OTT

Alison Gillwald

Research ICT Africa and University of Cape Town

Parliamentary hearing on OTT services Cape Town, 26 January 2016

To identify the challenges of evidence–based policy formulation and safegaurding public interest outcomes in relation to OTT services

Content

‣ Locating OTT in new ICT system

‣ Harnessing the economic and social impacts of the Internet

‣ Regulation, investment and innovation

‣ What are OTTs?

‣ Why are operators concerned about the rise of OTTs.

‣ Net neutrality and zero rating

‣ What are the harm issues?

‣ What if the evidence, if there is any?

‣ What is to be done?

Internet as a global distribution network, stimulated by convergence between media, telecommunications and IT, facilitated the provision of content (audio visual) over converged IP networks, across multiple devices, with layers of governance at the international, regional and national level.

…To complex ICT ecosystem

Complex adaptive systems/regulation‣ ICT ecosystem characterised by exponential technological

development and increasing dependency on connectivity for positive social and economic national outcomes.

‣ Complex adaptive systems that innovative to circumvent bottlenecks often through disruptive competition

‣ Need to move from regulation of static linear value chain to adaptive, flexible regulation that does not stifle adaptiveness, and innovation.

‣ Competition regulation (static efficiency) to understanding dynamic, complementary relationship between different elements in ICT ecosystem

‣ Unintended outcomes of instrumental regulation for one objective (competition) produce negative outcomes in other (eg. Innovation)

4

Changed market conditions‣ Saturated voice markets shifting to data

‣ Introduction of low-end smart phone driving data

‣ Declining revenue from traditional services

‣ Operators face becoming ‘dumb pipes’

‣ Multiple new business models emerging from data competition to retain and attract new customers

‣ Zero-rated services, social media bundles, blended bundles, build-your-own-bundle.

‣ Multiple user strategies to access and use Internet –substituted voice and text data services, public wifi for updates, YouTube.

5

Key regulatory issues?‣ Where will market power, possibly even monopoly, be a persistent

problem? ‣ Higher likelihood in fixed than in wireless markets.‣ Higher likelihood in access than in services markets (civil engineering

infrastructure as a bottleneck) ‣ Higher likelihood in sparsely populated and/or regions with low purchasing

power‣ Significant market concentration in new economy and information markets

(e.g. search)

‣ Where can regulation improve outcomes? Enforceability?‣ Which instruments (especially in fast-changing markets)? ‣ How can regulation best cooperate with other policy makers (e.g.,

competition authorities, economic development)?

- – Bauer, J (2015) RIA Research Design

Changed market conditions‣ Saturated voice markets shifting to data‣ Introduction of low-end smart phone driving data ‣ Declining revenue from traditional services‣ Operators face becoming ‘dumb pipes’‣ Multiple new business models emerging from data

competition to retain and attract new customers‣ Zero-rated services, social media bundles, blended

bundles, build-your-own-bundle.‣ Multiple user strategies to access and use Internet –

substituted voice and text data services, public wifi for updates, U-tube.

‣ More users, more devices, more services, more demand7

Data revenue as a % of total rev

8

10%13%

16%18%

23%

29%

10% 12%15%

24% 26%

32%30%

48%

42%

0.0%

9.1%

18.2%

27.3%

36.4%

45.5%

54.5%

63.6%

72.7%

81.8%

90.9%

2010 2011 2012 2013 2014 2015

Datarevenueasa%oftotalrevenue

Vodacom MTN TelkomMobile

Note: Telkom Mobile Data for 2015 not reported

Voice vs. Data Revenue: Vodacom

9

27,42228,584 29,395 29,151

28,13525,855

4,3636,180

7,6398,882

10,974

13,538

0

2,917

5,834

8,750

11,667

14,584

17,500

20,417

23,333

26,250

29,167

32,083

35,000

2010 2011 2012 2013 2014 2015

Data revenue as a % of total revenue

Vodacom Voice Revenue Vodacom Data Revenue

Voicevs.DataRevenue- Vodacom(Rmns)and(%)

90% 87%84%

72%77%

71%

10% 13% 16% 28% 23% 29%

Voice vs. Data Revenue: MTN%

10

90% 88%

75% 76% 74%68%

10% 12%15%

24% 26%32%

0%

13%

25%

38%

50%

63%

75%

88%

100%

2010 2011 2012 2013 2014 2015

Data revenue as a % of total revenue

MTN Voice Revenue MTN Data Revenue

Note: Comprehensive data only publicly available for Vodacom

Data revenue increased 40,0% and now contributes 31,5% to total revenue. This was driven by attractive segmented data bundles and an increased uptake of digital services Q3/2015 Financials.

