Financial Conditions and Borrowing Behavior of University Students during the COVID-19 Pandemic: Evidence from Bangladesh

Abstract

The COVID-19 pandemic presented significant challenges to university students. This study explores the financial conditions and borrowing behavior of university students during the pandemic in Bangladesh. The study used a sample of 840 students from major public universities in Bangladesh and applied bivariate analyses and mean comparison tests. The findings reveal that a substantial portion of the students experienced job loss and financial problems during the pandemic. Approximately 50% of the students had substantial loan burdens, 16.31% borrowed from formal institutions, and 39.17% borrowed multiple times. Furthermore, about 20% perceived borrowing costs as high, and most struggled to make regular installment payments. Analyzing the subgroups, women leaned more on family support for income, while men had independent income sources. Job loss affected a greater percentage of females than males. Females also borrowed more from formal sources, while more males perceived borrowing costs as negligible. In terms of urban and rural comparisons, rural students relied more on family support, while urban students had independent income sources. Financially literate students encountered more job loss and financial trouble and borrowed more from informal and low-cost sources compared to their less financially literate counterparts. The study suggests grants, subsidies, and reduced educational expenses for students who faced job loss, financial trouble, and burdensome debt.

1. Introduction

The global impact of the COVID-19 pandemic has influenced the socioeconomic circumstances of individuals worldwide, including students [1,2,3,4,5]. In Bangladesh, a developing nation, the pandemic has taken a substantial toll on various aspects of development, most notably the economy. Research has shown a significant increase in poverty levels and food insecurity during this period [6,7]. The initial response of the Bangladesh government to the pandemic included the imposing of extended lockdowns, which required university students to leave their dormitories and private accommodations around the campuses [4,8,9,10,11]. This relocation had ramifications not only for their education but also for their financial earnings and overall livelihoods [5,12]. A substantial portion of university students in Bangladesh rely on part-time income to support their educational and daily needs, compounded by the fact that many come from middle-income households that experienced income reductions due to the lockdown and economic turmoil. Additionally, the transition to online classes during lockdown introduced new financial burdens associated with the purchase of communication devices and internet access [11]. The sudden contraction of income sources and the simultaneous increase in educational expenditures led numerous students to exhaust their limited savings and resort to borrowing from both formal and informal sources. Although government financial aid programs were designed to mitigate pandemic-induced economic losses, they rarely focused on supporting students [13]. Remarkably, there has been a dearth of comprehensive studies that examined the financial circumstances of Bangladeshi students during the pandemic.

This study aims to bridge this research gap by providing information on the financial conditions and borrowing tendencies of university students in Bangladesh. Furthermore, this research delves into students’ financial and debt literacy and its association with their financial management skills. Specifically, the study hypothesizes that the financial conditions of university students deteriorated significantly during the COVID-19 pandemic, for which they needed to borrow from formal and informal sources. Moreover, students’ socioeconomic background and financial literacy determine their financial management skills. The contribution of this study lies in two key aspects. First, to the best of my knowledge, it represents the first effort in Bangladesh to systematically assess the financial conditions and borrowing behavior of university students amid the pandemic. Second, it furnishes evidence illustrating how financial and debt literacy can empower students to effectively manage their financial needs and concerns.

The impact of the COVID-19 pandemic on Bangladeshi society and economy has been extensive. Despite significant economic growth in recent decades, the pandemic posed significant challenges for the country. In particular, more than 90% of the total population of Bangladesh earns less than BDT300 (approximately $3) per day, according to World Bank statistics [14]. The economy’s growth was highly dependent on the success of the garment industry and remittances from Bangladeshi workers [15,16], both of which were negatively affected by the pandemic. Furthermore, the economy lacked the resilience to absorb the shock, as Ahamed [17] highlighted the decline in production in all sectors, including agriculture, industrial (especially the ready-made garment industry), foreign remittances, and the service sector. The education system suffered due to widespread school closures, and UNICEF’s [18] recent study revealed a 19% reduction in average household income. Roberton et al. [19] estimated the possible severe consequences for around 30,000 children under the age of five. UNICEF [18] further noted that the working-age population, including students, experienced significant income losses. The IMF [20] projected similar challenges for the Bangladeshi economy, which could lead to social unrest. The prolonged economic shutdown negatively affected both formal and informal sector production [21]. Bangladesh also observed a decrease in employment rates and income for the working-age population, comprising 55% of the total population [21]. Given the reduction in family income and the loss of formal and informal employment opportunities, university students faced financial difficulties and found it challenging to maintain their education. Consequently, they were left with limited options—either utilizing their small savings or resorting to borrowing from formal and informal sources—to navigate these unprecedented financial challenges. The deteriorating financial conditions and subsequent borrowing behavior of students in Bangladesh share similarities with those of other countries. Russel et al. [22] and Cornett and Fletcher [23] reported that the reduction in the financial well-being of students is a global phenomenon. They had to rely more on family income, which also decreased during the pandemic.

