Summative evaluation of the Budget 2008 Canada Student Loans Program (CSLP) Enhancements

Enhancements relevance

To what extent are the CSLP enhancements responsive to the needs of students? ^{Footnote 4}

Focus group participants aware of the CSLP enhancements introduced in Budget 2008 for part-time/married students perceived the enhancements to be valuable. Similarly, those who were aware of the RAP (the new repayment assistance measures) perceived it as being easier to apply for and more flexible than the previous provisions.

Regarding the CSGP, key informants in general agreed that it is important to offer grants alongside student loans. Grants help make PSE more accessible for disadvantaged students such as students from lower income families, students with children and those with disabilities. The literature suggested a need exists to help students overcome financial barriers, to help debt-averse students, for equity/societal reasons, and because they may be more effective than tax-related measures. Further, a survey of grant recipients indicated that 17.6% would not have continued with their studies if student grants were eliminated immediately and replaced by student loans, with lower income students being more likely not to continue.

CSGP disbursement and costs

What are the annual program costs and administrative costs of the CSGP? ^{Footnote 5}

After the introduction of the CSGP in the 2008-09 loan year, annual program costs (as measured by the total amount disbursed through grants and bursaries – including those under the former Canada Millennium Scholarship Foundation (CMSF)) increased from $421.2 million to $593.4 million in 2009-10, and to $715.3 million in 2013-14. The number of grants and bursaries disbursed increased almost every year between 2000-01 and 2013-14, with a major increase of roughly 180,000 after the implementation of the CSGP.

Administrative costs (including those related to the CMSF) increased from $42.8 million in 2008-09 to $53.0 million in 2013-14. As a percentage of annual program costs, administrative costs dropped from 10.0% to 7.3% over this period.

Impact of CSGP on CSLP

What has been the impact of the CSGP on student needs for the CSLP? ^{Footnote 6}

The introduction of the CSGP was expected to lower loan disbursements by $213.4 million between 2008-09 and 2009-10. However, little change in loan disbursements was seen ($2.08 billion in both 2008-09 and 2009-10), but the number of students accessing student loans increased by over 36,000. This increase could have been driven by a number of factors, including the introduction of the CSGP and/or the 2008 financial crisis, the latter of which may have compelled some to return to school – increasing the need for student loans. Given that for most students the CSGP grant amounts reduce the portion of financial assessed need covered by loans, it was estimated that the reduction in student loan amounts due to the CSGP was in the range of 11.5% to 17.0%. Further, statistical estimation analyses showed that student loan indebtedness levels at consolidation (the sum of all federal loan disbursements) were lower after the introduction of the CSGP. In the future, it is expected that the increasing use of Registered Education Savings Plans (RESPs) will also lead to lower student loan indebtedness levels.

Impact of CSGP on PSE access and persistence

Has the CSGP led to an increase in PSE access rates or in PSE persistence rates for the targeted groups? ^{Footnote 7}

Evidence collected for the CSG for students from low-income families and the CSG for students from middle-income families did not reveal an impact on PSE participation of either grant. However, from a more personal perspective, a survey of grant recipients suggested that those in receipt of the low-income grant were more likely to indicate that they would not have continued with their studies in the absence of the grant. One possible factor related to the lack of impact on PSE participation was the corresponding reduction in student loan amounts caused by the grant amounts and, hence, overall student financial aid remaining constant in most cases. Other potential factors that could have contributed to the lack of impact are the size of the grants relative to PSE costs and/or awareness issues. Further, the literature review suggested that if an existing student aid program switches from providing primarily loans to providing primarily grants or a combination of loans and grants, then there appears to be no statistically significant enrolment response except among low-income students. This last point could be an indication that the CSGP is an appropriate means for assisting Canadian families who struggle with the cost of higher education.

Although the literature indicated mixed effects of grants on PSE persistence, no strong evidence was found of a correlation between CSGP eligibility (for the two grants noted above) and a student remaining in school longer. It was again suggested that this finding was due to total student financial aid remaining constant (i.e. in most cases, little to no increase in overall financial liquidity).

