Chicago Booth Review Podcast How Takeout and Delivery Boost Productivity

Chad Syverson: The full service restaurants, for a while, that was the only way they could sell anything was to do takeout, but through the sheer desperation of the pandemic, they figured out how to do it. Later, people did start returning, but they kept the takeout bit because they had figured out how to do it. And also the customers realized that, "Yeah, I never thought of getting takeout from The Drake before, but I guess they could do it."

Hal Weitzman: For the 30 years leading up to the COVID pandemic, productivity in US restaurants was essentially flat. Since the pandemic, restaurant productivity has soared. Why? Because of the rise of takeout and delivery. Welcome to the Chicago Booth Review Podcast, where we bring you groundbreaking academic research in a clear and straightforward way. I'm Hal Weitzman, and today I'm talking with Chicago Booth's Chad Syverson about his research into the restaurant productivity boom. So how has takeout and delivery transformed the economics of the food industry?

Chad Syverson, welcome back to the Chicago Booth Review Podcast.

Chad Syverson: Thanks. It's good to be here.

Hal Weitzman: We're here to talk about real labor productivity in the restaurant industry. What is restaurant productivity, what does that mean? You're not talking about dining there, you're talking about the people who work in the restaurant.

Chad Syverson: Right. We're talking about how well restaurants perform when they're putting together meals to serve to people basically.

Hal Weitzman: Okay. So real labor productivity means that, it means how many meals can you put out per hour?

Chad Syverson: Yeah, that's a good way to think about it. How we actually measure it is we inflation adjust total spending at restaurants, and then we divide it by the number of workers or the number of worker hours that go into serving that amount of inflation adjusted spending.

Hal Weitzman: Okay. And you document what you call the curious, it sounds like a novel, the curious surge of productivity. It could be a novel.

Chad Syverson: That would be a great mystery, yeah.

Hal Weitzman: The curious surge of productivity in US restaurants. Why is it curious?

Chad Syverson: Because it reflects a pattern that's a big break from what used to be. So if you look at the data, we've got good data on this stuff going back to the early '90s. Basically, productivity as we just described it, the number of meals sold per employee, productivity in restaurants had been absolutely flat for 30-plus years coming into COVID. When COVID hit, there was a sharp drop of about 20% productivity and that's basically because people stopped eating a lot of restaurant meals, and they've let a lot of workers go, but didn't let workers go quite as fast as eating restaurant meals did.

But very quickly as we emerged from the initial wave of the pandemic, productivity rose back up again, but it didn't stop at its old level where it had been for 30-plus years, it kept going. It eventually stopped at a level about 15% higher than it had been in those prior decades and it stayed there since then. So we're coming up on about four years now of productivity in the restaurant industry being about 15% higher than it was for at least 30 years prior to COVID.

Hal Weitzman: Is that a typical pattern? If you get an industry where there's a lot of layoffs and then demand starts to go back up again, they're probably slower to hire people, so does productivity always go up in those circumstances?

Chad Syverson: You can get those kind of blips from just lags in leads in the amount of hiring you do relative to output, but that tends to work itself out pretty quickly as the business cycle does. But we're well past the end of the business cycle portion of the pandemic and productivity-

Hal Weitzman: And we're still way up.

Chad Syverson: ... is staying high, yeah.

Hal Weitzman: Okay. What's interesting about this is how you got this result because you analyzed mobile phone data for 100,000 restaurants. So what kind of data did you get? What information did you get from it? Why is that better than traditional government statistics on this kind of stuff?

Chad Syverson: Well, to be clear, the pattern I just described, the flat for 30 years, then a 15% rise and stay there, that is in the government statistics, so we knew that was going on in the industry overall. What we couldn't figure out from the government statistics was why did productivity go up and that's where we used this phone tracker microdata that you were mentioning.

The way to think about what this data is is it collects the location at a pretty high frequency of a given mobile phone. It doesn't track a person, so we can't look at Chad and walk him walk around Chicago during the course of the day, it doesn't tell you that. What it does tell you is this phone that I happen to be holding was in this establishment for 45 minutes or an hour and 45 minutes, or what have you. And then later that day, it was in this establishment and it was there for 10 minutes. So we know, we basically have a counter on the door of every restaurant, so we know how many people walk into that restaurant and we know categorically at least how long each of those people stayed. We can't any individual over time, but again, we know someone was in there and we know how long they were there for. That's the data we used to measure productivity at the restaurant level to try to figure out why productivity rose.

