Vermont Legislature SmartTranscripts:

[Grace Miller (Guest Presenter)]: We're live.

[Sen. Seth Bongartz (Chair)]: Hi, Grace. You are still muted.

[Grace Miller (Guest Presenter)]: Hi, everyone.

[Sen. Seth Bongartz (Chair)]: Hi. So I did kind of do a quick introduction, but you why don't you, obviously, you can do it better than me, introduce yourself and talk a little bit about what your thesis was and then the conclusions that you had. You and I talked over the weekend, so I have a sense. It's why don't we, we'll quickly introduce ourselves around the room. If you've been you've been watching. So Mhmm. Out of the room right now, but probably coming back in is senator Ram Hinsdale, and then next to her. Nader Hashim, Windham County. Good afternoon. Dave Weeks representing Rollins County. And Seth, I represent Bennington County. Terry Williams representing Rutland. Steve Heffernan, Addison County District.

[Grace Miller (Guest Presenter)]: Awesome. It's nice to meet all of you. If you don't mind, I'm going, I prepared a little bit of a presentation. I'll go ahead and share my screen.

[Sen. Seth Bongartz (Chair)]: So Grace, just actually for your record, mean, for the record, your name?

[Grace Miller (Guest Presenter)]: Yes, for the record. My name is Grace Miller. I am presenting my research that I produced as a part of my undergraduate senior essay at Yale University. And now I work as a mathematics teacher in Tennessee.

[Sen. Seth Bongartz (Chair)]: And just for the picture of this gets worked in, you are from, grew up in Newport.

[Grace Miller (Guest Presenter)]: Oh, of course. Will always loop that in. Don't worry.

[Sen. Seth Bongartz (Chair)]: And you leave it in. All right. You're welcome. Okay.