Loan access – Part-time and married/common-law students

To what extent has there been an improvement in loan access for part-time students and for married students or students in common-law relationships? ^{Footnote 8}

Budget 2008 increased the maximum annual part-time student loan from $4,000 to $10,000. Following this change, there was a significant annual increase in the number of part-time student loan borrowers (rising from 1,425 in 2008-09 to 11,254 in 2013-14). It was estimated that 80.2% of the increase was related to the Budget 2008 change. Nevertheless, only about 1% of all part-time students access student loans.

Budget 2008 also announced a reduction in spousal contributions (from 80% to 70% of a spouse’s/partner’s disposable in-study income) to narrow the gap between contributions from spouses and parents of students. Although there was a 35% increase between 2008-09 and 2013-14 in the number of married/common-law student loan borrowers, this increase was identical to that for single students and was correlated with other factors such as the overall increase in the number of full-time married/common-law PSE students. Many key informants indicated that the decrease was insufficient.

Repayment Assistance Plan (RAP)

To what extent has the RAP made it easier for students to manage their loan debt? ^{Footnote 9}

The RAP was perceived by key informants and some focus group participants as a positive change, with the number of borrowers benefitting from repayment assistance measures increasing from 132,574 in 2008-09 to 160,572 in 2009-10 after the introduction of the RAP. Although the new parameters (e.g. affordable payments) can explain a portion of the increase in the number of beneficiaries, the impact of the 2008 financial crisis and resulting recession cannot be underestimated, as it negatively impacted employment outcomes for youth, making it more difficult for them to repay their student loans. The cost of loan repayment assistance decreased from $97.3 million in 2008-09 to $73.1 million in 2009-10, but eventually increased to $167.5 million in 2013-14.

Overall, the introduction of the RAP showed little impact on the repayment of loans compared to repayment under the old provisions. However, the full impact of the RAP had not yet been fully realized by 2013-14, and it is important to note that the recession likely had a tangible impact on loan repayment.

Awareness

To what extent are eligible students aware of the different measures? ^{Footnote 10}

Focus groups participants indicated that the canlearn.ca website has had minimal impact on their awareness of the RAP, since there was confusion between the canlearn.ca website and the National Student Loans Service Centre (NSLSC) website. However, among those who used the canlearn.ca website, there was a strong appreciation of the tools and a deeper understanding of the RAP than those who had not visited the website.

The CSGP survey conducted for the evaluation indicated that 39% of grant recipients were aware that they would qualify for a grant at the time of applying for a student loan and 30% knew of the variety of grants available when they first applied. Among those who were aware that they would qualify for a grant, 72% found that the eligibility criteria were clear at the time they applied for a loan.

Recommendations ^{Footnote 11}

Consider revising the formula used to calculate student loan repayment (e.g. basing repayment on net income instead of gross income) so that it is equitable for students in different circumstances. However, further analysis of the RAP may be required before implementing any major changes, given that the RAP has not yet fully matured.

1. Continue to use the CSGP to target disadvantaged students and consider increasing the CSG amount and expanding CSG eligibility for lower income students to help reduce debt and provide extra liquidity to improve their PSE access and persistence.
2. Further improve student loan access for married and common-law students, as the Budget 2008 changes were noted to have been insufficient and were shown to have not had any impact on loan access.
3. Promote awareness of the information on the canlearn.ca website, as those who used the site found it to be of great benefit. In doing so, further awareness and understanding of the CSGP would also be enhanced for prospective PSE students.

Recommendation 1

Consider revising the formula used to calculate student loan repayment (e.g. basing repayment on net income instead of gross income) so that it is equitable for students in different circumstances. However, it is understood that further evaluation of the RAP is required before implementing any major changes.

As part of Budget 2016, changes will be made to the RAP income thresholds to ensure that no borrower with student loans will be required to make any repayment until they are earning at least $25,000 per year (beginning in 2016-2017). This change will make student debt more manageable. The Program is mindful of the recommendation to consider basing repayment on net income instead of gross income so that it is equitable for students in different circumstances; however, such major changes to the program design elements need to be informed by further evidence in order to ensure a balance of program design with costs and simplicity of program operations. The CSLP strives to ensure the equitability and simplicity of use of the RAP and will continue to monitor uptake to ensure this debt management benefit fulfills its purpose.

