publisher colophon
abstract

We examine the likely implications of switching from a prior year (PY) financial aid system, the current practice in which students file the Free Application for Federal Student Aid (FAFSA) using income data from the previous tax year, to prior-prior year (PPY), in which data from two years before enrollment is used. While PPY allows students to gain information about the cost of college earlier, financial aid awards could be less targeted if annual household income fluctuates dramatically. We study this question using data from more than 70,000 students at nine institutions over a five-year period spanning the Great Recession. We find that 60%–70% of students do not see any changes to their Pell awards under PPY in this case study, and the neediest students rarely see large changes. About 7%–15% of students see large declines in their Pell awards under PPY, and the effects vary by institutional and student characteristics. Our simulation of the potential costs of switching to PPY results in an estimated increase in Pell program costs of four percentage points.

introduction

After decades of increases in tuition and fees, more than seven in ten students now borrow for their baccalaureate education, with the average borrower taking out $29,400 in loans to finance his or her bachelor’s degree in 2012 (Reed and Cochrane 2013). The increased financial burden many students and their families shoulder to pay for college makes early financial planning imperative, [End Page 253] but current financial aid policies require students to apply to college without a clear picture of their aid eligibility. This lack of information means that students and their families do not know how much they will be expected to pay for college until they are on the brink of payment (Luna De La Rosa 2006; Perna 2006; Tierney and Venegas 2009). As a result, some talented students from low-income families decide to forgo even applying to college (Hoxby and Avery 2012).

Students become eligible to receive federal financial aid, as well as many types of state or institutional aid, by completing the Free Application for Federal Student Aid (FAFSA). The FAFSA asks for standard income information that is found on tax forms, but also collects information on student and parent (for dependent students) investments and assets that are not a part of a tax return. Income information from the prior tax year (PY) and current asset information are used to determine the student’s financial need by calculating an expected family contribution (EFC) for the award year, which represents a measure of a family’s short-term financial ability to pay for college.

While students can file starting January 1 for the following academic year, most students wait until at least February or March, because employers do not have to provide employees with W-2 forms (needed to file taxes) until the end of January.1 An additional advantage of waiting to file is that students can use the Internal Revenue Service Data Retrieval Tool, which allows tax return data to be directly transferred to the FAFSA two to three weeks after the tax return is filed; this considerably cuts down on a student’s filing burden. However, the current financial aid timeline does not allow students in some states to use the Data Retrieval Tool, because the filing deadline for state aid is too early. At least six states recommend applying for their grant aid programs “as soon as possible after January 1, 2014,” as aid is given on a first-come, first-served basis (Federal Student Aid 2014).2 At least ten additional states have deadlines before March 2. To provide earlier and more useful financial aid eligibility information to students and their families, some researchers and advocacy groups have proposed using financial information from two years prior (“prior prior year,” or PPY) instead of PY information (e.g., Advisory Committee on Student Financial Assistance 2005; Dynarski and Wiederspan 2012; TICAS 2013). The idea has gained bipartisan support in Congress, including multiple pieces of legislation supporting PPY introduced in the summer of 2014. For students who intend to enroll in college [End Page 254] for the 2014-15 award year, the FAFSA would be based on income data from the 2012 tax year (PPY) instead of 2013 (PY). Most students and families will have completed their income tax returns for the 2012 tax year by the spring of 2013, meaning the PPY approach would allow students to potentially get their federal aid package one full year before beginning college (see Figure 1).

Figure 1. Comparing PY and PPY timelines
Click for larger view
View full resolution
Figure 1.

Comparing PY and PPY timelines

research questions

To examine the potential effects on students and the federal government of changing the financial aid system from PY to PPY, we examine detailed data for more than 70,000 students at nine institutions between the 2007–08 and 2011–12 award years, which notably include the effects of the economic recession. The sample for this case study includes two community colleges as well as five four-year public and two four-year private institutions with various missions and selectivity levels. We answer the following research questions:

  1. 1. To what extent does using PPY income and asset data versus PY affect students’ EFCs, and how would this affect Pell Grant eligibility and financial aid awards?

