Governments That Get Results

Policymakers spend enormous energy debating what government should do. But who’s paying attention to whether it actually gets done? Gloria Gong runs Harvard’s Government Performance Lab, where she’s spent 11 years sending teams into more than 100 state and local agencies to work on that exact question. In this episode of “After Hours with Jamie Rubin,” Gong explains why government defaults to process over outcomes, how a Detroit violence intervention program cracked the code on results-driven contracting and why her lab mostly skips New York City — which she says has the best government talent bench in the country but makes a terrible model for everywhere else. She also tells the story of how her husband’s carefully maintained list of her passions saved her from a career in corporate law.

As always, you can listen to the episode on Apple Podcasts or Spotify.​​​​​​​​​​​​​​​​

════════════════════════════════════════

Jamie: Gloria, thanks for being here.

Gloria: Happy to be here.

Jamie: Why don't we start by you describing what you do for a living?

Gloria: Absolutely. I run a lab at the Harvard Kennedy School called the Government Performance Lab. We have an unusual model for an academic setting. We focus specifically on questions around design and implementation of social services, not just the policy, in part because the thing that we noticed when we started this is that there was so much work that goes into that initial question on what should government do. But a lot of times people are asking that in this complete ivory tower way of, in a perfect world, what should government do? And no one seemed to be noticing that those things didn't actually happen once all the policymakers walked away and handed the initiative to some agency to actually implement.

So we really focus on the questions around when you hand something to an agency and ask it to get it done, what actually happens in that black box and how can you make it happen better so that services actually get delivered for people? And so we do this by actually sending people out to partner with state and local governments to help them design and test innovative new fixes to their services or their operations, to actually see whether we can improve service delivery.

Jamie: And so you do it on a consulting basis?

Gloria: Harvard is absolutely not allowed to use the word consulting, but we do a thing that might look similar to people who have seen consultants, which is that we send really talented, early career staff out to sit with government side by side with their staff and actually help them figure out what to do, how to do it, how do you actually do the design?

Jamie: What's the scope and the spec for doing this? Do they come to you and say, look, we want to solve homelessness in Des Moines or something? You then take that away and come back with a plan and then go help them implement? Or is it they have the plan and you help them? What's the model?

Gloria: What we typically do is we'll put out a call for applications on a particular topic that we've heard from a bunch of governments represents a common concern. So we did this recently around places that were trying to improve their coordinated entry systems for homelessness, places that are trying to improve their placements into permanent supportive housing. We just ran a big cohort of governments that are trying to set up alternative emergency response programs, and we're very oversubscribed, so we might get like 70 applications for five slots in a cohort or something like that. But then we'll take that group of governments and work with them intensively, maybe over one to one and a half years to actually figure out what does it take to do this intervention and what do we learn from it? And then we'll try to take the learnings from that and hand it, spread it to the other governments who weren't able to get the direct assistance.

Jamie: Got it. Just through education and writing and all that kind of stuff.

Gloria: We've run custom executive education courses at Harvard, but we also do just a lot of very practitioner-focused creation of materials and tools. And we also run these communities of practice. So for example, on alternative emergency response, we've done cohorts of like 15 to 16 governments that were all working on setting those programs up. But then we also created this community of practice that has like 450 government leaders from other cities who aren't necessarily in the cohort, but we're sharing out the information that we learn with them as quickly as we can.

Jamie: Having been in government, it sounds like the kind of thing that you always hope somebody will do. I worked for the governor for five years. I don't remember ever getting plugged into something like this.

Gloria: Our big insight, which is not going to sound interesting to anyone who knows anything about government, but it turned out to be exciting in the academy, is that there's just almost no support for anyone past that first level of executive leadership. And it's all the folks who are doing things. So everyone who has a question of how will we create these, expand these pre-K seats, how will we send out these new teams? There's nothing for them. And so that's what we're trying to focus in on: those doers who have very little support from external places.

Jamie: So it's what people who have not been in government call the gigantic bureaucracy. Basically, as you said, government is starved for resources. There is no level of government that has the resources it needs. Period.

Gloria: The way we describe what we do when we talk to mayors or governors about coming in is that we say, look, we really deeply are not on that team that thinks government's incompetent. You have a bunch of staff who are very capable. They can do really, really good work. They are probably maxed out keeping the trains running. These are big, hard things to run. It's very hard to ask the same people to design new train tracks, and that capacity to think about what's a system that we should shift to a lot of times also needs to be protected. They need to be firewalled from the emergencies in the day to day. So that's often the role that we're trying to play for folks, is to be able to come in and say, let us be a team that's firewalled a little bit from the daily emergencies that can help you try to figure out how are we going to go from the system we have today to the system we want to get to.

