Background to the Invention The management and administration of the trade in credit products is significantly more expensive then it can and should be, with consequent problems for both lenders (tighter margins) and for borrowers (higher costs). This invention addresses one element of this problem by reducing management and administration complexity.

The buying and selling of credit products such as home or personal loans is a huge and lucrative business. This business is, however, somewhat different from the buying and selling of physical products in that a credit product is really no more than an exchange of legal obligations. As such the assessment of a credit product to determine if the legal obligations it incorporates are desirable ones is at the heart of the trade in these instruments.

From the borrower's point of view a loan has a variety of attributes that make it more or less desirable, such as the interest rate, the repayment term and the level of administration fees. Although competition between lenders has led to the addition of attributes over the years (such as redraw facilities, bundled check accounts, low cost credit cards, etc.) the number of attributes of interest to a borrower remains relatively small and manageable.

The lender's perspective is, on the other hand, much more complex and difficult to manage. The attributes of a loan that matter to a lender generally revolve around the rules that determine with what kind of borrower the lender is willing to enter into a legal relationship. For a given product or group of products

a lender will typically specify extensive eligibility criteria designed to ensure that a loan is offered only to borrowers who will be able to repay it.

The eligibility criteria will commonly include such things as the borrower's income, assets, liabilities, dependents, credit history and security offered. While all loan products will tend to have similar eligibility criteria they will vary in important ways depending upon many factors. For example, many lenders will only lend to borrowers for property purchases in major towns and cities-this may be a sensible policy but it leaves the rural market open to a competitor that is willing to offset the additional risk by, for example, offering a product at a slightly higher interest rate. Conversely, a lender may attempt to target low-risk borrowers by offering cheaper rates to, for example, individuals who can demonstrate a high net worth.

The variety in type and extent of borrower risk is reflected in the variety of eligibility criteria applied by the lenders. This leads to considerable complexity in the management of the lending process, thereby increasing overhead costs to the lenders and decreasing their competitiveness. Much of this management complexity derives from the need to ensure that the right eligibility criteria are consistently applied to applications for the appropriate products.

The fact that credit products have no physical existence considerably complicates the management process because it can be hard to determine exactly what it is that is being managed. Typically a loan is identified merely by a product name and the key distinguishing attributes that are of interest to a potential borrower, such as the interest rate etc. This ignores the fact that the eligibility criteria applied to applications for that loan are crucial to the profitability or otherwise of selling that product. It would seem sensible to regard a credit product conceptually as the sum of both its features (e. g. interest rate) and its constraints (e. g. only available in major urban areas). Typically, however, this is not the case. Credit products are managed as if the eligibility criteria are a completely separate issue, and said criteria are commonly managed and administered as if they were only tangentially related to the loans themselves.

This approach to managing eligibility criteria has several undesirable consequences:

1. Introducing a new product, or changing an existing one, can be more difficult than it should be due to problems in ensuring that the right eligibility criteria are applied to it.

2. Even existing products may be assessed inconsistently due to the lack of an explicit link between the product and the relevant eligibility criteria.

3. Introducing new products, or changing existing ones, can be much slower than it should be because the entire application assessment process may have to be examined to confirm that the correct eligibility criteria will be applied in the correct manner.

Summary of the Invention According to a first aspect of the present invention, there is provided a method of managing the complexity of multiple variable features associated with a plurality of credit products, including the steps of: (a) for each credit product, identifying each of its features and constraints; and (b) for each credit product, creating a document which incorporates each of the features and constraints in a machine-readable format.

According to a second aspect of the invention, there is provided a document describing a credit product, for use in a system for managing the complexity of multiple variable features associated with a plurality of credit products, the document having: (a) the features of the credit product listed in a machine readable format; and (b) the constraints of the credit product listed in a machine readable format.

The first and second aspects of the invention solve the complexity problem essentially by defining a credit product as the sum of all of its attributes, including the eligibility criteria that will be applied in assessing applications for the product. A loan product can be defined as consisting of: * Features-name/value pairs such as"Interest rate"and"6.95%", and Constraints-name/rule pairs such as"Loan to valuation ratio"and"Must be less than 80%".

This approach makes it possible to describe completely every attribute that either the lender or the borrower might need to consider in order to decide

whether or not to enter into the legal relationship that the loan agreement represents.

