Commercial bank interest rates on credit card plans in the United States were over six percent higher in early 2025 than in the same period in 2022. In February 2025, the interest amount on credit card plans amounted to 21.37 percent. Alongside this development, the overall amount of credit card debt in the U.S. reached an all-time high in Q4 2023. Credit cards are considered one of the most common ways to pay in the United States, so potential changes on credit card debt are closely tied to both the inflation figure and central bank interest rate of the country.

Graph and download economic data for Commercial Bank Interest Rate on Credit Card Plans, Accounts Assessed Interest (TERMCBCCINTNS) from Nov 1994 to Feb 2025 about consumer credit, credit cards, loans, consumer, interest rate, banks, interest, depository institutions, rate, and USA.

United States - Commercial Bank Interest Rate on Credit Card Plans, Accounts Assessed Interest was 21.91% in February of 2025, according to the United States Federal Reserve. Historically, United States - Commercial Bank Interest Rate on Credit Card Plans, Accounts Assessed Interest reached a record high of 23.37 in August of 2024 and a record low of 11.96 in February of 2003. Trading Economics provides the current actual value, an historical data chart and related indicators for United States - Commercial Bank Interest Rate on Credit Card Plans, Accounts Assessed Interest - last updated from the United States Federal Reserve on June of 2025.

The interest rate for credit cards in the UK grew to an all-time high in May 2025, even though the base rate for the Bank of England grew at a slower pace that month. Credit card interest rates tend to be significantly higher than other forms of lending, and the United Kingdom is no exception to this. By May 2025, the average interest rate had increased to ***** percent. The Bank of England base rate stood at **** percent since April 2025 – which was not yet the highest value observed. Nevertheless, the central bank's interest rate grew slower than that of credit cards.

Graph and download economic data for Delinquency Rate on Credit Card Loans, All Commercial Banks (DRCCLACBS) from Q1 1991 to Q1 2025 about credit cards, delinquencies, commercial, loans, banks, depository institutions, rate, and USA.

Brazil Lending Rate: per Month: Pre-Fixed: Corporate Entities: Prepaid Credit Card Bills: Agiplan Financeira S.A. - CFI data was reported at 0.000 % per Month in 03 Jul 2019. This stayed constant from the previous number of 0.000 % per Month for 02 Jul 2019. Brazil Lending Rate: per Month: Pre-Fixed: Corporate Entities: Prepaid Credit Card Bills: Agiplan Financeira S.A. - CFI data is updated daily, averaging 0.000 % per Month from Jan 2012 (Median) to 03 Jul 2019, with 1867 observations. The data reached an all-time high of 0.000 % per Month in 03 Jul 2019 and a record low of 0.000 % per Month in 03 Jul 2019. Brazil Lending Rate: per Month: Pre-Fixed: Corporate Entities: Prepaid Credit Card Bills: Agiplan Financeira S.A. - CFI data remains active status in CEIC and is reported by Central Bank of Brazil. The data is categorized under Brazil Premium Database’s Interest and Foreign Exchange Rates – Table BR.MB013: Lending Rate: per Month: by Banks: Pre-Fixed: Corporate Entities: Prepaid Credit Card Bills. Lending Rate: Daily: Interest rates disclosed represent the total cost of the transaction to the client, also including taxes and operating. These rates correspond to the average fees in the period indicated in the tables. There are presented only institutions that had granted during the period determined. In general, institutions practicing different rates within the same type of credit. Thus, the rate charged to a customer may differ from the average. Several factors such as the time and volume of the transaction, as well as the guarantees offered, explain the differences between interest rates. Certain institutions grant allowance of the use of the term overdraft. However, this is not considered in the calculation of rates of this type. It should be noted that the overdraft is a modality that has high interest rates. Thus, its use should be restricted to short periods. If the customer needs resources for a longer period, should find ways to offer lower rates. The Brazilian Central Bank publishes these data with a delay about 20 days with relation to the reference period, thus allowing sufficient time for all Financial Institutions to deliver the relevant information. Interest rates presented in this set of tables correspond to averages weighted by the values of transactions conducted in the five working days specified in each table. These rates represent the average effective cost of loans to customers, consisting of the interest rates actually charged by financial institutions in their lending operations, increased tax burdens and operational incidents on the operations. The interest rates shown are the average of the rates charged in the various operations performed by financial institutions, in each modality. In one discipline, interest rates may differ between customers of the same financial institution. Interest rates vary according to several factors, such as the value and quality of collateral provided in the operation, the proportion of down payment operation, the history and the registration status of each client, the term of the transaction, among others . Institutions with “zero” did not operate on modalities for those periods or did not provide information to the Central Bank of Brazil. The Central Bank of Brazil assumes no responsibility for delay, error or other deficiency of information provided for purposes of calculating average rates presented in this