The borrowing behavior of university students in Bangladesh, even before the pandemic, has received limited attention in research. In fact, the income and spending patterns of university students remain relatively unknown. Unlike developed nations, Bangladesh lacks government-initiated student loan schemes. Formal financial institutions offer student loans with high-interest rates and stringent terms and conditions, making them less accessible to students [24,25]. Consequently, borrowing from personal sources becomes the main viable solution for financing educational and personal expenses when income falls short. As a result, borrowing is a prevalent practice in Bangladesh, primarily due to students’ insufficient savings to address unexpected financial challenges. Furthermore, access to formal financial institutions is limited, and the borrowing process is often complex and time-consuming. This leads students to opt for informal sources of credit. Among informal options, friends and family members constitute the main sources of credit. Although the specific terms and conditions of these loans, including interest rates, are not well documented, they are generally considered more flexible and convenient. Alternatively, some students may borrow from pawnbrokers and informal lenders, often at exorbitant interest rates. Pawnbrokers lend money on a daily or weekly basis, with or without collateral, at much higher interest rates compared to formal financial institutions. Studies have highlighted the potential exploitation of borrowers by informal lenders. Another significant source of informal credit is microfinance institutions (MFIs). Particularly in rural areas, people often borrow from MFIs for business purposes, although they frequently redirect the funds toward personal needs such as education expenses for their children. Although the interest rates and terms offered by MFIs are not as harsh as those of pawnbrokers, they remain considerably higher than those of formal financial institutions. In some instances, individuals may borrow from multiple sources, including those with exceedingly high-interest rates, and may borrow amounts that exceed their repayment capacity. Over time, this can lead to overborrowing and eventually defaulting on payments.

How students can manage the funding requirement to complete higher education is a matter of concern. Given alternative sources of funds with their pros and cons, students need to understand which source will provide them with lower costs. Financial literacy, which has long been conceived as a rational decision-making instrument, could be helpful for students in deciding the sources of funds [26,27,28,29]. Financial literacy enhances people’s money management skills and opens alternative investment opportunities. Moreover, financially literate people understand the value of information and make profit-maximizing decisions. Lusardi and Mitchell [30] provided a theoretical background on the value of financial knowledge accumulation in economic decision-making. Thus, financial literacy could be an effective instrument for students to decide which loan sources to take. Moreover, literacy helps them build sound savings behavior, which could ultimately reduce the need to take loans. Lusardi and Tufano [31] and Sevim et al. [32] found that debt literacy reduces people’s tendency to overborrow. Therefore, a study of students’ borrowing behavior from the point of view of financial literacy seems important.

2. Data and Methods

2.1. Variables

The main focus of this study revolves around the examination of the financial conditions and borrowing behavior of students during the pandemic. Assessment of financial conditions encompasses various financial indicators, including household income, asset balance, sources of income, experiences of financial difficulties, and instances of job loss during the pandemic. The study also examines student borrowing behavior using a variety of debt-related indicators, such as current debt status, borrowing sources, primary reasons for borrowing, frequency of borrowing, cost associated with borrowing, and their ability to make regular installment payments. The financial conditions and borrowing behavior variables used in this study were designed following the studies of Sevim et al. [32] and Russel et al. [22].

In addition to investigating financial conditions and borrowing behavior, the study aims to explore potential variations in these aspects based on the students’ level of financial literacy. To measure financial literacy, the study adopted a conventional approach similar to the methodology outlined by Lusardi and Mitchell [33,34]. This involved utilizing three distinct questions, with each question contributing equally to calculate the financial literacy scores of the participants. This approach allowed for a comprehensive evaluation of the interaction between financial literacy, financial conditions, and borrowing behavior among students. The questions and options were as follows:

• Suppose you had Tk100 in a savings account, and the interest rate was 2% per year. After 5 years, how much do you think you would have in the account if you left the money to grow?

(i) More than Tk 102

(ii) Exactly Tk 102

(iii) Less than Tk 102

(iv) Do not know

2.

Assume that the interest rate on your savings account was 1% per year and inflation was 2% per year. After 1 year, how much would you be able to buy with the money in this account?

(i) More than today

(ii) Exactly the same

(iii) Less than today

(iv) Do not know

3.

Buying a company stock usually provides a safer return than a stock mutual fund. True or false?

(i) True

(ii) False

(iii) Do not know

In addition, this study also examines various demographic, socioeconomic, and psychological attributes of the students. These include factors such as gender, age, marital status, place of residence, level of education, employment status, overall satisfaction with their financial situation, concerns about future life, and their perspective on future prospects. By considering these diverse characteristics, the study aims to provide a comprehensive understanding of the students’ profiles and how these factors might interact with their financial conditions and borrowing behavior during the pandemic. The demographic, socioeconomic, and psychological variables used in this study were designed following the studies of Khan et al. [27,28], Sevim et. al. [32], and Russel et al. [22].