Merely making use of this approach to construct a single document to completely describe each loan product is sufficient to resolve the problem of missing or ambiguous links between products and their eligibility criteria. It is possible to take the approach much further by encoding the product description document using a markup language-this can make feasible the full computerised automation of the processes of finding and applying for a loan and of having the application approved.

According to a third aspect of the invention, there is provided a method of identifying one or more credit products which are suitable for a particular customer, including the steps of : (a) for each of a plurality of credit products, identifying each of its features and constraints; (b) for each credit product, creating a document which incorporates each of its features and constraints in a machine readable format; (c) entering into a computer information about the customer's financial position and credit requirements; and (d) instructing the computer to compare the customer's information with the features and constraints of one or more of the credit product documents to identify any of the credit products which have features and constraints matching the customer's financial position and credit requirements.

According to a fourth aspect of the invention, there is provided a method of assessing the eligibility of customers for particular credit products, including the steps of: (a) for each of a plurality of credit products, identifying each of its features and constraints; (b) for each credit product, creating a document which incorporates each of its features and constraints in a machine readable format; (c) entering into a computer information about the customer's financial position, credit requirements, and the credit product which the customer desires to acquire; and (d) instructing the computer to compare the customer's information with the features and constraints of the desired credit product to determine

whether the customer's financial position and credit requirements comply with the eligibility criteria for the credit product.

Markup languages make use of embedded tags to provide machine- readable meaning to text. It is possible to define tag sets for specific purposes such as loan processing. For example a tag set with tags pre-defined for such things as"interest rate"and"loan to valuation ratio"makes it possible to create a document that both completely describes every important fact about a loan product and is machine-readable-that is, a computer could automatically extract information from the document and act on that information.

Similarly a loan description document could include constraint names tagged and paired with the business rules that apply to them. This means that the assessment of loan applications can be greatly simplified by having a computer-generated list of all of the business rules that have to be applied in order for an application for that particular product to be accepted. Changing the product description document can have a literally instantaneous effect on application processing by automatically updating the processing checklists.

The inclusion of the approval rules in the load description document itself means that the document can drive its own approval workflow. The document itself determines, for example whether a Titles Office search or a security valuation or a credit check is required; these determinations are not made by some separately defined process. As a consequence, it becomes much easier to ensure that the approval assessment steps for an application for a particular financial product are taken, but that unnecessary and thus wasteful steps are not. Introducing a new product with features and constraints tailored to some perceived market needs becomes a simple matter of writing a suitably complete product document and loading it into the system. In comparison, introducing new products into other loan processing systems requires labour-intensive and error- prone manual changes to the general purpose loan approval process.

The business rules themselves need not be limited to simple textual instructions. They can point to or initiate small computer programs that perform calculations or manipulate data in response to data submitted by a borrower in support of a loan application. For example, a borrower might indicate that he or she wishes to borrow $400,000 to help buy a property valued at $600,000. A nerson could read the business rule relatinc to accentable loan to valuation ratios

and perform the specified calculations, but the use of a machine-readable markup language allows a computer to identify automatically the correct rule to apply and the action to be taken depending upon the calculation's outcome (that is, to approve or reject the application).

It is important to note that the present invention does not pre-suppose or require complete automation of the loan application process. Some, all, or none of the constraints associated with a product may have their business rules assessed automatically, with the remainder defaulting to human assessment.

The invention will hereafter be described in greater detail by reference to an example. It is to be understood that the particularity of the example does not supersede the generality of the preceding description of the invention.

Detailed Description of Preferred Embodiment Hyper Text Markup Language (HTML) has been used extensively in applications on the World Wide Web. However, HTML is not suitable for the purposes of the present invention because its set of tags are limited and are not adaptable for the purposes of identifying the important elements of a loan application. On the other hand, eXtensible Markup Language (XML), which is a relative of HTML, specifically allows the"extension"of the tag set to include terms of value and interest to a specialist field. A tag set specific to the processing of applications for credit can be defined using XML.

A document encoded using a markup language is just plain text. The standard convention is that the tags that label particular words or groups of words are identified by being enclosed in < > brackets. An end tag is identified by placing a/before the tag name.

