Evaluation of the Canada Student Financial Assistance Program: Loan Repayment

Management response

The Learning Branch agrees with this recommendation.

Currently, the Learning Branch regularly engages with internal and external stakeholders, as well as provinces and territories, as part of broader awareness-building activities and information sharing on student financial assistance supports. Going forward, the Learning Branch will explore opportunities for targeted approaches to enhance the awareness of available student financial assistance supports, and to help improve financial literacy.

Within the Learning Branch, the Canada Student Financial Assistance Program undertakes several activities to increase awareness of Canada Student Grants and Canada Student Loans, as well as supports that may be available to those in repayment, such as the Repayment Assistance Plan. The Canada.ca website provides clear information on the support available, as well as information on how to apply, depending on one's province or territory of residence. In addition, from time to time, Employment and Social Development Canada runs various communications and social media campaigns to promote these supports.

Concerning the repayment of Canada Student Loans, targeted awareness strategies help to maintain the status of borrowers whose loans are in good standing, facilitate access to repayment assistance for those who need it, and help those who are delinquent bring their loans back into good standing. Through its third-party Service Provider, the Program communicates and promotes repayment options, including the Repayment Assistance Plan. Targeted communications also provide information to delinquent borrowers to help prevent them from defaulting on their loans.

The Program is also collaborating with the Financial Consumer Agency of Canada, provinces and territories, and other stakeholders to develop new tools and resources to improve the awareness of student financial assistance support. This includes developing an approach to test the impact of financial literacy training modules on applicants and borrowers entering repayment.

Going forward, the Canada Student Financial Assistance Program will continue to gather and analyze information on student loan borrowers and will explore opportunities to leverage findings from this analysis to develop targeted strategies to promote and enhance awareness of support.

Management action plan

Recommendation 1.1: The Learning Branch will explore opportunities for targeted approaches to enhance the awareness of available student financial assistance supports, and to help improve financial literacy.

Completion Date: March 2025

Recommendation 1.2: The Canada Student Financial Assistance Program will continue to gather and analyze information on student loan borrowers and will explore opportunities to leverage findings from this analysis to develop targeted strategies to promote and enhance awareness of supports.

Completion Date: September 2026

Management response

The Learning Branch agrees with this recommendation.

The evaluation indicates an overall improvement in repayment outcomes from a student loan borrower's perspective. However, the evaluation also suggests through early evidence that borrowers who recently started repaying still face challenges meeting their student loan obligations.

The Canada Student Financial Assistance Program provides several supports for those who face challenges in repaying their loans. Through the Repayment Assistance Plan, the Government assists eligible borrowers who cannot make their monthly loan payments. In November 2022, the Government implemented several enhancements to the Repayment Assistance Plan to help more borrowers in need. Key changes include increasing the income eligibility threshold for a single individual from $25,000 to $40,000, making commensurate increases to borrowers with families, and indexing the thresholds to inflation. In fact, as of August 2023, the threshold for a single borrower is now $42,720. In addition, monthly payments have been reduced from 20 % to 10 % of a borrower's family income for Repayment Assistance Plan recipients who are above the no-payment income thresholds.

In addition, as of April 1st, 2023, the Government has eliminated the accrual of interest on Canada Student Loans. This builds on a 2-year, temporary waiver of interest accrual, and represents a significant investment to reduce the burden of loan repayment on students, while providing predictability in the face of fluctuating interest rates.

Given that these significant changes have only recently been implemented, the Learning Branch's view is that time is needed to measure the impact of these changes on the repayment behaviour and outcomes of borrowers.

The Learning Branch assesses repayment trends on an ongoing basis as part of regular program management. The Canada Student Financial Assistance Program monitors key indicators, such as the percentage of loans repaid each year, the 3-year default rate, and uptake of the Repayment Assistance Plan. The Program also works with its third-party Service Provider to conduct detailed analysis by demographic groups, track repayment behavior and patterns, and review default rates. In addition, the Office of the Chief Actuary produces an annual Actuarial Report, which includes trends and forecasts on the amount of student loans issued, the utilization of the Repayment Assistance Plan, and information on defaulted loans. That report is also used to monitor trends in repayment for the Program.

In the short term, the Canada Student Financial Assistance Program will continue to monitor trends in repayment using these existing mechanisms and processes.

It is anticipated that, in 5 years' time, there will be sufficient data to draw evidence-based observations and conclusions about the impact of these investments. This is because default is measured retrospectively, and it takes 3 years to measure a single repayment cohort's default rate. The 5-year timeframe will allow the analysis of 3 consecutive repayment cohorts, which will account for the multiple factors that could influence default rates. Therefore, 5 years from now, the Canada Student Financial Assistance Program will conduct a more thorough analysis to assess the impact of recent program changes on repayment, including repayment length, delinquency, and default.