Table 1. Definition and measurement of variables.

Variables	Definition and Measurements
Financial conditions and borrowing behavior-related variables
Household income (hhincome)	Continuous variable: income of the entire household measured in Bangladeshi Taka (BDT).
Household assets (hhassets)	Continuous variable: balance of assets of the entire household measured in BDT.
Source of income (income_source)	Categorical variable: sources of income of a student where 1 = own income, 2 = family support, 3 = part-time income, 4 = scholarship, and 5 = others.
Financial trouble (fin_trouble)	Binary variable, which equals 1 if a student or family experienced financial trouble during the pandemic and 0 = otherwise
Loss of job (job_loss)	Binary variable, which equals 1 if a student a lost job during the pandemic and 0 = otherwise
Borrowing (loan_balance)	Categorical variable: the amount of loan taken during the pandemic where 1 = excessive loan with difficulty in payment, 2 = moderate loan with no difficulty in payment, 3 = low level of loan, and 4 = no loan or do not want to answer.
Sources of borrowing (loan_source)	Categorical variable: sources of loan taken during the pandemic where 1 = family and friends, 2 = financial institutions, 3 = microfinance institutions, 4 = informal lenders, and 5 = others.
Frequency of borrowing (loan_frequency)	Categorical variable: frequency of loan taken during the pandemic where 1 = once, 2 = twice, 3 = thrice, and 4 = more than three.
Cost of borrowing (loan_cost)	Categorical variable: cost of loan taken during the pandemic where 1 = low cost, 2 = moderate, 3 = high, and 4 = very high.
Regularity of installment payment (loan_reginstallment)	Binary variable, which equals 1 if a student failed to pay installment of loan during the pandemic and 0 = otherwise
Reasons for borrowing (loan_reasons)	Categorical variable: reasons for taking a loan during the pandemic where 1 = family expenditure, 2 = educational expenses, 3 = personal expenses, 4 = purchasing electronic devices, and 5 = others.
Demographic and socioeconomic variables
Male	Binary variable: 1 = male, 0 = female
Age	Continuous variable: age of students in 2022
Years of education	Continuous variable: number of years of education
Married	Binary variable: 1 = currently married, 0 = otherwise
Children	Binary variable: 1 = have at least 1 child, 0 = otherwise
Unemployed	Binary variable: 1 = currently not employed, 0 = otherwise
Area of residence	Binary variable: 1 = rural area, 0 = urban area
Financial literacy	Continuous variable: average score for the number of current answers from three financial literacy questions
Financial satisfaction	Ordinal variable for the statement “I am happy with my financial status”.
	1 = completely disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = completely agree
Anxiety about future	Ordinal variable for the statements “I have anxieties about life after 65 years of age” and “I have anxieties about life in the future” for individuals less than 65 years old and for those who were aged 65 years or above, respectively.
Myopic view of the future	Ordinal variable for the statement “Since the future is uncertain, it is a waste to think about it”.
	1 = completely disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = completely agree

shows the definitions and measurements of variables of this study.

2.2. Data

Data for this study were collected through a survey conducted by the Department of Banking and Insurance within a specific timeframe, from 5 June 2022 to 10 August 2022. The survey aimed to understand the socio-economic dynamics of students during the pandemic. The survey gathered information from students attending major public universities in Bangladesh, including Dhaka University, Jahangirnagar University, Rajshahi University, Chittagong University, Jagannath University, and Khulna University. While there are a total of 43 public universities and 103 private universities in Bangladesh, the chosen public universities are considered the oldest and largest, attracting students from various regions of the country. These universities were strategically selected to cover various geographical areas of Bangladesh. On the other hand, private universities are primarily located in major urban centers and typically have a smaller student population. Eligible participants were regular students aged 18 years or older enrolled in the major public universities in Bangladesh. In the first stage of the sampling procedure, a convenience sampling method was used to select six large public universities in the country from a population of 54 public and 112 private universities, with efforts made to ensure representation from various socioeconomic backgrounds. In the second stage, a simple random procedure was followed to select students from each university. The number of students from a university was determined based on the total number of students at that university. The total sample size of the study was determined by the standard formula:

$$n=\frac{Z^2·p·(1-p)}{E^2}$$

where Z stands for the level of confidence (95%), p stands for the estimated proportion (0.5), and E stands for the margin of error (5%). Considering these values and a design effect of 2, the minimum sample size stands at 770.

The questionnaire used in the survey was carefully designed, incorporating relevant factors and established principles. Where possible, existing validated scales were used to measure concepts. The survey included attention-check questions to verify the participants’ attentiveness and accurate responses. Responses that were incomplete or failed the attention checks were excluded from the analysis. The final dataset consisted of 840 complete responses, which accounted for 84% of the initial 1000 responses collected. This final sample was well distributed among different university students across the country, providing a balanced representation.

