Impact of a Tax Credit for College Savings. TIAA-CREF and ScholarShare. January 15, 2014 (Update) Prepared for

Transcription

1 Impact of a Tax Credit for College Savings Prepared for TIAA-CREF and ScholarShare January 15, 2014 (Update) Presented by The Blue Sky Consulting Group 1

2 Presentation Overview Background on 529 plans, including what other states do Incentive effects of creating a tax credit for college savings Modeling the impact of a tax credit for college savings Conclusions 2

3 Overview of 529 plans 529 plans created to help families save for college Gains accrue free from federal and state income tax In many states, including California, contributions are made after tax 3

4 Participation in 529 Plans Varies with Income Percent of U.S. Households with 529 or Coverdell Accounts by Income Percentiles Source: US Department of the Treasury Income Percentile 4

5 Many States Offer Incentives 5

6 California s Experience Californians accumulate less debt on average In 2009, the median debt accumulated by California students was about 88% of the median debt accumulated by all students in the country, largely due to the large number of California students in low-cost public education But, the share of students with loans has been increasing The share of first-time students in all California colleges who take out loans has risen from a third to 45% in the last ten years, corresponding to a period of rising tuition due to reduced state funding And, average student debt has been increasing According to PPIC, between 2005 and 2010, average loan amounts among full-time freshmen rose 36%, after adjusting for inflation 6

7 California is one of 15 states that do not offer any sort of tax incentive or matching program for participation in a 529 plan (Perhaps as a result) participation in California s 529 plan lags the national average Accounts per Capita and Per Child California and US California US Accounts per Capita Accounts per Child Source: Blue Sky analysis of data from CSPN and the Census Bureau 7

8 Likely Effects of a Refundable Tax Credit to Encourage College Savings More participants in the ScholarShare Program Increased contributions Increased likelihood that Californians attend and complete college More Californians would move their account from other states plans 8

9 Economic effects Short-term reductions in economic activity In the short run, increased savings results in reductions in economic activity as resources are moved from consumption to savings Long-term increases in economic activity In the long run, increased savings and the investment returns on those savings results in reductions in student debt load after attending college The resulting lower debt payments contribute to increased economic activity, as students have more to spend in the state s economy Increased college going would also boost economic activity 9

10 Research on the Effects of State Tax Incentives Michigan SEED Experiment low-income families were offered a 1:1 match on contributions to Michigan s 529 plan 22% of families made contributions But, total savings increased by just 45%, with 55% coming as a reduction in other savings Oklahoma SEED Experiment Similar to MI experiment; low-income participants were offered a match for contributions to OK s 529 plan 16% of participants opened an account Participants saved more than those in the control group who were not eligible for a match, though amounts saved were small (about $47 on average) 10

11 Modeling the Effects of a State Tax Incentive for College Savings Building on the available empirical research, we developed a model to estimate the effects of a tax credit for college savings The model estimates Account growth Increase in contributions Change in student debt Increase in college going Economic effects General Fund impact 11

12 Model s Empirical Foundations Responsiveness of families to a tax credit based on evidence from research on retirement savings This research suggests that: Savers respond to matching incentives Larger incentives are correlated with more savings The higher the income of the saver, the lower the response to the incentive (since higher income people are generally already saving) Overall increase in participation from such incentives of about 5% to 20%, depending on income 12

13 Model s Empirical Foundations, cont. Evidence from college savings data in states with and without incentives 70% Average Recontribution Rates 65% 60% 55% 50% 45% 40% KY CA MS GA OK MN* VT OR MI WI CT States without Incentive States with Incentive Avg Rate Non-incentive States (52.2%) Avg Rate Incentive States (57.7%) * MN does not currently offer any incentives but until 2011 had matching grants up to $400/yr for families with AGI below $40K. Data for for most states. CA, OR and WI based on , depending on data availability. Source: Blue Sky analysis of data from TIAA-CREF. 13

14 Model s Empirical Foundations, cont. Evidence from college savings data in states with and without incentives Same-State 529 Accounts per Person Under 24 Yrs Old KY CA MS GA OK MN* VT OR MI WI CT States without Incentive States with Incentive Pop-Wtd Avg for Non-Incentive States (0.015) Pop-Wtd Avg for Incentive States (0.034) * MN does not currently offer any incentives but until 2011 had matching grants up to $400/yr for families with AGI below $40K. Source: Blue Sky analysis of data from TIAA-CREF. 14

15 Estimates of the % of Existing 529s Out of State Hart Survey Survey of California households with children regarding college savings Results imply that approximately 36% of accounts are out of state Comparison to National Average CSPN data tracks accounts in each state and nationally If Californians overall own accounts (in CA and other states plans) at the national average rate, results imply that 80% of accounts are out of state Comparison of data for incentive and non incentive states Comparing data for TIAA-CREF managed plans in states with and without incentives Results suggest that if Californians overall own accounts at the average rate among incentive states, approximately 54% of accounts are out of state 15

