Much attention was given this week to a recently granted Facebook patent that may open the door for the social network to create an alternative credit score system based on your social graph. In other words, the creditworthiness of your Facebook friends, could dictate your own creditworthiness. This system would leverage the old idiom that “birds of a feather flock together,” suggesting that people are friends with people that behave similarly to them.
There certainly might be some truth to this in some circumstances, but it is still a terrible idea with small chance of passing regulatory muster.No matter what you think about credit scores and the credit system in America, it is relatively fair. It factors in positive and negative behaviors, meaning both good payment history and defaults can count towards your score. It contrasts positively against many countries that have negative attribute systems, where only defaults count towards your credit rating. There is no predetermination in the American credit system. You reap what you sow.A consumer in America might not have a credit file, but this is a distinctly different problem than having poor credit, which is generally a reflection of missed payments or defaults.Social data, on the other hand, judges a person based on many biases and relationships that can be gamed but are also unfair. The regulatory ramifications of a system like this could be huge. The American credit reporting system has a number of checks and balances to protect various classes of people. Credit must not discriminate based on age, gender or race. Redlining is illegal and key players in the credit system are frequently audited for compliance.Now imagine the simple case of social credit data. Social and economic classes are often networked. In this context, it would most certainly not be fair if you were judged by the gender, income, or ethnicity of your friends.
Not only would it be less fair, it would likely be much less predictive than our current credit system. Over the last few decades there has been a growing chorus of entrepreneurs who claim that social and big data will be better than credit data. Many of the people advocating for this claim have never worked in credit or financial services. If they had, they would realize that this is the equivalent of the financial services holy grail. Lenders have been investigating alternative credit models for decades. Nothing has come even close to the predictive value of the credit score.Our credit system is not perfect but it is pretty good when compared against alternatives. Improvements could be made. Error rates on credit reports can definitely be improved. Medical debt is a large source of financial distress and bankruptcy in America and the credit system needs to change how it views this. Issues like these can create a spiral of poverty and indebtedness in our society. There’s work that needs to be done.Winston Churchill once said about democracy that it was the worst form of Government, except for all of the other forms. Even with its specific flaws I think lenders, regulators and statisticians would generally acknowledge a similar truth about our credit reporting system.
Best,
Ken