Management action plan

Recommendation 2.1: 5 years from now, the Canada Student Financial Assistance Program will conduct analysis to assess the impact of recent program changes on repayment, including repayment length, delinquency, and default.

Completion Date: March 2029

Repayment outcomes: The share of borrowers who paid in full is higher in recent^{Footnote 15} cohorts than in earlier cohorts, suggesting improved repayment ability

Figure 7 below shows the cumulative percentage of borrowers who paid their student loan in full at the end of each year since they began their repayment. The shorter curves represent more recent cohorts, while longer curves represent earlier cohorts. A shift in the shorter curves upward indicates that a higher percentage of recent borrowers rapidly paid their student debt in full relative to earlier cohorts. This suggests that borrowers' ability to repay student loan may be stronger among recent cohorts.

For instance, by the end of the first year of entering repayment, 24% of borrowers who began repayment in 2020 to 2021 paid their student debt in full, compared to 19% of borrowers in 2007 to 2008. Three years after the start of the repayment, one-third of the most recent cohort of borrowers (2018 to 2019, 2019 to 2020) had repaid their student debt in full compared to 26% for the 2009 to 2010 cohort. (For more information, see Figure G1 in Appendix G).

• Source: Canada Student Loan Repayment Patterns: Insights of Administrative Data from 2007 to 2022 (2023)

On the other end of the repayment spectrum, loans in default represent the opposite outcome compared to loans paid in full.

Downward trend: Default rates decreasing among borrowers since the 2009 to 2010 cohort

In addition to the increased rate of full loan repayment observed over time, more recent cohorts demonstrated a decrease in default rates compared to earlier cohorts. Many factors may have contributed to the decline in default rates, including:

• a favourable macroeconomic climate and robust labour market conditions for graduates since 2008^{Footnote 16}
• improvements in the management of student loan portfolios
• better outreach and communication with borrowers
• successive enhancements in student financial assistance policies, in particular the Canada Student Grants and the introduction of the RAP (2009)

A detailed breakdown of CSFA Program expenditures spanning academic years 2009 to 2010 through 2021 to 2022 reveals that a main factor driving the escalation in total Program outlays has been the increase in grant disbursements (Figure G2 , Appendix G).

Figure 8 below shows a noticeable trend across cohorts, with earlier cohorts experiencing higher default rates each year than the recent cohorts. Cohort analysis also reveals that borrowers who entered repayment in the wake of the 2008 financial crisis (the 2009 to 2010 cohort) had less ability to repay their loan than other cohorts. This may be due to the long-lasting adverse effects of the crisis on economic and labor market conditions at the time of graduation and throughout the repayment period. Overall, these data demonstrate better student loan repayment outcomes over time (Figure G3, Appendix G).

• Source: Canada Student Loan Repayment Patterns: Insights of Administrative Data from 2007 to 2022 (2023)

This evaluation compared this default rate to the CSFA Program's key performance indicator, the 3-year cohort default rate. It shows the value of loans that enter repayment in a given academic year and default within 3 years to the value of all loans that entered repayment in that academic year.

This result is consistent with the 3-year default rate for direct loans for full-time students which has been on a downward trend over the last decade, indicating greater repayment ability. The 3-year default rate decreased from 15% in 2009 to 2010, to a low of 7% in 2019 to 2020.

Figure 9 below shows the 3-year default rates for direct loans (overall, based on loan amounts) and its distribution by institution type. It suggests that borrowers who attended university have much lower default rates compared to borrowers from other types of institutions (college and private institutions).

Studies have shown that borrowers’ income after graduation is strongly related to graduates’ level and field of study, with some exceptions. For example, bachelor graduates earned more than college graduates, which can explain their lowest default rate.^{Footnote 17}

• Source: Canada Student Loan Repayment Patterns: Insights of Administrative Data from 2007 to 2022 (2023)

Interaction between RAP and CSL default rate

Figure 10 below compares the percentage of borrowers in default from four different cohorts (2009 to 2010, 2012 to 2013, 2015 to 2016, and 2018 to 2019) to understand the interaction between the RAP and default rate. Most borrowers in default have never used RAP. There are two borrower groups in this analysis: those who have used RAP and those who have never used RAP.

• Source: Canada Student Loan Repayment Patterns: Insights of Administrative Data from 2007 to 2022 (2023)

Please note that borrowers who have used RAP have done so at any time after entering repayment until the end of the reporting year, compared to those who have never used RAP during repayment.

