Affiliate Spotlight: Christopher Knittel on researching firm and consumer responses to environment and energy policies
Christopher Knittel is the George P. Shultz Professor and a professor of applied economics at the MIT Sloan School of Management. In this J-PAL affiliate spotlight, Chris sheds light on the research questions he’s pursuing to inform policies on greenhouse gas emissions and carbon co-pollution reductions.
What first drew you to study economics, and why did you choose to focus much of your research around environmental economics and how firms and consumers respond to policies?
I expected to be a lawyer when I entered undergrad, but then I took my first economics class and fell in love with it. I was interested in firm behavior and consumer behavior around firms. My PhD work was in industrial organization. I was doing a lot of work on banking and finance, but when my son was born, I started worrying more about environmental issues. That drew me toward research on local pollution and climate change, and I’ve focused on this ever since.
What research questions are driving your current research agenda?
The overarching question driving my research is how consumers and firms respond to policies and changes in market conditions: whether this is energy prices or policies that impact prices, or policies more generally. I like to come full circle and ask what those responses mean for policy effectiveness and efficiency. I try to understand what firms and consumers do around the energy space and what products lead to pollution, and then write papers to help policymakers design better policies. We’ve focused on energy efficiency because that has been a bipartisan focus for many policymakers. Going forward, I hope that policymakers will enact broader measures around climate change. Much of my work aims to understand climate policies that aren’t on the map today but hopefully will be considered in the future.
What is an example of a randomized evaluation measuring the impact of an environmental or energy policy program that particularly resonated with you? Why did you choose to do this evaluation?
I am wrapping up a study on industrial energy efficiency. Energy efficiency policies have bipartisan support, and a lot of that work focuses on residential consumers—yet questions remain about commercial and industrial facilities and what the impacts of energy efficiency policies might be for them. Catherine Wolfram, Michael Greenstone, and I are finishing an experiment in California that outfitted one hundred industrial facilities with sub-metering. We measured how much energy the compressed air systems were using.
Compressed air might sound boring, but it’s an important consumer of electricity: twenty percent of industrial electricity is used by compressed air. We worked with large firms whose monthly electricity bills were tens of thousands of dollars. To conduct the study, we teamed up with a startup that installs sub-meters and uses data analytics to understand what’s going on. Sometimes, savings were as simple as a notification from the startup saying: “did you know your compressed air machines are running on the weekends, even though you’re not here on the weekends?” Right now, we’re analyzing the data. I don’t want to give away the results, but it’s one of the first experiments that I’m aware of in the industrial space, and we’re excited about this.
What have been some of the most meaningful results or outcomes of your work so far?
Hunt Allcott and I ran a randomized evaluation about how consumers think about fuel costs when buying a car. If you look at the cost-benefit analysis of fuel economy standards, it assumes consumers are myopic (shortsighted), and standards try to correct this. The belief that consumers make mistakes is often conventional wisdom among policymakers. The ideal evaluation to illuminate whether this is accurate would be randomizing loan rates and seeing whether consumers invest more in fuel economy when offered lower interest rate loans. Although we could not do that, we instead asked how consumers respond to information about fuel economy and costs. One reason why consumers might be myopic is if they don’t have such data easily available.
We teamed up with Ford and developed an iPad app about fuel economy and randomly assigned customers walking into seven dealerships across the country to either use the app or see some unrelated information. The app asked what car they drive, what they think gas prices will do in the future, and what car they’re considering. The app took that information and showed them what would happen if they switched to a more fuel efficient car. It translated savings into more tangible terms, such as how many iPads or pairs of jeans they could buy with that money. We then observed what cars those consumers bought.
The results showed that this information does not have an effect on purchases. This is consistent with the idea that consumers were already making accurate assessments of fuel savings, at least on average. We saw both overestimates and underestimates of fuel costs. On average, the information isn’t affecting them, although there is likely to be heterogeneity in these effects (variation in how the treatment affects people), and we were under-powered to test that.
How have your research findings been received and used by the non-academic community?
Questions about how consumers think about fuel economy and the best ways to decarbonize transportation are front and center in the minds of many firms and policymakers. I am fortunate to run a research center at MIT that connects with industrial affiliates and policymakers about research. I’ve heard this study cited in DC and Sacramento and have presented it to policymakers and the Auto Alliance. I think it is one more piece of information for an important policy question.
I think that policy in DC and Sacramento is like a huge boat that takes a long time to turn. The focus on fuel economy standards is going to take a long time to turn towards a focus on something like carbon taxes or alternative policies. I can’t say I’ve seen “the end result” of my work, but I think this work has at least made some policymakers question conventional wisdom, in terms of how consumers make decisions. On the margin, I think my work has helped move them away from what economists have found to be inefficient policies toward more efficient policies.
