Russ Roberts

Canice Prendergast on How Prices Can Improve a Food Fight (and Help the Poor)

EconTalk Episode with Canice Prendergast
Hosted by Russ Roberts
PRINT
You're Not the Boss of Me (Are... Peanut Butter, Pasta, and Rela...

If you have 250 million tons of food to give away every year to local food banks how should you do it? Canice Prendergast of the University of Chicago's Booth School of Business talks with EconTalk host Russ Roberts about how he and a team of economists created an artificial currency and a daily auction for the national food bank Feeding America so that local food banks could bid on the types of food that were the most valuable to them. Prendergast explains the results of the new system and the cultural and practical challenges of bringing prices, even artificial ones, to a world accustomed to giving things away.

Size:28.7 MB
Right-click or Option-click, and select "Save Link/Target As MP3.

Readings and Links related to this podcast episode

Related Readings
HIDE READINGS
About this week's guest: About ideas and people mentioned in this podcast episode:

Highlights

Time
Podcast Episode Highlights
HIDE HIGHLIGHTS
0:33Intro. [Recording date: November 17, 2015.] Russ: Our topic for today is a bit unusual. You were part of a team that introduced market forces to how food is distributed from a national food bank to local food banks. So, what was the problem to begin with that got the team involved? Guest: So, the not-for-profit organization that asked us to see if we could help is called Feeding America, which is a very large not-for-profit based here in Chicago. And part of their mandate, I should say, is to receive foods, typically from distributors or manufacturers--think Kraft, Walmart, whatever--and their job is to allocate a large amount of food to regional food banks all around the country. This is a pretty complicated animal. The food that we got involved with is about a quarter of a billion pounds of food. So, think of the typical example of being a Tyson plant in Kansas says, 'We've got an extra truckload of frozen chicken.' What do you do with it? So, it's a national organization; they have to decide who gets a share and what the algorithm is to decide how that should be assigned. And the way they tended to do it was the following: which is, they have reasonably good measures of poverty levels across different parts of the country. So, what you can do is rank different regional food banks in terms of their need. And what they do is, they have a big, long list, and essentially if you are at the top of the list, you get offered the first thing. And if you say yes--often if you say no, what happens is you get put to the bottom of the list [?] you get your thing. So, essentially it's an algorithm that's based on their perception of what an individual food bank needs. So, they called us and said, 'Look, there's some issues here.' And the issues largely were on the demand side, which is-- Russ: Before we get to those, can we just work a little bit on how the old system works? Guest: Sure. Russ: So, you said, they use the poverty rates in different areas to determine--to determine what? Guest: To determine need. Russ: So, if I'm at the top of the list now, it's my turn. Is the question then how much I get of what's at the top of the list? Let's just say I'm at the top of the list; I'm a Washington, D.C. food bank. And what is on the top in terms of product? Guest: Okay. So the way it works is that on any given day, there's probably across the country let's say 50 truckloads of food. So a truckload of food is probably, say, 30,000 pounds of food. So, what happens is, Feeding America has 50 truckloads of food. And based on everything that's happened up to now, you at Washington, D.C. are at the top of the list. What they will do is, they will offer you one of those truckloads. Russ: And if I were a different place I might get a half a truckload? Guest: No. Truckloads basically--there is an indivisibility, which is, you need a truck to bring something there. At times what will happen is two small food banks will get together and share a load. But for the most part you will get a truckload of food, which then goes to your distribution center; and you then assign it to your local church, your local community center, or whatever. Russ: So that's--did you say 30,000 pounds? Guest: Yeah, that would be the typical truckload. Russ: So, that sounds like a lot of food. So, I get 15--I'm going to call it 'tons' cause it sounds more fun. It's 15 tons, right? Did I do that right? Guest: Yeah. Russ: So, 15 tons of food come my way. And of course I don't get to choose what's in that particular truckload. Or do I? Guest: Very little. Almost no choice. Russ: In the old days. We're talking about the old days. Guest: In the old days. Very little choice. And that was largely the problem. The caricatured example that we would deal with on our committee was where a foodbank director in Idaho would say, 'Look, I have a warehouse full of potatoes and you just give me more potatoes. I don't need potatoes. I need something else.' Russ: And even though there were 50 trucks available that day, I didn't even get to choose among the 50 trucks? Guest: Rarely. Rarely. Essentially, what they do is they assign them to you. Russ: Okay. That was the luck of the draw. And after I accepted it, or if I didn't because really I couldn't do anything with them--in that case, the potatoes, say--I went back to the bottom of the list. How many people are on the list? How many foodbanks were being supplied? Guest: 215. Russ: Okay. So, they are pretty--as you said, they are all over the country. And as you say, it's always going to get trucked to me? Guest: No. I have to pay for trucking. Russ: Okay. Guest: Eventually, the way it works is individual food banks pay, for the most part, for transportation. Occasionally there are subsidies given for transportation. But the cost to me is that I have to go get it. Russ: But it's always trucked. It's not flown, or-- Guest: Correct. Correct.
5:32Russ: So, that's a pretty amazing thing. So, every day, in the old system--this is pre-2005, if I remember correctly from your article, that of course we'll put a link up to--every day, before 2015, 50 trucks headed out from Chicago, centrally situated-- Guest: Not from Chicago. No. Not from Chicago. From wherever the manufacturer was. The manufacturer could be in Oregon, could be in Maine. They could be in wherever. So, part of what they were trying to do was also match in terms of the geography. Russ: Matching meaning which of the 50 trucks. Guest: Exactly. Russ: So, if I'm in far-off Florida-- Guest: If I'm in Louisiana, it doesn't make sense to offer me stuff from Alaska. Russ: Because it's not going to be worth it for me. It's too expensive. Guest: Exactly. Now, one thing I should be absolutely clear about is that, one thing that Feeding America felt and still feels is extremely important is that the system be seen as fair and transparent. So, you know, there are situations where they could use their discretion to give you something that they didn't offer to somebody else. But that sometimes led to hard feelings. So one of the things that they tried to do was, they wanted a system that was very transparent for all the food banks. And at times that meant they couldn't use their discretion in a way that we might think, 'Well, it could have made a foodbank better off.' Because they were afraid that it would be seen as unfair. Russ: And one more thing to clarify, before we talk about the reform. Where--you said a Tyson's factory might have some leftover, frozen chicken breasts--why would they be giving those away? So, talk about some of the reasons that Feeding America would be able to receive a donation of, as significant--these are large. It isn't somebody dropping off some cans. These are large, corporate sources typically. So, what are some of the reasons? Guest: Largely it's inventory management. You know, that, you know, firms, whether it's producers, whether it's distributors, make guesses as to what demand is likely to be. All from their own. There are certain products that they think are going to sell, but they don't. That would be one thing. Second thing, in the context of produce, is the size of the crop. Sometimes crops are a lot bigger than we think they are going to be. So, they don't have use for them largely in terms of inventory and storage. And the final reason, actually, is--and this is a non-trivial issue with respect to dairy products, meat products, fish, and for produce--is the danger of spoilage. Which is getting close to [?] its sell-by date. So for many of these reasons, that's sort of where the donations come from. Russ: They get a charitable deduction for that? Guest: They do. Russ: And what determines the size of that deduction? I never thought about this. What's the amount they can take? I assume it's some kind of market value? Guest: You're well beyond my area of expertise on that one, I'm afraid, Russ. I wish I knew. Russ: It's just a question. Guest: [?] Very much so. Russ: And some of it can be corporate--good will--they are hoping they are going to get some-- Guest: Absolutely. And in fact let me make one thing very clear because it will be important for the way that the new market-based system has worked: which is that there are many, many cases where donors will not give it to Feeding America, say, you assign it as you see fit, but rather they will specify that it has to go to a particular food bank. And usually that's the local food depository at the local foodbank. So, there's a lot of direct donations that actually occur in addition to these non-direct donations. And that's very important for the market. Russ: And that's for community, developed community relations. Guest: Of course. Russ: Feeling of efficiency on the part of the donor, as well. Guest: Yes. I mean, we all have preferences for local stuff. You know, we live in these neighborhoods. Russ: Yep.