Blended Revenue Per User (ARPU)

11

184 183

157

128 125

113112

92

8070

87

23

6960 63

75

0

20

40

60

80

100

120

140

160

180

200

2010 2011 2012 2013 2014 2015

BlendedARPU- FinancialYears2010-2015

Vodacom MTN TelkomMobile

Note: MTN Data for 2011 not reported

OTT‣ Historically referred to the delivery of film and TV content via

the Internet, without requiring users to subscribe to a traditional cable or satellite pay-TV service.

‣ Three different revenue models in tradition broadcast/cable environment:

• SVOD -Subscription-based services such as Netflix, Showtime Frontrow (in SA case),

• AVOD - ”Free” ad-supported services such as Hulu• TVOD – transactional services such as iTunes, Vimeo On Demand and

Amazon instant video• OTT Devices: Apple TV/ Xbox

‣ Classical Competition/regulation issues:• Vertical integration/dominance: Downstream/upstream competition,

price squeeze (zero rating data -Front Row and MTN)• Symmetrical regulation between licensees and OTT - obligations/local

content 12

OTT in mobile broadband environment?‣ different evolution in mobile broadband and prepaid

environment that characterises internet access and use in Africa

‣ underlying networks not fibre/cable but mobile broadband networks that have been primarily responsible for delivering broadband, as they did voice service.

‣ in mobile pre-paid environment it generally refers to third party services, applications, platforms that are dependent for delivery on the underlying network.

‣ They are ‘free’ only in the sense that users are not required to pay an additional fee over and above their data charges (but obviously the more intensive they are, the more bandwidth users need buy).

13

14

Example of zero rated services in South Africa

15

Examples of Social media bundles

MTC Namibia

MTN Nigeria

16

Example Blended Bundles

Operators lobby for regulation

‣ A business model which requires infrastructure based service providers to constantly increase their performance (or cut their costs of doing so), while being limited by economic regulation on the revenue side, may not be sustained in the long run

‣ WICT 2012 -ETNO, Brussels-based lobby group representing large telecommunication companies in 35 European countries, proposed that ITU to designate Internet content providers as "call originators" and subject them to a "sending party network pays" rule that would allow telecommunications operators to charge them rates they believe are commensurate with the bandwidth their content consumes.

‣ United States based Internet content provider giant Google's 2011 revenue was US$37.9 billion with about US$2 billion of it believed to be revenue streams generated by the user-generated video site YouTube's 161 million users 2011, while Facebook is believed to have made about US$4 billion in 2011.

‣ Since early 2010 European telecommunication giants and Google have been battling an increasingly heated commercial war, where the telecom companies are trying to get Google to pay for the bandwidth YouTube and its other websites are using.:

17

Zero rating and net neutrality

‣ Net neutrality: every packet of data on the Internet should be treated equally when it is routed irrespective of what content is in that packet, who sends it and who is going to view it. Nothing should be throttled, or downgraded in speed, or blocked.

‣ Fear that telecom operator will have the incentive to speed up and give preferential treatment to packets of zero rated content, at the expense of content that is not zero rated.

‣ Net neutrality advocates argue Freebasics, not free and open Internet and those who can only access will get stuck in walled garden.

‣ Generally, this has been applied as a competition issues, on technical (quality of service) and price grounds. (FCC)

‣ In practical terms it can never the applied absolutely. Simply managing network congestion requires, time-sensitive data like voice, video treated differently from e-mail data.

‣ If strictly applied as in QoS to access, effect to deny users access.

18

What evidence is there of harm?

‣ a number of marketing techniques that mobile operators need to employ in competitive marketplace and have done for decades.

‣ ZR is the result of the operator’s competitive situation …operators don’t deploy zero rating because they can, but because they must. (Baumol)

‣ Exclusive and non-exclusive products different considerations.

‣ Wikipedia Zero (non exclusive) tends to be across dominant and competitors.

‣ Freebasics, primarily Facebook Zero platform that late entrant operator zero rates to distinguish itself and attract customers with the intention of migrating them into data paying customers

‣ Gateway to open internet (Facebook claims that 50% of zero rated users migrate to paying service, but big data analytics not transparent so difficult to tell. )

19

ZR – competition enhancing for MVNO market

‣ inconsistent that zero rating is rampant across Internet applications and services (e.g. advertising supported games, search, social networks, music streaming etc) but arbitrarily prohibited on mobile broadband services.