Hal Weitzman: Okay, and you call that customer dwell time. Is that what we do when we go to McDonald's?

Chad Syverson: Yeah, you're dwelling.

Hal Weitzman: We dwell there.

Chad Syverson: Don't you like to dwell at ... You don't want to dwell too long.

Hal Weitzman: No way.

Chad Syverson: That's why it's fast food.

Hal Weitzman: Definitely.

Chad Syverson: Short dwell times, yeah.

Hal Weitzman: Yeah, you're right. Yeah, yeah, that's true. They don't need to put the no loiter sign because you definitely don't want to dwell there. So there's a reduction, or a strong correlation in the reduction in this customer dwell time and the increase in productivity. Okay, so customers are spending less time in a restaurant. Why does that mean higher productivity?

Chad Syverson: Yeah. So our finding was the restaurants that saw the biggest increase in productivity had the largest drops in average customer dwell time. In other words, they were moving their customers through the restaurant faster than they were before. And if you could do that, if any given customer is moving through more quickly, you can serve more customers per unit time. You do that and that's an increase in productivity because essentially, again, what the overall data shows, they're serving 15% more meals per employee than they used to. We're finding in this microdata that seems to be because they're able to get customers in and out with their meals faster than they used to.

Hal Weitzman: I see. So the shorter the customer dwell time, the higher productivity?

Chad Syverson: Exactly.

Hal Weitzman: Okay. So this is all about takeout and delivery, isn't it? It's all about ... This is the short of Uber Eats, Deliveroo, DoorDash kind of economy.

Chad Syverson: Yes, we think so. And the reason why is because we see dwell time categories. So we know whether you were there less than 10 minutes, whether you were there 10 to 20, 20 to 40, and so on and so forth. One possibility, and this is not what happened, is just people come in, sit down, have their meal, but they push them out the door a little sooner, they're able to increase their turnover that way. But people are still tending to stay, whatever, 30, 60 minutes on average for a meal. That's not what happened.

All the decreases in average dwell time were coming from a lot more customers being there less than 10 minutes. And that suggests to us that this is about takeout and delivery meals. It's either someone coming in to pick up their meal and walking out pretty quickly, or a driver coming in to pick out an ordered meal that they're then going to deliver to the person who ordered it.

Hal Weitzman: And that makes a lot of sense. So if we think back to the pandemic, even the fanciest restaurants in the world started doing-

Chad Syverson: Takeout.

Hal Weitzman: Right.

Chad Syverson: Yes.

Hal Weitzman: You go to The Drake Hotel, whatever, you get takeout tea. When I was out, they would send you little cakes and whatever on a-

Chad Syverson: The little sandwiches-

Hal Weitzman: ... lazy Susan or whatever.

Chad Syverson: ... in the little takeout boxes, yes.

Hal Weitzman: Right, exactly. So everyone started doing it. The question I would have is, I don't go to tea at The Drake, we should do an episode there. That would be a good use of Chicago Booth's money. I'm guessing that if I were The Drake, I want people back in there. I don't want people doing takeout. Everything people were saying during the pandemic when we had behavioral scientists and videos with Chicago Booth Review, people would say we would go back to normal. So we're now at normal, but normal is different. We changed our behavior because of the pandemic it sounds like.

Chad Syverson: Yeah. And I think in some ways, the restaurants were able to have their cake and eat it, too, no pun intended as I say that. I think the full service restaurants, for a while, that was the only way they could sell anything was to do takeout. Now, they had maybe never done takeout before, but through the sheer desperation of the pandemic, they figured out how to do it. Later, as the pandemic receded, they're able to have people come in, people did start returning, but they kept the takeout bit because they had figured out how to do it. And also, people, the customers realized that, "Yeah, I never thought of getting takeout from The Drake before, but I guess they could do it and I kind of liked it." So maybe four out of five times, I'm going to go to The Drake in-person, but for that fifth time I'm going to do takeout.

So you have people choosing to consume meals, and maybe consume more restaurant meals than they ever did before, it's just some of them are now delivered in a different form than they were before COVID. So with the full service restaurants, it was most of them were going from no takeout and delivery to some takeout and delivery. Whereas the limited service restaurants probably were doing those things, but they just got better at doing those things because of COVID, and many of them have actually reconfigured their operations to be more takeout and delivery heavy.

