Munger on Milk
Sep 2 2013

Mike Munger of Duke University talks with EconTalk host Russ Roberts about why milk is in the back of the grocery store. Michael Pollan and others argue that milk is in the back so that customers, who often buy milk, will be forced to walk through the entire story and be encouraged by the trek to buy other items. Munger and Roberts argue that competition encourages stores to serve customers and that alternative explanations explain where milk is found in the store. The conversation also discusses restaurant pricing, government "nudging" and related issues of grocery economics.

Explore audio highlights, further reading that will help you delve deeper into this week’s episode, and vigorous conversations in the form of our comments section below.


Sep 2 2013 at 7:02am

Not sure about Munger’s Yeltsin story. The same is said of Khrushchev’s visit to the US in 59.

Mike Munger
Sep 2 2013 at 8:51am

MP: Yep, I think the Yeltsin story is an urban legend. Here is a reference to the Yeltsin version. But the overall point is more general, and there is no specific reference to the story itself….

Russ Roberts
Sep 2 2013 at 9:04am

I hope everyone enjoyed this conversation and learned something. In my enthusiasm to get some of my points across and to engage Mike, I neglected an important punchline. I spoke about how competitions constrains the opportunity of businesses to exploit consumers but I should have made it clearer that we consumers, we buyers of milk want the milk to be in the back of the store. Not because we like walking through the store to get to it, but because if the milk were in the middle of the store or near the front, it would have higher costs thereby resulting in higher prices we aren’t willing to pay.

In other words, we prefer cheaper milk that’s less convenient to convenient milk that’s more expensive and that’s what competition among stores gives us. Of course we’re not all the same. For people in a hurry or who are only buying milk, convenience stores offer an alternative. And as Mike and I pointed out, a number of supermarkets have found a way to keep milk cool relatively cheaply in the front of the store.

Mike Munger
Sep 2 2013 at 10:23am

Russ makes a good point, but of course it’s a subtle one. Consumers actually “want” people to bring groceries to their house, for free. But that’s not possible. Consumers “want” the milk to be in the middle of the store, for free. But that’s not possible. Given the actual menu of choices that are feasible, and given a choice between milk in the middle of the store for a 20% price premium or much cheaper milk at the back of the store, consumers “choose” milk at the back of the store. Competition can’t give consumers what they want, in an abstract sense. Because they want more, and they want it for free.

But competition gives consumers choices between different configurations of prices and amenities that are feasible. And then consumers decide what they want. What’s subtle about this is that consumers might, on a survey, complain about the milk being at the back of the store, even though milk is at the back of the store precisely because that is what people really DO want, given the costs of doing anything else. But observers who don’t understand competition might conclude (as MP did) that markets aren’t providing what people want.

If that were true, then Michael Pollan has an important business plan: start your own grocery! You’ll make a killing, bud, IF you are right. But I won’t be buying any stock in that outfit anytime soon…

Sep 2 2013 at 10:32am

Why milk in the back?

Probably because the walls have electrical outlets. Sure, modern building can be built/modified to have electrical outlets anywhere, but originally they were only built into walls.

All the other points make sense too, but the location of outlets is the most obvious to me.

Why take out is not priced lower?

This can be an inefficiency that’s not widely exploited. As was pointed out by E. Thompson, the mark up of food prices in restaurants isn’t well understood and restaurant goers don’t think too deeply about what’s in the price of food.

So they see the menu price as the food price and don’t react to the take out price as too high because the table rent is included. They are unaware they are not renting the table when they go for take out just as they are unaware they are renting the table when they dine in. Restaurant goers are largely oblivious to the table rent and think food price is just food price.

Nudging: Why did stores with increased fruit/vegetable sales have higher profits?

Because although on an individual unit basis (ie one box of food stuff vs 1 lb of produce) processed/packaged food may have higher margins than produce, when more of the produce inventory is sold before it spoils, the store sees increased margins because previously spoiled produce inventory now enters the sales/revenue column instead of the cost column (ie disposal, unprofitable self space, stocking, etc).

So each unit of produce has higher incremental impact on margin.

Great show Econtalk!

Sep 2 2013 at 11:11am

I remember the Frit-o-Lay guy saying they stocked the shelves because they’ve perfected an art, making it look perfect and not breaking any chips.

Steve Sedio
Sep 2 2013 at 12:10pm

The power of something as simple as the cost of milk. Socialists should be made to define how they would optimize the price of milk. Then tell us they can manage an entire economy, better than the free market, with a straight face.

I asked a guy that owned a small chain, why the store didn’t send me a grocery list each week. I had their card, so they had my buying history. He said they make most of their profits from my impulse buying. That could be true, but only if my wife put impulse items on the grocery list.

Many times an answer rings true, until you look into it.

As far as milk, even though we don’t understand the details (unless you are an omnipotent socialist), I can assure you it is most profitable in the back – otherwise it wouldn’t be there….

It could be the innovators lost business, because customers went to the back, and when they didn’t find milk, they left (in a rapidly changing environment, and organism will do as little as possible to maintain a former way of life – from physical anthropology about evolution).

It could be as simple as we want all our stores laid out the same, and we already have 1000 stores (going back to 1947).

That said, Walmart doesn’t have milk on the wall (at least not here), because the food based groceries are in the middle of the store.

About lower prices for take out. Can anyone explain why a dinner for 60 people is so expensive? Only 2 or three meal choices, higher priced, gratuity added to the bill (no matter the quality of service). And, the bill has to be $X, even if half the people don’t show. The restaurant has none of the unknowns that contribute to their costs, yet their they charge far more. Maybe we linger longer – now we see the cost of renting the table….

Pietro Poggi-Corradini
Sep 2 2013 at 12:18pm

Maybe the reason take-out is not priced lower is because there’s a limited amount of burners in the kitchen. Too much take-out activity would lengthen the time people wait for food at the table.

Great podcast, but, no mention of the high prices in airports?!?!

Pietro Poggi-Corradini
Sep 2 2013 at 12:29pm

I asked my eleven-year old daughter why is it that milk is stocked on the back of the store. Her answer: “because nobody wants to feel cold when they first enter the store.”

Mike Munger
Sep 2 2013 at 12:39pm

Pietro P-C: Your daughter is on to something. Stores always have flowers and fresh produce first.

The flip side to “people don’t want to feel cold” is that the coolers would have to run harder near the door.

As for airports, that just seems like the old salt monopoly as a revenue raiser. The airport sells local monopolies, and then people don’t compete on price because they have to cover average costs. It seems to me that there is very high turnover in some of those shops, because they can’t cover their fixed costs. Prices are too high, but lowering their prices wouldn’t help. Even though the prices are high, demand is inelastic. I don’t go the airport to shop for shirts. I only shop for shirts if I forgot one. And then I’m going to buy, even if it’s expensive.

Sep 2 2013 at 1:10pm

Why do most restaurants charge the same price for take out, when that customer is not renting a table?

Because the dine in customer is more likely to order beverages. The additional sales, at a much higher margin, cover the table rent.

Scott Small
Sep 2 2013 at 1:36pm

There are several additional aspects of the take out business that are not obvious but crucial to note.

Firstly take out customers de facto pay a lower price. Typically they do not purchase wine, deserts, and are less likely to purchase a drink. Restaurants operate on the total profit of the customer and not individual items.

Second is that take out orders on average tend to come at lower average volume. There are many more single “seat” orders for obvious reasons. That makes the transaction costs relative to the items purchased slightly higher.

Third is that even take out customers are actually still occupying a “table” in some sense. In many restaurants that offer takeout there is a dedicated room and staff to handle these orders. Even though someone is not physically sitting in the restaurant that inventory is still being occupied.

Fourth the profitability of the customer is also somewhat diminished by the loss of opportunity to up sell the customer consistently through the dining experience.

I think the core point that is missed is that menu prices tend to reflect the totality of the business model for the restaurant, not individual pieces. The installation of a take out service may have feedback effects that alter the price of the sit down business.

Sep 2 2013 at 2:00pm

If positioning on a wall is necessary, why must it be the back wall? Why not the wall nearest to the door or on the sides of the store?

On coffee- is not the issue of refill also a factor? The costs of beverages must factor in refills. How might the cost of cheesecake be affected by an “endless slices” feature.

Can we have a Econ Talk on the economies of buffets?

Sep 2 2013 at 2:05pm

One reason I can think of as to why takeout is not cheaper is because the restaurants make more money on drinks if you eat in. I can imagine that the profit margins on pop are very large. If the restaurant sells liquor and wine anyways, they will make much more money against the costs of their liquor license if you get drinks there, instead of taking your food home and buying your own. The cost of drinks is what covers the table rent.

Jim Ellison
Sep 2 2013 at 2:05pm

There is a price differential of 15 to 20% for take out: no tips (at least in the US).

Luke J
Sep 2 2013 at 3:12pm

Many groceries have freezers and coolers (reach-in and with doors) right in or very close to the center, so I am not convinced that energy conservation and outlet convenience are complete explanations.

Cowboy Prof
Sep 2 2013 at 4:03pm

Best podcast since the great North Carolina hurricane that resulted in ice trucks being chased away.

Two points.

1. As per grocery loyalty cards. Mike noted that there are cashiers at Piggly Wiggly that would scan their own cards if you forgot your card. Russ responded that this would be beneficial in promoting “friendly service.” Indeed, but here in WA a few grocery store chains are cracking down on that because of ye-olde principal-agent problem. Safeway loyalty cards can earn you gas discounts at their gas stations, and it is possible to get upwards of $1 per gallon off. Some stores were worried that cashiers were scanning their own cards to get the gas discounts. One might think you would be indifferent as to who received the gas discount if the customer wanted their privacy yet also get the discounted price, but apparently this was creating some behaviors among agents that the principal wanted to stop.

2. As for the same price of take-out food. I think one of the explanations proffered was partially correct — the time it takes to cook the take-out food uses valuable cooking time that could be used to get food faster to the dine-in customers, and hence turn around the table.

But, there is also the issue of customer expectations. If you purchase lower-priced take-out food, enjoyed it and then said, “Hey, let’s go get a sit down meal there next week” only to find that the meal is 25% more expensive, you would be likely to only to take-out food. (Most customers don’t think about the cost of the table.) Over time, this may lead to fewer people dining in and fewer used tables, which are a sunk cost.

Finally, you do pay more for dine-in food in terms of the gratuity. I don’t know about anybody else, but I rarely tip when I get take-out, but do tip when I dine-in. So you do pay more for dine-in, but it just might be distributed differently among the different component costs of the meal — food, table space, service.

Tim M
Sep 2 2013 at 4:18pm

I recently opened a wine store with a two door refrigerator, and we planned on putting it next to the sales counter. Once we got it and plugged it in, we realized that the noise level was a bit too high for our own comfort and for our customers, and so we relocated it to the back. No one ever complained about having to walk to the back of the store for chilled wine.

Greg G
Sep 2 2013 at 5:00pm

Russ and Mike

You guys are obviously having so much fun here that it makes it fun to listen to. Nobody is better than you two at explaining how market pricing works to benefit consumers and constrain the greed of sellers.

I would love to hear a podcast where you guys tackle the topic of the pricing of the management of retirement accounts by the financial industry. This is a very interesting issue. Vanguard will charge you five basis points to manage a Total Stock Market index Fund that will outperform the vast majority of the equity based mutual funds in the average IRA, 401K or 403B.

In many of these cases people pay thousands of percent more in fees for worse performance. The companies charging more do everything they can conceal these fees but Vanguard does everything they can to publicize the difference.

When I was a local retailer the biggest premium I could get for offering better selection and service was about 10%. When it comes to managing retirement accounts many companies can charge immensely larger fees for worse worse performance.

This topic cries out for a podcast!

Sep 2 2013 at 5:27pm

Here in London, a LOT of restaurants in high foot traffic areas have one price for eating in and another for takeout. Why? Well, it really is the case that if you sit down they have to turn away another potential customer.

In the US, where land is more plentiful and building restrictions are not so onerous, that is far less often the case.

Sep 2 2013 at 5:37pm

Was a study done to measure the effectiveness of NYC Health Dept nudge sugar adds?