With respect to fluctuating wages, it is the Program’s view that the RAP is sufficiently responsive to a borrower’s changing financial situation. RAP eligibility is assessed based on the applicant’s most recent monthly income. If an applicant is assessed as being able to make an Affordable Payment, and their income drops during their six-month RAP term, the applicant can ask for a re-assessment of their RAP application. This re-assessment would take the new income into account, and if eligible, the borrower’s Affordable Payment would be adjusted. Furthermore, if a borrower is ineligible for the RAP when they apply, but their financial situation changes in the future, they can re-apply and have their eligibility re-assessed. In lieu of revising the formula to account for fluctuating wages, steps could be taken to better inform applicants about the re-assessment process within the RAP.

Recommendation 2

Continue to use the CSGP to target disadvantaged students and consider increasing the CSG amount and expanding CSG eligibility for lower income students to help reduce debt and provide extra liquidity to improve their PSE access and persistence.

Canada Student Grants provide targeted, up-front, non-repayable financial assistance to PSE students from low- and middle-income families, students with dependants and students with disabilities. The grants are flat amounts that count towards the assessed need of students. Therefore, depending upon a student’s level of assessed need, the grants can contribute to the reduction of loan disbursement and amount of repayable student debt; it may also provide extra financial liquidity to the student.

As part of Budget 2016, grant amounts will be increased by 50% for the 2016-2017 academic year to ensure that students receive help that reflects the rising costs of PSE. Additionally, the Program will work with the provinces and territories to expand eligibility thresholds for grants starting in 2017-2018, so that even more students can receive non-repayable assistance.

The Program agrees with the report’s conclusion that the CSGP leads to a lower debt amount. This is consistent with the policy intent of the CSGP. Although currently there is no evidence to indicate that the CSGP has a positive impact on PSE access, this could be due to the fact that the amount of the grants (up to $2,000 for CSG-LI recipients) may still not be sufficient to create a financial incentive for under-represented groups to overcome the barriers to PSE participation. It is the Program’s view that enhancing grants could further improve PSE access. The recommendations from this evaluation will serve to inform the changes in CSGs.

Recommendation 3

Further improve student loan access for married and common-law students, as the Budget 2008 changes were noted to have been insufficient and were shown to have not had any impact on loan access.

Currently, students and parents are expected to contribute to the student’s education from income earned during their study periods. Parental contributions are assessed by determining the amount of discretionary income by subtracting the Moderate Standard Of Living cost from after-tax income. For married and common-law students, the spouses and partners are also expected to make a contribution. As part of Budget 2008, the requirement for spousal contributions was reduced from 80% to 70% of a spouse’s/partner’s disposable in-study income to narrow the gap between contributions from spouses and parents of students. Contributions for all groups of students are comparable to an objective standard rather than to each other.

Budget 2016 proposes to introduce a flat-rate student contribution to determine eligibility for Canada Student Loans and Grants to replace the current system of assessing student income and financial assets. This change would allow students to work and gain valuable labour market experience without having to worry about a reduction in their level of financial assistance. It would also benefit adult learners, many of whom may work while studying or have significant financial assets. The Program will be working collaboratively with provinces and territories to finalize the flat-rate contribution model for 2017–2018 academic year.

The Program recognizes that more could be done to improve access for married and common-law students. Treatment of married and common-law students and spousal income within the flat-rate design will be a focal point for discussions with provincial/territorial partners, to ensure the eligibility assessment remains transparent and predictable for all borrowers.

Recommendation 4

Promote awareness of the canlearn.ca website, as those who used it found it to be of great benefit. In doing so, further awareness and understanding of the CSGP would also be enhanced for prospective PSE students.

The program recognizes that the content of canlearn.ca proved valuable to those who use it and wants to find ways to make it more accessible for students. All correspondence to borrowers from the National Student Loan Service Centre (NSLSC) referred to canlearn.ca for further information on student loans and grants. Additionally, web searches for the NSLSC website automatically directed individuals to canlearn.ca. As of January 18, 2016, the content of canlearn.ca has migrated to Canada.ca. This content currently offers a number of interactive tools that help students save, plan, and pay for PSE. The CSLP will continue to work with partners to promote and increase use of the content, which can now be found on Canada.ca. As part of the transition to the new service provider contract, the Program is looking at ways to enhance borrower communications and promote awareness of program offerings. The new communication strategy will also include references to the content formerly found on canlearn.ca.

