Digital is trans­form­ing Lend­ing De­ci­sions

Banks and fi­nan­cial ser­vices in­sti­tu­tions can now take easy lend­ing de­ci­sions. Dig­i­ti­za­tion has not only trans­formed the de­ci­sion-mak­ing process, it has also changed the way risk man­age­ment is han­dled in these in­sti­tu­tions

Banking Frontiers - - Contents - Mo­han@bank­ingfron­tiers.com

Banks and fi­nan­cial ser­vices in­sti­tu­tions can now take easy lend­ing de­ci­sions. Dig­i­ti­za­tion has not only trans­formed the de­ci­sion-mak­ing process, it has also changed the way risk man­age­ment is han­dled in these in­sti­tu­tions

Banks and fi­nan­cial ser­vices in­sti­tu­tions are reeval­u­at­ing their de­ci­sion-mak­ing pro­cesses in the re­tail loan seg­ment with the ad­vent of dig­i­ti­za­tion and avail­abil­ity of data. Digital trans­for­ma­tion in credit risk man­age­ment as well as in the way in­for­ma­tion is be­ing made avail­able to take de­ci­sions bring greater trans­parency to risk pro­files. These in­sti­tu­tions are now able to ex­pand their busi­nesses through tar­geted riskbased pric­ing, faster client ser­vice without sac­ri­fice in risk lev­els and more ef­fec­tive man­age­ment of ex­ist­ing loan port­fo­lios. They can also ef­fec­tively meet grow­ing cus­tomer ex­pec­ta­tions us­ing strong data man­age­ment and ad­vanced an­a­lyt­ics.

“The dif­fer­en­ti­a­tion in prod­ucts and ser­vices of­fered by banks today are marked by their speed of de­liv­ery. For the nextgen cus­tomers, today is in fact yes­ter­day,” says Vi­jay Anandh R, head, Re­tail Risk at RBL Bank. “Only dig­i­ti­za­tion can ac­com­plish this. To a great ex­tent, In­dian banks are not lag­ging in dig­i­tiz­ing var­i­ous spheres of bank­ing ac­tiv­ity,” says he.

He adds: “I will de­scribe the year 2017 as the year of dig­i­ti­za­tion as far as banks and fi­nan­cial in­sti­tu­tions are con­cerned. There are sev­eral ini­tia­tives and ef­forts in this re­gard and all have to a great ex­tent im­pacted the way bank­ing is done. This is al­most akin to the sce­nario in the early 2000s when ATMs were first in­tro­duced in In­dia, which in fact marked the be­gin­ning of the new age bank­ing in the coun­try. At that time, only a priv­i­leged few could en­joy the ben­e­fit of this au­to­ma­tion, but sub­se­quently ATMs be­came one of the pre­ferred chan­nels for bank­ing. Sub­se­quently mul­ti­ple chan­nels evolved and today we can claim trans­ac­tions can hap­pen through var­i­ous chan­nels, de­pend­ing on the pref­er­ence of the cus­tomers. There has also been a ma­jor trans­for­ma­tion wherein

the role of di­rect sales agents is just wan­ing as the in­ter­ac­tions with cus­tomers are made pos­si­ble through var­i­ous ways and at dif­fer­ent lev­els.”

SCORE­CARD MOD­ELS

Govind Sankara­narayanan, COO, Re­tail Busi­ness and Hous­ing Fi­nance, Tata Cap­i­tal, says in the last 2 years, with the ex­plo­sion of the digital, more and more fi­nan­cial in­sti­tu­tions have rec­og­nized the im­por­tance of cre­at­ing digital-fo­cused/ digital-first fi­nan­cial prod­ucts to cater to the in­stant and vary­ing needs of the dig­i­tally-savvy cus­tomer. “With cus­tomers now spend­ing more time online, be it for bank­ing trans­ac­tions or online pur­chas­ing, the easy avail­abil­ity of non-tra­di­tional cus­tomer in­for­ma­tion through their so­cial media pro­file and other digital foot­prints has al­lowed banks and NBFCs to cre­ate unique score­card mod­els to carry out a credit assess­ment of an in­di­vid­ual. We re­cently launched a prod­uct along sim­i­lar lines called the myLoan app, which uses tra­di­tional and non-tra­di­tional data sets like a cus­tomer’s so­cial pro­file etc, to as­sess the credit wor­thi­ness of an in­di­vid­ual and af­ter giv­ing a #myS­core, gives a con­di­tional ap­proval for the loan amount.”