Sep 2 2013 at 7:15pm

In continental Europe, UHT milk (that does not need to be cooled) is a lot more popular than in the US. It is my perception – without having tested this in any systematic way – that the milk is mostly in the middle of the store, at least in Germany. The pasteurized (cooled) milk is in the coolers which are located at the end or the side of the store (towards the outer walls). This seems to support Prof Munger’s argument…

Sep 2 2013 at 9:55pm

At last! An EconTalk podcast where I feel I might know more about the topic than Russ and his guest. First, Russ is right about why milk is in the back of the store: the milk comes in through the back door and is stored in a cooler in the back room. The dairy case is placed in the back of the store to be near the storage cooler because, as manager after manager told me back when I used to stock the dairy case, milk loses one day of shelf life for every degree it gains in temperature (I don’t know if this is actually true, but I can say for certain that when I worked in a grocery store, the managers believed it). So, you have to get the milk from the storage cooler to the dairy case fast (I worked in a grocery store in the late 1970s, before the days of dairy cases that could be loaded from the back).

However Pollan also has a point. Today, I work in marketing and I can tell you that it’s just good merchandising to place high-volume, highly desirable, high mark-up items along the route from the front door to the dairy case. So, it’s not that the milk is placed in back as part of a conspiracy to make us buy those other goods, it’s that the merchant makes a virtue of necessity by placing those other goods along the path buyers must take to get to where the milk ought to be.

Why, then, can ice cream be placed in the middle of the store? Here, the answer requires a little bit of knowledge about grocery store storage systems. In that back storeroom is a zero degree cooler, in which all of the frozen goods are stored. Those goods don’t start to thaw until they get to around 32 degrees. So, ice cream can be removed from the zero degree cooler, carted out to the center of the store and placed in the display freezer before it starts to lose shelf life.

Pollan is certainly right about premium shelf space being at eye level. Not only do food suppliers pay extra for that space, they’ll also engage in less than ethical practices to get it. For example, I currently have a client whose product is sold in grocery stores. In this year’s placements, a competitor got my client’s products moved to a less attractive shelf by taking a store chain’s buyer to the Super Bowl. My client clearly had the better business case for the premium placement – his was the best selling product in its category. He did not, however, have tickets to the Super Bowl. So, he got the boot (you can believe, though, that he’s closely monitoring the scanner data to show the buyer that the other product doesn’t perform as well in the premium space as his did, so we expect the loss to be temporary).

Finally, I noticed what I thought was mostly just a disagreement over semantics. According to Russ, Pollan asserts that supermarkets are trying to maximize profit but, in fact, they are actually trying to maximize revenue. No, they’re trying to maximize profit (that’s what good managers do). However, in a low margin business, the best way to maximize profit is to maximize volume (revenue). So, again, when viewed from the proper perspective, Russ and Pollan are both right.

Thanks for a fun podcast. As always, I’m looking forward to next week’s edition.

Sep 3 2013 at 1:03am

My thinking on why takeout menus are often the same as dining in.

The main thing in my mind is that the real “time rental” foods like wine,coffee and dessert are not ordered off of takeout menus. Therefore they want you to order the sitdown food, as you could then buy wine, coffee and dessert as well. If the takeout menu was substantially cheaper, people would be incentivized to avoid buying wine/coffee/desert inside and instead get them cheaper at home.

Mike Tolhurst
Sep 3 2013 at 1:51am

I agree with Cowboy Prof. With conventional gratuity at 15% that is a substantial mark-up for dine in, even though it is not printed on the menu. I think suggestive evidence that this accounts for the hidden difference is the posters from the UK who do note that there is a difference in price. That makes sense because the tipping conventions in the UK generally are different than those in the US. (Lower-to none.) So you have different price lists where there is no gratuity (UK) but the same price list where there is. (USA)

Sep 3 2013 at 11:09am

Supporting David on milk temperature vs shelf life :

I bike everywhere, including to the supermarket.

Milk would never be good for even as long as its sell-by date, until I started always buying frozen vegetables with it and packing the milk so it’s surrounded by the frozen veggies.

Now the milk is good almost two weeks past its sell-by date, just by keeping it from warming at all during a half hour bike ride.

Roy Haddad
Sep 3 2013 at 11:55am

Munger gives a convincing explanation for something that just happened recently:

A few days ago I went to a nice restaurant, and was so impressed with a wine I had there I decided to buy a bottle to take out, and was surprised when the waitress informed us that in that case they take $10 off the menu price. Like a reverse corkage fee.

Sep 3 2013 at 12:03pm

Regarding the grocery store layout. I am wondering if inventory turnover have an major impact the decision. If the grocery store can turn over the items faster than the account payable period, they will make more money with less investment. So an item that occupies the best location should have high margin and high turnover. I am not sure what a store would do if they have to choose between high margin and high turnover, they may even choose high turnover given its will be more capital light given the supplier is actually paying for the inventory (and even stocking the shelf as mentioned).

As some one already mentions, maybe restaurants know they can up-sell customer on extras such as desert and the tip they would get. So maybe the sit down price is actually artificially lowered because of this.

Sep 3 2013 at 2:26pm

Haven’t even gotten to the part where Mike comes in, but am thoroughly enjoying this one already. Good job Russ!

Bill Turnier
Sep 3 2013 at 5:14pm

First, refrigerated and frozen goods are commonly placed at the back and along walls to make it cheaper to outfit the store with freezers and refrigeration. Second, milk is especially difficult to move from the refrigeration units in the store warehouse area to the customer area refrigeration. It is very heavy for employees to move (I worked in the dairy department when I was in high school).

Supermarkets are well run enterprises that do not subscribe to any economic orthodoxy. Some things are placed for profit such as the candy at checkouts. Most of that is there for the kids to grab, not the adults. Other items are placed fir access and shopping convenience of adult shoppers. I wish Russ and Mike had worked in a few as teenagers. The management at ours in Teaneck explained much placement to any employee who showed an interest.

I also wish Mike and Russ had shopped at Trader Joe’s. It is an exciting different model. The average revenue for a TJ is $1700 per square foot. It is $450 for Harris Teeter, $650 for Kroger and $950 for Wholefoods. The average supermarket carries close to 50,000 different items; TJ has 5,000 on average with house brands overrepresented. Yet, their stores are packed with people who are happy with limited choices ( and cheap decent wine). TJ like Costco and Wholefoods also pay their employees much better than their competitors and one consequence is that they ate mkre helpful and pleasant with customers. There are several different models of grocery stores competing for our dollars I wish Russ and Mike had discussed some of these different models besides 7-11 and the conventional model.

Michael Munger
Sep 3 2013 at 5:28pm

Dear “S”: That’s pretty funny.

An undergrad in one of my classes listened, and then told me I should just change me name to Ed McMahon.

I’ve been practicing saying, “You are CORRECT, sir!”

The implication being that, as we all know, Russ is Johnny.

I think that means that Alex Tabarrok gets to be Doc Severinsen.

Sep 3 2013 at 6:02pm

You have a relatively heavy, perishable item with a high turnover rate. I had always assumed the milk was in the back because it’s easier to replenish the shelves from the back of the dairy case. Just imagine what the cost of milk would be if you had to send a couple of workers out into the dairy aisle every few hours to shift all the milk around just to add new stock to the back of the shelves.

If you had shelf stable milk, you could place it on a display up front but then Pollan would probably see the UHT pasteurization as some sort of evil to be avoided.

Robert Promm
Sep 3 2013 at 8:04pm

All this talk about cost. Cost is irrelevant, mostly. The price is the market clearing price regardless of cost. Cost is only an issue as to whether you will offer a product or not. A high-end restaurant offers take-out at the same price as their eat-in product cuz they are best trying to hit the market clearing price for their product. Am I missing something?

Sep 3 2013 at 8:22pm

What do you guys make of casinos?

I was always told that casinos engage in lots of shady business practices to keep gamblers playing and therefore losing money. They pump in oxygen to keep you awake, they don’t put up clocks so you’ll lose track of time, they make it hard to find exits, they pump you with alcohol to impair your judgment, and so on.

First of all, is any of this true?

If it is true, would you defend these practices as strongly as you do the supermarkets’ right to push less healthful food on consumers in the interest of profitability?

Apparently people enjoy fat, salt and sugar as much as they do losing money over cards. Libertarians will say, personal choice! But if 1/3 of all Americans are overweight, does that not impact our health-care costs nationally? Do compulsive gamblers not endanger their own financial futures and the well-being of their families? Where do we draw the line on personal choice and say, in effect, OK you can drive a motorcycle, but you have to wear a helmet.

No man is an island, entire of himself. Our personal choices frequently effect others in ways we fail to appreciate.

Sep 3 2013 at 9:16pm

With regards to the question of carryout prices being the same as dine-in prices for most restaurants: I was surprised that neither Mike nor Russ offered up the possibility that people who take advantage of the existing offer might assume that all dine-in costs, such as service and time, are addressed in the gratuity. While this is certainly not the purpose of the gratuity, it may be a common misconception. Since one is less likely to offer a gratuity on carryout orders, there is at least the perception of savings.

Sep 3 2013 at 10:03pm

Grocery store owners sell what their customers want and also sell premium shelf space to companies with products.

Companies with products get best spots on shelves when they pay for it and also their items sell! They can pay indirectly with branding/advertising and/or directly to the store.

Mainstream products (most popular items) get most space and best visual spots if the owner is smart because those items need to be found easily, sell fast with good margins and they never want to run out.

If a company is willing to pay a ridiculous amount up front – you may see crappy items that don’t sell in a highly visible spot on the shelf taking up lots of room. Most grocery owners learn quickly that this is a terrible strategy though and it’s not necessarily a good idea to receive upfront payments for items that don’t sell and just take up space.

Chain grocery stores don’t have much say in the layout of store in general. If they are smart, they pay people like Herb Sorensen to do the design.

Nick Payne
Sep 3 2013 at 10:07pm

Some of Pollan’s work like the final section of his popular book where, recounting his first time hunting, he acquires “the hunters eye”, makes me cringe with embarrassment, but his writing elsewhere highlights many issues I think an economist should appreciate.

Principle among these, his excoriation of corn subsidies and the resultant cheapness of unhealthy food additives and general market distortion seem laudable.

In the Podcast, I also heard a slight dismissiveness of behavioral economists’ endeavors to combat obesity, which have shown significant potential to influence health decisions for the better. I see some humor in the mirrors on shopping carts, but healthcare costs, costs that arise substantially due to obesity and diet/lifestyle related ills, are no laughing matter!

Sep 3 2013 at 10:13pm


Thats hilarious!… as usual

Evan Downie
Sep 4 2013 at 5:22am

I’m pleased to able to contribute on this issue with some basic practical information on commercial refrigeration systems.

Though I’m not a designer, nor engineer, I do install automated control systems in large commercial properties. There are a variety of factors in placing a refrigeration system such as thermal load (windows and other heat sources), humidity, and other environmental issues. But in my experience there are basic determining facts surrounding placement that I will try to briefly sum up.

Most commercial refrigeration systems tie together to one location. Because of this all displays and storage units are placed close to that central location. Which reduces runs of lines and cables, and simplifies labor. This is the first factor of placement, to reduce initial cost. Installation of refrigeration systems represent one of largest initial expenses in this type of construction. The second factor is accessibility/serviceability. Very important in reducing long term costs. The third, and most important, factor is safety. Safety above all determines where displays are located.

Refrigerant as well as being an environmental hazard, is also a poison. It travels at extreme temperatures, under pressure, through the system. So the goal is for the lines, and equipment, to be as isolated from the general public as possible, but also accessible to service personnel in the event of a leak. Thus the back of the store is the most practical, and safest, location for the support equipment to retail milk.

Shayne Cook
Sep 4 2013 at 7:55am

Commenter Robert Promm (above) is the only one I’ve read or listened to here who gets this right.

The price of anything/everything that is available (milk, takeout, Boeing 747s, …) is based solely and exclusively on what people are willing to pay for it.

The availability of anything/everything is based solely and exclusively on whether the combined costs of production, transportation, warehousing, selling, etc., are less than the price people are willing to pay for it.

Since the central topic here is ‘milk’, one might ask why it is that it has become extremely uncommon to see “milk men” and milk delivery trucks beavering around the suburb neighborhoods in the mornings. That used to be an incredibly common sight back in the 1960s. Could it possibly be that the reason that isn’t a common service anymore is due to the fact that the costs of providing that service – in addition to the costs of the ‘milk’ – exceeds the price people are willing to pay for it?