The default trends among borrowers who have never utilized the Repayment Assistance Plan (RAP) show a significant positive shift from earlier cohorts relative to recent ones. To illustrate this shift, consider that the default rate for the 2009 to 2010 cohort was 20%, whereas it dropped to 13% for the 2015 to 2016 cohort in the sixth year of repayment (See Figure G4 in Appendix G for more details).

This decrease in default rates among borrowers without RAP experience in recent cohorts is a notable improvement. Much of this improvement can be attributed to the fact that many low-income borrowers have become RAP users. This strengthens the overall repayment performance of those who have never used RAP.

Borrowers who have used RAP tend to maintain stable default rates initially but experience an upward trend in later years, particularly by the sixth year of repayment. Conversely, for borrowers who have never used RAP, default rates increase significantly in the first 2 years of repayment but then level off after 5 years.

Analyzing cohort data reveals that borrowers with any history of using RAP exhibit lower default rates in the early years of repayment compared to those without any RAP experience. For instance, in the second year of repayment for the 2009 to 2010 cohort, borrowers with RAP experience had a default rate of 4%, while those without any RAP history had a default rate of 15%.

However, this gap between the two groups gradually narrows over time, and by the sixth year of repayment, their default rates become more similar. At the end of year 13, the gap has narrowed to 2 percentage points from the initial gap of 11 percentage points at the end of year 2.

Similar patterns are observed in the 2012 to 2013 and 2015 to 2016 cohorts. In summary, borrowers with RAP experience^{Footnote 19} tend to utilize this assistance early in their repayment journey, preventing them from defaulting during that period. However, some of them may still fall into default once they are no longer on RAP. The faster increase in default rates in later years is due to their lower incomes during this time. This indicates that borrowers who have used RAP are more prone to default in the long term compared to those who have not.

Additionally, borrowers experiencing financial difficulties may become eligible for RAP, causing them to transition from the non-recipient group to the recipient group. This dynamic contributes to the increasing slope of the default curve in the later years of repayment. In some cohorts, the percentage of borrowers in default becomes higher among the recipient group in the later years of repayment.

Further analysis is necessary to fully understand why some borrowers who used RAP early in repayment later fell into default. One question to consider is whether the income threshold for RAP eligibility during the study period was too restrictive to provide effective assistance to borrowers with long-term repayment difficulties. The enhancements to RAP introduced in Budget 2021 present a valuable opportunity for conducting such an analysis.

Some borrowers may have defaulted early in their repayment period because they were not aware of the RAP or its benefits, leading them to miss payments and later default

The focus of this evaluation was not whether borrowers were aware of CSL repayment options. However, some surveys conducted by the CSFA Program indicate that most student loan borrowers are aware of the existence of the RAP. These results also show that certain groups are still unaware of this assistance and would be at risk of default.

Pulse surveys conducted by the CSFA Program in 2021 and 2022 that included RAP awareness components reported that most borrowers (76%) were aware of the RAP. Of those, one quarter (25%) claimed to have applied for RAP between April 2021 and June 2022. The CSFA Program 2021-2022 Annual Client Satisfaction Survey reports that 38% of respondents were not aware of the repayment options offered through the CSFA Program. Among those who knew about the options, well over four out of five (83%) indicated that they were aware of the RAP, making it the most widely known repayment option among borrowers. However, data from the annual client satisfaction surveys (2008 to 2022) and the Pulse surveys do not provide an understanding of the reasons why defaulting borrowers have not applied for RAP.

Debt-to-income ratios

The debt-to-income ratio is calculated as the ratio of the consolidated CSL to the borrowers’ gross household income. The distribution of this ratio examines whether borrowers still owe on their government student loans at the time of leaving school and at the time of survey, as presented on Figure 12 below.

• Notes:
• • n =2,420
  • This figure presents income as a question with scales where the respondent was asked to choose their approximate total household income in 2022 from a series of income bands. For calculating the ratio income bands were converted to numerical values using the mid-point of the scale for estimated household income.
  • Excludes borrowers who did not report their income. Excludes those with a government loan debt of $0 or who did not report their debt.
• Source: Survey for the Evaluation of the CSFA Program (2023)

Please note that there is no specific universally accepted threshold for an acceptable student loan debt to household annual income ratio. However, lower debt-to-income ratios would suggest that borrowers could manage their debt repayment over a shorter period while maintaining a reasonable level of financial flexibility.

