Tuesday, November 09, 2010

Poverty and food security bill

What is the definition of poverty ? Is it that person who does not have enough money to buy food for himself & family or a house or a car ? Even if they are given money to buy these things they will not become "rich". This money will be spent on buying the basic needs and when that money is spent they will rejoin the ranks of poor. I think poverty is dearth of skills that allows a person to earn money and lead a good life and the only way to make people non-poor is to provide them with the resources & skills that help them earn money to fund basic and not-so-basic-needs of life.

Lets look at a family in a village right now and try to find the reason why they are poor. There is enough food available to remove hunger in most of the villages. I believe it more to do with the lack of healthcare and education infrastructure. They can earn enough money to feed themselves. But thats it. They spend all the money in food, clothes and other day-to-day needs. They manage to save very little and these savings are spent when somebody falls ill in the family. There are not good doctors in the villages so people are compelled to get healthcare in expensive private clinics. There are not good schools in the villages and they can either spend a fortune to send their kids to city schools or let them play all day long. This leaves a trail of poor people who are ready to migrate to cities and sleep under the over bridges. It is not as if our cities do not have enough jobs. On the contrary there is huge dearth of people with skills. However we do not have enough people with these skills because these people never managed to learn these much needed skills.

How does food security act fit into all this ? Why is government trying to feed people for almost free ? Won't it make people dependent on government and hence lazy to work ? Is it not better to tackle the problem of skills and let people earn their bread themselves?

It is not such a simple problem. Food is a very basic and immediate need. People cannot be compelled to send their children to schools when their very existence is in question. They take a shortcut to earn their living without gaining enough skills due to poverty which in turn keeps them poor for life. What availability of food is not going to do is to make people rich. However it will kill hunger and bring extremely poor people out of their misery. It will take their mind off this problem. It will make children of our country not beg for food. It would stop families from sending their children out in the cities to be servants and stop people migrating from villages to cities in such large numbers.

If combined well with NREGA, right to education and a massive upgrade in healthcare infrastructure then it would trigger people to think long term, send their children to schools & colleges and finally come out of poverty forever. I know it's a far fetched dream and given the track record of our government in past one is justified in being skeptical. Yet I am glad to know that the government is working on this problem and their is some strategic thinking and movement on the ground. I hope to see a poverty free India in next 10-20 years.

Thursday, November 04, 2010


Another century from Sehwag! The player who has no foot movement and who could have been almost written off by the greats of Cricket, had it not been for his aggressive run making. I believe he has been tagged as lucky batsman quite a few times by likes of Gavaskar. Sehwag is a prime example of why hard work and determination are more important than talent.

It's not about the size of the dog in the fight but the size of the fight in the dog!

Tuesday, October 26, 2010

The microfinance debate: SHG vs JLG

For last few weeks a huge debate has been brewing in the microfinance space. AP government has cracked down upon MFIs and passed an ordinance to stop them from lending to people who are part of SHGs(or self help groups), unless those SHG members obtain a NOC from their SHG.

To the uninitiated, MFI use a model called JLG (joint lending group) and government supports SHG. The basic difference between these two is the fact MFIs lend money from the banks to make loans to poor people of the JLG model. In SHG model, on the other hand, participants are encouraged to open a bank account themselves and put their joint savings into it. When they need money and SHG approves it, they get a loan. They can also get money from bank, where they have a bank account, if they need. However in most cases, such external borrowings are very little as bank doesn't reach out to them. In the end, very little money from outside the economy is pumped into the local economy.

In my opinion, SHG model doesn't work that well when it comes to poverty alleviation. Here's how I'd like to present my point:

Diwali, to many, seem like a waste of time and money. How can such a poor(?) country put up with such spend thrift traditions! In spite of all the arm-chair musings, every time it comes along, people end up spending more money than last year. In the end, it ends up circulating a lot of money from the banks and pockets of the people to the deepest & darkest(TM) corners of our country. People buy clothes, sweets, fireworks, electric lightings, earth lamps, tisi oil, shares (on diwali night) etc. Most of these goods end up pushing money from people who have it to people who need it as these goods come from the toil of the poorest of our brothers. Lets look at these goods in a little detail and look at their value chain.

