Bangladesh Overview - Banking

ANALYSIS
Country overview: Bangladesh

August 2014
GSMA Intelligence
Mobile for
Development Impact
gsmaintelligence.com/m4d [email protected] @GSMAi
GSMA Intelligence Country overview: Bangladesh
Contents
Executive summary............................................................................................................................. 3
National context................................................................................................................................... 6
Mobile for Development opportunity............................................................................................. 8
Understanding the past to envisage the future................................................................................ 8
Disaster response....................................................................................................................................... 10
Agriculture......................................................................................................................................................15
Mobile money................................................................................................................................................ 17
Education and employment...................................................................................................................20
Investment in innovation........................................................................................................................... 21
The mobile market............................................................................................................................ 24
Low disposable income, but relatively high mobile penetration.............................................24
The mobile internet is on 2G..................................................................................................................29
Regulation......................................................................................................................................................31
Appendix............................................................................................................................................. 33
Benchmarking Bangladesh: methodology........................................................................................33
Country environment................................................................................................................................36
Glossary............................................................................................................................................... 39
About
Mobile for
Development Impact
GSMA supports the digital empowerment of people in emerging markets through its Mobile
for Development Impact programme, used to inform investment and design decisions for
mobile services. Our work is freely accessible through support from Omidyar Network and
in partnership with The MasterCard Foundation at gsmaintelligence.com/m4d
Executive summary
Bangladesh is, in many ways, a country ahead of its time in terms of mobile access. Despite
being ranked as a low income country, over 50%1 of the population subscribes to mobile
services and it has outpaced all its peers in terms of network coverage. The combination
of limited disposable income and more mature mobile usage means the customers of
Bangladesh are more discerning. Mobile network operators are looking to innovative valueadded services (VAS) to remain competitive and respond to the slowing growth in core
mobile services. Services that add value to and improve on the livelihoods of the consumer
are more likely to support these goals.
1. Bangladesh is one of the most populous and most densely populated countries in the
world. Mobile penetration levels are relatively high, even in rural areas (something not
seen in most other emerging markets). However, consumer spending levels are among
the lowest in the world. Given the reality that there will be less reliance on growth from
new subscribers over the next 45 years, mobile operators will need to develop new
revenue streams beyond core mobile connectivity services that support basic human
needs around agriculture, education and employment provide a key opportunity.
Bangladesh has a myriad of social challenges, including low literacy rates, child malnutrition,
poor access to electricity and a significant urban-rural divide. However, mobile penetration
is higher than would be expected given its low-income status. Subscriber penetration
reached 40% at the beginning of 2014 and it is expected to grow to 50% by 2020.
Bangladesh has a unique pedigree given the transformative impact of the Grameenphone
Village Phone programme a pioneering initiative started in the 1990s to empower rural
women through mobile services and subsequent rapid roll-out of network infrastructure.
Between 1997 and 2002 mobile coverage swiftly spread to the majority of the country,
many years ahead of other emerging markets. While 3G auctions have only recently been
conducted, the provision of basic 2G coverage is widespread. However, the prepaid nature
of the market and low incomes of new subscribers means that ARPU levels are low (among
the lowest in the world). This underlines the need to develop new revenue streams, we
believe the key opportunities are in mobile data and VAS that have both a commercial and
social impact.
2. Bangladesh is predominantly a prepaid and 2G market, as 3G has only recently

been deployed following delays in the auction process (3G today makes up just 2% of
connections). However, mobile internet penetration is over 20%, mostly from 2G feature
phones. In other words, there is a latent desire for internet access that is growing. The only
question is how fast this occurs, which will depend on affordability and tariff structures.
Bangladesh is a prepaid and 2G market; 97% of connections are prepaid and 98% are 2G.
In the first half of 2013, smartphones accounted for only 6% of total handset shipments (a
local brand, Symphony Mobile, is the largest vendor with a 37% handset market share of
which 92% are feature phones and 8% smartphones). The low 3G penetration is due to the
late release of 3G spectrum in 2013. However, mobile internet has reached 20% penetration
1
Excluding youth under 16

3
by delivering mobile data services using 2G networks. The combination of 3G spectrum

now in play, solid existing site coverage, and consumer appetite for internet use in urban
and rural areas means this trend is accelerating (we expect 3G connections to surpass 2G
by 2020). The only question is pace: this could very well occur faster than expected, if the
number one challenge of affordable access (both airtime and handsets) can be overcome.
3. The use of mobile in driving socio-economic improvement is on the rise. From its
pioneering roots in microfinance 30 years ago, the country now has a relatively solid
mobile consumer base, much of which are underserved in core life needs. Mobile operators
are demonstrating the potential for social VAS, and while we expect this to continue to
grow, it will take time. There is still an opportunity for public and private investment in
providing seed capital for the innovation that is not yet market-led, with a key role for
government in facilitating this process.
While it is true that the majority of unconnected individuals reside in rural areas, this is due
to the fact that the majority of population are in rural areas. The opportunity to address
the underserved in urban settings is far from insignificant we estimate six million city
dwellers are yet to own a mobile. There are a host of sector opportunities that result
from the widespread access to mobile phones yet the lack of access to basic services
including responding to natural disasters, driving gains in agricultural productivity,
improving educational and employment outcomes, or increasing financial inclusion.
So, how should the various players in the Mobile for Development ecosystem respond to
this opportunity?
Mobile operators are increasingly involved in leading commercial VAS that carry social
impact and, although there are challenges, we expect this to continue over the next two to
three years. This is, on its own, a positive story given the implicit commercial value of such
services attributed through operator investment. However, in order to drive increases in
ARPU, new approaches to marketing and pricing these services would help, particularly as
personal incomes rise. For example, use of flexible credit scoring techniques; experimenting
to include VAS as part of bundle packages alongside traditional access services (voice,
SMS and data); or using VAS as a more defined tool for customer retention (e.g. brand
loyalty).
Donors and NGOs have a growing presence providing seed capital for services that have
not yet scaled and are not market-led (BRAC is a notable example). However, we believe
there is still an important untapped opportunity for the venture capital (VC) and impact
investor community. Dhaka is not yet an innovation hub on the scale of regional peers
such as Bangalore, Colombo or (further afield) Nairobi. However, this is not due to a lack
of entrepreneurs. There are clear challenges faced by private investors seeking to place
capital in Bangladesh ICT, including the market operating environment and the lower
likelihood of exit opportunities. To accelerate the process of these barriers being overcome,
we believe it is important investors play a direct role by fostering an enabling environment
of education and mentorship in addition to capital. This is not a short term game, but the
long term rewards for early entrants are attractive.
Government also has an important role to play. Bangladesh has an opportune landscape
for Mobile for Development services high mobile penetration and a number of social
challenges that can demonstrably be helped through mobile technology. The government
set out laudable and ambitious goals around expanding digital empowerment to the mid
and low income underserved population as part of its flagship Digital Bangladesh vision.
In order to execute this successfully, public policy priorities should be focused on ensuring
investment security (particularly for foreign capital), and liquidity in public markets, both
of which would help would-be investors (and indeed mobile operators) with more reliable
decision making.
National context
Located in South Asia, Bangladesh is the worlds eighth-most populous country and
among the most densely populated countries with 1,188 people per square kilometre. The
population is very young, with just over half under 25 years of age, and there is an equal
male-female split. Bangladesh was part of what was formerly known as East Pakistan
and became independent in 1971, when the two parts of Pakistan split after a bitter war
which drew in neighbouring India. After spending 15 years under military rule and restoring
democracy in 1991, Bangladesh has had a period of relative calm and economic progress,
however it continues to struggle with natural calamities owing to its low-lying geography
and vulnerability to floods and cyclones together with other socio-economic problems.
Since independence, Bangladesh has made significant improvements in human and social
development indicators, including gender equality, universal primary education, food
production, and population control. It has also made improvements in healthcare: life
expectancy is higher than any other country in the region (except Nepal), while infant,
under-five and maternal mortality rates are better than other countries in the region. Yet,
despite improved survival rates, nearly half of children have chronic malnutrition. One
of the biggest problems in Bangladesh is the deep and widespread poverty levels since
approximately 50 million people live in poverty, on less than $2 per day. This it is primarily a
rural phenomenon and 85% of the countrys poor live in rural areas. The urban-rural divide
is evident in many socio-economic aspects (see Figure 1).
100%
Urban
Rural
88%
90%
80%
71%
70%
60%
55%
57%
50%
46%
43%
40%
30%
38%
35%
29%
21%
20%
10%
0%
Population
share
Poverty2
Access to
electricity
Access to
sanitation
Accounts at a
formal financial
institution
Figure 1: Urban vs rural gap, 2010

