Online Lending – a ‘Dynamics’ chat with Niraj Verma

Online lending in India, has so far been limited to P2P lending. It is a platform, where the borrower and lender come together to meet mutual requirements, and the platform (unlike a bank), only charges a fee for the services. The ROI on loans is reported to be as high as 36% in some cases.

Online lending is at a very nascent stage in India. About 20 P2P online lending firms are reported to have sprung up in 2015, but they do not occupy prominent visibility in the market. The RBI did publish a consultation paper in April, 2016, classifying them as NBFCs for regulatory purpose, but the overall direction remains hazy.

We continue the discussion with Niraj Verma on Dynamics, to explore the world of online lending, and the risk/dangers/limitations involved.

Previous discussion:


Hi, Niraj … Welcome back. I noticed a rather sceptical view on digital banking, in both online and offline discussions on the subject, after our last chat. Perhaps, it stems from an ingrained fear of transparency. It could also be the increased risk of frauds with enhanced transparency. Online lending, thus, becomes a very sensitive subject.

I have read some very technically informative blogs by you on the subject. Can you brief us on the retail lending scenario in India?


Let us begin with a brief recap of evolution of lending systems in general.

All lending transactions need a lender, a borrower and a touchpoint. In the pre-industrial era, peasant borrowers would seek out a moneylender in the village for loans. It was impossible to borrow from a remote lender. Moneylenders, too, were comfortable in dealing with local borrowers. Lender-borrower relations were based on personal equations. The State had a negligible say, and the village was the only geographical touchpoint.

Later, brick-and-mortar banking offered an additional, reasonably priced avenue. The geographical constraint remained, and institutional lending was regulated, formal and impersonal. Yet, it managed to erode the influence of moneylenders.

Implementation of Core Banking System (CBS) and Loan Origination System (LOS) technologies provided an interesting twist. Banks offered loans to a target market of pre-qualified potential loan applicants. These systems use two touchpoints – telecom/Internet for the preliminary offer, and a brick-and-mortar branch for processing the loan application.


With this background, how do you visualize the future of online lending shaping up in India?


Online lending systems are part of the ongoing Fintech revolution. Strictly speaking, FinTech simply refers to a technology platform that connects consumers and providers of financial services. Types of financial services supported, depend on design of the platform.

Online lending platforms use cyberspace as the primary touchpoint. These platforms

  1. Empower credit seekers to explore more lending options, with less geographical constraints.
  2. Enable non-institutional lenders like HNIs to participate in the loaning process and
  3. Help institutional lenders to expand their outreach at minimal incremental costs.

Incidentally, personal details of credit applicants need not be, and should not have public exposure on the platform, without the applicant’s explicit consent. Such information should ideally be disclosed to a P2P lender only after a firm credit offer has been made and accepted, or to an institutional lender on expression of interest. Privacy concerns should override any pretense to total transparency.

Fintech opened its innings with money transfer and payment applications. Crowd-funding and online lending applications appeared later. Initially the FinTech platforms were designed to support P2P lending to salaried employees and professionals only. The current trend, however, is to enter into collaborative partnerships with institutional lenders as well for increased traction and accelerated growth.

One would have expected the FinTech companies to move on to MSME and Agri Business/Rural clientele. However, progress is slow, with the lack of predictable cash flows, a paucity of reliable credit rating agencies and consequent problems in designing an appropriate scoring system.

Another major problem with online lending systems, is the lack of user interfaces (UI) in local languages, and it excludes a large non-English speaking population.

Fintech platforms will take some more time cater to MSME and Agri Business/Rural segments.


Where are we on detection and prevention of frauds? Are we capable of tracing a defaulter online, if s/he has disappeared from the recorded address? Since security systems are not fool proof, will changes be necessitated in the law?


A comprehensive lending system has three components – Loans Origination System (LOS), Loans Management System (LMS) and NPA Management System (NPAMS). LOS typically starts with lead generation and ends with disbursal. LMS take over after disbursal and is usually limited to interest-computation and transaction-processing. NPAMS takes over after a loan turns bad, and encompasses the complete cycle of repayment reminders, negotiations, re-amortization, compromise settlements, interest funding, interest remission, rehabilitation, legal proceedings and write-offs.

