Figure 1: Percentage (%) of MFI Loans Referred By Agents
Across Rural and Urban Areas in Five Clusters (MFIN-Sponsored NCAER Study)

Now, please note that in Hyderabad (which had a reasonable number of MFI loans) and Jaipur (which had a small sample of MFI loans), agent-referred loans constitute a significant percentage. However, in Chennai, according to the NCAER study, agents are completely absent and hence, there are no
agent-referred loans.

A comment is in order here. I am glad that the MFIN-sponsored NCAER study talks about the significant presence of agents (at least) in Hyderabad. Interestingly, it also cites these agents as one of the major reasons for client-level resentment with MFIs. It has taken almost a year after the crisis for the Indian microfinance sector to get some official acknowledgement about the presence and negative impact of agents in Indian microfinance. I have all along been talking about agents – well before the 2010 microfinance crisis in Andhra Pradesh—as I had seen them operate on the ground^v . That said, let us look at what other stakeholders have said with regard to the use of agents in Indian microfinance, especially in Andhra Pradesh.

N Srinivasan noted in the State of the Sector Report (2010),

“The pressure to achieve performance targets and breakeven within a short period of time has pushed the relatively new staff of MFIs to look to centre leaders who are in the know of MFI operations. These centre leaders have become a critical rallying point and are today termed as ‘ring leaders’. In state after state (Madhya Pradesh, Rajasthan, Orissa, West Bengal, Andhra Pradesh, Karnataka and Tamil Nadu), stories abound of how ring leaders informally register new customers promising loans for a fee. Most new MFIs setting up operations in such areas approach these centre leaders as an easy and natural entry point. This provides the necessary influence to the ring leaders to deliver on the promise made to several registrants for loans. The centre leaders are also in a position to obtain loans in the name of others, advantageously using the relative unfamiliarity of new field staff and new MFIs. The resultant ghost loans have a tendency towards default. The clients that pay the registration fee in order to get a loan feel justified in holding up repayments. This behaviour has an adverse effect on repayment rates and necessitates stronger recovery efforts. Some MFIs (including those in the list of top 10) had to wind down operations in some pockets of states such as West Bengal, Chhattisgarh, Rajasthan and Maharashtra without making an attempt to consolidate.”^vi

Likewise, Microfinance Focus wrote (22 December 2010):    

“Moulding business models to meet their growth targets, some of the largest microfinance institutions are using group leaders as interface agents between borrowers and loan officers. Popularly called as ‘Ring Leaders’, these agents are responsible for conducting meetings in their premises and collecting weekly repayments from the borrowers. Borrowers of microfinance institutions in townships of Mehndipatnam, Begumpet and Dilkhushnagar of Hyderabad (capital of Andhra Pradesh) told (the) microfinance focus team that now these ring leaders have become a major cause of distress for them. The principle of ‘Know Your Customer’ is one of the keystones around which microfinance practices have been evolved. However, with the introduction of the ‘ring leaders’ into the process, it seems that this essential requirement of lending is being compromised. The end borrowers interact with the ring leaders who maintain their passbooks and repayments. The loan officers, in turn, collect these from the ring leaders, reducing the amount of their interaction with the borrowers to almost negligible levels. In fact, as microfinance focus learnt, the interactions between the borrowers and the loan officers have reduced to such alarming levels that the borrowers are not even aware of which microfinance institutions they have taken the loan from! Another disturbing practice which came to light was the charging of ‘membership fees’ by the ring leaders from the borrowers to join an MFI group. Ranging in the amounts of Rs300- Rs500, this membership fee is over and above the interest paid to service the loan. This fee was pocketed entirely by the ring leaders and is their ‘commission’ for allowing a prospective borrower to be part of the group. “Ring leaders have become a major cause of distress for us but as we need money and don’t have any better sources, we give in to their demands”, one of the borrowers said.”^vii

Similarly, Daniel Rozas and Karuna Krishnasamy have argued that,
 
“But besides heavy presence of multiple borrowing, an even more insidious factor is the extensive presence of loan agents, who act as intermediaries between the clients and the MFIs. Of the six borrowers interviewed, two were agents/group leaders. To make matters worse, both of the agents were also heavy multiple borrowers (with 3+ loans), and at least in one woman’s case, these multiple loans all came from the same MFI, with the agent having used fictitious borrowers for the purpose. Such a situation further increases the agent’s incentive to default on her own loans, while influencing her group members to default on theirs. In a couple of cases, the agents had such a dominant role that the borrowers didn’t even know which MFI their loans had come from. For an MFI seeking to restart collections, this makes the matter that much more difficult. In effect, the MFI will be forced to bargain with the agent if it is to have any hope in collecting not just that agent’s several loans, but also those of the group members.”^viii 

Taken together, all of these suggest that agents have now arrived as intermediaries in Indian microfinance and they are being used in the decentralised microfinance model that is currently still in vogue. While there could be many reasons for use of agents, the aspects of “building scale quickly”, the “pressure to reduce interest rates” and the “desire to be a cost leader and maximise profits and value to shareholders” appear to be the major ones for their (increasingly widespread) use in the Indian microfinance industry. Therefore, given the above discussion, it seems difficult to accept the findings— from the MFIN sponsored NCAER report—which seems to imply that there are no agents in Chennai (Tamil Nadu). But hold your breath, the next article in this series provides concrete evidence (from the horses’ mouth) with regard to presence and use of agents by MFIs in Tamil Nadu (including Chennai).