He says the gov­ern­ment’s ‘Digital In­dia’ ini­tia­tive is also lend­ing a help­ing hand to the fi­nan­cial in­sti­tu­tions by open­ing data through eKYC and other plat­forms, which will push more fi­nan­cial in­sti­tu­tions to­wards adopt­ing new-age credit assess­ment meth­ods.

CREDIT RISK MOD­EL­ING

Vi­jay Anandh fore­sees a sce­nario where core de­ci­sions in banks and fi­nan­cial in­sti­tu­tions will be made based on data an­a­lyt­ics. “There will also be a lot of em­pha­sis on credit risk mod­el­ing, which will en­able banks not just to avoid risks but of­fer credit to more peo­ple, who were hith­erto re­mained unat­tended. Avail­abil­ity of data and tools to process and take out ac­tion­able in­for­ma­tion from this data will also help banks to of­fer dif­fer­en­ti­ated pric­ing for their prod­ucts de­pend­ing on the cus­tomer pro­files. I am of the firm view that such a sce­nario will also help im­prove the bond­ing with the cus­tomer, which in turn will broaden the scope of cus­tomer ac­qui­si­tion. Nat­u­rally, this will jus­tify the end use of funds. This is not just for banks but for NBFCs as well, who will have bet­ter busi­ness op­por­tu­ni­ties,” he avers.

Ac­cord­ing to him, dig­i­ti­za­tion can also cre­ate a bet­ter sup­port sys­tem for fi­nan­cial in­sti­tu­tions and they can re­spond to cus­tomer queries in a more ef­fi­cient and faster man­ner be­cause all the in­for­ma­tion about a cus­tomer would be avail­able at one place. This, he says, will change the very role of call cen­ters and the agents at these cen­ters.

“An­other ma­jor ad­van­tage of dig­i­ti­za­tion is the op­por­tu­nity to cross­sell. For ex­am­ple, if a person has availed of a home loan from a bank or a NBFC, the con­cerned in­sti­tu­tion can, based on the credit pro­fil­ing of the cus­tomer, of­fer him a loan for buy­ing fur­ni­ture or house­hold goods,” says Vi­jay Anandh.

RISK-BASED PRIC­ING

Is it fea­si­ble for In­dian fi­nan­cial ser­vices in­sti­tu­tions to cat­e­go­rize bor­row­ers and of­fer risk-based pric­ing of loans?

Govind Sankara­narayanan feels yes. Says he: “Us­ing risk-based pric­ing, the bor­rower with a good credit score can get a lower rate of interest on any kind of loans. Risk-based pric­ing of loans is a sys­tem of of­fer­ing credit at a rate de­pend­ing on the cus­tomer’s credit score. For in­stance, lenders may of­fer a higher rate of interest to you if you are viewed as a higher risk bor­rower and your credit score is low as per the other banks’ lend­ing poli­cies. For the same loan amount, lenders are likely to of­fer a lower rate of interest if your score is high. These benefits thus en­sure that the cus­tomers work to­wards keep­ing their scores and credit wor­thi­ness high.”

Vi­jay Anandh is of the view that the sys­tem is yet to take off in In­dia. How­ever, banks should adopt this be­cause now in­for­ma­tion to seg­ment cus­tomers on the ba­sis of risks they pose - say prime, sub­prime or Alt A bor­row­ers - is avail­able. “Banks may choose to of­fer a higher rate of interest if a cus­tomer is viewed as a higher risk bor­rower, de­pend­ing on the banks’ lend­ing poli­cies. Sim­i­larly, bor­row­ers with lesser risk can get lower interest rate. Riskbased pric­ing helps both the lenders and bor­row­ers. This also leads to en­sur­ing that the cus­tomers al­ways at­tempt to keep their credit his­tory clean and we can pre­vent delin­quen­cies.”

Kal­pana Pandey, CEO amd MD, CRIF High Mark Credit In­for­ma­tion Ser­vices, too main­tains In­dian banks can adopt risk based pric­ing. “They can seg­ment cus­tomers us­ing credit scores and other in­for­ma­tion cap­tured through ap­pli­ca­tions. Based on the seg­ment char­ac­ter­is­tics and risk pro­file of the bor­rower, the terms of the loan of­fered could vary. For a seg­ment of cus­tomers with low risk could be of­fered a higher loan-to-value as well as lower interest rate, whereas for cus­tomers with higher risk a higher interest rate.”