[Grace Miller (Guest Presenter)]: You guys should be able to see my screen now. Okay, and hopefully you can hear me all right. So again, my name is Grace Miller. Grew up in Newport, Vermont, and I'm a proud graduate of North Country Union High School. Like I said, I completed this research on the fiscal impacts of Vermont's acts 01/5346 as a part of my economic senior thesis at Yale University. But now I teach in a very large district in Tennessee that serves over 40,000 students, which is obviously a very different educational context from the Northeast Kingdom. And that has definitely shaped how I think about scale and governance more in practice. Since completing this study a year and a half ago, as I'm sure you all know, there's been a couple articles that cite my work in ways that do not fully reflect the findings or their limitations. So I appreciate the opportunity to be here today with you guys to clearly explain what this analysis does and does not show, and to discuss what those findings mean moving forward. So of course I was looking at acts one hundred fifty three and forty six, which was responding to the very interesting district structure that Vermont had and kind of still currently has. By 2010, we had one of the largest numbers of school districts in the country, over two seventy serving roughly 85,000 students, which was correlated with high increases in inflation adjusted per pupil expenditures. So the AOE and the legislature, y'all pursued district mergers explaining your goals to provide substantial equity, lead students to achieve or exceeds the state's educational quality standards, maximize operational efficiencies, and promote transparency and accountability, all while delivering education at an efficient cost. And so of course that idea behind school district mergers is rooted in economies of scale, that assumption that as districts get larger, they're providing the same services at a lower per student cost by reducing redundancies among administration and contracted services. Economists around the country have done pre consolidation cost analyses. It's pretty obvious why we would expect this to lead to savings in the aggregate, but to date only a very small number of empirical studies have actually examined post merger fiscal outcomes with very mixed results. And at the same time, a lot of qualitative feedback from administrators and educators points to diseconomies of scale, so that as districts grow, newer costs are emerging like transportation, coordination challenges, and other bureaucratic layers. So I set out to explore these main three questions surrounding district mergers in Vermont. Specifically, can economies of scale be achieved without consolidation? I think it's very important that in Vermont, district mergers often have not involved in school closings, despite that being a common fear. And so that raises the important question of can meaningful savings come purely from bureaucratic restructuring? Also, how does the merger process affect student equity? Improving equity is one of the most frequently cited justifications for mergers by the AOE, both before and after they occur. Yet there is little clarity around how equity is defined or measured within Vermont. And in this study, I examined changes in specific spending allocations and drawn interviews with superintendents and principals to better understand how equity is perceived and operationalized post merger. And then finally, how do very small rural school districts benefit from mergers and what characteristics of mergers impact their success rate? Given Vermont's rural geography and then also coming from the Northeast Kingdom myself, I find this question especially important because organizational decisions that may work in larger regions, maybe Chinden County can look very different in sparsely populated communities. So I wanted to briefly explain how I collected and structured the data for this study. I was provided district structured data and detailed fiscal expenditure spreadsheets by Brad James, the former director of education finance. And it's important to know that for the context of this study, I treated merged districts as a single unit over the entire observation period. In practice, that means if two districts later merged, I aggregated their spending, enrollment, and tax data before the merger, as if they had already been operating as one district, which allows for a clean comparison of outcomes both before and after the merger. So just a simple example, if two districts spent a combined 20,000 per pupil prior to merging, I can then directly observe whether that figure declined, increased, or stayed constant after consolidation relative to non merged districts. Of course, I was not only interested in how total education spending, but also where money moved within district budgets. So I used the detail function object and project level fiscal codes provided to me by the AOE to kind of create my own spending buckets, spending categories across years. So this included stuff like contracted services, student and teacher support, special education services, food, transportation, to name a few. And then of course I contacted all superintendents across the entire state and all principals of Merge districts to try to match my quantitative analysis with qualitative data to inform the results that I saw later. So obviously the main goal of my analysis was to find the average fiscal impact of experiencing a district merger. But to articulate how I did this, I want to give a very quick example. So let's say instead we want to find the average increase in wages from getting a college degree. If we have college educated Sarah making $90,000 and high school educated John who makes 60, can we deduce that a college degree increases your wages by $30,000 Of course not. There are all these confounding variables like family background, baseline ability, gender gap that are impacting their wages behind the scenes. What we would want is to compare two identical versions of Sarah, one who attended college and one who did not, so that education is the only difference between them. Obviously, can't do that. So what can we do in the context of mergers? Instead, we can compare changes over time between two groups that are similar on average across all known and unknown confounding factors. That's the intuition behind a difference in difference analysis. So rather than comparing levels, we compare trends before and after a policy change between a group that experiences the intervention, merge districts and a group that does not. This approach relies on two things. One, that districts that experienced a merger do not differ on any known or unknown factors that would influence their experience post merging. And also something called parallel trends, which you can see in that graph right there. That in the absence of the intervention, this case a merger, merged and non merged districts would have followed similar spending trajectories over time. The levels don't need to be identical themselves, but the trends need to move parallel. So then any systematic deviation after the merger can be attributed to the merger itself. Vermont is great for this type of analysis because we're relatively racially homogenous, predominantly rural, and importantly, many mergers were either mandated strongly incentivized rather than what I would call purely voluntary. And so it reduces the concern that districts self selected into mergers based on unobservable characteristics. So to prove this random assignment of mergers, I compared merged and non merged districts on several pre merger characteristics like average daily membership, household income, per pupil spending levels, growth rates, and I found no statistically significant differences between the two groups. So this supports that assumption and can help verify our results later on. The idea of parallel trends that absent a merger spending in merged and non merged districts would have increased at the same rate over time was also verified by the data, almost beautifully, might I add. So what you're looking at here is the average spending per people over time for merged districts in red and non merged districts in blue. And so prior to the merger period and after, which is important, the two trends moved almost perfectly in parallel, which gives us confidence that the districts were on similar fiscal trajectories before consolidation and any divergence afterwards can reasonably be attributed to the merger itself. And a really important implication of this approach that may be hard to understand is that it allows us to subtract out factors that affect all districts at the same time. Things like inflation, statewide policy changes, or macroeconomic conditions, because those forces influence both merged and non merged districts. So the remaining difference isolates our effect of the merger. Here's an example in the second set of figures of what happens when spending deviates from that parallel trends assumption. So on the left is average per people spending on administrative support. The trends you can tell are pretty parallel prior to the merger, and then around 2020, we see a clear drop in administrative spending for merged districts relative to non merged districts after consolidation, which is a pattern consistent with one of the primary policy goals of mergers and what we would expect to see that we reduced administrative redundancy. On the right, which is very interesting is average transportation spending. So again, the trends are pretty parallel prior to the merger, but then following consolidation transportation spending increases for merged districts actually. So what did I find overall? I find sorry

[Sen. Seth Bongartz (Chair)]: go back one slide just bud. What happened on the slide on the letter? What happened with that rapid rise in 2021?