Source: World Bank
Poverty headcount ratio at urban and rural poverty lines

6
Bangladeshs GDP has been growing at an average of 6% each year for the past ten years.
However, it is still a low income economy with a GDP per capita of just over $800. But there
is great potential for growth Bangladesh is committed to becoming a middle-income
country by 2021, its 50th year of independence, and economist Jim ONeil believes that
Bangladesh is part of the next eleven, a set of eleven countries with a high potential of
becoming the worlds largest economies in the 21st century, along with the BRIC countries.
Bangladesh is one of the worlds leading exporters of textiles and garments, as well as
fish, seafood and jute. The majority of employment is in agriculture, comprising nearly
50% of the workforce and contributing 17% of the countrys GDP. The country has
diversified its economy through a growing industrial sector which contributes 29% of
GDP; in 2005 more than three-quarters of Bangladeshs export earnings came from the
garment industry. Another significant contributor to the development of the economy has
been the propagation of microcredit by the Grameen Bank (see Mobile for Development
opportunity section).
Mobile for Development opportunity

Understanding the past to envisage the future
To understand the modern opportunity in Bangladesh of mobile in driving socio-economic
improvement, it is helpful to first recast the discussion back 30 years to its roots in
microfinance. In 1976, Professor Muhammad Yunus launched a research project to examine
the possibility of designing a credit delivery system to provide banking services for the
rural underserved and named this institution Grameen Bank (meaning rural in Bangla).
The objectives of the project were focused on using financial access to empower isolated,
predominantly low income communities and, in turn, catalyse a micro entrepreneurship
market through self-employment. The project was extended to different districts in
Bangladesh in 1979, and finally in 1983 the Grameen Bank Project was transformed into an
independent bank (which still exists today as a member-led cooperative).
Jumping forwards and following in Yunuss footsteps, in 1994 Iqbal Quadir, a US-based
native Bangladeshi entrepreneur, approached Grameen Bank to use its micro credit
mechanism to establish a nationwide telecommunication network. In parallel with the
launch of commercial mobile services for Grameenphone in 1997, the Grameenphone
Village Phone Program was founded to target micro credit customers (who were mostly
women) in a symbiotic micro economy entrepreneurs were loaned capital for mobile
phones, which they could rent out to others, generating an income flow (and the ability to
repay the cost of the loan).
The significance of this is that, despite it being a low income country, Bangladesh
is now a market in many ways ahead of its time. While we believe that unique mobile
subscriber penetration is around 42%, a third of the Bangladesh population are under 16.
Adjusting for this (few will own mobiles outright, with most accessing through sharing
arrangements), ownership among adults is actually above 50% (see Figure 2). Here, a
more interesting perspective of the urban-rural divide comes into view, from which we
observe three important points. In rural areas, penetration is around 50%, with this enabled
from widespread coverage having been established much earlier than in neighbouring
countries. Second, while it is true the majority of the unconnected are in rural areas, there
is a not insignificant untapped audience of 6 million individuals in urban centres who do
not own a mobile there is no lack of network coverage, but access remains unaffordable.
Finally, and perhaps most important, the relatively maturing position of Bangladesh means
that operators have a shorter time period to rely purely on new subscribers for growth,
with the impetus for new sources of revenue a nearer term reality than many comparable
emerging markets.
Million
120
Total
Rural
Urban
100
80
60
27
6
40
20
38
36
0
Adult
population
Total
Split
Total
Connected
(mobile subscribers)
Split
Unconnected
(non-mobile subscribers)
Figure 2: The opportunity is not just in rural areas

Source: GSMA Intelligence, IFPRI
Indeed, operators have a large footprint in the value added services space, with Robi and
Bangalink most active. Most of this is concentrated in entertainment and social media, with
a limited, although growing, presence in commercial services targeting core life needs in
banking, agriculture and education amongst others (see Figure 3). This is also reflected in
our own tracker, with Bangladesh having a relatively low number of active M4D services
(53, or roughly one per million mobile subscribers) compared to its benchmarked peers.
Entertainment
General entertainment
Teletalk
Robi
Social media
Grameenphone
Citycell
Ringtones
Banglalink
Airtel
Other
mAgri
M4D
Education
Employment
Money
Health
0
10
15
20
25
30
35
Figure 3: Commercial VAS offered by operators in Bangladesh

Source: Operator websites
Note: general entertainment includes news, religion, games, music, etc. Other includes services to recharge
prepaid, receiver pays for calls/ SMS, missed called alerts, etc.
9
However, these products and services are increasingly being led by operators. Market forces
are evidently at work operators have limited budgets both in investment and corporate
finance, and allocating these to M4D VAS is indicative of the perceived return opportunity.
We believe operators will continue to grow their position as the dominant players in M4D
given weakening subscriber growth, declining ARPU and a drive for revenue diversification
coupled with consumer demand for services adding value to livelihood, and rising mobile
internet penetration. However, this takes time. There is an opportunity for seed capital to
bolster innovative services that have not yet scaled. NGOs are also increasingly playing
in the space, accounting for around 25% of services (see Innovation and Investment
section), considerably higher than in other emerging countries (Nigeria, as one example,
NGOs account for less than 10%). This leaves a still largely unrealised opportunity for other
investors donors, VCs and impact investors to enter, particularly those based in the
country or region.
For its part, the Bangladesh government has been heavily involved in promoting ICT,
fronted by the flagship initiative (Digital Bangladesh by 2021) as part of the 2008 election
manifesto. The vision proposes to mainstream ICTs as a tool for digital empowerment
through human resource development, increasing mobile penetration, digitising government
services to increase access, and integrating ICT in business. It has complemented this with
the Access to Information (A2I) programme from 2007, with the goal of leveraging ICT in
public service delivery and promoting the use of ICT in social sectors such as education,
health and agriculture. This project was supported by UNDP and USAID. The programme
first focused on increasing access to service, and then on simplifying service delivery.
The programme has been successful in decreasing time, cost and barriers in accessing
government services; in just a few years, the number of underserved people benefitting
from e-services went from zero to 4.5 million every month. Access centres now exist in
4,545 rural local government institutions and 727 locations in urban municipalities. Moving
forwards, we believe the key challenges for government are in attracting investment to
ICT, especially by creating a transparent regulatory environment for investors to operate
within.
We will discuss the investment and innovation market, but first we will highlight opportunities
for commercial engagement in some of the key sub-sectors within the M4D space.
Disaster response
Bangladesh is one of the most environmentally vulnerable countries in the world, ranking
in the top 15 for number of natural disasters per year. The most common natural disasters
in the world are hydrological (floods) and meteorological (storms). Bangladesh is exposed
to a roughly even split of droughts, floods and cyclones amounting to an average of 8
events per year, concentrated in the annual monsoon season (see Figure 4). On average
5% of the population (or 7.5 million people) in Bangladesh are affected by a natural disaster
each year, among the highest in the world (for comparison, Somalia and the Philippines
are around 10%, with the US at 1%). Between 2008 and 2012 nearly three million people
were displaced by natural disasters related to weather and geophysical hazards3. The
government now estimates that 2030 million people could be displaced by the effects of
3
These do not include droughts, nuclear or industrial accidents and epidemics

10
climate change by 2050.

During natural disasters it is very difficult to get accurate and timely information on
population movements, which is crucial for ensuring that interventions are more efficient
and provided to the right places. Today, given the widespread availability and use of mobile,
this technology can be used to monitor population movements and contribute to this
humanitarian data post-disaster. An example of an organisation doing this is Flowminder
(for more information see case study).
30
Climatological
Hydrological
Meterological
Geophysical
25
20
15
10
Myanmar
Sri Lanka
Somalia
Australia
Haiti
Brazil
Pakistan
Russia
Vietnam
Bangladesh
Mexico
Japan
Turkey
Colombia
Afghanisatan
India
Indonesia
Philippines
United States
China
Figure 4: Average number of natural disasters by disaster type per year (10 year average), 200313
Source: GSMA Intelligence, EM-DAT
Note: climatological includes drought, extreme temperature and wildlife; geophysical includes earthquakes,
mass movement dry and volcanos; meteorological represents storms and
hydrological includes flooding and mass movement wet
The opportunity in mobile is multi-pronged, underpinned by the relatively high level of

mobile access even in exposed areas. First, the basic provision of 2G connectivity for calls
and SMS leading up to and in the immediate aftermath of a disaster. Second, the use of
call data records (CDRs) means that individual migratory patterns can be tracked over
the same period, providing evidence to civilian authorities of, for example, urban overdensification. Finally, mobile money platforms allow remittances to be sent to displaced
family members in the absence of physical connections or modes of transaction. For the
mobile operators, this is not a revenue opportunity, but one of solidifying customer trust
and loyalty. Government support for disaster response services is growing, and there are
a number of organisations active in the space. The key challenge is in forming effective
partnerships with operators to take advantage of their network scale.
11
Special feature: In conversation with Linus Bengtsson

Co-founder and Executive Director, Flowminder
Can you tell us what Flowminder is and what you do?

