The Colloquim was conducted by Arusha Cooray from Australia.
Joint study by Arusha Cooray, Nirmali Sivapragasam and Harsha de Silva
- Remittance inflows into the developing economies have increased tenfold from US $31,058 million to US$ 327,591 over the 1990 to 2008 period
- Initiatives taken by policymakers of developing economies to promote the use of information and communication technologies (ICTs) in particular mobile phones specifically targeting individuals outside the reach of traditional banking services.
Benefits of ICT are well documented by the literature.
Jensen (2007) – the use of mobile phones by fisherman contributed to a fall in price dispersion and reduction in waste, leading to increases in consumer and producer welfare in Kerela
de Silva (2005) – mobile phone based ICTs can reduce information asymmetries in the agricultural market, enabling farmers to reduce transaction costs and search costs associated with locating sales outlets.
Sivapragasam, Aguero and de Silva (2010) – the use and awareness of the mobile phone as a mode of remitting money. They find that, respondents from higher income and education brackets, and also those who have access to personal bank accounts are more likely to be aware of the mobile phone as a remittance transmission device.
A binomial logit model is estimated to investigate the awareness of the mobile phone as a mode of remitting money.
Ai = X1ib1 + e1i
where Ai = awareness of the mobile phone by individual, i
X1i = vector of control variables
e1i = a random error term
A = 1 if respondent is aware of the mobile phone as a mode of remitting money (A >0) and A = 0 if the respondent is not aware of the mobile phone as a mode of remitting money
A logit model is estimated to investigate the potential use of the mobile phone:
Ui = X2ib2 + e2i
Ui = potential use of the mobile phone by individual, i.
X2i = is a vector of control variables
e2i = a random error term
U = 1 if the respondent will potentially use the mobile phone as a remittance transmission mode (U >0) and U = 0 if the respondent will potentially not use the mobile phone as a remittance transmission mode.
Third step of awareness
•A multinomial logit model that simultaneously investigates both:
(1) awareness and (2) potential use of the mobile phone as a mode of remitting money.
The first two stages of the estimation can then be combined involving three possibilities:
Y = 0: Ai ≤ 0, Ui ≤ 0 (the individual is not aware of the mobile phone as a mode of remitting money, the individual does not use the mobile phone as a mode of remitting money)
Y = 1: Ai > 0, Ui ≤ 0 (the individual is aware of the mobile phone as a mode of remitting money, the individual will potentially not use the mobile phone as a mode of remitting money)
Y = 2: Ai > 0, Ui > 0 (the individual is aware of the mobile phone as a mode of remitting money, the individual will potentially use the mobile phone as a mode of remitting money)
Next, the three categories are arranged in descending order from a welfare point of view. The welfare ranking –
- aware and will potentially use (Y = 2)
aware but will not potentially use (Y = 1)
not aware and do not use (Y = 0).
Nirmali : Survey data on information and communication technology (ICT) use at the “bottom of the pyramid” (BOP) conducted by LIRNEasia in 2008 and 2009.
Bangladesh, Pakistan,, India, Sri Lanka, The Philippines* and Thailand:
1500+ quantitative purposive sample
HG: Why were local migrants excluded?
NS: Decision taken
In the Reasons that Have Prevented Use of Mobile Phone as Money Transfer Mechanism graph
HS: Sample for the overseas worker is around 1000 and almost all is using money transfering services.
AZ: Something wrong with the data
Anusha: People use more than one transmission method.
Comparative Percentages– only some of the percentages are presented as table too long
Results show that age, age squared and education are statistically significant for awareness.
For potential use a age, age sq and frequency with which money sent is statistically significant.
SL: Bank account ownership and deposit money into bank account. any additional value in having both?
Arusha: Is it better to take one out and rewrite it?
1 question is whether you own a bank account and whether you are transmitting money in a bank account?
HG: Just do internal tests to clarify it.
Multinomial Logit Estimates are in two slides.
Aware of the Mobile Phone as a Remittance Mode but Will Potentially Not Use the Mobile Phone as a Remittance Mode only age and age sq appears to be significant.
Aware of the Mobile Phone as a Remittance Mode and Will Potentially Use the Mobile Phone as a Remittance Mode
only Frequency of Communication, Frequency with which Money Sent, Deposit Money into Bank Account are significant.
Aileen: what the number of observations?
Arusha: 1200 interviewed, will check on it
- Older respondents are more likely to be aware and use the mobile phone as a mode of remitting money, however, that this effect weakens in the higher age groups.
- The higher the level of education of the respondent, the greater the likelihood of being aware and using the mobile phone as a remittance transmission device.
- Income is not an important factor determining the awareness or potential use of the mobile phone.
- The frequency of communication is important for the awareness and potential.
- The frequency with which money is sent home has a positive impact on awareness and use.
- The use of a bank account as a remittance transmission channel has a negative impact on awareness and use.
People were satisfied with current mode of transmission so was not looking for alternatatives.
This result is different to study done by de Silva, Nirmali and Aileen.
HG: explain Older respondents are more likely to be aware and use the mobile phone as a mode of remitting money, however, that this effect weakens in the higher age groups. what is older? what is higher?
So there is a peak age, not a linear relationship?
HD: Bank account ownership is positive influence on awareness but your less likely to use mobiles to remit money.
An inverse relationship. in new study bank ownership has a negative influence. So we need to check the need for both bank acc variables.
concerned about the income co efficient. expenditure is better estimate of income at this level because poor people will want to over estimate their income.
AZ: which of the 2 models have more robustness?
Arusha: not sure of
NS: r aquared is almost same.
HD: income is difficult to change so leave the income and focus on education and aware. we know for sure if more aware more usage so thats the what we can use for policy
SL: not comfortable with the literature review of the paper, very generic not the correct reading for an awareness papers.
HG: remittance literature is relevant not ICT4D papers.
Aileen: what do you mean welfare point of view?
Arusha: it is quick method and increase saving so there is welfare.