9:11Russ: So, that system, obviously, if you think about it for 10 minutes, or 30 seconds, depending on your background in economics, you'd say, 'Well, it's not going to work so well.' But it's all "free." Meaning it's donated without a charge. There is this trucking charge; so that causes some issues. But in general: Don't look a gift horse in the mouth. This is a pretty good deal. So, what was the impetus to do something to change it? What was the--obviously there was discontent. There was discontent for 50 years or however long it existed. But what do you think provoked the dramatic change that you ended up being part of? Guest: I don't think--so, I was part of a group of 14 that was involved in the changeover. And I think it would be wrong to say that there was a groundswell of a desire for a major change when we got together. I think the general sentiment was that, you know, Feeding America is trying to do lots of things simultaneously. They are trying to make sure that the poorest areas get the most food. They are trying to introduce a system that is seen as transparent and fair. And both of those things were largely done through the old system. But the thing that it didn't really satisfy, which is where I would say grumbling occurred rather than widespread discontent, was that there was very little ability of the food banks to tailor the loads that they got to their individual needs. And they did see that as important. And they saw it as important on two dimensions. One, I would say, is transitory. Which is: You happen to have got yoghurt already. You don't need more yoghurt. In fact, getting yoghurt could be terrible because you don't have refrigeration to basically keep it from spoiling. So, I would describe those as kind of transitory issues associated with demand. They wanted something that would allow them not to get the yoghurt if they already have it. Probably an even more important thing, though, actually has to do with permanent differences in demand across neighborhoods--or different food banks, I should say. Which is: There are some food banks that have a lot of other sources of food. You know, they manage to get donated, direct donations, either from Feeding America or from other sources. And those are sort of in the language of food banking called 'food-rich areas'. So, one of the things that they really want to do is try to level the playing field in some sense, whereby those areas that are not food-rich might get access to more food. So they list those permanent and transitory demand sites saying that they kind of want to change. And there was another issue, I think which turned out to be important, too, which is: Often, the sentiment was that at times donations would be turned down because there was a fear it couldn't be placed quickly. And maybe we could kind of assist somewhere: by making that system more liquid, we could get more donations. So, people entered into this process, I think believing that we would tweak an algorithm in some way that might make things a little bit better on those dimensions. And really it was only through a very long conversation that really took a year, did we end up converging on something that looks closer to what economists would consider a market. Russ: Well, when you said that some of the food banks were not food-rich--so, they were in, say, more rural areas, or areas I guess more likely that there wasn't a large food distributor or [?] nearby that would make a donation. Guest: That would be fair. Russ: That Feeding America felt uncomfortable using discretion to hand out, to solve that problem. Of course, they could have. Guest: They could. But I think they were very reluctant to do something to do something that might be perceived as non-transparent. Russ: Or favoritist. Guest: And that's--economists are often very good at complaining about centrally-administered systems, bureaucratic kind of things. But they were walking a very fine line here, I think. Russ: Meaning? Guest: Meaning that they were trading off what they saw as efficient use of their discretion against the perception that it [?] as favoritism toward somebody. Russ: Sure. That makes sense.
13:26Russ: So, after about a year you said you worked on it--and we should just mention: Feeding America used to be called something else. Guest: It was called America's Second Harvest. In fact, it was America's Second Harvest while we were actually involved in this process. Russ: And it's been renamed since. So if you Google and find, as I did, some other story, it's the same story. Guest: Correct. Russ: So, what was your proposal? Guest: So, we as a group met together for, you know, the best part of a year, on and off. And I think relatively early in the process we raised the possibility that--and I think this came initially from the academics here--that we could think of using some market-based system, not using real money, which they felt very uncomfortable with, but rather using some kind of constructed money, which we called 'shares.' So, those conversations, or that suggestion, I would say, was initially responded to somewhat skeptically. And I would say it in the following sense, which is: Many of the people who run food banks feel that they are catering to people that have been marginalized. They perceive it as marginalized by the market. And they were reluctant to do this in the context of the food banking industry because largely they thought it might offer certain advantages to certain food banks--particularly the big ones relative to the smaller ones. So, for example, you know, Chicago Food Depository has many, many employees. You are in rural West Virginia; you have very few; so it may give advantages to that one relative to the other. But at the end of a year's conversation with a whole series of checks and balances, everybody got on the same page. And then, an important part of this is that we--meaning the committee--did not have the right to mandate what the institution would do. Rather, the decision was actually a vote, actually held by all 215 food banks. And it was the food bank directors who did all the heavy lifting in terms of persuading the other food bank directors to sign on for this. Russ: So, before we get into the details of what you actually proposed--which are fascinating--how did you educate those directors about what they were about to get into, given that most of them weren't economists? Most of them, I think in your paper you have one of them quoted as saying, 'Hey, I'm a socialist. This doesn't smell right to me. I don't like this.' How did you get them to vote for it? Guest: I would say off, first off, education happened on both sides. I learned at least as much from them as they learned from me, I suspect. And I think it was not hard to educate them on the benefits of choice. The harder part was showing them that choice needn't sacrifice assignments based on need. And a lot of that discussion and a lot of the slightly more complicated issues have to do with money supply, strangely enough. Which is, one of the things that we have done as part of this process is we construct this fake currency. So as the underperson in charge of the fake currency, I have to decide two things. One is how much of this fake currency do I print. And secondly, who gets it. Russ: And how much. Guest: So, a lot of the discussions were--and how much. Exactly. How much each gets. And I think we relatively quickly got to the point of everybody understanding that we could assign this, say, currency in a way where the most needy would be in our metric the richest. And we could sort that problem out. And much of the education at my end was: They are on the ground. They know how this thing works day to day. They kept throwing counterfactuals at me like, 'Hang on. We set up this option and you have a big food bank like Los Angeles (LA) staring at the screen all the time. You have somebody in rural West Virginia who is so busy, they are lucky if they check it once a day. How do we make sure we don't get an advantage for LA relative to the rural West Virginia case?' So there were a series of things like that, that actually cropped up. Another good example of something that cropped up is that a hugely important thing for Feeding America is donor relations. They have to keep donors happy. So, one way of annoying a donor is by either refusing something or even worse saying you are going to take it and then two days later it's still sitting on their dock. So, one of the other issues that we spent a fair bit of time talking about was: How do we move what's called 'hard to move' product. And the way we did it largely was by subsidies: we have negative prices as part of this. So, you know, to say that I, we educated them on the merits of markets may be true in a very broad-brush sense, but certainly I learned as much from them about how to design a bunch of safeguards. And you know, I would say the following, which is: The idea that sometimes you can use a constructed form of currency to do better than a central form of assignment is, you know, a relatively straightforward idea in a-- Russ: Textbook. Guest: Textbook. Exactly. Russ: Almost. If you've got the right textbook. Guest: Exactly. But there are very, very, very few examples where it's actually happened. And I think the lesson I learned from this is that, you know, that the hardest part here is actually sorting out the details to get it to work, and I think we've been lucky in that we've managed to do it. But for that I relied on both the staff, or we relied on both the staff and even more so I would say the directors of these food banks. Russ: Well, the best markets emerge. This is a special kind of market. You had an original top-down solution where Feeding America allocated stuff with their own rules. It's still a top-down solution but they decentralized a big chunk of it, which is how people--the mechanism by which Feeding America decided who got how much and what got changed to a system that has more incentives. So it's not a "real" market, but it's got market elements, which is often the case in these kind of markets. As we've talked about many times on this program, those kind of markets because they don't bring all the other pieces with it often fail or have very unattractive attributes. Guest: Sure. Russ: But this one actually seems to be fairly successful. Before we get to the actual reform in more detail about how it was implemented, I want to ask a question about the decision-making process. You said you had a vote among the 215 food banks. Is that right? Guest: They did, yes. The 215 food bank directors voted on whether to implement this change. Russ: So I assume it was one food bank, one vote. Guest: To the best of my knowledge that's correct. Russ: So that obviously meant that you had to, even if you didn't want to, you had to create a system where a lot of the small--I assume that the top 20% of the food banks are something close to 80% of the food that gets handed out in America by foodbanks. I assume. As you said, there are giant ones in LA and smaller ones rural West Virginia. So I assume that the one foodbank, one vote meant that whatever system you did choose had to be very attractive to the smaller food banks. It couldn't cater to the larger, more powerful ones. Correct? Guest: Absolutely. It did pass overwhelmingly so I think we managed to persuade not just the median voter here but almost all of them. Russ: Good. Guest: As I say, it was the food bank directors--like, you mentioned that one food bank director, his name is John Arnold, from Michigan, he was probably the person that was most resistant to moving toward a market-based system. He became its greatest advocate almost by the end, and I think his views managed, I suspect, to sway quite a lot of people. I wasn't in the room, but I would be willing to bet that was true. Russ: Didn't you have some kind of simulation beforehand or is that once they voted? Guest: So, after they voted, what we did was we had a 3-month simulation, so that people would get used to the notion of bidding. Russ: So cool.