Jamie: And how did you get to this place? Were you in government at one point or were you just a junkie?

Gloria: Just a junkie.

Jamie: Okay, great. So how did you get to this place?

Gloria: So I was actually in law school. I had a clerkship and an offer from a law firm. What they say to everyone is, well, you have something you want to do, but what you should do first is you should clerk for one to two prestigious judges. Then you should go work for five to six years at a law firm. And at that point you should start doing the thing that you want to do.

Jamie: I just want to pause for a second. I don't know who's still telling people that kind of stuff, but we need to figure out who that is because that's extraordinarily bad advice. It wasn't good advice when I was in law school and it's not good advice now.

Gloria: I went to the career services folks my first year and I was like, I'm really interested in public service. And they're like, oh, you know what you should absolutely use your summers to work at corporate law firms to make sure that isn't what you want to do. And I was like, well—

Jamie: Such bad advice.

Gloria: I think I just said it's not what I want to do.

Jamie: It's like no other area of your life would somebody tell you that. Once you know what you don't want to do and what you want to do, what you should do first is do more of what you know you don't want to do, just to make sure.

Gloria: So the funny thing about this is this specific advice, which was given to me by so many people, and also taken by so many people in my cohort. They have some stat that they tell you, which is like 90% of you came in saying that you want to do something in public interest, but 90% of you will exit to the top corporate law firms. And the thing that was funny for me about it now sitting where I am, is that a bunch of people took that advice because the thing that was sitting behind it was people telling them, oh, you'll be able to lateral from that to anything you want to do. And now that I'm sitting on the side of hiring, I'm like, that is not true. Five years at a corporate law firm does not prepare you to run an agency or do many of these other things.

Jamie: You're definitely right. I mean, it doesn't prepare you to run anything really.

Gloria: It might, but I think everyone just got told you will definitely be able to go from this into anything and nobody said, yeah, you will have spent six years doing something that you at least said at the beginning was not what you wanted to do. That's very expensive.

Jamie: And honing a bunch of skills that probably are not going to translate necessarily unless what you wanted to do is be a government lawyer or something like this.

Gloria: That just never was said to me.

Jamie: Yeah. Okay, got it. All right. So you dodged the bullet or what happened?

Gloria: Yeah, so the short version of the story is I have a little bit of contrary instinct, so I'd been working internationally previously on human rights China stuff, and I was like, I think I want to work on U.S. domestic things. And so everyone was like, great: White House, DOJ. That's where our connections are. And I was like, oh, well what about state and local? And everyone I talked to was just like, I don't think that's a thing, or we don't know anyone there. So that made me be like, okay, I want to go that direction. When people would tell me, we don't know anything about that, I was like, well, great. That means there's more of a need there. So I started looking into state and local, no connections out of YLS—

Jamie: Yale Law School.

Gloria: Yeah. For years.

Jamie: Well, by the way, not 100% true. So for example, I regularly tell the people in the career services office that if anybody comes through wanting to talk about state and local government, they should absolutely send them to me, and I've been doing that for 10 years.

Gloria: There you go. So I was like, I'm going to go this other direction. Then the other thing that basically happened is my husband saved me. So I had signed up to do this clerkship law firm thing and he was like, hey, just tell me really quickly about what you're excited about about this. And I said whatever you gear yourself up to say in the interview. So I was like, something, mergers, acquisitions. And he's like, okay, that's awesome. Well, because I care about you so much, I keep a list of the things you're passionate about so that I can support you and whatever your dreams are. So I'm just going to open it up real quick and write mergers and acquisitions right here under women and children and intergenerational poverty. And I was like, yeah, yeah, yeah. Okay, I got it. I'm not okay.

Jamie: So parenthetically, your husband sounds like—I mean, I'm embarrassed that I don't think I've ever really done that for Gretchen, but what does he do just out of curiosity?

Gloria: He's an awesome person. He's amazing. He does strategy stuff.

Jamie: He is a saint. Say he is a partner to love.

Gloria: Get out of here. All right, fine.

Jamie: Anyway, the long and short of this is I asked my judge if I could have a hot second to look at other things, which I did not realize at the time—massive. Very—

Gloria: You're not supposed to do that. You're not supposed to do that. He was incredibly generous. He was like, yes, please feel free to look around. I found Jeff Liebman in this moment where he'd come back from the White House and I was like, hey, I want to start the state and local thing. We got coffee and by the end of it he was like, hey, why don't you come and just start this with me right now? Don't take the bar, don't clerk. Just let's just do the thing we want to do right now. So that's what I've been doing.