One quibble with the second-to-last sentence in Robert Promm’s comment: “A high-end [or any other] restaurant offers take-out at the same price as their eat-in product ….”, solely and exclusively because there are obviously consumers who are willing to pay the same price for take-out as eat-in.

Sep 4 2013 at 8:33am

Such a lively conversation over milk (that isn’t even spilled).

In “Defense of Michael Pollan”, I think he simply found facts that supported his point of view, and didn’t research it further. That’s the danger of observational studies vs. the empirical world of retailers. It might sound good on paper, but how will the customer respond?

My take on the milk is that you want to pick it up last and then add the ice cream even later to minimize thawing. No one wants to pick up milk and then snake through the aisles as it warms up. When I was a bagger at a grocery store, we were taught to pack all the cold items together. These days that “skill” seems lost.

Evan Downie
Sep 4 2013 at 8:47am

There was a question posed about placement of ice cream freezers. I have no direct knowledge for that product. However stand alone units can have self contained cooling systems, and in those circumstances would be more open to strategic placement.

Russ Roberts
Sep 4 2013 at 10:12am

What is missing from some of these comments is the interaction between price, cost, subjective value (or willingness to pay) and competition.

Competition drives price toward cost. If that cost is above what people are willing to pay, the product will not be available.

But the fact that people are willing to pay a lot for say sit down or take out or milk usually won’t explain the price that you actually see. You have to take into account competition. What you actually pay is usually well below what you are willing to pay because of competition.

Sep 4 2013 at 10:51am

Here in the UK there is a systematic discount for takeout at all places that offer takeout. This has always bothered me because the negative externality of the clean up is handled by the city rubbish services. Maybe the rental side is more important than the cleanup side however; it is considered unethical to sit at the outside seating if you have ordered take out, although you would bus your own table.

Shayne Cook
Sep 4 2013 at 11:53am

To Russ Roberts:

I suspect you may be making some incorrect assertions, and I suspect that is based on a bit of imprecise use of terms on your part – to wit:

Competition = Availability of Alternatives (and that ONLY.)

The “subjective value” you reference in your first statement has, as a fundamental component, availability of alternatives. Competition provides alternatives, and is therefore a core component of “subjective value”, not an “and” in addition to and separable from “subjective value”.

I dearly love competition, as you apparently do. But the benefit I, as a consumer, gain from competition is greater availability of alternatives (choices), not necessarily or inherently lower prices. And I suspect it is provable that competition has done nearly as much to reduce costs-of-production as it has to reduce purchase-price – via the phenomena of economies of scale, if for no other reason.

Regards “prices/costs”, I learned in an econ class once that opportunity cost – not “dollar price/cost” – is the only thing that is relevant. I’ve found that concept critically important, robust, and unequivocal in both academic work and “real life” (operating successful businesses).

Zach Adams
Sep 4 2013 at 12:24pm

Relevant to the discussion of “table rent:”

In my younger days, I rode the Greyhound bus a lot all over Texas. Within a few blocks of a bigger city bus terminal, like San Antonio or Houstonfast food places ALWAYS had a sign indicating that you could only sit for an hour, and usually only if you bought food (not soda/coffee.) The increased demand for table space causes them to be a lot stingier in collecting the rent.

Sep 4 2013 at 4:16pm

On why milk is in the back, lets not forgot that it’s not just milk in the back. It’s milk, half and half, cream, buttermilk, sour cream, yogurt, and a whole host of other dairy products that need to be refrigerated. The decision is not “does milk go in the front or the back”, it’s “does milk, and all this other stuff, go in the front or the back”. So the front of the store gets a small stand alone fridge with some milk for those truly in a hurry just for milk, while the bulk is in the back, which makes more sense for the dairy section overall.

Eliot St John
Sep 4 2013 at 11:51pm

How do the results of the nudging studies fit in with the basic analysis that grocery stores are providing consumers what they want because of competition? If what consumers “want” can be influenced by apparently superfluous things like placards and mirrors, the idea of consumer sovereignty seems harder to apply. This doesn’t mean that supermarkets are deliberately manipulating consumers into buying unhealthy food. It does suggest that consumers don’t necessarily have an immutable preference for unhealthy food that would require unduly intrusive government paternalism to change.

It sounded like Russ and Mike didn’t like the idea of stores having mirrored carts and whatnot. But why not, if the stores do it voluntarily because it actually helps their bottom line? If you think enough people share your preference to not see their chin fat, why not start a competitor that will charge a premium for mirror-free carts?

Sep 5 2013 at 5:13am

I enjoyed the podcast.

A couple thoughts that may have been mentioned, but I didn’t catch.

Milk is is big and stores want large inventories so they don’t run out or they will go the way of the basketball-less Venture. Large coolers are needed for a large inventory of a large item.

Second, it’s usually in the back on the side where you finish shopping for the reason David mentions – temperature sensitive shelf life. Being close to the last thing you put in your cart minimizes the time between grocery store cooler and home fridge. That’s worth something.

Sep 5 2013 at 10:23am

Turnabout is fair play:

Michael Pollan is just trying to manipulate reader into reading his writing.

BTW Milk yuk I never drink the stuff. IMO it tastes OK for calves not for adult humans.

Evan Downie
Sep 6 2013 at 7:12am

Although this issue has generated a lot of debate. The fundamental reason for milk being in the back of the store is; that is where the refrigerators are.

Milk placement may corolate with interesting economic conspiracies, however the cause is a series of engineering banalities.

I suppose there is wiggle room for where, in the coolers, milk is placed.

Dave Lochtefeld
Sep 6 2013 at 8:55am

I really like the podcast in general, but have to say I was not impressed with this one. It was just 2 guys speculating/ranting. I can get that at the local bar.

Why not include Michael Pollan (or someone with his views) on the show to defend himself, or a grocery store manager to explain why the store is run the way it is.

Sorry to be negative. I generally love the show. The guests and ideas are very interesting. Like the recent show on how violence, or the threat of violence, effect economic policy and growth. I never thought of that connection before.

The Milk episode reminds me of a previous episode where the guest talked about how liberals, conservatives and libertarians have different vocabularies and talk past each other. Michael Pollan is the liberal and sees the world as oppressor vs oppressed. Russ and Mike are the libertarians seeing the world as freedom vs tyranny.

Sep 6 2013 at 9:37am

There is another price differential for take out vs dine in in some places. Where I live, sales tax is charged on all items ordered for dine-in, but only soft drinks if ordered for take out. Thus at the drive through, my wife’s $0.99 tea is $0.99, while my $0.99 soft drink is $1.06. I know this extra does nothing for the retailer, but it does mean consumers can save another 7% on take-out.

Matthew Fieger
Sep 6 2013 at 10:27am

Thanks for another great podcast Mike and Russ.

This entire debate seems to be based on one assumption – that businesses are always trying to maximize profit.

I am not so sure that is true. A business could maximize profit by maximizing risk, but maximum risk is not what the owners/investors want. If the grocery store were to maximize profit in every possible way, including so called deceptive methods such as manipulative product placement, they would also be maximizing risk that a new market entrant could be more competitive by offering lower prices.

Perhaps the business is not trying to maximize anything, but rather trying to achieve stability by balancing profit and risk.

This could explain those befuddling moments when the business has a clear opportunity to maximize profit but chooses not to do so. Russ gave some examples of that – when the cashier volunteers a coupon or when they swipe their discount card when you don’t have one. It could be that the reason those moments seem irrational is because the theory is wrong.

Sep 6 2013 at 3:12pm

For those who are interested, this will give you an idea of just how much thought and effort goes into figuring out where and how to place products on store shelves:

Retail Next Learning Labs, BTW, is not some out-on-the-fringes weirdo consulting group. It’s part of Rock Tenn, the largest U.S. supplier of merchandising displays. My point is simply to suggest that Pollan is quite right to suspect that your grocer is working hard to get arrange his store so that it sucks as much cash out of your wallet as possible (without offending you, of course).

Sep 6 2013 at 3:16pm

Let me start by saying that I have listened to virtually every EconTalk podcast since I found you guys back in late 2006. I’m a big fan and very much appreciate all the hard work you put in to create this for us.

Given your wonderful track record, I was disappointed in the Milk podcast. I hope not to come across too harshly, but I do want to critique this one.

I am startled by the naïve assumption expressed early on that firms can be relied upon to cater to their consumers and never attempt to take advantage of them. (Your dismissal of the possibility that grocery stores might place higher margin items at eye level to maximize profits is based entirely upon this assumption.) The idea that competition will enforce this behavior is laughable. In fact, you yourselves laugh at it later in this very same podcast when you discuss the difficulty discount supermarkets have in opening new locations.

I am all for libertarian ideals. But please don’t ignore realities. Economists focus so intently on equilibrium pricing that they sometimes forget that vast swathes of our economy are in disequilibrium, sometimes for protracted periods. Indeed, many firms strive mightily to create brief dislocations during which they can extract monopoly profits. Witness the exorbitant pricing of text messages when they first came out–across all cell carriers. Or the efforts of your local Giant store to keep out Harris Teeter or Super Walmart.

On a less technical note, while the tone of most of your podcasts is very professional, this one really started to devolve toward the end into something closer to a couple of college dudes hanging out in their basement. Java the Hut? Lard Butt? Lending library? J Lo? Really?!

There was one jewel hidden in there: the suggestion that policy makers should focus solely on the consumer, completely ignoring the desires of the producers. That’s an intriguing idea and one I would love to hear you delve into.

Sorry to be so critical. But I really do care about your product. Otherwise I would have just hit delete and moved on. Keep up the good work.


Andre Kenji
Sep 7 2013 at 3:01am

Mike and Russ are right that retail, specially grocery stores, is a really tough market, because it´s highly leveraged. On the other hand, that just means that you have to maximize every profit possible, and that´s not necessarily means being good to the customer.

On the other hand, foreigners knows that in many countries it´s *easy* to spot produce in supermarkets, because that´s high value items that can easily deteriorate. If produce is hidden in the US, that may just mean that there is little demand for them.

Sep 8 2013 at 2:46am

>> You have to ask another set of questions: Why is the store so clean? […] Why do they hire friendly checkout people? […] Why do they bother being open 24 hours sometimes […] Why do they give away samples?

Have you ever been in an Aldi?

joe B
Sep 8 2013 at 11:43pm

i am disappointed to not see (in the week of Coase’s death) a TRANSACTION COSTS explanation for take out costs prices.
let’s start with the premise (mentioned here) that take out prices will be lowered only if there is competition.
and I would argue that because of the hassles of lowering takeout prices (printing two menus, explaining things to customers etc) there is a transaction cost for offering this lower price
and (empirically, i would infer) the costs associated must be high enough that it makes no sense for restaurant to offer it
this is a falsifiable claim: in the future, when there are no printed menus (iPads etc dominate), the costs of two menus will drop such that there will be more restaurants offering dual prices

Sep 9 2013 at 7:06am

I am not sure if anyone has already commented on this point, but I recently saw an interesting case where take out was actually charged less than dine in at multiple restaurants in London. Basically, in bakeries and cafes where people buy lots of goods to go, there is usually a higher price for dine in than take out, which was about 10%-20%. This price difference is clearly displayed in front of the foods. The price difference is because a VAT is applicable to food that is eaten in and not for food that is taken out. For some small items (chips, drinks, etc.) there is not a difference, but for the main items (sandwiches, salads, and soups) the difference applies.

I know that this difference is not because of some calculations by the cafe, but it might provide an interesting case to study. For example, after seeing this a couple of times, I abused the system. My wife and I purchased several items. When the cashier asked whether it was dine in or take out, I told her that one items was dine in and the rest were take out. We then sat down in the corner, which was difficult to observe, and ate all the items. Ultimately we saved about 3 pounds on a 16-17 pound purchase. I guess that I was equally interested in saving the money and seeing what would actually happen.

