Alt­man’s Z-score gets re­ju­ve­nated for new busi­nesses, even for In­dia

Z-score-famed Alt­man to de­velop model to pre­dict bank­ruptcy in SMEs

The Free Press Journal - - CORPORATE - Edited ex­cerpts:

He is a man who loves bank­rupt­cies and he does not shy away from ad­mit­ting it any­where. It is not be­cause he is a sadist but mainly be­cause he en­joys col­lect­ing data from failed en­ti­ties to recog­nise the po­ten­tial de­fault­ers and save them. Here we are talk­ing about

Pro­fes­sor Ed­ward I Alt­man, best known for de­vel­op­ment of the

Z-score to pre­dict bank­ruptcy.

Alt­man, who is Pro­fes­sor of

Fi­nance, Emer­i­tus, at New York

Uni­ver­sity's Stern School of

Busi­ness, has ded­i­cated more than 50 years study­ing bank­ruptcy. He was in

Mum­bai re­cently. Be­fore one of his speeches, he chats ex­clu­sively with Jes­cilia Karayam­para­m­bil and RN Bhaskar about his plan in In­dia.

Your model was based on as­sets and most of the com­pa­nies are tra­di­tion­ally based on as­sets. But if you look at tech­nol­ogy com­pa­nies like Uber, they are not based on as­sets. How do see those com­pa­nies?

Apart from Z-score, I have another model that I have de­vel­oped—Z dou­ble prime. It was built in 1995. That is for non-man­u­fac­tur­ers, even though it has as­sets among the vari­ables it con­sid­ers. We found Z dou­ble prime to be ac­cu­rate for re­tail­ers. But I can­not re­ally say it will be ac­cu­rate for tech­nol­ogy firms. How­ever, this model is ap­pro­pri­ate for re­tail, en­ergy, ser­vices sec­tor etc. It fo­cusses on four vari­able rather than five. We elim­i­nated the fifth vari­able—‘sales to to­tal as­sets’. We find the vari­able most sen­si­tive in case of such in­dus­tries, and found that we had some trou­ble with it.

If you ap­ply the Z model, to a typ­i­cal re­tailer, you will have on an av­er­age much higher scores com­pared to the man­u­fac­tur­ing model.

In case of as­set­s­light com­pa­nies like Uber, both mod­els will be a prob­lem. I don’t have an answer to that. I will have to build a sep­a­rate model for tech­nol­ogy com­pa­nies like Air B&B and Uber so on. Ad­di­tion­ally, we would need a sig­nif­i­cantly large num­ber of such com­pa­nies be­fore we build a sep­a­rate model. We would need de­faults in this space to build the new model. I love de­faults and bank­rupt­cies. Not be­cause I am a sadist but be­cause I get data. I have said that over years. My model will not be ap­pro­pri­ate for firms with less as­sets or no as­sets. I can­not do any­thing about it.

I have ap­plied the model to firms es­pe­cially when they have value of eq­ui­ties—or mar­ket cap­i­tal­iza­tion—which is one of the vari­ables. For in­stance, let us take another firm that has very high eq­uity and that is Tesla. The model is per­fect for Tesla as it does have as­sets and also, even though it has huge mar­ket value rel­a­tive to other au­to­mo­bile com­pa­nies, it has a rel­a­tively low num­ber or score. This is mainly due to the high level of debt of the com­pany and the lack of prof­its. We rate Tesla in terms of credit risk as B(-). Even though this com­pany could have a good mar­ket value.

I have run my model on 10 large re­cent In­dian bank­rupt­cies. Ev­ery­one was pre­dicted to be bank­rupt two years be­fore the bank­rupt­cies ac­tu­ally hap­pened. Even though their bond rat­ings were in­vest­ment grade. So com­ing to In­dia was pre­fect for me as it al­lows me to test the model against the back­ground of in­creased de­faults.

If your model were ap­plied, maybe it would have been a dif­fer­ent sce­nario in In­dia?

The model is to­tally un­emo­tional, no col­lu­sion and no cor­rup­tion. People who used the model have saved a lot of money by not in­vest­ing in Bhushan Steel, IL&FS etc. A lot of in­dus­tries are get­ting to know that. There was a re­al­i­sa­tion that this model pred­i­cated bank­ruptcy or de­fault and so they did not in­vest.

Do you see people in In­dia ap­ply­ing this model very rig­or­ously?

I am im­pressed that there are people who have been ap­ply­ing the model rig­or­ously in In­dia. At times, there have been some mis­takes. This is one of the pur­poses for my visit to In­dia. I would like to ed­u­cate people on the way to use the model on the ba­sis of what we have learned over years.

One of the pur­poses of the speech is to make the ap­pli­ca­tion more rig­or­ous and ac­cu­rate. It turns out that it is not only the In­dian pop­u­la­tion but the people around the world need to un­der­stand the model. The model is out there on the in­ter­net. It is in the pub­lic do­main. But the cri­te­ria for de­vel­op­ing the model and clas­si­fy­ing an as­set as dis­tressed have changed over years.

How many speeches are you giv­ing in In­dia?

I am giv­ing roughly seven speeches in In­dia. This is to teach people how to use the model in the mod­ern con­text.

In what ca­pac­ity, can we see you in In­dia?

We are de­vel­op­ing new mod­els that we will be sell­ing in In­dia and else­where, con­cen­trat­ing on small and medium size firms. In this case, it will go be­yond gen­eral model; and have spe­cific mod­els for dif­fer­ent sec­tors and dif­fer­ent re­gions of the coun­try. We have built mod­els for the SME area in 19 Euro­pean coun­tries. Now, we have de­cided to en­ter po­ten­tial mar­kets like In­dia.

I am also aware that we might just not find much of a dif­fer­ence in In­dia but we know that we will face some dif­fi­culty in get­ting re­li­able data in the quan­ti­ties that we re­quire.

I am here to ex­plore the pos­si­bil­ity to build a model for In­dian com­pa­nies but par­tic­u­larly in SME do­main.

What has been the out­come of your visit to In­dia?

The crowds have been en­thu­si­as­tic. My gen­eral feel­ing is that the model should be used by rat­ing agen­cies in ad­di­tion to their tra­di­tional tech­niques. I have been ap­proached by banks and other in­sti­tu­tions which I can­not talk about.

How do you see Bank­ruptcy Code in In­dia?

I am very im­pressed by the In­dian Bank­ruptcy Code. But it is too young to say whether it will work or not. For the first time, the con­cept of re­struc­tur­ing and re­or­gan­i­sa­tion has be­come rel­e­vant in In­dia. Most firms that go bank­rupt in many coun­tries usu­ally liq­ui­date and do not re­struc­ture. You have been ob­jec­tive to try to re­ha­bil­i­tate funds. So, your new code and my visit to In­dia is for­tu­nate for me. I hate to say it but I have seen a lot of in­ter­est in my work due to the bank­ruptcy that you’ll had. If there were no bank­rupt­cies or non per­form­ing loan is­sues, there will be low in­ter­est where my work is con­cerned.

Newspapers in English

Newspapers from India

© PressReader. All rights reserved.