Jamie: Great. Oh, so you've been doing it really since graduation?

Gloria: Yeah.

Jamie: Okay. Having done this now for 11 years, I guess, and having seen all of the stuff you've seen and having advised governments, there's a very obvious question, which is when do you decide that you need to go put it into practice?

Gloria: Someone asked me the other day if I was keeping a list of all the people I'd want to work for.

Jamie: Oh, that's an interesting question.

Gloria: It was a great question. I don't know the answer to it, but it did make me think there are some people who are really inspiring. I think I want to work for a team and that's why I've been where I am. Where I've been so long, my team is amazing.

Jamie: Yeah. How many states and locals have you worked with?

Gloria: Over 100.

Jamie: Yeah. A hundred. So I mean, you must have seen every flavor of competence and not competence.

Gloria: Yes.

Jamie: Okay, got it. I mean, all the other things aside, do you at least feel like you could do this? You could get into the game directly?

Gloria: I mean, of course. That's what I'm saying. If you see a team that you want to support, you're like, I'd love to support that person, go ahead and put me in.

Jamie: So you've been there for 11 years. Give me one of your favorite teams to work with, and we'll stipulate that you love everybody equally, but if you had to go back in time and start again, you would start with?

Gloria: Lisa Morrison Butler when she was running the Chicago Department of Family and Social Services just ran a great team. They had great strategy people. They really were trying to figure out how can we realign all of our spending and contracting around clear outcomes. We're working with Governor Wes Moore's senior cabinet right now. They have just pulled in a ton of very competent people. They've been dealt a tough hand, just in terms of the things happening in their state, but they're really trying to push on his vision around child poverty and how to help families in the state.

Jamie: Oh, interesting. So he's given them a mission generally.

Gloria: He announced that he was going to eliminate child poverty. And he gives these funny talks where he's like, every person on his staff is like, don't say eliminate—

Jamie: Don't say it. Don't do it.

Gloria: Exactly. And then of course, what always happens, you come in, there's the bridge collapse, all these things happen. But he's pulled together a really strong group of people who are doing really good work on how do we move our infrastructure forward in terms of AI, digital, the actual core capabilities of the state so that we can layer on top questions like how can we get people benefits faster and quicker? How can we serve more families who are at risk of falling into poverty and those kinds of things?

Jamie: So one from Chicago, one from the state of Maryland. Okay. Have you worked in New York? I'm just curious.

Gloria: We've done a little bit here. We tend not to work in New York for two reasons. One is New York actually has an unbelievable talent bench. From what I've seen, the folks in government in New York are the top talent bench in government in the U.S. out of some of the top federal roles. You guys have a great talent bench here. And so we tend to think, don't take water to the sea. You don't need extra people necessarily. The second reason is pilot projects done in New York tend to not translate into scalable solutions elsewhere, because literally every city in the country would be like, hmm—

Jamie: Yes.

Gloria: That means nothing to us. We have none of those resources, none of that infrastructure. And we are a tiny fraction of the size. You do a project in New York if you want to have direct impact, because it's such a huge direct impact in New York is phenomenal because it's just so big. But if you're trying to say, can we do it here because it can be a model for what's done elsewhere, it's just often not the case.

Jamie: I think it's interesting. I mean, there are institutions in New York, so for example, the Furman Center, which is at NYU, the housing center run by people who I know well, all of whom have been in government in the housing world. They're spectacular. When I became the board chair of NYCHA a couple years ago, I thought, okay, great. I know a lot of people. I will be the bridge to NYCHA for lots of people who should be helping and never have. Great. So I went and sat down with my friends at Furman and I explained what the strategy was going forward, which made a lot of sense. And they were like, that's awesome. And I said, I'm sure you guys have been helping NYCHA along the way because you have all these resources and you have analytics and you're thinking about things to do. Blank face. And I said, how is it possible that you've got a third of all the public housing in the country within two feet of you in the five boroughs? And they said the same kind of the same thing you did, which is it's so big, it's kind of not relevant to anybody else. And so if we want to be relevant to everybody else and be sort of scalable, we have to go elsewhere. And also, it's not that NYCHA doesn't need help, but they really do have a lot of resources that other places don't have.

Gloria: We try to do these knowledge exchange things between different cities. And the trick if you bring New York in is they'll be like, okay, well the first thing you should do is have your internal analytics team. And everyone's like—

Jamie: Okay. And that would be who?

Gloria: Exactly. They're like, take your 20-person data team.