Sep 10 2013 at 7:17am

I firmly agree with Tom. Econtalk is fantastic and I really enjoy the podcasts. However this podcast delved into the land of fantasy, that is, the land of perfect competition. It is easy to forget as an economist that perfect competition doesn’t exist in any form, let alone in the supermarket industry,

Stores do put Milk in the back because it causes people to walk through the shops and look at other products. The fact that convenience stores put their milk out the front is no proof at all that competition is fierce and supermarkets are “putting consumers first” by their placement of milk. Convenience stores put their milk out the front because they sell convenience – and the milk is 2-3x the price for that convenience.

Supermarkets are more of an oligopoly market, and once you arrive in the store you are there because of lower prices or lower search costs or a loyalty program or a combination of all three. Once you arrive at the supermarket the cost of leaving because you do not like the position of milk is very large and so you are now in the hands of the supermarket. Even Costco – which is a huge store keeps milk at the back of the store with other staple items.

You only have to look at a few examples. And I shall give two from my country, Australia. We have roughly 4 major chains of gas stations and while in the US you have enjoyed pay at the pump for many years, we have never introduced it here. Why? Because gas stations prefer you to walk through the store as they make high margins on the food and drink items, while margins on fuel is low. And where is the checkout – it is ALWAYS at the back of the store. Whoever comes out with pay at the pump may temporarily gain some market share in the short term, but once the others respond everyone loses profit in the long run. Firms understand this.

We have two retail giants in the supermarket business, and they compete fairly fiercely. Recently they had a price war on milk, but the milk still stayed at the back of the store. The two chains were making losses on each purchase of milk but it did not matter as it got you in the store.

As far as product placement goes the two chains began promoting their “own” brand items on which they enjoy higher margins of profit by placing them at prominent places in the shop. Because they control a large portion of the food market there was little upset suppliers could do, except for Coke who supplies a huge majority of Australia’s (and the worlds) favourite drinks. Faced with losing prominent space they simply stated they would not supply a chain with special offers if they did not get the prominent shelf space. Because of the demand for their product they were able to dictate terms. Supermarkets will do everything they can to attract customers, that is true, but they also use their market power to maximise profits and maximise sales to consumers.

All this to say, that we live in a world or imperfect competition and heterogeneous goods – it seems that in this last podcast that simple fact seems to have been set aside for the simpler world of pure competition – a world that we do not live in.

A final word on the items near the checkout. They place candy and magazines because they are impulse buys and highly desirable with a good markup. The other items are not there as a public service but a last attempt to sell you low priced items you may have a need for regularly. The supermarket is neither your friend or your enemy, it is a firm trying to maximise profit, which sometimes works in your favour, and sometimes doesn’t.

Sep 10 2013 at 3:51pm

A response to the question of why take-out prices are usually the same as sit-down prices: The total bill when sitting down is usually higher than take-out bills, so the restaurant wants to encourage sit-down customers by providing “free rent” as it’s called in the podcast.

The reasons sit-down bills may be higher are because the food is for a larger group (averaging probably 2-4 people vs. 1-3 with carry-out – think of a large birthday party compared to a husband picking up Chili’s on his way home from work).

Another reason I can think of is, and seems to make even more sense, is that the restaurant has a higher chance of upselling you when you come in and sit down, including drinks, desserts, extras, etc (plus tip – which the restaurant doesn’t see directly but decreases the amount they have to pay their wait staff).

Russ Roberts
Sep 10 2013 at 9:43pm


This has nothing to do with perfect competition. That’s an extreme case. At the other end is monopoly. Neither describes the world of supermarkets. But supermarkets do compete. The question is how much. The more they compete–and the easier it is for new entrants to come into the market–the more customers get served by both price and quality.

You assert that the milk is in the back to make people walk through the store. If you listened to the podcast, you heard a bunch of other reasons that might explain the placement of milk without presuming market power of the supermarkets allowing them to exploit customers. There are other reasons given in the comments for the placement of the milk that don’t rely on consumer exploitation. I don’t know if I’m right. All I’m suggesting is that competition encourages firms to do what customers want rather than what might maximize profits in the absence of competition.

The real point I was trying to make is that if it is costly for the milk to be more conveniently placed, then customers want the milk in the back.

How competitive is the supermarket world in reality? Tough question. They do seem to constantly try to improve, making nicer stores with more selection and new services. Do you think grocery stores have gotten more profitable over time as they’ve introduced new ways to make their stores more attractive? Or have they been forced to share their gains with customers in the form of lower prices. I think the latter but I’m happy to hear evidence to the contrary.

I like your Australian examples. There may be very little competition in the gasoline market there and that may explain the lack of pay-at-the-pump. Or there may be costs that are not obvious in Australia that don’t apply here. That’s the value of presuming competition. It helps you look for other factors you may have ignored.

Sep 11 2013 at 12:00am

Thanks for your reply Russ. I really do find your Econtalk podcasts of huge benefit. So thanks for the work you put in! I agree that there may be other explanations as to why milk may be in the back, and it could be a combination of many issues. I guess I felt that during the interview the message being sent was that we didn’t have to worry about such “marketing tricks” because as long as the market is relatively competitive consumers best interest would be served. I agree that exploit is a very harsh word, I think supermarkets are just doing what they can to sell more. I also agree that prices over the years have gone done and that is due to competition, my argument is that they are still looking for extra profit at the margins.

part of the fuel market is actually linked to the supermarkets in a strange way in Australia. A few years ago one firm started offering petrol discounts if you shopped at their store. This was very successful and so the other firm bought Shell Oil’s retail business in Australia and offered similar discounts. Our competition watchdog is concerned that this could lead to less competition.

Thanks again for the great work and I look forward to more Econtalk!

Allan Stokes
Sep 11 2013 at 9:43am

The episodes I find most frustrating are the ones where the probing questions are flying at half mast. Where’s the discussion of whether “back of the store” has ever actually been a big deal to anyone, and if so why? Polio? Diabetes? Fallen arches? The more frustrating mystery, I think, is why the empty parking spaces are located so far away from the entrance that I’m already dreading having to walk to the back of the store by the time I get to the maw, even though it’s such a small distance further.

If the flag in this discussion had been flying a little higher, there might have been some discussion about why so many grocery stores are engineered to make it nearly impossible to get from point A to point B at any decent walking speed (which some of us can still attain) without flinching inside over taking out some flimsy cardboard product display stuck into the middle of a grocery aisle that wasn’t all that wide in the first place. Pardon me, ma’am with the two screaming children, if you just moved your cart six inches I could proceed today around this Lindt outcropping.

What else is on that cardboard display that creates a choke point when any shopping cart comes to rest within three feet on either side? Sometimes potato peelers (one can never have too many). Far more often potato chips. The same darn potato chips you tripped over getting into the store, and for which there’s already an entire aisle devoted to hand-fluffed pillows of salty goodness. I know from a previous, excellent episode that these mini displays are there to create “product impressions”. They are there so that when you finally get to the chip aisle you think to yourself “chips seem to be on my mind a lot lately, I must really be hankering for these salty treats” and reach out to grab a bag.

The upshot of these product impression mini displays is that actually getting to the back of the store with any pedestrian efficiency has become a royal PITA. There are days when the grocery store is a little more crowded than average where I would almost pay good money to be able to trot with my cart down one side of the aisle in a straight line, come what may to any misplaced enticements. I ended up becoming mainly a Costco shopper where the cart lanes are wide and unhindered by craggy Frito Lay and Uncle Ben’s hooking for Snow White’s ankles (for the most part). Hoofing to the back of the store is a minuscule consideration compared to the impediment of shopping aisle arteriosclerosis.

Come on guys, this is blatant psychological warfare designed to wear the consumer down until the consumer ends up with a cart full of lizard-brain. Our slow brain knows the value of things, our lizard brain doesn’t. I have a lot of trouble believing that neither Russ nor Mike has ever heard of Kahneman or Baumeister. Or is it that you have you firmly concluded that people who weigh 300 unhealthy pounds actually, at some level, want to weigh 300 unhealthy pounds? No, our preferences depend on what brain circuits are activated in the moment. In many people this flip-flop between now and later is never properly reconciled, but is instead medicated with a motile elbow.

I’m aware that the lizard brain is sneaky and that many people subconsciously seek out the toxic stimulus environments which contribute to lizard-brain ascendancy and gratification even after explicitly stating the opposite goal (especially after stating the opposite goal).

In a different venue this would be referred to as consumer/vendor co-dependency or perhaps as institutionalized obesity enablement. We now live in a world where one can’t buy milk without walking past potato chips six times (in most aisles you’re lucky if you’re actually moving so fast as that).

That society has fallen into this particular dependency loop has been systematically engineered by the geniuses at Pepsi and Frito Lay (I’m 90% not speaking rhetorically).

True enough, within this larger morbidity, competition is a harsh mistress. There is such a thing as robust competition situated within a morbid equilibrium carefully nurtured by the profit motive with the largest bankroll. Yes, the morbid equilibrium would be even worse if robust competition in the narrow frame somehow faltered.

Is that good enough? Is that the society we wish to build?

Here’s my own story. I love chips. I had a bit of an addiction for a while and it was starting to show. It’s a tough battle. You’re busy, you skipped lunch, there they are, delectable single servings in large, larger, and largest. Then I started to consciously focus on how much I hate getting bogged down over those arteriosclerotic mini displays and I’ve hardly bought a bag since. I even switched stores to avoid the worst aisle-display offenders. I just had to make the right conscious connection, then it was easy.

Between the wider aisles and the narrower me, distance to the milk cooler never crosses my mind as being of the least concern. Interesting how convenience turns out to be a plastic thing, easily melded by others if you don’t meld it yourself.

Bevis Schock
Sep 12 2013 at 1:10pm

Sometimes we make simple things complicated. As the conversation went along I became more and more convinced that when some consultant says the milk should be in front for impulse buyers, or says it should be in the back to get impulse buyers to buy something else as they walk around, the guys with their feet on the ground cite the hassle of moving heavy milk and the ease of loading cases from the rear and say leave it where it is. (I thought we should be attentive to the writer above who said he used to work in one of these joints).

My conversation with local restauranteurs, who operate without fancy business models and/or efficiency studies but make a lot of money, indicate that they sell take out for the same prices as eat in do bc it is simple. In fact, in a conversation with the mgr of a local sushi eatery a few nights ago he was salivating at the fact that on most nights his take out revenue is around 20% of the total take for the evening, and it was easy money.

I suspect that he intuitively knows that he must be driven by profit maximization, but on any given evening he is driven by trying to increase gross revenue.

I think that when one looks at the gigantic number of entrepreneurs in any city in the country, (including the “one horse” operators), and the tiny number of graduates of MBA programs, one will quickly realizes for the majority of those who somehow have it in their genetic make up to successfully make money by buying and selling on their own account, the process is really based on three pretty simple concepts – none of which require deep analysis or schooling: deliver value to the customer, get as much revenue as you can, and keep improving. That is what I do in my little business, and I am going on 30 years.

Sep 14 2013 at 6:19pm

I think Bevis has it right with the simplicity argument and I’d add that having the same prices also provides simplicity to consumers.

There are lots of areas where consumers seem to react poorly to variable pricing. Keeping it with restaurants, try making a reservation at a popular/upscale place on Friday or Saturday night. It’s tough. On Tuesday not so much. On Valentine’s Day? Really tough. The raw economics would indicate that these places should charge more on Saturday and/or less on Tuesday and way more on Valentine’s Day. But they don’t…maybe some promotions on weekdays, but in the main, the prices are the same.

Why? Because people like it that way. Because the “I’m paying more on Saturday” outweighs “I’m paying less on Tuesday.” And that calculus annoys people, even if it makes sense when if they really think about. It’s similar to why people get mad about price gouging. And I think it’s the same reason why middle seats on planes cost the same as window and aisle seats.

Getting back to the take-out/dine-in, many people will see it as “they’re charging more for dine-in” than “they’re charging less for take-out”. And that’s going to generate the difficult to define not-so-good feeling for potential customers. Once they have that feeling, they’re more likely to go somewhere else.

Michael Byrnes
Sep 15 2013 at 10:15pm

A bit late to the party here (just listened today), but I have a couple of comments:

1. I liked Russ’ point that part of what you pay for in a restaurant is “renting” the table. Great point that we often don’t think about. But the value of “renting” a table isn’t a fixed price. When the restaurant is crowded and there are people waiting to be seated, the value of a table is quite high. People who linger over coffee are directly cutting into potential business. However, if no one is waiting to sit down, during off hours, etc., restaurant owners might rather have customers linger, in the hope that they might order something else. Also, they might prefer that people who do walk in off the street do not see an empty dining room. The real “cost” of renting a table could be negative! I would assume that the food prices charged by the restaurant are a compromise.

