e-news

Credit cycle considerations in a non-standard lending environment

Lending in the non-standard or non-conforming market has many unique business challenges, however, many of the traditional risk techniques used by risk managers can be efficiently used to improve decision making.

This article examines the uniqueness of the non-standard market and examines core processes in managing the customer relationship.

What is non-standard lending?
The term non-standard lending covers a wide range of business as well as a wide range of customer quality. This will include the prime lenders dealing with non-conforming lending to self employed all the way through to  the extreme sub-prime lending of micro-finance.
As an example of the range of lenders the non-standard lending will include:

1. Home Credit - Typically a mixture of customer quality ranging from prime customers who require ease of access to merchandise through to sub-prime. Borrowing is typically short term with 12 months being typical. Payment may be monthly, four weekly or weekly and typically has fairly high APR.
2. Credit Unions - Owned and controlled my members and typically aimed at supporting non-profit lending. Values are typically small and terms short. 
3. Payday Lending - Extreme Short Term lending that may cover loans until next pay cheque (example 2 weeks). These are typically branch linked and due to their short loan period have an extremely high APR. This is a rapid growth area and during 2007-2008 grew by 130% in the UK.
4. Mortgages - This can be both first & second charges on a property and typically have higher than prime APRs.  This may include high ‘Loan to Value’ ratios, Buy-to let, Self employed, first time buyers, credit impaired  and debt consolidations.
5. Pawn-broking - Typically Loans against products or security, however, not currently registered with bureaux.

Value chain considerations in non-standard lending
When considering the credit cycle the non-standard lending industry has many unique features

Prospecting
The identification of new applicants is typically aimed at generating high volumes of applications; however, targeting tends to be fairly loose looking mainly towards low income sectors or stressed area markets. The larger players may pre-screen and use automated direct mail approaches along with tactical branch driven solutions. The smaller players will be more tactical in nature such as word of mouth, local mail drop, local press, direct mail and branch network.

Many businesses are becoming more sophisticated by combining internal and bureau data to develop their model building capabilities and then linking response, accept & take-up models to improve selection criteria.  One lender improved their targeting using this technique leading to twice as many loans being taken up for the same volume mailed.

New Business Acquisition
Historically this has been heavily manual in nature with key focus on the ability to repay the loan granted. Decisioning was heavily subjective; however, decision engine technology is rapidly being introduced to this sector to establish control and consistent lending criteria.  

The larger players have developed data capture mechanisms and have developed basic scoring models. The smaller businesses are more reliant on generic bureau score and segmentations although many processes are still manual in nature.  In many cases the take on of a customer is based on light data and therefore the initial lending can be very conservative.  The initial bad debt on the first loan is occasionally viewed as a"marketing type" cost and therefore cross selling or up selling is key once a regular proof of repayment is established. 

Cross-sell & up-sell
An essential part of the business model of non-standard lending is the continuance of the relationship over time by means of addition lending by up-sell or cross sell processes.    Many larger lenders will use and have developed internal behaviour segmentations that are mainly policy rule or payment made led although analytical scoring models are fairly rare. The other lenders are extremely policy rule led with manual type processes used to identify and contact customers. Examples of eligibility rules for cross sell may be as simple as the customer making four to six contiguous payments. Increase in up-sell will be fairly modest (e.g. £250-500) however up-sell process would be fairly frequent (e.g. increases every 4-6 months)

Customer Management
Customer Management is limited in scope and if it happens is focussed mainly on either profiling aimed at aiding the cross sell process or at preventative collection activity.  Direct customer contact prior to a payment becoming due date is extremely common with the aim to be at the front of any payment demands. The larger lenders use profiling in a fairly coarse manner and this feeds into typically manual set of procedures, however, scoring is becoming more common using both internal and external data to rank order customers with a short term delinquency outcome. Smaller lenders have little ongoing processes and are nearly all subjective exercise driven. 

Collections
This is the KEY element for success in the non-standard & micro-finance environments.  It is typically very branch based along with a manual approach to collections. Home visits to collect payments are common although larger players have access to diallers and letter sequencing. Due to the typical branch catchment area and prioritisation would typically geodemographically or postcode based.  

Collections are mainly focussed on cash based processing although keys issues include obtaining up-to date contact information and gone away tracing. Many lenders are now using external bureau data help solve these key collection areas. The larger lenders will use letter and phone to support the branch based collections and are starting to use more direct debit processing on the better risk customers.  

MIS
The management information is highly targeted towards operational control and assessing the efficiency of the branch or staff. Strategically the data tracking tends to be fairly fragmented with little continuity from operational monitoring at collector / branch level on a weekly basis through to financial accounts. From a risk perspective there is typically little or no risk and strategy monitoring and any that exists is predominately manual in nature.

Acquisition & collections reporting is typically ring-fenced and not linked to understand the influences between each area; fraud definition is poor or not formally recognised.   Databases not sophisticated for many players although larger players becoming more organised.

What is the future for the sector?
The development and change being seen in the medium and large players over the last two to three years is impressive with many organisations investing in flexible tools, automation and the extensive use of analytics.   

The volumes in many organisations are similar to many Banks. There is an Increasing use of decision and scoring technologies tied in with a better understanding of the strengths and weaknesses of both internal and external data. Although this information is being used for segmentation there is still a heavy bias towards still manual execution using the legacy ledger systems.

The business infrastructure requires efficient collections normally focussed around an extensive branch network, however, the increasing footprint of internet access as the primary communication channel for acquisition is causing the industry to fragment.    Automation and cost efficiency becoming more important and the manual execution focus is therefore slowly changing

For the smaller lenders the volumes are sub-critical in size for large stand alone decision engines and the increasing use outsourced solutions or segmentation is being seen. This is usually linked to bureau data append as well as data cleaning of addresses and phone numbers.  This part of the market is typically very niche and are extremely cost conscious and due to high bad debt rates are fairly low margin businesses. 

Overall the non-standard lending environment consists of many small businesses and a few large lenders. They cover a very large proportion of the population that traditional lenders are unwilling to lend to and provide a fairly flexible service. The increasing use of decision technology, analytics and automation will drive these businesses to expend their domain areas and in the longer term may challenge the traditional lenders within the prime and near prime sectors.

Mark Keyworth
Principal Consultant
Decision Analytics
Experian