Table 2. Descriptive statistics.

Variable	Obs	Mean	Std. Dev.	Min	Max
Male	840	0.55	0.5	0	1
Age	840	22.96	1.51	19	27
Married	840	0.08	0.28	0	1
Residence_rural	840	0.48	0.5	0	1
Education	840	16.18	1.35	13	18
Employment	840	0.12	0.32	0	1
HHIncome (BDT)	840	35,698.94	29,112.99	15,000	135,000
HHAssets (BDT)	840	1,946,886.00	1,601,665.00	500,000	4,500,000
Finliteracy	840	0.71	0.33	0	1
Fin_satisfaction	840	3.06	1.13	1	5
Anxiety_future	840	3.99	0.98	1	5
Future myopia	840	3.66	1.15	1	5

shows the descriptive statistics of the key variables. The characteristics of the respondents are the following: 55% are male, the average age is 22.96 years (SD = 1.51), 8.00% are married, they have on average 16.18 (SD = 1.35) years of education, and they have a financial literacy score of 0.71 out of 1 (SD = 0.31). Statistics also show that 48.00% of the respondents live in rural areas, and 12% have part-time or full-time employment (SD = 32.00%). The average household assets and income of the respondents are BDT1,946,886 (SD = BDT1,601,665) and BDT35,699 (SD = BDT29,113), respectively. Regarding behavioral issues, respondents were found to show a mediocre orientation to the future versus the present (mean = 3.66, SD = 1.15) and financial satisfaction (mean = 3.06, SD = 1.13) but a relatively higher degree of anxiety about life in old age (mean = 3.99, SD = 0.98).

2.3. Methods

In conducting the data analysis for this study, mainly univariate descriptive statistics were utilized to examine the main variables of interest. This approach allowed to understand the distribution of financial conditions and borrowing behavior among university students, providing insight into the impact of the pandemic. Additionally, cross-tabulation was used to explore the relationships between financial conditions, borrowing behavior, and key socioeconomic variables such as gender and area of residence. This bivariate analysis allowed to observe how these factors intersected and potentially influenced students’ financial situations and borrowing patterns during the pandemic. To further deepen the analysis, mean comparison tests were conducted. Comparing means using a t-test is a common statistical method to determine if there is a significant difference between the means of two groups or conditions. An independent sample t-test was used to test the hypothesis that financial conditions and borrowing behaviors of university students differ by gender, area of residence, and financial literacy. The formulae used for the independent sample t-test are as follows:

$$t=\frac{\overline{X_1}-\overline{X_2}}{\sqrt{\frac{s_1^2}{n_1}-\frac{s_2^2}{n_2}}}$$

These tests facilitated a comparison of various aspects of financial conditions and borrowing behavior between respondents who were financially literate and those who were not. This comparison aimed to uncover any differences in financial management practices between these two groups, shedding light on whether financial literacy played a role in improving individuals’ control over their financial matters. Using these analytical methods, the study was able to provide valuable information on financial conditions, borrowing behavior, and the role of financial literacy among university students during the pandemic. Russel et al. [22] employed a similar method to assess student financial well-being, need satisfaction, and college persistence during the pandemic. However, the study’s limitation lies in the absence of control for endogeneity, which remains a concern. Advanced statistical tools like IV regression were not utilized due to the unavailability of an instrumental variable in the dataset.

3. Results

This section delineates the financial conditions and borrowing behavior of university students during the pandemic in Bangladesh.

3.1. Financial Conditions of University Students during the Pandemic

The financial conditions of university students during the pandemic were measured by using six variables. The first two variables include household income and household assets, which show the financial strength of the students’ families. The next three variables include the sources of income of the students, the loss of work during the pandemic, and the major financial problems faced by family members. Finally, the overall satisfaction of students with financial conditions during the pandemic were evaluated.

Table 3. Financial conditions of university students during the pandemic.

Financial Conditions Variables		Mean
HHIncome		BDT35,699
HHAssets		BDT1,946,886
Sources of income
	Salary	5%
	Family support	47%
	Part-time income	43%
	Scholarship	3%
	Others	2%
Job_loss		33%
Fin_trouble		66%
Fin_satisfaction		3.06

provides a univariate description of household income, assets, sources of income, job loss, financial difficulties, and financial satisfaction of the sample university students during the pandemic in Bangladesh. The results show that the mean household income of the students is BDT35,699, and the standard deviation is BDT29,112, while the mean household assets are BDT1,946,886, with a standard deviation of BDT1,601,665. Regarding the sources of income, the salary constitutes 5% of the income of the students. That means that only 5% of the students were employed in a formal job or business during the pandemic. As universities offer many executive programs, students with a full-time job or business are not rare. However, most students rely on family support (47%) and part-time income (43%). Moreover, 3% have scholarships, and 2% have other sources of income. As many industries were shut down for several months and physical movement was restricted, the loss of formal and informal jobs was measured in this study. The results show that 33% of the students experienced loss of job during the pandemic, which is a significant number given the percentage of students engaged in part-time and full-time jobs. Due to the long-term shutdown of universities and the restriction on physical movement, most of the students had to leave their university residence, leading most of them to lose part-time jobs in that area. The results also show that about 66% of the students have experienced financial trouble during the pandemic. As the household income and assets show, most households have a mediocre income capacity and a wealth base. Thus, a sudden loss of jobs or restriction in economic activities caused huge problems for most of the households. Despite this fact, their level of satisfaction with financial conditions, given the circumstances, is enumerated. The results show that they are moderately satisfied (mean = 3.06, SD = 1.13) with their current financial conditions.