Flowminder is an NGO based in Stockholm focussing on research into disaster response,
infectious diseases and socio-economic analysis. All of the team members are academics
with a faculty position at a university (e.g. Harvard University, Karolinska Institute and
Stockholm School of Economics). In addition, we have a number of collaborators from
other institutions. As academics, we always publish our work so all the results and the
methodology are publicly available for other researchers.
We have pioneered a number of applications using anonymous call data records (CDRs)
from mobile operators to solve public health issues. In 2009 Andy Tatem at Flowminder
launched a project to use operator data in Tanzania to understand how malaria was
imported into Zanzibar from the mainland, thus improving malaria elimination planning
on the island. We initiated the first work on using mobile operator data in disaster
response with Digicel Haiti to follow population displacement and mobility in Haiti after
the earthquake in 2010 as well as during the cholera outbreak the same year.
Can you talk about the project you are currently working on in Bangladesh?
We are undertaking a pilot study on the impacts of the cyclonic storm Mahasen, which
struck Bangladesh in May 2013. There is a lot of interest from the different ministries on
how to take this forward operationally. During pilot projects, we usually work by engaging
operators, government ministries and NGOs to get everybody on-board and understand
the methodology. We can then use this as a platform for long-term work. In Bangladesh,
we work with Grameenphone, Telenor, ICCCAD and United Nations University. We are
looking both at the short term effects as to how people react and move in response to the
cyclone, and also at the long term and wide-area effects, primarily on migration.
Can you share some of your findings from your current study in Bangladesh?
In terms of preliminary findings, what we see in Bangladesh is consistent with findings from
previous studies of cyclone impacts. For the most part we dont see large displacements,
such as those we documented after the earthquake in Haiti. The low level of displacement
is likely due to Mahasen, being much weaker than the largest super typhoons. We can see
patterns of how people move in preparation for the cyclone and we look at movements
during the cyclone. In Bangladesh, there are a lot of cyclone evacuation centres around
the coast. As the cyclone approaches, we can see in detail how people are moving and
we see that in certain areas there may be a tendency to move to the cyclone centres too
late. By having this type of information the government can investigate if this is related
to the number of shelters or to the communication patterns in the warning messages, or
other reasons.
12
Looking at the long-term changes, we see both significant gains in population in some
areas (mostly urban), potentially through a decreased economic viability in others (mostly
rural), although these analyses are not finalised. This is a common coping mechanism;
if rural economic resources are damaged, people tend to temporarily migrate to urban
areas to work as construction workers or in garment factories, and returning after a
period of time. These are patterns that are poorly understood on a large scale. It is well
known that there is a lot of cyclic temporary migration, and that some of these people will
migrate permanently, but this is difficult to quantify. We help to quantify these movements
and understand them on a wider scale. Hopefully in the future it will provide a better
understanding of what makes certain people migrate.
From what you have seen, are migration patterns specific to a country or a type of
disaster?
The value of mobile phone data is that we are able to look at these patterns as they are
happening and very rapidly analyse the data. There are no generalised patterns that apply
to every country or every disaster. Traditionally survey data collected after disasters shows
very different migration patterns in different countries. Migration patterns also depend on
the disaster, but more crucially they depend on the resilience of the population and the
post-disaster habitability of the area, which differ widely from one place to another and
between families. Moving also demand resources so some people who have experienced
very heavy losses may migrate less, as they have fewer resources to pay for the move. It
has a lot to do with your livelihood, resources, previous experience and what you expect
from the future.
What data do you look at for your analysis?
We look at CDR data, which provides information on location and movement. Currently
we just track the tower locations of anonymous calls. In addition, we also use top-up data,
thus we can see how consumption patterns change over time. We use this data as a proxy
for peoples financial situation after a disaster. We see that people top-up a lot before
a cyclone, evidence of how important it is to communicate during a disaster. A policy
implication could be to allow people to communicate either by providing credit or certain
free calls before, during and after such an event.
Have you seen operator willingness to collaborate change since your project in Haiti?
We collaborate with several operators globally. It is easier than before, but by no means
easy and I dont think it should be either. Although the operator data is very valuable
for disaster response, it is sensitive data, and it naturally takes some time to build trust
between collaboration parties. In Bangladesh we are currently at a pilot phase, but we
hope we can continue to do work here, so it is important to build long-term relationships.
We have an official endorsement from the ministry of disaster response and relief, as well
as from the regulator, which I think has made it easier for Grameenphone to take part
in the work. The minister for disaster response is very interested in this project since for
them this is a completely new area and they see the potential in our analysis. There is a lot
of excitement from different parties, so now we want to finalize the results and get a good
follow-up dialogue. Although Bangladeshi researchers are already part of the project,
we would very much like to be able to work with Bangladeshi researchers and analysts
13
so that this work in the long run can get integrated into the national standard operating
procedures for disaster response.
How accurate is your data compared to other sources?
Based on our work in Haiti we published a comparative analysis contrasting mobile phone
data and a large retrospective household survey, and we found a very good correlation.
In the long run we aim to build up similar validation studies in multiple contexts, including
Bangladesh. Although networks are very resilient, there will also be cases where the
network completely collapses for extended periods of time, for which no data will be
provided. In some settings with skewed user base and limited radio-coverage, this method
will not be of much value. Getting the estimates as good as possible is a long-term research
question that will keep us busy for the foreseeable future. It is especially important to
develop methods to adequately assess the movements of the most vulnerable people like
the poor, children, pregnant and lactating women and the elderly. The alternatives to this
method are however very limited. It is simply extremely hard to manually count individuals
in sometimes-unsafe settings, moving at different speeds, in different directions, across
thousands of square kilometres.
What are the main challenges and opportunities in using big data to address disaster
response?
The opportunities are very clear you can get a rapid picture of population movement
after disasters to be able to locate people in need and to understand better which areas are
severely affected. There are two sets of challenges, firstly a policy one, getting regulators
on board so that operators feel safe. The second challenge is on the scientific side; to
optimise the estimates and account for biases.
14
Agriculture
kg/hectare
With 71% of the population and 85% of the countrys poor living in rural areas, agriculture
is an important source of employment and contribution to GDP in the country. However,
there are some challenges in agriculture in Bangladesh related to weak rural institutions and
vulnerability to natural disasters. Another challenge, which is common to most developing
countries, is the low agricultural productivity considering the amount of labour force in
agriculture. On average, in developing countries 40% of the labour force is in agriculture,
compared to just 3% in the developed world. However, the productivity in developed
countries is 1.5 times greater than that of developing countries (see Figures 5 and 6). The
same is true for Bangladesh; if we compare Bangladesh with the UK, we see that in the
former nearly 50% of manpower is in agriculture compared to just 1% in the latter, but the
productivity in the UK is more than double that of Bangladesh.
50%
40%
5,000
4,000
30%
3,000
20%
2,000
10%
1,000
0%
Developed
countries
Developing
countries
Bangladesh
Figure 5: Labour force in agriculture, 2012

Source: World Bank
Developed
countries
Developing
countries
Bangladesh
Figure 6: Agriculture productivity, 2012

Source: World Bank
We believe there are four main drivers behind the productivity gap:
Lack of information on agricultural inputs and nutrition; prices for crops across
markets and accurate weather information
Access to markets is problematic; there is a gap in matching supply and demand
since intermediaries usually act in silos and there are poor logistics which cause
wastage
Financial exclusion; the availability of loans, payment facilities, savings and insurance
for protection against crop failure, and
Availability of modern agricultural machinery
The first three causes can be addressed through mobile agriculture services. Farmers
currently refer to a variety of sources for their information, (see Table 1) which can be
time consuming. Using mobile to collate information and advisory services focussed on
livestock and nutrition, market prices and weather forecasts would greatly streamline
this process for farmers, at the same time as offering operators and service providers
a chance to create social benefits for their users, enhancing customer loyalty. Secondly,
15
using mobile supply chain services can provide real time visibility of supplier networks and
track and trace products in supply chain. Finally, mobile financial services for farmers, such
as savings, credit products and micro insurance for crops can increase financial inclusion.
Principal information source