Hal Weitzman: So all this growth in productivity, or much of this growth in productivity is coming at the middle to lower end of the price market, is that fair?

Chad Syverson: Actually, I wouldn't say that because while the microdata we look at, for reasons we don't need to go into, are better for limited service restaurants, that's why we focus that phone tracker stuff there. If you just split the industry into those two components, they both exhibit the exact same pattern of flat and then about a 15% increase, whether it's full service or limited service restaurants.

Hal Weitzman: If you're enjoying this podcast, there's another University of Chicago Podcast Network show that you should check out. It's called The Pie. Economists are always talking about the pie, how it grows and shrinks, how it's sliced, and who gets the biggest share. Join host Tess Vigeland as she talks with leading economists about their cutting-edge research and key events of the day. Hear how the economic pie is at the heart of issues like the aftermath of a global pandemic, jobs, energy policy, and much more.

Chad Syverson, on the first half we talked about your research into the rise of restaurant productivity and how it's all about takeout and delivery, and even the fanciest places started and still haven't completely stopped the ability to go and get takeout. So something that's interesting is you note in this research that takeout and delivery can lead to what you call a substitution of home production for restaurant labor. Is that like me microwaving my Dominoes pizza? Does that mean we're all contributing to the boost in restaurant productivity?

Chad Syverson: It means it's a little tricky to try to do an apples-to-apples comparison of what a meal is before and after COVID, so this is something we had to wrestle with in the paper. It's true, when you think about the full service situation. When you have an in-house dining experience, you go there, someone comes and takes your order. Then they come bring you your food, then they take your dishes away, and someone in the back washes your dishes for you.

Now if you, from that very same restaurant, get the same food, you're going to take that home, it's going to be eaten on your dishes which you're going to clean, you're doing all the carrying and busing. So in some sense, you are doing part of the work the restaurant used to do. And that's kind of you have to think about, okay, is that really the same product, or how is that a different product and should we call that productivity growth or just something different? We talk about that in the paper.

In think in the end, we came down, it is true productivity growth, but it's a little muddled. I think it's less muddled in the limited service restaurant-

Hal Weitzman: What do you mean it's true productivity growth? What is true?

Chad Syverson: Well, a true productivity growth is you're delivering exactly the same thing, the same experience-

Hal Weitzman: Rather than transferring the productivity to me?

Chad Syverson: Yes, rather than having you do some of the work and not counting that-

Hal Weitzman: Here's a frozen pizza, heat it up. Right.

Chad Syverson: In principle, we should count the hours that you're spending at home eating this takeout meal in the hours that restaurants-

Hal Weitzman: In other words, it might not be an increase in productivity, it might just be a distribution?

Chad Syverson: Yes, exactly. So again, we think through that in the paper. In the end, we think nevertheless, it's still real productivity growth. I think this is just less of an issue in the limited service context anyway because most of the time, all that work which I just described, someone coming and taking your order, bringing it, that was never happening. Now it's just who's doing the travel. It used to be that you had to drive to the restaurant to get the meal, now someone might be driving to you, but it's not free for you to go to a restaurant. That was a cost of getting a meal that we didn't count before, now it is an extra cost. On the other hand, we're not counting the cost of the driver, we're just basically trading who's driving for what, so maybe that just washes out in the limited service context and it's a little more apples-to-apples in that case.

Hal Weitzman: Okay. So the one thing I'm thinking about when you're talking is a lot of productivity in normal manufacturing let's say, which I know you write about a lot, is driven by automation. There has been some automation, right? You go to McDonald's, many McDonald's you just order on a screen, there's not a person. Is that helping?

Chad Syverson: There's been a little bit, but probably not that much. Yeah, it's-

Hal Weitzman: Well, there's Starbucks where you have to order online, and then you have to go and pickup.

Chad Syverson: I think a lot of the automation has been the ordering systems. It used to be it spoke an order to someone and they typed it in, now you're typing it in yourself and there's a little bit of, again, trading who's doing the work here. There has been a little bit ... We did not see actually a big trend break in how capital intensive the industry is, so it's not like all of a sudden they've installed a bunch of electronics that they never had before and that's creating this huge increases. And it certainly didn't have this spike, step function that the productivity did.

Hal Weitzman: But it's a change and it's a change that's not going away.