Credit card delinquency reached its highest level since 2019 in the first quarter of 2024, whereas mortgage delinquency declined to its lowest level. This is according to consumer data supplied by large banks that have to report such figures when handling over 100 billion U.S. dollars worth of assets. 3.56 percent of credit card balances were 30 days late - the highest percentage since tracking began in 2012. First-lien mortgage origination remained historically low, likely due to high interest rates and housing prices. Note the graphic shown here is different from another source on credit card delinquency rates in the U.S., as those figures are aggregates.

A revolving credit card in Brazil had an average of APR that was over 440 percent, as the country dealt with inflation and a lack of credit card regulation. This is according to information from the country's central bank, which tracked the Annual Percentage Rate or APR for both installment credit cards and revolving credit cards. The interest rates in Brazil are sizable: Out of 63 regulated institutions that deal with credit cards in the country, only seven had an APR that was below 100 percent. This 100 percent is part of a proposal in late 2023 to impose of a maximum interest rate. Credit cards long ranked as Brazil's most popular payment method for online shopping.

Graph and download economic data for Consumer Loans: Credit Cards and Other Revolving Plans, All Commercial Banks (CCLACBW027SBOG) from 2000-06-28 to 2025-06-18 about revolving, credit cards, loans, consumer, banks, depository institutions, and USA.

Argentina Lending Rate: Monthly Average: Domestic Currency: Credit Cards data was reported at 84.860 % pa in Mar 2025. This records an increase from the previous number of 83.890 % pa for Feb 2025. Argentina Lending Rate: Monthly Average: Domestic Currency: Credit Cards data is updated monthly, averaging 39.390 % pa from Jul 2002 (Median) to Mar 2025, with 273 observations. The data reached an all-time high of 125.950 % pa in Mar 2024 and a record low of 25.660 % pa in May 2007. Argentina Lending Rate: Monthly Average: Domestic Currency: Credit Cards data remains active status in CEIC and is reported by Central Bank of Argentina. The data is categorized under Global Database’s Argentina – Table AR.M006: Lending Rate: Non Financial Private Sector.

Mexico Interest Rates on Household Credit: Credit Card data was reported at 35.680 % pa in Aug 2009. This records a decrease from the previous number of 36.380 % pa for Jul 2009. Mexico Interest Rates on Household Credit: Credit Card data is updated monthly, averaging 34.205 % pa from Jan 2004 (Median) to Aug 2009, with 68 observations. The data reached an all-time high of 41.870 % pa in Dec 2008 and a record low of 31.390 % pa in Apr 2007. Mexico Interest Rates on Household Credit: Credit Card data remains active status in CEIC and is reported by Bank of Mexico. The data is categorized under Global Database’s Mexico – Table MX.M006: Household Credit Interest Rates.

During the past years, the interest rate on consumer credit in Italy increased a lot, reaching 8.97 percent in May 2024. The interest rate hike of the past years has started to reverse, with the interest rates of consumer loans decreasing in the second half of 2024.

Graph and download economic data for Finance Rate on Personal Loans at Commercial Banks, 24 Month Loan (TERMCBPER24NS) from Feb 1972 to Feb 2025 about financing, consumer credit, loans, personal, consumer, interest rate, banks, interest, depository institutions, rate, and USA.

Credit Cards Market size was valued at USD 14.31 Billion in 2023 and is projected to reach USD 17.50 Billion by 2030, growing at a CAGR of 4.2% during the forecast period 2024-2030.

Global Credit Cards Market Drivers

The growth and development of the Credit Cards Market can be credited with a few key market drivers. Several of the major market drivers are listed below:

Spending Patterns of Consumers: The credit card market is largely driven by the spending patterns and preferences of consumers. Credit cards are in greater demand as a practical payment option as consumers move more and more toward cashless transactions and online shopping.

Situation of the Economy: A number of economic indicators, including inflation, GDP growth, and employment rates, have an impact on disposable income and consumer confidence, which in turn have an impact on credit card usage. Consumers tend to spend more and sometimes use credit cards when the economy is expanding.