2. Russ and Mike, have you ever done a podcast on the economics of tipping? I recently read about a restaurant that abolished “voluntary” tipping in favor of an 18% service charge, refusing to accept any additional tips from customers. The owner argued that it improved the service quality.

Sep 17 2013 at 12:15pm

Here is a new question on this old topic… I wonder if taller or shorter people buy cheaper items on average. This would be the inverse effect of the “eye level” norm.

Most isles have the less expensive or store brand items on the very top or very bottom self. I’m tall and my “eye level” is not the statistical norm and so I tend to buy the less expensive stuff. Do I do this because of my height or am I just price conscious?

Dan Hanson
Sep 17 2013 at 2:05pm

I find the milk debate interesting because it is a good example of how local knowledge influences choices, and how even economists tend to frame the debate in terms of their own experience.

Most of the people debating where milk should be vs where it is approach this with the experience of consumers, and frame the question around that experience and make assumptions based on that experience.

But a large grocery store is part retail and part industrial. Anyone who has worked in such a store would have no trouble explaining why the milk is at the back. The obvious parts were mentioned in the podcast – the volume of milk, the rate at which it sells, the heat from the freezer, etc. All good points, but under-stated.

Milk turns over fast. You don’t stock a milk cooler once a day – you stock it continuously. Sometimes it’s an almost-constant activity, with customers pulling milk out of the front of the cooler while someone in the back constantly replenishes it. There’s just no way to do this at the front of a store – not without disrupting shoppers with constant trips from the back to the front with large dollies full of milk.

Also, the most efficient way to do this from a time and energy standpoint is to have a cold room behind the freezer. That way, you aren’t releasing heat while you’re stocking the cooler. You put on a jacket, go inside, and stay in there for a fair bit of time while restocking/facing the cold goods.

The room is also used for storage of the milk inventory that is still in crates. It’s much more efficient to make the coolers part of the large milk storage area than to have separate cold storage and have to move the milk from one cold room through a warm space to a cold cooler. It also helps maintain a longer shelf life for the milk if it doesn’t fluctuate in temperature.

Produce has similar limitations. Produce is along the periphery because most of it is stored in the back in cold rooms, and then it’s wrapped and prepped in a work area, loaded on carts, and moved out into the store. These are perishable, bulky items. The amount of restocking required is considerable, so keeping produce along the periphery keeps the disruption caused by restocking to a minimum and makes restocking faster and more efficient. There’s also more manpower required for produce because the store stock as to be constantly rotated, spoiled goods removed, etc. Putting it along the periphery keeps the disruption of all this activity to a minimum.

In addition, the produce coolers often have water spray systems, so you need to run plumbing as well as power. That’s hard to do if the cooler isn’t along a wall.

Then once you get past these factors, there’s the industrial design requirements. Mike Munger mentioned heat removal, and that’s important. There’s also power requirements, noise, and maintenance to consider. Large milk coolers require huge compressors and heavy power lines and auxiliary equipment that’s easier to maintain and inspect if it’s out in the open instead of buried under the floor or hidden behind decorative panels. It’s easier to mask the noise of the compressors if you can put them in a back room. Compressors that are in the public space need to have noise insulation, and that is difficult when the compressors are large and powerful.

Packaged frozen goods are another matter. They’re lighter, smaller, and have longer shelf lives. Therefore the restocking costs are lower, and the cost of energy per product is lower. So, it’s easier to put such products in the middle of the store.

Milk is a low margin, high volume product with substantial transportation and storage costs. The result is that small gains in efficiency of production and distribution are critically important – especially in a large grocery store that has a lot of price competition.

Sep 22 2013 at 12:02am

I really enjoyed the podcast. I don’t have anything to offer to the discussion about why milk is where it is. I just wanted to point out to Russ that there’s a Walmart in Montgomery County, in Germantown, although I don’t know if it’s a “Super” Walmart. But also of note, a Wegman’s also opened in Montgomery County (also in Germantown) about 2 weeks after this episode was recorded, so even in highly regulatory Montgomery County, these things can get built. It’s true, though, that there are a lot more Wegman’s and Walmarts just across the river in Fairfax County. There’s a lot more sprawl there to. Just sayin’.

Comments are closed.

About this week's guest: About ideas and people mentioned in this podcast:

Podcast Episode Highlights
0:33Intro. [Recording date: August 29, 2013.] Russ: Topic for today is milk. Want to alert listeners that later this week we expect to be releasing a bonus, mid-week episode.... EconTalk All-stars. Audio and video of the event to be released.
1:56Russ: Now, on to milk. The idea for this week's episode came from a brief video interview with Michael Pollan, who described the supermarket as "a treacherous environment if you are trying to eat healthily." He then went on to say, "It's designed"--the supermarket--"to extract as many dollars from your wallet as possible. So for example the milk will always be the maximum number of paces from the door." The idea being that you have to get to the milk; they are going to put all these items along the way. And I mentions I think that they are the high-margin items. And that way, through impulse-buying, you are going to make more money for the grocery store. When I heard that, I stopped listening, actually. For years, when I taught microeconomics or price theory I would almost always ask my students the following question for homework. "True, False, or Uncertain: Milk is at the back of the store because that way you have to go through the whole store and are likely to buy lots of other stuff along the way." Now, a lot of students--and you out there listening--might also give the same answer. Which is: A lot of people would say, True, that way the firm can maximize its profits. But that's a bad answer. It's true, firms do want to make a lot of money. And it might even be true that milk is at the back of the store to make you go through the whole store. I don't think so, and I'm going to give you an alternative explanation in a minute, and then Mike's going to chime in. Well, probably more than chime in. But if it is true that that's why they put the milk in the back of the store, you have to ask another set of questions: Why is the store so clean? Why do they clean the restrooms? Why do they hire friendly checkout people? Couldn't they make more money if they didn't sweep the floor, and hire ruder people? Why do they bother being open 24 hours sometimes--when they are selling very few items? Why do they give away samples? Why don't they charge for them--because you'd be willing to try something at least at a small amount? And is it even true that the milk is always at the back of the store? Or better yet: Is it true that milk is the most popular item, guaranteed to make you wander around the whole store? Does Barnes and Noble put the most popular books at the back of the bookstore so that you have to go through the whole store? That way you'll buy a lot more books? I don't think so. They put them, strangely enough, at the front, where it's most convenient. What about 7-11 and other convenience stores? The whole idea of a convenience store is to make it more convenient. It's small. You don't have to travel as far for items that you buy frequently. There's really no 'back' of the store. So, yes, the milk is in the back of the store in the cooler, but that's very close to the front. And they even put sodas sometimes on ice right in front of the store to make it more convenient so you won't have to walk through the whole store. Even though it's tiny, they still want to make it even easier. So one view of the world, which is Michael Pollan's, is that the grocery store exploits you. It makes you walk through the store so you'll be suckered into filling your cart. Now I have a different view. Yes, the grocery store wants to make a lot of profit. But that urge, as it often is--almost always--is constrained by competition. And competition forces the greediest of grocery store owners to serve its customers. Or we'll go somewhere else! Sometimes, of course, firms don't compete very hard. They don't want to. They'd rather not. They'd rather have a cartel so they don't have to work so hard to serve the customer. But in general those things lack a competition, as cartels are hard to sustain without government. They tend to break down. Firms are always looking for an edge. The way you look for an edge is to make the consumer happier. And then they are more likely to come into your store. Now, Ari Indek[?], who follows me on Twitter, writes that you can follow me on EconTalk and other things at EconTalker, Ari Indek writes that ShopRite, the grocery store chain, actually puts the milk up front in a mini-cooler, to make it more convenient. Now, Pollan also claims that the highest markup items are at eye-level. He argues that the 'real food'--and by that he means that the unprocessed food, the produce--is lower margin, and that's around the edges. So if you want to eat healthy and to save money, you should shop the perimeter of the store. But as Brendan O'Donohoe argued in our potato chip episode here on EconTalk, only 15% of the people are like me, who tend to snake up and down all of the aisles. Wouldn't it be true then, that it would more profitable to put the vegetables, the low-profit items, in the middle of the store, and put the high-profit, high-margin, high-processed items--which are things like the cereal and other things--wouldn't that go on the perimeter, according to Pollan? So, if I'm right, why is milk in the back of the store? If I'm right that grocery stores don't do it there as a conspiracy to sucker you into buying stuff you don't necessarily want, or that's high profit, that's at eye level, what's my explanation? And before I do it, I'm going to let Mike react. Mike, what's my explanation? What are your thoughts? Guest: My first thought on listening to that, I can just see you in a retirement home wheelchair shouting at the television, 'No, it's not true!' You're like two years away from that. Russ: No doubt. When you and I get together, I do tend to somehow lose my EconTalk hat. It's not on right now. And as a result I did get a little ranty there. I should have warned the listeners.
7:22Guest: I think that's just fine. We'll think of me as the interviewer here early on. I went and tried to find, because you had warned me, that this was going to be your concern. And that you had said that you were going to try to give an answer. What interested me about it was the confusion that I think people have between firms that are trying to maximize profits, and therefore firms that are trying to exploit consumers. And I think that the explanation that you gave, I think you said it was 'constrained.' I would say, 'Directed.' So, both von Mises, when he talks about the consumer being the captain of the ship, when he talks about grocery stores and producers being the crew of the ship--and yes, it's true that they go around making the ship change direction. But the captain chooses that direction. Now he may not know much about the operation of the ship, but if there is something wrong, he'll switch ships. The actual person who came up with the phrase 'consumer sovereignty' was William Harold Hutt, W. H. Hutt. And I went back and looked at some of the things that he wrote. And it was surprising to me how much of an argument he had with Joan Robinson and Arthur Pigou. And some of the other people who argue about welfare economics. What I think is interesting is the way that Joan Robinson and the opponents of the view that you are probably going to give, would want to think of stores as being little, local monopolies. So, sort of the oligopoly firms operate at the local level. Like restaurants. Because you can't really compete if it's a convenience store. It's late at night. This is the only one I can go into, and so I'm stuck. Russ: I think that's a silly argument, but carry on. Guest: Right. So, and Hutt thought so, too. How can you possibly thought that's the nature of competition? But they were absolutely stuck on it. So I think in interpreting Pollan and the people, the sort of intellectual history of that, starts on a very strange view of the monopoly power of grocery stores. And so, checked the factual basis, what you had said. It is actually true, the dairy section of most grocery stores is on the far said, away, or at least in the far corner away from the doors. And I think I have an explanation of that also. But maybe we'll compare that in a moment. More and more though, convenience stores, even grocery stores to some extent, have these coolers near the door that are called 'Grab and Goes.' And Grab and Goes are understood to be a response to a consumer's desire, sometimes, to get a gallon of milk and go. And they are actually quite frequent. So, Pollan, regardless of the interpretation that he would attach, is just factually incorrect. So, if it ever was a problem, and that's what consumers wanted, then consumer sovereignty, or von Mises, the consumer being the captain of the ship, said, "I'd like to have milk near the door, please.' And there it is.
10:46Russ: And you could argue, though, to give Pollan his due, you could argue that, okay, it took 70 years. That's not very effective. So it could be that now the grab-and-go phenomenon, and the convenience store phenomenon, have finally broken the grip of the monopolistic grocery store. Guest: But if the legacy of that Joan Robinson, monopolistic competition view, that is so hard to get rid of because it's just ingrained in the way people think about economics. Russ: I think it's a fascinating phenomenon; I think we've talked about it before as to why people are so eager to assume they are being exploited. And of course, by the way, this secret is so secretive, that, is so well kept, that of course Michael Pollan knows it. So, he's alerting the world to danger. So be careful, keep your eyes down, don't look at the shelves. Keep 'em down, actually keep 'em down at the level where the oatmeal is, in his worldview. He suggests that the processed cereals are up at eye view; and oatmeal is down at the floor. In my grocery a lot of the processed cereal is down at the floor also. And I don't buy either one of them, so I don't pay a lot of attention to it except to notice that cereals are all up and down. But it's very comforting. I don't really understand. I think you'd sleep better at night thinking that the grocery story is your friend rather than your enemy. And I don't want to mislead you, listeners. I'm not here to tell you, oh, it's okay, it's really your friend. I don't think they are literally your friend. But I think it would comforting to know that that competition protects them from the rapacious grocer. Guest: I think it also would be interesting to be compared to what? Because, famously, when Boris Yeltsin visited a Texas grocery store in 1989 when he was visiting from Russia--you know, he's a wealthy guy. Actually accused the store of being a 'Potemkin store'--that we had added a bunch of different products. So, if the complaint is there's too much product differentiation and that's somehow wasteful, okay, let's put that to one side. That we'd be better off if we had fewer choices and it's wasteful to advertise all these different things. That's not the complaint. The complaint is: Product placement is inherently at least duplicitous and probably manipulative because it gets people to buy things they don't really want and they have impulse purchases that they don't really want. It so often happens--I went and spent nearly two days looking at product placement. And it turned out that by learning things I turned out learning a lot less than I thought I did out outset. I was actually sympathetic to this view that you could manipulate people, most of the time were in a hurry. The fact is, most of the time what they are trying to do is maximize revenue. And maximizing revenue, given that the coats are relatively fixed, it's like maximizing profits. It is true that people buy more of they eye-level stuff--stuff that's eye level on the shelf. Russ: No doubt. Guest: But it's also true that high volume things get moved to eye level on the shelf because it's more convenient. So it's over-determined. It's not that they can make you buy whatever is at eye level. You probably would probably have this vague sense of unease--I don't like this store. And it might be that they have a bunch of stuff you don't want to buy at eye level. You might not be able to identify that, because you are so used to having your desires anticipated and fulfilled by product placement. I'm not coming back to this store. I'm not really sure why, but they don't seem to have anything. Russ: Before I forget, I have to quote Walter Williams, who, when we interviewed him ages ago, I'm pretty sure I quoted this story of his because I like it so much. His story is, he talks about your relationship with your grocer: I don't tell my grocer when I'm coming; I don't tell my grocer what I want to buy; but if they don't have it when I get there, I fire them. I'm not going to come back. I remember shopping at a Venture store which was a chain now defunct in St. Louis, with one of my children. We went shopping--it was a Target, K-Mart, Wal-Mart competitor that didn't make it. And I was taking one of my kids to buy a basketball or something, and they didn't have any. The shelf was empty in the basketball section. I think my child, who was probably 6 at the time, said: I think this store is in trouble. Yeah. You're on to something. But it might have been me. I don't have the transcript. But usually when a store is out of stuff it's a death sentence. And actually Venture was in a death spiral at the time. I don't think they were trying very hard, or they weren't doing very well at it. Whichever. But a store that systematically puts the most profitable items at eye level, as you suggest, is probably not going to be as pleasant a place to shop, and whether I notice that or don't is a different question, I guess. You can debate that. I don't deny the fact that people have impulse issues. I'm sure that people make mistakes; I'm sure there are consultants who try to exploit those mistakes. They are looking very hard at what people do, and I'm sure that there are listeners here--I hope one of them is Brendan O'Donohoe, who used to be deeply involved in this, for Frito Lay--I'm sure he and others perhaps listening know a lot more about what grocery stores are trying to do. But what they are trying to do and what they are able to do, or what they'd like to do, those are not the same thing. Guest: That's what I think is interesting is their description of what they actually do is often very consistent with trying to anticipate and fulfill desires that the consumers have that they may not have really consciously thought out. If you look at the end caps, the end of the aisles, the things that are on sale-- Russ: Very valuable real estate-- Guest: or right by the place where you are pushing your cart where you wait in line, so you are sort of trapped there because there are two carts ahead of you, what sort of stuff is there? Well, there's candy. And candy not in big packages--it's individual packages because that's the highest profit way to sell it. Okay, fair enough; that's an impulse purchase. We'll give them that one. But fingernail clippers? Batteries? You look and, Oh, I need that. You can call that an impulse purchase but it's more like they are trying to make a list of stuff you might need, and you think, oh, my wife would have been so mad if I'd forgotten this; I told her I'd pick some up; thank goodness it was here. They actually try to anticipate. These aren't impulse purchases; these are remind-you, you need them purchases. Russ: And you used the phrase, 'over-determined.' That's an econometrics phrase. What you meant by that is that it's hard to distinguish different hypotheses given the data. The data are consistent with some of these different views. I am sure that the stuff they put at eye level is the most popular. The question is: Is it the most popular because they put it at eye level, or do they put it at eye level because they are trying to make it easy for us to find the stuff we want? Guest: I found a surprising number of studies that said it was the second. That I think is something it's easy to lose sight of. If you think of the store inherently as trying to manipulate you, the fact that they look to see what it is people want and then they put it at eye level, that's actually consistent with convenience. Russ: I also want to concede--often, when I go to Costco, I have a very good time. I almost have a good time. But often when I get home, I regret some of the pleasure I overindulged in. Do I really need two [?] Guest: In fact, your lovely wife may point out--why did you get that? Russ: You bought another pair of reading glasses? Well, I'm always thinking what if they are lost. We have 19 now. So, I now have 240 oz. of ketchup, 60 peaches. So there is sometimes buyer's remorse. And if I found that to be a common problem at Costco, I would try to go less often. Or I wouldn't go there at all. Guest: But to the contrary, it's so much fun. Russ: That too. I enjoy shopping there. I actually get pleasure from it. Which is interesting. It's an obvious mix of perimeter and snaking. I do a lot of--I do both. I sometimes just do perimeter and sometimes I go down most of the aisles. If I have a list in my hand. In my experience the snaking phenomenon is because I don't shop there often enough, whatever it is. It's the most efficient even if it's the most horrible. I'm going up and down every aisle; I'll find everything on the list if I have to. Guest: It's like the salesman problem--you'll only cover each aisle once. So it's efficient given that you don't know where anything is. Because the alternative would be to hop around, and you'd end up covering a lot of aisles three or four times. So it makes sense to go and look on each one. It's efficient given that you don't know the store that well.