Example 1 In this example, a document is written in XML to describe a mortgage home loan product. For the sake of clarity and simplicity, the constraints and features of the loan product have been greatly simplified. The loan is available in all States of Australia and the ACT. It has an initial interest rate of 6.65%, no establishment fee, no valuation fee, no legal fee, ongoing fees of $8 per month, and a discharge fee of $250. If there is no mortgage insurance taken out, the loan is available for up to 80% of the valuation of the property offered as security,

whereas if mortgage insurance is taken out, the loan is available for up to 95% of the valuation of the security.

The XML code which embodies this information reads as follows : <!--xml version="1. 0"--> <! DOCTYPE MortgageClass PUBLIC"-//LTS//DTD for various mortgage products//EN"> <MortgageClass> <HomeLoan> <HomeLoanProposal<BR> Institution="ADELAU" Country="AU" Local="ACT NSW QLD SA TAS VIC WA" Type="fixed" term="1" Offset="yes" Redraw="no" Combination="yes" InterestOnly="yes" ID="HLP0001"> <InitialIntRate>6. 65</InitialIntRate> <VarDefRate>6.55</VarDefRate> <AddRepayNIAMax>20k/y</AddRepayNIAMax> <BR> <BR> <EstabFee>0</EstabFee><BR> <ValFee>0</ValFee><BR> <LegalFee>0</LegalFee><BR> <OngoingFee>8/m</OngoingFee> <DischargeFee>250</DischargeFee> <BR> <BR> <MaxPCLentNMI>80</MaxPCLentNMI><BR> <MaxPCLentWMI>95</MaxPCLentWMI><BR> <MinLent>10k</MinLent><BR> <MaxLent>neg</MaxLent> <Valuation State="Unmet">Required</Valuation> <ExposureCalc State="Met">Not required</ExposureCalc>

<TOSearch State="Met">Required<TOSearch> </HomeLoanProposal> </HomeLoan> </MortgageClass> This marked-up format allows a computer program to search through a text document and identify the various features and constraints of the home loan.

Note that this example only represents a sample of the features and constraints that would exist for a real credit product. A full description would typically be a much larger document.

In the above example, a separate"document"is created for each credit product application, based on the generic document for that product, In the separate document relating to a single application, particular constraints have a status of"Met"or"Unmet". The credit product approval workflow is conceptually no more than ensuring that all of the constraints for the product are"met"by the applicant. If any attributes remain"unmet", the loan cannot be approved. This process can be manual (paper checklists ticked off by clerical workers as each task is completed), or computerized (unmet constraints automatically kick off processes to resolve themselves ; e. g. a"valuation required"constraint could initiate a valuation request), or any combination of the two.

In the above example, the code indicates that (a) a valuation is required for approval, but that it has not yet been received, (b) an exposure calculation is not required and thus has a default state of"Met", and (c) a Titles Office search is required and has been performed.

Applying the same principle to a document that is a truly complete description of a credit product makes every salient feature of the product accessible to computer applications. In other words, the product becomes self- describing. Provided care is taken to ensure that every feature and constraint is included, the product document becomes all that is required to represent every attribute of that product and is thus all that is needed to define and manage the processing of applications for that product.

Advantages offered by the present invention include the following : 1. Making credit products easier to manage, and

2. Making it possible to automate the processes relating to managing such products, and offering such products to consumers.

Completely describing a credit product by listing all of its features and all of its constraints in one document makes it possible to treat the product description as if it were the product itself-a key advance because otherwise the fact that a credit product is really no more than a disembodied concept makes it difficult to"pindown". In other words, the product description can be managed in the same way more conventional products with physical existence are managed, and this is logically equivalent to managing the product itself.

More importantly, the fact that the product is described in a way that is machine readable means that it is possible to automate and computerise: Creating new products and changing existing ones 'Distributing products electronically (e. g. via the Internet) Informing consumers of product features, and of the constraints they need to meet in order to qualify for particular products 'Assisting tenders and borrowers to enter into legal relationships with each other that meet the needs and expectations of both parties.

By way of contrast, the conventional approach of, in effect, distributing the attributes of credit products across a variety of documentary sources, written and unwritten procedures, and other forms of implicit knowledge within the lender organization, is impossible to manage reliably or to automate.

It is to be understood that various alterations, additions and/or modifications may be made to the parts previously described without departing from the ambit of the present invention.