She points out that Bank of Bar­oda has al­ready started do­ing it for home loans and many lenders do it for per­sonal loans. “It is high time when In­dian banks be­gin to in­cen­tivize cus­tomers with bet­ter credit score with bet­ter terms, and move away from a port­fo­lio-based pric­ing of loans. The lenders can max­i­mize prof­itabil­ity by ac­cept­ing more cus­tomers but at the same time, at pro­por­tional risk rates,” says she.

Har­shala Chan­dorkar, COO,

Tran­sUnion CIBIL, points out that credit in­sti­tu­tions in In­dia are grad­u­ally adopt­ing risk-based pric­ing of credit prod­ucts. “Some pro­gres­sive in­sti­tu­tions like Bank of Bar­oda have started of­fer­ing credit score-based lend­ing to re­tail mort­gage loan seek­ers, which in­volves pro­vid­ing dif­fer­en­tial rate of interest based on the bor­rower’s CIBIL score. Cus­tomers with a good loan re­pay­ment track record and strong fi­nan­cials may get loans which are at least 50-75 ba­sis points cheaper than a cus­tomer with a bad credit score,” says she.

AD­VAN­TA­GEOUS

She adds that risk based pric­ing would be ad­van­ta­geous in many ways. “For in­stance, if a lender starts of­fer­ing com­par­a­tively lower interest rate based on this model, it would at­tract bor­row­ers with good credit scores. This is sym­bi­ot­i­cally ben­e­fi­cial to both lenders as well as the bor­row­ers. The lender will be able to main­tain a good port­fo­lio of bor­row­ers with least prob­a­bil­ity of de­fault to avoid stressed as­sets and on the other hand the bor­rower will be able to make use of the lower interest rate which would fur­ther boost the ob­jec­tive of main­tain­ing a good fi­nan­cial be­hav­ior in all bor­row­ers.”

KEY ROLE OF BU­REAUS

That high­lights the role of credit in­for­ma­tion bu­reaus. Credit scores were one of the ma­jor tools for bankers in their de­ci­sion mak­ing. Has this changed? What do bankers look for in un­der­stand­ing the be­hav­ior of a prospec­tive loan seeker?

Chan­dorkar pro­vides an in­sight: “You know CIBIL Score has al­ways been an im­por­tant tool for lenders for as­sess­ing the risk pro­file of the bor­row­ers be­fore tak­ing any lend­ing de­ci­sion and it still re­mains so. The CIBIL Score and Re­port helps the lender as­sess three vi­tal pa­ram­e­ters of your fi­nan­cial health - credit lever­age, credit be­hav­ior and credit ap­petite. CIBIL Score and Re­port show­cases to the credit in­sti­tu­tions the bor­rower’s credit ex­po­sure like how many loans and credit cards has the con­sumer availed, what is the type of credit the con­sumer has availed - se­cured (home loans, car loans, loan against prop­erty) and un­se­cured (per­sonal loan, credit card) and what is the com­po­si­tion of se­cured vs. un­se­cured credit, how much is the to­tal amount of credit taken. The re­port also re­flects the bor­rower’s credit be­hav­ior on how he/she is ser­vic­ing availed loans and credit cards and his or her credit ap­petite how much credit has the con­sumer al­ready taken and what is the ex­ist­ing debt ex­po­sure and whether the con­sumer can man­age ad­di­tional credit based on his/her in­come to debt ra­tio.”

Pandey says the us­age of the credit score still plays an im­por­tant part in credit de­ci­sions. “Now, with mul­ti­ple credit bu­reaus be­ing avail­able, many lenders use credit scores from two credit bu­reaus to take a de­ci­sion. Most large lenders also have their own ap­pli­ca­tion score­cards, which take into ac­count in­for­ma­tion com­ing in ap­pli­ca­tion as well apart from the credit score. This in­for­ma­tion may in­clude in­come, monthly ex­penses, pos­ses­sions, em­ploy­ment etc. If the loan seeker doesn’t have a foot­print in the credit bu­reau, then the lender needs to de­pend on al­ter­na­tive data sources for pro­fil­ing him. In case the cus­tomer has an es­tab­lished credit his­tory in credit bu­reau, such al­ter­na­tive data sources are used to re­fine the cus­tomer’s pro­file,” she ex­plains.

