In summer 2019, the Financial Conduct Authority (FCA) has started a Credit Information Market Study. In the consultation on its terms of reference, the FCA asked:
Do you consider that the credit information market is working well?
I think it has grown up over the years into a poorly understood, complex system which has many problems. It isn’t working well for consumers or lenders.
The issues fall into three main groups:
- The fact there are three main Credit Reference Agencies (CRAs) creates difficulties for consumers and lenders in seeing the complete picture.
- The data held – and therefore the calculations on it – can be inaccurate and errors are hard to correct. A lot of the problems come from the SCOR Principles which are supposed to govern how arrears, defaults and arrangements are reported by lenders. These are not well understood and not applied in the same way by all lenders.
- The credit score calculations tend to encourage some consumers to take on too much debt and fail to encourage consumers to tackle debt problems.
There is a summary at the bottom of this article on what has changed since 2019.
This matters – credit scores affect people’s lives
The purpose of CRA data for lenders is to help them assess new applications for credit. It provides:
- information on how the person has managed their credit in the past; and
- a snapshot of their current credit and its monthly commitments.
So a credit score can be seen as a summary of how likely someone is to default in the future.
But that description suggests that CRA reporting is neutral, just setting out the facts. It misses a key feature of the credit records and scores – they affect the interest rates someone will be charged in future.
Lower interest rates make future credit easier to manage well. High rates mean someone is more likely to run into new problems or find it hard to solve existing problems.
Credit ratings do not merely describe someone’s situation, they actually influence what the future risks will be. Academics call this performativity.
This is one feedback loop. There is a second one – people change their behaviour to try to improve a credit score. The ways in which credit scores are calculated encourage some behaviours and discourage others.
Problems caused by multiple CRAs
There is widespread confusion about the fact there are three main CRAs, with different data and which each calculate their own “credit score”.
It is common in forums for someone asking why their Experian score is so low to be advised to get a ClearScore report as it’s “better”.
Few customers understand that a report may only be looking at one CRA, why a different CRA may have different data or how often the data in their report is updated.
The multiplication of credit reports in the last few years hasn’t helped at all. Consumers want a complete report on all their data and a single credit score. Encouraging more competition in the reporting sector may well increase confusion rather than reduce it.
Advising people that they have to look at three different reports to find out what is happening is unrealistic. Anyone who has had a lot of credit accounts can get dozens of pages for each report. There is no easy way to work out where the differences are to account for what can often be very different scores.
This is a natural monopoly and artificial segmentation makes a difficult topic less easy to understand.
It’s also important to explain to consumers that many lenders don’t use the calculated credit score. This is made harder not easier by having three different CRAs and many different credit reports, all shouting that their report and score is the best.
Cost for lenders
Lenders have to choose between only using one CRA to minimise their costs and getting access to two or three in order to improve their decision making.
This is a particular problem for start-ups with little or no revenue. The FCA wants to encourage new entrants into the ethical/alternative lending segment, and it is just these people that most need good information to assess credit applications that can least afford to pay for access to all three CRAs.
And some bad credit lenders make an apparently perverse decision about which CRA to use. A bad credit lender can decide to only use Experian, when almost all payday lenders report to TransUnion. This “hear no evil, see no evil” approach to credit record checking may well be profitable for the lender, who keeps its CRA costs down and has a fig leaf of respectability for what are essentially inadequate affordability checks.
Problems caused by inaccurate data
Defaults the consumer didn’t know about
There is no legal obligation for a firm to notify a customer when a default is added to their credit record. Often consumers only find out about a default on an old utility or mobile bill from a previous house when they are declined for a mortgage or cheap car finance.
It is sometimes argued that it is the consumer’s own fault for not informing creditors when they move house. But in many of these cases the consumer was unaware there was a debt, so saw no need to tell a firm they no longer had a relationship with that they had moved.
And where a customer made a mistake and cancelled a DD too early, it seems an extreme punishment that they should have a default on their record for 6 years.
It should be pretty simple to tell the difference between a consumer who is trying to avoid paying their debts and someone who has a good credit record who has accidentally not paid a debt and paid it as soon as it was discovered. But this argument often does not result in someone’s credit record being corrected.
CCJs the consumer did not know about
A similar same issue arises with CCJs. In December 2017, the Ministry of Justice started a consultation on the problem of default CCJs because of:
concern about the potential adverse impact of a County Court judgment on individuals who, unaware that a judgment had been made against them, found months or years later that their credit rating had been damaged.
People can find out several years later, often when they are turned down for a mortgage or for a tenancy agreement, about a CCJ for a parking ticket or an old utility bill where the court papers were sent to a previous address.
A CCJ paid within a month doesn’t affect your credit record, but there is no mechanism to delete a CCJ from a credit record when someone has paid a CCJ when they found out about it. The set-aside procedure to cancels a CCJ is expensive, cumbersome and may not work.
How long should problems show on a credit record?
When a default has been added on a credit record, the debt disappears 6 years after the default date. This happens whether it is fully paid, partially paid, it will soon be cleared in a DMP, token payments are being made or it isn’t being paid at all.
In contrast, a payment arrangement will continue to show for 6 years after the debt is settled, which could be 10 or 20 years.
Why is this sensible? If a problem that is over 6 years old is too old to be interesting, why should it be retained if it was less serious? Deleting defaulted debts which are currently being repaid in a DMP or a payment arrangement also means that other lenders cannot get a good picture of someone’s outstanding credit commitments.
