Editor’s note: This story is part of the Monitor’s Inside EFAs series, which examines New Hampshire’s Education Freedom Account program.

If recent estimates by dueling policy organizations are to be taken at face value, removing income eligibility requirements for participation in New Hampshire’s Education Freedom Account program could cost the state as little as $6 million or as much as $102 million.

That massive range prompted the Monitor to take a closer look at how the two organizations – the right-leaning Josiah Bartlett Center for Public Policy and the left-leaning Reaching Higher NH – arrived at their numbers.

The projections come as the state faces a budget crunch and as lawmakers consider expanding the program as soon as next school year. The program is currently only available to families who make up to 350 percent of federal poverty guidelines – $112,525 for a family of four.

Both policy organizations’ estimates considered a fully universal program that would be open to every student regardless of whether they were currently enrolled in public or private schools.

Next year, the program is set to provide families between $4,266 and $9,675 depending on their qualifications – money that comes from the state’s education trust fund.

The Monitor’s review of the two groups’ projections found:

■ The organizations are comparing different things. The Bartlett Center’s estimate includes only the potential additional cost for students who don’t currently qualify, while Reaching Higher’s estimate provides a full projected cost estimate of a universal program, including adding in the $27.7 million in current spending. 

■ A significant portion of the gap between the two assessments boils down to differences in what portion of newly eligible students would participate. Reaching Higher estimates that every newly eligible student currently enrolled at a private school or participating in home education would participate in the program, while the Bartlett Center estimates the take-up rate would be 19% in the first year. In other words, Reaching Higher provides a worst-case cost scenario, while the Bartlett Center tried to offer a more realistic figure based on its analysis of past participation.

■ Both groups focus solely on students who have already left the public school system and don’t account for any students who would exit the public schools to participate in the program. In 2024, 32% of newly-participating students exited public school immediately prior, according to data from the Department of Education.

■ Both groups under-count the number of students currently participating in home education in the state due to limited state data.

■ The Reaching Higher analysis assumes that a shift to a universal program would also increase participation from those who are already eligible, policy director Christina Pretorius said in an interview this week. The Bartlett Center does not make that assumption and focuses solely on newly-eligible students.

Take-up rate

Pretorius said her organization’s goal was to give lawmakers a sense of their “cost exposure” if the program eliminates income eligibility restrictions.

“I think it’s really important that the state understands if 100% of these students enroll in the school voucher program – which whether or not is likely, it is legal – they would be on the hook for this amount,” Pretorius said.

“If they don’t go in eyes wide open, … it’s kind of like writing 10 checks and hoping only four people cash those checks,” she added.

Drew Cline, the president of the Bartlett Center and the chair of the state Board of Education, called that approach “reckless and irresponsible”.

“I would push back hard against that and say it’s preposterous to claim that the state would be on the hook for a number that does not exist in any state,” Cline said in an interview this week.

Cline’s group projects that 19% of newly-eligible students would enroll in the program in the first year following the change and 32% would participate in the second year, numbers which he said are equal to participation rates in the first two years after the income-restricted program was rolled out – 2021 and 2022, respectively.

Cline did not immediately share the calculations the Bartlett Center statistician did to arrive at those figures.

Newly-eligible private school students

Two main factors pose challenges to determining how many students would become eligible under a universal program.

First, the Department of Education does not track whether current EFA participants attend private schools versus participate in homeschooling, so it is not possible to determine how many students from each category are already participating.

Second, there is no publicly-released income data about private school and home educating students, so it is not possible to determine what portion are already eligible.

Reaching Higher did not contend with those issues because the group assumed that all students would enroll in the EFA program.

The Bartlett Center started with the number of private school attendees in the state this year – 17,670 – and subtracted the number who attend pre-school – 1,516 – and the number who are non-residents – 4,990. To calculate the latter number, Cline said they received data from the Department of Education that private schools self-report to the state. (Cline noted they found some data errors that the group’s statistician attempted to remedy.) The Bartlett Center did not add in the number of New Hampshire residents who attend private schools out-of-state.

The Bartlett Center then assumed that half of current EFA recipients attend private schools and subtracted that figure, as well.