  2. 2. Are there differences in the proportion of students who would be affected by a switch to PPY by student characteristics?

  3. 3. What are the potential costs to the federal Pell Grant program of a switch to PPY? [End Page 255]

Pros and Cons of PPY

The additional time to financially prepare and plan for college costs using PPY could induce more students to fully participate in the college application process, which has the potential to increase college enrollment and/or completion rates. For example, Bettinger, Long, Oreopoulos, and Sanbonmatsu (2012) conducted an experiment in which tax preparers in low-income communities automatically transferred potential students’ tax data to the FAFSA, submitted the application, and provided information about the net price of attendance. This intervention resulted in an increase of 4.8 percentage points (from a base of 48.5%) in any college enrollment during a three-year period for dependent students. Advancing the FAFSA filing timeline could have similar effects, especially if coupled with additional filing assistance.

The PPY approach could also help reduce the number of low-income students who enroll in college and would be eligible to receive Pell Grants, but do not complete the FAFSA. Kantrowitz (2009) used data from the 2007–08 National Postsecondary Student Aid Study to estimate that 2.3 million students would likely have been Pell eligible if they had only filed the FAFSA, 1.1 million of whom would be expected to have a zero EFC (and the maximum Pell Grant). These low-income students who enroll in college but did not file the FAFSA are much more likely to drop out of college than their low-income peers who filed the FAFSA, conditional on observable characteristics (Novak and McKinney 2011).

However, PPY also has some potential drawbacks. The PPY income may not accurately represent a family’s current economic situation compared with the PY income, especially as the annual volatility of family income has risen over time among lower-income families (e.g., Gottschalk and Moffitt 2009; Kopczuk, Saez, and Song 2010). If a student’s family income increased between PPY and PY, then their Pell award could be larger under PPY. Other students could see a smaller Pell under PPY, but they could be held harmless by asking their institution’s financial aid office for a professional judgment to be performed on their FAFSA. Professional judgments allow aid administrators to adjust FAFSA elements to better reflect a family’s current ability to pay for college. For example, if a dependent student’s mother lost her job several months before the student planned to enroll in college, the financial aid office could adjust parent income on the FAFSA to reflect current earnings instead of those from when she had a job. Just over 1.5% of Pell recipients in the 2011–12 academic year received professional judgments (US Department of Education 2013), but the number of students who receive professional judgments and the corresponding workload for financial aid offices could increase considerably under PPY. [End Page 256]

A switch to PPY could result in a different set of rules for state and institutional financial aid programs if they continue to use PY for their aid programs. For example, the use of PY is currently written into Section 663 of New York’s education laws regarding financial aid disbursement. This is a particular concern with highly selective institutions, many of which use the College Board’s College Scholarship Service Profile as an additional financial aid application to gather supplemental financial information on a family’s assets beyond what is collected on the FAFSA. It is currently used by more than 400 institutions and scholarship programs, a number that is likely to grow if PPY is used and aid awards are viewed as being less reflective of current resources.

Literature on PPY

Although existing literature suggests the promise of advancing the financial aid timeline, few studies to date have investigated the use of PPY. This could potentially be because of the difficulty in obtaining student-level financial aid data for multiple years at multiple institutions. The earliest study to examine PPY was conducted by the US Department of Education’s Advisory Committee on Student Financial Assistance (1997), which compared income data from 1996 (PY) with that of 1995 (PPY) and found that income changed by more than $10,000 for 45% of all FAFSA filers during those two years; they concluded that PPY should not be implemented. An additional analysis of US Department of Education data by Madzelan (1998) found that PPY income predicts current income with 82% accuracy, while PY income is only slightly more accurate at 87%. However, the focus on family income may not reflect changes in Pell awards, as large changes in family income may result in no changes to a student’s Pell award if he or she has very low or very high income.