Buying of clothes -> cloth shop owner -> cloth weavers -> cotton farmers
Buying of sweets -> sweet shop owner -> poor people making these sweets -> sugar seller -> sugar companies -> farmers of sugar cane etc

Had there been no diwali, any other festivals or marriages, we would end up killing our clothing industry, cotton farmers, retail cloth shops and lively hood of other millions.

Currency only makes sense if it can be spent. Similarly SHGs are not doing any good by persuading people to put most of their money in the bank. As more and more people do it, the local economy suffers from lack of liquidity.

JLG model on the other hand *pumps* money in the local economy, which spurs *real* growth. People buy more, work more to earn more and this increased economic activity in turn empowers a lot many other people in the local economy to contribute effectively and feed their families. At the end of the day when these people have enough money they will turn to banks to save it. Pressurizing them to use SHGs exclusively is just going to make more people poor. In my view government is sticking to a model which is pre 1991 and doesn't make much sense in a globalized, connected economy.

Note: This post is about my thoughts & ideas and not of my employer's.

Sunday, October 24, 2010

The Milk Network

India is going through a major bout of malnutrition. The government has been trying to fight this with schemes like mid day meals etc. They have been partly successful with these schemes. Still, there is a lot of poverty and malnutrition. Come 2013-14, we will have food security act providing 35 kg of wheat, rice and millet to almost 75% households in the country at Rs 1-3. This is a very good scheme to remove hunger. I believe this may remove a lot of hunger from this nation.

However, we cannot build a great nation just by removing hunger. We need to feed our young brigade with good nutrient food which should catalyze their growth as an individual, empower their minds to take us to the top. It is important for us to realize that feeding and educating the young India is like investing in the future. These kids today will make us a real superpower tomorrow. We need to feed kids with nutrients which makes their minds sharp such as vitamins, calcium, protein, sodium and potassium etc. Milk has all of these in very generous quantities. India is also a biggest producer of buffalo milk in the whole world with yearly production of 59 million tonnes.

How do we make sure all the kids of India have regular access to milk ?
I believe we need to build a milk network. A network of people who provide kids in India with free milk. 250 ml of a day to each and every kid in India. There are about 35 crore children in India below 15 years of age. This may exclude those who do not need free milk. This leaves us with about 15 crore such children (assuming a 40-50% poverty). Bottom line is, how much it is going to cost us ? 15 crore children, 250 ml each, at Rs. 3/packet for 300 days is 900 * 15 = 13500 crore/year or $3 billion/yr.

How big or small is this number ?
It is small if you consider that defense budget of India is about $32 bn/year. or that the NREGA this year was 39000 crore or about $9 bn or that Mr. Mukesh ambani's new home costs about $2bn! At the end of the day if we can get 10000 companies to donate an average of 50 lakh and 10 lakh rich Indians (in India and world wide) to donate an average of Rs. 50,000/year, we can surely make it work. We should also be able to get some grants and funding from big donor agencies. Govt. funding this is also a reality. However, this is a massive effort.

we can look at a completely distributed and volunteer based model in which people(or housing complexes) start donating milk to few kids in the neighbourhood. This way may be a little fragile to start with but could be more sustainable and empowering. To integrate this to the milk network we can build an online space where people running this leaf of the network can feed data back to the network, ask for monetary help & guidance and share experiences & data. I believe it would be important for this network to monitor statistics such as weight of kids, school attendance etc to make a strong case that such a network really works in uplifting the society and get funding. I also believe that kids who will benefit from this scheme would later in their life would be more compassionate and able citizens of our country. It goes without saying that these kids would surely feed the network when they become able.

For now I have started giving milk to my dhobi's 2 kids and plan to put 3-4 more such kids in the roaster! I plan to pursue this idea if I get like minded people so please ping me if you are interested.

Wednesday, October 20, 2010

Ruby: Convert numbers to Indian currency format using regex

10000.to_s.gsub(/(\d+?)(?=(\d\d)+(\d)(?!\d))(\.\d+)?/, "\\1,")

Works on decimals too

Sunday, October 17, 2010

I fear the future

Where robots drive the cars and do house chores, individual privacy would be non-existent, poor & unskilled will not have a place in the world, medical care would be super expensive, everybody would drive a car, media will be powerful enough to topple a government if the government doesn't favour the media and study of history, languages, cultures etc would be deemed unnecessary.