Authored By
As the new school year begins and COVID-19 cases continue to spread across the United States, questions remain about how to educate students effectively and safely. We sat down with affiliate Sarah Cohodes to discuss the school reopening process, the impact of remote learning on students and parents, and the potential paths forward.
As the new school year begins and COVID-19 cases continue to spread across the United States, questions remain about how to educate students effectively and safely. We sat down with Sarah Cohodes––Associate Professor of Economics and Education at Teachers College, Columbia University––to discuss the school reopening process, the impact of remote learning on students and parents, and the potential paths forward. For more information on Professor Cohodes’ recommendations for reopening K-12 schools, read her column in The Atlantic.
Could you tell us a bit about the focus of your research and work before COVID-19? How has it changed since COVID-19?
My work focuses on policies that help students access high quality education. I can’t help but laugh when thinking about the beginning of the pandemic, when I was so overwhelmed with the enormity of school closures and lack of childcare, I thought that this had nothing to do with the issues I study. I now realize that COVID-19 is clearly connected to my research. What does access to high quality education mean when many typical modes of education are not safe? How do you promote access for the students who need it the most?
I think we're still in emergency mode and research questions are still developing, but I’m particularly interested in how families are adjusting to remote learning. The most important education work right now is probably understanding and describing the experiences of students, families, and teachers.
Can you speak a little bit about how your work and expertise as an education researcher has shaped your understanding of the challenges that schools are facing right now?
My work has shown me the impact that schools can have on students. So much of the “school reopening” conversation is about the risks and costs. We need to discuss the benefits of schools too, to understand how they can best serve students and families. COVID-19 is changing the conversation around the role schools play in society in terms of childcare and our social safety net.
We need to ask ourselves: Why have we been asking schools to play these roles? Is school the best place to do those things? We know that quality education can change the trajectory of young people's lives––we need to consider what must be done to maximize access to these opportunities.
As schools are grappling with reopening and exploring how, and to what extent, they can integrate technology, what lessons have emerged from evidence that can inform how these decisions are made?
We know technology interventions are not necessarily going to save anybody money or reduce the burden placed on teachers. I think there is incredible promise in technology to help us through this moment, but there's no getting around the fact that interaction with technology needs resource-rich investments in teachers to facilitate it.
We don't have the infrastructure for remote learning in large parts of the country. This goes beyond getting a student an iPad or setting up a WiFi hotspot. This moment is highlighting infrastructure that we should have been investing in over the past decade, including school building improvements, a reliable social safety net, and widespread internet access. While technology offers promise, more will need to be done to make that technology work. How can we help students to engage and interact with technology?
As schools face pressure to reopen and some are reopening, can you walk through the impacts that this will have on parents, students, and teachers?
There is, of course, the difficulty that schools as institutions are being asked to act alone, and a pandemic is not something that acts only in one sector. Just as we task schools with taking care of children's food needs and detecting abuse and neglect, we're now asking them to deal with a public health crisis.
It’s clear to me that the quickest path towards reopening school successfully is a nationwide virus suppression effort. In places with high community transmission, that would involve not reopening schools, as well as wider closures and lockdowns. Everywhere, we need more masking, testing, and contact tracing.
If we want schools to open as normal, we should prioritize schools over other activities. In order for those strategies to work, you have to pay people to stay home. All of this requires government action, which we are not seeing.
Regardless, schools are reopening. With social distancing requirements, they won’t be able to come back at full capacity. There are several models being suggested––some combine remote and in-person learning, some focus on young children for in-person instruction––but we also need to think about who needs support the most. Young children definitely need support, but so do students with special needs, English language learners, and students that are homeless or in foster care. The problem with all these plans is that there's no way for everybody to get everything.
As an economist, I think about face-to-face instruction as a resource that we can choose to allocate. I wish we were thinking about how we can allocate this precious resource in the way that it is going to have the most potential benefit with the least potential risk and cost.
A more thoughtful reopening would help to gain the trust of educators, ensuring that they are only being brought back to classrooms when the need is great, and that remote learning would be an option for the teachers and students that need it. A more thoughtful reopening would also give parents more certainty about their situations, which would facilitate work.
A number of other education tools, including tutoring, could also be considered scarce resources. What recommendations do you have for how we can target these resources to reach students who need them most? How can we use them to prevent widening of gaps by income level or racial group?
First off, we have made a choice to make tutoring a scarce resource. It doesn't have to be a scarce resource, as we could invest in ways to make tutoring a valued part of how we cope with and eventually recover from this time.