More than one third^{Footnote 22} of borrowers reported having paid off their student loans at the time of survey. In addition, the data shows that among respondents who still had student loans at the time of the survey, around half (47%) of borrowers have a ratio below 20%, suggesting that they can manage their debt repayment while maintaining a reasonable level of financial flexibility.

77% of borrowers still owing on their CSL had a debt at the time of survey that was 50% or less of their annual household income in 2022. 24% of borrowers had a debt-to-income ratio of 10% or less, compared to only 12% when considering their CSL at the point of leaving school.

Scott-Clayton (2017) emphasizes the importance of Grade Point Average (GPA), family wealth, loan amount, and their debt-to-income ratio. Higher GPAs and greater family wealth are linked to a lower likelihood of default. While the likelihood of default decreases with the loan amount, it increases with the debt-to-income ratio, highlighting the individual's role in repayment.

Figure 13 below shows the distribution of CSL debt-to-income at the time of the survey based on the extent of which the borrower was using the RAP for making payment. The majority of borrowers (88%) had a debt-to-income ratio of 50% or less, while this was the case for borrowers under the RAP making affordable payments (50%) and zero payment (35%).

These data also highlight the effectiveness of government assistance, the RAP, in focussing on borrowers with lower financial capacity.

The purpose of the RAP is to support borrowers facing financial difficulties in repaying their student loans. The figure below shows that at the time of survey, fewer borrowers benefiting from the RAP (making affordable and zero payment) had a debt-to-income ratio of 50% or less, compared to borrowers who were not benefitting from the RAP.

• Notes:
• • n Paying without assistance = 1,125, n RAP - Making payments = 139, n RAP - Zero payment = 138
  • This figure presents income as a scale question with the respondent asked to choose their approximate total household income in 2022 from a series of income bands. For calculating the ratio income bands were converted to numerical values using the mid-point of the scale for estimated household income.
  • Excludes borrowers who did not report their income. Excludes those with a government loan debt of $0 or who did not report their debt.
• Source: Survey for the Evaluation of the CSFA Program (2023)

Repayment-to-Income ratios

The repayment-to-income ratio is the percentage of monthly gross household income used for student debt repayment. This ratio provides more information about the extent to which borrowers can afford to repay their loans without experiencing financial stress or difficulty.

Borrowers' repayment-to-income ratios were approximated by dividing reported monthly CSL payments at the time of the survey by monthly gross household income in 2022. Please note that a low ratio indicates a higher ability to repay, while a high repayment-to-income ratio may mean less flexibility in meeting loan obligations.

Survey data indicated that almost three-quarters (72%) of borrowers were making payments that were 5% or less of their monthly household income, while nearly two-thirds (65%) of RAP recipients were also making payments that were 5% or less of their monthly household income. Thus, these results indicate that a significant majority of borrowers well positioned themselves in terms of their ability to repay their student loan.

Figure 14 below shows the distribution^{Footnote 23} of monthly CSL payments at time of the survey to monthly household income based on whether the borrower was making payment with the support of the RAP or not. It excludes borrowers benefitting from the RAP and not making payments.

• Notes:
• • n Paying without assistance = 1,368, n RAP - Making payments = 144
  • This figure presents income as a scale question with the respondent asked to choose their approximate total household income in 2022 from a series of income bands. For calculating the ratio income bands were converted to numerical values using the mid-point of the scale for estimated household income.
  • Excludes borrowers who did not report their income. Excludes those with a government loan debt of $0 or who did not report their debt.
• Source: Survey for the Evaluation of the CSFA Program (2023)

In addition, Figure 15 below shows the distribution of total repayment-to-income ratios, calculated by dividing monthly payments on total student loans (CSLs plus loans from banks, other financial institutions, family members, friends, and other acquaintances) at the time of the survey by monthly gross household income in 2022.

The results indicate that more than half (56%) of borrowers were making payments that were 5% or less of their monthly household income at the time of the survey, followed by RAP recipients making payments (48%) and RAP recipients making zero payment (38%). Note that RAP borrowers making zero payment may be reporting payments for loans other than CSLs since their government student loans were being fully covered by RAP.

• Notes:
• • n Paying without assistance = 1,382, n RAP - Making payments = 149, n RAP - Zero payment = 42
• Source: Survey for the Evaluation of the CSFA Program (2023)

Similar analysis for previous cohorts of borrowers using most recent linked data^{Footnote 24} at Statistics Canada corroborate and point to a similar distribution. The analysis examines the ratio of CSL student loans payments to family income over the first year of repayment following loan consolidation. It also found that for most borrowers (more than 60%), their annual student loan repayments represented less than 5 % of their annual family income.