Extension
officer
Private
sector
Peer
group
Lead
farmer
Television
Other
High-yielding crop/species
35%
22%
13%
7%
19%
4%
Cultivation techniques
21%
11%
19%
11%
25%
15%
Soil condition
35%
4%
10%
7%
14%
30%
Seed usage
20%
39%
21%
8%
4%
8%
Pesticide usage
14%
65%
5%
7%
3%
6%
Fertiliser usage
16%
56%
9%
8%
2%
10%
Irrigation methods
12%
12%
24%
14%
5%
33%
Market access
7%
20%
41%
15%
5%
12%
Weather forecast
3%
1%
4%
1%
62%
28%
Type of information sought
Table 1: Principal sources of agricultural information among farming population in Bangladesh, 2011
Source: Orgquest, Katalyst
Note: sample population of 500 cereal, staple and vegetable farmers
There is room for significant growth in operator involvement in reaching the 22 million
agricultural workers that have mobile phones; in addition, by offering mobile agriculture
services operators have the potential to attract 14 million new customers to their
subscriber base (see Figure 7) by giving them a reason to connect that they may not
have had previously. Robi, Banglalink and Grameenphone are currently offering mobile
agricultural services, mainly providing market information, agricultural news and weather
information via interactive voice response (IVR) or native voice services in both Bangla
and English. However, these services have not scaled yet; a study carried out by Katalyst
and The Springfield Centre estimates that 200,000 farmers benefitted in 2012 from the
two mAgri services offered by Banglalink and Grameenphone4. This is a significant number,
but it represents only 1% of the total labour force in agriculture in Bangladesh. Offering
voice-based services is an important step given the high rates of illiteracy among the
target audience. However, we believe there is still a lack of awareness of these services and
their value proposition (especially important for individuals with low disposable incomes).
Operators and co-operatives can play a larger role in improving this, such as through rural
distribution centres and below the line advertising.
Source: Making ICT work for Bangladeshs Farmers, November 2012

16
Agricultural workers (million)
25
Have a mobile
Do not have a mobile
20
15
10
Haiti
Rwanda
Cambodia
Zimbabwe
Burkina Faso
Afghanistan
Mozambique
Nepal
Kenya
Tanzania
Bangladesh
Figure 7: Market opportunity for mAgri VAS

Source: GSMA Intelligence
Mobile Money
Today 40% of Bangladesh adults have an account at a formal institution, leaving over
half without one5. In order to ensure access to financial services to the unbanked and
given the increase in mobile adoption, in September 2011 the Bangladesh Central Bank
issued guidelines for mobile financial services (MFS). The central bank has chosen to allow
only banks to provide the services, with mobile operators and microfinance organisations
active partners. The role of operators is limited only to technology provider to the banks
which have been approved to offer mobile financial services. Operators will need to secure
approval from BTRC to allow USSD connectivity to the banks they have a partnership
with; once approved by BTRC, subscribers of the respective mobile operator will be able
to access MFS of that bank. All banks are licenced through the central bank (Bank of
Bangladesh) and can offer different types of services, such as:

Cash in /out using mobile account through agents, bank branches, ATMs, MNO outlets
Disbursement of inward foreign remittances (outward remittances are not allowed)
Person to person (P2P) payments
Person to business (P2B) payments (utility bill payments and merchant payments)
Business to person (B2P) payments (salary disbursement, dividend and refund
warrant payments, vendor payments, etc.)
Government to person (G2P) and person to government (P2G) payments
Other payments (like microfinance, overdrawn facility, insurance premium, DPS, etc.)
Banks are responsible of selecting, training, equipping and monitoring agents. Agents,
together with banks, are allowed to acquire new customers, however, the bank is
responsible for ensuring that the mobile account is activated, that know your customer
(KYC) protocols have been followed and that the documentation provided to register is
verified. In addition, banks are responsible for ensuring the technological infrastructure is
5
Source: Global Findex, World Bank

17
in place to protect customer privacy and data exchange. Banks can decide how much to
charge for services and can link their MFS with those of other banks.
MFS have increased rapidly; the number of agents went from 9,000 to over 414,000 in
just over 2 years and they cover all regions in Bangladesh (see Table 2). Two leaders have
emerged, BRAC bank/bKash and Dutch Bangla Bank; both have formed partnerships with
mobile operators and have access to 88% and 75% of mobile subscribers respectively. For
bKash in particular, this has translated into a strong set of numbers: 90,000 agents, 11.6
million registered users and five million active customers (or around 75% of the total in the
country)6.
bKashs growth has been driven by three main factors7. Firstly, BRAC is the only bank
whose business model is solely based on delivering mobile financial services. The focus of
bKash is only on building its core business and is not considered as an alternative delivery
channel to reach customers. Secondly, bKash was not born out of a pilot project, but it
aimed to scale from launch. Finally, the 2011 regulation created a favourable framework for
bKashs success to the exclusion of mobile operators or their subsidiaries. The regulatory
design has given bKash the competitive edge overt mobile operators, which have proven
to be very successful in providing mobile financial services in countries were regulation is
more enabling.
Description
March 2012
December 2013
June 2014
9,093
188,647
414,170
Number of registered accounts (million)
0.4
13.2
16.7
Number of active accounts (million)
NA
6.5
6.7
$25.9
$857.4
$1,100.1
Number of agents
Total transactions ($, million)

Table 2: Mobile financial services
Source: Bangladesh Bank
At a market level, there are more than 16 million mobile money subscribers in Bangladesh
today, of which 40% are active accounts (see Table 2); this is above the global average of
30%8. However, there is still much room for growth as only 16% of adults are mobile money
users, and it is here that we believe operators are well positioned to provide leadership, and
that there is therefore a strong case for market liberalisation to permit this. For starters,
most operators in the country are already involved in the value chain through partnerships
with banks to facilitate MFS by providing connectivity and distribution. Network scale is a
key asset in reaching the still financially unserved audience but who are mobile subscribers
indeed, GSMAs annual global survey of mobile money services has consistently shown
operator-led services to be among the fastest growing.
Source: Your Neighbour is Your Banker, Your Community is Your Platform, ACCION, April 2014
Source: bKash Bangladesh: A Fast Start for Mobile Financial Services, CGAP, July 2014
8
Source: Mobile Financial Services for the Unbanked: State of the Industry 2013, GSMA, February 2014
6
7
18
Security concerns have arisen given that the majority of transactions currently go over
the counter (OTC) via an agent, circumventing registration and KYC checks9. Indeed 85%
of mobile money users have not registered their own account and the majority prefers
to conduct transactions through an agents account10. While OTC via an agent properly
trained on the necessary anti-money laundering (AML) risks and due diligence may, in
special circumstances, be permitted, the failure of agents to carry out any KYC checks
exposes MFS providers to money laundering and terrorism financing risks and could
result in Bangladesh facing international reprisals for non-compliance with internationally
accepted AML/CFT (countering financing of terrorism) controls.
Finally, operators can draw upon their local area and customer expertise to more effectively
communicate value proposition of mobile financial services. All of these come into play
given the importance of speed and scale if the government is to achieve its vision of
driving universal financial inclusion. Under the current regulatory framework, the Bank
of Bangladesh carries out a prudential assessment of the banks (or its subsidiarys)
preparedness to offer MFS together with on-going reporting obligations. This prudential
assessment can be extended to MNOs (or their subsidiaries) as the risk profile of the
MFS service remains the same for bank subsidiaries and non-bank subsidiaries11. Indeed
by adopting a non-discriminatory approach to the regulation of MFS under which banks
and non-banks, particularly MNO subsidiaries, are permitted to issue e-money, the Bank
of Bangladesh will have boosted access to financial services in Bangladesh accruing the
following benefits:
Improving efficiencies and scale to the national payments system
Addressing any regulatory conflict or arbitrage concerns as the regulated institution
will be a special purpose vehicle specifically incorporated to provide MFS
Unlock capital held back by MNOs, bringing investment and spurring innovation
Enhance efficiency in distribution of MFS through existing wide-scale distribution
networks
Inspire public confidence in MFS as MNOs will utilise their resources to educate the
public and drive mass recruitment campaigns, reversing the current lack of customer
awareness on the product and general customer apathy.
Some mobile money services are being offered primarily over-the-counter (OTC). In such cases, a mobile
money agent performs the transactions on behalf of the customer, who does not need to have a mobile
money account to use the service and does not necessarily need to provide KYC. Services that dont offer
OTC register users by requesting KYC and a mobile money account (wallet) is opened in the users name.
Some services offer both OTC and mobile money accounts
10
Source: Bangladesh - Quicksights Report FII Tracker Survey, Financial Inclusion Insights, April 2014
11
A functional approach to regulation focuses on applying the same rules for the same kind of service
(whether offered by a bank or non-bank) while an institutional specific approach focuses on regulating the
institution providing the services. For a more detailed explanation of this point and for references see Mobile
Money: Enabling Regulatory Solutions, Simone di Castri, GSMA, 2013
9
19
Education and employment

In recent years, education levels have improved in Bangladesh, especially the ratio of
children completing primary education, which went from 57% in 2008 to 75% in 201112.
However, literacy rates are still quite low 55% for females and 62% for males. Among the
youth demographic, literacy is better (around 80%), but unemployment rates are nearly
double compared to the national average (see Figure 8).
25%
Overall
Youth
20%
15%
10%
5%
Rwanda
Benin
Cam bodia
Nepal
Burkina Faso
Tanzania
Zimbabwe
Bangladesh
Mozambique
Kenya
Afghanistan
0%
Figure 8: Unemployment rates