In my mind, the appropriate strategy to keep gaps from widening would be to make a tremendous effort to suppress the virus now, so that students can return to normal schooling as soon as possible. Then, schools can assess where students are, address their needs, and turn to options like tutoring, longer school days, summer school, vacation academies, credit recovery, or other strategies to help students catch up.
While we focus on virus suppression, we may need to relax our standards and expectations for the moment, especially for younger kids who have time to catch up. On the other hand, we’ve set up our school system in a way that if you struggle with reading by the end of second grade, it’s very hard to succeed.
I also worry a lot about high school students who may have been on the verge of dropping out before the pandemic, and other high school students who are now not returning to school because they are supporting younger siblings or working to help their family. Helping these students with targeted counseling to graduate high school is an imperative now.
Figuring out which children need support isn’t rocket science. If a teacher hasn’t been in touch with a child since March, they need support. If a student needed support in the classroom before schools closed, has special needs, is learning English, or doesn’t know how to read––they need support.
When students are back in school, there will need to be a time of assessment, followed by investment in additional supports for many years to come, if we want students to recover and be resilient. Right now, the biggest focus should be on connecting students with the school system who were not connected in the spring, especially given the high likelihood of remote schooling across the country.
What role do you think education researchers should play as the new school year begins?
We should not be doing any research that is putting additional burden on schools and students at this time, and I don't think that we should put adults in schools that don't need to be there. The question then becomes, what data can we gather and what research can we do in this context?
We know that the pandemic is detrimental to learning—you don't need research to tell you that. But research can inform our understanding of how it’s impacting different students. Looking at data that’s passively collected while students interact with learning tools can tell us about which students need the most help.
Research can also try to identify strategies that help students recover, but again, I don't think those questions are too complex. I think that things that have worked in the past will hopefully work again, even if we have to think about new formats and ways of doing it.
To me, the most important thing that can happen now is just keeping track of what is happening to students––who is going to school, how much school they're getting, and what type of school they're getting. This will help with targeting interventions––which will need to be ongoing for years.
Can you speak a bit about how reopening schools is going to affect parents?
Parents are struggling right now. Some of the families that this is going to affect the most are single parent families, because many single parents have found ways to make child care work within intergenerational family networks. The terrible thing about this particular disease is that by bringing grandma into your bubble, you're putting grandma at risk. If you’re relying on grandma to care for your child and you are an essential worker, there are extremely difficult choices to be made, and our society’s failure to invest in childcare, both before and during the pandemic, limits those options even more.
Something else to consider is how this will affect women. In heterosexual relationships, it is most often women picking up the slack in terms of home production and remote learning. This is also a structural issue since women are paid less than men, making them the partner that’s more likely to take a step back from the workforce if necessary. Once you leave the workforce and time goes by, it's much harder to re-enter the market.
For women who keep their jobs, there’s a worry that child care tasks will need to interrupt or take priority over work. In these instances, who is going to get promoted? Who is going to get a raise? Who is going to be seen as a leader? This is going to scar women's labor market prospects for years to come.
As a mother of a young child, you may have confronted many of the issues that you described. How has this motivated your research and your outlook on the overall educational crisis?
Even outside of the COVID-19 crisis, being a mom made my work both more and less important. It was less important because I had my family and wanted to spend time nurturing it and watching it grow. It also became more important because every single child who is a data point in my research is a child who has somebody who loves them and wants to watch them grow in that same way.
It was hard in April and May when it was just me and my daughter watching PBS Kids all day while I tried to work on emails and classes and my husband went to work as a health care provider. I’m lucky to have a spouse who has prioritized my career, which is rare for women. But this time, my husband’s work was clearly more important than mine, since he was doing COVID testing, and I was trying to run regressions in Stata. So there was no choice when it came to who would do childcare and who would focus on work. And for us, that happened to fall along the typical gender lines you see in the labor market.
We also have the resources to get through this tough time, and don’t have to worry about job loss or access to food. Of course, even in academia, there are structures in place that disadvantage moms and working parents, and I worry about my research slowing down. But I have resources to be resilient in this moment. So many families, and so many moms, are working with fewer resources and fewer options.
Any final thoughts?
My final thought is this: The best way to help kids learn is to get them back into something that looks closer to normal school. The best way to get kids into normal school is to suppress the virus, have an actual federal response, pay people to stay home where needed, and give schools the money they need to transform into safe and enriching spaces for students and teachers. Until those things happen, I don't think we're going to be able to move forward in a coordinated way that helps kids.
Will Dobbie is a J-PAL affiliate and an assistant professor of economics and public affairs at Princeton University. He is a recipient of a 2019 Sloan Research Fellowship, awarded to “early-career scientists of outstanding promise” in recognition of “distinguished performance and a unique potential to make substantial contributions to their field.”
