Scoring model of the credit rating of the borrower

Almost everyone who has ever received a refusal to issue a loan, heard from the manager such a phrase: "The decision was made by the scoring system. Your reliability indicators as a borrower do not correspond to the norm ". What is the norm, what is scoring and how to pass the "credit detector" to "excellent"? Let's try to figure it out.

General information

So, what is scoring? This is a kind of system for assessing the reliability of the borrower, built on a number of parameters. When a person applies for a loan, the first thing that they propose to do is fill out a questionnaire. The questions of the questionnaire are not just invented. This is the scoring model for estimating the potential borrower. Depending on the answer, a certain number of points is assigned for each item. The more of them, the higher the probability of obtaining a positive decision on the issuance of funds.

There is one nuance. If you have a negative credit history, then the further answers to the questions and the number of points scored are often irrelevant. Already this one fact is enough for refusal.

The goals and objectives of scoring in modern banks

Any scoring model used in the credit system is introduced in order to obtain such results:

• Increase in the loan portfolio due to a decrease in the share of unreasonable refusals on loans;
• Acceleration of the procedure for evaluating a potential borrower;
• Decrease in the level of non-repayment of loan funds;
• Improving the quality and accuracy of the borrower's assessment;
• Centralized accumulation of customer data;
• A decrease in the provision for the amount of probable loss on loans;
• Assessment of the dynamics of changes in the individual credit account and the entire portfolio of loans in general.

Credit scoring: how does it work?

To achieve these goals, banks use a scoring model for assessing creditworthiness. It assumes minimal influence on the result of biased managerial or collusion of the bank's employees.

Almost all information entered into the questionnaire must be confirmed by the availability of documents. The bank manager executes in this case a purely technical role - contributes data to the program. When all the points of the questionnaire are swirled, the computer program calculates and displays the result - the number of points you have collected. Then the situation can develop in different ways.

If you scored too few points, you can be sure: the loan will be refused.

The number of points was much higher than average? If the loan amount is small, the decision can be made directly on the spot. In case you are applying for a rather impressive amount, you will be announced that you have passed the first stage of the audit, and the application was submitted for consideration to the bank's security service.

The number of points "floats" in the middle? The manager, most likely, will demand to lead the guarantor or will appoint a series of additional checks.

Types of scoring

In general, the scoring model consists of seven types of assessment, four of which are related to lending, and three to marketing. For credit practice, such types of scoring are typical:

1. On request (Application-scoring). This model is most often used to assess the reliability and solvency of customers. It was built, as already mentioned, on the evaluation of the questionnaire and assigning to each answer the corresponding number of points.
2. From fraud (Fraud-scoring). Helps to calculate potential scammers who managed to pass the first stage of testing. The principles, methods and methods of testing for fraud are the commercial secret of each bank.
3. Behavioral scoring. There is an analysis of the borrower's behavior in relation to the loan, the probability of a change in solvency. Based on the results of the assessment, the maximum loan amount is adjusted.
4. Work on returns (Collection-scoring). This model is applied to problem loans, at the stage of repayment of unpaid debts. The program helps to formulate a plan of measures for the repayment of the loan: from a warning to the transfer of the case to the court or collection firm.

The other three views look like this:

1. Pre-Sale (Pre-Sale) - reveals the potential needs of the borrower, allows you to propose an additional product.
2. Response - assesses the likelihood of the client's agreement with the proposed loan programs.
3. Attrition assessment is an assessment of the likelihood that a client will terminate his relationship with the bank at this stage or in the future.

Disadvantages of the scoring system

The assessment of the creditworthiness of individuals has its drawbacks. The main thing is that the system is not flexible enough and does not adapt well to real parameters. For example, a scoring model adopted in the US will put a high score on a person who has replaced a large number of work places. Such a person is considered a remarkable specialist, very much in demand in the labor market. We have this fact will play with the borrower a cruel joke. The greatest number of points will be received by a person who has only one entry in the workplace. If the borrower often changes the employer, then he is considered unreliable, unhappy and a bad specialist. His rating in the eyes of the bank is rapidly dropping, because the next dismissal may not follow the new work, which means that delays in payments will begin.

