Impacts, implementation and outcomes
The extraordinary growth of the Australian economy over the last decade has been driven by a number of factors. Receiving the most attention generally is the mining boom. Yet equally, and in some states more important, is the contribution of consumer spending.
Overall consumer spending accounts for approximately seventy per cent of Australia's gross domestic product, putting it front and centre in any discussion about the growth of the economy moving forward. Underpinning the strength and growth of consumer spending has been access to credit. This has been driven by both demand and the significant changes that have occurred in the consumer credit market as a result of financial deregulation. That deregulation has produced increased competition between lenders and innovation in the type of credit products available to consumers.
It is the significance of the contribution that consumer credit makes to the nation's economic growth that underlines the importance of the assessment and risk processes central to ensuring the free flow of that credit. In a nutshell, at the heart of every lending decision is the credit assessment process. And at the heart of the credit assessment process is credit reporting. It is this reality that makes the debate about Australia's credit reporting system so important.
This was true in 2004 when the issue was re-introduced to the political agenda. It is even more true in 2008 when the global credit markets are suffering from a severe bout of risk aversion and the key to unlocking those credit markets is transparency and quality data. The issue of credit reporting has been hotly debated in recent times. Yet it is one of those issues that people in the industry follow closely but can be lost on those living a normal life.
If you are unsure of what we mean when we talk about credit reporting, think credit checks. Essentially a credit report provides a lender with data and analysis on the credit worthiness of an individual. Depending on which country you are in, a credit report provides varying degrees of data on specific events such as credit applications, defaults, payment patterns and judgments. A credit report will also provide a risk rating or recommendation on whether credit should be granted to a specific applicant.
Credit providers across a range of industries including banking, finance, insurance, utilities and telecommunications purchase credit reports from a credit reporting bureau as part of their credit assessment process. The intent is to form a view on the risk presented by a particular applicant dependent on the number of credit facilities they have and their performance as a debtor.
Credit bureaus are central repositories of this type of data and provide an efficient form of credit assessment to a large and varied lending community. And because the bureau focuses specifically on credit risk they become leaders in new trends and innovation focused on improving risk assessment processes. The underlying principle of this system is that data sharing improves a lenders understanding of a consumers credit risk and helps to ensure appropriate lending decisions are made. All lenders benefit from sharing their data and all consumers have fairer access to the consumer credit market.
Different countries have different types of credit reporting systems which are broadly defined by the amount of data that can be placed on credit files. Australia has what is known as a negative-only credit reporting system. Under this system a credit report may contain some identification data; a listing of credit applications but not whether those applications have been approved or rejected; and negative performance data such as defaults, bankruptcies and court judgments.
The other end of the spectrum is known as a full-file positive reporting system and it is best demonstrated by the United States model. Under this system a credit report may contain all of the information in the negative-only system plus a range of payment performance data such as account balances and debt ratios. This additional information allows lenders to see, and consumers to showcase, a more comprehensive picture of each consumers' credit worthiness. This becomes particularly important for those consumers that may have had a negative credit event.
In the negative-only system a negative credit event is all lenders have on which to judge an application. Under the positive system there may be additional data of good payment performance that can off-set negative events and therefore provide access to the mainstream credit markets.
In the developed world Australia is joined only by New Zealand, France and Italy in having a negative-only system. The remaining countries operate under positive reporting systems with each having their own particular take on the amount and type of positive data that is permissible on credit reports.
So what is the impact of credit reporting and why is further reform necessary?
Broadly speaking there are three generations of credit reporting, each bringing with it a new degree of impact for consumers, lenders and the broader economy. The existence of these generations has been extensively researched and documented by organisations such as the World Bank.
- The first generation is the simple existence of credit bureaus which increases private sector lending and lowers national financial sector risk.
- The second generation introduces comprehensive data which leads to wider lending but lower default rates than negative only data. Wider lending is particularly beneficial to small business.
- The third generation involved broader participation by lenders and together with comprehensive data improves financial performance.
The documented contribution of the three generations is important. They clearly identify that credit reporting makes an important contribution not just too individual lenders for their credit assessment processes but also to the broader functioning of the financial services sector and the economy.