Jamie: It's incredible. Yeah. Now it's funny because you say that because if you—I mean just taking NYCHA or really, but this is true of any agency that I've had any interaction with. They will all tell you they're understaffed and I can tell you for sure NYCHA is understaffed. And I don't think you'd find another agency that would say different. We have 600,000 residents. We have 12,000 staff, and most of them are in the field, and that's just, you wouldn't, if you were running a private company that housed that number of people, you would have four or five times the number of people in the field. But for whatever reason, we do have a breadth of functionality. So to take your example, we do have an analytics department, but we also have other needs. It's just we can't hire—if we could hire five more people in the field and give up the analytics, that wouldn't really make any sense.

Gloria: Exactly.

Jamie: So it's just very, very challenging. So if you had to say, I don't know if you've done this, I've read some of the other stuff you wrote—spoiler alert, you and my mom—

Gloria: And my mom.

Jamie: And I didn't see anything in particular that was like, here's everything I've learned from a high level over my time here. If you were writing a book, you would say, okay, well what are the five magic keys to running a good government or something like that. Have you ever thought about it that way?

Gloria: I love that question. I can give you my version of it right now. And the funny thing is, it's going to be the same thing I'm saying in every piece. I only have two ideas, but one of them is we don't see people focus on outcomes enough. There's so much focus on process. And that's such an easy thing to say. It's a really hard thing to do. You just default immediately to being like, well, we had the meetings. We did the town hall. And the ability of a leader to say, we are trying to get something done. We're just going to measure if that gets done and work backwards from getting that done. Not to say, well, if we do a lot of meetings, that counts.

Jamie: I think that's a great—again, I co-sign that. I think I've never—it doesn't happen in government for the most part. Nobody in government gets rewarded for outcomes. Do you think that's why it is or is there something else going on?

Gloria: I think that's partly why you do process things because it's a risk mitigation thing.

Jamie: Mm-hmm.

Gloria: I just wrote a case about this thing that Mike Duggan did when he was mayor in Detroit, where they shifted their community violence intervention funding to be outcomes-focused on did you drop shootings in your zone. I think it's an incredibly innovative approach to a type of funding that people are doing all over the country but isn't really being connected to accountability around outcomes. Really innovative design. They let the providers design their intervention. So it was much less micromanaged.

Jamie: So this is for nonprofits?

Gloria: Mm-hmm. These tiny nonprofits.

Jamie: Yeah.

Gloria: Community violence interrupting groups. Community violence interrupters are super proximate to their communities, from the communities. They base it around outcomes. I taught this teaching case around it that I'm hoping we can teach to government leaders in the different programs we do at the Kennedy School. And I was teaching it to this class of students and I break them up into sections where I'm like, okay, you guys think from the mayor's team perspective, you guys think from procurement and legal, you guys think from the provider's perspective, and we use it to tease out what's hard about this. And the kids on the mayor's team—one kid raised his hand, he's like, sorry, let me just see if I understand this. I have all this money from ARPA and I could do two things. I could set up a thing that's called Office of Violence Prevention and have a ribbon cutting, and then let them just go spend the next three years setting up an office. And there is zero downside to me. And I get to say I started an office that has violence prevention in the name, or I could say to the community, I'm going to put this money towards bringing shootings down and we're going to measure it and put it out in dashboards.

Jamie: Yes.

Gloria: And he was like, that sounds like it's only political downside. So much risk. Why? Where do you get this kid?

Jamie: Well, but it's exactly to your point, right? Everyone we see who's doing outcomes-based work is taking a risk, because they really are trying to get the thing done. There's so much more safety around doing process, where you can say, yeah, I named a person and we did a task force, and then we have a committee. And the people who are like, look, I really want to push around this are doing it because they're actually trying to get it done and they care, but not because all the political rewards are set up that way.

Gloria: I hope you weeded them out quickly before they get into government.

Jamie: That was the core insight that they were pointing out, was like, hey, this is a hard thing. This is a hard thing for a leader to do.

Gloria: Yeah, it is.

Jamie: Actually, stay at Harvard kid. You're doing okay. Yeah, you go. Sorry. Look, that is, it's probably true because the one thing that politicians in general, elected officials are really good at, more or less, is figuring out what it is that's going to get them reelected.

Gloria: I do think there's a separate set of reasons we don't focus on outcomes, which is—and this is a lot of the stuff that we focus on—it's actually not what the structure is right now. So you say you're trying to do something as an elected official. People get together and do this policy work or whatever. You then hand it off to an agency to implement and then it goes through a set of standard processes. The agencies, where typically—unless it's a thing that the agency has direct control over, which is the minority of things in government—you're going to send it out through procurement and a contracted provider is going to do it. Typically, a nonprofit. That process is almost never oriented around outcomes.

