Evidence Analysis of the 2025 Ballot: Propositions LL & MM
Regarding the Healthy School Meals for All Program and Supplemental Nutritional Assistance Program
Executive Summary
Why this Report?
Ballot measures reach voters through the General Assembly, known as referred measures, or when citizens initiate measures. Leading up to election day, Colorado voters typically receive information from the Blue Book voting guide, which provides a nonpartisan summary of the measure, and from supporters and opponents of these ballot measures through paid media and direct voter contacts. Each side makes claims about the positive or negative consequences that will result if the ballot measure is passed.
Many of these ballot measures have significant consequences. But too often missing from the public discussion is an objective analysis of the evidence behind the supporters’ and opponents’ claims about the impact of the measures.
This report is designed to fill that gap by equipping voters with the best available unbiased information to make their decisions.
In keeping with established frameworks for characterizing levels of evidence, this report will identify evidence levels on a 1 through 5 scale, with 5 being the most rigorous, and evidence certainty as “high,” “moderate,” “low” or very low,” so voters can put the information about ballot measures and the impact they may have in context.
The Ballot Measures
Two referred measures are on the ballot in November 2025, Propositions LL and MM.
Proposition LL would allow the state to keep the $12.4 million that was collected over the amount allowed by Proposition FF, the ballot measure in 2022 that passed to initiate the Healthy School Meals for All program.
If approved, the state would spend this additional revenue on the Healthy School Meals for All program instead of refunding the funds to taxpayers. This would impact households earning $300,000 and above annually who are taxed to fund the Healthy School Meals for All program.
Proposition MM would increase the amount of revenue collected for the Healthy School Meals for All program by reducing tax deductions for taxpayers earning $300,000 or more annually, effectively raising their taxes. The measure also expands the allowable uses for the revenue collected to include costs incurred by the state for implementing the Supplemental Nutrition Assistance Program (SNAP). Both of these measures require voter approval in accordance with the Taxpayer Bill of Rights (TABOR).
The Healthy School Meals for All program provides:
state reimbursements to local schools that provide free meals, including breakfast and lunch, for all students at public schools that participate in the National School Lunch program and School Breakfast program;
a grant program offering the greater of $5,000 or 25 cents per lunch to participating schools for a local food purchasing program;
technical assistance grants for local food purchasing; and
increased wages for school employees who prepare and serve lunch, the greater of $3,000 or 12 cents per school lunch.
The goal of the Healthy School Meals for All program is to provide school meals to all students at no cost.
What does the evidence say?
Analysis of the claims of positive impacts from the program:
Increased participation in school meals and nutritional quality
| Evidence: | High
| Multiple studies suggest that offering free school meals increases participation in both breakfast and lunch, which improves nutrition particularly among schools that prioritize healthy meals.
| | |||
Improves test scores and attendance
| Evidence: | Low
| There is low evidence that suggests that test scores improve, and when improvements to test scores are observed, they are very minor. Attendance improvement is also low, but may improve a couple years after universal free school meals has been implemented.
| | |||
Reduces obesity
| Evidence: | Low
| Research suggests minor improvements in body weight through participation in free school meals.
| | |||
Improved nutrition decreases disciplinary action at school
| Evidence: | Moderate to Low
| Evidence suggests that changing students’ overall diet (rather than just adding supplements) has the biggest positive impact on negative behavior, especially for boys, showing a small effect on aggression, a medium effect on antisocial behavior, and a large effect on criminal offending.
| | |||
Increased nutritional content of lunches prepared at school versus home
| Evidence: | Moderate
| Despite parents’ concerns about the quality of school meals, the study shows that school-prepared lunches are generally more nutritious than lunches brought from home.
| | |||
Positive mental health impacts from reduced food insecurity
| Evidence: | Varied
| Adequate nutrition can have a positive impact on the mental health of both students and parents:
| Depression: | High | Food insecurity is strongly linked to depression across populations, with especially high risks among non-Hispanic Black and Hispanic families, children facing suicidality, and those experiencing more frequent or severe food insecurity, creating a cycle where depression and food insecurity reinforce one another.
| Anxiety: | Low | Food insecurity shows a less consistent relationship with anxiety than with depression, appearing more clearly in adults than children and varying by race and ethnicity, with stronger evidence in non-Hispanic White populations than in Hispanic populations.