Canada Student Grants for full-time students ^{Footnote 21}

• Grant for Students from Low-Income Families (CSG-LI): if a student is from a low-income family and qualifies for a full-time Canada Student Loan, he/she may qualify to receive $250/month (half at the start of the year and half midway through). This grant is available for all years of undergraduate university, college or trade school programs, provided the entire program is a minimum of 60 weeks in length.
• Grant for Students from Middle-Income Families (CSG-MI): if a student is from a middle-income family and qualifies for a full-time Canada Student Loan, he/she may qualify to receive $100 per month of study. This grant is available for all years of undergraduate university, college or trade school programs, provided the entire program is a minimum of 60 weeks in length.
• Grant for Full-Time Students with Dependants (CSG-FTDEP): if a student is from a low-income family, qualifies for a full-time Canada Student Loan, and has children under the age of 12 as of the start of the study period, he/she may be eligible to receive $200 per month for each child.
• Grant for Students with Permanent Disabilities (CSG-PD): if a student has a permanent disability as defined in the Student Financial Aid regulations and qualifies for a full- or part-time Canada Student Loan, he/she may receive $2,000 per academic year to help cover the costs of accommodation, tuition and books.
• Grant for Services and Equipment for Students with Permanent Disabilities (CSG-PDSE): if a student is eligible for the CSG-PD and has a permanent disability which requires exceptional education-related costs for things such as tutors, note-takers, interpreters, brail assistance or technical aids, he/she may receive up to $8,000 per academic year to help pay for these expenses.

Canada Student Grants for part-time students

• Grant for Part-Time Studies (CSG-PT): if a student is a part-time student, qualifies for a part-time Canada Student Loan and is from a low-income family, he/she may be eligible for up to $1,200 per academic year.
• Grant for Part-Time Students with Dependants (CSG-PTDEP): if a student qualifies for a CSG-PT and has up to two children under the age of 12 as of the start of the study period, he/she may be eligible for $40 per week of study ($60 per week for those with three or more children under the age of 12).

Furthermore, following the introduction of the CSGP and the announcement that the “CMSF would close at the end of the 2008-2009 loan year”, ^{Footnote 22} transition grants were also introduced for bursary and award recipients to ensure that each student continued to receive the same level of funding for the duration of their current PSE program as was the case under the CMSF.

4.1.1 Total disbursements

Table 1 compares annual CSGP costs to those under the CAG, Canada Study Grant and CMSF. In 2000-01, the number of CSLP-related grants combined with CMSF grants and bursaries totalled 125,192 and amounted to $290.0 million, eventually rising to 186,604 and $421.2 million, respectively, by 2008-09. ^{Footnote 34} Both figures continued to increase following CSGP implementation in 2009, with 425,860 grants and $715.3 million disbursed in 2013-14. It is noteworthy that there was a particularly sizeable jump in 2009-10, coinciding with the first full year under the CSGP. ^{Footnote 35}

The average grant amount declined from $2,257 in 2008-09 to $1,616 in 2009-10, mostly due to the more generous CMSF awards/bursaries (which averaged about $3,000 per year) ceasing to be offered after 2008-09. Note that since a student can receive more than one type of grant, award or bursary at the same time, the average amount received per grant/award/bursary recipient will be higher than the average grant amount presented in Table 1. Using the number of unique grant recipients provided in the CSLP Annual Reports, Evaluation calculated that the average grant amount per grant recipient declined since the implementation of the CSGP, from $2,010 in 2009-10 to $1,946 in 2013-14. ^{Footnote 36} Thus, these results confirm that the average grant amount received per grant recipient has declined since the introduction of the CSGP.

Source: CSLP Annual Reports from 2000-01 to 2013-14 and Millennium Scholarship Annual Reports from 2000 to 2009.

¹ Excludes CMSF Awards and bursaries disbursed in non-CSLP participating provinces and territories.

² A loan year is equivalent to an academic year (i.e. from August 1 to July 31).