3.2. Borrowing Behavior of University Students during the Pandemic

Borrowing behavior indicates the patterns and habits that individuals exhibit when it comes to borrowing money or obtaining credit. Due to financial trouble and loss of work, many students and their families had to take loans to support family and educational expenses. The borrowing behavior of the university student was measured through several variables such as borrowing position, sources of borrowing, frequency of borrowing, cost of borrowing, regularity of installment payment, and reason for borrowing.

Table 4. Borrowing behavior of university students during the pandemic.

Borrowing Behavior Variables		Percentage
Loan position
	Excessive	8.33%
	Moderate	41.19%
	Low	14.88%
	No debt or no answer	35.60%
Sources of borrowing	Financial institutions	16.31%
	Microfinance institutions	6.07%
	Family and friends	72.86%
	Informal lenders	0.60%
	No answer	4.16%
Borrowing frequency	One time	29.05%
	Two times	19.64%
	Three times	8.81%
	More than three times	39.17%
	No answer	3.33%
Cost of borrowing	Low	80.36%
	Moderate	12.86%
	High	2.27%
	Very high	4.52%
Regularity of installment	Yes	23.05%
	No	76.95%
Reasons for borrowing	Family expenditure	27.50%
	Educational expenditure	18.93%
	Personal expenditure	37.38%
	Purchasing devices	6.07%
	Others	10.12%

shows the scenario of borrowing behavior of university students during the pandemic. The results show that 8.33% of the respondents have an excessive loan balance, 41.19% have a moderate loan position, 14.88% have a low loan position, and the remaining 35.60% either have no debt or did not provide an answer regarding their debt status. Thus, a significant percentage of students have borrowed heavily during the pandemic. Regarding borrowing sources, 16.31% of the respondents borrow from financial institutions, 6.07% from microfinance institutions, 72.86% from family and friends, 0.60% from informal lenders, and 4.16% did not provide an answer regarding their borrowing sources. Thus, most of the students borrowed from family and friends, presumably at low or no cost. However, about 22% of the students had to borrow from formal sources with high-interest rates. To understand the cost of borrowing, they were asked about the magnitude of the cost of borrowing. The results show that 80.36% of the respondents perceive the cost of borrowing to be low, 12.86% perceive it to be moderate, 2.27% perceive it to be high, and 4.52% perceive it to be very high. Furthermore, 29.05% of respondents have borrowed only once, 19.64% have borrowed twice, 8.81% have borrowed three times, 39.17% have borrowed more than three times, and 3.33% did not provide an answer on their borrowing frequency. Thus, a significant percentage of the students had to borrow several times during the pandemic, indicating a sustaining financial problem suffered by them. The financial problem facing the household has been reflected in the regularity of the installment payments. The results show that 23.05% of the respondents paid their installments regularly, while 76.95% could not pay their installments regularly. Finally, regarding the reasons for borrowing, 27.50% of respondents borrowed for family expenses, 18.93% for educational expenses, 37.38% for personal expenses, 6.07% for purchasing devices, and 10.12% for other unspecified reasons.

3.3. Financial Conditions and Borrowing Behavior of Male and Female Students

The financial conditions and borrowing behavior of the male and female students are presented in

Table 5. Financial conditions and borrowing behavior of male and female students.

		Male	Female	Significance of Difference (Chi2)
Financial conditions
Sources of income	Family support	46.09%	54.01%	5.18 **
	Own source	53.91%	45.99%
Job_loss	Yes	35.60%	30.00%	2.90 *
	No	64.40%	70.00%
Fin_trouble	Yes	64.15%	67.92%	1.31
	No	35.85%	32.08%
Borrowing behavior
Loan position	Excessive	8.86%	7.75%	0.33
	Not excessive	91.14%	92.25%
Sources of borrowing	Formal	19.87%	25.67%	3.99 **
	Informal	80.13%	74.33%
Borrowing frequency	One time	26.78%	31.55%	2.29
	More than one time	73.22%	68.45%
Cost of borrowing	Negligible	82.29%	77.81%	2.63
	Not negligible	17.71%	22.19%
Regularity of installment payment	Yes	23.76%	22.93%	0.08
	No	76.24%	77.07%

Note: ** p < 0.05, * p < 0.1.