| Hyperactivity and Behavioral Issues: | Moderate | Food insecurity, particularly in younger children and in cases of persistent or severe food insecurity, is associated with hyperactivity and behavioral issues.
| Stress: | Low | Food insecurity may act as a chronic stressor tied to poor health outcomes, though evidence directly connecting food insecurity to psychological stress is mixed, possibly due to differences in how stress is measured and the unique ways parents experience food insecurity compared with the general population.
| | |||||||||
Analysis of the claims of negative impact from the program
Increased food waste
| Evidence: | Low
| Longer lunch lines without increasing the allotted time to eat lunch results in students who opt into school lunches having less time to eat. One result is the potential for increased food waste. Results were inconclusive, however the study notes that average food waste is on par or below
| | |||
Increase in state tax rates impact on economic activity
| Evidence: | Moderate
| An increase in state income tax is found to reduce personal income growth and, in the long run, state income tax collections. There was no indication that the increase in state income taxes impacted unemployment or GDP.
| | |||
The goal of the SNAP program is to ensure reliable access to food for individuals below a certain income level
What does the evidence say?
Analysis of the claims of positive impacts from the program:
Poverty reducing and non-nutritional effects of SNAP participation
| Evidence: | Varied
| Adequate nutrition can have a positive impact on the mental health of both students and parents:
| Healthcare utilization: | Moderate to High | SNAP participation has a positive effect on access to healthcare for both preventative and needed care.
| Family allocation of resources: | Moderate | SNAP participation improves families’ ability to manage monthly household expenses, including housing, utilities, and food costs.
| Impact on child development and behavior: | Low | SNAP participation is tied to some improvements in child behavior and academic performance.
| Mental health: | High | There is a strong link between SNAP participation and improved stress and depression for parents and improved mental health in children.
| Abuse and neglect: | Moderate | Most studies associated participation in SNAP benefits with lower risk for child abuse and neglect.
| | ||||||||||
Dietary sufficiency and quality for SNAP participants
| Evidence: | High
| Research shows that participating in SNAP ensures that families have sufficient caloric intake, but the quality of the nutrition is lower for people on SNAP than for people not receiving SNAP benefits.
| | |||
Analysis of the claims of negative impacts from the program
Barriers to SNAP Participation
| Evidence: | Moderate
| There are various policy, administrative, environmental, access, and household barriers to SNAP participation that result in many households that are eligible for benefits not applying for or receiving benefits.
| | |||
Why are we doing this analysis?
Ballot measures reach voters through two means. The General Assembly can refer a measure to the ballot, or the citizens can initiate measures. Ballot measures are currently characterized for the public by advocates and opponents through paid and earned media and other distribution channels, along with a descriptive summary and analysis from the Legislative Council Staff of the Colorado General Assembly. This analysis is called the Ballot Information Booklet, more commonly known as the “Blue Book.” The Blue Book voting guide provides an important resource for voters, but it does not include information about the efficacy of the components of the measures.
The objective of this report is to provide Colorado voters with clear, unbiased details about ballot measures earlier in the election cycle about the proposals they will consider in November. Too often missing is a thorough evidence-based analysis of each ballot measure. What is the level of evidence supporting the proposal? What is the likelihood that the proposal will achieve its intended objectives or impact?
This analysis aims to fill that gap.
What is evidence-based policy?
Evidence-based policy, which has its roots in evidence-based medicine, has been defined as “the practice of informing public policy decisions through the use of scientific evidence and rigorous research…ensuring that policies are grounded in strong and credible empirical data, which can…enhance overall effectiveness.”1F [LINK]
The strength of research evidence is commonly assessed through classifying levels of evidence. Two of the most widely recognized frameworks for making that assessment in medicine which have more general application are the Oxford Centre for Evidence-Based Medicine (OCEBM) and the GRADE Working Group frameworks, as described in the Appendix in further detail.
Significant progress has been made in interpreting these medical research levels of evidence into the policymaking process..
In keeping with these established frameworks for characterizing different levels of evidence, and for clarity and accessibility, this report will refer to evidence levels as defined below; and to evidence certainty as “high”, “moderate”, “low” or very low”. Additional information can be found in the Appendix.
Next Steps
XXXXXXXX
Conclusion
XXXXXXXX