By type of grant delivered under the CSGP, Table A-4 in Appendix 4 indicates that three grants started under the Canada Study Grant (the grants for Students with Dependents, for the Accommodation of Students with Permanent Disabilities, and for High-Need Part-Time Students) and one grant started under the CAG (the grant for Permanent Disability) and that these continue to be provided. Since 2009-10, the number of grants and the amount disbursed for the CSG-FTDEP, CSG-PD, and CSG-PT have increased significantly, mainly due to the change in eligibility parameters and the maximum amount available per grant, while the number of grants for the CSG-PDSE has remained relatively constant.

The CAG-LI that was targeting first-year students from low-income families was eliminated in 2009-10 and a new CSG-LI was introduced, which provided 238,513 grants to students from low-income families in 2013-14 (amounting to $455.0 million) including grants to non-first-year students. The CSG-MI was also introduced in 2009-10, and provided 95,493 grants to students from middle-income families in 2013-14 (amounting to $72.3 million).

The total amount disbursed in transition grants was $52.1 million in 2009-10 (35,089 grants), $24.0 million in 2010-11 (15,884 grants), and $6.2 million in 2011-12 (4,365 grants) – the last year that transition grants were disbursed, as these grants were only introduced “to ensure that students would not be impacted negatively by the CMSF’s closure ^{Footnote 37} ”, as explained in Section 2.2.

4.1.2 Administrative costs

To examine the administrative costs of disbursing grants before and after the CSGP, it is necessary to look at the different types of administrative costs (i.e. the statutory administrative cost of the grants, their operating costs, and the CMSF costs) to estimate total administrative costs over time.

Since no specific information on the administrative costs of the Canada Study Grant, CAG and CSGP exists, Evaluation pro-rated grant disbursements based on total CSLP disbursements in order to determine the portion of administrative costs attributable to grants. ^{Footnote 38} The technical evaluation report related to CSGP costs estimated that statutory administrative costs (i.e. administration fees to provinces and territories and service provider costs) attributable to grants were $5.2 million 2008-09, rose to $14.7 million in the year after, and increased to $19.0 million in 2013-14 (as shown in Table 2).

To calculate the grant operating budget the Treasury Board Main Estimates (Part II) were used. These estimates provide information on operating expenses based on the departmental strategic outcome of “Learning”. Given that the CSLP and the related grants are part of “Learning”, Evaluation estimated the grant operating budget assuming that the same level of dollar effort is needed to operate all of the Learning programs. ^{Footnote 39}

Using the above assumption, the estimated costs for operating the grants are estimated to have been $4.3 million in 2004-05 and $34.0 million in 2013-14, with a significant increase between 2008-09 and 2009-10 (as shown in Table 2). Further, factoring in CMSF administrative costs, Table 2 shows that total administrative costs to deliver all grants and CMSF awards and bursaries rose from $27.6 million in 2004-05 to $53.0 million in 2013-14, although it is important to remember that the CMSF was not administered by the CSLP. Total administrative costs as a share of grant/CMSF expenditures rose from 7.4% in 2004-05 to 10.0% in 2008-09, but fell back to 7.3% by 2013-14.

Since the CSLP is delivered through the NSLSC, meaningful delivery cost comparisons with other ESDC programs that are not delivered by a third party are more problematic. ^{Footnote 40} Nonetheless, the technical evaluation report concluded that the costs for delivering the CSGP and the other types of non-repayable student aid were not out of line compared to other ESDC programs, especially given the variance in the estimates for different ESDC programs. ^{Footnote 41}

Sources: CMSF Annual Reports from 2000 to 2009, CSLP Annual Reports from 2000-01 to 2013-14, and Main Estimates, Part II, Treasury Board, from 2004-05 to 2013-14. Assumes CMSF payments were disbursed between January and March of each year.