The only published empirical study, to our knowledge, that examines the distributional effects of PPY is by Dynarski and Wiederspan (2012), who used the National Postsecondary Student Aid Study to compare PY data from 2007 with PPY data from 2006 and found that 77% of continuing students would see a Pell Grant of within $500 of their current award. However, their analysis does not answer important details about how PPY could actually work. They did not examine potential changes to Pell awards by dependency status, although volatility in Pell awards may differ between independent and dependent students. By focusing on the dollar value of Pell changes, they did not explore the percentage of students who would gain or lose a Pell Grant under PPY. This is important because Pell eligibility is used to determine eligibility for many other grant programs. They only had data for two years, which does not allow for deeper examination of income and Pell eligibility fluctuation in a PPY system [End Page 257] over time. Finally, they did not examine the cost implications of PPY, which are important to consider from the federal government’s perspective.

data and methods

In this case study, we leveraged student-level financial aid data from the 2007–08 through 2011–12 academic years provided to the National Association of Student Financial Aid Administrators by nine partner institutions. Details about our dataset and analytic methods follow in this section.

Sample Characteristics

The nine institutions in our sample included two public community colleges, five public doctoral-level universities, and two private four-year colleges. The demographic characteristics and graduation rates of the institutions in this case study can be found in Table 1, along with a comparison with other institutions in those sectors using data from the Integrated Postsecondary Education Data System. Although the participating institutions were chosen because of data availability, they appear to be reasonably representative of their broader sectors.

We received data on nearly 160,000 undergraduate students who filed the FAFSA at least once between the 2007–08 and 2011–12 academic years. To be included in the analytic sample, students must have enrolled and filed the FAFSA for at least two years under the same filing status (dependent, independent without any dependents, or independent with his/her own dependents), not received a professional judgment on their aid package in either year, and had a calculated EFC within $50 of the actual EFC in either year.3

The restrictions resulted in an analytic sample of 73,009 students. Students excluded from the analytic sample differ from those who are included along several important dimensions. The analytic sample disproportionately consists of dependent students at four-year institutions. Dependent students were ten percentage points more likely to be in the analytic sample, and students attending two-year colleges were nearly twice as likely to be excluded from the analytic sample as they were to be included. Black and Pell-eligible students were less likely to be in the analytic sample, while Asian students and students whose parents graduated from college were more likely to be in the sample.

Table 2 shows summary statistics of the analytic sample by prior year, using characteristics from the PPY. There were approximately 30,000 students in the data set during each of the four PY/PPY periods, for a total of 120,927 student-year [End Page 258] observations. Compared with nationally representative data from the 2011–12 National Postsecondary Student Aid Study, our sample had a much larger fraction of dependents (75% vs. 49%) and much smaller proportions of independent students both without any dependents (14% vs. 24%) and with any dependents (11% vs. 28%). Part of this difference was driven by the lower likelihood of independent students to be observed in our sample for two consecutive years. Only 11% of students attended a community college—far below the national average.

Table 1. Summary Characteristics of Institutions in the Data Set.
Click for larger view
View full resolution
Table 1.

Summary Characteristics of Institutions in the Data Set.

About two-thirds of our sample was white, 56% was female, and 65% had parent(s) who attended college; these percentages are consistent across years. The effects of the Great Recession can be seen on parent income (dependents only) and student income (independents only), which drop by 5%–10% between [End Page 259] the 2008–09 and 2011–12 PYs. The percentage of Pell-eligible and zero-EFC students both increased by about 15 percentage points during this period, which is a function of both declining income and changes to the Pell program during the 2009–10 academic year (2010–11 PPY) that resulted in more students becoming eligible.

Table 2. Summary Statistics of Students with PPY Data by Year
Click for larger view
View full resolution
Table 2.

Summary Statistics of Students with PPY Data by Year

Calculating PY and PPY Pell Awards

We began by manually recalculating the student’s current EFC using the PY methodology for each year data was available from the 2007–08 to 2011–12 academic years. Because the federal formulas and data requirements differ by dependency status, we calculated EFCs using each individual data element and the applicable formula for each of the three dependency statuses described above. These federal formulas can require a large number of data elements about student and spouse (if independent and married) or parent (if dependent) income, assets, and household size; the EFC formula for the 2013–14 academic [End Page 260] year has up to 51 questions for dependent students and 29 questions for independent students in addition to a bevy of worksheets required to generate values for certain elements.