LEND­ING DE­CI­SIONS

Pandey out­lines the fun­da­men­tal prin­ci­ples be­hind a lend­ing de­ci­sion – the classic 5 C ap­proach in­volv­ing assess­ment of char­ac­ter, ca­pac­ity, cap­i­tal, con­di­tions, and col­lat­eral, which he says con­tinue to gov­ern the lend­ing de­ci­sions. “What has evolved over last few years is how these are as­sessed, es­pe­cially the in­tent, iden­tity and the ca­pac­ity of the bor­rower. The data sources for such assess­ment have come up strong like the data en­riched and a ro­bust credit bu­reau en­vi­ron­ment than what it was 5 years ago, avail­abil­ity of pub­lic reg­istries, in­fu­sion of Aad­haar etc. Lenders are in­creas­ingly us­ing com­bi­na­tions of data points such as own data (like sav­ings bal­ances of cus­tomer), tra­di­tional credit data (such as credit bu­reau data), al­ter­na­tive data sources (such as mobile data, so­cial media data etc), in­ter­nal credit poli­cies and an­a­lyt­ics and score­card in the lend­ing de­ci­sion. Such data points are now be­ing used in real-time lever­ag­ing va­ri­ety of tech­nolo­gies to take lend­ing de­ci­sions in a few min­utes,” says she.

She adds that sev­eral lenders are also ex­per­i­ment­ing with the use of con­sent based data sources (such as Face­book pro­files, mobile data cap­tured through bank’s apps, spend­ing pat­tern on e-com­merce plat­forms etc) for bet­ter de­ci­sion for new to credit or thin file cus­tomers.

3 FAC­TORS

Ac­cord­ing to Chan­dorkar, lend­ing de­ci­sions at credit in­sti­tu­tions for re­tail credit are largely de­pen­dent on 3 fac­tors - in­come-todebt ra­tio, KYC de­tails and the bor­row­ers’ CIBIL Score and Re­port. These fac­tors have more or less re­mained con­stant over the last 5 years, she says.

MON­I­TOR­ING BOR­ROWER

Govind Sankara­narayanan speaks about what could be the ideal method­ol­ogy in mon­i­tor­ing a bor­rower: “The ideal method­ol­ogy that we fol­low in pro­vid­ing loan to the ap­pli­cant is to strictly eval­u­ate the ca­pac­ity to re­pay the loan amount along with right doc­u­ments. The eli­gi­bil­ity cri­te­ria dif­fer ac­cord­ing to the type of loan re­quested. For ed­u­ca­tion loan, the eli­gi­bil­ity is checked on the ba­sis of the fac­tors like

the stu­dent’s past aca­demic record, qual­ity and rep­u­ta­tion of the in­sti­tute where the stu­dent wants to pur­sue the cho­sen course, his­tory of co-ap­pli­cants, etc. For auto loans, we pro­vide fi­nance for both new and used cars, there­fore the cri­te­ria are dif­fer­ent in terms of in­come of the ap­pli­cant. In case of used car loan, for the bor­rower, there are few in­come re­lated eli­gi­bil­ity cri­te­ria. A salaried in­di­vid­ual needs to have a min­i­mum an­nual in­come of `3 lakh and should have been in em­ploy­ment for at least two years (of which, one year should be with the cur­rent em­ployer). For the self­em­ployed pro­fes­sion­als, the min­i­mum in­come should be at least `2 lakh and the person should be in busi­ness for at least 3 years. In our re­tail loan seg­ment, we also pro­vide busi­ness loan and to of­fer loan dig­i­tally, we are associated with Biz2Credit. The eli­gi­bil­ity process pri­mar­ily in­cludes min­i­mum of 25 years of age and max­i­mum 65 years of the ap­pli­cant, busi­ness should be in profit for three con­sec­u­tive fi­nan­cial years and turnover should be on a pos­i­tive trend.”