4.3.1 PSE participation

Although evidence collected from key informant interviews with federal/provincial student loan representatives and academics did not suggest an optimal combination of grants and loans for dealing with PSE access and persistence, grants were identified as more effective in assisting low-income students. Nevertheless, the literature review completed for the 2011 CSLP Summative Evaluation suggested that multiple forms of student financial assistance are critical to increasing PSE enrolment levels. ^{Footnote 53}

The literature review completed specifically for this evaluation indicates that budget/liquidity constraints impact PSE participation, especially in the U.S. For instance, it was determined that increasing liquidity through student financial aid or reducing budget constraints by $1,000 led to an increase in PSE participation rates of 3 to 4 percentage points, as noted in Dynarski (2000) and Dynarski (2003). ^{Footnote 54} The literature review also suggested that if an existing student aid program switches from providing primarily loans to providing primarily grants or a combination of loans and grants, then there appears to be no statistically significant enrolment response except among low-income students. This last point could be an indication that the CSGP is an appropriate means for assisting “Canadian families who struggle with the cost of higher education.” ^{Footnote 55}

Related to this, three studies developed for the 2011 CSLP Summative Evaluation demonstrated that each additional $1,000 in loans increases the enrolment of students from low-income families by a range of 0.4 to 1.0 percentage points. It should be noted that one of the studies did not find any grant effect on PSE participation. However, the study indicated that “the smaller impact of grants than loans is counter-intuitive. It probably reflects the fact… that in many provinces grants are subtracted from need and reduce the amount of loans”, thereby corroborating U.S. results that switching from loans to grants without increasing liquidity has limited impacts.

A more recent study (Frenette 2011) estimated the impact of prospective debt load on the probability of PSE attendance among youth using data on the CAG for low-income youth and the Millennium Access Bursaries. ^{Footnote 56} The study found no evidence that the grants helped raise PSE enrolment. Frenette (2012) focused on estimating the cost reducing ability of the CSG‐LI and CSG‐MI. ^{Footnote 57} Unlike their predecessor (the CAG for youth from low‐income families) and the concurrent Millennium Access Bursaries – both of which offered loan remissions but no additional liquidity, the CSG‐LI and CSG‐MI offer both loan remissions and additional liquidity. Findings indicated that the CSG‐LI was successful in reducing direct costs of students but had no impact on out-of-pocket costs (the study found no impacts of the CSG-MI on either direct costs or out-of-pocket costs). ^{Footnote 58} A technical report by Frenette (2015) that was developed for this evaluation provided little to no evidence of a statistical association between CSGP grant eligibility and PSE access. ^{Footnote 59}

Using a combination of several databases aggregated at the provincial level, two models were developed to capture the effect of loans and grants on PSE enrolment. ^{Footnote 60} One model indicated a statistically significant positive effect of the total loan amount disbursed on the PSE participation rate, showing that an increase of 1% in the total amount of loans disbursed leads to an increase in PSE participation of between 0.119% and 0.135%. The other model that included total financial aid (loans and grants disbursed grouped together) suggested a similar impact on enrolment rates, showing that an increase of 1% in total financial aid leads to an increase of between 0.104% and 0.119% in the PSE participation rate. However, neither model showed a significant impact of grants on their own, corroborating results from previous studies.

In dollar terms, this implies that an additional loan disbursement or total financial aid increase of between $143 million and $162 million would lead to a 1% increase in PSE participation, assuming an average annual loan amount disbursement equal to the average loan disbursement between 2000-01 and 2012-13 ($1.93 billion) and holding other factors constant.

4.3.2 PSE persistence

In addition to the impact of the CSGP on PSE participation, the degree to which the presence of student financial aid contributes to students remaining in school and completing their studies is also important. The literature review noted that students whose financial aid package was not adequate to cover the actual cost of studying or who accumulated high levels of debt were less likely to complete their studies, suggesting that within a financial aid package, the non-repayable grant component can be a key component to encourage PSE persistence. A U.S. study showed that “need-based aid improves retention, but that its level of effectiveness varies with need. Specifically, grants… increase the retention probability by 1.3 percent per $1,000, whereas subsidized loans increase the retention probability by 4.3 percent per $1,000”. ^{Footnote 61}

Two other U.S. studies (Dynarski (2008) and Scott-Clayton (2011)) found that scholarships had a large impact on college completion. ^{Footnote 62} In a more recent study, Castleman and Long (2013) found that an additional $1,000 in grant aid eligibility was associated with a 3.5 percentage point increase in the probability of earning a bachelor’s degree within six years. ^{Footnote 63} As pointed out by Frenette (2015), the grant in the Castleman and Long study did not reduce the loan amount and, therefore, led to an increase in budget liquidity.