Source: ILO, World Bank
The main challenges Bangladeshis face when looking for a job are lack of skills and
information13 most job searches are informal (word of mouth) rather than formal (such
newspapers or magazines). Mobile phones and computers are not used frequently when
looking for a job. However, Bangladeshis have shown a willingness to use their mobile to
access training and to pay for job search and job matching information if delivered via
phone: the preferred communication method is over voice calls rather than SMS (see Value
of voice an unsung opportunity). Younger generations have shown willingness to pay for
employment related services for three main reasons: they can use the service when they
like, they dont have to travel to use this service and this would be the only way they would
have access to this service. The market is significant we estimate 6.5 million youth will
have access to mobile employment services by 201813.
From a business model perspective, we expect most of the opportunity to be B2C.
Planned services are collaborative efforts, with one targeting a mobile service for women
entrepreneurs in Bangladesh by providing information on starting a business and accessing
finance, and another a partnership between BRAC, Robi and the British Council to launch
a mobile education service to help adolescent girls develop English skills to improve their
employment prospects. Mobile operator interest is evident in both of these, and we would
12
13
Source: World Bank

Source: Mobile Services for Youth Employment, GSMA/Alcatel-Lucent, July 2013
20
expect that to continue over the next 23 years given the size of the target market and
the favourable trust perceptions held of mobile operators in this space. The monetisation
question is challenging in that there is little precedent for paid-for services in this area.
Clever thinking in tariffs such as by bundling in employment VAS with connectivity
services would help, as would more concerted marketing efforts because services in
this area carry a specific demographic bent. However, demand is likely to grow both as a
result of more university graduates seeking work and a greater proportion of these having
the digital literacy skills that make the use of mobile an attractive option (ICT education
has been made compulsory since 2013 at the secondary level under the Digital Bangladesh
vision, with an equivalent target at the primary level for 2021).
Investment in innovation
While there are clear market opportunities for driving scalable VAS that have both a
commercial and social impact, it is important to link the discussion back to the who. In
Bangladesh, one proxy for its growing maturity is the increased involvement of mobile
operators in delivering M4D services. The country is well ahead of others on this measure
(see Figure 9), and while operators have identified challenges in continuing to invest in this
space (notably regulatory boundaries and a lack of innovation with clear business models),
we expect these to be gradually overcome over the next 23 years. Where operators
are not in a leadership position, NGOs are increasingly filling the gap, especially in the
health, agriculture and learning sectors. For example, BRAC launched the innovation fund
challenge for entrepreneurs focused on mobile money to improve its existing programs
and establish new ones with the objective of increasing the adoption of digital financial
services in the country.
MNO led
NGO led
Other
100%
25%
80%
55%
60%
40%
20%
71%
30%
18%
7%
45%
27%
22%
0%
Bangladesh
Mozambique
Nigeria
Figure 9: Stakeholder involvement in products and services

Source: GSMA M4D Impact
This is a positive story given that seed capital is a necessary ingredient in the transition
from pilot to scalable market services. However, Dhaka is not yet an innovation hub to the
extent of other regional centres such as Bangalore, Colombo or (further afield) Nairobi.
The relative lack of involvement is not, however, due to a lack of investable opportunities.
21
VCs in Bangladesh face two main challenges: a lack of exit strategies and opportunities,
and regulation. The first challenge is aligned with our recent analysis on the different
challenges that investors face when investing in ICT in emerging markets (see Figure 10
and Financing innovation). Most commonly, companies exit through an IPO or by being
bought out. Neither are considered reliable options in Bangladesh, as the stock market is
not vibrant and there is little practice of M&A.
The business model is not scalable

Lack of appropriate experience and exposure
Lack of appropriate business skill
There are few/no attractive exit options
The business model is not replicable
The idea is not ambitious enough
It is difficult to form partnerships
The idea is not innovative enough
Fears of the bigger players (eg. mobile operators) wiping out the business
The venture is too impact focussed and not commercial enough
It is difficult to maintain partnerships
Difficulties in valuing companies in the ICT sector in emerging markets
Entrepreneur does not fit in with the culture of the investing organisation
Lack of suitable co-investment/co-intervention partners
Lack of appropriate technical skills
Lack of confidence in projected social impact
The ICT sector is relatively new and uncertain
Political risk
Macroeconomic risk
1
Quality of entrepreneur (average 2.93)
Business model (2.87)
Quality of idea (2.73)
Return on investment (2.79)
Ecosystem (2.48)
Country risk (2.13)
Figure 10: Risk perceptions of investors active in ICT in emerging markets

Source: GSMA M4D Impact
Note: 1 = not a relevant risk; 4 = highly relevant risk
On the regulatory front, there is no framework in Bangladesh that secures the investment
of VC firms. This leads to a higher risk in investment which deters investment in the country.
According to the IESE Business School study on venture capital and private equity country
attractiveness index, Bangladesh ranked 80th out of 116 countries (see Figure 11), this is
much lower compared to other countries in the region such as India and Pakistan which
ranked 30th and 66th respectively. Bangladeshs poor score can largely be traced to issues
of the market operating environment and governance (specifically investment protection
and corporate governance, human and social environment, and entrepreneurial culture and
deal opportunities drivers). We believe that as these challenges are tackled there is still an
opportunity for other investors to be involved (currently there are a few active players such
as Venture Investment Partners Bangladesh Limited (VIPB), Asian Tiger Capital Partners
and the Equity & Entrepreneurship fund (EEF)).
22
VC/PE attractiveness index
More attractive
0
20
China
India
40
60
Kenya
Pakistan
80
Cambodia
Bangladesh
100
Benin
120
200
Burkina Faso
Zimbabwe
Less attractive
150
100
50
Ease of doing business rank
Figure 11: Ease of doing business (2014) and VC/PE country attractiveness (2013) indices
Source: IESE Business School, World Bank, IFC
Note: VC/PE country attractiveness index ranked out of 188 countries with data for 2013
Ease of doing business ranked out of 189 countries with data for 2014
Lower scores are more attractive
At an innovation level, the government is planning to create hi-tech parks to create a

suitable business environment for technology industries, with the Ministry of Science
and Information & Communication Technology (MOSICT) having now established an ICT
incubator to develop the ICT sector in Bangladesh. All of these are positive signs and
indicative of the conviction behind fostering an enabling environment for sustainable
innovation in ICT. We therefore believe efforts should be focused on mitigating concerns
on the market operating environment, particularly those around investment protection,
promoting the free flow of capital and exit opportunities on the public markets.
23
The mobile market

Metric
2011
2012
2013
2014
Connections (million)
85.7
97.6
114.3
125.1
100%
100%
100%
100%
% prepaid
97%
97%
97%
97%
SIMs per subscriber
1.65
1.70
1.81
1.86
Unique subscribers (million)
51.7
57.2
62.9
67.1
Penetration, connections
56%
63%
73%
78%
Penetration, unique subscribers
34%
37%
40%
42%
Connections growth (annual)
25%
14%
17%
9%
Unique subscriber growth (annual)
16%
11%
10%
7%
ARPU, by connection ($)
$2.40
$2.27
$2.01
ARPU, by subscriber ($)
$3.85
$3.81
$3.55
$2,221
$2,493
$2,556
18%
12%
3%
% active
Recurring revenue ($, million)

Recurring revenue growth (annual)
Table 3: Bangladesh, key mobile indicators
1989
Licence issued
to Bangladesh
Telecom Limited
(CDMA) and
Sheba Telecom
1996
1997
Bangladesh
Launch of
Licence
Telecom Limited Grameenphone,
issued to
rebranded to
AKTEL and
Telecom
Pacific Bangladesh
Sheba
Malaysia
using the brand
Telecom
International
Citycell
Bangladesh and
Grameenphone
2004
2005
2007
2010
2012
Launch of
state-owned
Teletalk
Orascom
acquires Sheba
Telecom,
rebranded to
Banglalink
Launch of
Warid
Bharti Airtel
acquires Warid
Teletalk
launches
first 3G
network in
Bangladesh
AKTEL
rebranded
as Robi
Figure 12: Mobile network operator launch timeline

Low disposable income, but relatively high mobile penetration with room for further
growth
Bangladesh is a low-income country, with a GNI per capita of $900 (we are using GNI per
capita as income levels are based on it), however, it is the tenth largest market worldwide
in terms of unique mobile subscribers. Based on the global mobile penetration average of
the different income levels, Bangladesh has a higher mobile penetration compared to the
low and lower-middle income economies (see Figure 13). The number of mobile phone
users in Bangladesh has grown rapidly; in 2003 subscriber penetration was only 1%, and in
ten years this grew to 40%. It is expected to grow to 50% by 2020. There is still significant
room for growth as more than 40 million adults still do not subscribe to mobile services.
24
Penetration (%)
100%
78%
80%
80%
48%
40%
38%
Low income
Bangladesh
(low income)
Lower-middle
income
Upper-middle
income
High income
GNI amount
<$1,045
$900
$1,045$4,125
$4,125$12,746
>$12,746
Example
countries
Cambodia
Ethiopia
Nigeria
Bolivia
Indonesia
Zambia
Brazil
China
South Africa
Chile
Israel
United Kingdom
40%
32%
20%
0%
Figure 13: Average mobile subscriber penetration by income level, globally