21:41Russ: Okay. Let's talk about how the system actually works. So, one of the things you did was, you were going to auction off those 50 trucks now. And one of the punchlines of the story, which I'm going to reveal early, is, if I get this right, it was more than 50 trucks after a few years. Because the system was able to attract more donations--for whatever reasons. Possibly because it was more attractive. So this 50-plus trucks that are now being allocated, used to be: You were at the top of the list, take it or leave it; you move to the bottom of the list. How did it work now? Guest: So the way it works now, is the following. So, you log on in the morning, say. So there's a website. So, everybody logs on, and you will see the 50 truckloads that are there on a given day. You will have the opportunity to bid with our fake money, which we call 'shares'. It works in a standard auction setting. You pay the price that you bid. The way it works, though, is there is not a continuous auction. There's a sealed-bid auction that occurs twice a day. So, at 12 o'clock and at 4 o'clock Central Time, the market basically clears. And you can't see the bids of other people. It's sealed bid. And the reason that we did that was because we didn't want to set up a system that could benefit the ability of somebody to wait until the last second and, like, snipe, as on e-Bay. Because there was a sense that this would benefit the larger food banks relative to the smaller ones. Russ: Because they can't have a dedicated person sitting at the terminal. Guest: Exactly. Exactly. So, twice a day, [?] bid. So market closes at 12 o'clock; within 5 minutes you get an email telling you whether you've got the good or not. You can then re-equilibrate, decide what to bid on the second option during the day. Russ: But it's--the 12 o'clock option, when it ends--those trucks are gone. All 50 or whatever 25 or whatever it is, they are done for. Right? Guest: They've been sold. Exactly. Russ: And they start off on their missions, ideally. And then 4 o'clock, there's another group that's available. Guest: Correct. That's exactly right. So, a few additional things with respect to this. Which is: You can actually do what is called 'joint bidding.' So, one of the things we talked about is these large trucks, if you are a small food bank, you may not need a whole truck. So, what can happen is food banks can join together to jointly bid--fractional bids, as we call them--based to be on a truck. So that's one thing. Second thing is that--another thing that was selled to be important was that the smaller food banks would have enough funds to potentially buy the most desired object. So, one of the other things that we have is a credit system, whereby the smaller food banks--the bottom half, basically--can access credits. Which means they are given additional shares and they pay it back over a relatively short period of time. But it's a way of making all those accessible to them. And then the other thing that has happened is that, because one of the problems or one of the difficulties I would say with the old system was it's hard to move product. So what we do is we allow negative shares--negative prices, I should say--for individual units. So, you could actually receive shares for taking certain things, which would free up your funds to buy something else as well. Russ: The donor's happy now that it wasn't wasted, sitting around. Guest: Exactly. It's a way of trying to basically pay people to keep donors happy. That essentially how it works. Russ: Now, I have to say, I think the word 'shares'--choosing that word to describe the unit of currency is a genius bit of choice there. 'Loves' might have been better; but I think 'shares' is just about as good as it gets. As you didn't choose--it's Monopoly money, it's fake money. But it's not called funny money or dollars or green dollars or food dollars. Just out of curiosity: Was that somebody's idea? Was it like a eureka moment, or just kind of something [?] Guest: I believe that was one of our food banking members who came up with that one. I would say a more general point, which is: Share has this notion of ownership. And in fact that share of ownership, that sense of ownership and sharing was key to actually getting a different issue to work. Which is: One of the concerns that many of the staff and food banking raised, was the following. Which was: Suppose that there are some food banks that are, you know, food rich. Food rich means I have all the staples. You know--I have enough in vegetables; I have enough produce-- Russ: From my other sources-- Guest: I have enough dairy. From other sources. That, what I'm going to do is I'm going to hold my shares back and I'm only going to bid on the most desirable things. The most desirable things. The caricatured example is we would talk about were peanut butter and frozen chicken. Peanut butter is great--ton of nutrition value relative to its transportation costs-- Russ: Lasts forever-- Guest: Lasts forever. Exactly. So the concern was that a big food bank--let's take Los Angeles or Chicago--would hold their shares back, and what they would do is they'd spend all of their money on the really expensive stuff. Russ: And they'll always get it. Guest: They'd always get it. Russ: Just like the rich get all the TVs. No, I'm kidding. But I understand that's the kind of concern people would have, realistically. Guest: What actually turned out to be extremely important was the following. Which is: The way the money supply works is that whenever people spend--so suppose Los Angeles or Chicago spends 30,000 shares on a truckload of frozen chicken. That 30,000 shares gets given back to everybody that evening at midnight. Russ: Except for them. Or no? They get a pro-rated share? Guest: They get a pro-rated share, too. They get back 1/215 of the money that they pay. And that, it turned out, gave individual food banks the sense that they are the sellers of this food. They are the owners of this food. And if Los Angeles decides to pay through the nose for a truckload of chicken, we get their money. And that's a good thing. Russ: Well, it reminds me a little bit of the luxury tax in baseball. Right? So, if the Yankees or the Red Sox spend more than a certain level, it's true they have an advantage as a big market or popular fan base or rich owners, whatever it is. But at least some of that largesse gets spread back around to the other teams. Guest: Exactly. Now, sometimes there's a little bit of that that's pure psychology. Russ: Yes, it is. Guest: But another part of it actually turned out to be hugely important, which is the general equilibrium effect, which is: If it turns out that the food-rich tend to focus their expenditures on the very expensive stuff, in equilibrium it makes the staples really cheap. And that opens up a lot of food for the food-poor places. And that's really been one of the huge implications, I think, of the market.
28:50Russ: So, I want to understand the allocation and shares a little bit better. You didn't just create a fake currency--and of course before I say this, before I get into that--the total amount of currency in some dimension is unimportant. It's only going to determine the total price level--the average price of various goods. So, once you've set that, which is a mechanical choice to make sure there are enough zeroes to make it easy to calculate, and I'm sure you thought about all that--you didn't get a pro-rated--everybody didn't get 1/215th. Guest: Absolutely not. The allocation that you get of these shares, the simplest way to think about it is, it's how many poor people there are in your catchment [?] area relative to the total number of poor people that there are in the United States. So, if you happen to be in a particularly large area or you happen to be particularly poor, you will a larger share of the shares, as we call them. Russ: But aren't there more food banks in the larger areas, too? Guest: True; that's why we take catchment area. Essentially you each have your own catchment area. So it could be that, for example, San Francisco has three : in which case, their shares will be divided by 3. Russ: So you went from a system of your allocation being in pounds, which is remarkably blunt and weight driven--so there's heavier foods that probably everybody wants or doesn't want; there are light foods that are fabulous or awful; but instead you now went to a place you could actually choose what you buy. Guest: That's correct. Russ: So, that's the shares. So, each bank has a certain amount of shares. They can go online once a day to buy stuff, by auction. And I thought I saw in an article that actually you didn't pay what you bid--you paid the second-highest bid. Did that change? Or was that an error by the journalist? Guest: Yeah, that changed. No, no, actually; that actually changed from the end document which the committee put together to what was actually implemented. People, for reasons that have remained somewhat opaque, a lot of people like to pay what they bid. And in