Students in all three dependency categories can qualify for simplified EFCs (in which household assets are not considered) if the adjusted gross income of a student (independent) or his/her parent(s) (dependent) is less than $50,000 per year and if a household member receives means-tested benefits, did not have to file the Internal Revenue Service Form 1040, or was laid off from their previous job (dislocated worker). Dependent and independent students can qualify for an automatic zero EFC (where no additional information is required) if the adjusted gross income is below a federally set income threshold ($20,000 or less in 2007–08 and 2008–09; $30,000 in 2009–10 and 2010–11; and $31,000 in 2011–12) and any of the additional criteria required for simplified EFC eligibility were met. Independent students without dependents were not eligible for an automatic zero EFC.

We began by recalculating students’ PY EFCs using available income, asset, and demographic data elements, using separate formulas for each year/dependency status combination to ensure the data provided to us by the institutions were accurate, getting clarifications from colleges when necessary. We were able to calculate PY EFCs that exactly matched students’ actual EFCs in nearly 94% of cases. We allowed for a $50 deviation between the actual and calculated EFCs to account for rounding differences across a large number of elements; this boosted our accuracy to almost 95%.

We then calculated the EFC for a given year using PPY data in the PY formula. All elements were used as reported in the PPY, with the exception of student and parent ages (used in the asset contribution calculations). Because ages can be carried forward to the PY without any error, we added one year to the PPY age to calculate the PY age. All other elements, such as household size and the number of family members in college, came from the PPY instead of the PY because the PY values were not perfectly known as of the PPY. These calculated EFCs from the PY and PPY were then converted to the Pell Grant award using the US Department of Education’s conversion guidelines for full-time students. We used the full-time guidelines because neither enrollment intensity data nor the actual values of Pell Grants received were available at all institutions, but this resulted in larger changes to Pell Grants than would actually occur for part-time students. Using data from the 2011–12 award year for the roughly 85% of students with Pell award values, the calculated Pell awards appear to be overstated by 14% for dependent students, 36% for independent students without dependents, and 53% for independent students with their own dependents. We adjusted the cost estimates downward by these amounts to take part-time students into account. [End Page 261]

Simulating PPY Costs

We next estimated the potential costs of a switch from PY to PPY at a national level by extrapolating the results from our nine institutions to all colleges and universities receiving federal financial aid funds. Because we were examining a potential change to a program, there was a substantial amount of uncertainty in the potential costs that require a sensitivity analysis (Levin and McEwan 2000; Hummel-Rossi and Ashdown 2002). Our estimates have some uncertainty owing to a relatively small number of colleges. We employed a Monte Carlo simulation (Rubinstein and Kroese 2008) to estimate costs across 100,000 trials, allowing for a wide range of values regarding cost assumptions. We considered three ways in which program costs could change:

  • • Case 1, changing from PY to PPY, holding all other factors constant,

  • • Case 2, adding an increase in professional judgments that would switch Pell awards from PPY back to PY, and

  • • Case 3, adding a potential increase in enrollment rates as a result of students receiving earlier notification of their aid awards.

To estimate the costs of PPY, we used three means from our sample: the average PY Pell award for the full sample, the average PY Pell for students who would lose their entire Pell by a switch to PPY, and the percentage of students who would lose their Pell award. We combined this with national data on Pell recipients for the average award, the number of recipients, and the percentage of Pell recipients who received professional judgments in the 2011-12 academic year (US Department of Education 2013). All of these elements were examined separately by dependency status.