‣ campaigns against zero rating are waged as a way to pressure mobile operators to change their pricing in favor of users who consume high volume video and against those users who have never used the Internet but need an incentive to try.

‣ Inverted competition argument: Access to WhatsApp is free then it effectively harms other competitors because to access to them must be paid

20

‣ Zero Rating plans enable mobile wireless customers to download and upload online content without incurring data usage charges or having their usage counted against data usage limits.

‣ Number of studies (Layton & Calderwood 2015, Nera 2015) demonstrating that Zero Rating an economically efficient mechanism for increasing consumer welfare

‣ given unique characteristics of information technology markets, which make it beneficial to offer lower prices and other incentives to expand the size of the market, especially in developing countries where incomes, and market penetration, are low.

‣ broad-based bans or restrictions on Zero Rating plans are far more likely to harm consumer welfare than improve it(NERA)

21

Zero rating

Arising competition/regulation issues

‣ Traditional business models and conditions enjoyed by incumbents challenged by introduction of broadband technologies, networks and devices (smart phones) that have enabled low and no-cost voice services (in particular)

‣ Hit traditional mobile voice and SMS business models that have driven the revenues of mobile models for two decades.

‣ Like fixed line incumbents, some mobile incumbents have resisted shift to new data models but other have embraced these inevitable developments to positive effects.

22

Countries offering of new pricing models

23

Zero rated services Social media bundles Blended bundles

19

13

24

Changing business models‣ SA like the rest of Africa, although voice still makes up for a significant

portion of revenues, data revenues are growing much faster.

‣ Operators still feeling the pinch from declining voice revenues, as data revenues still not exceeding voice, which are being substituted with OTT ‘free’ voice and text services.

‣ Underperforming operators (often incumbents) unable to develop their own content or enter into complementary relationships with the platforms that drive internet use, have been lobbying through industry associations, the ITU for OTT services to be forced to pay operators (on the grounds that they are benefits from operator investment risk).

‣ Operators that have embraced new data driven business models and are seeking to optistie net works appear to be moving toward global trends, first seen in fixed market, of flat rate models

24

Pricing‣ Although data prices, plummeted with increasingly

competitive products and services, as new models emerged an operators sought to attract new (data) customers and retain customers, prices in SA have now stabilised.

‣ These remain, and still remain far too high for the average pre-paid user to be downloading U Tube or doing software updates off their regular data packages that would drive the take off data services, as they have in jurisdictions were the prices are lower or peoples incomes are on average higher

25

0

45

90

135

180

225

Q1 2014 Q2 2014 Q3 2014 Q4 2014 Q1 2015 Q2 2015 Q3 2015 Q4 2015

Cell C MTN South Africa Telkom Mobile

SA 1GB price between 2014-2015

345556778888999

1010101010

111212121314

1517171718

2021

2424

2526

3033

3535

4775

0 20 40 60 80

EgyptMozambique

KenyaTunisiaGhana

Cape VerdeGuinea

TanzaniaMadagascar

Burkina FasoNiger

SenegalMalawi

MauritiusSao Tome and …

EthiopiaAlgeriaUganda

BeninMoroccoRwanda

South AfricaLiberia

NamibiaMali

CameroonZambia

ChadCongo Brazzaville

TogoNigeria

D.R CongoAngolaGabonSudan

LesothoMauritaniaBotswana

Cote d'IvoireZimbabweSwaziland

SeychellesLibya

Q3 2015 USD Price 1GB basket by dominant operator

Bundled Value for Money Index Q3 country comparison

88.40

68.52

63.10

61.81

57.42

55.70

43.65

42.89

36.12

33.53

23.39

13.92

11.39

11.21

10.34

9.59

7.11

4.12

4.02

3.75

3.46

3.22

2.92

0.03

Uganda

Tanzania

Namibia

Kenya

Camero…

Rwanda

Senegal

Sudan

Mauritius

Morocco

Madag…

South …

Angola

D.R …

Algeria

Egypt

Maurita…

Gabon

Nigeria

Chad

Cote …

Botswana

Ethiopia

Ghana

Q3 BVMI comparison by all operators

88.4069.75

68.5263.10

61.8157.42

55.7053.49

43.6542.8942.19

38.5336.12

33.5323.39

20.3516.2716.00

15.0313.92

11.2110.8010.349.809.598.818.057.497.116.156.035.675.244.124.023.753.463.082.922.491.960.030.020.02