Interest Rates: The cost of borrowing and the allure of credit cards are impacted by fluctuations in interest rates set by central banks. While higher interest rates may cause consumers to cut back on spending and become more concerned about repaying their debt, lower rates may encourage consumers to use credit cards more frequently.

Rewards and Incentives: To draw in new business and keep existing ones, credit card companies provide a range of rewards, cashback plans, travel perks, and incentives. Attractive rewards programs have the power to increase credit card usage and sway customer decisions.

Technological Innovation: New developments in digital banking, contactless payments, and smartphone wallets are transforming the credit card industry. Credit card issuers are investing in technological innovations to improve security and convenience as a result of consumers' increasing adoption of digital payment methods.

Regulatory Environment: Market dynamics are influenced by rules that govern the credit card industry, such as data security standards, interchange fee laws, and consumer protection laws. The pricing strategies, product offerings, and profitability of card issuers can all be impacted by changes in regulations.

Demographic Trends: The adoption and use of credit cards are influenced by demographic factors such as urbanization, population growth, and shifting lifestyles. Younger generations—Gen Z and millennials in particular—are more likely to use mobile banking and digital payments, which is increasing demand for credit cards with cutting-edge features.

Bank A issues Credit Cards to eligible customers. The Bank deploys advanced ML models and frameworks to decide on eligibility, limit, and interest rate assignment. The models and frameworks are optimized to manage early risk and ensure profitability. The Bank has now decided to build a robust risk management framework for its existing Credit Card customers, irrespective of when they were acquired. To enable this, the Bank has decided to create a “Behaviour Score”. A Behaviour Score is a predictive model. It is developed on a base of customers whose Credit Cards are open and are not past due. The model predicts the probability of customers defaulting on the Credit Cards going forward. This model will then be used for several portfolio risk management activities.

Problem statement

Your objective is to develop the Behaviour Score for Bank A.

Datasets

You have been provided with a random sample of 96,806 Credit Card details in “Dev_data_to_be_shared.zip”, along with a flag (bad_flag) – henceforth known as “development data”. This is a historical snapshot of the Credit Card portfolio of Bank A. Credit Cards that have actually defaulted have bad_flag = 1. You have also been provided with several independent variables. These include: • On us attributes like credit limit (varables with names starting with onus_attributes) • Transaction level attributes like number of transactions / rupee value transactions on various kinds of merchants (variables with names starting with transaction_attribute) • Bureau tradeline level attributes (like product holdings, historical delinquencies) – variables starting with bureau • Bureau enquiry level attributes (like PL enquiries in the last 3 months etc) – variables starting with bureau_enquiry You have also been provided with another random sample of 41,792 Credit Card details in “validation_data_to_be_shared.zip” with the same set of input variables, but without “bad_flag”. This will be referred to going forward as “validation data”.

Requirements

Using the data provided, you will have to come up with a way to predict the probability that a given Credit Card customer will default. You can use the development data for this purpose. You are then required to use the same logic to predict the probability of all the Credit Cards which are a part of the validation data. Your submission should contain two columns – the Primary key from the validation data (account_number), and the predicted probability against that account. You are also required to submit a detailed documentation of this exercise. A good document should contain details about your approach. In this section, you should include a write up on any algorithms that you use. You should then cover each of the steps that you have followed in as much detail as you can. You should then move on to any key insights or observations that you have come across in the data provided to you. Finally, you should write about what metrics you have used to measure the effectiveness of the approach that you have followed.

Evaluation

As detailed in the previous section, you are required to submit the Primary key and predicted probabilities of all the accounts provided to you in the validation data, as well as a documentation. We will only evaluate submissions that are complete and pass sanity checks (probability values should be between 0 and 1 for example). Submissions will be evaluated basis how close the predicted probabilities are to the actual outcome. We will also evaluate the documentation basis it’s completeness and accuracy. Extra points will be granted to submissions that include interesting insights / observations on the data provided.