Pandey main­tains that the method­ol­ogy of mon­i­tor­ing a bor­rower dif­fers de­pend­ing on the na­ture of loans. “For mon­i­tor­ing, the de­sired end use is im­por­tant. Is it for manag­ing risk or cross-sell/up-sell or tar­gets for col­lec­tions and re­cov­ery? The se­lec­tion of cus­tomers, ap­proach of mon­i­tor­ing, fre­quency of mon­i­tor­ing and anal­y­sis of out­put may also dif­fer ac­cord­ingly. Watch­list mon­i­tor­ing called Alerts is de­ployed usu­ally for high-delin­quency cus­tomers or for very at­trac­tive cus­tomers sus­cep­ti­ble to poach­ing by com­pe­ti­tion es­pe­cially on un­se­cured prod­ucts. An off­line mon­i­tor­ing works well for pri­or­i­tiz­ing cus­tomers for col­lec­tions and re­cov­ery or for iden­ti­fy­ing whom to and what to cross-sell or to re­view ex­ist­ing credit lim­its or to as­sess health of en­tire port­fo­lio,” says she.

She is also of the view that lenders should be look­ing to use loan orig­i­na­tion and/or loan man­age­ment sys­tems, which are flex­i­ble and so­phis­ti­cated enough to man­age a va­ri­ety of an­a­lyt­i­cal score cards, credit poli­cies and de­ci­sion flows to sup­port full au­to­ma­tion in de­ci­sion mak­ing, at the time of cus­tomer on­board­ing, cus­tomer life cy­cle man­age­ment and col­lec­tions. Such sys­tems today are avail­able in SaaS (soft­ware-as-aser­vice) or li­censed mod­els, she adds.

“Mid-to-large and di­ver­si­fied lenders should also eval­u­ate de­ploy­ing an ef­fec­tive de-du­pli­ca­tion sys­tem to be able to iden­tify a cus­tomer uniquely across the prod­uct port­fo­lios,” se says.

Pandey men­tions that so­cial media be­hav­ior and its re­la­tion­ship with credit risk is yet to be seen in In­dia. “The us­age of so­cial media for a reg­u­lar In­dian cus­tomer is very dif­fer­ent from the west­ern world. Is post­ing pics etc. or fol­low­ing a thread good enough to as­sess a credit risk? How­ever, it can be used for ver­i­fi­ca­tion and check­ing con­sis­tency of in­for­ma­tion pro­vided by the ap­pli­cant to some ex­tent,” says she.

IN­FOR­MA­TION LINK­AGES

Vi­jay Anandh says there is whole lot of in­for­ma­tion link­ages like Aad­haar be­ing made manda­tory for in­come tax re­turns fil­ing or PAN be­ing made com­pul­sory in house pur­chases, or KYC nec­es­sary for open­ing a bank ac­count. This is go­ing to cre­ate a huge in­for­ma­tion pool and this in­for­ma­tion pool is now avail­able to fi­nan­cial in­sti­tu­tions, which can process this data at gran­u­lar lev­els and take de­ci­sions based on this data. This link­age in all the iden­ti­fi­ca­tion doc­u­ments, he says, will ben­e­fit the cus­tomer by elim­i­nat­ing the risk of fraud and bet­ter pric­ing on fi­nan­cial prod­ucts. “Dig­i­ti­za­tion is also go­ing to cre­ate a level play­ing field for banks in the coun­try, be­cause the data as­sets would be avail­able to all and they would all be mak­ing use of the same tools to process this data to base their de­ci­sions. As this sit­u­a­tion in­creases com­pe­ti­tion, it will be the cus­tomer who will be the fi­nal ben­e­fi­ciary,” he says.

But, this work is al­ready be­ing un­der­taken by credit in­for­ma­tion bu­reaus and banks and fi­nan­cial ser­vices in­sti­tu­tions have ac­cess to the pro­cessed data on cus­tomers from these bu­reaus. Be­sides, there is whole lot of pro­cesses in­volved in dedu­pli­ca­tion of the data, which the bu­reaus are in any way do­ing. So, it is a du­pli­ca­tion of work?