. The results of financial conditions between male and female students show that among males, 46.09% receive income from family support, compared to 54.01% among females, and 53.91% of males have their own source of income, compared to 45.99% among females. Among males, 35.60% have experienced job loss, while 30.00% of females have faced job loss, and 64.15% have faced financial trouble, while 67.92% of females have experienced the same. Regarding financial behavior, 8.86% of males have an excessive loan position, compared to 7.75% among females; 19.87% of males borrow from formal sources, while 25.67% of females do the same; 26.78% of males borrow only once compared to 31.55% among females; 82.29% of males consider the cost of borrowing negligible, compared to 77.81% among females; and 23.76% of males pay installments regularly, compared to 22.93% among females.

3.4. Financial Conditions and Borrowing Behavior of Urban and Rural Students

The financial conditions and borrowing behavior of urban and rural students are presented in

Table 6. Financial conditions and borrowing behavior of urban and rural students.

		Urban	Rural	Significance of Difference (Chi2)
Financial conditions
Sources of income	Family support	44.72%	54.13%	7.36 ***
	Own source	55.28%	45.87%
Job_loss	Yes	35.46%	30.95%	1.89
	No	64.54%	69.05%
Fin_trouble	Yes	63.54%	67.88%	1.74
	No	36.46%	32.12%
Borrowing behavior
Loan position	Excessive	6.53%	10.05%	3.35 *
	Not excessive	93.47%	89.95%
Sources of borrowing	Formal	19.10%	25.34%	4.69 **
	Informal	80.90%	74.66%
Borrowing frequency	One time	34.92%	23.52%	13.20 ***
	More than one time	65.08%	76.48%
Cost of borrowing	Negligible	84.92%	76.26%	9.94 ***
	Not negligible	15.08%	23.74%
Regularity of installment payment	Yes	19.85%	26.42%	5.05 **
	No	80.15%	73.58%

Note: *** p < 0.01, ** p < 0.05, * p < 0.1.

. The results of financial conditions between urban and rural students show that among urban students, 44.72% receive income from family support, compared to 54.13% among rural students, and 55.28% of urban students have their own source of income, compared to 45.87% among rural students. The results further show that among urban students, 35.46% have experienced job loss, while 30.95% of rural students have faced job loss, and 63.54% of urban students have faced financial difficulties, while 67.88% of rural students have experienced the same. Regarding borrowing behavior, 6.53% of urban students have an excessive loan position, compared to 10.05% among rural students, 19.10% of urban students borrow from formal sources, while 25.34% of rural students do the same, 34.92% of urban students borrow only once, compared to 23.52% among rural students, 84.92% of urban students consider the cost of borrowing negligible, compared to 76.26% among rural students, and 19.85% of urban students pay installments regularly, compared to 26.42% among rural students.

3.5. Financial Conditions and Borrowing Behavior of Financially Literate and Illiterate Students

The pandemic underscored the importance of financial literacy. Individuals needed to make informed borrowing decisions, especially when navigating various lending platforms and dealing with complex terms and conditions. This study measures the financial conditions and investment behavior of financially literate and less literate students. For this purpose, two subsamples were created based on financial literacy scores. Students in the financially literate group are those who scored more than 0.66, while the financially less literate group is comprised of students who score 0.33 or less on the financial literacy index.

Table 7. Financial conditions and borrowing behavior between financially literate and less literate students.

		Financially Less Literate	Financially Literate	Significance of Difference (Chi2)
Financial conditions
Sources of income	Family support	51.96%	49.00%	0.49
	Own source	48.04%	51.00%
Job_loss	Yes	16.48%	37.60%	27.89 ***
	No	83.52%	62.40%
Fin_trouble	Yes	53.07%	69.31%	16.48 ***
	No	46.93%	30.69%
Investment behavior
Loan position	Excessive	6.04%	8.98%	1.61
	Not excessive	93.96%	91.02%
Sources of borrowing	Formal	33.52%	19.33%	16.50 ***
	Informal	66.48%	80.67%
Borrowing frequency	One time	32.42%	28.01%	1.35
	More than one time	67.58%	71.99%
Cost of borrowing	Negligible	75.27%	81.74%	3.77 *
	Not negligible	24.73%	18.26%
Regularity of installment payment	Yes	20.88%	24.16%	0.86
	No	79.12%	75.84%

Note: *** p < 0.01, * p < 0.1.

shows the financial conditions and investment behavior of financially literate and less literate students. The results show that 48.04% of financially less literate students have their own source of income, compared to 51.00% among financially literate students, while 51.96% of financially less literate students have received family support, compared to 49.00% of financially literate students. The chi-squared value shows that there is no significant difference between financially literate and less literate students in terms of sources of income. Regarding job loss, 16.48% of less literate students have experienced job loss, while 37.60% of financially literate individuals have faced job loss, and 83.52% of less literate have not experienced job loss, compared to 62.40% among financially literate students. The chi-squared value shows that there is a significant difference between financially literate and less literate students in terms of experiencing job loss. Financially literate students comparatively lost more jobs during the pandemic. Regarding financial trouble, 53.07% of less literate students have experienced financial trouble, while 69.31% of financially literate individuals have experienced financial trouble, and 46.93% of less literate have not experienced financial trouble, compared to 30.69% among financially literate students. The chi-squared value shows that there is a significant difference between financially literate and less literate students in terms of experiencing financial trouble. Financially literate students comparatively faced more financial trouble during the pandemic.