One of the reasons behind Bangladeshs high mobile penetration rate is due to the impact
of the Grameenphone Village Phone programme, a pioneering initiative that, among other
things, has led to the empowerment of rural women in Bangladesh. Grameen Bank provides
women entrepreneurs in rural areas with credit to buy a phone from Grameenphone
which is then used to provide mobile pay phone services, allowing the women to charge
a markup agreed with Grameenphone. Underpinning this is the rapid roll-out of mobile
infrastructure. By 2000 nearly half of the country was covered by mobile networks, and
this rapidly increased to nearly the whole country by 2002 (see Figure 14). In June 2014,
Grameenphone reported that 2G coverage was 99.17% by population and 89.50% by area14
and that all 64 of the countrys district headquarter cities are covered by its 3G network15.
In this regard, Bangladesh is significantly ahead of countries within its region or level of
development (see Figure 15).
14
15
Source: Q2 2014 Business Highlights, Grameenphone, 2014

Source: Grameenphone 3G reaches all 64 district headquarters, TeleGeography, April 2014
25
1997
3/64 districts
2000
30/64 districts
2002
50/64 districts
2004
61/64 districts
Present
64/64 districts
Figure 14: Mobile coverage evolution

Source: Association of Mobile Telecom Operators of Bangladesh (AMTOB)
100%
99%
97%
95%
91%
90%
89%
87%
85%
85%
85%
Zimbabwe
Tanzania
80%
75%
Bangladesh
Rwanda
Kenya
Afghanistan
Cambodia
Figure 15: 2G population coverage16

In deploying network infrastructure, operators have needed to extend their coverage to rural
areas with limited access to the grid. Operators need to find an alternative energy solution
to reduce their dependency on diesel generators (DG) to power up their base stations,
as this leads to a significant increase in operational expenses (opex). By depending on a
DG solution, operators spend 65% of their opex on diesel for off-grid sites. Operators can
implement a green technology solution for 3,622 problematic sites and claim some savings;
in 2012 there were only 171 green-power sites. GSMA Green Power for Mobile (GPM) has
estimated that for the problematic sites, operators could save around $41 million in opex
with total investment of up to $100 million in 2012. This can increase to a $90 million saving
per year by 2015 with an investment of $184 million for 6,660 problematic sites17.
16
17
For most countries the figure represents the coverage of the largest operator
Source: Extending the Grid: Bangladesh Market Analysis, GSMA, 2013
26
Connections (million)
Since 2007 there have been six mobile operators in Bangladesh18. Grameenphone is the
dominant operator, with a 41% market share at the end of 2013 (see Figure 16). Bangladesh
has a relatively competitive market which has had a clear consumer benefit in terms of
falling mobile phone prices; the effective price per minute has fallen from $0.12 in 2002 to
$0.01 today which in turn has led to Bangladesh enjoying a lower cost of ownership as a
share of income compared to its peer countries (see Figure 17 and 18).
120
Citycell
Teletalk
100
Airtel
80
Robi
Banglalink
60
Grameenphone
40
20
0
2004
2005
2006
2007
2008
2010
2009
2011
2012
2013
Figure 16: Operator market share evolution

$0.14
20%
19%
18%
$0.12
16%
$0.10
14%
12%
$0.08
9%
10%
$0.06
9%
8%
8%
7%
7%
6%
6%
6%
6%
$0.04
5%
4%
4%
$0.02
3%
2%
Bangladesh
Rwanda
Nepal
Cambodia
Tanzania
Afghanistan
Kenya
Benin
Zimbabwe
Burkina Faso
Mozambique
2013
2011
2012
2010
2009
2007
2008
2006
2005
2004
2002
2003
Figure 17: Effective price per minute ($)

Haiti
0%
$0.00
Figure 18: Airtime cost (ARPU) as a share of income, 2013

Source: GSMA Intelligence, World Bank
Bangladesh has one of the lowest average revenue per user (ARPU) levels in the world;
however this has not always been the case. In 2001, ARPU was nearly $30 and declined to
$2 in 2013. Over the same period the prepaid subscriber base grew from 62% of connections
in 2001 to 97% today. While contract ARPU is still 34 times higher than prepaid ARPU in
There are other two operators Qubee and Banglalion Communications with WiMAX licences and a market
share of less than 1%
18
27
Bangladesh, this change in the prepaid mix is the dominant factor for the decline in blended
ARPU (see Figure 19). Another reason for the declining ARPU is a mix effect from new
subscribers that are low income. Finally, adverse regulation surrounding the introduction
of per 10 second billing on voice calls (as opposed to per minute) have impacted ARPU
growth. This decline in ARPU has had a negative effect on mobile service revenue growth
which now sits at 3% per year. Another factor has been the decline in subscriber growth;
in 2012 the Bangladesh Telecommunication Regulatory Commission (BTRC) deactivated
over one million VoIP customers from Banglalinks subscriber base and cut incoming voice
call termination rates by 50% to fight the illegal calls market, placing pressure on revenue
growth.
Blended ARPU
Prepaid ARPU
Contract ARPU
Prepaid connections
$60
100%
$50
80%
$40
60%
$30
40%
$20
20%
$10
$0
0%
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
2011
2012
2013
Figure 19: ARPU evolution

Bangladesh is still a market dominated by voice. In 2013, on average, 94% of Grameenphones

recurring revenue was generated via voice services. This is one of the highest rates in
the world, compared with an average of 85% in Zimbabwe, 64% in Kenya and 60% in
the US, Europe and Asia. Of non-voice revenue, VAS account for only 3% of operators
recurring revenue. In addition, we expect ARPU to continue to decline as a majority of new
subscribers will continue to be low income prepaid rural customers. This underlines the
need to mitigate this by developing new revenue streams. We believe the opportunity is
in data and VAS.
28
The mobile internet is on 2G

Bangladesh is a prepaid and 2G market; at the end of 2013 97% of connections were
prepaid and 98% of connections were on 2G. 3G licences were only awarded at the end
of 2013 after a series of delays from the regulator. As such, Bangladesh is one of the last
countries in Asia to be awarded 3G licences (see Bangladesh: Asias untapped mobile
broadband opportunity). However, the top four mobile operators have been offering data
services on 2G networks. Hence, while 3G penetration is low, mobile internet connection
penetration is over 20% (see Figure 20). For example, in Q2 2014, Grameenphone reported
that five million active Facebook users were on its network. The uptake of 3G connections
is expected to be relatively fast; by 2020 we forecast 3G connections will surpass 2G, even
if this is slower than the regional average.
Mobile internet penetration
3G penetration
25%
20%
15%
10%
5%
0%
Q2 2012
Q3 2012
Q4 2012
Q1 2013
Q2 2013
Q3 2013
Q4 2013
Q1 2014
Figure 20: Mobile internet connections19 vs 3G penetration

Source: GSMA Intelligence, BTRC
In the first half of 2013 smartphones accounted for 6% of total handset shipments into
Bangladesh, which has steadily increased but is still very low (see Figure 21). The largest
vendor in Bangladesh is a local brand, Symphony Mobile, with a 37% handset market
share, of which 34% accounts for feature phones and 3% for smartphones (see Figure 22).
In just five years, Symphony Mobile has out-powered international giants like Nokia and
Samsung. Nokia dominated the Bangladesh handset market for around a decade with a
50%+ handset market share annually until the end of 2010, slipping under 50% in 2011, and
was surpassed by Symphony Mobile in 2012. Symphony Mobiles unique offerings and firstmover advantage have given the firm a distinct competitive edge; for example they were
the first to launch dual-SIM handsets and the countrys first phone with QWERTY keyboard
and trackball. Another selling point is given by their pricing structure, handset prices start
from BDT 1,000 ($13) and go up to BDT 20,000 ($250). In addition, their smartphones run
on Android, which is the most popular operating system in Bangladesh. Lastly, Symphony
Mobile has established customer care centres in all major districts and collection points in
the smaller districts to ensure service to customers based in remote areas.
19
Figures for mobile internet represent the number of connections using the internet over the last 90 days
29
Million
Smartphone
Feature phone
8%
3%
0
Q1 2013
Q2 2013
Figure 21: Feature phone vs smartphone shipments, H1 2013

Source: CMR India
Symphony Mobile
3%
Samsung
1%
Other
2%
Other
43%
Smartphones
6%
Symphony Mobile
34%
Feature phones
94%
Micromax
5%
Nokia
6%
Maximus
7%
Figure 22: Handset sales share, H1 2013

