fairly easily and are dedicated to remaining as agents. However ... resources to buy enough ..... amount of transactions
Agent Network Accelerator Survey: Uganda Country Report 2013 January, 2014 Contributing Authors: Kimathi Githachuri, Mike McCaffrey, Leena Anthony Annabel Lee, Anne Marie van Swinderen, Graham A. N. Wright
1
Project Description Through the financial support of the Bill & Melinda Gates Foundation, MicroSave is conducting a four-year research project in the following eight focus countries as part of the Agent Network Accelerator (ANA) Project:
Africa
Kenya Nigeria Tanzania Uganda
Bangladesh India Indonesia Pakistan
Asia
Research findings are disseminated through The Helix Institute of Digital Finance. Helix is a world-class institution providing operational training for digital finance practitioners.
2
Focus of Research The research focuses on operational determinants of success in agent network management, specifically:
Quality of Provider Support
Agent & Agency Demographics
Core Agency Operations
Liquidity Management
Business Model Viability
3
The Research Is Based On 2,028 Nationally Representative Agent Interviews Data collection occurred in June/July 2013, using a random route methodology based on the displayed agent census.
Achieved Sample 373, 18% 540, 27%
Red points represent a census of agents conducted by Brand Fusion in 2013. Blue ones are the ones interviewed for this research.
1115, 55% Kampala
Non-Kampala Urban
Rural
Sample Profile* Key Providers*** A B C
Location Kampala
NonKampala Urban
322 102 66
838 208 167
Exclusivity Rural
392 96 117
Exclusive
1,364** 148 134
NonExclusive
188 258 216
Dedication Dedicated
NonDedicated
713 170 137
*Note this table shows results only for the top three providers. It sums to 2,308 as it represents all providers served by agents. i.e. if an interview was done with an agent serving three providers, it is counted three times in this table. ** Note the high concentration of Provider A exclusive agents (1,364) has significant impact on the overall findings. ***Provider names have been anonymized to maintain confidentiality.
839 236 213
4
Uganda Overview The quantity of agents in Uganda is expanding rapidly and profits are high; however, the next phase of development needs to focus on the quality of service provided.
In general agents are satisfied with their profits, rebalance fairly easily and are dedicated to remaining as agents. However, agents primarily face challenges on fraud and security, communication with their provider, and system downtime
5
Providers’ Market Share Of National Agent Network Warid 17%
Airtel 14%
MTN 63%
Kampala City AMS
Ezee Money is the first third-party provider. It offers ALL mobile money services and aggregates them using terminals
Non-Kampala Urban AMS Others Warid 6%
Others Warid 7% 19% Airtel 13%
During this research period, Airtel and Warid announced their acquisition and merger plans
Others 6%
Others Warid 6% 16%
15% MTN 61%
Airtel 17%
Rural AMS
MTN 62%
Airtel 13%
MTN 65%
Agent market share is defined as the proportion of cash-in/cash-out (CICO) agents by provider.
6
Daily Transaction Levels* Show Healthy Business For Agents Total Transactions Per Day
Percent of Respondents
30%
Median Transactions Per Day
The most common band is 21 to 30 transactions a day
25%
Kenya Tanzania
46 31
Uganda
30
20%
50% of agents are now doing at least 30 transactions a day
15%
10%
This value shows growing market maturity
5%
% 130+
121-130
Kampala Non-Kampala Urban Rural Total * Numbers represent transactions per day by selected provider, not overall volumes for the agency.
111-120
101-110
91-100
81-90
71-80
61-70
51-60
41-50
31-40
21-30
11-20
1-10
Number of Transactions
7
Largest Stated Barriers To Daily Transactions* 6.00
5.00
Many locations in Uganda seem saturated
This represents opportunities for product development and marketing communication
Rank
4.00
3.00
2.00
1.00
0.00 Too many other Lack of Individual Lack of Too often have Doing more Too busy to do agents resources to buy clients demand awareness of only either cash business means anymore competing for enough float for service is not service among or e-float when too much more business business very regular potential the client is risk of fraud or (already have customers in the asking for other robbery lines of clients) area * These scores are weighed averages of rankings, so that higher scores represent dimensions receiving a higher ranking.
8
The Lack Of Offerings Means Potential For Product Innovation Products And Services Offered In The Country 120%
Percentage of Respondents
100%
Agents report the process takes too long, and many are not even sure what the commission earned is. This indicates a very 99%
100%
high level of agentassisted OTC transactions processed from the agents’ phone (direct deposits).
80% 60% 40%
33%
30% 17%
20%
Banking services (credit, savings, insurance) are practically absent.
