Insights: Loans

Loans insights provides detailed information on the applicant’s loan activity, segregating loan types and lenders, based on the account statement data. Loans insights depict tendencies and estimations of overall liabilities and can be useful when doing affordability checks for various use cases. All output keys from Loans Insights can be used as simple business rules to check certain criteria holds true, and can also be used as independent features in scoring models to boost general predictiveness of a model.


Glossary

Keywords used throughout the product response and API documentation are listed and explained below.

  • Incoming loan - all income transactions for an individual that have been categorised as loans (Nordigen parent category ID 22). For example, mortgage loan transactions.

  • Outgoing loanor repayment - all expense transactions for an individual that have been categorised as loan repayments (Nordigen parent category ID 79). For example, mortgage loan repayment transactions.

  • Loan sources - each unique lender. For example, a specific bank that issued the mortgage.


Field purpose and definitions

Loans insights response holds 10 key-value pairs out of which only 1 value is an object with nested elements describing each individual category summary, while all other fields are float or integer values describing overall loan activity. Key-value pairs and objects are described below.

Loans insights

Fields like  calendar_months, calendar_months_with_incoming_loans and calendar_months_with_outgoing_loans give insights of statement’s length in full calendar months. In the same example as above, that would be all months in between February to August, i.e, 7 months. Count of calendar months with incoming and outgoing loans is always equal or smaller than the count of months. 

Field average_monthly_outgoing_loan_payment describes the average amount that is spent on all loan repayments. At first a monthly aggregate is created (sum for all repayment amounts) and then applied mean. For example, let’s imagine a statement with 5 calendar months - each month an individual pays 300 EUR towards a mortgage and in the last two months additional 50 EUR is spent on repaying a payday loan. Calculated value for average monthly outgoing loan repayment is \( \frac{3 \cdot 300 + 2 \cdot (300 + 50)}{5} = 320 EUR \).

Field days_since_last_incoming_loan gives an insight how recent (in days) is the last loan received.

Note: as outlined in API documentation, the days_since_last_incoming_loan field takes into account last incomplete month’s transactions as well. For example, if we use the same dates from example above - an uploaded statement’s start date is 15th of January and end date is 20th of September, and the last incoming loan is received on 29th of August, then the response value will be 22 days.

Monthly incoming (monthly_incoming_loan_trend) or outgoing loan trend (monthly_outgoing_loan_trend) is defined as a slope of linear approximation of monthly loan transaction data. Namely, the more positive the trend value is, the steeper is the increase in data (and vice versa for negatives). Trend is weighted on the last 3 calendar month data to emphasize most recent changes, and outliers are corrected (smoothed) before approximation. Trend calculations are limited to not more than the last 12 calendar months of data in order to minimize impact of old data points. Field is calculated only if the statement contains at least 3 calendar months, otherwise null is returned.

Both ratio fields, i.e., ratio_of_incoming_loans_to_all_income and ratio_of_outgoing_loans_to_all_expenses, summarizes the fraction of income and expense related to loans to the sum of all income or all expense transactions respectively. In essence that gives an insight into what fraction of income comes from loans (for example, incoming payday loans) and what fraction of expenses is spent on loan repayments, indicating the existing debt burden.

Fields in loans by category object

The purpose of each element in this object is to give a more granular view of a client's loan activity per loan type.

Nordigen’s category tree currently divides loans into 10 types for incoming loans and 11 types for loan repayments. The table below is a summary of category_id object values and category_title values.

Category titleIncome category IDExpense category IDComment
Fines-562For all fields that are calculated on income, category values will be null as there aren't any incoming loan fines transactions.
Payday loans9194 
Consumer loans9395 
Leasing17496 
Car leasing795796 
Mortgage24297 
Credit cards30282In Nordigen category tree the corresponding expense category title is “Credit card repayment”, but for the Loans Insights product the key is named “Credit cards”
Credit line346347 
Student loans9298 
Overdraft656657 
Other loans11781 

As most response fields overlap with fields described earlier, then examples and descriptions for average_monthly_outgoing_loan_payment, days_since_last_incoming_loan, monthly_incoming_loan_trend and monthly_outgoing_loan_trend won’t be provided in this section. However worth to mention and emphasize that for each field the word loan is associated with the respective categories listed in category_id object. For example, all values below are calculated on all incoming or outgoing payday loan transactions.

   
      ...
      {
         "average_monthly_outgoing_loan_payment":150.02,
         "category_id":{
            "expense":"94",
            "income":"91"
         },
         "category_title":"Payday loans",
         "days_since_last_incoming_loan":3,
         "monthly_incoming_loan_trend":-10.3,
         "monthly_outgoing_loan_trend":0.2,
         "number_of_incoming_loan_sources":1,
         "number_of_outgoing_loan_sources":9,
         "ratio_of_outgoing_loan_category_to_loans":0.972
      }
      ...
 

And as for other fields, number_of_incoming_loan_sources and number_of_outgoing_loan_sources gives an insight into how many unique lenders appear in the bank statement. If one (or both) values are 0, then respective other fields will be null. For example, if there aren’t any Payday loan repayment transactions, then number_of_outgoing_loan_sources will be 0 and all outgoing loan calculations like average_monthly_outgoing_loan_payment will be null for this particular category.

And the field ratio_of_outgoing_loan_category_to_loans gives an insight on loan type distribution for the account statement. If an individual has only mortgage payments, then the value for the Mortgage element is going to be 1, and null for all other elements. An obvious conclusion - the sum of this field over all object’s elements will always be 1.