Will Dobbie is a J-PAL affiliated researcher and an assistant professor of economics and public affairs at Princeton University. He is a recipient of a 2019 Sloan Research Fellowship, awarded to “early-career scientists of outstanding promise” in recognition of “distinguished performance and a unique potential to make substantial contributions to their field.”
Congratulations on being awarded a 2019 Sloan Research Fellowship! What does this fellowship mean to you, and what was your reaction when you first learned you were awarded?
My initial reaction was a combination of feeling very honored, lucky, and a little shocked. The award makes me feel like I should work harder to live up to the potential people see in me.
It also gave me a chance to reflect on how incredibly lucky I have been throughout my career with the numerous opportunities I have received. I have had great advisors along the way in graduate school.
At Princeton, my colleagues have also been very giving of their time and energy. My growth as a researcher and an academic due to this setting has been incredible. I would say my real talent has been listening to the terrific advice and guidance I have received during the course of my career.
Taking a step back, how did you first become interested in the field of economics and researching the causes and consequences of poverty in the United States?
Like many people, my interests have evolved over time.
My original interest in economics started when I was living in Kenya as a college student studying abroad. I was struck by the contrast I saw in cities like Nairobi, where you have incredible wealth and poverty at the same time.
It was puzzling on an economic and intellectual level, as I met all these people who were incredibly bright and hardworking, and still experiencing poverty. I wanted to understand the barriers preventing such talented people from living better lives. In graduate school, my eyes were opened to the same puzzles in the United States.
Much of your research focuses on important policy issues, such as racial bias in the criminal justice system, the labor market consequences of bad credit reports, and the long-run effects of charter schools.
What guides your process for selecting a research question, and what interests you about these particular themes?
There are so many challenging and interesting topics in the world, so I try to choose topics in which I am personally invested. This personal investment is important because I like to dedicate myself to research topics for at least 10-15 years at a time.
I also try to choose topics that are policy relevant so that my work can influence policy, or at least add to the broader debate around those policies.
So, I try to ask myself if this research is important to me, and whether it could move the needle on a particular policy issue, and make my decisions based on that.
What have been some of the most meaningful results or outcomes of your work so far?
Three in particular stand out.
The first was my early work on charter schools with Roland Fryer, where we learned that the racial test score gap could be closed with high-quality schooling. That was very impactful for me early on because it helped underscore the possibility that research can meaningfully change the debate on an important issue.
The second was my work on bankruptcy with a number of coauthors, where we learned how the bankruptcy system can function as social insurance and really help individuals climb out of poverty (see the papers here and here).
Before, many saw the bankruptcy system as a way for individuals to take advantage of the system, and believed that individuals who filed for bankruptcy were only in that situation because they were financial failures who would continue to fail in the future. My work has shown that we should really see bankruptcy as a social support—a lot of people benefit from the current bankruptcy system and are able to rebuild their lives after getting a financial fresh start.
The third was my most recent work on the criminal justice system, again with a number of coauthors, where we learned how incredibly harmful the current system can be for families—particularly for minority families (see the papers here and here).
Our research has documented exactly what the costs of the criminal justice system are for these families, which hopefully will help move the needle on policy. I do think that the policy environment is ready for a new pre-trial system, in particular, and it’s been incredibly meaningful to be part of the conversation on how we can do that in a responsible and effective way.
What new projects are you most excited about, and are there any projects you feel are more feasible now that you have this flexible source of funding as a Sloan Research Fellow?
I’m particularly excited about the future of two projects.
The first is with Katie Coffman at the Harvard Business School and Crystal Yang at the Harvard Law School on reducing racial disparities in the bail system by improving bail judges’ decision-making. We are focusing on two aspects of judicial decision-making. The first is consistency, as we want to increase the overall fairness in the pre-trial system. The second is reducing negative stereotypes judges may have, as we want to reduce what we see as unwarranted racial disparities in pre-trial outcomes.
The second project I’m really excited about is with Sara Heller at the University of Michigan and David Deming at the Harvard Kennedy School. We are interested in addressing the problem of high dropout rates in education and job training programs.
For this project we are partnering with Job Corps, a US government program that provides education and job training for youth, to introduce a curriculum based on cognitive behavioral therapy. The curriculum aims to decrease “problem” behavior and improve employability among youth living in high-risk situations. It is based on Sara’s earlier work but adapted to the Job Corps context.
We are designing the project in a way that, if the pilot program is successful, can be rolled out nationwide.
I am excited to move forward with these types of projects in which we are moving beyond the use of existing data and programs—we are actually using research to create new programs and approaches that can help reduce poverty and inequality in the United States.