Where did it all go wrong with credit scoring for consumers?
Credit scores & consumers.
This story draws together a topic that I am very passionately interested in personally, but also something that seems to be a worldwide problem. Let me write out some of things I have discovered from my research over the past years and where I think some answers are emerging to save consumers.
Some might say that it’s my manifesto for launching Aire.
What exactly are credit scores?
Credit scoring, for individuals is the process in which a rating is assigned to each consumer describing their likelihood of defaulting on any of their credit lines in the near future. To some extent, the vagueness in the definition is almost true of the field as well.
Credit scores, as the final output is called, are compiled by credit bureaus, which are typically private enterprises that collect data, run statistical models and produce these scores (notably Experian, Equifax, Transunion, Call Credit — the major ones in the US and UK).
When banks make lending decisions, they rely on these third-party credit scores and factor in other inputs from their records (income, expenses, cash-flow etc) to make a lending decision. So these credit scores are quite important inputs to compute how much to lend, at what rate and for how long.
But unfortunately it doesn’t just stop there with credit scores. In places like US and Europe, it almost forms the bedrock of our socio-economic progression. Any young professional will tell you anecdotes of how a poor credit score didn’t allow them to get a decent mobile-phone contract, perhaps had to pay up higher deposits on their apartment rental and worse yet, might even have got rejected from a potential job despite doing well on the interviews.
Yes. Even employers are checking credit scores to determine whether or not to hire a person.
And on the deeper end of the pool, those with low or non-existing credit scores are also denied banking services by banks. Even if it is to use the bank as a deposit-point and use debit-card only transactions (i.e.: “Just hold my money, and let me use it from a debit card — I don’t need credit or lending facilities” — the answer is still no). And this leads to large parts of the population being unbanked or underbanked.
While illegal immigration and criminal records are also a major contributor to being un(der)banked — low credit scores often block legitimate law-abiding citizens as well. Several of them. NYTimes reported that in the US, nearly 54 million adults (‘Credit Invisibles’) cannot access financial services owing to credit file issues (i.e.: reasons other than legal residence or criminal records). Globally this is 4.5 billion people who do not have a formal credit score (data by WorldBank).
Bottomline: In the 50+ years that credit scores have been around, our society has somehow started labelling those people who have a bad credit score, as bad people.
Who generates these scores?
The initial methodology for calculating a credit score was conceived by Fair Isaac Corporation in the 1950s (many might recognise this company under its new branding, FICO, which is listed on the stock exchange). Their method takes into account 5 core parameters that they felt accurately predicted the likelihood of default of a person on any of their credit lines in the coming 6 months:
This method (FICO) is used by a majority of the main credit scoring bureaus around the world. Credit scoring bureaus such as Experian, Equifax, TransUnion and CallCredit are private for-profit companies that specialise in collecting data about individuals for those 5 parameters and then producing a FICO score (ie: Credit Score).
Typically, a higher score allows you to borrow larger amounts, at lower interest rates from wider range of lenders (who also do some further analysis on top of your FICO credit score).
But as many have tried, it is not easy to find out your own scores. After figuring out the confusions that exist, if you do manage to log into your account (after paying) there is no guarantee that they will actually show you the true credit score (yes, there exists the “educational score” or FAKO score!).
Bottomline: Very little innovation has happened in this space in the last 60 years owing to a near-monopoly. Customers have no information transparency and they almost have no rights in this game!
So are they all correctly calculated?
Would like to give you a different answer, but it’s no.
Infact a recent report in the US, published by the FTC stated that nearly 26% of all credit reports carry errors. That is more than 1 in 4! And worse yet, some the errors might even allow people to have significantly better credit scores such that their mortgages and other interest rates might be better.
Imagine something so core and fundamental to our lives, but only operating at a 74% accuracy! Wouldn’t take those odds on our seatbelts, would we?
Bottomline: The companies who have taken on the responsibility of being credit scorers aren’t doing their job with the rigour and perfection that is needed for such a critical score!