Humanity as we know it would cease to exist. I fear this future!

Sunday, October 10, 2010

Google building self driving cars!

The latest funny/frightening news from Google: Building self driven cars. Toyota prius running with help from cameras, radars, GPS and Google's massive data centers. How does it fit into the whole "organize the world's information" thingy ?

Even if you stretch it a mile it won't.

I think Google has distracted itself from it's original path of being everything information. It shows in things like this and efforts to sell phones. The problem with that is that the world's information is still not organized. Not even close. They have not made search any better than it was (sorry instant doesn't count). Won't it be better if they could answer simple questions directly ? or do better job at image search or catalog searching etc ? In fact the google maps itself is not fool proof which these cars depend on!

Tuesday, September 28, 2010

Wednesday, September 08, 2010

NoSQL can get you fired!

Digg is facing some issues with it's new launch of V4. The core of this issue is use of Cassandra in the new version instead of MySQL which was running the site in V3. As a result the VP of engineering has to pay with his job!

As with all new, shinny "world domination" technologies, Cassandra is too immature to be used in the wild. I am not saying that Cassandra is a bad project but that it needs time to mature. World needs NoSQL databases but you cannot just shift a multi million hits website's core system from a stable, trusted, tested workhorse to a new, talented yet unproven and expect everything to just work.

Thursday, August 12, 2010

Time for my IE revenge

Since the birth of internet we all have been tortured by website which work only in IE 6. It is time for revenge! Mostfit doesn't work in IE6. We tell the companies using it specifically to use Firefox or chrome. If they insist on M$ we tell them to "use IE 8, but no IE6"!

Die IE6! DIE

open source

Open source is super contagious. I just my office admin saying to a M$ guy: "We are against microsoft! we are building one open source products and we do not appreciate M$"! Mind you this is a social sector company and not a technology company!

Tuesday, August 03, 2010

Google ME

Rumour has it that Google is planning yet another social product "Google Me". I would like to take this early opportunity to put it out there:


like others viz Buzz, Wave, Knol, Open social, Orkut etc.

Monday, July 12, 2010

Why *NIX is the future ?

"perfect replication is enemy of any robust system" - DAEMON by Daniel Suarez

This is the single biggest reason why windows systems/networks are captured so easily by worms, hackers and viruses. Each windows network is made of exact clones and if an attacker can pawn one system in the network he can annihilate each and every one of them. A typical corporate windows network is made up of identical systems running workstations, laptops, routers and servers. Once a box is compromised each and every one of them is compromised.

Compare this to a linux system/network. Even though it is possible to write a virus/worm for linux based systems, writing a virus which can affect every posix complaint system is at best improbable. Almost every *nix box is different from each other. Kernels, Desktop environments, distributions, servers, browsers etc all are be different and this makes is so much difficult to run an exploit on each of these machines in one go.

Saturday, April 17, 2010

Microfinance as a business

Microfinance as an industry has started to grow up. However it is not all rosy. An impending IPO of SKS has added more fuel to the naysayers calling MFIs yet another loan sharks, sum-prime like bubble, profiteers and what not.

A lot of these are coming from western media like WSJ who have started covering MFI sector almost every week and obviously a lot of it is negative. They generally pick up on small village and show how MFIs have flushed them with loans or how they have all defaulted as a community. A lot of it, I think, is plain BS! These Americans have no freaking idea of whats happening on the ground. I think what western media clearly fails to comprehend is the sheer size of the market and diversity of this country...or they don't want to. Paint a bleak picture and their readers would get some kind of sadistic pleasure out reported failing of yet-another-poverty-elimination-experiment.