Source: GSMA Intelligence, CMR India
30
Regulation
The current regulatory framework in Bangladesh is based on the National Telecommunications
Policy White Paper from 1998 and the Telecoms Regulating Act of 2001. The 2001 act laid
out the framework for an independent regulator (BTRC) as well as a broad framework for
operator licensing and spectrum management. The act was amended in 2010 to remove
BTRCs power to set tariffs and issue licences without central government approval. In
addition, various sanctions were imposed with regards to licence violations. VoIP services
were also legalised and provisions for a universal service fund (USF) made.
In 2012, the government commissioned an expert from the ITU to look at reforming the
legislative and policy framework, with nine main recommendations, of which the most
relevant are to: determine presence of market dominance and need for regulation to
promote competition; introduce a market-based spectrum allocation, trading and sharing
process; and review taxation policy.
Yet there has been no formal adoption of the draft policy proposed in 2012. There has
been some progress in implementing a few recommendations, but in other areas things
have actually worsened such as mobile number portability (MNP; a BTRC Directive was
issued to operators to provide this service for ~$1 per request in mid-2013, with a deadline
of January 2014 to comply, yet this deadline was not met for technical reasons) and
infrastructure sharing (the regulator looks set to approve the countrys first mobile network
infrastructure agreement between Teletalk and Airtel, however, there is no evidence that
fixed/mobile infrastructure sharing is currently occurring). In May 2014, the current minister
of telecommunications stated that the telecom policy will be revised with all operators
calling for a review such that policy reflects evolving technology, particularly in mobile.
The Bangladesh mobile market faces two main regulatory challenges: spectrum and
taxation. The amount of spectrum assigned to operators in Bangladesh is low compared
to developed markets (see Table 4). In addition, an auction for 3G spectrum was only held
in September 2013 after a series of delays from the regulator. BTRC does not yet have a
long-term strategy and roadmap in place for the allocation and planning of spectrum,
making it difficult for operators to plan future deployments. In licensing, there appears to
be a different treatment to different licensees; BTRC issued WiMAX licences without any
competitive auction, despite strong opposition from operators and denying a high court
ruling.
31
Frequency
(MHz)
800
850
(CDMA)
2G
3G
DD20
900
1800
2100
700
WiMAX/LTE
2300
2500
(TDD)
(TDD)
2600
Grameenphone
2 7.4
2 14.6
2 10
Banglalink
25
2 10
25
Robi
2 7.4
2 7.4
25
Airtel
25
2 10
25
10
TeleTalk
2 5.2
2 10
2 10
Citycell
2 8.82
Qubee
35
Banglalion
35
2 10
Always On
26
Other
2 5.04
2 7.5
20
25
2 2.86
2 15.5
2 44.34
2 36
35
15
2 70
2 15 2 16.72
2 30
2 75
2 54.34
2 42
100
50
2 70
Olli
Vacant
Total
Table 4: Spectrum portfolio, Bangladesh

In addition, Bangladesh has one of the highest taxation rates for mobile services in the
world, with operators paying tax at nearly 60% of revenue, this has been increasing since
2008 at an annual rate of 8%. Just over 70% of mobile services taxes are sector-specific
taxes (i.e. corporate taxes that are only paid by companies within the mobile sector),
which is higher than the estimated average of 40%21. In addition to these taxes, the finance
minister has proposed an additional 15% value added tax (VAT) on the import of mobile
handsets; this could lead to a further illegal import of mobile phones. Another tax that has
been a longstanding issue in Bangladesh is the tax on SIM cards and SIM replacements.
The industry has been keen on removing the BDT 300 ($3.90) tax on SIM cards and the
BDT 100 ($1.30) tax on SIM replacement given the risk that the tax hike could reduce SIM
sales and cut growth in the sector.
All these taxes and regulations will impact the growth of investments and quality of
services; in the current situation operators will not be able to continue offering services at
the current rates. This will impede the growth of the industry and the competition to get
new subscribers especially from the bottom of the pyramid.
20
21
Digital dividend
Source: Mobile Taxes and Fees - A Toolkit of Principles and Evidence, GSMA/Deloitte, 2014
32
Appendix
Benchmarking Bangladesh: methodology
Nepal
Afghanistan
Haiti
Bangladesh
Burkina Faso
Cambodia
Benin
Kenya
Rwanda
Tanzania
Zimbabwe
Mozambique
Figure 23: Benchmarked countries

For the purpose of this report we have chosen a selection of 11 other benchmark countries
to compare to Bangladesh (from a total 236 countries world-wide). The selection of
countries has been made on the following four criteria:
1.
2.
3.
4.
Wealth GDP per capita (2013)

Economic growth GDP growth rate (201013)
Mobile market maturity subscriber penetration (Q4 2013)
Human development Human Development Index (HDI, 2012)
Taking Bangladesh as the base we have selected countries which fall under a variance
range chosen for each of the four criteria. Table 5 shows the range of variances chosen to
arrive at a group of countries which represent major regions across the globe.
Criteria
Variance
Lower value
Nigeria
Higher value
GDP per capita ($)
30%
$580
$829
$1,078
GDP growth rate (%)
3pp
3.3%
6.3%
9.3%
~ 15pp
25%
40%
55%
0.2
0.32
0.52
0.72
Subscriber penetration (%)

HDI (1 is the highest)
Table 5: Benchmarking criteria

33
GDP per capita ($580$1,078)

21 countries (including Bangladesh): Asia Pacific (4),
Latin America (1), MENA (1), SSA (15)
GDP growth rate (3.3%9.3%)

Subscriber penetration (25%55%)

HDI (0.320.72)
12 countries*: Afghanistan, Bangladesh, Benin,

Burkina Faso, Cambodia, Haiti, Kenya, Mozambique,
Nepal, Rwanda, Tanzania and Zimbabwe
* Four countries (Lesotho, Mauritania, Tajikistan and Togo) filtered owing to very low population
Country
Sub region
GDP per
capita ($)
GDP
growth
rate (%)
Subscriber
penetration
(%)
HDI
Population
(million)
Afghanistan
South Asia
$678
8.2%
37.7%
0.374
30
Bangladesh
South Asia
$829
6.3%
39.9%
0.515
156
Benin
Africa
$805
4.8%
42.8%
0.436
10
Burkina Faso
Africa
$684
6.8%
31.6%
0.343
17
Cambodia
East Asia/Pacific
$1,008
7.3%
50.8%
0.543
15
Haiti
Latin America
$820
4.2%
39.8%
0.456
10
Kenya
Africa
$994
4.6%
39.8%
0.519
44
Mozambique
Africa
$593
7.3%
25.0%
0.327
26
Nepal
South Asia
$694
4.0%
36.5%
0.463
28
Rwanda
Africa
$633
6.9%
30.3%
0.434
12
Tanzania
Africa
$695
6.8%
29.8%
0.476
49
Zimbabwe
Africa
$905
5.6%
47.2%
0.397
14
Table 6: Data used for benchmarking

Whilst 16 countries show similar profiles after applying the four benchmarking criteria,
we saw that four of these have very low population (less than 10 million) when compared
34
to Bangladesh which has around 155 million inhabitants. We therefore filtered these out
leaving us with 12 countries including Bangladesh along with Afghanistan, Benin, Burkina
Faso, Cambodia, Haiti, Kenya, Mozambique, Nepal, Rwanda, Tanzania and Zimbabwe.
Note that while we have benchmarked Bangladesh against these peers for the purposes of
aligning markets at similar stages of development, we also provide regional comparisons
at relevant points in the report to group markets influenced by similar regulatory climates
and spectrum plans.
35
Country environment
Relevant groups and organisations
Government bodies and trade

associations
Bangladesh Telecom Regulatory Commission

Association of Mobile Telecom Operators of Bangladesh
Data: regulation, business environment, demographics and economics

2014
ICT regulation (1 = nonexistent,
7 = well developed)
Government emphasis on ICT
(1 = weak priority, 7 = high priority)
Bangladesh
Kenya Mozambique
Nepal
Tanzania
3.0
4.1
2.9
2.7
3.2
4.1
4.7
3.4
2.9
3.7
Kenya Mozambique
Nepal
Tanzania
Table 7: Government and regulation

Source: Global IT Report 2014, World Economic Forum
2014
Bangladesh
Venture capital availability (1 = very

difficult; 7 = very easy)
2.0
3.0
2.1
2.3
2.6
Impact of ICT on new products, services & business models (1=not at all;
7 = significantly)
3.8
4.8
3.6
3.5
3.9
Impact of ICT on access to basic

services (1 = do not enable access at
all, 7 = enable access)
3.7
4.3
3.3
3.3
3.4
Ease of doing business rank
130
129
139
105
145
Business entry density rate22
0.1
0.9
Number of days to start a business
10.5
32
13
17
26
Corruption perception index (0 =

highly corrupt, 100 = highly clean)23
27
27
30
31
33
Table 8: Business environment and entrepreneurship