Chad Syverson: It is a change. I agree. I think where you mostly see the physical changes are in the configuration of the restaurant. So I like to talk about my neighborhood McDonald's, which used to look like a lot of McDonald's. It had a counter with six to eight cash registers, you would stand in line and go up to one of the people standing behind one of those cash registers. Well, now there's a little peninsula with one register and it's unusual to go in and make an order live. What you've probably done is you've made an order online before and they've replaced that long counter with basically shelving for takeout orders. So they have reconfigured the physical space in a way that's made them more efficient for this increase in takeout and delivery, and less of a come in and order.

Hal Weitzman: Yeah, that's right. I wonder if you compared it to Home Deport or wherever, somewhere else that gets a lot of online orders, whether you would find that they've also become a lot more productive.

Chad Syverson: Yeah, that's an interesting question that we speculate a little bit about in the paper. Well, we just speculate it'd be interesting to look at, we don't actually have any leads yet.

Hal Weitzman: Okay. So we've talked a lot about, or we've implied this is a lot about fast food restaurants because they're quick in, quick out. But do you think it's generally when you say the restaurant industry, how much of the increase has been at the McDonald's type end and how much at The Drake type end?

Chad Syverson: It looks like it's happening at both ends. So the microdata analysis, we focus on limited service restaurants because the data's just better in that case. But if you split up the nationwide aggregates that come from government statistics which do allow you to separate into full service versus limited service restaurants, you see the exact same productivity pattern of no change for 30 years, a quick drop, and then a recovery that goes to a level about 15% higher than its prior level and it's stayed there for the last five years. And that's true in both limited service and full service restaurants.

Hal Weitzman: Okay. Are there some chains or sectors that have had the bigger boost in productivity?

Chad Syverson: Yeah. So we talk about that a little bit in the paper. There's some variation in how big the productivity growth was across different chains. We take the largest chains in the data and break it out one-by-one. I think the most interesting pattern there, and this gets back to what we were talking about with takeout and delivery, is the chains that saw the least increase in productivity were the pizza places. And that's because basically that's all they were doing before.

Hal Weitzman: They were already doing it.

Chad Syverson: They were already doing takeout and delivery before, so they didn't have much change to experience, and therefore the productivity growth that came with it. It was really seems to be focused on the restaurants that had the biggest change in takeout and delivery, and the con commit, and drop in dwell time.

Hal Weitzman: Okay. So it really could be The Drake, if you're starting from nothing and you're doing a little bit, that could be bigger than Dominoes or whatever.

Chad Syverson: I think that's exactly right.

Hal Weitzman: Okay. You advertised a second ago, at the end of this paper you're thinking about other industries that might have experienced similar productivity bursts. What were you thinking of there?

Chad Syverson: Well, we were thinking about maybe some other retail industries have shifted more from the pure bricks to bricks-and-clicks, grab-and-go type operation. So we've started to look into that a little bit, we haven't gotten far enough to announce any findings yet.

Hal Weitzman: But are you thinking about the same pattern, in other words, what happened during the pandemic, or do you just mean in general?

Chad Syverson: Well-

Hal Weitzman: Is it the online ordering phenomenon we talked about?

Chad Syverson: Right. In specific, we were speculating about whether changes in the way business was done and maybe the business had to be done because of the pandemic might have led to some different ways of doing things that have stuck since and that's where our curiosity is.

Hal Weitzman: Certainly, if you think about clothing, the clothing, there's so much clothing that has no physical stores at all now.

Chad Syverson: That's true, too. Now, that's a measurement issue when you've got sector delivering something that you used to go get yourself. That goes back to this are you just swapping home production for in-store production. That's even trickier to handling in clothing than it is in restaurants, but that would be something you'd want to think about when looking at it.

Hal Weitzman: Okay. Chad Syverson, this is fascinating. Love talking about food with you.

Chad Syverson: I love eating food, too.

Hal Weitzman: We should have it cooking and writing, because it's the curious surge of productivity. Thank you very much for coming.

Chad Syverson: Sounds good.

Hal Weitzman: Thanks very much for coming back on the Chicago Booth Review Podcast.

Chad Syverson: It was my pleasure. You're welcome.

Hal Weitzman: That's it for this episode of the Chicago Booth Review Podcast, part of the University of Chicago Podcast Network. For more research, analysis, and insights, visit our website at chicagobooth.edu/review. When you're there, sign up for our weekly newsletter so you never miss the latest in business-focused academic research. This episode was produced by Josh Stunkel. If you enjoyed it, please subscribe, and please do leave us a five-star review. Until next time, I'm Hal Weitzman. Thanks for listening.