. 112
( 208 .)


same channel in each of the ¬ve regions and thus have the coverage of a nationwide li-
cense without actually bidding on one. MTAs would have a total of 561 licenses available
(51 MTAs — 11 licenses) and BTAs, 2,958 licenses (493 BTAs — 6 licenses).
The number of providers in a market was technically limited by the number of oper-
ating licenses issued for that market. But since license holders could sell portions of their
spectrum to resellers, the number of providers a market could physically accommodate
and the spectrum™s capacity de¬ned the true number of ¬rms operating in a market. Li-
censes could also be bought by a group of companies, so that multiple providers could
operate on the same license in the same area.
The minimum provider investment necessary to start a paging company varied by the
technology, protocols, and licenses used. In 1996 the minimum outlay for a simple one-
way nationwide network was approximately $200 million dollars.10 A nationwide
license would cost an additional $25 million dollars. A regional network of the same

9. MHz (megahertz) is a unit of frequency comprised of a million hertz. 1 hertz = 1 cycle per second. 1 MHz = 1,000
kHz (kilohertz) .
10. This scenario assumed the network used FLEX protocol and 1,000 radio transmitters to broadcast messages.
Equity Security Analysis

13-25 Part 2 Business Analysis and Valuation Tools

speci¬cations, with licenses, would cost approximately $40 million. Larger, better cap-
italized providers, therefore, had advantages.
While the six largest providers served over 60 percent of the market, the remainder
of the industry was highly fragmented. Competition was intense at all levels, and since
service was hard to differentiate, it rested on the linked elements of cost, data delivery,
Arch Communications Group

and price. Low costs, in a high ¬xed cost industry, were achieved by “loading” infra-
structure, that is, piling as many subscribers as possible onto an existing network. Allo-
cating costs over a large subscriber base lowered per-unit costs and could be re¬‚ected in
pricing. Loading, however, swelled the number of messages that had to travel the net-
work, increased transmission time, and delayed messages to the end user. In paging,
rapid message delivery was critical. Low prices drew subscribers, but long term loyalty
was a function of the ability to deliver data immediately. Providers could manage these
components by upgrading to faster protocols or adding spectrum.
In the pursuit of scale and spectrum, a pan-industry rush of mergers took place in the
1980s and 1990s. In one decade, providers consolidated from 1,000 to 500 with 8“10
large regional or national companies. By 1996, however, merging pains, management
mismatches, concern over the merger wave ending, delays in deployment of NPCS net-
works, and fears that PCS would cannibalize paging and that no room was left for inter-
nal growth, caused sector morale to go down.

Arch™s Largest Competitors, 1990s

PAGENET. Founded in 1981, PageNet was the largest and fastest growing provider in
the U.S. with service in all 50 states, the U.S. Virgin Islands, and Puerto Rico. PageNet
was considered a trend setter (¬rst to add a million units via internal growth), the low
cost leader, and the most successful company in the industry (160 percent the size of its
closest competitor). PageNet had 7.8 million subscribers (20 percent of the industry
base), adding 2.3 million subscribers in 1995 (350,000 by acquisition).
PageNet™s growth strategy focused primarily on addition through internal growth.
Due to this strategy, PageNet did not have the acquisition problem of consolidating net-
works of different frequencies or back of¬ce systems. PageNet™s strategy of aggressive
pricing policies, reliable service, and emphasis on direct sales (largest industry sales
force with 1,000 people and 6,000 resellers) was executed by decentralized manage-
ment. PageNet owned the most spectrum of any provider. The company built a 24-hour
support/distribution center and a National Accounts Division to provide one contact
point to large, national clients and to forge and manage such alliances.
PageNet regularly entered partnership and distribution agreements to expand its cli-
ent base. Sprint, MCI, and GTE were PageNet clients who resold services to their own
clients. Partners could market services under their own or PageNet™s brand and could
customize agreements. With Sprint and MCI, PageNet handled shipping, customer ser-
vice, and pager leasing. Sprint and MCI oversaw advertising, billing, and marketing.
Subscribers were unaware they were dealing with PageNet at all. GTE was a typical re-
528 Equity Security Analysis

Equity Security Analysis

seller, owning its pagers and handling its own customer service and billing. Reseller
churn impacted PageNet™s subscriber numbers but not its revenues.
PageNet™s future projects included the ¬rst commercial wireless pocket answering
machine, dubbed VoiceNow, which was in its ¬nal testing stages. PageNet was also ex-
panding its business overseas through its recently formed international division.