Brazil Loans: Outstanding: Financial System: Arrears: 15-90 Days: Household: Nonearmarked: Credit Card: Financing data was reported at 8.070 % in May 2019. This records an increase from the previous number of 7.780 % for Apr 2019. Brazil Loans: Outstanding: Financial System: Arrears: 15-90 Days: Household: Nonearmarked: Credit Card: Financing data is updated monthly, averaging 5.620 % from Mar 2011 (Median) to May 2019, with 99 observations. The data reached an all-time high of 8.470 % in Mar 2019 and a record low of 2.890 % in Dec 2013. Brazil Loans: Outstanding: Financial System: Arrears: 15-90 Days: Household: Nonearmarked: Credit Card: Financing data remains active status in CEIC and is reported by Central Bank of Brazil. The data is categorized under Brazil Premium Database’s Monetary – Table BR.KAB004: Loans: Outstanding: Financial System: Arrears from 15 to 90 Days: by Modality. Brazilian Central Bank has made changes in methodology of Financial System Credit Data in February of 2013 after 13 years following the same methodology. These changes are necessary face the expansion of credit, favored by the improvement of the indicators of employment and income, continuous and sharp reduction of the interest rates and by important institutional advances. It is essential the availability of new information, in particular, which allows more detailed monitoring of credit arrangements with targeted resources, especially real estate financing, whose dynamism has contributed to reducing the housing deficit in the country. The main change includes coverage of data on concessions, interest rates, terms and default rates that were extended to the segment of directed credit and also became necessary to further detailing the statistical framework, to enable identification of the terms most relevant as well as reduce the relative share of loans not classified - embedded in 'other receivables'. Banco Central do Brasil fez mudanças na metodologia de Dados de Crédito do Sistema Financeiro, em fevereiro de 2013 depois de 13 anos seguindo a mesma metodologia. Essas mudanças são fundamentais face a expansão do crédito, favorecido pela melhora dos indicadores de emprego e renda, redução contínua e acentuada das taxas de juro e por importantes avanços institucionais. É imprescindível a disponibilidade de novas informações, em particular, que possibilitem o acompanhamento mais detalhado das modalidades de crédito com recursos direcionados, sobretudo os financiamentos imobiliários, cujo dinamismo tem contribuído para a redução do déficit habitacional no País. A principal alteração compreende a cobertura dos dados relativos a concessões, taxas de juros, prazos e índices de inadimplência que passam a serem estendidos ao segmento de crédito direcionado e também se fez necessário aprofundar o detalhamento do arcabouço estatístico, de modo a possibilitar a identificação das modalidades mais relevantes, bem como reduzir a participação relativa das operações de crédito não classificadas – incorporadas em “outros créditos”.

Total credit card debt in the UK grew by **** billion British pounds between October and November 2023, now reaching a similar level of debt as seen in early 2017. The annual growth rate of credit card debt stayed about the same in March 2025, reaching *** percent when compared to March 2024. The growth rate in 2023 has been relatively consistently since May, which may potentially be attributed to growing interest rates and the cost of living crisis.

About one out of four of all Mastercard credit cards issued worldwide were found within the United States, as of the first quarter of 2025. The value inside the U.S. grew noticeably over the past few years. By mid-2023, the number of credit cards issued in the United States grew by more than ** million when compared to the same period in 2022. Compared to Visa credit cards, Mastercard is issued relatively more often outside the U.S. than inside the country. Mastercard in the United States A sizable share of Americans use Mastercard credit cards regularly. Mastercard and Visa dominate the purchase card market, both in the United States and worldwide. Other leading credit card companies are Chase, American Express and Discovery. Credit card market One way that credit card companies earn money is by collecting a fee. These fees include interest on balances and transaction fees, both of which depend on the number of purchase transactions.