16 Data Sources ScholarShare historical data for number of new accounts, total accounts, asset balance, contributions, account holders, accounts with CA addresses vs. out-of-state addresses, etc. Re-contibution rates for 529 accounts in states with and without state incentives for (provided by TIAA-CREF) Hart Research Associates 2012 and 2013 survey data (provided by TIAA-CREF) California 2011 tax filing data by income category for joint, single, head-of-household, and married (filing separately) returns from the Franchise Tax Board US Census data for CA households regarding household size, number of children, etc. US Department of Education data regarding the percentage of college graduates receiving Associate s Degrees vs. Bachelor s Degrees California General Fund data from the Department of Finance (DOF) California Economic Output data from the Bureau of Economic Analysis (BEA) Earnings by educational attainment from the College Board (from 2012 Census data) IMPLAN data to estimate impact of changes to savings on statewide economic output Sallie Mae 2013 survey data regarding college savings activity Numerous academic studies and peer-reviewed articles 16

17 Scenarios Refundable tax credit of 20% of amount contributed, up to $2,500 for a maximum credit of $500 per household (except as noted below) Scenarios modeled: A. No income limits B. Income limited to $75K single/$150k Joint C. Income limited to $50K single/$100k Joint D. Phase out, no income limits - Credit amount and match reduced from $500/20% to $400/16% for incomes between $75K and $150K; tax credit amount reduced to $300/12% for incomes above $150K E. Phase out, income limits - Tax credit phased out (by reducing amount) by $75K single/$150k Joint F. Phase out, income limits - Tax credit phased out (by reducing amount) by $100K single/$200k Joint 17

18 Results Our modeling suggests that a refundable tax credit for college savings would Modestly decrease short term economic activity More substantially increase longer-term economic activity Decrease student debt Increase college going Increase General Fund costs, which would be partially offset by increases in economic activity 18

19 Number of Accounts Would Increase (Scenario B - $75/150K Income Limit) 700, , , , , , , California 529 Accounts Note: Includes in-state ScholarShare accounts only Baseline Accts Migration Accts New Accts

20 Economic Activity Would Increase Over Time (Scenario B - $75/150K Income Limit) (50) (100) Estimated Change in Annual Economic Activity ($M)

21 Other Effects (Scenario B - $75/150K Income Limit) Number of College Graduates Would Rise Over 20 years, more than 8,500 additional students would graduate from college Student Debt Would Decrease Over 20 years, total student debt would decrease by nearly $600 million 21

22 General Fund Costs Would Be Partially Offset (Scenario B - $75/150K Income Limit) 20 Impact on State General Fund ($M) 0 (20) (40) (60) Tax Credit (New Accts) Tax Credit (Existing ScholarShare) Net Impact Tax Credit (Migration) GF Impact of New Econ Activity 22

23 Scenarios Compared: Economic Impacts Change in State Output ($M) A. No Income Limit or Phase-out B. Income Limit = $75K Single, $150K Joint C. Income Limit = $50K Single, $100K Joint (50) (100) D. No Income Limit, Phase-out to 12% (max $300) at $75K/$150K E. Income Limit = $75K/$150K, Phase-out starts at $50K/$100K F. Income Limit = $100K/$200K, Phase-out starts at $75K/$150K 23

24 Scenarios Compared: General Fund Impact 0 (10) Net GF Impact ($M) A. No Income Limit or Phase-out (20) (30) B. Income Limit = $75K Single, $150K Joint (40) (50) C. Income Limit = $50K Single, $100K Joint (60) (70) (80) (90) (100) D. No Income Limit, Phase-out to 12% (max $300) at $75K/$150K E. Income Limit = $75K/$150K, Phase-out starts at $50K/$100K F. Income Limit = $100K/$200K, Phase-out starts at $75K/$150K 24

25 Scenarios Compared: Benefits by Income Group 100% 90% 80% Distibution of Tax Credit Benefits by Income Group 70% 60% 50% 40% 30% 20% $200,000 and Over $150,000 to $199,999 $100,000 to $149,999 $75,000 to $99,999 $50,000 to $74,999 Less than $50,000 10% 0% A. No Income Limit B. $75/150K Income Limit C. $50/100K Income Limit D. No Income Limit, Phase-out E. $75/150K Income Limit, Phase-out F. $100/200K Income Limit, Phase-out 25

26 Sensitivity Analysis: Impact of Changing Model Assumptions +/-10% Percent of existing 529s out-of-state (50%) Final Level of take-up/penetration - Migration (80%) Re-contribution rate w/credit (57.7%) Annual baseline participant growth rate (2.0%) Years to full take-up/penetration - Migration (10) Tax Credit Impact - Impact from any tax credit (24.3%) State GF impact per $ of state output (0.047) Percent of 529 beneficiaries who end up in CA (90%) Reduction in other savings (%) (55%) Final Level of take-up - New Accounts (100%) Marginal Increase in college attendance (27%) Student loan average term (years) (10) Tax Credit Impact - Impact per $K Income (-0.21%) Years to full take-up - New Accounts (5) Student loan interest rate (7.0%) Multiplier for economic impacts (2.00) Investment Return Annual Avg. (6.0%) (Scenario B: 20-year Avg Net General Fund Impact = $-34.3M) Change in 20-yr Avg Net General Fund Impact ($M) 26

27 Conclusions Creating a refundable tax credit for contributions to California s ScholarShare plan would Increase the number of participants and the amount saved for college Increase the number of Californians attending and graduating from college Increase economic output and jobs in California Result in an annual average state General Fund cost of between $21 and $57 million, depending on how the tax credit is structured and who is eligible Cost of program is driven by the large number of Californians currently saving in another state s 529 plan who would likely open a ScholarShare account 27