Some fintech companies, reluctant to enter transaction processing, limit the online lending functionality to a subset of LOS, which stops just before disbursal. Though, most companies cover the entire LOS functionality and a part of LMS :

  • Registration of Lenders
  • Registration of Credit Applicants
  • Posting of Credit Request by Credit Applicants
  • Pre-approval of Credit Request
  • Pre-verification of Credit Request
  • Publication of Credit Requests
  • Posting of Credit Offers by Lenders
  • Sharing Credit Offers with Applicant
  • Loan Documentation
  • Loan Disbursal
  • Repayment Transactions

The ‘Pre-approval of Credit Request’ sub-process checks the submitted data to ensure that the applicant profile matches the one specified by some institutional lender, or as deemed acceptable by a majority of P2P lenders. The ‘Pre-verification of Credit Request’ sub-process next, checks the veracity of the submitted data. This sub-process includes both online and offline checks. The risk of identity theft (a major cause of online frauds), thus stands minimized. TAT, though, would depend on the number and nature of checks.

Chances of detection of a defaulter are actually higher than that in the traditional banking model, as a  software routine can be used to run a periodical check on all related social media profiles. The LMS functionality can be made more robust with analytic and predictive functionality, coupled with offline checks.

Detection of fraudulent diversion of funds, though, is possible only during field inspections, and hence is out of the purview of online lending systems.

This is only about design and implementation of processes that minimize risk of default.  It reminds me of an interesting conversation with a consultant. She is in discussion with some prospective clients who wish to launch a fintech platform, for the same client segment that is normally served by NBFCs and MFIs. Her apprehension was that finding credit ratings from credible rating institutions would be rather difficult, as  such clients play havoc with the standard scoring system. I suggested a variant of Self Help Group (SHG) concept, to introduce an element of distributed and shared liability. Another way could be to create a pseudo credit history based on payments of utility bills. The point is that a customized solution can be devised to reduce risk.


Great insights! Expanding on the same concept, which type of loans do you think, can be covered with just a future projection of information and resources available today? Do you visualize a system for consortium lending to industry being developed online?

Online lending platforms can be classified in three major categories:

  1. Pure P2P platforms
  2. Platforms promoted by an institutional lender as an extension of its other platforms and
  3. Hybrid platforms that can support P2P lenders as well as one or more institutional lenders.

The pure P2P platforms, for reasons enumerated earlier, accept personal loan applications from salaried employees, and a few professionals, only. The functionality can be easily extended to cover loan requests for purchase of consumer durables and vehicles, by the same client base, provided a legal framework on creating charge on financed assets is developed first. Small business loans to MSMEs may also be handled if the platform incorporates functionality like ratio analysis, cash/funds flow generation & analysis, working capital assessment by 1st/2nd method of lending, what-if analysis etc. to enable a P2P lender as well make an informed credit decision.

In case of an institutional lender, whether on an owned or a hybrid platform, theoretically all loan products can be supported, given that the detailed credit appraisal will be done offline. Such offline appraisal should also take care of qualitative assessment requirements. The only major issue will be handling LMS processes for a running credit facility.


Right. I was thinking about the same issue of a running credit facility. Maybe, some ‘if this, then that….’ logic can be infused in the existing transactional functionalities.


One possible solution could be issuing smart cards to avail running limits, and routing the transactions through the platform. The scoring systems, designed for the salaried employees and professionals, will not work for other segments. A new scoring system would be needed.

Consortium finance can very well be handled thru’ a fintech platform, which has all consortium lenders as its members. Some additional functionalities would of course, be needed.

  1. Secure channel of communication between the FIs to allow formation of a lenders’ consortium to finance a project,
  2. Functionality to handle several credit facilities sanctioned and disbursed by more than one FI linked to a single credit request,
  3. Functionality to download transaction data from a 3rd party CBS & to upload transaction data to a 3rd party CBS,
  4. Functionality to accept, as upload MSOD & QIS, data for onward transmission to the lender FIs in the consortium and
  5. Push notifications from one lender FI to other lender FIs in the consortium.

On account of high amount of funds involved and multiplicity of parties, many more offline activities related to pre-sanction verification, negotiations, field inspections etc. will happen outside the purview of online lending system.


What is the overall outlook for the future?        


Online lending is still a work-in-progress. Customer acquisition has emerged as a major problem for pure P2P lending platforms. Single FI sponsored platforms would just end up as an extension of their LOS systems in cyberspace. Hybrid platforms, that support multiple FIs and P2P lenders operate in the same loaning space, may be more successful in the years to come. It will be even more interesting if some such platform were to allow preapproved dealers of consumer durables and vehicles to come on board as vendor-members. Ad revenue from such vendor-members would also help reduce the transaction costs.


IMO a lot is needed. of work by the regulators, FIs and application developers, before we can even think on online credit system.

Links to Niraj Verma’s blogs on the subject:



2 thoughts on “Online Lending – a ‘Dynamics’ chat with Niraj Verma

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s