Arch Communications Group
MOBILECOMM. MobileComm was the second largest provider of paging services
(4.4 million subscribers) with local, regional, and national service in the 50 states, Can-
ada, and the Caribbean. MobileComm served 97 of the top 100 largest metropolitan
markets. In 1996 MobileMedia acquired (and renamed itself) MobileComm for $930
million (the largest industry acquisition), netting 1.8 million subscribers and consolidat-
ing revenues of $323 million.
MobileComm™s acquisitions aimed at establishing national presence in one-way and
two-way networks (it had two nationwide PCS licenses), growing sales distribution ca-
pability to retail channels, and adding spectrum. MobileComm™s internal growth plan
emphasized high sales productivity and strategic alliances. After the MobileMedia
merger, MobileComm began centralizing back of¬ce functions (all credit and collection
tasks in one place) and building two service centers (in Texas and Maryland) that pro-
vided 24-hour customer service and billing support.
By mid-1996 MobileComm was the victim of high churn: 3.8 percent (industry aver-
age: 2“3 percent). This was due to network congestion11 that delayed message delivery
during peak hours in major markets and a rise in resellers on its networks. MobileComm
was also still trying to cut duplicate back of¬ce/support expenses from the merger. In late
1996 MobileComm changed its entire upper management. The restructuring slowed the
already troubled integration and Texas center project. Standard & Poor™s downgraded its
rating on MobileComm™s $460 million debt. MobileComm was bound by its creditors
to raise $100 million in equity capital by year™s end.

OTHER PROVIDERS. The fourth largest provider, Metrocall, gained scale by a series
of fast acquisitions, but would expend considerable resources to mold the various parts
into one entity, while also attempting to integrate new management. American Paging,
the seventh largest player, was undergoing a large restructuring which included a man-
agement turnover. The restructuring was blamed for the drop in subscriber additions and
weak operating performance the ¬rm experienced.
The ¬fth largest provider, ProNet, had followed a fast grow strategy that focused on
dense urban markets. Despite acquisitions, doubling subscribers in one year, creating re-
seller programs to drive long-term internal growth, and one of the lowest cost structures,
ProNet announced mid-1996 that, due to price concessions to resellers, it would be un-
able to grow cash ¬‚ow for several quarters. Standard & Poor™s lowered ProNet™s credit
rating. Consequently, ProNet saw its stock dumped.
11. MobileComm had the industry™s largest alphanumeric subscriber base (14 percent of its subscribers used alphanu-
merics). High revenue alphanumeric paging, however, uses four to ¬ve times the capacity of numeric paging,
congesting and over-traf¬cking the paging networks they occupy.
Equity Security Analysis

13-27 Part 2 Business Analysis and Valuation Tools

Arch offered local, regional, and nationwide service, every pager type, and also special
services such as voice mail. In 1995 87 percent of Arch™s in-service pagers were digital
display, 7 percent were alphanumeric, 2 percent tone only, and 2 percent tone/voice.
Arch owned, leased, and provided service to 45 percent of its in-service pagers and pro-
Arch Communications Group