We then simulated values for other parameters. To reflect the uncertainty of our estimates regarding the difference between PY and PPY Pell awards (Case 1), we introduced a perturbation in the estimate with a standard deviation equal to the average standard error across the four years of PY/PPY comparisons for each dependency status. We assumed that half the students who would lose their entire Pell award under PPY would request professional judgments (Case 2) in the median simulation, and that they would receive the average PY Pell award for students who lost the grant in PPY. Finally, we assumed an average effect on Pell receipt rates of two percentage points (Case 3). This is approximately half the estimated effect of FAFSA completion and college access programs (Harvill et al. 2011; Bettinger et al. 2012). The parameter values for Cases 2 and 3 were simulated using a binomial distribution with 100 draws and the specified means.

To estimate the costs under PPY without any additional professional judgments or Pell recipients (Case 1), we used the (perturbed) change in Pell awards between PY and PPY and divided by the average PY Pell in our sample [End Page 262] to estimate the percentage of change by dependency status. This percentage was then multiplied by the number of Pell recipients nationwide and the average award. We estimated the costs with additional professional judgments (Case 2) by adding Case 1 costs to the result of multiplying the estimated professional judgment rate by the number of Pell recipients nationwide and the average Pell award students who would lose the grant in PPY received in PY. Finally, we estimated the cost of PPY with professional judgments and additional Pell enrollment (Case 3) by adding the product of additional enrollment, national Pell enrollment, and average Pell receipt nationally to the results from Case 2.

Limitations

The greatest limitation to this study is that only nine institutions are represented in the data set, which was extrapolated to represent the more than 7,000 institutions participating in federal student aid programs. While the institutions are at least somewhat representative of their sectors, we do not have data from any for-profit institutions. Also, while we have data on one Hispanic-serving institution (Florida International), we do not have any historically black colleges in our sample. We only observed students in the years in which they filed the FAFSA, which does not allow us to observe the characteristics of students who remained continuously enrolled but did not file the FAFSA each year. The data set also does not include information on students’ academic outcomes or graduation rates, meaning we do not know why students left the data set or how they fared in college. With those limitations noted, individual-level panel data on FAFSA measures across multiple institutions is uncommon and makes this study valuable. The cost estimates also have a substantial amount of uncertainty, but represent an initial estimate of what a switch to PPY might cost. Further research is needed in this area.

results

We first presented the changes in Pell awards using PPY instead of PY by year and dependency status (Table 3). The first key finding is that the estimated effects of a switch to PPY vary considerably by dependency status and year, with all comparisons but one being significant across years at p < .01. Three in ten dependent students (panel A) were affected by a switch to PPY, although slightly more students saw a change (either an increase or a decrease) in their Pell award between the 2008–09 and 2011–12 academic years. More independent students without dependents (panel B) were affected, with about 40% seeing at least a slight change in Pell awards. Only three in ten independent students (panel C) [End Page 263]

Table 3. Change in PPY Pell Award by Dependency Status and Year
Click for larger view
View full resolution
Table 3.

Change in PPY Pell Award by Dependency Status and Year

[End Page 264]

had their Pell awards change by using PY. Unlike dependent students, there is a sharp drop in the number of independent students who saw a change in their Pell award over time, as more of these students received a zero EFC each year.

The average change in the Pell award amount from PY to PPY is close to zero across the period for both dependent students and independent students with dependents, suggesting that Pell program costs should stay relatively constant for these groups before taking professional judgments into account. However, independent students without dependents consistently see smaller PPY Pell awards than PY awards across each of the years. The average PPY Pell is $88 lower than the average PY Pell, or about 2.5% of the typical award. This difference is likely owing to the fact that independent students may work full-time for a period before entering (or re-entering) college. This is less of a concern for independent students with dependents, who are likely to have a zero EFC each year. [End Page 265]

About 15% of dependent students and independent students with dependents would see changes of more than $1,000 in their full-time equivalent Pell award using PPY, with similar numbers experiencing upward and downward changes. Independent students without their own dependents see much more volatility, with 26% of students seeing a change of $1,000 or more. Roughly 10% of dependent students and independent students without dependents would gain or lose Pell eligibility by switching to PPY, compared with 5% of independent students with dependents.