ONE VARI­ABLE

Vi­nay Anandh agrees that credit in­for­ma­tion bu­reaus are un­der­tak­ing the task of pro­cess­ing data on cus­tomers and banks de­pend on their prod­ucts for de­ci­sion­mak­ing. “But, these prod­ucts are just one vari­able, which add to the in­for­ma­tion fi­nan­cial in­sti­tu­tions cre­ate on their own about cus­tomers. To that ex­tent, we at banks im­prove our de­ci­sion-mak­ing. The ad­van­tage we will have by un­der­tak­ing data min­ing and an­a­lyt­ics is that we will have the ca­pa­bil­ity to un­der­stand the dif­fer­ence be­tween a low risk cus­tomer and high-risk cus­tomer. It will not just be de­ter­mined by a score pro­vided by the bu­reaus. As re­gards dedu­pli­ca­tion, there are el­e­men­tary pro­cesses like data scrub­bing which can take care of this. Of course, the work can also be out­sourced. The whole process of min­ing and anal­y­sis can now be com­pleted in just 10 steps,” says he.

Chan­dorkar says the RBI had man­dated all credit bu­reaus to pro­vide credit score in the range of 300- 900 where a score with propen­sity to 300 would im­ply high risk­i­ness of the bor­rower whereas, a score closer to 900 would mean low risk­i­ness of the bor­rower. Today, all bu­reaus in In­dia pro­vide credit scores on this range. How­ever, each bu­reau has its own pro­pri­etary al­go­rithm of score cal­cu­la­tion.

“Avail­abil­ity of credit in­for­ma­tion in­sights and so­lu­tions from Tran­sUnion CIBIL has sig­nif­i­cantly con­trib­uted to driv­ing growth in the re­tail credit seg­ment while fu­el­ing credit pen­e­tra­tion and fi­nan­cial in­clu­sion,” she says, adding “Tran­sUnion CIBIL Data Anal­y­sis re­veals that re­tail loans have grown at an av­er­age CAGR of 28% over the last 3 years while at the same time there is sig­nif­i­cant re­duc­tion in re­tail NPA rates. Credit in­for­ma­tion so­lu­tions have also helped re­duce the av­er­age time re­quired for ap­proval of a loan ap­pli­ca­tion from ap­prox­i­mately 7-9 days 3 years back to around 3-4 days today. Us­age of in­for­ma­tion so­lu­tions have con­trib­uted to the growth in ru­ral lend­ing - en­quiries on the bu­reau for ru­ral lend­ing was around 25% over 5 years back and has steadily grown to al­most 35% today. We would be evolv­ing and cus­tomiz­ing our so­lu­tions to the mar­ket dy­nam­ics for ably sup­port­ing the in­dus­try on credit pen­e­tra­tion and fi­nan­cial in­clu­sion.”

NEW PROD­UCT

Chan­dorkar also says credit bu­reaus are today of­fer­ing prod­ucts, which are be­yond the tra­di­tional credit score, and which of­fer clear in­sight into a cus­tomer’s be­hav­ior over a pe­riod of time - say 5 years. “For ex­am­ple, we have re­cently launched Cred­itVi­sion to ex­pand the el­i­gi­ble con­sumer base and drive ac­cess to fi­nance for many more de­serv­ing con­sumers. Cred­itVi­sion is a trans­for­ma­tional way of look­ing at the past credit be­hav­ior of a con­sumer. Cred­itVi­sion al­go­rithms pre­dict risk and ex­pand credit op­por­tu­ni­ties by in­ten­sively study­ing the trended data that un­locks the pat­terns in pay­ment, ex­po­sure and spend be­hav­ior. These al­go­rithms, which are based on the past 36 months of trended data, en­able iden­ti­fi­ca­tion of com­pre­hen­sive and spe­cific cus­tomer be­hav­ior and are de­liv­ered in an eas­ily us­able and quan­ti­ta­tive for­mat to en­able fi­nan­cial in­sti­tu­tions to use these cus­tomer in­sights for mak­ing more pre­cise lend­ing de­ci­sions. This would pave the way for pos­si­bil­i­ties to ex­pand the re­tail credit mar­ket and pro­mote ac­cess to cheaper, eas­ier and faster credit op­por­tu­ni­ties, which is a ma­jor as­pect of fi­nan­cial in­clu­sion in this digital age.”

She men­tions that Tran­sUnion CIBIL’s anal­y­sis of con­sumer credit de­mand and be­hav­ior us­ing Cred­itVi­sion, shows that it could en­able credit ac­cess to an in­cre­men­tal 15 lakh bor­row­ers ev­ery year without com­pro­mis­ing on risk. In ad­di­tion, the al­go­rithms have iden­ti­fied an­other 20 lakh bor­row­ers who cur­rently have ac­cess to bank­ing credit, but would be el­i­gi­ble for higher lines of credit or higher loan-to-value.

