Q&A with Elliot Regenstein About Improving State Early Childhood Systems

Back in 2022, you released a book focused on education policy and the issues of accountability, teacher pay, and school choice. What made you decide to focus your next book on state early childhood systems?

I really enjoyed writing that first book, so in 2023 I was chewing on what might make a fun project for a next book. At the time I was doing some interesting work on governance (in partnership with Watershed Advisors) and on data systems (in partnership with Third Sector Intelligence), and I realized it had been over a decade since there had been a book focused on early childhood systems. As a process person I wanted to write a process book, and fortunately found a publisher who was interested.

The primary audience for this book is people who already have some connection to early childhood systems and policy, and I’m trying to offer my assessment of where the field really is at this moment in time. We’ve come a long way in the last couple of decades but still have a lot of work ahead of us, so it seemed like a good time to take stock of the progress we’ve made and what we might learn from it. My hope is that people who work in the early childhood field will read this book and find themselves mostly nodding along, although if occasionally they come across a point they disagree with that’s great too.

While there are certainly plenty of challenges when it comes to K-12 education policy, the K-12 system seems extremely straightforward when compared to early childhood education. Why have state early childhood systems historically been so fragmented and difficult to navigate?

The whole concept of early childhood systems is comparatively new. Head Start was established in the 1960s but specifically bypassed states. When subsidized child care was established by the federal government, states generally housed the program in a human services agency, which made sense because those agencies had the expertise to manage and distribute those funds. Then when state pre-K started expanding in the early 2000s that was typically housed in a state education agency, in part because it was pitched as an educational intervention. These services are delivered by a complex mix of providers, and there’s a lot of overlap – and we know that parents aren’t choosing settings based on the public funding stream that supports it, they’re choosing it based on fit and affordability.

As these services continued to expand in the last couple of decades, states realized that they didn’t have anybody whose job it was to actually help all of these pieces fit together – nor did they even have basic data on how these services were interacting at the community level. States started making various efforts to develop focused expertise and authority, and also to develop a more holistic understanding of what’s actually happening out there. Not all of those efforts were entirely successful, but they were at least identifying a real problem and trying to address it.

Throughout the book you emphasize the importance of unified leadership for successfully managing state early childhood systems. The past several years have seen a growing number of states moving towards consolidating the governance of early childhood programs into a single department or agency. What do you see as the potential positive impacts of these moves?

There are a few potential positive impacts of these changes – and indeed, in some states they have had an actual positive impact!

One is having elevated leadership. In the old model of early childhood, the seniormost people fully dedicated to early childhood within state government were often division managers who might be three or four rungs on the org chart below the agency head. These division managers have often been very skilled at overseeing their programs, but they have also been excluded from high-level policy conversations in both the executive and legislative branch. Having higher-level political leadership in the executive branch can help early childhood fight for airspace in a crowded policy environment.

Another potential benefit is policy coherence. Historically, states have had state pre-K, child care subsidy, and child care licensing in three different agencies, which can make things very challenging for a provider trying to offer both pre-K and any provider who’s trying to offer both pre-K and subsidized child care. Having unified oversight can make it easier for policymakers to address the regulatory incoherence that often emerges from having three different oversight agencies. Ideally over time states will shift their mindset from managing a set of funding streams to supporting an ecosystem of providers.

Chapter Six of the book is focused on the importance of data for evaluating and improving early childhood programs. You point out that, “Much of the data in the early childhood system is, for lack of a better term, terrible.” For someone unfamiliar with the intricacies of early childhood policy, can you explain why it can be so difficult for states to answer fundamental questions such as the total number of children served by various programs or which children are eligible for which services?

From a technical standpoint, part of the challenge historically has been that it’s been really hard to integrate siloed data. This has been a challenge that goes beyond early childhood, but it’s been particularly impactful in early childhood given our historical fragmentation. Fortunately, new technology makes it easier to integrate data across services, which is really helpful to understanding how Head Start, child care, state pre-K, home visiting, and special education services fit together. States may not have fully developed those integrated data systems, but they’re making more headway than ever before.

It certainly helps that the technological tools to integrate data are advancing rapidly, but the cultural capacity to integrate data is even more important. If nobody in a position of authority is curious about data – or empowered to act on data once it’s produced – then the ability to crank out better reports doesn’t really move the needle for children. States with senior leaders who are thinking about the system as a whole are going to be in a position to use data more effectively to drive policy change, and that’s really the key. Plus from a legal standpoint it’s easier to integrate data across multiple services in the same agency than it is to integrate data across agencies.

In the past, you’ve written about how cloud-based technology can help states improve their data integration processes. Do you have similar hopes for the use of AI in the future or is it too early to know?

Oh, man, good question, and I’d say it’s too early to know. As someone who focuses on state processes, I’d love to see AI used effectively to help analyze data and inform policymakers – that could be really powerful. But it will take substantial capacity and expertise for state governments to effectively leverage AI, and not all state governments are well positioned to do that. I hope we see some successful early adopters that create good models other states can follow. It’s also worth noting that this is a field where a shocking amount of reporting is still being done by paper and pencil; there are a lot of places where even advancing to 10-year-old technology would be a huge leap forward, so for better or worse I’m not expecting us to be on the bleeding edge of adopting AI.

Quality Rating and Improvement Systems (QRIS) were developed in the 1990s to provide families with helpful information for selecting an ECE program and helping to improve the quality of those programs. While you point out that QRIS have often not been successful at achieving these two goals, you also praise Louisiana’s reforms to improve and simplify their system. What steps did Louisiana take to make their QRIS more helpful to parents and providers?

The main thing Louisiana did was focus on the quality of the classroom experience. We actually know a great deal about what makes for a great experience for children in their early years. Louisiana focused on that and made it central to its QRIS, which feels correct to me. In systems work there are a lot of decisions that are very contextual: how state government works, how local infrastructure should be built, and how advocates will engage can vary enormously across states, and that’s totally fine. But great experiences for children should be at the heart of the conversation in every state, and ideally QRIS will reinforce that focus.

After closely examining state early childhood systems for this book, how do you generally feel about the direction in which state-level early childhood policy seems to be headed? Is there more reason for optimism or concern when thinking about the next few years?

I’m optimistic for sure, even though the concerns are real. I got into this field a little more than 20 years ago, and in that time we’ve clearly made a lot of progress – which is exciting! That progress built on decades of foundational work, and fortunately it seems like we’re continuing to accelerate. Advancement is never linear, but over time this is a field that’s accomplished a lot and has much to be proud of.

It doesn’t get easier from here, of course; many of the early adopters of unified governance still have substantial work to do to build capacity, and some of the states that haven’t adopted unified governance have faced significant headwinds to early childhood systems development. So the road ahead is probably uphill. But we know a lot more than we did 20 years ago, and there are a lot of terrific people who’ve come into the field and are making an impact around the country. I hope my book will be helpful for everybody in the field, and that when we look back in 10-20 years we’ll see that we’ve made even more progress on all of the issues it discusses.