# of Cell Tower Sites per State in the USA

Discussion in 'Tower Component, Lighting , Accessory Discussions' started by Backup Power, Jan 22, 2012.

  1. Backup Power

    Backup Power Friend of the Community

    We are doing some research and are looking for the following information:
    1. # of Cell Tower Sites per State in the USA
    2. # of Roof Antenna Sites per State in the USA
    3. Typical DC & AC load of Cell Tower Sites
    4. Typical DC load of Rooftop Antenna Sites
  2. David Ward

    David Ward Friend of the Community

    Using the Tower Maps database the cell tower and rooftop counts by state are in the attached file.

    - David Ward www.towermaps.com

    Attached Files:

  3. JPHFC

    JPHFC Friend of the Community

    Hi David,
    Thank you for some very useful data. Quick question: is there a definition for "Other" compared to Towers & Rooftops - the total of 88,159 looks quite high.
  4. John Sandford

    John Sandford Frequent Poster

    Your total of towers, 110,228, and rooftops, 64,089, might be representative of the country, but what are the 88,159 others? Can't be that many DAS.
  5. David Ward

    David Ward Friend of the Community

    The number of "other" sites is dominated by billboards. We break down sites into one of about 8 "types". When a site owner doesn't specify, we make our best effort to assign the "type", however, there are still many which are simply unknown.
  6. UCI_IE22

    UCI_IE22 Friend of the Community

    Hi David,

    A couple follow up questions and comments about your data:

    Does the total tower number in your spreadsheet include collocated sites or is that number actual towers?

    A JP Morgan report that came out last year performed the following calculation to determine the total number of cell towers in the US.

    "We have used the CTIA's published figures to track cell sites, which at the end of June 2010 totaled 251,618 cell sites in the US. These are total cell sites and we assume 25% are on rooftops and other alternative structures to towers, leaving about 75% on towers. We estimate the average tenant per tower in the US ~1.7."

    The result of that equation is 111,000 actual towers.

    The total number of towers is inline with your numbers if they are actual towers but if your numbers are cell sites located on towers, then your actual tower count would be signfigantly lower. JP Morgan also estimated that 75% of all cell sites were on towers ,while your data shows 42%. Could you provide any commentary on the process that the JP Morgan analyst used to size the market?

    The CTIA lists total cell sites as about 256k but I believe that includes collocated sites which means towers and rooftops with multiple carriers are double or triple counted. Does your company consider the information coming from the CTIA to be accurate?

  7. Wireless Estimator

    Wireless Estimator Administrator Staff Member

    Hi UCI_IE22,

    Trying to identify an accurate count of towers and collocations in the US is almost as daunting as trying to discover what humankind should expect on Dec. 21, 2012, when the Maya's "Long Count" calendar might signal the end to all cellular service and everything else.

    Thanks for providing the JP Morgan report:

    "We have used the CTIA's published figures to track cell sites, which at the end of June 2010 totaled 251,618 cell sites in the US. These are total cell sites and we assume 25% are on rooftops and other alternative structures to towers, leaving about 75% on towers. We estimate the average tenant per tower in the US ~1.7."

    In order to use the equation of 111,000 cell towers, one must rely upon JP Morgan's considerable research to identify the tenants per tower rate of 1.7.

    Or was there a thorough examination?

    Although year-end reports come out next week for American Tower and Crown Castle, at the end of 2010, both companies said they averaged 2.6 tenants with SBA Communications and other major towercos nipping at those numbers.

    Using JP Morgan's average, 35% is a pretty impressive amount for "others" since wireless internet service providers such as Clearwire are tallied in CTIA's cell site count.

    But if we presume that their tenant count is spot on, then the 111,000 cell towers would not be correct since CTIA has previously stated that their data gathering is a largely volunteer effort of carriers and not all of their members respond to CTIA's inquiries.

    More importantly, CTIA counts each carrier's site, so if four carriers were co-located on one tower, it would still be considered as four individual cell sites.

    In effect, even if JP Morgan could identify that they have accurate tenancy information, any tower counts derived from it would be based upon inaccurate information.

    CTIA and PCIA have previously stated that there were no companies or representative data within the industry that could provide a reliable total of towers.

    We're hoping that more accurate information will be available on Dec. 22, 2012.:)
  8. rclaxton

    rclaxton Friend of the Community

    Why in the world would anyone give credibility to J.P. Morgan's guesstimate?

    One of their analysts said that the big four made up 75% of the revenue of the tower companies. Throw in Metro, Leap, Clearwire, US Cellular and the RCs and it puts their 111k tower count in the toilet.

    This is the company that didn't jump into seriously following the wireless sector until the mid 90s.

    These are the same fine folks who advised AT&T for the T-Mobile deal and lost their multi-million dollar fees because the deal didn't go through and their client lost $6 billion.

    Then they kicked their client while they were down by downgrading AT&T's stock because the deal didn't go through.

    Ironically, when they did, the stock went up.

    I think you would be better off using a Ouija board rather than J.P. Morgan's analysis.
  9. Ken Schmidt

    Ken Schmidt Friend of the Community

    The problem with tower databases (we maintain and sell our own database as well) is that by nature of the manner that cell site location data is made available (FCC, FAA, tower company data, carrier tower data, and other misc sources), it is impossible to estimate the number of rooftops or actual cell sites. I believe that RBC Capital does the best job of estimating the number of cell sites by aggregating counts by the carriers. In fairness though, I don't agree that because the tower companies average 2.7 tenants per tower, that the average for all towers is remotely near 2.7. The tower companies are aggregators of multi-tenant sites- and in some of the previous transactions with carriers did not purchase single tenants sites that would always be single tenant towers. For example, when the T-Mobile tower portfolio is sold, the number of tenants per tower will be significantly less than 2.7. Portfolios like AT&T tower will also be substantially less primarily because AT&T's inbound collocation process has historically been lengthy and difficult to work through. (It is getting better.)

    Since most rooftop cell sites are directly leased by the landowner to the carriers and the carriers have little incentive to actually disclose those locations, any data regarding rooftops is suspect. Furthermore, the vast majority (80%-90%) of sites marketed as rooftop sites by the tower companies do not actually have carriers using them. They are simply buildings with whom the building owners have seen fit to agree to a site management agreement with the tower companies. The same is true for "other" structures.
  10. JPHFC

    JPHFC Friend of the Community

    Thanks to everyone who contributed infomation here, very useful. It has been surprising to learn how ambiguous some of the cell tower/antenna data is, I would have thought FCC, etc. would be more stringent about this topic.
    Just to return to the original question from "Backup Power" on Jan 22, does anyone have a feel for the AC/DC loads on Urban rooftop antennas compared to the larger cellular towers (such as 500 watts, 1 kW or)? I'm also trying to understand if rooftop installations typically have any kind of a backup power system using batteries or generators in case of grid failure? Sounds like available space and landlord issues might be a concern....
    Any thoughts would be much appreciated!
  11. Mohammed333

    Mohammed333 First Time Poster

    I joined this site after I read this post; thanks everyone for the great discussion.

    Is there any update on this regards

Share This Page