprint will be able to get them.

 

Ever wonder why spectrums are getting more expensive and why carriers are now trying to limit data usage?

 

It won't be getting any cheaper as congestion and increased capex will ensure that pricing going forward will remain relatively elevated for price per /gb.

 

And if you think competition between the carriers will lower prices, then think again. Ever wonder why Sprint has spotty services and T-Mobile can only offer limited data?

 

Sprint is filled with high frequency spectrums. They are good only when there's a tower near by and to build a tower requires a lot of capex but the capacity is very high, hence why sprint can offer unlimited data but shitty coverage.

 

T-mobile has low frequency spectrums but lack high frequency ones so they can offer decent coverage, but very low data usage and much slower speed.

 

The only two that are really competing are Verizon and AT&T. So data usage will go up over time, which makes me to believe that data usage in car will only make the use of spectrums more and more expensive.

 

 

[Picasso]

Okay. What made you use 50% overal EBITDA growth? Back of the napkin must have a reason.

 

Did you look at the total data cost vs current SIRIUS XM subscription cost? What about spectrum capacity for the carriers to offer data at an affordable monthly price? If cellphone usages have already clogged up most of the current spectrum usages, and cars require even lower frequency spectrums, what makes you think that Internet radio would be the affordable way to listen to music or exclusive contents?

 

Because cars will travel in areas where it will require low frequency spectrums, it would be very very expensive for carriers to buy these spectrums. The current carriers that have these low frequency spectrums are AT&T and Verizon and from current data plans, they are not cheap. The last remaining low frequency spectrums are in for auction early 2015. The last remaining ones and I suspect sprint will be able to get them.

 

Ever wonder why spectrums are getting more expensive and why carriers are now trying to limit data usage?

 

It won't be getting any cheaper as congestion and increased capex will ensure that pricing going forward will remain relatively elevated for price per /gb.

 

And if you think competition between the carriers will lower prices, then think again. Ever wonder why Sprint has spotty services and T-Mobile can only offer limited data?

 

Sprint is filled with high frequency spectrums. They are good only when there's a tower near by and to build a tower requires a lot of capex but the capacity is very high, hence why sprint can offer unlimited data but shitty coverage.

 

T-mobile has low frequency spectrums but lack high frequency ones so they can offer decent coverage, but very low data usage and much slower speed.

 

The only two that are really competing are Verizon and AT&T. So data usage will go up over time, which makes me to believe that data usage in car will only make the use of spectrums more and more expensive.

 

FY16 EBITDA estimates are for $1.8B versus $1.3B for FY13.  That is pretty close to 50% total EBITDA growth in three years.

 

I have a couple views on the subscriber costs vs. data costs.

 

First, the United States is ranked very low in terms of speed and reliability versus the rest of the world.  I was comparing the returns that T and VZ have versus other worldwide carriers and it seems to me that there is a wide gap there.  Part of the reason why Softbank was trying to merge S into TMUS to get the scale to compete and improve the speeds.  Higher costs to T and VZ may not have to flow down to the consumer in a 1:1 basis given how profitable they already are.  There will need to be massive investment in the infrastructure that will come to the detriment of T and VZ shareholders and those juicy 5-6% dividends.

 

Second, consumers are paying an awful lot in subscriptions these days.  Netflix, Hulu, HBO, Showtime, Starz, wireless plans, cloud storage, cable, broadband, etc.  There is only room for so much and we have started to see it in cable.  If you are paying for broadband, there are people willing to simply keep the broadband and add Netflix on top.  It is already showing in the cable industry numbers and now the cable stocks.  If consumers have another option to keep the data plan and cut off a service, there is a decent likelihood of that occurring.

 

You know more about the strain this extra data usage would cause than I do.  I am just observing what seems to be a high probability headwind for 60% of the value in LMCA.  But these are probably opinions the market may share and could contribute to a LMCA discount.

[Picasso]

I think irregardless there should be a larger discount for this kind of stock.  I can point out other large cap, well covered stocks with 50-60% discounts to NAV.  For example Softbank is at a massive discount to NAV($76 billion versus $120 billion SOTP).

 

Personally, I don’t know enough about Softbank, therefore I cannot say why its discount to NAV is so massive.

What I know is that I always think about the business, and never compare something I think I know, like Liberty Media, with something I don’t know, like Softbank, on some kind of financial metrics, like their discounts to NAV.

Without thinking hard about the business and understanding it well, the comparison of financial metrics doesn’t make much sense to me.

 

Gio

 

I agree, the quality of the business should be a large factor.  I know Softbank quite well and frankly it shares a lot of qualities to Liberty Media.

[giofranchi]

I agree, the quality of the business should be a large factor.  I know Softbank quite well and frankly it shares a lot of qualities to Liberty Media.

 

Then Softbank today is an unbelievable bargain! ;)

 

Gio

[Picasso]

I agree, the quality of the business should be a large factor.  I know Softbank quite well and frankly it shares a lot of qualities to Liberty Media.