20:18Russ: So, it's nice that there are some studies that back up my view. I didn't go to the trouble--this was my suggested topic for a change; I wanted to tell the listeners. Usually when Mike and I talk, Mike proposes an idea that's in his brain, that's bugging him or he's interested in or that he wants to talk about. This one, as you can tell, is mine. It was my idea, but you did the homework, which I appreciate. Of course, those studies could be wrong. They could be pawns of the industry. Who knows where the truth lies? My goal here, though, is to try to get listeners to imagine that there's an alternative and that imagining that alternative, whether it's true or not--I think it's true, but even if I'm wrong--when you start, as I do, I often start with the presumption that the result is a competitive result rather than an exploiting, monopolistic result, it helps you see things you wouldn't otherwise see. Now, the underlying hypothesis could still be wrong. Maybe they don't compete very hard. I want to mention--you earlier alluded to Joan Robinson and you mentioned restaurants. Sometimes there's a restaurant that's very convenient for you, so it has some advantage in keeping your business and it can do an imperfect job and you'll still keep coming because it's so convenient. That can be true about your grocery; it can be true about your gas station. But it's not convenient for everybody, is one of the problems with that kind of analysis. And these businesses, groceries and restaurants, etc. are unbelievably competitive. They are constantly struggling to stay afloat. They have very small margins. They are very high cost. And in the case of the restaurant business, most of them don't make it. So the idea that a restaurant can exploit you is very hard for me to understand. Now, a successful restaurant, a restaurant with a reputation, a restaurant that's been in business for a long time and has a great brand name, they get a window of opportunity to take advantage of that for a while. They can serve you a bad meal and you'll come back. They might serve you two or three bad meals and you'll come back, because you had 50 that you liked. Eventually, you'll stop going there. But certainly new restaurants, and most restaurants, are constantly struggling against the competition. Do you want to react to that? Guest: Well, what's interesting about the monopolistic competition model is that it is true that the restaurant can decide what price to charge. And they are probably not literally charging price = marginal cost. They don't do enough volume to get to the point where they are going to charge price = marginal cost. That is, the amount that it costs for them to make and serve that one meal is not what the price is. The price is very substantially higher than that. Now, some nights there might be nobody here: and we ought to cut our prices. But they don't. They may have happy hour or something like that. But the thing is, it's a relatively homogeneous product. Maybe they have a few things that are specials, or happy hour from 6-8. Grocery stores have a more complicated pricing model, and they do it in a number of ways. They have sales that they advertise--loss leaders that they try to get people in on. And even convenience stores may have milk that's pretty close to the price in a grocery store. Whereas toothpaste is 5 times as much. So they have really high prices because they don't sell that much at the 7-11, the Circle K; they don't sell that much, it's a really high price, but it's 11 o'clock at night, I really need it, there's no where else for me to go--I'm actually glad that it's open. Some people are mad that the price is so high. Russ: And they have low turnover of the items, so they have to carry some of those fixed costs per item, it's going to be a little bit higher. So it looks like the margin is very high. But it may not be in reality. Guest: The margin is very high on that one thing. If they knew they were going to sell it, it would be different. But this is basically an insurance policy. What they've done is they've bought a bunch of things, and every once in a while one or two people are going to buy a couple of them. They may end up throwing it away, but they have this inventory and they have to carry this inventory. It's a really terrific service. Grocery stores can use some things like that that are loss leaders, but generally they turn things over quite a bit more. The interesting thing about grocery stores lately is they move more and more to loyalty programs, where you have a card and there are substantial discounts on all kinds of products, if I show my Kroger card, my Piggly Wiggly card, my Harris Teeter, here in North Carolina. If I show that card I get substantial discounts. I think there's just a bunch of odd things about that. Russ: Yes, there are. Guest: One of them is that they are using the information about the patterns of your consumption, your purchases, to try to market things better. But I find it fairly common--if I don't have my card, the person at the checkout will just use theirs. Russ: Yeah, I know. Guest: That seems bizarre to me. Russ: Well, that hurts the price discrimination explanation. The claim is that they are trying to give a good price to the loyal, or the ones who are smart enough or the ones who are concerned enough about price to have the card but they exploit everybody else. Especially the people with the allegedly relatively inelastic demand who don't care so much about price, and therefore can be exploited by the high marked price and don't get the discount with the card. The same is true of coupons, by the way. 'Do you have a coupon?' 'No, I don't.' 'Don't worry--we have some here.' That's bizarro. It's certainly bizarro for the price discrimination model. It's just hard to understand, period. Guest: Well, I think there are two functions for this. One is: We announce sales and we announce reductions for loyalty programs, like those cards are going to get people in the store to begin with. But if someone is in the store they seem to say, let's provide them with friendly service. I don't think that anyone has gone to the checkout person and said, You will be fired if you use your card. It seems to be, it's at least tacitly accepted.
26:28Guest: Like I said, it turned out I knew less than I thought I did. I still want to hear, Professor Roberts, if I can be the interviewer for a second, your explanation, for why milk is in the back. Russ: Because I'm going to put you off for a minute. Guest: Okay. Russ: I'm going to duck the question, because I want to jump onto something you alluded to about restaurants that's too fun not to mention. But don't let me get away with this. Because it could be I don't have an explanation. And I just want to talk about the Red Sox. Or the Cardinals. Because it's August 29th and they are both in first place. Guest: Both in first place. Russ: I think we should schedule an interview for September 30. I know causation and correlation can get confused, but who knows? Why take a chance? Anyway, the thing I want to jump to is something I heard from Earl Thompson, UCLA economist, that's so interesting and if you've never heard the idea or the argument before, I think you'll find it intriguing out there. Which is that if you want to know the markup on something--and a lot of times people say, oh, the liquor is so expensive, that's where they make all their money; they lose money on everything else but they get it back on the liquor. Or, they make it up on the desserts. Or whatever it is. And one of the things that ignores--it ignores a couple of things. It ignores the fixed and sometimes not so fixed cost of a liquor license, which makes the actual profit of the item different from what the markup is. I think people think, I know what a bottle of wine costs, and this is so much more than that in the restaurant. But of course the steak is a lot more, too. You say, well, okay, but there are labor costs; there's not much labor cost in a bottle of wine. How could that be the legitimate, competitive markup? And Earl Thompson's insight was, if you want to understand the markup on restaurant items, you should think about how long it takes to eat them. Because what you are doing at a restaurant is two things. You are eating the food, which has fixed and variable costs--excuse me, an up-front and a labor cost. But you are also renting a table. And--this is another topic--the table is rented for $0. The rent for the table is implicit in the items. Now, why that is, is an interesting cultural phenomenon; maybe we'll talk about it; put it aside for the moment. You can sit at the table for a long time. There's no meter. There's an argument for putting a meter there. But, for either cultural or convenience reasons--I think it would hurt the pleasure of the meal to see that clock ticking; I think that's the reason. Guest: At some point they may start to say, look, you either order stuff or go. Russ: They will. So there is a meter, but it's very casual. There's a lot of looseness in that meter. So, basically, you pay for your table rent in the retail price of the item. So, coffee--oh, gosh, coffee is so expensive; it costs them pennies to make it; they make a killing on the coffee. They don't. Coffee is what you linger over. The wine is what you linger over. The dessert adds real minutes to the meal. And all of that reduces the number of times a table can turn over in the course of an evening. And so you have to pay for the privilege of consuming those long-time items. And I think that's an incredibly deep insight into pricing in restaurants and how to think like an economist. I think it's beautiful. Guest: Yeah. I had not thought about that before. So, the idea of average cost is very complicated; and marginal cost is sort of facile. So, just complaining about that, you can't mean that. Russ: So, if Earl is right--and I think he is--and if I'm right in agreeing with him, that raises a question. And I'm going to just raise this question; and then you can press me on the question or you can just answer the question and we can go on. But we would spend a long time in my micro class talking about this. Which is the following. If it's true that the markup for the item, the price of the item, includes, is affected powerfully by the amount of time it takes to consume it because you were implicitly renting a table, why is it that so many restaurants have take-out menus that have the same prices as their sit-down in the store menu? And that's a very good question to challenge Earl Thompson's view and my view. And listeners, I challenge you to think about how I would answer that question. Because if I'm right, then take-out should be cheap. And eat-in in the restaurant should be the more expensive price. The fact that they are together seems to refute the claim that the price of the item reflects the time it takes to eat it. And that would be true certainly for drive-by windows at fast food restaurants, but even just regular restaurants that offer you take out. That would seem to be a refutation of that. But it's not. Of course, there's an answer. It may not be right, but there's an answer. You want to think about that, Mike, or do you want to press me? Guest: Nope. I think that is interesting. I think for fast food, surely the answer is that they are rarely completely full. They don't have the same, not exactly the same peak load problem. And people don't linger as long at fast food restaurants anyway. Russ: Great point. Guest: The question is more interesting for relatively expensive restaurants that have a take-out menu. And I've noticed more than a few--it's not half, but I've noticed more than a few that do have a separate take-out menu where it's cheaper. There are discounts. But by and large you are right, and that does seem like a mystery. If the explanation is correct. If Earl Thompson's explanation is correct. And it's intuitively plausible. Then you immediately run into a problem: Why would they have the same prices for take-out? So I'm really just restating it. I think there are some places that have different take-out menus, but not many. Russ: A lot of them have the same. And forget the complicated how long it takes to eat a thing. If I'm not going to be sitting at a table, shouldn't I get a break? And a better way to say it: If I'm going to be sitting at the table, shouldn't I have to pay a premium? And yet another way to say it is: A restaurant that has lots of take-out can have a smaller eating space, in theory, and have lower prices because they have lower rent. So shouldn't they distinguish between those? And wouldn't a restaurant that offered a bargain on take-out monopolize the take-out business? Shouldn't there be competition among restaurants for take-out so that those who refuse to offer a discount should be in trouble relative to those that do? Guest: Yeah, I think that's the--it's not the: I should get a break. Because nobody cares what you should get. But: why wouldn't they give me a break in order to attract my business? And the peak problem then--you'd have to rebuild everything because the peak problem is going to be in the kitchen. I can't provide service to the people who are paying a premium to be at the tables if I'm also really backed up on take-out orders. Russ: Yeah. That's another issue about space. By the way, that of course plays another role in the price of the good: it's not just how long it takes to eat it. It's how long it takes to cook it. Obviously an item that takes longer, we would expect it to have a bigger markup over its raw material cost than an item that can be cooked very quickly. Guest: Yes, the time and preparation effort. Russ: Well, I'll give you my answer. It's a guess. Obviously we don't know the real answer. Maybe it's a mistake on the part of restaurants. People say there's a mistake; you always want to say, well, they kind of have an interest in it. Guest: If there's a mistake, you should be in the restaurant business. Russ: That's one answer. That was one of my thoughts for Michael Pollan, too: why don't you open a grocery where the milk is all in the front? Guest: He'd make a killing. Russ: And the answer, he's going to make a killing, oh my gosh, with all the healthy food? Maybe later we'll get to some nudging, because I know that's something you've been thinking about. Because that all can get wrapped up into this. But my explanation, which is the best I can do, is that if there was a big differential, which there is in terms of cost--I think--between take-out and sit-down--I think sit-down is more expensive. Now take-out has its own costs that maybe aren't so obvious. So that could be part of the problem, too; maybe there are some aspects of take-out that I'm not aware of--that it's very costly to bundle it all up, put it to the side; maybe people don't show up sometimes for their take-out orders, food gets thrown away. So there may be some issues here I'm not aware of. But my other thought was that when you come in for take-out, what if you said: I changed my mind; I'd like to eat it here? Are they going to say: Oh, we're going to charge you more for that. So if you drive through the McDonald's take-out line and say the take-out price at the window is half, and you've got the little bag. Are they going to have to now monitor that you don't want to come in to the restaurant and sit down with it? It just adds a whole level of other costs and monitoring that I don't think they want to deal with. But that's just an answer; it's the best I can do.
36:33Russ: So, let's talk about the milk. So, the reason I think it's in the back--it's always been my presumption, and I've got a little bit of evidence on this; it's anecdotal but it's a little bit of evidence--is milk has to be in coolers. And it's hard to put a cooler in the middle of the store. Obviously we're finding ways technologically to do it; that's part of the reason why it's taken 50 years to put the grab-and-go mini coolers in the front. I think that's a technological change. But it's hard to put a cooler in the front. And it's very inconvenient to fill a cooler from the front. So, milk is brought to the store in trucks. It's cold. It needs to stay cold. You don't want it wandering around. And it's just hard to have a cooler in the middle of the store. Or the front of the store. So the coolers tend to be on the back wall of the 7-11 and the Harris Teeter. There are some freezers in the middle of the store now--for ice cream--so that does challenge my claim. But my argument would be it's not to inconvenience customers to make them buy other stuff. It's that it's the cheapest way to provide the milk. Guest: Well, I think that's exactly right. And actually I think there's evidence for it, from, again, the time that I spent, more evidence than I would have expected. Milk is not only heavy and bulky and relatively fragile, but the coolers that you have the milk in produce quite a bit of heat that has to be vented. Russ: That's a better point. Guest: So that it has to be against a wall. To start with it has to be against a wall. And it would be easy if you could deliver it from the back. And so almost all milk--if you see someone putting more milk in from the grocery store, it's not from the front. There are these slightly angled metal sort of tray things and they are loaded from the back. And it has to be in the back to do that. And that's true for grocery stores, not for convenience stores, but as you said, convenience stores aren't that big anyway. So the grocery stores do that for milk. Freezers you load once and they may turn over every two or three weeks. You have to put new milk out almost every day. And you do it from the back-- Russ: So, the inventory part is a big part of it. But how is it, on this venting issue, how does the ice cream part of the store that's not against the back wall, how do they manage that? Because it's not as frequent? Guest: A lot of them are horizontal, and the ice cream ones, they have fans that keep the cold air in. The ones that are big, that have the heavy doors aren't open nearly as often. And so what they have is pipes that run under the floor. The milk gets opened so often--they are three times as powerful. It actually takes a more powerful cooler for the milk than it does for the freezer because people don't open the freezers as much. Milk is constantly being turned over, and the doors are constantly being opened. It just produces a lot more heat. The question is not how cold does it have to be. The question is how often does warm air come in. And milk, just a lot more. Russ: You said you didn't learn anything. You said you got dumber. You got smarter! Guest: It wasn't so much that I really felt like I got dumber before. What happened was that everything I thought was true just turned out to be more complicated. So, maybe that's what getting smarter is--finding out that you don't know the things that you think you did. I just assumed that it was true. It isn't so bad--you have to walk to the back to get the milk, to look at the dairy. In fact there's almost no other way that they can do it. And if you were going to design a store, it would be awfully expensive for you to do anything except the way that all the stores do it. And so it's not a conspiracy. It's just technology. Russ: And this grab-and-go think, this new innovation of the freezer at the front: what do you think about that? Guest: Well, it's a smaller cooler, and it produces heat. But it's not the massive one. They just re-stock that by hand; if it's out they restock it by hand. They restock it from the front. There aren't actually that many people that just want to get milk. That's just a myth. It's not true that you run into a grocery store and just want milk. If you want that, you go into the Circle K--you go into a convenience store.