order to facilitate this through the election process, we switched back to it, first-price bidding, rather than second-price auction. Russ: Why did you--talk about the economics for a second of the second-price bid being attractive. Why would you have proposed that at all? Guest: The appeal of the second-price auction is largely it cuts out strategic bidding in a much wider set of circumstances. It will cut out strategic bidding compared to a first-price auction. Ex post I don't think it's turned out to be that important. There are just so many bidders and there are so many opportunities for substitution, both across time and across food, that I think the strategic elements have been minimal. But conceptually we thought the second-price bid would be cleaner, but it didn't resonate with many of the food-bank directors. Russ: So, I don't know if I'm doing the math right here, but in the old days when you allocated food by getting in line, essentially, and then getting a fixed amount of pounds depending on your need, defined by poverty area and other factors--in the old days, every day, say, 50 trucks of food were allocated to 20 of the 215 banks. So, once you got your truckload, that day, if you got it, you went to the bottom of the line, back of the line; and it took roughly another 4 days till you'd get another delivery that started from Feeding America. How did that change? Because you could lose every auction now--if you have bad luck, if you're not good at it. Was there any pattern after--of course there's always learning factors; it takes a while for the market to settle down; it takes a while for people to figure out what is a reasonable price or pasta versus chicken. But once things sort of--and I'd be interested in how long that took from your perception. But once things did settle down, how often did you win? Guest: So, on average you still win just as much as you did before-- Russ: You'd think. Guest: You think. But it depends on what you bid for. That's sort of the natural thing. Let me undo the two parts. First is: how long did it take? I would say--let me start by doing the following, which is within the first 7 months, almost everybody had at least won something. So, 97% of people had won something within the first 6 or 7 months. So, people were engaged. Having said that, I think there was a lot of learning that happened at the beginning. And, one of the ways which I can see this, we can see this, is, we talked about these negative prices? We also, one of the other things is that there is a lot of stuff that sells for zero. A lot of produce actually sells for zero. They don't allow negative prices for produce. Russ: Zero, meaning you only have to pay the transport costs. Guest: Exactly. So, first 2 years, if you do negative prices, about 10-12% of all loads sold for negative prices. And it was largely because a lot of people weren't bidding. If you look 2 years, over the succeeding 5 years it's gone down to about 4%. So it's kind of converged. Zero, non-positive price, so include all the zeroes, the first year we did it, 50% of things sold for zero. Russ: Wow. Guest: Now it's converged--it's very steady. It's converged to about 20%. So that there appears to be--there was a certain amount of reticence to bid on stuff, but now it appears that everybody is up and running. So, if you look now, the main thing that's actually happening now is not that people are unlucky, they don't get a bunch of stuff--I mean, you can bid every single day. It's not like if you don't get something today you are out of luck. You just change your bid a little bit for tomorrow. So, that's not usually the main sorting that's actually happening. The main sorting that seems to be happening is some people have decided to become what we call the bottom feeders--which is, they want a lot of pounds of food buy relatively cheap things. And then there are other people who are sorting[?] up the quality dimension and saying, 'We want the peanut butter. We want the chicken. We want the cereal. We have no interest in the stuff down at the bottom because we already have that stuff.' Russ: So cool.
35:12Russ: Do you have any feel--this is maybe getting outside your expertise, but do you have any feel for how the variety of food changes day to day? Obviously there's seasonal differences. There are seasonal differences in demand as well--winter and summer are going to be different, I assume, and [?] in certain areas. More in Chicago than southern California. But if it's a March day and I didn't get the potatoes I wanted, or the pasta or the cereal, whatever it is, is that likely to be available tomorrow? Is tomorrow just another random set of stuff. Guest: Depends what it is. So, here's sort of the thing, which is: One of the things we've looked at is we've looked at the distribution of different kinds of food is. Certain kinds of food don't come up that often. Fish, for example. If you really want fish and there's something on the market today, it could be 3 weeks before you see fish again, in anywhere that's close to your area. By contrast, if you want carrots, you just wait till tomorrow; there's going to be something that's either carrots or a close substitute for carrots. So, it very much depends on the kind of food. In fact, probably the single most desired thing, if you look at prices, is pasta. Pasta doesn't come up that often. That's one kind of--if you see it today, you've probably got to do it. Okay? But the other thing I would say is the following, which is: I think part of the reason why this system--and I believe it has worked--has worked is the following. Which is, it's like an infinitely repeated game, which is: You can bid on something, but if you don't get it, you just wait till tomorrow and bid on something. So, your money still has value. Your fake money still has value. Which is one of difficulties you get with these kind of fake money systems. But the other thing I will say is, there's loads of variability in supply on a given day, or a given week. You look over a month, over two months, then you are going to get a lot of kind of consistency in what the supply is going to be. And what food banks are good at, is they can substitute across time--which is: Well, we didn't get chicken for this time, but you know, as long as we get chicken or pork or hamburgers next time, that's okay. Russ: Yeah, because it's protein. Guest: It's protein. Russ: It's not to fill out the menu to make it match whatever they promised. They want protein. They want carbohydrates. They want fat--whatever it is. Guest: In our language, this is what we call a hedonic, which is: There are quality differences across different kinds of foods but we can kind of aggregate these things up. Like, fish is not that unique, because there are substitutes for fish in terms of nutrition. Russ: Talk about maroon pounds, which is fascinating, and how that changed when the new system was put in place. Guest: So, one of the things that we were interested in is obviously efficient use of food. And your first order of thought is: Use the food that Feeding America, or at the time, America's Second Harvest, has, in an efficient way. But one thing we also wanted to do was use their other sources of foods in an efficient way. So, let's take the example of before, which is: You already have yoghurt sitting in your warehouse. Suppose a local distributor offered you more yoghurt. All else equal, you'd have to say, 'Look, sorry, we haven't got a 'fridge. We're not going to be able to move this thing.' So, can't take it. Now what we have done, through this system, is we allow you to take that truckload of yoghurt, put it on the market itself, our market we call the choice system. You put it on the choice system; you get the returns from this. So in some sense, what it is, is it's an attempt to increase supply. But increasing supply not from the usual sources of food that Feeding America would get, but from your own sources of food. So, it's actually resulted in--if you look on average on over the 5 or 6 years after this system was introduced--on average it's about 12 million pounds of food. Russ: Extra. Guest: Extra food. Exactly. Russ: Going through the system, per what? Guest: Per year. Twelve million pounds per year. Exactly. And the other thing that--sorry, go ahead? Russ: Well, what I was going to say is what's powering that, just again to clarify: So, if I'm a foodbank in Florida, and I get lots of oranges, so I never want to take oranges, I might decide to take some of those oranges, put them on the system and get some shares to buy some more of other stuff. Right? So-- Guest: Absolutely right. Russ: So, it's a--a sharing economy kind of--it's a brilliant, it's just a brilliant idea. So, before, all you were doing was allocating the donations made directly to Feeding America. But once you had the system in place, why not allow donations from local places to local food banks be put into that national economy that you've created? And presumably that encourages local donors who aren't part of Feeding America to donate locally, because they know it will get used even if normally it might just get turned down. Because they've already got plenty. Because they've already got that natural source. Is that accurate? Guest: That was the hope. Which seems to have occurred. It certainly has occurred. One thing I will say, though--it's sort of related to your example about oranges--which is: If you look at, I told you that 12 million pounds of food tends to be put on the choice system--if you look at the quality of food that gets placed through these maroon pounds, it's about twice as high as the average quality in the normal system. And part of that is that food like certain kinds of produce--broccoli, for example--tends not to sell for very high prices. So people tend not to put it on. Instead, what they will usually do with those kind of foods is they'll share them informally with somebody else. So, you know, you are in Jacksonville; what you'll do is you call up the guy in Orlando and you say, 'You want my oranges?' Because truthfully, you are not going to get very many shares of oranges. Where you will get a lot of shares if you have a truckload of cereal. So, quality-adjusted, the number of maroon-pounds is probably twice as high as the 12 million pounds. Russ: Talk about peanut butter for a second. Guest: Sure. Russ: You said pasta doesn't sell. Guest: Doesn't[?] sell. Russ: Does sell. [?] One of the reasons it [sells/doesn't sell ?] is it lasts forever. Like peanut butter. But the fact that it lasts forever means, in general, one would think, you wouldn't get a lot of donations of it. Because you are not worried about spoilage. You are not worried about it going bad. Guest: Absolutely right. Russ: And I would also think that peanut butter--there are only a handful of peanut butter manufacturers. Why would they ever be making donations to Feeding America? Are they--is it inventory control? I don't know. I can see it would be a big problem to gauge the national demand for peanut butter so that the national producers, like Jif or Peter Pan [Skippy?--Econlib Ed.]--you'd think it would be easy for them and they wouldn't be making donations. Or is there something else going on there with peanut butter that I don't know about? Guest: It's a good question. Peanut butter is such a rare product that perhaps you are absolutely right on this one. And the more common--I would say, if you wanted a good which is very valued, which actually comes up relatively frequently is actually cereal. So, and those have quite high storage costs. Peanut butter doesn't have huge storage costs, I mean from manufacturers. So it may be that storage costs are sort of non-trivial for manufacturers for things like cereal boxes.