IN­STANT LOANS

Sev­eral banks of­fer loans al­most in­stan­ta­neously and through online plat­forms? How risky is this? What are the in­puts that should nor­mally go into while tak­ing such de­ci­sions?

Vi­jay Anandh re­sponds: “Banks and fi­nan­cial in­sti­tu­tions today have tech­nol­ogy plat­forms that can elim­i­nate risk and cre­at­ing cus­tomer pro­file al­most in­stan­ta­neously. There are sys­tems like eKYC, which en­sure that the rel­e­vant in­for­ma­tion is avail­able. Banks are also hav­ing in their pos­ses­sion the in­for­ma­tion on the credit be­hav­ior of the ap­pli­cant. And these are avail­able to the de­ci­sion maker, who can in al­most all cases take in­stant de­ci­sions. There are ba­sic ad­van­tages like im­proved cus­tomer ex­pe­ri­ence, bet­ter bond­ing, lit­tle doc­u­men­ta­tion, and elim­i­na­tion of mid­dle­men or agents. What all these leads to is the fact that data an­a­lyt­ics is go­ing to be­come the core in de­ci­sion-mak­ing in banks.”

Pandey says ex­ten­sive use of data and de­rived an­a­lyt­ics through ro­bust and flex­i­ble tech­nol­ogy plat­forms have made this pos­si­ble. “These are not nec­es­sar­ily risky as the heuris­tics of man­ual un­der­writ­ing have been cod­i­fied in ad­di­tion to other so­phis­ti­cated sta­tis­ti­cal ap­proaches. The online plat­forms and real-time assess­ment tools have en­abled a quick ac­cess to cus­tomer from any­where and at any time, and avail­abil­ity of de­ci­sion in a few sec­onds to a few min­utes. A cus­tomer is ex­posed to so many things ev­ery day be­cause of hand-held de­vices en­abled with in­ter­net con­nec­tions. The loy­al­ties are short and at­ten­tion span shorter. Quick de­ci­sions and online seam­less ser­vice, there­fore, have be­come fo­cus area for fi­nan­cial in­sti­tu­tions too,” she says.

TRUSTED CUS­TOMERS ONLY

Govind Sankara­narayanan’s views dif­fer slightly. Says he: “Most banks of­fer in­stan­ta­neous loans (in­clud­ing dis­bur­sals) only to their own cus­tomers ie, those cus­tomers who have an ex­ist­ing ac­count with them and con­se­quently, have con­formed to all KYC norms by shar­ing all their in­for­ma­tion with the fi­nan­cial in­sti­tu­tion. Tra­di­tion­ally, most NBFCs and other fi­nan­cial ser­vices com­pa­nies only of­fer in­stant loan ap­provals, which is ful­filled only once the cus­tomer shares ad­di­tional doc­u­ments with them. While the for­mer isn’t risky as the bank has full dis­clo­sure on the cus­tomer via their bank ac­count de­tails, the lat­ter has its risks associated with fraud­u­lent bank ac­count state­ments and other doc­u­ments, which could give the lender a wrong im­pres­sion about a cus­tomer. But with the ad­vent of strin­gent KYC norms in­volv­ing the use of PAN and Aad­haar has helped mit­i­gate such risks and of­fer a clear pic­ture.”

Vi­jay Anandh R fore­sees a sce­nario where core de­ci­sions in banks will be made based on data an­a­lyt­ics and credit risk mod­el­ing will be the norm

Govind Sankara­narayanan feels risk-based pric­ing will en­sure that the cus­tomers work to­wards keep­ing their scores and credit wor­thi­ness high

Kal­pana Pandey be­lieves assess­ment of char­ac­ter, ca­pac­ity, cap­i­tal, con­di­tions, and col­lat­eral will con­tinue to gov­ern the lend­ing de­ci­sions

Har­shala Chan­dorkar points out that CIBIL Score and Re­port helps the lender as­sess 3 vi­tal pa­ram­e­ters of a bor­rower’s fi­nan­cial health

Newspapers in English

Newspapers from India

© PressReader. All rights reserved.