Lending Rate: per Annum: Pre-Fixed: Corporate Entities: Prepaid Credit Card Bills: Banco Commercial Investment Trus data was reported at 0.000 % pa in 03 Jul 2019. This stayed constant from the previous number of 0.000 % pa for 02 Jul 2019. Lending Rate: per Annum: Pre-Fixed: Corporate Entities: Prepaid Credit Card Bills: Banco Commercial Investment Trus data is updated daily, averaging 0.000 % pa from Jan 2012 (Median) to 03 Jul 2019, with 1867 observations. The data reached an all-time high of 0.000 % pa in 03 Jul 2019 and a record low of 0.000 % pa in 03 Jul 2019. Lending Rate: per Annum: Pre-Fixed: Corporate Entities: Prepaid Credit Card Bills: Banco Commercial Investment Trus data remains active status in CEIC and is reported by Central Bank of Brazil. The data is categorized under Brazil Premium Database’s Interest and Foreign Exchange Rates – Table BR.MB038: Lending Rate: per Annum: by Banks: Pre-Fixed: Corporate Entities: Prepaid Credit Card Bills. Lending Rate: Daily: Interest rates disclosed represent the total cost of the transaction to the client, also including taxes and operating. These rates correspond to the average fees in the period indicated in the tables. There are presented only institutions that had granted during the period determined. In general, institutions practicing different rates within the same type of credit. Thus, the rate charged to a customer may differ from the average. Several factors such as the time and volume of the transaction, as well as the guarantees offered, explain the differences between interest rates. Certain institutions grant allowance of the use of the term overdraft. However, this is not considered in the calculation of rates of this type. It should be noted that the overdraft is a modality that has high interest rates. Thus, its use should be restricted to short periods. If the customer needs resources for a longer period, should find ways to offer lower rates. The Brazilian Central Bank publishes these data with a delay about 20 days with relation to the reference period, thus allowing sufficient time for all Financial Institutions to deliver the relevant information. Interest rates presented in this set of tables correspond to averages weighted by the values of transactions conducted in the five working days specified in each table. These rates represent the average effective cost of loans to customers, consisting of the interest rates actually charged by financial institutions in their lending operations, increased tax burdens and operational incidents on the operations. The interest rates shown are the average of the rates charged in the various operations performed by financial institutions, in each modality. In one discipline, interest rates may differ between customers of the same financial institution. Interest rates vary according to several factors, such as the value and quality of collateral provided in the operation, the proportion of down payment operation, the history and the registration status of each client, the term of the transaction, among others . Institutions with “zero” did not operate on modalities for those periods or did not provide information to the Central Bank of Brazil. The Central Bank of Brazil assumes no responsibility for delay, error or other deficiency of information provided for purposes of calculating average rates presented in this

Lending Rate: per Annum: Pre-Fixed: Corporate Entities: Prepaid Credit Card Bills: BIORC CFI data was reported at 0.000 % pa in 03 Jul 2019. This stayed constant from the previous number of 0.000 % pa for 02 Jul 2019. Lending Rate: per Annum: Pre-Fixed: Corporate Entities: Prepaid Credit Card Bills: BIORC CFI data is updated daily, averaging 0.000 % pa from Jan 2012 (Median) to 03 Jul 2019, with 1867 observations. The data reached an all-time high of 0.000 % pa in 03 Jul 2019 and a record low of 0.000 % pa in 03 Jul 2019. Lending Rate: per Annum: Pre-Fixed: Corporate Entities: Prepaid Credit Card Bills: BIORC CFI data remains active status in CEIC and is reported by Central Bank of Brazil. The data is categorized under Brazil Premium Database’s Interest and Foreign Exchange Rates – Table BR.MB038: Lending Rate: per Annum: by Banks: Pre-Fixed: Corporate Entities: Prepaid Credit Card Bills. Lending Rate: Daily: Interest rates disclosed represent the total cost of the transaction to the client, also including taxes and operating. These rates correspond to the average fees in the period indicated in the tables. There are presented only institutions that had granted during the period determined. In general, institutions practicing different rates within the same type of credit. Thus, the rate charged to a customer may differ from the average. Several factors such as the time and volume of the transaction, as well as the guarantees offered, explain the differences between interest rates. Certain institutions grant allowance of the use of the term overdraft. However, this is not considered in the calculation of rates of this type. It should be noted that the overdraft is a modality that has high interest rates. Thus, its use should be restricted to short periods. If the customer needs resources for a longer period, should find ways to offer lower rates. The Brazilian Central Bank publishes these data with a delay about 20 days with relation to the reference period, thus allowing sufficient time for all Financial Institutions to deliver the relevant information. Interest rates presented in this set of tables correspond to averages weighted by the values of transactions conducted in the five working days specified in each table. These rates represent the average effective cost of loans to customers, consisting of the interest rates actually charged by financial institutions in their lending operations, increased tax burdens and operational incidents on the operations. The interest rates shown are the average of the rates charged in the various operations performed by financial institutions, in each modality. In one discipline, interest rates may differ between customers of the same financial institution. Interest rates vary according to several factors, such as the value and quality of collateral provided in the operation, the proportion of down payment operation, the history and the registration status of each client, the term of the transaction, among others . Institutions with “zero” did not operate on modalities for those periods or did not provide information to the Central Bank of Brazil. The Central Bank of Brazil assumes no responsibility for delay, error or other deficiency of information provided for purposes of calculating average rates presented in this