vided service only to the remainder (30 percent of which were subscriber owned pagers,
25 percent reseller owned).
Arch owned and was developing two nationwide channels (acquired primarily to ex-
pand its regional services), but had followed a unique service strategy of offering nation-
wide paging through a network of af¬liates. When a subscriber using Arch™s nationwide
paging left one af¬liate™s market and entered another, he/she called a toll-free number
that would prompt the user™s pager to become active and receive messages (on the af¬l-
iate™s channel) in the new market.
Arch acquired 60 percent of its subscribers through direct distribution (direct sales
and ¬rm owned stores). Direct distribution was a more expensive channel by which to
add subscribers, but gave Arch an ARPU higher than the industry average. The indirect
channel (comprised of low cost, low ARPU resellers and high ARPU retailers) contrib-
uted the remaining 40 percent of subscribers.
Over the past few years, Arch had shown a decline in monthly ARPU. This was be-
cause the number of subscriber or reseller owned pagers for which Arch received no re-
curring rental fee had increased more than 25 percent over the past few years. Secondly,
over the same period, the percentage of new pagers in service added through indirect
channels (mostly resellers who purchased bulk airtime at discount) had increased. Fi-
nally, the decline in paging service retail prices, resulting from pressure on pricing due
to increased competition and growth, also drove revenues down. Arch™s revenue decline
mirrored an industry-wide decline. While some observers were alarmed by the sustained
declines, others measured operating performance by EBITDA rather than revenues and
paid little attention to the drop. Revenue declines ignored both the differences in oper-
ating margins from different distribution channels and the fact that paging was a volume
driven, ¬xed cost business and that spreading those costs over a large base had a positive
impact on margins.
Arch generated most of its revenues by charging subscribers ¬xed periodic fees. As
long as subscribers remained in service, the recurring payments constituted an income
stream free of additional selling expenses. Arch™s net losses were mostly due to the in-
terest on debt incurred to ¬nance growth, and the large depreciation and amortization
charges related to assets. Arch required considerable funds to service debt, ¬nance ac-
quisitions, fund expansion and upkeep of existing operations, and cover pager and pag-
ing system expenditures. The company™s capital expenditures had increased from $10.5
million in 1992 to over $60.6 million in 1995 and were expected to reach the $100 mil-
lion mark in 1996. These expenditures were supported by cash from operations, equity
issues, and debt. At the end of 1995 Arch had assets totaling $785.3 million. The com-
pany expected to generate positive cash ¬‚ow by 1998.
530 Equity Security Analysis

Equity Security Analysis

During 1995 Arch had several important accomplishments: it made six acquisitions
of which the $540 million USA Mobile (second largest industry acquisition) and
Westlink (substantially adding to Arch™s nationwide presence) helped push it from the
industry™s tenth to third largest company; added 1.1 million subscribers (acquisition) and
366,000 subscribers (internal growth”tripling quarterly internal growth); expanded ser-
vice from 13 to 40 states; grew EBITDA from $18 million in 1994 to $47.2 million; grew

Arch Communications Group
total revenues 114.2 percent to $141.8 million (and grew net revenue by 124.7 percent);
and raised over $300 million in new debt and equity capital. These results crowned sev-
enteen consecutive quarters of net revenue and cash ¬‚ow increases in an industry where
such measures were expected to remain weak for the foreseeable future. Table A shows
selected Arch ¬nancial highlights.

Table A Arch Financial Highlights
Year Ended Year Ended Year Ended
12/31/95 12/31/94 12/31/93
Net revenues $141,809,000 $63,116,000 $41,277,000
Earnings before interest, taxes,
depreciation and amortization (EBITDA) 47,186,000 17,969,000 11,315,000
Net income (loss) (36,602,000) (6,462,000) (5,725,000)
Per share data:
Weighted average shares 13,498,000 7,183,000 7,125,000
EBITDA $ 3.50 $ 2.50 $ 1.59
Net Income (loss) $(2.72) $(0.90) $(0.80)
Ending subscriber units in service 2,006,000 538,000 254,000
Source: Arch 1995 Annual Report

Arch followed a strategy that emphasized low prices, proven technologies, and reliable
delivery of service. Arch™s management team, headed by CEO C. Edward Baker Jr., was
considered to have the longest successful management track record in the public paging
industry. Arch™s decentralized management structure allowed it to control costs, smooth-
ly consolidate acquisitions, and respond to subscriber needs quickly.
As one of the industry™s lowest cost providers, Arch was able to price competitively,
sustaining its cost structure by consolidating operating functions, using fast transmission
systems, and spreading costs by taking advantage of economies of scale arising from
pursuing large scale.
Arch offered its subscribers no frills paging services based on standard and tested
technologies that could be depended upon to deliver messages reliably and quickly with
none of the potential hiccups that services based on new or experimental technologies
might present. Arch did, however, keep up with emerging paging technologies by invest-
ing in a consortium that was developing advanced paging services. Scott Hoyt, Arch
Equity Security Analysis

13-29 Part 2 Business Analysis and Valuation Tools

Marketing V.P., explained, “This industry will eat you alive if you™re wrong as a technol-
ogy innovator. We prefer to take advantage of other people™s mistakes.” Hoyt added that
Arch considered itself a “fast follower.”12


. 112
( 208 .)