Using the same linked data, complementary analysis also examined the average annual CSL repayment to the family income ratio over different repayment periods. For example, for the cohort of borrowers who consolidated their loan in 2010 to 2011, the ratio was estimated over a period of up to 5 years following consolidation. The same analysis was conducted for the 2014 to 2015 cohort, but up to 2 years following consolidation. Overall, the average ratios of annual CSL repayment to family income ratio remain about the same at 15% over the repayment period (15.5% for the 2010 to 2011 cohort and 14% for the 2014 to 2015) cohort).

Student loan repayment-to-income ratio - Recent measures

Recent measures can potentially improve the student loan repayment-to-income ratio going forward. While not covered by this evaluation, these include enhancements to the RAP introduced in the 2021 Budget and effective November 1st, 2022. A borrower earning $40,000 per year or less (adjusted for family size) is no longer required to make any payments on their student loans while on RAP. Moreover, for those receiving assistance from RAP, the maximum monthly CSLs payment has been reduced from 20% to 10% of their household income. These changes aim to improve the effectiveness of RAP in providing relief to borrowers and preventing defaults.

Starting April 1st, 2023, the Government of Canada has implemented interest-free loans for students, turning the 2020 temporary suspension of interest accrual on Canada Student Loans and Canada Apprentice Loans into a permanent measure. This new change will address one of the suggestions made by focus groups participants to permanently remove interest from student loans to make repayment easier.

It is worth noting that international comparisons showed that Canada's interest rates (Prime plus 2.5% until 2019) for repayment were less generous compared to those of the United States, New Zealand, Australia, and the United Kingdom. Therefore, the implementation of interest-free loans can have a positive impact on borrowers' repayment ability.

Demographic factors and student loan debt to income ratios

Certain demographic groups tend to have a higher debt-to-income ratio, indicating a lower ability to repay their student loans on time. Notably, single, and older borrowers are consistently more likely to have a higher debt-to-income ratio than their counterparts.

In addition, borrowers who are likely to have higher student loan debt, such as borrowers from private colleges or universities, or who studied in high-cost programs such and medicine and dentistry are more likely to have difficulties repaying their student loans.

To gain insight into borrowers' loan-repayment difficulties, the evaluation performed 2 separate analyses that examined factors impacting student loan repayment.

The first analysis used current survey data to calculate the debt-to-income ratio already introduced in this report. A multiple linear regression analysis was conducted to identify demographic groups with higher debt-to-income ratios. The predictor variables were identified by responses to the relevant survey questions and included, among others, type of institution, age group, region of residence, marital status, and other statuses. The results of this analysis highlighted that:

• borrowers who attended university had higher ratios compared to those who attended public colleges
• individuals aged 25 to 44 also had higher debt-to-income ratios than those aged 18 to 24 years
• the ratio of student loan debt to household income was higher for residents of the Central or Atlantic/Quebec regions than for those living in the Eastern (Ontario) region
• single borrowers with no children had higher debt-to-income ratios compared to married or common-law (with or without children) borrowers at the time of the survey
• finally, borrowers who self-identified as having a disability or as Indigenous also had higher ratios than their counterparts, respectively

Complementary to the debt-to-income ratio, the second analysis seeks to inform borrowers' ability-to-repay, by examining the ratio of average family income during the repayment period to the consolidation amount.^{Footnote 25} This analysis used data^{Footnote 26} linkage to inform borrowers' capacity to repay student loans and identified distinctive patterns among various socio-demographic characteristics, as shown in Figure 16 below. The analysis found that the average ratio is 13, which would be equivalent to an average family income of $65,000 over the repayment period divided by a consolidated CSL debt amount of $5,000. Therefore, below average ratios would suggest lower capacity to repay student loans.

• Source: Evaluation of the Canada Student Financial Assistance (CSFA) Program using Linked Administrative Data (2023)

Data linkage analysis also shows that single parents encountered significant limitations in repayment, and the presence of dependents further reduced their ability-to-repay ratio.

• Immigrants exhibited a lower ability-to-repay ratio despite having slightly smaller loan amounts than Canadian citizens
• Borrowers in the lowest age category (15 to 25) have a much higher ratio than any other age group
• Recipients of disability grants also displayed a higher ability-to-repay ratio than their counterparts

In addition, "field of study" emerged as a factor. For instance, medicine and dentistry are associated with lower ability-to-repay ratios due to higher student debt. Yet, borrowers who did not graduate had "higher ability-to-repay ratios" than those who did because of their lower incomes and debt loads.