[Grace Miller (Guest Presenter)]: I would assume that is respective to COVID-nineteen. I don't specifically know why that would increase administrative support spending so heavily, but it like what happened in 2020, it's happening to both merged and non merged districts. So we would have to talk to district officials to properly identify what was going on in the budget there. So overall, I find no evidence that district mergers led to significant savings in per people spending or importantly slowing of its growth rate. Several administrators mentioned to me in conversations that without the merger, their spending would have grown faster, but the data doesn't necessarily support this. However, and this is very important, specific budgetary allocations were significantly altered. There was a decrease of $386 per people on administrative support spending and $2,168 on contracted services. But that money was then met with an increase in spending per people on salary and benefits, materials, teacher and student support spending and transportation. So these results indicate to us that although there were neutral aggregate savings, Merge districts were able to invest more in budget sectors that directly impact students and teachers, an investment they wouldn't have been able to make without increasing their spending had they not merged. And maybe transportation is a little bit of an asterisk on that conclusion. If this money was not reallocated, the entire savings represents 6.5% of the average budget for merged districts in 2024. So if you have my thesis, you can see this table a little bit more clearly in there, but my conversations with superintendents and principals were very interesting and particularly important for understanding why the quantitative results looked the way they do. If we looked only at the fiscal data and assume that the primary goal of mergers was to save money, we might conclude somewhat naively that the mergers overall failed because they did not generate substantial cost savings. But I'm more curious as to why. Do mergers inherently lack the ability to produce savings? Was it the way they were implemented? Do educational leaders in Vermont conceptualize the goals of mergers differently than policymakers do? These are kind of the questions I was thinking about. And so the results were incredibly mixed. Roughly, I think I talked to about 18 people in total, and roughly half of the administrators reported experiencing fiscal benefits following the merger, while the other half reported seeing no cost savings. But among those who described positive experiences, the benefits they emphasized focused specifically on equity. Several administrators described a notable shift in how decisions were discussed at the board level and meetings moved from being framed around individual districts saying, oh, I am from this school to a district wide perspective, rather I'm a board member representing this district. And the administrators told me that that shift made disparities in stuff like curriculum, materials, and programming across schools more visible. And in some cases, those inequities were addressed directly. Administrators also cited the ability to share personnel across schools as a benefit. One superintendent described hiring a single French teacher shared across multiple schools rather than leaving each school to staff the position independently. And others noted that merging allowed them to fill support roles that had previously fallen entirely on individual principals. Another frequently cited benefit that I was not able to explore throughout my analysis was the idea of distributing the tax burden of high fixed cost, particularly special education services and capital construction across a broader tax base. And we would expect for small rural districts with limited enrollment that spreading these costs is viewed as a meaningful form of financial relief, even if total spending did not decline. But on the other hand, half of administrators I talked to were more skeptical or did not see any realized benefits from the merger process. They cited things like a decrease in personal connections and community ties. But many also noted that the concern within the public over loss of local control tended to diminish diminish in the post merger period. Several noted that any savings that did occur were absorbed in the reallocation of distributing shared materials and new personnel, which aligns with that increase in transportation expenditures we saw, although I don't know what directly increased transportation costs If you're compensating the gas of that shared French teacher across several schools, then you would assume transportation costs would go up. Also some administrators viewed the merger process as largely bureaucratic, reporting little change in instructional practices or operational strategy beyond operating under a larger budget. And that larger budget part is also important because some superintendents explain that consolidation made budget approval more difficult as voters were asked to pass a single high dollar budget covering multiple schools rather than a smaller school specific budget. Then perhaps the most important testimony I was given came from two principals within the same district, two principals from different schools within the same district. They responded to the question, has the merger resulted in financial benefits for your district? And the first principal answered, yes, we are able to share resources and finances to help support all of our schools and student needs. But the second principal responded, we have seen minimal gains in financial savings, in parentheses, buying bulk supplies, which were reallocated. So I think that this glaring inconsistency should definitely caution both the AOE and the legislature against relying solely on quantitative or qualitative testimony to evaluate merger outcomes. I think administrator perspectives are essential for understanding implementation and lived experience, but the divergence in responses even within a single district definitely highlights the need for systematic post merger quantitative evaluation alongside our more qualitative evidence. Okay, so what does this all mean? So I think that overall, there is very little evidence that district mergers on their own will substantially mitigate Vermont's education finance challenges and the potential for large immediate cost savings definitely appears limited at this point. But that being said, and I want to make this clear, this is not necessarily an argument against merging, nor is argument that mergers offer no value. I think that Vermont's current district structure is clearly inefficient in some places. For example, I don't really think there's any justification for Newport City and Newport Town operating two separate elementary school districts when they're like roughly ten minutes apart and serve practically overlapping communities. So redistricting in cases like this is definitely reasonable. However, I think if the state is going to redraw district boundaries or pursue more district mergers, the findings of the study point strongly towards the need for clearer directives and more measurable goals from policy leadership. Because the data reveals a very wide range of merger experiences that don't necessarily correlate, or systematically vary by district size. Excuse me. Let me get a drink of water really quick.