 

Then Softbank today is an unbelievable bargain! ;)

 

Gio

 

Ha!  Not to derail this thread, but I saw you talking about TPRE in another thread.  I believe around 10% of their book is in Softbank.

[Wilson-TPC]

Okay. What made you use 50% overal EBITDA growth? Back of the napkin must have a reason.

 

Did you look at the total data cost vs current SIRIUS XM subscription cost? What about spectrum capacity for the carriers to offer data at an affordable monthly price? If cellphone usages have already clogged up most of the current spectrum usages, and cars require even lower frequency spectrums, what makes you think that Internet radio would be the affordable way to listen to music or exclusive contents?

 

Because cars will travel in areas where it will require low frequency spectrums, it would be very very expensive for carriers to buy these spectrums. The current carriers that have these low frequency spectrums are AT&T and Verizon and from current data plans, they are not cheap. The last remaining low frequency spectrums are in for auction early 2015. The last remaining ones and I suspect sprint will be able to get them.

 

Ever wonder why spectrums are getting more expensive and why carriers are now trying to limit data usage?

 

It won't be getting any cheaper as congestion and increased capex will ensure that pricing going forward will remain relatively elevated for price per /gb.

 

And if you think competition between the carriers will lower prices, then think again. Ever wonder why Sprint has spotty services and T-Mobile can only offer limited data?

 

Sprint is filled with high frequency spectrums. They are good only when there's a tower near by and to build a tower requires a lot of capex but the capacity is very high, hence why sprint can offer unlimited data but shitty coverage.

 

T-mobile has low frequency spectrums but lack high frequency ones so they can offer decent coverage, but very low data usage and much slower speed.

 

The only two that are really competing are Verizon and AT&T. So data usage will go up over time, which makes me to believe that data usage in car will only make the use of spectrums more and more expensive.

 

FY16 EBITDA estimates are for $1.8B versus $1.3B for FY13.  That is pretty close to 50% total EBITDA growth in three years.

 

I have a couple views on the subscriber costs vs. data costs.

 

First, the United States is ranked very low in terms of speed and reliability versus the rest of the world.  I was comparing the returns that T and VZ have versus other worldwide carriers and it seems to me that there is a wide gap there.  Part of the reason why Softbank was trying to merge S into TMUS to get the scale to compete and improve the speeds.  Higher costs to T and VZ may not have to flow down to the consumer in a 1:1 basis given how profitable they already are.  There will need to be massive investment in the infrastructure that will come to the detriment of T and VZ shareholders and those juicy 5-6% dividends.

 

Second, consumers are paying an awful lot in subscriptions these days.  Netflix, Hulu, HBO, Showtime, Starz, wireless plans, cloud storage, cable, broadband, etc.  There is only room for so much and we have started to see it in cable.  If you are paying for broadband, there are people willing to simply keep the broadband and add Netflix on top.  It is already showing in the cable industry numbers and now the cable stocks.  If consumers have another option to keep the data plan and cut off a service, there is a decent likelihood of that occurring.

 

You know more about the strain this extra data usage would cause than I do.  I am just observing what seems to be a high probability headwind for 60% of the value in LMCA.  But these are probably opinions the market may share and could contribute to a LMCA discount.

 

What source is that for 1.8 billion EBITDA and what assumptions are those using? What sub numbers and what EBITDA margins? Penetration ratio and conversion ratio and what churn rate? Lastly, what ARPU and SAC?

 

I personally don't care for the discount in LMCA.

 

Now before you compare profitability of Verizon and ATT to other global competitors, you have to look at the scale economics. Countries like Japan can build out more towers because of the density of the population and only utilize low frequency spectrums in rural areas while using high spectrums in cities. In the US, it's not feasible giving the population size and the way the population is spread out.

 

My model currently indicates to me that for a feasible data plan to stream music, carriers will need to charge at least 25-35 per month. Unless a consumer feels that listening to stream music/content costs 2x the current price of Sirius XM then I think I will be wrong giving my projections.

 

The only reason SoftBank wants to buy T-Mobile for sprint is because T-Mobile has second tier low frequency spectrums. And sprint wouldn't need to build the capex needed to use these, so cost saving wise it makes sense.

 

If you think ATT and Verizon shareholders are going to get the shaft while consumers benefit from lower price data packages, then I think you need to study the economics of why telecom companies buy spectrums and spend so heavily on capex. An investment is only worth while if the NPV of the project is greater than the cost. And no where do I see signs of data packages getting lower in the near, medium, or long term.

 

[giofranchi]

Ha!  Not to derail this thread, but I saw you talking about TPRE in another thread.  I believe around 10% of their book is in Softbank.

 

Yes, you are right. Therefore, I indirectly own a bit of an unbelievable bargain! :)

 

Gio

 

[Picasso]

Okay. What made you use 50% overal EBITDA growth? Back of the napkin must have a reason.

 

Did you look at the total data cost vs current SIRIUS XM subscription cost? What about spectrum capacity for the carriers to offer data at an affordable monthly price? If cellphone usages have already clogged up most of the current spectrum usages, and cars require even lower frequency spectrums, what makes you think that Internet radio would be the affordable way to listen to music or exclusive contents?