In Table 4, we examine changes in Pell award by gender, race/ethnicity, and Pell eligibility in the prior year (zero EFC and Pell-eligible with a nonzero EFC).4 Men and women are similarly affected by a switch to PPY across all dependency statuses. Hispanic students see more volatility in their Pell awards, particularly among dependent students. This change is likely a result of Hispanic students in our sample being closer to the Pell eligibility threshold. White and black dependent students are affected at similar rates, which is owing to a large percentage of white students having household incomes not qualifying them for the Pell Grant and most black students having incomes making them Pell eligible.

The group of students most affected by PPY is those who are eligible to receive a Pell Grant in the PY but do not have an EFC of zero. These students are relatively close to the maximum allowable EFC for Pell Grant eligibility, and thus can go from receiving a small Pell Grant to no award with a relatively small change in household income. Only about 5% of these students saw the same Pell award in the PPY as the PY, and some of the changes were quite large. About one in six dependent students and independent students without dependents had a Pell in the PY, but not in the PPY; this drops to less than one in ten when considering independent students with dependents. About one in four Pell recipients with a positive EFC lost $1,000 or more from their Pell award by using PPY data (varying slightly by dependency status).

simulated cost estimates

Panels A and B of Table 5 contain the parameter estimates used in the Monte Carlo cost simulation of the percentile distributions of each simulated variable. The greatest amount of uncertainty is with the rate of additional professional judgments assumed to occur under PPY. Assuming that half of the students who [End Page 266]

Table 4. Change in PPY Pell Award by Dependent Status and Student Characteristics
Click for larger view
View full resolution
Table 4.

Change in PPY Pell Award by Dependent Status and Student Characteristics

[End Page 267]

lost their entire Pell award under PPY would receive a professional judgment (and their PY Pell award) results in increased professional judgment rates of 5.45% for dependent students, 3.90% for independent students without dependents, and 1.45% for independent students with dependents. The 10th–90th percentile range of estimated professional judgment rates varies by three to five percentage points by dependency status, reflecting the potential range of values we feel are plausible.

In our median cost simulation (Panel C), there are slight cost savings ($37 million per year) to the federal Pell program, owing to a switch to PPY if no changes are assumed to professional judgments or Pell enrollment (Case 1). Dependent students received larger Pell awards under PPY, but were these outweighed by the smaller awards given to independent students who likely worked more in the PPY tax year than the PY tax year. Ninety percent of our simulations had cost estimates between a $249 million savings and a $175 million additional costs, a relatively small change based on the Pell Grant program’s annual average costs of about $34 billion (US Department of Education 2013).

We estimated that the median cost of PPY after accounting for the likelihood of additional professional judgments (Case 2) would be approximately $605 [End Page 268] million, with 90% of simulations showing increases between $172 million and $1.11 billion. Nearly all of the cost increase is owing to dependent students. Adding the potential of increased Pell enrollment owing to PPY to the other costs (Case 3), the median cost is $1.35 billion per year, with 90% of simulations between $629 million and $2.15 billion. About two-thirds of the additional cost would be owing to dependent students.

Table 5. Monte Carlo Simulation Results
Click for larger view
View full resolution
Table 5.

Monte Carlo Simulation Results

Putting this potential cost increase into perspective, an annual increase of $1.35 billion would reflect a 4% increase in overall Pell expenditures. Recent quasiexperimental evidence from Florida (Castleman and Long 2013) and experimental evidence from Wisconsin (Goldrick-Rab, Kelchen, Harris, and Benson 2014) suggest that need-based financial aid can result in substantial increases in graduation rates. While we did not model the potential benefits to federal revenues as a result of increased educational attainment and the associated earnings increases, they should be kept in mind when considering the likelihood of increased program costs. [End Page 269]

discussion

We explored the extent to which students’ Pell Grant awards would change in a switch from PY to PPY using data for more than 70,000 students in a five-year period spanning the Great Recession. We found that PPY yielded similar results to PY for the neediest students—very low-income students, many of whom are independent students with dependents. Giving students advanced notice of their aid awards has the potential to increase college enrollment and persistence rates, as prior research has suggested (Kelchen and Goldrick-Rab forthcoming). However, two groups of students saw more volatility and thus would be more likely to lose Pell Grants under PPY: students whose EFCs narrowly qualified them for a Pell Grant in the PY and independent students who worked before entering college. [End Page 270]