Similarly, in Canada, the study The Price of Knowledge (2009) indicated that receiving need-based student assistance in the form of loans or grants can improve PSE persistence. ^{Footnote 64} Findings from another study related to the CMSF suggested that higher levels of persistence were positively correlated with low student loan debt levels and financial aid received in the form of grants. ^{Footnote 65} However, two additional studies pertaining to the CMSF found no evidence of an increase in persistence after 2000, following the introduction of the CSMF. ^{Footnote 66}

The CSGP survey indicated that 67% of survey respondents would have continued with their studies without any changes even if student grants no longer existed. Another 15% would have changed an aspect of their enrolment decision such as working while studying, studying part-time, registering in a different program or in a different institution, and another 18% indicated they would not have continued with their studies, mainly due to having to borrow too much money. CSG-LI recipients were more likely to indicate that they would not have continued with their studies, perhaps reflecting the greater importance of grants for students from low-income families.

Frenette (2015) did not find any strong evidence that CSGP eligibility was related to PSE persistence. Evaluation also developed several models to estimate the impact of student grants and loans on the graduation rate. ^{Footnote 67} In all models, the statistical estimation analyses indicated that student loans, grants, and total financial aid were not significantly correlated with graduation rates. Even when Evaluation measured the cumulative effect of several years of financial aid receipt on graduation rates, results again indicated no evidence of a strong correlation between financial aid and a change in graduation rates.

5.2.1 Repayment Assistance Plan use and costs

Prior to the introduction of the RAP, borrowers who faced repayment difficulties received on average $100 million in loan repayment assistance per year through the IR and DRR provisions (see Table 5 below). After the introduction of the RAP, loan repayment assistance decreased from $97.3 million in 2008-09 to $73.1 million in 2009-10, but had increased to $167.5 million by 2013-14. Even after accounting for inflation, the cost of the RAP in 2013-14 was significantly higher than the costs of IR/DRR in any given year. The annual cost of the RAP should continue to increase based on projections contained in the 2014 CSLP Actuarial Report. ^{Footnote 73}

The number of borrowers who benefited from loan repayment assistance increased from 132,574 in 2008-09 to 160,572 in 2009-10, and to 233,864 by 2013-14. The average amount of loan repayment assistance per student loan borrower dropped from $734 in 2008-09 to $455 in 2009-10, and reached $716 in 2013-14.

It should again be reminded that the RAP was introduced shortly after the 2008 financial crisis. The unemployment rate of those aged 15-24 rose from 11.4% in 2007-08, to 13.5% in 2008-09, and to 15.2% in 2009-10, while the average weekly wage rate fell slightly (by 0.7%) between 2008-09 and 2009-10, possibly making student loan repayment more difficult for borrowers just entering the labour force. ^{Footnote 74} On this regard, Oreopoulos et al. (2012) showed that “young graduates entering the labor market in a recession suffer significant initial earnings losses that eventually fade, but after 8 to 10 years”. ^{Footnote 75} This suggests that the large increase in the number of borrowers requesting loan repayment assistance in 2009-10 and in subsequence years might be partially correlated with the 2008-2009 recession.

As indicated in Section 3.1, the RAP was perceived by key informants and some focus group participants as a positive change compared to the IR/DRR provisions. An international comparison revealed that Canada provides greater assistance to borrowers in repayment facing financial hardship than most of the countries examined in the report. ^{Footnote 76}

Source: CSLP Annual Reports from 2000-01 to 2013-14 and CSLP Actuarial Reports from 2001 to 2014 by the Office of the Superintendent of Financial Institutions (OSFI).

Note: ¹ A loan year is equivalent to an academic year (i.e. from August 1 to July 31).

² Note that RAP disbursements in 2009-10 included $15.8 million in Interest Relief payments and $2.3 million in DRR payments approved before August 2009.

The vast majority of early RAP participants received RAP-Stage 1, where affordable payments are made by the student towards the loan principal while the government pays a portion (or all) of the interest on the student loan. In 2009-10, 89.6% of all cases involved the government paying the entire portion of interest (87% in 2013-14).