42:28Russ: So, let's talk about the overall reaction. So, as an economist, my first thought when you tell me this system--why don't you tell me that the software worked and the interface was fairly understandable, attractive, easy to use? You tell me there's a lot of bidding going on. I'm pretty confident that what's going to result is a better allocation. So, one of the natural effects is: You get what you want. If you are a local food bank. Instead of getting what you are stuck with. So, that's pretty obvious. You could--it's hard to measure, as you point out in one of your articles. But, there's an interesting piece of this which somebody might not think of, which is the supply effect. So, every day you had roughly 50 truckloads to allocate, Feeding America did. But one of the ideas of this was to improve the attractiveness of giving to Feeding America. Because donors would now know that it would be used more effectively. So my question--so one of the things you claim in the paper was that there was a supply effect: that they were able to attract more donations. I want you to talk about that. And if you know anything directly about how Feeding America marketed their system to potential donors now that this was in place? Because the other thing that strikes me about it is it's pretty complicated. Usually when you make a reform like this, one of the challenges is, the iconicoms[?] of this really clever system--and everybody goes, 'I don't get it; it's too complicated.' So, you made it transparent and simple enough that you can sell it. Did Feeding America use that as a selling point to donors; and do you think that's why their supply has increased, and by how much? Guest: So, I think that is an excellent question. I think supply has increased. If you look around the time of the change, to give a sort of very narrow window, from about 2000-2004, just before the change was introduced, I would say about 200-220 million pounds of food was being assigned through this. Within the first 7 months, I would say, of the new change, we were getting 60, 70 million extra pounds. And that's largely maintained. The only caveat I have to give to that is we got hit--'we' meaning Feeding America--got hit by the financial crisis. The financial crisis both affected supply and demand. So there was a blip that happened around 2008. But, the numbers still look very good, I would say. And then on top of that, you also have these maroon-pounds that are coming in--this food that's being brought in from different sources. So, I think in terms of aggregate supply, I think it's been quite [?], I would say. Russ: But the maroon pounds--by the way, they are called maroon pounds because you gave a designation of yellow and blue for the other kinds of food that came to the system. It's nothing about the color of the food. It's not about-- Guest: It's not at all. [?] In fact it's a much cheesier icon, which is maroon is the color of the U. of Chicago. Russ: Oh, right. Go, go, Maroons. Guest: Exactly.Russ: My alma mater. Guest: In terms of how Feeding America has marketed, the honest answer is: I don't know. If you ask me to guess, I would be willing to bet that a lot of the change has not occurred through explicitly marketing to donors, that this will go to better end use. I could be wrong. I suspect a lot of it comes from Feeding America now accepting donations that they previously wouldn't have taken because they were afraid that they weren't going to be able to move it. But now the market is more liquid, so I could imagine in their minds--and I'm imputing here--them thinking, 'This total [?].' Like, people take stuff. They'll take it. You know, especially now that we have these negative shares. Russ: And I guess just as a technicality: The maroon pounds--the locally donated pounds that end up on the national distribution market through the share system: They weren't really an augmenting, fully augmenting supply, increase in supply. Because they just were not being measured before as part of the system. But they were going to people and feeding them.
46:26Russ: And it's also imaginable that local food banks are now accepting more donations than they did before, because they can do something with them. But it's probably not 12 million pounds. Guest: Absolutely. I have no idea what the alternative use for those 12 million pounds would have been. Less than their use, I guess, is what we know, because they are choosing to put it on. Russ: So, as I said, and before; and you point out in your paper, you point out: Well, this is going to be better. Talk about some of the anecdotal reactions you got from executives--5, 6, 7 months, a year or two years into the program, after they had gotten the hang of it. What was some of the anecdotal evidence you got that this was an improvement? Guest: I think that the overwhelming response, I would say, 'This is working as you told it would work. We get to pick the one'--I would say, the one thing we really worried about was the following, which was there would some particularly smaller food banks that either would be so intimidated by this they just wouldn't get involved. So, one thing that we allowed them to do was to delegate bidding to Feeding America itself. Just give a broad mandate for what you want. We'll do it for you. No food bank has chosen to do that. Okay? Second thing we did was we spent a lot of time setting up what was effectively an appeals committee. Which is: If you think you've been screwed by this system, okay, what you can do is you can make an appeal and we'll change the way that we're assigning shares to you. Now, it was set up in two senses. One was for, like, transitory things that could happen. So, take an extreme case, which is Katrina hits New Orleans. How do you make sure New Orleans gets more food? There's nothing in the formula that deals with New Orleans. So, second thing would be, a more common example would be, like suppose a local plant closes. Big increase in demand. So we set up this apparatus largely to deal with these exceptional circumstances. As Feeding America tells me, this committee has never, ever, ever met. Because nobody has ever complained. Russ: That's pretty amazing. Guest: So, instead, I think the really fundamental upshot of this is that, unlike the old system, any food bank that wants food on a given day can get food on a given day. Because you can buy inexpensive food very inexpensively. As long as you have a truck, you can get it. And I think that sort of availability has kind of overwhelmed any other issue that might have come up as a result of a lot of the problems. The second thing we worried about a little bit is that maybe the small guys wouldn't know how to bid--so they wouldn't be able to work it out. And the big guys would know how to bid because they do it more frequently. But I haven't empirically tested that yet. But it doesn't seem to have been a major concern. Russ: I heard you say something--I think I understood it and fully understand--so, when we talk about 50 truckloads of food, in my mind, it's just sitting on the truck waiting to go. But that's not actually what happens. Guest: No, no. Russ: So you are saying you have to go get it. Meaning you have to rent a truck; you have to-- Guest: They usually own trucks. They own trucks. So what they will do is they will either own or rent, and they will send it to wherever the donor is. Russ: So that's the sense in which they pay the transportation--the literally have to go get it. Guest: Right. Russ: So it's not like they get billed after the delivery. Guest: Correct.