Care must be taken to assist the students who could potentially receive less in grant aid under PPY to make sure their aid awards do not change. These students can request that their financial aid office perform a professional judgment on their aid package to substitute PY income data if it is lower than PPY income. However, this is likely to disproportionately affect the community colleges and less selective four-year colleges that serve more of these students, as they may see their understaffed financial aid offices facing an overload of professional judgment requests. Additional concern should be given to for-profit colleges and minority-serving institutions, as they serve a disproportionate number of first-generation students and Pell recipients who may not have knowledge of the professional judgment process. More outreach and funding efforts are needed for these colleges by providing specialized assistance to help make the professional judgment process known and accessible to students who would otherwise be adversely affected by PPY.

PPY has the potential to help many students, and it appears quite possible at this time that Congress will pass legislation requiring a shift to PPY. If that is done, we urge an evaluation of the implications for students, colleges and universities, and federal finances. If time permits, a demonstration program using several regions or states should be considered. Careful attention in either case should be paid to when students file the FAFSA under PPY, the number of professional judgments requested, and the resulting implications for financial aid offices and program costs.

Robert Kelchen

Robert Kelchen is an assistant professor of higher education in the Department of Education Leadership, Management and Policy at Seton Hall University.

Gigi Jones

Gigi Jones is an education research scientist with the National Center for Education Statistics.