Source: Global IT Report 2014, World Economic Forum, Transparency International, World Bank
22
23
Data for 2012

Data for 2013
36
Topic
2014 rank
2013 rank
Change
130
132
Starting a business
74
83
Dealing with construction permits
93
80
-13
Getting electricity
189
189
Registering property
177
177
Getting credit
86
82
-4
Protecting investors
22
21
-1
Paying taxes
100
98
-2
Trading across borders
130
126
-4
Enforcing contracts
185
185
Resolving insolvency
119
121
Overall ranking
Table 9: Ease of doing business, by topic

Source: World Bank Ease of Doing Business Rankings
Bangladesh
Kenya
Mozambique
Nepal
Tanzania
South
Asia
158
45
26
28
50
1,760
Urban population
29%
25%
32%
18%
28%
33%
Literacy rate22
59%
72%
51%
57%
68%
70%
Median age (years)24
24.3
19.1
16.9
22.9
17.4
26.2
Mobile penetration, connections
73%
70%
48%
77%
54%
74%
Mobile penetration, subscribers
40%
40%
25%
37%
34%
34%
2013
Population (million)
Table 10: Demographic data

24
Data for 2014

37
Bangladesh
Kenya
Mozambique
Nepal
Tanzania
South
Asia
6.0%
4.7%
7.1%
3.8%
7.0%
3.7%
FDI (% of GDP)22
1.1%
0.6%
36.4%
0.5%
6.0%
1.2%
Unemployment22
4.5%
9.2%
7.5%
2.7%
3.5%
4.1%
Inflation
7.5%
5.7%
4.2%
9.0%
7.9%
10.9%
2013
GDP growth
Table 11: Economic data

Source: World Bank, IMF, GSMA Intelligence
38
Glossary
ICT regulation
How would you assess your countrys laws relating to the use of information and
communication technologies (e.g., electronic commerce, digital signatures, consumer
protection)? Key: 1 = nonexistent; 7 = well developed, 20102011 weighted average.
Government emphasis on ICT
How much priority does the government in your country place on information and
communication technologies? Key: 1 = weak priority; 7 = high priority, 20102011
weighted average.
Venture capital availability
In your country, how easy is it for entrepreneurs with innovative but risky projects to find
venture capital? Key: 1 = very difficult; 7 = very easy, 201011 weighted average.
Impact of ICT on new products, services and business models
To what extent are information and communication technologies creating new business
models, services, and products in your country? Key: 1=not at all; 7 = significantly, 2010
2011 weighted average.
Impact of ICT on access to basic services
To what extent are information and communication technologies enabling access for
all citizens to basic services (health, education, financial services, etc.) in your country?
Key: 1 = do not enable access at all, 7 = enable access significantly, 20102011 weighted
average.
Business entry density rate
Recurring (service) revenue generated in the period, including revenue generated from
the use of the network (voice, messaging, data, VAS), but excluding non-recurring
revenue such as handset or equipment revenue.
Unique subscribers
Total unique users who have subscribed to mobile services at the end of the period,
excluding M2M. Subscribers differ from connections such that a unique user can have
multiple connections.
Mobile penetration, subscribers
Total subscribers at the end of the period, expressed as a percentage share of the total
market population.
ARPU, by subscriber
Average revenue per user (ARPU). Total recurring (service) revenue generated per
unique subscriber per month in the period. Different from ARPU by connection, ARPU by
subscriber is a measure of each unique users spend.
Mobile termination rate (MTR)
Charges which one mobile operator charges to another for terminating calls on its
network
39
About GSMA Intelligence

GSMA Intelligence is the definitive source of mobile operator data, analysis and forecasts,
delivering the most accurate and complete set of industry metrics available.
Relied on by a customer base of over 800 of the worlds leading mobile operators, device
vendors, equipment manufacturers and financial and consultancy firms, the data set is the
most scrutinised in the industry.
With over 22 million individual data points (updated daily), the service provides coverage
of the performance of all 1,140 operators and 1,153 MVNOs across 3,505 networks, 65
groups and 236 countries worldwide.
gsmaintelligence.com [email protected] @GSMAi
About the authors

Barbara Arese Lucini Analyst, Emerging Markets
Barbara is an analyst at GSMA Intelligence focusing on research for emerging
markets. Before joining GSMA in April 2013, Barbara worked for FrontlineSMS
in London and at Accenture in Italy. She holds an MSc in Development Studies
from SOAS, London and an undergraduate in Mathematics from Universit
Statale di Milano, Italy.
Tim Hatt Lead Analyst

Tim is the head of research for emerging markets at GSMA Intelligence, having
joined the team in October 2012. In this capacity, Tim has responsibility for the
team producing research reports and presenting externally at conferences and
public speaking engagements. Prior to joining the GSMA, Tim spent 6 years in
London as an analyst covering telecoms and a variety of other sectors.
40
gsmaintelligence.com [email protected] @GSMAi
Whilst every care is taken to ensure the accuracy of the information contained in this material, the facts, estimates and opinions stated are
based on information and sources which, while we believe them to be reliable, are not guaranteed. In particular, it should not be relied upon
as the sole source of reference in relation to the subject matter. No liability can be accepted by GSMA Intelligence, its directors or employees
for any loss occasioned to any person or entity acting or failing to act as a result of anything contained in or omitted from the content of this
material, or our conclusions as stated. The findings are GSMA Intelligences current opinions; they are subject to change without notice. The
views expressed may not be the same as those of the GSM Association. GSMA Intelligence has no obligation to update or amend the research
or to let anyone know if our opinions change materially.
GSMA Intelligence 2014. Unauthorised reproduction prohibited.
Please contact us at [email protected] or visit gsmaintelligence.com. GSMA Intelligence does not reflect the views of the GSM
Association, its subsidiaries or its members. GSMA Intelligence does not endorse companies or their products.
GSMA Intelligence, 5 New Street Square, New Fetter Lane, London EC4A 3BF
41

Bangladesh Overview - Banking

Uploaded by

Copyright:

Available Formats

Bangladesh Overview - Banking

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Bangladesh Overview - Banking

Uploaded by

Copyright:

Available Formats

ANALYSIS

Country overview: Bangladesh

gsmaintelligence.com/m4d [email protected] @GSMAi

GSMA Intelligence Country overview: Bangladesh

GSMA Intelligence Country overview: Bangladesh

2. Bangladesh is predominantly a prepaid and 2G market, as 3G has only recently

Excluding youth under 16

GSMA Intelligence Country overview: Bangladesh

by delivering mobile data services using 2G networks. The combination of 3G spectrum

GSMA Intelligence Country overview: Bangladesh

GSMA Intelligence Country overview: Bangladesh

Figure 1: Urban vs rural gap, 2010

Poverty headcount ratio at urban and rural poverty lines

GSMA Intelligence Country overview: Bangladesh

GSMA Intelligence Country overview: Bangladesh

Mobile for Development opportunity

GSMA Intelligence Country overview: Bangladesh

Figure 2: The opportunity is not just in rural areas

Figure 3: Commercial VAS offered by operators in Bangladesh

GSMA Intelligence Country overview: Bangladesh

These do not include droughts, nuclear or industrial accidents and epidemics

GSMA Intelligence Country overview: Bangladesh

climate change by 2050.

The opportunity in mobile is multi-pronged, underpinned by the relatively high level of

GSMA Intelligence Country overview: Bangladesh

Special feature: In conversation with Linus Bengtsson

Can you tell us what Flowminder is and what you do?

GSMA Intelligence Country overview: Bangladesh

GSMA Intelligence Country overview: Bangladesh

GSMA Intelligence Country overview: Bangladesh

Figure 5: Labour force in agriculture, 2012

Figure 6: Agriculture productivity, 2012

GSMA Intelligence Country overview: Bangladesh

Principal information source

Type of information sought

Source: Making ICT work for Bangladeshs Farmers, November 2012

Agricultural workers (million)

GSMA Intelligence Country overview: Bangladesh

Do not have a mobile

Figure 7: Market opportunity for mAgri VAS

Source: Global Findex, World Bank

GSMA Intelligence Country overview: Bangladesh

Number of registered accounts (million)

Number of active accounts (million)

Total transactions ($, million)

GSMA Intelligence Country overview: Bangladesh

GSMA Intelligence Country overview: Bangladesh

Education and employment

Figure 8: Unemployment rates

Source: World Bank

GSMA Intelligence Country overview: Bangladesh

Figure 9: Stakeholder involvement in products and services

GSMA Intelligence Country overview: Bangladesh

The business model is not scalable

Quality of entrepreneur (average 2.93)

Business model (2.87)

Quality of idea (2.73)

Return on investment (2.79)

Country risk (2.13)