49:48Russ: Let's talk for a second about inflation. So, obviously, if for whatever reason, let's say we got an increase in the supply of donations over time, that would have an effect on the average bid. Or, vice versa: if there was an increase in the demand for food because of the financial crisis that would normally increase the prices overall. But you wanted to hold--you wanted there to be zero inflation. So, did you change over time the total? Why was that, for one? And number two, did you change the total number of shares that were allocated daily? Guest: So, the way it works is that--let me start with the objectives. So, the objective was that in order to help people to bid, seeing the history of prices for similar goods would give you some notion as to what to bid. I mean, a value to everybody I could imagine; but it was selled[?] to be of a special value to the smaller food banks in the following sense. Which is: Suppose I am a small foodbank; I really want a truckload of cereal. I haven't bid on cereal for, like, a year and a half, so I'm not really sure I should be paying for it. But what you can do on the website, you basically click a link and when you click that link it says: This is what the history of prices is for cereal over the last 5 years. And what we wanted to do is set up a system whereby by observing that history of prices, it gave you a reasonable instinct for what you should be bidding. Now, as you say, one of the things that can actually change the price level--forget relative prices, but change the price level as if, suppose that quantity of shares didn't change but the amount of food doubled. Okay? Then prices are going to get cut in half, all else equal. And that would be confusing for people. Russ: Just like life. They wouldn't know whether their cereal had gotten cheaper because nobody wanted it any more, or there was a lot of it, or whether everything had--it was just a deflation problem. Guest: Precisely. So, essentially what we did was set up a system where annually, basically what we do is the change in food basically from one year to the next basically changes the aggregate supply of shares. That's how the entire system was set up. Russ: And that's what you've been doing from the beginning. Guest: Yep. So that's essentially the objective, largely to keep prices stable. Now, to go to the second point that you mentioned: There are times when you want prices to change. And that's if relative demand and supply conditions change. And you mentioned the financial crisis. We set this system up to essentially keep, all else equal, prices relatively flat. Prices have remained pretty flat except for what happened in 2008. In 2008 there was a huge shock where demand went way up at food banks; supply went down at food banks. And what we see in the data is that the price level went up by about 25%. Russ: Wow. Did relative prices change, by the way? If you looked a time series of cereal, or, correcting for inflation, was there much of a change in relative prices over the whole thing? Guest: Okay, that has yet to be tested. I have recently got all the data on the auctions that we have had. Let me tell you my guess. My guess is that the relative price of low quality food went up for the following reason. Which is: The way it works in a typical year is that people who have loads of other food only focus on the expensive stuff. I think what happened during the financial crisis--and this is part of the reason the price level rose--is that a lot of these larger food banks felt they didn't have enough of the staples, so they started to bid on the less expensive stuff. So my guess is relative prices compressed, but pure guesswork at this stage.
53:54Russ: Is there a reason that you just wrote about this? It happened a while back, this change, if I remember right--it was 2005. This is so interesting, especially to economists but I think even to normal human beings. Is there a reason it's taken a while to write about it? Or did I miss something? Am I just late to the party? Guest: No, it was something I had always meant to do, but you know, something always managed to get in the way. And largely it was one of my colleagues persuading me. Truthfully, people persuaded me on the following basis, which was to say, 'Look, this might be useful for other people.' Russ: Sure. Guest: And essentially that's what I found persuasive. So I ended up getting around to it. So, other than, you know, me being just trying to [?] the other kind of work that I try to do, I have no other excuse, I'm afraid, Russ. Russ: That's okay. Do you see any direct applications you think might be happening in response to this? Are there other areas where things are being given away that might be allocated more effectively through a created market? Guest: Not yet, to be honest. Let me give you one thought I have--it's probably 4 weeks since I wrote the paper, 6 weeks since I wrote the paper, so probably these things take a little time. The one case I think that is of interest, to me at least, is: Remember the piece that we have been involved in. The piece that we have been involved in is getting food from a manufacturer to the intermediary Feeding America, to the Chicago Food Depository. The Chicago Food Depository is one example. The Chicago Food Depository then has to decide how it gives food to your local church, your local community center, your local school. Russ: Sure. Two middlemen there. Guest: Yes exactly. And they do not use this fake money, at that decentralized level. And I actually think it could be useful. One of the things--I don't think you want to get into individual churches and food banks bidding on stuff. But one thing that could be done is that you essentially set up these fake money accounts and you give them credit cards. They come in and they basically buy how much broccoli they want, how much chicken they want, at the local food depository. The way it's done in many food depositories--it varies, but the way it's done in many food depositories is you actually pay real money to get food. Now, it's usually subsidized and some things are free, but you actually use real money. And part of the problem that Feeding America always had with real money is, they were afraid that it was--that it would be better-off neighborhoods that would be better at fund raising than the ones that are in the poorer neighborhoods. So, they felt it could be really counteracting their desire to allocate based on need. I don't know how substantive that problem is at the local level; I think that's at least something worth further considering. Russ: And of course all of this is an application of what Hayek called the information problem--that it's very difficult to know what people actually want. We're talking about food banks. Of course, taking that to the next level would be people--the people who actually eat the food or take the food from the food bank. We live in a welfare system that has a lot of in-kind attributes and a lot of people--going back to Milton Friedman and others have suggested, 'Let's just give them money.' And of course it's a big trend in international development, fighting poverty--the failure of a lot of other systems has encouraged people to say, 'Let's just start giving them cash.' You have any thoughts on that? Guest: At a local level--I would say in the following sense. Which is, there's a lot of disagreement--this actually post-dates our change in the following sense. So, I use the the food banks as a metaphor for a larger point, which is: There is a big disagreement among the food banking community now over the issue of nutrition. So, probably over the last 4 or 5 years--and most food banks agree on most things, but one of the things they disagree on is the extent to which you should only give out food which has high nutritional value. And I think a lot of this comes down to, there's a degree of paternalism, which is: 'We ain't giving you fizzy drinks because fizzy drinks aren't good for you.' And yet on the other hand, there are a lot of people that say, 'Look, most people go buy fizzy drinks. Why shouldn't they have them, too? You know, people want these things, they should be offered them.' Russ: It's fascinating. Guest: So, I would say the local example that I'm kind of dealing with, sort of has that problem--'problem' may be the wrong word, but that discussion, I would say, that's currently going on. But there have been quite a few food banks that have sort of changed their demands. And this is one of the things the market system allows you to do, which would have been really difficult to do under the centralized system, which is: How do we make sure that certain food banks get more nutritious food and other food banks say, 'Look, I just want to make sure that somebody has food when they go to school in the morning'? And I think one of the things the market allows you to do is sort on that dimension as well. Russ: I just want to mention before we close that this whole experience reminds me a little bit of the Radford article we've talked about on EconTalk before--we'll put a link up to that episode. I think it was with Mike Munger, about the use of currency and allocation in a prisoner of war camp. It's a wonderful paper to go back and read, after you've listened to this episode you can go back and listen to that episode of EconTalk.
59:20Russ: I want to close with a question about your personal experience, having gone through this, which was a lot of hours of work--there was a lot of uncertainty whether it was going to happen at all; it finally happened. It seems to be working; it seems to be working well. How rewarding has that been for you? And, has it changed your views on markets or prices in any significant way? Guest: So, let me do the personal piece first. I think I learned an enormous amount--about myself, through this process. I have sort of really kind of sketched over one piece of it, which is, it took us a year of meetings to sort this out. I think one of the things, and I really [?] at one of my colleagues here, Harry Davis, this, which is: This would never have happened if we hadn't had listened. And I think to a large extent, I learned to listen, during this entire process. The second thing I will say is that I feel, I actually feel extremely lucky that I was involved in this for the following reason, which is: many, many people donate their time--millions and millions of people donate their time to all sorts of worthy causes. The thing that I got lucky about is I managed to donate my time for something I had a comparative advantage. I could use my skills to leverage to something--we, I should say, use our skills to leverage to something that's been valuable. And in that sense it sort of oriented my choices from then on: these are the kind of things I should do, things I can do better than other people or relative to my own inadequacies[?] in other areas, I should do. So, I think it's been extremely valuable to me in that sense. About the merits of a market? You know, I'm a faculty member at the U. of C. so I may not be the most representative person on the planet here. But I actually think I learned that with a relatively small number of tweaks, one can simultaneously satisfy some kind of efficient allocation with some desire to allocate based on the metric of need. You know, through the use of allocating more shares to the food banks that have the more poor; through all of these safeguards. You know, if you had told me 10 years that it would have been as seamless as it appears to be, I probably wouldn't have believed you. I mean, I fully expected some really unintended consequence to wriggle out that we'd have to like change something. But I've been surprised that it seems to have gone as smoothly as it has.