Lastly, the Bartlett Center multiplied the total number of students who would be eligible by the percentage of families in the state who earn above 350% of the federal poverty guidelines in order to determine how many people would be newly eligible.

Ultimately, the Bartlett Center estimated that roughly 6,000 private school students would become newly eligible. Reaching Higher, which did not focus on new eligibility, concluded that 16,154 private school attendees would be eligible in all.

The Bartlett Center claimed in its analysis that Reaching Higher included preschool students among its private school total, but Pretorius said that was incorrect. It appears the discrepancy between the state data, which shows 16,009 eligible private school attendees, and Reaching Higher’s number – 16,154 – was due to the erroneous inclusion of 136 “PG” students. PG, or “post grad”, students typically attend the state’s prep schools after having graduated high school and are not eligible for EFAs.

Newly-eligible home educating students

If counting private school attendees is challenging, getting a handle on the number of students who participate in home education is far harder. The Department of Education does not track total number, according to spokesperson Kim Houghton.

Both policy organizations relied in their projections on the number of students who notified the state in the past year that they would begin participating in home education, a one-time requirement. The number used – 3,971 – is surely a significant underestimate of the total number of home educating students because it doesn’t account for students who submitted their notifications in prior years.

A Department of Education employee speculated last year that as many as 38,000 students are participating in home education across the state, though he acknowledged he “does question some of the data,” according to minutes from a Home Education Advisory Council meeting.

The employee, Tim Carney, the administrator of educational pathways, did not respond to a request for comment. Houghton said in a statement that the annual notification numbers are “the best data collected” by the department.

Cost-per-student

The amount of money any one student receives depends on whether they qualify for thousands of dollars in additional aid based on qualification for free and reduced lunch, special education needs, and status as an English language learned.

The organizations took different approaches to determining the amount of additional aid newly enrolled students would receive.

Reaching Higher estimated that 26.9% of all the estimated 20,125 students enrolled in the program would qualify for the $2,392.92 in free and reduced lunch aid. That percentage is equivalent to the public school eligibility rate for free and reduced lunch. Currently, 37.1% of EFA participants qualify for the additional stipend. None of the newly-eligible families wouldn’t qualify.

For that reason, the Bartlett Center’s analysis – which focuses just on newly-eligible students – does not set aside any additional aid for free and reduced lunch families.

Both groups projected that the percentage of students who qualify for the $2,184.84 in special education aid would stay the same as it is currently.

Neither group projected any new families would qualify for the aid for English-language learners. Currently, only 2 of 5,321 enrollees qualified.

Under their assumptions, Reaching Higher estimated the cost-per-student for newly-enrolled students would be $4,974.54, while the Bartlett Center estimated it would be $4,410. (The Reaching Higher cost-per-student estimate employs a standardization the Monitor ran to compare “apples to apples,” which assumes linear growth in the program from this year to next.)

Already-eligible students

Pretorius said that she believes that transitioning to a universal program would spark growth not just among newly-eligible students but also among students who can already enroll.

“I think the state can expect a pretty high take-up rate because organizations like the Children’s Scholarship Fund, like Americans for Prosperity, like the [Department of Education] itself, put a significant amount of money in advertising the program and making sure the public is aware of it,” she said.

The number of students participating in the program has risen by an average of 50% year-to-year from 2022 to 2024, though each year the growth has slowed.

Cline said that while “lots of things are in the realm of possibility,” he does not see concrete evidence that the program’s expansion would lead to a faster rate of growth among the population that is already eligible.

“You can speculate about whether more people would enroll or fewer people would enroll, but unless you have some sort of way to gauge that, you’re really just kind of guessing,” he said.

Governor’s proposal

The widely divergent cost estimates may prove academic – at least this legislative session.

Governor Kelly Ayotte’s proposal involves retaining the current eligibility requirements for the next school year before expanding the program the following year to include all students who come directly from public schools. Ayotte projected that change would bring the total program cost to $44.1 million in 2026-27, an increase of $15 million from her projection for 2025-26. A spokesperson in her office did not respond to a question about how that number was calculated.

House Republicans have yet to indicate whether they will adopt Ayotte’s proposal or instead continue to chart ahead with their effort to make the program fully universal as early as this fall.

Jeremy Margolis can be contacted at jmargolis@cmonitor.com.