 

Because cars will travel in areas where it will require low frequency spectrums, it would be very very expensive for carriers to buy these spectrums. The current carriers that have these low frequency spectrums are AT&T and Verizon and from current data plans, they are not cheap. The last remaining low frequency spectrums are in for auction early 2015. The last remaining ones and I suspect sprint will be able to get them.

 

Ever wonder why spectrums are getting more expensive and why carriers are now trying to limit data usage?

 

It won't be getting any cheaper as congestion and increased capex will ensure that pricing going forward will remain relatively elevated for price per /gb.

 

And if you think competition between the carriers will lower prices, then think again. Ever wonder why Sprint has spotty services and T-Mobile can only offer limited data?

 

Sprint is filled with high frequency spectrums. They are good only when there's a tower near by and to build a tower requires a lot of capex but the capacity is very high, hence why sprint can offer unlimited data but shitty coverage.

 

T-mobile has low frequency spectrums but lack high frequency ones so they can offer decent coverage, but very low data usage and much slower speed.

 

The only two that are really competing are Verizon and AT&T. So data usage will go up over time, which makes me to believe that data usage in car will only make the use of spectrums more and more expensive.

 

FY16 EBITDA estimates are for $1.8B versus $1.3B for FY13.  That is pretty close to 50% total EBITDA growth in three years.

 

I have a couple views on the subscriber costs vs. data costs.

 

First, the United States is ranked very low in terms of speed and reliability versus the rest of the world.  I was comparing the returns that T and VZ have versus other worldwide carriers and it seems to me that there is a wide gap there.  Part of the reason why Softbank was trying to merge S into TMUS to get the scale to compete and improve the speeds.  Higher costs to T and VZ may not have to flow down to the consumer in a 1:1 basis given how profitable they already are.  There will need to be massive investment in the infrastructure that will come to the detriment of T and VZ shareholders and those juicy 5-6% dividends.

 

Second, consumers are paying an awful lot in subscriptions these days.  Netflix, Hulu, HBO, Showtime, Starz, wireless plans, cloud storage, cable, broadband, etc.  There is only room for so much and we have started to see it in cable.  If you are paying for broadband, there are people willing to simply keep the broadband and add Netflix on top.  It is already showing in the cable industry numbers and now the cable stocks.  If consumers have another option to keep the data plan and cut off a service, there is a decent likelihood of that occurring.

 

You know more about the strain this extra data usage would cause than I do.  I am just observing what seems to be a high probability headwind for 60% of the value in LMCA.  But these are probably opinions the market may share and could contribute to a LMCA discount.

 

What source is that for 1.8 billion EBITDA and what assumptions are those using? What sub numbers and what EBITDA margins? Penetration ratio and conversion ratio and what churn rate? Lastly, what ARPU and SAC?

 

I personally don't care for the discount in LMCA.

 

Now before you compare profitability of Verizon and ATT to other global competitors, you have to look at the scale economics. Countries like Japan can build out more towers because of the density of the population and only utilize low frequency spectrums in rural areas while using high spectrums in cities. In the US, it's not feasible giving the population size and the way the population is spread out.

 

My model currently indicates to me that for a feasible data plan to stream music, carriers will need to charge at least 25-35 per month. Unless a consumer feels that listening to stream music/content costs 2x the current price of Sirius XM then I think I will be wrong giving my projections.

 

The only reason SoftBank wants to buy T-Mobile for sprint is because T-Mobile has second tier low frequency spectrums. And sprint wouldn't need to build the capex needed to use these, so cost saving wise it makes sense.

 

If you think ATT and Verizon shareholders are going to get the shaft while consumers benefit from lower price data packages, then I think you need to study the economics of why telecom companies buy spectrums and spend so heavily on capex. An investment is only worth while if the NPV of the project is greater than the cost. And no where do I see signs of data packages getting lower in the near, medium, or long term.

 

I pulled the $1.8B from my Bloomberg terminal of which I am attaching.  I figure the market has digested these numbers so any cause to a valuation shift will be based on the result being above or below those values. 

 

If it costs $25-35 for the extra data, keep in mind that there will likely be other services included with this.  It is the internet after all which has a lot of different options for entertainment.

 

I think of it like owning the stock of America's Home Videos.  In case no one watched it, it was that one show that had all the home videos of dogs and babies hitting their heads and people falling in pools, etc.  The show probably made a lot of money and if someone said "one day you can just watch those videos on something called YouTube" the response would be "do you know how expensive/impossible it would be to view those on a 14k modem?"  Granted wireless is going to be different, but I worry about the "it's different this time."  Especially in something like technology.

 

There is also a reason why not just Softbank but Iliad is looking at the US wireless market.  Despite the difference in urban densities, the quality and speeds are still well below where they should be.  Whether T and VZ shareholders will be willing to bear the brunt of this depends a lot on what happens with TMUS and S.

 

The fact that AT&T has doubled their data cap at the same price point is a defacto drop in price.