For lenders the benefits are reduced delinquencies and defaults and sustainable and affordable growth into new markets. In essence lenders can lend more at a lower risk. This finding is most often demonstrated by what is known as the Staten research.
Professor Staten as Director of the Credit Research Centre at Georgetown University in Washington DC sought to examine the effects of credit reporting systems on default rates by comparing outcomes from negative-only and full-file positive systems.
Target approval rate |
Default rates |
|||||
Full model (%) |
Negative only model (%) |
% increase in default rate on loan with negative only model (%) |
||||
40 60 75 100 |
1.08 1.90 3.04 9.31 |
2.92 3.35 4.07 9.31 |
170.4 76.3 33.9 0.0 |
The Staten research, conducted in 2000, found that for every target approval rate - that is the amount of applications a lender seeks to approve - default rates were lower in a full-file positive reporting system than for a negative-only system. The only time this wasn't the case was when the target approval rate was one hundred per cent.
For a target approval rate of sixty per cent, which at the time of the research was the benchmark approval rate for most mainstream credit providers, a negative-only model would produce a default rate of 3.35% while a full-file positive reporting system would produce a default rate of 1.9%. This means that the negative-only system produced a default rate that was 76.3% higher than the full-file positive model.
In 2004 this research was extended to compare the impact of negative-only and full-file positive reporting in an economy other than the United States and for both small and larger loans. The findings were similar to that in the United States.
Target approval rate |
Argentina (loans of $20,000 and more) |
Brazil (loans of $300,000 and more) |
||||
Default rate (%) |
% decrease in default rate when using a positive + negative model |
Default rate (%) |
% decrease in default rate when using a positive + negative model |
|||
Negative only model |
Positive + negative model |
Negative only model |
Positive + negative model |
|||
40% |
2.45 |
1.53 |
-37.6% |
2.78 |
1.30 |
-53.2 |
60% |
3.81 |
2.98 |
-21.8 |
3.37 |
1.84 |
-45.4 |
80% |
6.03 |
5.70 |
-5.5 |
3.74 |
2.88 |
-23.0 |
100% |
12.19 |
12.19 |
0.0 |
6.77 |
6.77 |
0.0 |
Default rates - small and large loans
In Argentina the two reporting systems were tested for loans of $20,000 and more. Once again the positive reporting system produced better default rates for every target approval rate other than one hundred per cent. For the target approval rate of sixty per cent the default rate was 21.8% lower under the positive reporting model than for the negative-only model.
Similar research was conducted for Brazil, examining loans of $300,000 and more. Once again the same trend emerged. But importantly the research also demonstrated that the benefit of comprehensive reporting on the default rate improves for larger loans. This research showed that for a sixty per cent target approval rate the positive model produced a default rate that was 45.4% lower than the negative-only model.
The research, while now a little dated, was at the time seminal in illustrating that the amount and type of data permissible in a credit report could ensure better loan performance for lenders. And while at the moment we are looking at the benefit for lenders it is worth noting that lower default rates are also good news for borrowers. Essentially the lower default rates mean consumers are getting access to credit they can afford and manage. Or they are being denied access to credit that could end up being harmful to their financial health.
The findings of the research have been repeated time and again elsewhere around the world. Hong Kong is a good example. In the four years to 2002, Hong Kong experienced growth in personal bankruptcy of 1,900%. Around twelve per cent of all personal bankruptcy was caused by credit card debt. Credit card write-offs stood at 13.6% by the end of 2002. This was significantly higher than comparable Asian nations such as Singapore and Korea, which had write-off rates of 5.5% and 6.1% respectively. Defaulting customers in Hong Kong had acquired debts up to fifty-five times monthly income in 2000 and forty-two times monthly income in 2002.
A central component of the Hong Kong Monetary Authority's response was to require lenders to place more data on credit reports and to have lenders share that data with others through credit bureaus. Hong Kong Monetary Authority figures show that between December 2002 and December 2004, following the shift to more comprehensive reporting, credit card write-off ratios declined from 13.6% to 3.76%. Credit card delinquency ratios declined from 1.25% to 0.44%.
