Examine how to integrate location insights into your credit scoring models to reduce default rates and gain a competitive advantage.
Staying competitive in the modern consumer lending market can be challenging. But what if there was a solution that could enhance your credit scoring and give you a significant advantage over other lenders?
Location analysis is such a tool.
Understanding that not all locations are created equal is crucial. Economic conditions vary widely between urban and rural areas, and even within cities, there are neighborhoods that differ in prestige and economic stability.
This variation directly impacts credit risk, with stronger economies typically correlating with lower default rates.
By integrating location insights into scoring models, lenders can make more informed decisions, reduce default rates, and ultimately gain a competitive advantage.
The default rate on loans is influenced by the economic conditions in a specific region.
As an example, let's look at the credit card delinquency rate in various US states. According to VisualCapitalis, it ranges from 12.9 percent (in Iowa) to 39.1 percent (in Mississippi):
In developing countries, this ratio is usually higher. Let's look at Mexico's loan portfolio as an example.
According to a FINCA Mexico report, the level of arrears differs depending on the city where borrowers live. For example, in cities in the southeast of the country, more than 50 percent of transactions are in arrears.
While in other cities, the number of problem loans is much lower. For example, in Puerto Escondido, Oaxaca, and Cuatro Caminos the delinquency rate does not exceed 12-13%.
Similar dependence of borrowers' solvency on economic development in the region of residence can be traced in India.
A report by the Fintech Association for Consumer Empowerment (FACE) states that the default rate on consumer loans in India's megacities is 3.3 percent. While in rural areas it is noticeably higher at 4.1 percent.
The number of delinquent borrowers also varies depending on the state. For example, in Tamil Nadu, this figure is the lowest at 3.1%, while in West Bengal and Rajasthan, more than 4% of borrowers fail to repay loans on time and in full.
This statistics proves how crucial it is for lenders to enrich their scoring models with location information.
Location insights involve verifying the borrower's geographic location and comparing the obtained data with what is stated in the loan application.
This technique allows the assessment of an applicant's reliability and creditworthiness. The obtained data enable the lender to:
This technique is successfully used by alternative data providers. The borrower verification process using this technology involves several stages.
The collaboration between the lending organization and the alternative data provider begins with the transfer of the information provided in the loan application.
E.g. RiskSeal uses the following data for verification:
After location insights analysis, lenders receive an extensive borrower profile that includes:
A positive signal would be the submission of a loan application directly from the applicant's residence, meaning the current device location matches the home address provided.
This fact confirms that the borrower is indeed the person they claim to be, assuring the lender that it is not a fraudster.
At this stage, the alternative data provider performs geocoding, obtaining the precise coordinates of the applicant.
This allows identifying the exact point on the map where the potential borrower is located at the time of the loan application.
Lending organizations also have access to the reverse service—reverse geocoding. This involves determining the applicant's address based on the available coordinates.
Using reverse geocoding, the alternative data provider returns the following information about the borrower's location:
With location insights, it is possible to determine precisely where the applicant is currently located, whether in an office building, apartment building, park, prison, hospital, bar, etc.
The obtained data are also used to identify discrepancies in the provided locations.
If all addresses are in the same region (country, city, state, etc.), it is considered a positive sign, indicating the borrower's reliability.
Significant differences can be a reason for concern and may be indicative of fraud.
Based on the conducted verification, the lender compares the data obtained from various sources to conclude the borrower's creditworthiness and reliability.
The lending organization should measure the distance between the provided addresses and check the applicant's stated and determined locations. All the obtained results are important to consider in the scoring model for an objective consumer assessment.
The lender also needs to monitor where applications are coming from. For example, many applications from one location raises suspicion, a sort of red flag.
Checking borrowers' locations is not limited to comparing locations. Specialized platforms also analyze the region from which the loan application was submitted.
This is an essential feature because each region differs in economic and social indicators that directly affect the creditworthiness of potential borrowers.
When calculating the Quality of Life (QoL) Index, various data are considered, including income, cost of living, and purchasing power of citizens.
This indicator varies significantly by region.
In the Netherlands, Luxembourg, and Iceland, it is highest at 190–200 points, while in developing countries, it is substantially lower.
For example, in Mexico it’s 125, in India - 116, and in Nigeria - 49.5 points.
The schematic map below clearly demonstrates how the QoL Index differs across various countries worldwide:
In their scoring models, lenders should consider that residents of economically developed regions will demonstrate better creditworthiness due to higher income levels.
This is another critical indicator for credit risk management. The more employment opportunities available to a potential borrower, the more likely they are to make timely loan payments.
Unemployment rates also vary across different jurisdictions.
The highest employment difficulties are observed among residents of South African countries. According to Trading Economics, 32.9% of the population in these regions is unemployed.
In India, Nigeria, and Mexico, the situation is less dire. In these countries, 7.6%, 5%, and 2.6% of the population, respectively, are not employed.
This indicator is not as immediately obvious in its value for credit risk management as the previous two, but is still necessary to consider.
Higher crime rates in a region increase the likelihood of encountering fraud when issuing loans.
To determine crime levels, similar to the Quality of Life Index, there is a Crime Index, which represents the number of crimes per 100,000 residents.
Este índice varía para cada país del mundo.
De acuerdo con estadísticas, las tasas de criminalidad más altas se observan en Venezuela, los países sudafricanos, Afganistán y otros. En estas regiones, el índice de delincuencia supera los 75 puntos.
En la India, este índice es de 44, en México es de 54 y en Nigeria es de 65 puntos.
El sistema de puntuación de RiskSeal se especializa en enriquecer los modelos de puntuación de las instituciones de crédito con información de ubicación.
Realizamos la geocodificación y la geocodificación inversa para determinar las coordenadas y la dirección exacta de los posibles prestatarios.
Esta tecnología, combinada con otras herramientas como Soluciones de búsqueda de IP, permite a los clientes de RiskSeal:
RiskSeal's scoring system performs geocoding and reverse geocoding to determine coordinates and addresses of potential borrowers.
We provide lenders with detailed location information, which helps assess the applicant's creditworthiness and solvency.
Location insights allow credit organizations to verify whether the actual location of the borrower matches the address provided in the application.
This information also enables analysis of the economic situation in the borrower's region — assessing living standards, unemployment rates, and crime rates in the country or even the city.
All these factors directly impact the risk level assigned to the borrower.
Lenders can compare the information provided by the borrower with the data revealed during the verification process. Any discrepancies indicate a higher likelihood of fraud.
Additionally, location analysis allows for the determination of the applicant’s coordinates.
Finally, lenders can consider the economic indicators of the region from which the application was submitted.
All this data helps create a comprehensive picture of the borrower’s reliability.
Enriching the scoring model with location information involves several steps. The lender provides the alternative data provider with the information specified in the loan application.
The provider analyzes the borrower’s location and delivers the data to the credit institution, which should then incorporate this information into its scoring model.