Regarding borrowing behavior, the results show that 6.04% of less financially literate students had excessive loan burden, compared to 8.98% among financially literate students, while 93.96% of less financially literate students had normal debt burden, compared to 91.02% of financially literate students. The chi-squared value shows that there is no significant difference between financially literate and less literate students in terms of debt position. The results further show that 66.48% of less financially literate students borrowed from informal sources, including family and friends, compared to 80.67% among financially literate students, while 33.52% of less financially literate students borrowed from formal sources, including financial institutions and microfinance institutions, compared to 19.33% of financially literate students. The chi-squared value shows that there is a significant difference between financially literate and less literate students in terms of source of borrowing. Financially literate students borrowed more from informal sources compared to their less financially literate counterparts. Regarding borrowing frequency, 32.42% of less financially literate students borrowed once during the pandemic, compared to 28.01% among financially literate students, while 67.58% of less financially literate students borrowed multiple times, compared to 71.99% of financially literate students. The chi-squared value shows that there is no significant difference between financially literate and less literate students in terms of frequency of borrowing. The cost of borrowing situation shows that 75.27% of less financially literate students borrowed from least cost sources, compared to 81.74% among financially literate students, while 24.73% of less financially literate students borrowed from costly sources, compared to 18.26% of financially literate students. The chi-squared value shows that there is a significant difference between financially literate and less literate students in terms of cost of borrowing. Financially literate students borrowed less from costly sources compared to their less financially literate counterparts. Finally, regarding the regularity of loan installment payments, 20.88% of less financially literate students paid installments regularly, compared to 24.16% among financially literate students, while 79.12% of less financially literate students failed to pay installments regularly, compared to 75.84% of financially literate students. Although a significant percentage of both financially literate and less literate students did not pay their instalments regularly during the pandemic, the chi-squared value shows that there is no significant difference between the groups in terms of the regularity of instalment payment.

4. Discussion

The COVID-19 pandemic had a profound and far-reaching impact on the financial landscape of Bangladesh’s population [4,6,7]. Among the primary repercussions were reductions in household income and assets, with the most pronounced effects felt by those who experienced job losses or had their working hours cut. The long-lasting shutdowns of the pandemic essentially halted the operations of the major industries, leaving only a few essential and exporting sectors active [4,8,9,10,11]. A considerable number of households were forced to draw on their savings in order to navigate the economic adversities caused by the pandemic. This trend was especially prevalent among people who faced job loss or business closures. Consequently, a notable proportion of Bangladeshi households encountered increased financial pressures amid the pandemic. This predicament was further exacerbated by factors such as decreased income, increased expenses for essential commodities, and limited employment opportunities [4,10,11].

University students form a distinct demographic group that faced numerous challenges throughout the pandemic. They endured the loss of a pivotal academic year and the inability to adequately prepare for entering the job market. Furthermore, the extended closure of universities forced them to relocate, resulting in the forfeiture of part-time employment opportunities [11,35]. The findings of this study concerning the financial situation of university students during the pandemic underscore that the majority of these students come from middle-income families. These students rely primarily on family support and part-time earnings to cover their educational and daily living costs. Significantly, a substantial proportion of students lost their part-time income sources during the pandemic, and a significant portion encountered financial difficulties. This trend was more pronounced among male students and those who had a higher degree of financial knowledge. This is not unexpected, as male students and those more financially literate were more frequently engaged in part-time work, making them particularly susceptible to job losses. Nevertheless, although fewer in number than males, female students also experienced a significant loss of part-time jobs during the pandemic. This joblessness is particularly concerning for female students, as they have fewer opportunities to find employment and may face more difficulties in regaining their jobs when the situation returns to normalcy. These results are consistent with a comprehensive investigation conducted in Dhaka and Chittagong, two major cities in Bangladesh, which revealed that 68% of individuals experienced job loss during the pandemic and faced substantial financial strain [36].