The second benefit to lenders - increased sustainable lending - is also a core benefit for consumers. Given that we all understand why lenders benefit from increased and sustainable lending I'll examine this benefit from the consumer perspective. As we know, credit is a part of everyday life for consumers and indeed is critical to most people's wealth creation. This occurs most often through the use of credit to purchase the family home, the primary source of individual wealth in Australia. Therefore access to credit for consumers is critical and indeed essential. Given that for most people the use of credit can't be avoided, the aim is to ensure that they have access to appropriate, affordable and sustainable credit facilities.
And while the reality of a free society means we can't dictate that consumers only use credit for wealth creation, the data does show that comprehensive reporting produces an environment in which there is a greater uptake of credit for wealth generating purposes such as home ownership. Professor Staten looked at this issue as part of his 2000 research. In addition to looking at target approval rates and the consequent impact on defaults he also looked at the issue from the perspective of target default rates and their impact on approval rates. Essentially he examined the impact of credit reporting on a lenders ability to provide credit while maintaining their target default rate. Or more simply - how much could they lend consistent with their risk appetite? The research once again found that comprehensive reporting produced better outcomes. And importantly the research found that the lower the target default rate the greater their uplift in providing credit.
Target default rate (%) |
Approval rates |
||||
Full model (%) |
Negative only model (%) |
% decrease in consumers who obtain a loan with negative only model |
|||
3 4 5 6 7 Mean |
74.8 83.2 88.9 93.1 95.5 100.0 |
39.8 73.7 84.6 90.8 95.0 100.0 |
46.8 11.4 4.8 2.5 0.5 0.0 |
For a target default rate of seven per cent there was a half of one per cent decrease in the number of consumers who could access credit under a negative-only model compared to full-file positive reporting. For a target default rate of five per cent there was a 4.8% decrease in the number of consumers who could access credit under the negative-only model compared to full-file positive. For a target default rate of three per cent there was a 46.8% decrease in the number of consumers who could access credit under the negative-only model compared to full-file positive.
It is important to understand what this data is telling us. When a lender seeks to have a default rate of no more than three per cent the full-file positive model provides access to credit for a far larger number of consumers. There is no difference in the risk of default between those that are approved in a full-file positive system and those that aren't in a negative-only system with the same target default rate. Both the approved and declined represent a three per cent chance of default. The only difference is that under the negative only system the lender does not have enough information on 46.8% of lenders with which to make an informed decision.
The impact for lenders is obvious. They are being denied access to borrowers that match their risk profile.
For borrowers it means that the 46.8% of applicants that could not access credit must either go without or seek credit from lenders with a greater appetite for risk. Of course those lenders are likely to charge a higher price for access to their credit. This is an important point. Under the negative-only system consumers who can not access credit from a lender with a low target default rate are either denied wealth creating finance or must pay a higher price for their credit.
For consumers the benefit of positive reporting is clear. Positive data improves access to affordable and sustainable credit. The other major benefit for consumers is that comprehensive reporting can reduce over-indebtedness.
Currently the Australian system can be gamed. In a system where credit reports can not include data on the existence of credit facilities but only negative data such as defaults, consumers who are struggling with their debt level can acquire more finance in order to meet current credit commitments. As long as the consumer meets their minimum required payment they can easily access further credit, adding to their existing burden.
This absence of defaults on a credit report combined with an incomplete picture from an application increases the likelihood that a lender will approve a credit request. This information asymmetry allows a borrower to achieve short term relief to their credit problems but over the long term simply adds to their already problematic debt levels. This problem was also identified in the United Kingdom by a Treasury Select Committee inquiry into credit card charges and marketing in 2005 which included an examination of credit reporting. That committee found that:
- Lenders may be unable to assess accurately a consumer's ability to take on additional debt. A consumer making the minimum repayment across a number of different credit cards may be struggling with debt but may be seen to have a good credit record.
The committee also found that:
- lenders need to share full information regarding credit card accounts and to assess ability to repay using full credit data and an examination of consumer's full credit commitments in relation to income.
And the Chairman of that committee was reported as saying,
- Sharing more data will stop people being allowed too much credit and will prevent unmanageable levels of debt from building up.