[Sen. Seth Bongartz (Chair)]: That's fine. That's fine. Don't worry.

[Unidentified Committee Member]: So to go back to slide 11 when you asked the question about the increase of spending, that was the infusion of ESSER money, ARPA money, you know, and schools and municipalities were ill for that money because they didn't have a plan of the expenses.

[Grace Miller (Guest Presenter)]: Yes.

[Unidentified Committee Member]: Yeah. So now we've got, you know, we've got in place as far as the school bus contracts and Mhmm. It it goes up quick, but it never comes down quickly.

[Grace Miller (Guest Presenter)]: Absolutely. That's why I think the looking at those specific budgetary allocations was really interesting. But of course, just saying, oh, we increased spending on salary and benefits by a 212. The that doesn't answer the question of like, what were we spending that money towards in salaries? The hedging report just released that I said that we spent that money on higher salaries. That is not what I said. Like, I think that testimony from administrators would point more towards, oh, they hired additional personnel that they weren't able to hire previously. Again, so like, I don't know specifically where that money was placed. But I am definitely not saying that all of a sudden we started paying a specific person in that district a thousand dollars more. Okay, so back to conclusions. I think something that's really interesting and very important to note is that districts that merged in 2019 were the only group that consistently reduced administrative spending without offsetting those reductions through increases elsewhere. And by contrast, districts that merge in earlier years experienced more volatile reallocation patterns for unknown reasons. And it raises an important question of what were districts that merge in 2019 doing differently and what constraints or expectations kind of shaped earlier mergers? And of course, I don't know, but I think that differences in administrative testimony suggest that initial strategy, expectations and approach to the merger process probably most likely play a critical role in determining outcomes. But to be clear, this is not a critique of district leadership, but I think it reflects a broader policy challenge that many administrators were never necessarily given clear guidance on what success was supposed to look like, or clear expectations about how that success would be measured. And I think a lot of key questions kind of remain insufficiently defined. Like what do we mean by student equity? They could take on so many different roles, so many different ideas. And how much are we expecting to save from district mergers? Where should the savings come from? Are we going to everything that we save, is it just going to go straight back to taxpayers? Should the funds be eliminated entirely or intentionally reallocated like we saw in those districts that merged in the earlier years? I think without explicit answers to these questions, it's gonna be difficult for districts to implement mergers in any way that produces consistent measurable outcomes. Obviously mergers are complex, particularly when school closures are not a part of the conversation. And so I think that merger, this study suggests that mergers are more likely to be effective when we're paired with narrower, clearly articulated objectives, smaller measurable goals, such as reducing specific administrative redundancies or standardizing curriculum and instructional materials across schools. This may be a more realistic and measurable path moving forward. Obviously, I think that district mergers are going to continue. So I see two plausible paths forward. And the first is incentivized organically driven mergers initiated and led by local communities. The advantages of this approach are pretty clear. It preserves local control, which remains highly valued. It could potentially reduce political backlash by placing that decision making and accountability in the hands of communities themselves. And it allows districts to tailor mergers to local relationships, geography, and institutional knowledge, which we know local leaders often have the best understanding of their communities needs and constraints and all of that nuance. But at the same time, I think this approach also carries meaningful risks like without statewide consensus on goals or clear definitions of success that implementation is likely to remain inconsistent across districts. And that inconsistency limits the potential for predictable fiscal outcomes and makes it really hard for us to evaluate their effectiveness. Or it could be longer transition periods, delayed impacts and continued public confusion, which may be organic mergers would kind of mirror the outcomes we've seen, we saw under 01/1953 '46. Modest or neutral savings, heavily, heavy reliance on qualitative narratives and heterogeneous experiences. I think on the other hand, you could have more state guided mergers with clear oversight. So this would allow for potentially allow for measurable and actionable objectives with clear expectations set at the state level and implementation can be more consistent and outcomes more predictable. It also facilitates post merger evaluation, which I think is particularly essential if we're going to use mergers to inform policy decisions going forward. State guidance also potentially can allow consolidation efforts to align more directly with Vermont's statewide education funding system, which I know is also going to change very, very soon. Rather than leaving districts could kind of navigate mismatch independently. But then of course, there are important cons to this method as well. No one likes being told what to do, especially not Vermonters. And so state guided mergers are likely to face strong political and community resistance. And I think they definitely also require a substantially greater upfront capacity from state leadership and introduce more bureaucratic complexity. If poorly designed or implemented, I think it could amplify inequities or create new inefficiencies rather than resolving existing ones. And then of course, there is a legitimate concern among many Vermonters, especially in our more rural communities about whether state level leadership can design mergers that are sufficiently informed by the nuanced needs of individual communities. I know for many Vermoners, especially where I come from, the fear is not just about losing local control, but more so losing local understanding. I would say finally, that obviously with all studies come limitations. I did not have data on school closures during the merger period, at least in 2024, Vermont did not have a list of schools that had been closed anywhere that I could find. I don't think very many schools were closed. And while transportation spending did increase following the mergers, I cannot definitively attribute that increase to longer busing times, nor would I be inclined to. Mergers were just purely more bureaucratic. I also can't really observe how the benefit of high fixed costs being spread over a larger tax base, because I aggregated all of the merged districts prior to merging. And then I didn't examine student achievement, which of course is important, but it probably, it of course requires a longer post merger window in different data sources to assess this rigorously. Finally, I just think that taken together, these limitations kind of point to a broader issue that extends beyond this study that Vermont has in the past relied heavily on qualitative feedback projections to evaluate major education reforms, but there has been far less systematic post policy evaluation, which definitely creates real challenges for learning from past reforms and improving future ones. I say this as a Vermonter who deeply deeply values the state's history of thoughtful and progressive education policy, and as someone from the Northeast Kingdom who deeply values and understands the benefits that small rural communities provide. The lesson here reform is misguided or that 01/1953 was misguided, but that reform should be iterative and evidence driven rather than exploratory. Without clearly defined goals and post policy evaluation, I think that large scale change risks becoming experiments with the costs being borne primarily by students and taxpayers. So obviously the question is not whether change is necessary or whether district mergers are coming. That is why we are all here. But I think the question is whether we are defining what success looks like in Vermont and for Vermont students, and then measuring it consistently before moving on to the next major reform. And yeah, that is my presentation. I can go back to any slide.