17% 1% Welfare/Social
Savings deposits to a bank
Insurance
Products and Services
Credit
Airtime top-up
Bill payments
Money transfer
Cash-out (withdrawals)
Cash-in (deposit)
Account opening
%
9
The Vast Majority Of Agents Are Profitable* Profit Per Month
35%
Percent of Respondents
30%
Median Monthly Profit ($US)
11% of agents are making a loss. 24.1% 23.0%
25%
40% of agents are making at least US$100 of profits a month
20%
Kampala Non-Kampala Urban Rural
86
Total
78
11.0% 10%
8.4%
5%
4.0%4.0% 2.3%2.3%
2.5%
0%
74
The high profitability in rural areas is a combination of health revenues with low OpEx.
14.8%
15%
78
.4% .4% .8% .6% .2% .0% .6% .0% .0% .2% .0% .2% 950.01+
900.01-950
850.01-900
800.01-850
750.01-800
700.01-750
650.01-700
600.01-650
550.01-600
500.01-550
450.01-500
400.01-450
350.01-400
300.01-350
250.01-300
200.01-250
150.01-200
100.01-150
51-100
1-50
Breaking even
Making losses
Profit in (US$) Kampala
Non-Kampala Urban
Rural
Total
* Is calculated by subtracting expenses from total earnings from all providers served.
10
Risk of Fraud Most Burdensome To Agents’ Business* 6
Biggest Agent Management Issues
First and third most prevalent issues are security related, with the risk of fraud being the biggest concern
5
Ranks
4
Although customer service does not often make the priority list for providers, it certainly does for agents
3
2
1
0 Risk of fraud
Dealing with Threat of Armed Not making Time spent Time spent on Time spent in customer service robbery enough money to teaching float training from when something cover costs customers about management service provider goes wrong the product 11 * These scores are weighed averages of rankings, so that higher scores represent dimensions receiving a higher ranking.
Comparison Of Median Profits* By Existing Dimensions 100 90
89
Median Profit (US$)
80
This is an interesting finding considering a majority of agents are female
The absence of differences along these attributes is intriguing. For Exclusivity specifically, it seems that the highest performing agents stay exclusive, balancing the fact that non-exclusive agents receive commissions from more than one source.
78
78
78
78
Dedicated
Non-Dedicated
Exclusive
Non-Exclusive
70 53%
60
58
50 40 30 20 10 0 Male
Female
* Is calculated by subtracting expenses from total earnings from all providers served.
12
Median Profits* Lowest In the First Year of Operations 140
Years Of Operations & Profitability
We expect the rate of increase to taper over time as the market matures
120
Median Profit (in US$)
100
80
Agencies in their second year of operation can expect to make 81% more profits
78
117
117
50%
60 43 40
81%
20
0 Less Than One Year
1
2
* Is calculated by subtracting expenses from total earnings from all providers served.
3 Or More Years
13
More Than Half Of Agencies Have Been Operating For Less Than One Year Years Of Operation Of Agencies
The amount of new agencies was quite surprising and could represent robust growth and/or a high agency churn rate
3 or more years, 9% 2 years, 12%
Less than 1 year, 52%
1 year, 27%
Given 78% of agents interviewed predicted they would continue with the business next year, it seems more likely that the majority of this is growth
14
Median Monthly Revenue Is Well Above GNI Per Capita
Median Revenue (US$)
Median Revenue In US$* 180 160
156
140
Agencies in urban areas outside Kampala are earning 16% lower than the country average.
136
136
117
120
Monthly GNI per capita = US$95
100
80 60 40 20 0
Kampala
Non-Kampala urban
Rural
Total
*Revenue reported here pertains to all providers being served. Therefore for non-exclusive agencies, their total revenue is reported here with regards to all the providers they serve.
15
Agents Report Diverse Operating Costs, From Almost Nothing To Well Over 100 US$ A Month 30.0%
Operational Expense Per Month
Operational costs are mainly driven by rental, wages and rebalancing costs
25.0%
Percent of Respondents
Median: Kampala: Non-Kampala Urban: Rural: Total :
78 US$ 58 US$ 52 US$ 58 US$
20.0%
20% of agents in Kampala are paying US$ 100 or more in monthly operational expenses
15.0%
10.0%
5.0%
.0% 0 - 19
20 - 39
40 - 59
60 - 79
Kampala
80 - 99 100 - 119 120 - 139 140 - 159 160 - 179 180 - 199
Non-Kampala Urban
Rural
Total
200+
16
Surprisingly Many Agents Have Balanced Needs For E-Float & Cash Regions: Central
About The Same, 44%
Cash, 24% EFloat, 32%
Northern
About The Same, 35%
Cash, 7%
EFloat, 57%
Eastern
About The Same, 50%
Cash, 19% E-Float, 31%
Western
About The Same, 49%
Cash, 17%
Except for the Northern region the majority of agents in each region are reporting a balanced need for e-float and cash This may represent a general misunderstanding of agent needs, as many providers assume that predominantly rural regions, such as the North, have more need for cash than efloat.