The data shows then that comprehensive reporting increases sustainable lending and reduces default rates, ensuring a more efficient and effective credit market for both lenders and borrowers.
It is also important however to recognise the benefit these gains provide to the broader economy. An ACIL-Tasman study conducted for MasterCard examined this issue. That report found that a shift to comprehensive reporting in Australia would result in a:
- One-off increase in capital productivity of 0.1%, which would translate into economic benefits to the Australian economy of up to $5.3 billion, in net present value terms, over the next decade.
Once the benefits of comprehensive reporting are recognised, the question becomes what level of comprehensive reporting is desirable? Asking this reflects the fact that the level of benefit derived from comprehensive reporting is linked to the level of comprehensive reporting allowed. On this issue there is much debate. That debate is built around three strands of thinking. The first strand is opposed to any form of comprehensive reporting and believes that no further data should be allowed on credit reports. The second strand believes that Australia should move to a full-file positive reporting model like that in the United States. The third strand is the Dun & Bradstreet view that the first step should be the development of a model that finds a balance between the two extremes of the negative only and full-file positive models.
However it is worth noting that there has been considerable movement on this issue since the debate was placed back on the political agenda in 2004. While a number of lenders continue to believe the introduction of the United States model would be the most desirable, Australia's major lenders and credit providers which form the Australian Retail Credit Association have come to a general agreement on a moderate model which balances the two extremes. Each party can speak for themselves but from Dun & Bradstreet's perspective the reasons for the moderate approach are clear.
Firstly it is worth understanding what the D&B model looks like. In addition to the negative data currently allowed in Australia we have argued for four additional data elements. They are:
- whether a credit application has been approved and if so what type of credit
- the lender
- date opened, and
- credit limit.
We believe this minimal additional data provides some critical elements that allow a lender to acquire a better understanding of a borrowers' financial position and their capacity to meet new credit commitments. While not ideal, D&B believes that demonstrating the benefits that can arise from even this minimal additional data will be critical to the success of any future reform. Quite simply, given the level of opposition in some quarters to any reform and the controversy that accompanies reform of the Privacy Act in general, we believe that an attempt to move to a full-file positive model would not be successful.
Furthermore such an approach would deny us the opportunity to take a small but significant step in such an important debate. It is also worth noting that this has been the approach of many other countries. It took the United States three decades to produce the system it has today. And it is a system constantly under attack by those opposed to reform.
Other countries like the United Kingdom and Japan have also opted for reform on an ongoing basis rather than an immediate shift from negative-only to full-file positive reporting. However it is also important to understand that our more moderate model is not just about what is possible. There are very real benefits that can be derived from an intermediate model.
In Japan a more comprehensive reporting system, similar to the proposed D&B model for Australia has reduced the probability of delinquencies exceeding sixty days by 34.1% for the mean loan. The value of more comprehensive data in Japan increases as the loan amount increases. The Japan model reduces the probability of delinquency by 41.3% for the mean large loan.
Recently Staten refreshed his research at the request of the Australian Finance Conference to include an analysis of the impact of an intermediate model in addition to negative-only and full-file positive. The research showed that while a full-file model had the greatest impact on reducing default rates, an intermediate model brings significant improvement. With a target approval rate of sixty per cent the default rate under a negative-only model is 3.35%. Under an intermediate model the default rate is 2.46% and under the full-file model it is 1.9%. This research is important in illustrating to those sceptical of the moderate approach that there is a real benefit, particularly when examined through the prism of responsible lending.
So if the Australian Law Reform Commission does recommend reform of Australia's credit reporting laws and the Federal Government implements those changes, how does the industry prepare for change?
First lenders must use the data. It seems an obvious point but the reality is that there are a number of lenders who currently are not availing themselves of the data available under current laws, while others are availing themselves of only a limited amount of this data. This includes the need to consider alternative data, particularly as the credit cycle turns. Alternative data is data provided by utilities and telecommunication companies in addition to the traditional data provided by the banking and finance industry.
Alternative data is highly predictive of future defaults. In particular, it draws clear links between those that default on low value and non-bank debt and their eventual default on high value bank debt such as home mortgages.