Viability of Microfinance: A thought experiment
Microfinance institutions(MFIs) lend to poor people who actually do some 'real' work and repay it back to MFIs overtime. For this MFIs charge a good amount of interest rate. Now lets take an example of a farmer taking a loan for cow. A good cow(not a jersey) which costs about INR 10,000 gives about 10 lts of milk. This in rupees terms comes to about INR 80 per day. Even if we take 20% of this as costs of medicines etc and assume that a cow gives milk only half the year then he earns about INR 30 or per week INR210. Now a typical MFI loan of Rs.10000 with 50 installments has an EMI of about Rs. 230. I don't understand who it is difficult for a farmer to service this loan when cows are just a side business for a farmer. Moreover he'd get good returns on his 10K loans after he has paid of the loan in one year.

Why scaling up is a necessity for MFIs
Scaling up is not a luxury but a necessity. Those who do not scale their operations cannot ever make significant dent in the bleak universe of poverty. A typical MFI starts with about 100 clients and quickly grows to about 1000 clients. The typical loan size is about 10000 which means the amount of money disbursed is about 1 crore INR. That is a large village or a small town to whom they can reach. After this they start to stagnate in need of better management, IT, manpower etc.

India has 110 crore people. 30-70% of which are in dire need of funds to scale up whatever they do and become "non-poor". The market size is about 100 million clients. If an MFI just reaches 1000 people then it is not doing enough. In my view when they reach about a million clients then all they have captured is 1% of the market which still is clearly not enough. Even if we have 100 such players still there would be room for more MFIs to service cycle 2 and 3 loans of these people. In India MFIs of size serving more than a million customers was 3 according to this report in October 2009(page 11). Therefore saying MF sector is a bubble is a joke! There is a clearly a lot of room to grow.

There are many more challenges for smaller MFIs. Fraud by their employees due to weak monitoring, difficulty is securing credit lines, higher cost of loans etc. Larger MFIs with good governance can tide over these difficulties with ease.

It is About time we started looking at MFIs as proper for profit businesses. We have reached a point where MF sector needs to kick off the training wheels and switch to corporate model and government should respond with creating solid guildlines and rules for this sector. RBI has very clearly shown that with the banks and now is the time that they do it for MFIs/NBFCs. They also need good governance and that needs talent. Without money this sector cannot bring in talented people. This sector needs good IT, solid reporting tools, CIBIL like credit bureau, professional consulting firms etc. All of this can only come along if these MFIs start scaling up, bringing to the table economics of scale to pay off corporate salaries and put in good solid systems.

Profits not extraordinary profits
What Md. Yunus started was a NGO like Bank for poor but clearly the self governing, non profit model is not for everybody and that model cannot scale for everybody. Not everybody is Md. Yunus. He has very clearly shown us a path. Not everybody is a as talented or motivated. Moreover greed is the primary reason why farmers have suffered for so many years. Allowing it all over again would be foolish. This is a capitalist economy and everybody needs to earn his/her bread. If poor are benefiting from the loans MFIs provide them and earn profit on it then why should MFIs not be allowed to earn some profit. Having said that extra-ordinary profits is something what good RBI regulation should prevent from happening.

Note: This post is about what I think & feel and not of my employer.

Monday, April 05, 2010

Featuring MOSTFIT

I have been working on Mostfit for about 6 months now. It is an open source MIS (management information system) for MFIs(Microfinance institutions). For the uninitiated MFIs provide loans to the poor without any collaterals. It is a social lending model where a group takes a joint responsibility of timely repayment of loans. MFIs are almost like the infamous money-lenders with much larger scale and lower interest rates. This idea was first started by Md. Yunus of Bangladesh who later received Noble peace prize for it.

MFIs now have started to scale up and have started adding hundereds and thousands of clients to their books every week. Most of the MFIs are managed by excel sheets and due to the size of operations it quickly starts becoming a problem for them. Let me give you an idea of this size: SKS microfinance which arguably is India's largest MFI has about 1 crore clients and about Rs. 4000 crore in loan outstanding. They are even going for an IPO very soon.

This is where MOSTFIT comes into picture. It is designed to suit the needs of a diverse model MFI with many loan products. It is built from grounds up to be completely flexible. It is written in ruby, datamapper, haml and rspec. Code resides here. We even provide an instant MIS instance on http://mostfit.in

So any MFI out there looking for a MIS should give it a try at http://mostfit.in