Another problem: missing data
Some readers might have noticed this in their reports, if they have accessed them — missing data causes massive drops in the score.
Unfortunately while the FICO method seemed like a breakthrough in the 1950s and 60s — it hasn’t evolved much since. It still relies on a linear regression model, which means that all the inputs must be present or else it just crashes down the output.
And this is what happens for students. Typically they have none or very little financial history. Looking at the pie chart above, it’s clear that 90% of the score is built on backward looking financial history, which if not available, really penalises their scores. Why assume that all students are not trustworthy?
And it’s not just students who are affected: expats & migrants face this too (even if they had perfect scores in their prior country and work at top firms). Also this has been observed when a couple divorces, and the spouse who wasn’t on the financial bills has built up no credit history.
Bottomline: Newer mathematical models are needed that help bridge this gap of information inconsistency such that average citizens aren’t penalised unfairly.
What about other countries?
What I described above, is true for countries where proper credit bureaus exist that compile scores for the residents, and regularly try to update them (In places like US, UK, Germany etc — there is some coverage, with only about 70% of the legal residents covered with some basic score at the minimum).
But in other places around the world, the statistics are reversed. Emerging markets like Brazil, India, China and even large parts of Europe have fragmented credit scoring records for their individuals. In most of these places nearly 80% of the population doesn’t have a formal credit profile or credit score. Some government-backed bureaus exist in certain countries and use methods similar to FICO — but none of them have a wide coverage for their populations. And that means those people are either unable to access credit-based finance or worse, have to rely on the unfair shadow banking system.
Over the past several years, I have been very fortunate to be able to travel to various parts of the world including several emerging markets. And the one thing I noticed, is that the lack of these systems has led to a sub-economy that seems to profit off this lack of information. Naturally one also gets to observe the issues of predatory lending and loan-sharks operating in those ecosystems, thriving on the imperfect information about creditability.
What I have been doing to help change this?
The first step I believe is knowledge. I believe that the more information that people have, the better they can understand the situation around them. So articles like this one and links (below) are great starting points that hope to bring to the surface.
But the problem is deeper than just knowledge, and that is why I have spent several months meeting various people in law-making positions across the UK (and US) to try to see what can be done. It is great to see the positive response from our Treasury and Cabinet Office here in Whitehall (London). They have been actively looking at the ingredients to implement change for this issue.
And finally.. technology, which I believe can help improve the mechanics and process to our credit scoring. London is a fantastic hub for so-called FinTech startups that aim to look at problems of financial services via the lens of technology. I have met with some of the leading entrepreneurs in the space, who too have noticed the cracks in the credit scoring system. In various capacities, each FinTech startup in London is helping move forward the need for overhaul, and one such area is credit scoring.
I launched my third startup in early 2014 called Aire to address this very issue. I believe that by using self-declared data from consumers themselves and smarter (transparent) algorithms.. we can build a much more sustainable and enduring credit ecosystem. One that democratises the process and gives the power back to the consumer.
Our message to consumers.. “Go on.. take control of your credit profile”
It is interesting that in 1906, in front of the US Congress, John Pierpoint Morgan said that it is the “character of the person” that he uses to lend against, not the finances in his bank account. Perhaps there is a strand there to pick up from about how the future of credit scoring ought to evolve. We do have the technology to power these algorithms now.
So this long article, acting almost like a manifesto, lays out some of the areas that I am keen to help resolve for the people that we share society with. It is a big mission, and not an easy one either. But I am confident that support from readers like you (who have managed to reach this part of the article) will help move the cause forward.
For now this is an initiative from me. I hope you can show your support — I amon Twitter to receive your comments.
Some links that are useful:
- MoneySavingExpert UK talks about credit scoring
- US Congress tries to improve consumer access to their scores
- Burden of miscalculated scores on average households
- I will add some links about interesting companies in this space in upcoming updates.
Hope this helps you consumers out there. And we can collectively start making an impact to help financial inclusion.