[Sen. Seth Bongartz (Chair)]: Thank her for her time. In-depth.

[Grace Miller (Guest Presenter)]: Grace, would you like to run for the state senate? Please reach out to me. Thank you.

[Sen. Seth Bongartz (Chair)]: She's an economist. Maybe

[Grace Miller (Guest Presenter)]: when I come home from Tennessee. Right. Yeah

[Sen. Seth Bongartz (Chair)]: I think we have

[Unidentified Committee Member]: a job opening at JFO for an economist.

[Sen. Seth Bongartz (Chair)]: There you go. Grace I think the reason there's not that questions is because you were pretty clear, thorough and thorough and we had a very clear sense of what we're we heard you. Yes. And it was really helpful. Why I found I found our conversation really interesting the other day and and this is

[Grace Miller (Guest Presenter)]: Awesome. Thank you. I I'm just glad I got to I got to say my piece. I I've been I've been a little tired of random articles saying random things about what I wrote.

[Sen. Seth Bongartz (Chair)]: You do love what you actually said. Yeah. I understand. Well, thank you. And we know how to find out. We have we know how to get in touch with you and you may have once again.

[Grace Miller (Guest Presenter)]: Have you encouraged AOE to listen first?

[Sen. Seth Bongartz (Chair)]: No.

[Grace Miller (Guest Presenter)]: Pretty nice. They think they're listening now.

[Sen. Seth Bongartz (Chair)]: They they want me to listen. So have a great thank you so much. And with that we are adjourned. Have a good weekend. Thank you.