Jamie: Yes.

Gloria: So the contracts are not, we want kids in this community to have better reading scores and ultimately better economic mobility. It's you will create X number of seats and you'll invoice for those seats, and then all of the focus is around the compliance around that process. And so one of the issues that we focus on a lot is the translation between what you're trying to do and the way that it gets done. One of the main things that gets dropped out is what the thing was that you actually were aiming for. And you replace it with things where you're like, we will pay you to deliver showers at this shelter.

Jamie: Yep. We're already down a rabbit hole and we didn't even get to your two, three, and four or five.

Gloria: Same. Outcomes is pretty good.

Jamie: But this one is interesting. I'm going to guess that you spent time around the performance bond, social impact bond world.

Gloria: Yep. That's how we got started.

Jamie: My prior is I don't believe in that stuff. Tell me what you think about it.

Gloria: Social impact bonds?

Jamie: Yeah.

Gloria: So my short take on social impact bonds—the GPL started out as the Social Impact Bond Technical Assistance Lab. My short takeaway on it was the third-party investor stuff ended up being too complicated and government can borrow at a lower rate. So also not necessary, as in government should be more risk-taking with the money it has. It doesn't need to pay investors to offset that risk. And the transaction costs were incredibly high because it was these very sophisticated financial transactions that the governments weren't set up to do.

Jamie: Yep.

Gloria: So we essentially were, after we tested it in a bunch of places, we were like, we don't need that piece. And governments came to us and said, they're like, actually, we can pay for stuff. The piece that they really liked that we have kept, and it's the backbone of the work that we do now, is the stuff around measuring outcomes. Get your contracts organized right and then do data-driven performance management around the thing that you're trying to deliver to actually figure out what is stopping this from happening and problem-solving in a really agile, fast way. So that piece—

Jamie: Yeah.

Gloria: Of social impact bonds. I actually think it was incredibly successful.

Jamie: That makes sense to me. I had not followed the space really. We had some experience with it in New York State. We actually did the first, I think the first social services one we did was a recidivism one. Yeah. We did the CEO, right?

Gloria: Yeah, yeah, yeah. Exactly.

Jamie: So which did not work well.

Gloria: Worked in the sense that you now have evidence.

Jamie: We can talk about this some other time.

Gloria: No, no. But actually this is part of it. Government should be finding out what works.

Jamie: Totally agree. It just goes back to that—

Gloria: Right conversation we were just having around whether people want that or—

Jamie: It's great. It was great for us. But having said that, the missing piece there, which maybe doesn't matter, is the incentive structure, right or no?

Gloria: There is a thing that I think is really tricky, which is who is going to buffer the risk for the provider if they don't get their outcomes payments? But what we learned in subsequent projects, like the original social impact bonds put a lot of the payments at risk. What we've learned in subsequent projects is you don't have to have as much of it at risk to still incent some of the same behaviors. And you can still drive the same kind of adaptation. So this project in Detroit that I was talking about, I think is a really nice example of this. They took their ARPA funding. They made these super generous operational budgets for the social service providers who got selected to do the community violence intervention. They then made also very generous incentive payments on top of it if they hit target. So it was—

Jamie: Sort of a bonus. So you got the original payment—

Gloria: But then it was a bonus. Yes, but it was a bonus. So the operational funds, and I think that's the right way to structure—what I would think was the next generation of results-driven contracts, which is you can't be putting the providers' operational funding at risk so that they're like, well, at the end of the year we might find out we're not going to make any money from this year. But if you cover their operational costs, then using performance incentives as bonuses, I think is really effective.

Jamie: Are you seeing any of that in practice?

Gloria: This Detroit project's my favorite one that I'm seeing right now. And I heard the mayor talk about it and I was like, this is so smart. I called my friend who was in the mayor's office and I was like, hey, this is a really, really smart program design. I want to write a case about it. And she's like, oh, Gloria, that was actually my project and she had worked at the GPL, so I was like, oh, darn it. We're not seeing any independently seeded version.

Jamie: Oh, that's too bad. When was this?

Gloria: It's now. This was now.

Jamie: Okay. So all right, well, so it'll get out there. All right, so here's—we intended to talk about the article that you wrote for Vital City, and instead we ended up talking about the work that you do that led to that article and lots of other stuff. So I think we're just going to presume on your time and say, this is going to be part one of the conversation. And part two of the conversation is going to be about that article because that's free advice to the new administration, drawn on lots of other stuff. And if they're lucky, they'll find a way to drag you into helping. But one way or the other, it's very closely related. Okay. Great. Gloria, thanks a lot.

Gloria: Yeah, thanks.