References

Advisory Committee on Student Financial Assistance. Analysis of the Office of Postsecondary Education’s Prior, Prior Year (PPY) Income Proposal. Washington, DC: US Department of Education, 1997.
—. The Student Aid Gauntlet: Making Access to College Simple and Certain. Washington, DC: US Department of Education, 2005.
Bettinger, Eric P., Bridget Terry Long, Philip Oreopoulos, and Lisa Sanbonmatsu. 2012. “The Role of Application Assistance and Information in College Decisions: Results from the H&R Block FAFSA Experiment.” The Quarterly Journal of Economics 127 (3): 1205–42.
Castleman, Benjamin L., and Bridget Terry Long. Looking Beyond Enrollment: The Causal Effect of Need-Based Grants on College Access, Persistence, and Completion. Cambridge, MA: National Bureau of Economic Research working paper No.19306, 2013.
Dynarski, Susan, and Mark Wiederspan. 2012. “Student Aid Simplification: Looking Back and Looking Ahead.” National Tax Journal 65 (1): 211–34.
Federal Student Aid. Free Application for Federal Student Aid. Washington, DC: US Department of Education, 2014.
Goldrick-Rab, Sara, Robert Kelchen, Douglas N. Harris, and James Benson. 2014. “Income Inequality in Higher Education: Experimental Evidence on the Impact of Financial Aid on College Completion.” Working paper. [End Page 271]
Gottschalk, Peter, and Robert Moffitt. 2009. “The Rising Instability of US Earnings.” Journal of Economic Perspectives 23 (4): 3–24.
Harvill, Eleanor L., Rebecca A. Maynard, Hoa T. H. Nguyen, Claire Robertson-Kraft, and Namrata Tognatta. Effects of College Access Programs on College Readiness and Enrollment: A Meta-Analysis. Presentation at the Society for Research on Educational Effectiveness conference, Washington, DC, 2011.
Hoxby, Caroline M. and Christopher Avery. The Missing “One-Offs”: The Hidden Supply of High-Achieving, Low-Income Students. Cambridge, MA: National Bureau of Economic Research working paper No.18586, 2012.
Hummel-Rossi, Barbara, and Jane Ashdown. 2002. “The State of Cost-Benefit and Cost-Effectiveness Analyses in Education.” Review of Educational Research 72 (1): 1–30.
Kantrowitz, Mark. 2009. “Analysis of Why Some Students Do Not Apply for Financial Aid.” FinAid.org. Last modified April 27, 2009. http://www.finaid.org/educators/20090427CharacteristicsOfNonApplicants.pdf.
Kelchen, Robert, and Sara Goldrick-Rab. Forthcoming. “Accelerating College Knowledge: A Fiscal Analysis of a Targeted Early Commitment Pell Grant Program.” The Journal of Higher Education.
Kopczuk, Wojciech, Emmanuel Saez, and Jae Song. 2010. “Earnings Inequality and Mobility in the United States: Evidence from Social Security Data Since 1937.” The Quarterly Journal of Economics 125 (1): 91–128.
Levin, Henry M., and Patrick J. McEwan. Cost-Effectiveness Analysis: Methods and Applications (2nd Ed). Thousand Oaks, CA: Sage Publications, 2000.
Luna De La Rosa, Mari. 2006. “Is Opportunity Knocking? Low-Income Students’ Perceptions of College and Financial Aid.” American Behavioral Scientist 49 (12): 1670–86.
Madzelan, Daniel. HEA Reauthorization Issue: Using “Prior-Prior” Year Income. Washington, DC: US Department of Education, 1998.
Novak, Heather, and Lyle McKinney. 2011. “The Consequences of Leaving Money on the Table: Examining Persistence among Students Who Did Not File a FAFSA.” Journal of Student Financial Aid 41 (3): 6–23.
Perna, Laura W. 2006. “Understanding the Relationship between Information about College Prices and Financial Aid and Students’ College-Related Behaviors.” American Behavioral Scientist 49 (12): 1620–1635.
Reed, Matthew, and Debbie Cochrane. Student Debt and the Class of 2012. Oakland, CA: The Institute for College Access and Success, 2013.
Rubinstein, Reuven Y., and Dirk P. Kroese. Simulation and the Monte Carlo Method (2nd Ed). Hoboken, NJ: John Wiley & Sons, 2008.
Tierney, William G., and Kristan Venegas. 2009. “Finding Money on the Table: Information, Financial Aid, and Access to College.” The Journal of Higher Education 80 (4): 363–88.
US Department of Education. 2011-2012 Federal Pell Grant Program End-of-Year Report. Washington, DC: Office of Postsecondary Education, 2013.
The Institute for College Access and Success (TICAS). Aligning the Means and the Ends: How to Improve Federal Student Aid and Increase College Access and Success. Oakland, CA: Author, 2013. [End Page 272]

Footnotes

This paper was written when Gigi Jones was employed by the National Association of Student Financial Aid Administrators, and it does not necessarily reflect the views of the United States, the Institute of Education Sciences, or the U.S. Department of Education, where she is currently employed [5 CFR 2635.807(b)].

1. Although the US Department of Education does not make FAFSA filing rates by month available, data by quarter are available from Federal Student Aid. Nationwide, 54% of all FAFSA applications from dependent students were received by the end of March in the 2012–13 aid year, compared with just 34% of independent students. Filing rates by state varied from a low of 27% (Louisiana) to a high of 64% (Indiana).

2. These states are Illinois, Kentucky, North Carolina, South Carolina, Tennessee, Vermont, and Washington.

3. Dependent students are all under the age of 24. Students under that age can become independent if they are married, a graduate student, a veteran, have dependents of their own, or if they are emancipated or a ward of the court. Only the FAFSA for dependent students requires parental information.

4. We focus on EFC instead of family adjusted gross income because of the statutory relationship between EFC and Pell awards. However, most students with a zero EFC have household incomes of less than $30,000 and most other Pell recipients with a positive EFC (EFC greater than zero) have incomes between $30,000 and $60,000.

Additional Information

ISSN
1944-6470
Print ISSN
0098-9495
Pages
253-272
Launched on MUSE
2015-03-19
Open Access
No
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.