COMMENTS (16 to date)
Chris W. writes:

HI Russ,

I enjoyed the recent episode, however I had a question in my mind I kept hoping you would ask. How good are these food bank stewards at choosing food for their constituents? Thinking perhaps to the 'bulk purchasers', do these bidding strategies represent the culinary preferences of those they represent or do they represent the food preferences of those that are doing the bidding, which is entirely a different thing? So I guess it would be interesting to learn, if you have a feel for this sort of thing, given their ability to acquire food more selectively, if that is tailored to the tastes of the people that consume it, or if personal preference of the bidders (either dietary, or concept of health, or whatever) dictate the bidding outcome.

This week had a particularly good guest.

Mark Crankshaw writes:

Good podcast, very much enjoyed Mr. Prendergast and his delightful Irish accent...brings back some fond memories of living in the British Isles...

Wayne writes:

Very interesting topic this week. It does seem to be a bit complex & I can't help but think there is a more direct & less apparently "artificial" way to overlay free-market principles onto the food-bank realm. That being said, if it's working better than the prior system, I certainly can't argue with success. I'm fascinated by the emergent-order that developed but leary of Statist, top-down Planning schemes in general.
--Is there much research in how to efficiently distribute goods to people, for "free?" That is, what's going on with price-mechanisms in the non-profit realm? This appears to be all about incentives, above & beyond anything else.

I actually work for an organization that distributes food-bank items to the end-user. (It's not our primary activity but we are involved to the tune of about 400-600 lbs./week.) Every Monday we receive an email listing from our food-bank, of what is available that week. We then order what we want & schedule a time for pick-up, generally 2-3 days later. All items are "free" to us, but we do pay a "handling-fee" of 45 cents/pound & provide our own transport. I have no idea how they allocate the food to us, but we have never be turned-down on our order. Fresh & frozen goods are always popular but difficult to store & distribute, at least for us.

>Chris W. brings up excellent points. We primarily choose items 1)we know we can give away & 2)provide nutrition & variety. We do not however "edit selections" in the cause of Nanny-Statism, --last week for example we were able to acquire 60 lbs. of Halloween candy, which is a great treat for folks who sincerely have difficulty buying "food." (our primary activity is helping people find jobs, so they can buy their own food. The unemployment rate isn't "5%," it's more like 15%.)

We are fortunate our food-bank handles non-food items, such as toiletries, detergent, paper-goods & the like. We are also fortunate to be located near Battle Creek, Michigan & are able to acquire all manner of dry-cereals. (A note for Mr. Roberts on donations of cereal; the major companies are very efficient in balancing supply/demand but one of many bottle-necks in the distribution chain is storage & rotation of stock. They are also heavily regulated & if a production-run isn't "exact" or the box is missing a comma in the text, or if the product wasn't stored or handled "perfectly," it results in "wastage." Most but not all of that is given away to food-banks & the rest is destroyed. Donation tax-deductions are limited to the actual production costs. It's primarily just good business all around to donate what can be used.)