This is going to become an important issue in the next few years as the new Statutory Debt Repayment Plan details are worked out. How will this be recorded on credit records? This would seem to be a good opportunity to revisit what lenders, debt advisers and regulators think are desirable outcomes. This should not be a decision left to the lenders to decide,
Inconsistent application of payment arrangement / DMP reporting
The SCOR principles are intended to give a consistent approach to credit reporting. A lender can’t see who the existing lenders are on someone’s credit report, so they need to be able to assume that all lenders would report in a similar way.
The main area where this is not happening is around the decision to report a payment arrangement or a default. Most lenders adopt the SCOR “3-6 months in arrears” as the metric for adding a default.
But some credit cards, often the ones charging higher interest rates, and catalogue accounts will report an account as being in a payment arrangement for a very long while, even at the point where it is sold to a debt purchaser.
To the customer this is presented as an advantage – make the payments and we won’t mark your account as in default. It can be years later when the customer realises that if a default had been added early then it would soon drop off, but instead, the payment arrangement will harm his credit record for many more years.
To the lender (and subsequent debt purchaser) keeping the debt on someone’s credit record increases the pressure on the customer to repay the debt. So we effectively have a group of higher-cost lenders who are deliberately reporting in a different way in order to maximise their profit.
Lack of understanding of the SCOR rules
This is not a consumer education problem. A lot of the problems consumers have comes from the fact they are given inaccurate information by advisers and creditors. Three examples:
- As I was writing this, a reader left a comment asking when a debt with a default date in November 2013 would disappear, saying “A mortgage adviser said I should wait before settling the [debt] as it would appear like a brand new default on my file.” Which is wrong.
- Customers are given incorrect information about partial settlements so frequently (“if you accept this settlement offer it will show on your credit record for 6 years”) that it feels as though it is a deliberate ploy by some debt collectors to mislead people into settling in full.
- One customer took a case to the Financial Ombudsman when a mortgage lender – one of the major banks – did not use the bankruptcy date as the date of default. Only to be asked by the adjudicator at FOS to quote the relevant part of the Insolvency Act which says that the mortgage was a provable debt as the lender seemed to be suggesting that it wasn’t.
Poor consumer incentives
As more people become aware of credit reporting and hope to improve their credit score, there is the potential to move people towards “better decision making”.
Little incentive to repay debts in full
One area where the current SCOR principles are actively unhelpful is in encouraging people to repay debts where they have defaults.
At the moment a consumer’s credit score will only increase a very small amount if a defaulted debt is repaid. Although the overall debt balances fall a bit, the default itself remains after it has been settled and the effect of the default on the credit score calculation is not reduced at all by repaying it. When this is explained to a consumer it is usually met with disbelief.
Selecting risky behaviour in an attempt to improve their credit score
This may not always work in a positive direction for the system overall or indeed for individuals. For example, StepChange’s recent Subprime credit cards and problem debt report suggests that many people get “credit builder” cards to improve their credit rating but end up with more debt and a lower credit rating.
And the weighting given to credit utilisation in credit score calculations means that if someone has cleared a lot of credit card debt, their credit score is likely to decrease if they close the accounts.
From the point of view of a lender looking at an application, this is rational – a customer with a lot of unused credit doesn’t have financial problems. But it isn’t clear that it is good for consumers or the general financial system.
Free reports come with advertising
The multiplication of credit reports seems to a large extent to be financed by turning them into credit comparison sites.
A consumer who wants to check their credit record is offered a selection of loans or credit cards they may be approved for. It is far from clear that this benefits consumers:
- the selection offered may not include the cheapest options that consumer could access;
- someone in financial difficulty may be tempted to take “an easy option” when they may not have chosen to go and look for more credit.
The provision of free GDPR reports from each of the CRAs does not solve this problem by providing an advertising-free source because they do not contain credit scores. To a CRA, the concentration on “the credit score” is misplaced but that is what most consumers really want to see. Without the credit score a consumer has no way of assessing how important any changes are. And if little has changed, the headline credit score is a simple and quick way to confirm that.
It’s time to look at improving the whole credit information market
I’ve concentrated on some problems with the current system. It seems to me to be essential to understand these thoroughly before considering any changes. The interaction and feedback loops mean small changes designed to improve one area could have unforeseen consequences. For example, the recent multiplication of credit reports is not a benefit of competition so much as a source of confusion and advertising.
This FCA study is a good opportunity to reflect on where the current system is not resulting in good outcomes for consumers and lenders and how better outcomes could be designed in.
2021 & 2022 updates
The FCA study was understandably put on hold in 2020 as responding to Covid-19 took priority. But the FCA’s Woolard Review has highlighted the importance of this, saying:
There are significant opportunities to build a better credit information market to the benefit of consumers and lenders.
So I hope this will now be progressed speedily.
My review in “Lower interest if your credit score improves” – a good idea? points out that it is hard to design products to help consumers move from a poor credit rating to a better one because of the way scores are currently calculated.
In April 2021 Equifax changed its credit scoring to be out of 1000, not the previous 700, and it renamed all its rating bands. I think this sort of thing just increases customer confusion.
In summer 2022, BNPL data is being added to credit records, so lenders can take it into account. But it is not being used in calculating the credit score that a customer can see. Again this is unhelpful and creates more confusion.