For lenders to receive the full benefit of a comprehensive reporting system, even a moderate one, lenders must make use of the available data. And for lenders to have credibility in their push for more data they must demonstrate that they are maximising the potential of what is already available.
Second lenders must maximise their data sharing. The issue of data sharing is as important as the issue of the type of data that is contained in credit reports. The three generations of credit reporting were referred to earlier.
Australia currently resides in the first generation of credit reporting and the proposed reforms will allow Australia to progress to the second generation. However reaching the third generation, which brings the greatest uplift, does not require further legislative reform. Rather it requires broader participation from lenders in the credit reporting system. It is a reflection of the fact that data sharing brings benefits to lenders and borrowers in addition to the type and amount of data shared.
Once again we turn to approval and default rates to demonstrate the point. Research from Dr Michael Turner of the Information Policy Institute, one of the world's leading experts on credit reporting, demonstrates the benefit that comes from increased data sharing. His research shows that the greater the level of comprehensive data sharing with a credit bureau, the lower the default rates.
Data sharing and default rates
Where there is full-file positive reporting and one hundred per cent of lenders sharing data with a bureau produces a default rate of approximately seven per cent for a target approval rate of sixty per cent. The default rate jumps to nearly nine per cent when only seventy-five per cent of lenders are sharing data; and to approximately 10.5% when only twenty-five per cent of lenders share data. The gap between default rates grows the more selective lenders become with their approval targets. The point is also demonstrated when examining the issue from the perspective of improved access for consumers.
Acceptance rate |
||||
For a 7% default rate |
Share of lenders providing positive and negative information |
|||
100% |
75% |
50% |
25% |
|
Male |
64.92% |
51.40% |
44.31% |
33.68% |
Female |
63.20% |
42.24% |
33.43% |
22.30% |
Age categories |
||||
0-32 |
16.48% |
15.47% |
14.20% |
8.61% |
32-42 |
29.72% |
44.75% |
28.42% |
13.71% |
42-50 |
58.31% |
45.20% |
20.52% |
19.14% |
50-57 |
62.76% |
52.02% |
39.61% |
19.13% |
57+ |
77.13% |
72.98% |
69.54% |
66.49% |
In an emerging economy where minority groups have poor access to mainstream credit, the introduction of comprehensive data and lenders' willingness to share this data with credit bureaus provides greater access to mainstream credit for women and the young. When one hundred per cent of lenders share comprehensive data, males and females have very similar access to credit. However when the amount of lenders sharing data drops to twenty-five per cent there is a significant gap between men and women, in men's favour, when seeking access to credit.
The same is true when examining access by age. Access to credit for young people becomes most difficult when lenders have comprehensive data but do not share it with each other. Older people with more established reputation collateral suffer the least when lenders fail to share data.
So data sharing is as critical as the type and amount of data. It is an important point because just as there are lenders in Australia that want comprehensive reporting but aren't making use of already available data, there are lenders that are not maximising benefits for themselves and borrowers by sharing data. Third lenders must realise the full potential of any reform.
A number of organisations have argued that the benefit of an intermediate reporting model is not sufficient to entice lenders to upgrade their systems to report and receive this new data. This has been held out as an attempt to convince regulators to go for the full-file positive model on the basis that anything less would effectively result in no change. It is a spurious and dangerous line of argument. As we have already seen there are very real benefits to be derived from an intermediate model. But more importantly any future decision on a move to a full-file positive system by government will be largely based on whether lenders have sought to maximise the responsible lending benefits available from the first round of reform. A failure by the industry to do so because we didn't get all of what we wanted when we wanted it will ensure that the opportunity to turn this reform into the first of many will be lost.
A modern economy cannot grow without access to affordable and sustainable credit. It is vital to our long term future and our ability to ride out external shocks. Critical to credit growth being affordable and sustainable are risk management and assessment processes. Credit reporting is at the heart of this equation and reform that provides innovation and competition is critical to an effective system. With the Australian Law Reform Commission due to hand its report to government in just a couple of months we have a unique opportunity to underpin the future growth of the industry and the broader economy.
Christine Christian
D&B Chief Executive Officer
Credit Risk Conference March 2008