40:08Russ: Yeah. I wonder what proportion of customers actually do buy milk when they go? Guest: Do buy only milk. Russ: No, any milk. Is it 90%? 60%? I have no idea. That's another interesting question. Do you have anything else to say about coolers, milk, eye-level, margins, groceries? Do you have anything else you want to add? Guest: Well, it struck me that there is, like tables in restaurants, a scarce resource in grocery stores, and that is shelf space that people are actually interested in. So, if I am searching for a particular product, I can find it even if it's not on the top shelf. If I'm searching for something that's a particular size and I'm going to either compute the price per ounce or I am going to use the little tags that have price per ounce, I can put those on the lower shelf. So, if this is something that someone needs but is likely to have forgotten that they need, and thus will say, Oh, I remember now and pick it up--the shelf space that's available there is very expensive. One of the things that the studies have found consistently is that those things--you can't see, I'm making air quotes--things that are on that highest-value shelf space, they are "overpriced." They are much higher priced compared to the price per ounce for the lower-down products. But part of the reason for that is they are occupying valuable shelf space. Russ: Yeah. There's a rental fee. Guest: There's a rental fee, just like the table. Russ: But it's my understanding--and I apologize to Brendan O'Donohoe, for when we talked about this--but it's my understanding that they pay a premium to get in those--that food manufacturers pay a premium to get in those slots and for the end caps. And they stock them themselves. This is one of the most interesting things I learned from that potato chip podcast--and if you keep your eyes open at the grocery store you'll see it. You'll see people in the store who aren't working for Harris Teeter or Giant. They are working for Frito Lay. Guest: Well, they are working for them in a sense; they are just not being paid by them. Because they are doing the service that they would otherwise had to provide, but that's part of the implicit charge to the potato chip manufacturer or the soft drink manufacturer. It's an in-kind payment to the store because they'll do the stocking. Their return is that they get better shelf space. Russ: Yup. And the customers come in, and they like it. But it is a fascinating thing that in many ways a grocery store is just renting the space out to the manufacturers. We think of them as being something else. Guest: It's a mall. It's become more like a mall. Now the boundaries between stores in the mall aren't as clear as in an old-fashioned mall, but they are much more clear than might have been in a grocery store 50 years ago, where there were stock boys and stuff came in in trucks and you put it in. And I want to say one more thing about groceries, and then let's talk about 'nudging.' The other thing I want to say about groceries--and I'll be interested to hear your experience--but I live in suburban Maryland, in Montgomery County. And Montgomery County is a highly regulated environment. It's very hard to open a grocery store here. Grocery stores, unlike restaurants, have very large footprints. They require a lot of space to bring in the trucks for the store itself, for the parking lot. So there's a lot of rent-seeking activity around the fact that the politicians here restrain the free market's ability to provide grocery stores when and where you want them. So where I live, it took forever for a second grocery store to open nearby. There was a Giant Grocery Store nearby that was not very well kept, not very attractive. When Harris Teeter finally opened up, the Giant got overhauled, finally, and became a fairly pleasant place to shop. But it was only when Harris Teeter opened that they became a more competitive place. And the political environment here, the regulatory environment, makes it hard for competition to work any kind of magic at all. It's a little like the old phone company slogan: 'We don't care; we don't have to.' When you are in a competitive environment you have to care; you at least have to pretend to care, or you are going to lose your customers. The other thing that happens here in Montgomery County is that super-Walmarts--I don't think there's one here in Montgomery County--need a special permit. Any store of a certain size needs a special permit here in Montgomery County. I think the only store it's relevant for is super-Walmart; maybe Costco--I don't think so. But basically what it means is that for a super-Walmart to open, they have to jump through an enormous number of hoops. They have to be really nice to a bunch of legislators and county councilmembers. And as a result, there isn't one. That's part of the reason. It's also that land is expensive here. I don't want to suggest that this is the whole reason that there's not much grocery opportunity here. But my guess is that in regions of the country where the permitting and zoning process is a little more flexible and a little more liberal, then I think it's probably the case that you have more groceries to choose from, pricing is better, the store is cleaner and nicer. That's just my guess. We don't have a Wegman's here, for example. They are in northern Virginia. Wegman's is a phenomenal chain that would clean the socks off of Giant if it were here. But they don't have to worry about them. They are not here. And I don't think it's because Wegman's doesn't want to be here. Guest: This is another thing. I had not read W. H. Hutt on this. There was a battle between Hutt and Joan Robinson about sort of the status of welfare economics: how should you think about standing of harms. And what Hutt said is that you ought to try to choose policies that do the greatest benefit for large numbers, rather than try to protect producers or labor, where it's relatively a small number but it's a significant harm if they go out of business. Whereas Joan Robinson, Pigou, and others said: We should protect producers also. And so what you are talking about, to the extent that, with the caveat that there are many other reasons why it's hard to open a grocery store--but if you have to get the permission of your competitors to open a discount grocery store, they are going to do everything they can to withhold it. Because they are going to go out of business. The stakes for them are enormous. And there's no voice in that process for consumers, because the difference of a nickel, the tiny margins, the differences that I might pay for a few products at the grocery store--sure, I'd prefer a Wegman's. But there's a Giant; maybe there's another store. It would really change the nature of business for those. They are the ones who go down to the city council. They are the ones who show up at these meetings and say, no, no, there's all sorts of reasons we shouldn't do this. Hutt was really way ahead of his time, I think, in saying that these interest groups--a sort of a George Stigler theory. This was in the 1930s. A theory of interest groups that would say you're going to have very concentrated benefits that will be dwarfed by the costs that are imposed on many, many people--that's why you need to worry only about consumer sovereignty. Harms that are done to producers don't count. It's easy for us as economists to say that. Politicians don't feel that way. Russ: Yeah, that's a great point. I'm trying to find a way to work in something--we've been talking about milk, but milk sometimes does go into coffee, and coffee's called 'java', so maybe we can make an allusion to Jabba the Hutt. But it's probably cheap. Guest: I'm glad you didn't try to do that. Russ: But Hutt's an economist that--I don't hear his name very often. It's kind of cool. I'm happy for him. Guest: He was one of Buchanan's favorite economists. He was South African. He wrote in a way that was not all that clear. But he, if there's one thing to know about him--he's the guy who came up with the phrase 'consumer sovereignty'. Russ: I didn't know that. That's very cool.
48:21Russ: Let's close talking about nudging and various kinds of paternalism. What did you want to say about that? Guest: Well, one of the things about the nudge--and I've actually debated Cass Sunstein about this on occasion. The idea is: the way that we arrange things matters for the way people perceive them and compare and make choices. Russ: Which is certainly true. No doubt about that. Guest: Sure. Yep, there's no question. They also would say: Any way that they are arranged is a choice, so let's make a good one. One possibility is that stores are doing what Michael Pollan and others say, and they are arranging things in a way that's really manipulative. In that case, having a regulatory policy where we would arrange them better, maybe that's defensible. Suppose that we believe--as I do--consumer sovereignty drives this far more than manipulation. That is, consumers have certain things that they want, they are likely to buy; those are put at eye level, and it may benefit me because I buy it; I forgot that I needed it, I pick it up very quickly, it's very convenient for me to do it. But we may not trust consumers to make the best choices for themselves. Because they choose things that are maybe too expensive, are not very nutritious-- Russ: The candy. Guest: The candy, salty things, things that are high in fat, not enough fruits and vegetables. And so there have been a series of experiments around the country with making small changes in the environment. And we'll put up some references to these rather than go into details. But let me just give you some of the highlights. One of the things they did was they put a pretty large mirror on the grocery cart. Russ: Not a scale? Guest: Your big fat butt should go buy some apples. Russ: How about a flashing sign, a little neon, like a ticker at the bottom of the screen, across the handle bars of the cart: Get your rear end over to the produce section. In fact, when you are in the candy section, alarm bells should ring. Sorry. I interrupted you. Guest: And the more that you put in--there's a flashing light: Here comes lard-butt. It's a mirror. We're joking, but there's almost nothing that you can say that's as strange as reality. They put mirrors, as Russ just said, his voice rising with indignation--they put mirrors on these grocery carts so that you could look--and it's from underneath--so that you could see the fat under your chin. It's not just a view of your face--though it is--but it's also angled specifically so you can see how fat your face is. And this is supposed to make you go over the vegetable aisle. Now, it turned out that for a number of grocery stores, if they could increase the volume of sales in their vegetables, it would actually be a benefit because the marginal increase in price over what they pay is highest there. Because turnover is so high. So much of it goes to waste. Russ: I lost you there. Say that again about vegetables. Guest: All right. Let's say on a head of cabbage. Let's compare that to a box of cereal. Russ: A staple of my diet, by the way. Guest: Well, so a box of cereal lasts for quite a while, and it's very competitive. It's a very homogeneous commodity. I may go to one store or another for a difference of $.30 or $.40. I remember that my particular kind of corn flakes is cheaper. So that's very competitive. But if I'm going to buy cabbage, I'm going to buy cabbage. It isn't all that expensive. The cost from the farmer isn't that high. But it's about three or four times the cost that the grocery store paid for it. Russ: I don't know. By the way, that's exactly the opposite of what Michael Pollan claimed. Michael Pollan claimed that the high markup items are the processed cereals, etc.; that's where they make all their profit. Not on the vegetables. You are claiming the opposite? Guest: Well, let's compare apples and apples. Michael Pollan's claim, if I understand it, is that oatmeal is not very processed. And so oatmeal is much less expensive than corn flakes, which are very highly processed. And so oatmeal is down below and has a much lower price. So there you are comparing two kinds of cereals. Russ: He's not just claiming that corn flakes have a higher price. He's claiming they have a higher markup. A higher profit for the store. Guest: But compared to oatmeal. Which is a less processed kind of cereal, but both of them are dried and in boxes. Anything that's green has an enormous markup. Russ: You are saying because it spoils. You could lose it, throw it out. Guest: Yeah. Russ: So, again, that comes to my other point--just the costs are harder to see. That's fine. Guest: Yeah. And so, it seems like they didn't pay much for this cabbage; I shouldn't have to pay so much for it. But they throw so much of it away that that's [?]-- Russ: And they have to water it, by the way. They do. Guest: They have to handle it. Russ: No, they water it. They spray it every few hours to keep it fresh. Guest: Well, sometimes you reach in there and the sprinkler comes on. Russ: Yeah. Guest: It turned out that the stores were thrilled with this. And so a number of them have actually implemented this. Russ: The mirror. Guest: Yes. It increased their profits. Russ: Because it got people to buy the higher-margin cabbage? I find that-- Guest: It actually, well, I'm quoting an interpretation of a study. Who knows if that's actually correct. Russ: Interesting. Guest: The person mentioned it as something that was interesting. It was an unexpected result. The two things that they noticed were: overall, people did not spend more at the grocery store at the ones that had these mirrors. So it was a controlled experiment: some of them had the mirrors, some did not. The ones that had the mirrors spent about the same total amount per trip as the ones that did not have the mirrors. However, the ones with the mirrors spent more on fresh vegetables--cabbage, things like that. And the claim was that the profit is slightly higher for the ones that had the mirrors because everybody else was just buying packaged, relatively staple[?], lower-profit foods. Russ: Ehhh. Call me skeptical about that one. It reminds me of the Kingsly Amis line, which is immortal, which is: Inside every fat person is a fatter person trying to get out. So you look in the mirror and you think: I'm not as fat as I thought I was. I'm going to get some more candy. I don't know. Could go either way, it would seem to me. I'm not so sure. I will say for the record that I have not bought a piece of candy at a check-out counter in at least 2 years. So it's possible to avoid it. And also, by the way--I'm going to press you on this, Mike--I don't buy those magazines there, either. Do you? Guest: Well, there's no point. Usually I'm in line long enough I've already finished reading it. So I'm not going to buy it. Russ: You treat it as a lending library. That's nice. Guest: I have to know if JLo really is going to get back together with Marc Anthony. Russ: I can't sleep. Whoever they are. I know who JLo is; I don't know who Marc Anthony is. Guest: One of her many ex-husbands. Russ: Good to know. See? This show is full of value.
56:06Russ: So, on this nudge thing: some people would like, just like they have banned trans fats in many places, they want to ban large, sugary drinks. They want to ban all kinds of things. One argument is that: let's put the juices at eye level and the soda near the ground. Right? I guess. Let's put the fruits and vegetables as the impulse purchase. Let's put a bunch of plums at the checkout counter instead of candy. Guest: Yep. Even though that's impossible because some of them you have to keep cold, or you have to water, as we said before. It's just not set up for that. Russ: Well, they'd have to rearrange it. They'd put a little cooler there, just like they do with the other stuff. It would be nice. And actually, you know what else? We'd put economics books at the checkout counter instead of those trashy magazines, so you could get smarter. And maybe some Shakespeare. Guest: I don't think you should open a grocery store, Russ. That's an idea whose time will never come. Russ: I think you are right. Two things I want to mention before we close. There's a nice post at the blog Modeled Behavior on this topic. I think the poster was Adam Ozimek--I'm pretty sure--talking about his own personal experience in the grocery business and challenging Tyler Cowen, a frequent guest here, about his interpretation of the milk thing, which is similar to Michael Pollan's. If I remember correctly; I'll check. And I also want to mention one of my all-time favorite articles at the Library of Economics and Liberty, which is "Everybody Loves Mikey," which was written by Michael Munger, which is related to this issue of consumer sovereignty. When I went to search for it in the middle of this podcast, I put 'everyone likes mikey,' and I misspelled 'mikey' and I put 'likes.' But the actual title is "Everybody Loves Mikey," and Google is so smart and because everyone really does love Mikey and doesn't just like him, it said: Did you mean 'everyone loves mikey'? And it found it for me. Guest: It saw through it right away. Russ: Yeah. So I think we're done. Do you want to say anything else about groceries? Or milk? Guest: Yes, I do. In today's Wall Street Journal there's an article that Mexico, the entire nation, is considering putting up signs at the entry to grocery stores that will show the amount of sugar, and just a picture of 12 teaspoons of sugar, as you walk into a grocery store: This is how much sugar is in a soda. In the hopes that they will nudge people towards buying less soda. They have a big obesity problem; they are actually--Mexico, portions at least, have a bigger obesity problem than the United States. And they are hoping that just showing people, just information, about how much sugar is in soda will have an influence. That's an interesting question, if all you are doing is providing information. Now obviously you are competing with other kinds of attention; requiring grocery stores to do this, I think, is pretty controversial. But it's a nudge attempt. It's not really increasing the price in any way. They are not taxing it. They are not outlawing it, like New York City did. They are just saying: Do you know how much sugar you are eating? Russ: No, it offends me not because it lowers consumer wellbeing dramatically. Right? It offends me because it treats me like a child. It assumes I don't know that sugar is part of the problem. I think they'd be better off, by the way, driving trucks around town blaring the Gary Taubes podcast on EconTalk, the episode about why we get fat. Maybe that would have a more effective, positive outcome. The other thought I have is that warnings get forgotten very quickly. You see it over and over and over again and then it just becomes part of your mental landscape that you just edit out. Guest: Probably just 2 or 3 times and it's gone. Russ: So they'd have to change it constantly. Guest: Yeah. Yeah.

More EconTalk Episodes

Search Econlib