How Interest Rate Changes Affects The Collection of Non Performing Loans (NPL) ?

Ata Cengiz
5 min readMay 22, 2021

I have always thought that the impact of the macro economic indicators on debt collection is greater than we assume.

My study using Python shows that the impact is more than I thought.

The study consists of 5 parts:

1- Introduction
2- Methodology and How?
3- Result 1- Interest Rate and Collection Relationship
4- Result 2- Interest Rate and Paying-Borrower Relationship
5- Conclusion and Why?

Introduction

I’m Ata, working for South Capital, one of Argentina’s largest asset management companies. Our company purchases the non-performing loans of the banks through auctions and collects them from the borrowers.

In a country like Argentina where interest rates fluctuate frequently and are relatively high, interest rate decisions have a huge impact on the economy.

When the central bank cuts the interest rate, people queue up in front of banks and use the loan to buy a car / house or to invest. When interest rates rise, stagnation occurs in the economy and the rise of inflation slows down.

Consequently, it is unthinkable that a company like us, operating in the field of finance, would not be affected by this situation.

My goal is to show the effect of macroeconomic indicators on NPL collections.

Methodology and How?

This section covers details and technical work behind. If you are not interested, you can skip to the results section.

Macroeconomic indicators such as inflation rate, unemployment and value of the local currency are basically an effect of the interest rate change, you can adjust other macroeconomic indicators by changing interest rate. So, I used interest rate as my independent variable rather than other indicators.

My study covers from June 2018 to January 2021.

I have 3 data tables in hand.

1- Collections Table: It shows the collections from the debtors.

3 columns: year&month of the payment, the number of the paying debtors and the amount of the payment.
There are about 1 million payment information.
The total amount of payments made is about 1 billion dollars.

Let’s group by the table under Year-Month column.

2- Interest Rates Table: It shows the monthly interest rates.

3- Overall Data Table: The Company’s monthly status

The total remaining principal column shows the principal totals of the debtors that were active in the company during the month. While this number decreases as the borrowers pay, increases as the company purchases new portfolios. Since we use it when making agreements with borrowers, I also prefer to use it in the study. The Debtors column shows the company’s total debtors by month.

Let’s merge 3 tables.

I add 2 new columns:

A column that measures our collection success: The ratio of the collection realized to the total remaining principal of all active debtors within the month.

Another column shows the ratio of borrowers who pay to all active borrowers within the month.

Then i complete the study by using correlation analysis and Linear regression.

Result 1- Interest rate negatively affects the success of the collection!

Ready to hear the correlation score between?

-0.66!

Interest rate and collection has strong negative correlation. When interest rate rise, the ratio of collections realized to the remaining principal of all active borrowers drops drastically and vice versa.

According to the regression results, 1 point change in the interest rate affects the collection by 1.25%.

Result 2- The rise in the interest rate increases the number of paying borrowers!

Contrary to the relationship between interest rate and collections, there is a moderate positive correlation between the interest rate and the number of paying borrowers ratio.

According to the regression results, 1 point change in the interest rate affects the number of borrowers who make payments by 0.28%.

Conclusion-Why?

According to the results of the study, we discovered that the interest rate has a strong negative correlation (-0.66) with the collections, and a moderate positive correlation (0.30) with the number of paying debtors.

So what could be the reason?

  • In times of low interest rate, people want to get rid of their debts to buy a new home/car or to get a new loan? So, is it makes sense to assume when the interest rate is lower (the number of borrowers who pay is less, but the amount of payments made is more) people want to pay off their debts quickly?
  • In times of low interest rate, borrower’s family/friends help borrower by using loan on their behalf ?
  • In times of high interest rate, the number of paying borrowers is higher but the average amount of payments is lower means that the collection agents are more flexible by proposing debtor longer and smaller installments? Or should we ignore just because 1 point change in the interest rate affects the number of paying borrowers only by 0.28%?

What do you think?

--

--