Again, congrats on 500 episodes!
And, don't change the music!

Buzz writes:


Interesting interview. I'd love to know how much it cost to develop this system.

Planet Money had an episode on this too:

http://www.npr.org/sections/money/2015/11/25/457408717/episode-665-the-pickle-problem

SaveyourSelf writes:

I have tears in my eyes--this Econtalk was so good.

After listening to this episode and the kidney donation episode, I begin to wonder if economists have finally discovered their niche...or perhaps the field of Economics has finally become mainstream enough that laypeople know what questions economists can actually answer. Either possibility is exciting.

I really appreciated how this guest labeled what he knew vs. his guesses when answering questions. UoC is looking awfully good in this interview.

Mort Dubois writes:

Fantastic episode. My questions: is there any corruption in the system? It seems like it's a big endeavor, maybe totaling thousands of people participating. Is there some mischief going on? And how large does a closed monetary system have to be before corruption is inevitable?

David Zetland writes:

Wow. Amazing story of market design. In fact, so amazing that it resembled my (our) proposal for an auction market for (academic) journal articles, which would do a LOT for improving matching, marking value, rewarding good journals (we have an artificial currency too!) and changing referees from adversaries to allies.

http://www.springerlink.com/content/2q80214867370564/ (open access :)

Maybe the next challenge for Professor Pendergast?

Greg writes:

Interesting interview.

Does anyone know why they don't allow negative prices for produce?

Pseu writes:

Another fascinating episode. Made for good dinner-table conversation with the wife and kids. I, too, had heard the Planet Money version of this story and was happy to get some more details. In particular I had been curious why the food banks used a sealed-bid system, and now I know.

This story also reminds me of another Planet Money episode - theisone about how zoos and aquaria trade animals among themselves in a barter system rather than using money. I wonder if some kind of a "fake money" system (I prefer "quatloos" to "shares", but that may be because I'm a giant geek) would help that system work more ef-fish-antly (sorry) as well? Perhaps with zoo animals the market just isn't liquid enough for a system of this nature?

jw writes:

During the podcast, I was trying to imagine what the supply/demand curve would look like as supply was fixed (or more accurately, randomly reset at a fixed level). However, the introduction of the maroon supply would work to help balance the supply curve with the demand curve. Brilliant!(I think...)

Also, I wonder if the self described socialists that run a few of the food banks now think that "shares are the root of all evil". As Rand's Francisco d’Anconia said, money (or a share) is just a tool for exchange (or barter). Hopefully, they are learning something (I am ever optimistic...)

Daniel Barkalow writes:

I found the discussion of the winner/second-place price aspect of the auction interesting. When I'd heard that they used a sealed-bid auction to avoid giving an advantage to participants who could spend a lot of time on the site, my first thought was that the second-place price was the obvious mechanism. Paying the second-place price is equivalent to giving bidding instructions before the auction to an ideal bidder, who will constantly bid marginally over the top bid until reaching your limit; the winner pays exactly what was necessary to get everyone else to drop out, and it means that the sensible strategy is to say the most you're willing to pay, rather than offering less in hopes nobody else bids close to your limit.

On the other hand, the sealed-bid top-price auction means that you pay more for really wanting something, even if the rest of the market doesn't happen to care. If you want to make sure you get cereal, you're going to pay a high price for it, even if everybody else happens to be looking for other stuff that day.

Although nobody mentioned it, I thought this was a very interesting example of an economy where buyers got spending money irrespective of their contribution to the economy, and it would actually be somewhat inappropriate for buyers to try to increase supply. I think this is a possible destination for an economy that isn't limited by labor, which has come up in a couple of other recent episodes. A highly automated production system supplying a population that does meaningful work that doesn't generate revenue or affect supply would be very similar to the situation they were trying to address.

Vesall N writes:

I've been listening to econtalk for about a year and have really enjoyed all of the episodes, but this one was just simply amazing so I had to comment! Thank you so much for putting it together, Russ. And thank you Canice for getting around to writing the paper.

Three thoughts/questions come to mind:

1. Throughout the entire interview, I was thinking about how this setup/institution has to be considered both a political and economic institution, perhaps brought together through community governance. The fact that the entire body of people came together and consulted on the rules/institutions that would discipline the market exchange mechanism to produce the allocation distribution it desired cannot be decoupled from the construction of the exchange mechanism. There are many aspects of the process that really do separate it from how economists typically think of increasing market access, specifically in the involvement of those participating in the market in creating the "rules" of the exchange process. This is fascinating in and of itself.

2. Somewhat related to 1, the question of the 2nd middle-man mentioned at the end seems to be quite important. My father-in-law volunteers as part of the food-delivery team through his church, which caters to perhaps relatively affluent members of society. He is, however, connected through other volunteer endeavors with less affluent churches in historically black neighborhoods. These churches, if I understand both my father-in-law's experience and this episode's discussion of food bank institutions correctly, still rely on local relationships to procure food and hence do not get the best quality of food. I think it has something to do with not having the infrastructure (as per various regulations governing the system... such as refrigeration capacity) to participate in the food bank allocation process. The reason this is related to point 1 is that there were certainly parties interested in the process of "food bank" food allocation who were not able to contribute to generating the rules of the exchange process. Is there a reflection/evaluation process in place to adjust rules and institutions on the fly as we learn more about how food is being allocated?

3. I definitely need to read the paper, but my sense is there is something that is not "competitive" in the traditional sense about this market. I understand that all of the food banks are "owners" of the food, but the "firms" do not seem to be competing for profit with each other, or are they? The main source of competition is in procuring the best prices on the demand side (it seems), but not for firm survival on the supply side. In a way, there is monopolistic supply without monopolistic pricing partly because inflation of prices has been controlled in the system, but also because the suppliers are also the consumers. Does this make sense?

[Accidentally misspelled name fixed.--Econlib Ed.]

Russell writes:

I found the discussion pretty interesting. A system for allocating a scarce resource among alternate users with the objective of trying to do so in a more flexible way. I would guess that having a fairly fixed group of participants will help maintain the system. Extreme stress may cause it to breakdown but when that happens most systems have their problems.

Robert Swan writes:

Thought-provoking podcast -- as so many of them are.

At about 15 minutes in Russ says "the best markets emerge". My reflex is to ask "Best? Along what axis?" While I would be interested in any response to that, for this comment I'm more interested in how you decide whether or not a market has "emerged". It seems to me that all markets start with some design and, once launched, other aspects will emerge. Perhaps the real distinction is not in the design at all, but on how willing the market's designer is to tinker with what emerges afterwards, and whether that tinkering is to thwart or assist.

In this vein, the story of how the maroon goods entered the broader market was very interesting. I'd have liked to have heard about other ways the original design evolved to accommodate feedback.

northshore writes:

Fascinating discussion and system application. Thanks for a very good interview and prompt to read more about this.

Aaron Spear writes:

Regarding your and your guest's question about a food producers ability to make a donation and to deduct it from their taxes, although I am not an expert/CPA, I would expect the producer could deduct their expense equal to cost of goods sold. Further, the deductability would assume that the producer has Net Income to deduct at the end of the year. That said, it's possible there are local/state or other tax credits or incentives that I am not aware of. Great show, and keep up the good work!

Comments for this podcast episode have been closed
Return to top