The borrowing behavior of Bangladeshi individuals underwent significant changes during the COVID-19 pandemic. With the depletion of their limited savings, many turned to borrowing to navigate the financial hardships precipitated by the pandemic [37,38]. The combination of job losses, reduced income, and higher expenses created an increased demand for external financial assistance [38,39]. The purposes for borrowing varied, covering from daily expenditures and medical emergencies to educational costs and business investments. This period highlighted the diverse and varied financial needs of the population. A considerable portion of individuals sought financial relief from informal sources, including friends, family members, and local moneylenders [38]. While these sources offered swift access to funds, they often came with elevated interest rates and potential risks. The pandemic expedited the uptake of digital lending platforms, as many individuals turned to mobile apps and online avenues for loans. These platforms provided convenience and expedited processing compared to traditional banking routes. Microfinance institutions emerged as crucial players, extending credit to individuals with lower incomes and small business owners [38]. These institutions carefully adjusted their services to accommodate the economic challenges that arise from the pandemic. Government stimulus packages also played a role in supporting individuals and businesses affected by the pandemic [40]. Although these aids offered respite from immediate financial burdens, they were not sufficient to sustain daily livelihoods. Despite increasing borrowing, ongoing economic uncertainties resulted in repayment difficulties. Borrowers faced challenges in meeting their repayment commitments, prompting discussions about loan restructuring and relief measures.

The study’s findings indicate that approximately 64% of university students resorted to borrowing during the pandemic to cover both living and educational expenses. Among these, about 50% borrowed significant sums, predominantly from informal sources such as family and friends. A few students also borrowed from formal financial institutions, often characterized by high-interest rates. In particular, most of the students faced difficulties in making regular installments of loans, underscoring their financial difficulties. Subsample analysis revealed that female and rural students leaned more towards borrowing from formal financial institutions with elevated interest rates. Additionally, rural students experienced more irregular instalment payments. Interestingly, financially literate students exhibited positive borrowing behavior, primarily relying on informal sources with low or no interest rates. This is consistent with previous research suggesting that financially literate individuals make more informed borrowing decisions [31]. However, it should be noted that financially literate students also encountered more instances of job loss and financial troubles compared to their less financially literate counterparts. This could be attributed to the increased involvement of financially educated students in the work, possibly leading to job loss during the pandemic.

There are some limitations in this study that readers should bear in mind when interpreting the findings. First, the sample drawn from four national universities might not be representative of all student populations. Specifically, students from a large number of private universities could exhibit different financial circumstances and borrowing patterns during the pandemic. Second, despite unanimous consent, some students were reluctant to divulge their financial and borrowing data. Consequently, a portion of the observations had to be excluded from the sample. Nonetheless, this study offers initial evidence on the financial conditions and borrowing tendencies of university students. Future research should aim for greater inclusion in sample selection and conduct a more comprehensive exploration of risk factors linked to financial distress.

5. Conclusions

In the context of a lack of study on the financial conditions and borrowing behavior during the pandemic, this study investigates selected measures of financial conditions and borrowing behavior of university students of Bangladesh. The results show that most students rely on family support and part-time income to cover educational and living expenses. A substantial percentage experienced job loss during the pandemic and faced financial trouble. The results further show that about 50% of the students had burdensome loan balances, 16.31% borrowed from formal institutions, 39.17% borrowed multiple times, about 20% perceived borrowing costs as high, and most students could not pay installments regularly. The reasons for borrowing included family expenses (27.50%), educational expenses (18.93%), personal expenses (37.38%), device purchases (6.07%), and other reasons (10.12%). The subsample analysis shows that females relied more on family support for income (54.01% vs. 46.09%), while males had their own sources (53.91% vs. 45.99%). Job loss affected 30.00% of females and 35.60% of males. In terms of borrowing, more females borrowed from formal sources (25.67% vs. 19.87%), and more males perceived borrowing costs as negligible (82.29% vs. 77.81%). When comparing rural and urban students, family support was found to be more common among rural students (54.13% vs. 44.72%), while urban students had their own income sources (55.28% vs. 45.87%). Finally, financially literate students experienced more job loss and financial trouble and borrowed more from informal and least-cost sources compared to less financially literate students.

The persistent financial concerns and overwhelming burdens faced by students have had a lasting impact, leading to a decline in their educational performance and increased psychological distress. In particular, female students, who often lack sufficient opportunities for higher education, may find their aspirations for academic success diminishing due to income loss and resorting to borrowing. Unfortunately, despite the revival of economic activities in the country, there appears to be a lack of visible efforts aimed at addressing the financial challenges experienced by students.

The results of this study have important policy implications for the government and university authorities in Bangladesh and many other countries where students have faced similar challenges during the pandemic. The disruption of educational activities has been a widespread issue in many nations, resulting in students suffering from income loss and losing valuable educational years. Moreover, there have been instances of developing psychological distress due to financial hardships. As students have to go through a prolonged period of job loss and financial trouble along with burdensome debt, authorities need to provide them with financial incentives and reduce educational expenses so that they can continue with their studies. Furthermore, the ability of financially literate students to manage financial trouble better than their less financially literate counterparts suggests that financial literacy should be embedded in the educational curriculum so that all students can manage unforeseen contingencies successfully.