Although the number of borrowers benefitting from RAP-Stage 2 more than doubled between 2009-10 and 2010-11 (see Table 6 below), the average amount students benefitted from decreased significantly (from $1,563 to $963). This was likely due to the inclusion of $2.3 million in DRR payments approved before August 2009 but paid in 2009-10 (mostly in August 2009) as indicated in the 2010 CSLP Actuarial Report. Since 2010-11, the number of RAP-Stage 2 beneficiaries and total costs has increased steadily, with the average payment per beneficiary increasing from $963 in 2010-11 to $1,549 in 2013-14.

In the case of the RAP for those with a permanent disability (RAP-PD), the number of borrowers and the amount of repayment assistance disbursed increased substantially following 2009-10. The focus groups revealed that the RAP-PD had a positive impact on students with disabilities as it helped offset expenses related to their disability, thereby freeing up additional resources for educational expenses.

Source: CSLP Annual Reports from 2009-10 to 2013-14 and CSLP Actuarial Reports from 2010, 2011 and 2014.

Note: ¹ A loan year is equivalent to an academic year (i.e. from August 1 to July 31).

² Includes $15.8 million in Interest Relief payments approved before August 2009.

³ Includes $2.3 million in DRR payments approved before August 2009.

5.2.2 Impact of RAP on student loan repayment

Table 7 indicates that average student loan debt at consolidation increased steadily from 2005-06 ($11,323) to 2009-10 ($13,404), decreased for two years, and then increased again in 2012-13 ($13,250) and 2013-14 ($13,536). ^{Footnote 77} A higher percentage of borrowers used loan repayment assistance measures in 2008-09 and 2009-10 compared to 2007-08, regardless of the number of years after consolidation, corroborating the results from Table 5. However, this increase also happens to coincide with the 2008-09 recession, perhaps indicative that some borrowers experienced difficulties securing stable employment after their studies ended and needed help with paying their student loans.

Source: CSLP administrative database (tabulation ESDC). * RA stands for Repayment Assistance which includes IR/DRR and RAP.

Note: Borrowers consolidate their student loan(s) six months after completing their post-secondary studies (or ending full-time studies). Repayment begins once they have consolidated their loans.

Table 7 also reveals that there may have been a slight reduction in the percentage of borrowers who have repaid their loan in full since the introduction of the RAP, although it appears that this trend may have started before the RAP was implemented. This could again be a by-product of the recession in 2008-09. Nonetheless, from 2009-10 onwards, around 9% of borrowers paid off their loan one year after consolidation, 27% two years after consolidation, and over 35% three years after consolidation.

Finally, the percentage of student loan debt remaining to be paid was relatively stable during the period, particularly since 2008-09, at roughly 89% after one year, 59% after two years, and 46% after three years. This reduction in the percentage of student loan debt remaining to be paid confirms that borrowers continued to repay a portion of their debt every year. A technical report ^{Footnote 78} revealed that there has been a significant increase in the percentage of debt remaining to be paid by RAP users and non-RAP users compared to under the IR/DRR provisions. The technical report also confirmed a decrease in the likelihood of having entirely repaid a student loan by these two groups over the period. This may be due to changes in RAP eligibility criteria (more recipients) combined with the effect of being able to make affordable payments under the RAP. It is possible that some borrowers were “forced” to make full payment on their loan (to avoid default) during the years of IR due to their ineligibility for IR, while similar borrowers became eligible for the RAP and were able to avoid this situation. Nevertheless, a longer period of data would be needed to clarify the impact since the results are based on only a few years of data.

Borrowers who face financial hardship can decide to apply for the RAP instead of going into default. ^{Footnote 79} Table 7 shows that borrowers who have ever received repayment assistance had a significantly lower default rate than those who have never received repayment assistance. For instance, the former group had a default rate of around 4.5% two years after consolidation compared to about 16% for the latter group. Table 7 also reveals that, although default rates increased between 2008-09 and 2009-10 for both groups, default rates went down steadily after 2009-10. As mentioned previously, it is still too early to measure the full impact of the RAP.