To maximally adapt the system to our living conditions, experts for the evaluation should be developed by experts of the highest category and qualifications. But any results obtained in this way will still be dependent on the opinion and influence of the person. So, absolutely unbiased assessment still does not work.

So any scoring system has at least two drawbacks:

• High cost of adaptation to modern realities;
• The influence of the subjective opinion of a specialist on the choice of the client's assessment model.

In addition, the assessment system itself is also imperfect. The fact is that when you point out points, only the formal state of things is taken into account. The system is not capable of correctly assessing reality. For example, if a client has a room in a communal flat in Arbat, the system will give him a high score. After all, there is a Moscow residence permit and housing in the center. A chic mansion of several thousand square meters, located in a small village on the Black Sea coast, the system will designate as "housing in the village" and reduce the score for the lack of a Moscow residence permit.

What data is involved in building the model

In those cases when the credit rating of individuals is being evaluated, a bank employee must rely on a number of criteria. All of them can be divided into three large groups, each of which includes many indicators.

Personal:

• Passport data ;
• marital status;
• age;
• Presence of children, their age and quantity.

Financial:

• The amount of the basic monthly income;
• Place of work, position;
• Number of entries in the work book;
• Period of employment in the last firm;
• Presence of encumbrances (debts, outstanding loans, alimony and other payments);
• Availability of own housing, car, bank accounts and deposits.

Additional:

• The existence of additional sources of income not documented;
• The possibility of providing a surety;
• Other information.

Scoring model of the credit rating of a legal entity is built a little differently. Here the key parameters are financial indicators. But since they are calculated on the basis of the financial reports of the applicant campaign, in which case they can be adjusted. In view of this possibility, the objectivity of the assessment is greatly reduced. Therefore, scoring with dynamic indicators is used to assess legal entities.

The first step is based on gathering information that can not be calculated by material indicators. These include business reputation, market position, expert opinion on financial and economic sustainability.

The next step is the definition of financial indicators. Here we study the coefficients of liquidity, provision with own funds, objective indicators of financial stability, profitability, turnover of funds and so on.

Based on the results of two independent evaluations, the bank makes a decision to grant a loan.

Who can get high marks

If we talk about individuals, then the borrower's assessment is also carried out in many respects. There are many factors that can positively affect the rating:

• high salary;
• Presence of personal movable and immovable property;
• Long residence in a particular region;
• Availability of deposits;
• Documentary evidence of income;
• The presence of a landline phone at home and at work;
• Confirmation of official employment, especially at state enterprises and in the budgetary sphere;
• Presence of open accounts (deposit, pension, settlement) in a creditor bank;
• Presence of a significant amount of advance payment when obtaining a mortgage or a car loan;
• The possibility of providing recommendations, a surety or co-borrower;
• Excellent credit history.

How to cheat the system and can it be done?

It is believed that once the evaluation is conducted by a soulless machine, you can deceive it by discovering the "right" answers to the questions in advance. In fact, this is far from the case.

The scoring model of the client's evaluation is constructed in such a way that all the answers to the questions can be checked with the help of the relevant documents. In addition, banks often integrate into entire networks and discard the results of their audits into one common system. So if in the process of additional verification the fraud will be revealed, a bad cross will be placed on your reputation of the borrower. Nowhere and never will you get more credit.

To embellish reality, you can try only if the data is entered into the system only from the words of the client. However, it is rather difficult to find such a bank, and the interest there is so predatory that you yourself are unlikely to want to issue a loan there.

Scoring and credit history

If we consider that at least half of the residents of our country already had experience of applying for a loan, in the forefront is an indicator of the borrower's assessment, such as a credit history. Since the BCH has since been replenished with data on borrowers of microfinance institutions and other similar institutions, scoring models appeared on the market, adjusted for the presence and condition of the credit history.

These models assess borrowers for the probability of non-return of cash, the occurrence of delinquencies, the number of previously repaid loans and other parameters.

In addition, banks are offered automatic information about customers. Having connected such service, the bank will know:

• About opening by the client of accounts in other financial institutions;
• On obtaining new loans;
• On the occurrence of any delinquencies;
• New client passport data;
• About changing limits on accounts, credit cards and so on.

This will allow to further regulate the system of banking scoring and obtain maximum information about potential borrowers.