E-Float, 34%
17
Rebalancing Is Easy For Agents In Terms Of Time And Money Time Taken To Nearest Rebalance Point (Minutes) Kampala
Non-Kampala Urban
Rural
National Average
< 5.00
23%
34%
22%
29%
5.00-14.99
39%
46%
40%
43%
15.00-24.99
24%
11%
22%
16%
25.00-34.99
13%
5%
10%
8%
35.00-44.99
0%
2%
2%
2%
45.00-54.99
1%
1%
2%
2%
55.00+
0%
0%
2%
1%
Non-Kampala Urban agents have more convenient rebalancing points relative to the national average
72% of agents spend less than 15 minutes travelling to their nearest rebalancing point
9% of agents responses were not included here as they reported they did not travel to rebalance.
Agents tend to pay little or nothing to rebalance: 66% have costs of less than US$1
18
Frequency Of Monthly Withdrawals Seems Low
Percent of Respondents
The majority of agents withdraw cash five times or Monthly Cash Withdrawals For Rebalancing 18% less per month. 16% 14% 12%
A significant minority rebalances daily
Monthly cash withdrawal behaviour seems to be similar across all geographies
10% 8% 6%
4% 2% % 0
1
2
3
4
5
6
7
8
9
10 11 12 13 15 16 17 18 20 21 22 23 25 26 27 28 30
Number of Withdrawals Kampala
Non-Kampala Urban
Rural
19
Three Transactions Are Denied Each Day Due To Lack Of Float Transactions Denied Due To Unavailability Of Float
30%
Percent of Respondents
27% 25%
Only 27% operate consistent and optimal levels of float
Median:
This is equivalent to 10% of average daily transactions being lost across the country
Kampala: Non-Kampala Urban: Rural:
3 3 3
National Median:
3
20% 16%
15%
15% 11%
35% of agents lose at least five or more transactions per day due to lack of float
10% 7%
5%
5% 5%
3% 1%
3% %
%
0
1
2
3
4
5
6
7
6%
8
Number of Transactions Denied Per Day
9
10
More Than 10
20
Unpredictable Demand Seen As Greatest Impediment To Float Management* Much of the predictable behaviour is not effectively This is an indication that communicated by providers to the quality of agents 6.00 agents recruited have inadequate resources (poor selection) 5.00
Rank
4.00
This may be indicative of the challenges providers face in master agent recruitment and training
3.00
2.00
1.00
0.00 Unpredictable fluctuations in client demand
Lack resources in Have to shut store The cost incurred Time taken at Rebalance points Travel time to general to buy a to go get more is too much to do it rebalance point is often do not have rebalance point is sufficient amount float frequently too long cash/float too long that will last
* These scores are weighed averages of rankings, so that higher scores represent dimensions receiving a higher ranking.
21
Improving The Quality Of Agent Support Presents A Large Opportunity For Providers Training 94% report receiving training 46% from a provider 13% from master agent 45% from an employer 57% of agents have never undergone refresher training Operational Support: Only 33% of agents were visited by the provider, whereas 46% of agents report not being visited at all. Of those who were visited 35% report they were with no fixed frequency. Call Centre: 93% of agents were aware of a call centre and rated it a 4.6 out of 7 in terms of its ability to resolve its issues.
22
Recurrent Service Downtime Is Affecting Transaction Levels
Unreliable service is a challenge for most agents:
92% of agents report having experienced downtime in the past. Huge differences in providers, with some performing considerably better than others Only 48% of agents report receiving prior warning about downtime, however two thirds report that information given is inaccurate
There is a great reported variation per provider on amount of transactions lost per occurrence of server downtime with the median reported to be 10 transactions per day.
23
Outstanding Attributes Of Agent Network Management
Uganda is one of the most developed digital finance countries in the world with a robust agent network: Daily transaction volumes are high from Kampala through to the rural areas surveyed Healthy profits and revenues Strong competition and evidence of market saturation in some areas Cost-effective rebalancing systems Most agents intend to continue with the business
24
Opportunities For Improvement Agents are struggling with a number of issues. Here are some suggested areas of focus:
Large challenges on fraud and other security issues Lack of product diversity especially on the agent-level as the large majority of transactions are CICO Service down-time is greatly limiting number of transactions per day Customer service support is not adequately attended to Agents encounter frequent direct deposits (agent-assisted OTC transactions) as shown by the money transfer incidence High incidents of newly operating agencies needing close support that is not forthcoming Most agents close to rebalancing points resulting in limited geographical reach and financial inclusion Agent selection seems to be a significant challenge and barrier to increased transactions
25
Thank You www.helix-institute.com
[email protected] Helix Institute of Digital Finance Helix Institute 26