Smart UgandaUganda Telecom

ZanzTelMTC

Airtel Yu KenyaMTN Cameroon

Tigo RwandaVodafone Uganda

Tigo SenegalSudani

Smart TanzaniaOrange Kenya

Orange MauritiusInwi

Airtel MadagascarOrange Senegal

TelmaTigo Tanzania

Orange MadagascarMTN South Africa

Tigo DRCCell C

MobilisMTML (Now Chili)

Etisalat EgyptAirtel TanzaniaAfricell Uganda

Orange CameroonMauritel

Chinguitel MauritaniaOrange Botswana

MTN UgandaDjezzy GSM

Libertis (Gabon …MTN Nigeria

TigoMTN Cote d'IvoireOrange DR Congo

EthioTelecomTN Mobile

Movicel AngolaAirtel GhanaTigo Ghana

Vodafone Ghana

Market Share - Q2 2015

30

Cell C*24%

Telkom Mobile

2%

MTN35%

Vodacom

39%

Note: Excludes MVNO operators; Cell C’s 2014 end of year market share figures are used.

Operator Revenue

31

44,32446,392

48,427 48,234 48,31647,032

35,82238,597

41,35039,707 38,922

44,740

811,200 1,359 2,347

0

5,500

11,000

16,500

22,000

27,500

33,000

38,500

44,000

49,500

55,000

2010 2011 2012 2013 2014 2015

Revenue- FinancialYears2010-2015(Mn’s)

Vodacom MTN TelkomMobile

Note: Telkom Mobile revenue for 2015 not reported

EBITDA Margin

32

37% 37%37%

38%

37%37%

44%

35% 35%

34%

32%

33%

30.0%

31.4%

32.7%

34.1%

35.5%

36.8%

38.2%

39.5%

40.9%

42.3%

43.6%

2010 2011 2012 2013 2014 2015

EBITDAMargin- FinancialYears2010-2015

Vodacom MTN

Investment - Capital Expenditure

33

4,573

5,100

6,976 6,967 6,858

8,646

3,908 4,105

6,416

5,835 5,676

1,521 1,562 1,548 1,368

481

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

10,000

2010 2011 2012 2013 2014 2015

CapEx- FinancialYears2010-2015(Rmn's)

Vodacom MTN TelkomMobile

2015 MTN increased

investment in LTE and 3G colocated

sites

Creating conditions that facilitate high capital investment required for deployment of next generation networks to support innovation

Reform 0.2 - From static to dynamic regulation

‣ Static regulation transition from monopoly to open market (assumes core network infrastructure in place)

‣ Structural and conduct regulation at wholesale level (interconnection, unbundling, price regulation).

‣ Digitisation and convergence allows for multiple entrants, migration of services and content across platforms

‣ High levels of substitution - fixed, wireless, instant messaging, social networking

‣ New complementarities - content & apps drive data

Underperformance

‣ Dynamic systems such as the advanced ICT system may be “stuck” in an underperformance state (“attractor”)

‣ – Insufficient investment and innovation – Inefficiently high or low prices

‣ Multiple causes

• – Regulatory—regulation outside the workable performance zone (too strict, too lax)

• – Political—veto players capable of blocking change

• – Institutional—societal constraints and inertia

‣ Can be overcome but only after considerable costs

35

Creating enabling environment for next generation networks and services with positive consumer welfare outcomes ‣ Allocate high demand spectrum so operators can

efficiently provide services

‣ Service neutral licenses and services so operators and service providers can develop flexible complementary relationship

‣ Confirm consumer welfare and particularly pro-poor outcomes

‣ Ensure that rights and cybersecurity framework in place to created trusted environment

36

Need for new foundations

- Bauer (2015) Research ICT Africa Research Design Workshop

References:

38

Ø Bauer, J, M and Bohlin, E. (2008) From static to dynamic regulation: Recent development in US telecommunications Policy. Intereconomics, January 2008, Volume 43, Issue 1, pp 38-50

Ø Eisenach, J. A. (2015). The economics of zero ration. NERA economic consulting. http://www.nera.com/content/dam/nera/publications/2015/EconomicsofZeroRating.pdf

Ø Futter, A. and Gillwald, A. (2015). Zero-rated internet services: What is to be done? Policy paper 1, 2015: Broadband 4 Africa.

Ø Layton, R. and Calderwood, S. (2015). Zero Rating-do hard rules protect or harms consumers and competition? Evidence from Chile, Netherlands and Slovenia.

Ø Stalman, E and Adams R.S. (2015). Zero rating-A framework for assessing benefits and harms. Centre for Democracy and technology.