Bryan Caplan  

A Puzzle from the Blizzard

Problems with Means Testing... Survey on the FDA...
A blizzard is about to hit DC.  As reports of its magnitude spread yesterday, people unsurprisingly rushed to grocery stores to stock up.  Stores unsurprisingly failed to raise prices to cope with this sudden demand shock.  By the time I got to the grocery store last night around 11 PM, many of the shelves were unsurprisingly empty.

Many, but not all.  They were out of milk and bread, but there was still plenty of cheese and chocolate.  That was easily explained - people knew they could shop again in a few days, so they only needed to stock up on staples.  But the more I looked around, the more puzzled I was.

Here's what I noticed: For any given type of product, the most popular brand always sold out first.  There were no Eggo waffles, but plenty of Wegmans brand waffles.  All the national brands of hot dogs and sausages were gone, but there were plenty of obscure sausages still on the shelves.  If you broadened the categories, the pattern remained.  In produce, all the bananas were gone, but there were still plenty of apples.

You might say, "What's the puzzle?  Of course the most popular stuff sells out first."  But that's a feeble explanation.  After all, if X is ten times more popular than Y, then you'd expect stores to simply carry ten times as much X as Y.  Why would X sell out faster in a blizzard if stores have already taken its greater popularity into account?

Comments and Sharing

TRACKBACKS (1 to date)
TrackBack URL:
The author at ¬†Modeled Behavior in a related article titled The Economics of Supermarket Shelves writes:
    Bryan Caplan notices that during a blizzard the name brand items sell out ten times faster than the same item from a generic brand. This puzzles him: Of course the most popular stuff sells out first.”¬† But that’s a feeble explanation.¬† Afte... [Tracked on February 6, 2010 10:52 AM]
COMMENTS (39 to date)
John Thacker writes:

For non-perishables, just-in-time stocking means that the store doesn't have ten times as much on hand at any one point for something that sells ten times as much in a month.

Can't really explain bananas and apples, though.

mch writes:

Flight to quality.

masterleep writes:

A different kind of shopper is coming into the store. One that does not know how to shop.

Bob Murphy writes:

Not saying this is the primary factor, but my first guess is that the type of person who buys the more expensive name brand, is also the type to err on the side of buying too much when a blizzard is coming.

In contrast, the people who scrimp and buy the off brands aren't as likely to say, "Hmm let's buy 3x as much as we normally do."

Scott Wentland writes:

Is name-brand purchasing correlated with income? It might (or not, I don't know), and maybe higher income people plan better and show up disproportionately to the supermarket that day.

Another (possibly less plausible, though logical) economist explanation: planned consumption expenditure goes down over the next two or three days. Stuck in the snow storm, you're not going out to dinner, movies, or any other expensive places. You're staying home. So, why not eat the higher quality supermarket stuff as a substitute?

dWj writes:

Bob Murphy is on a similar line to where I was going; price sensitive folks don't respond in proportion to how price insensitive folks do.

I'm pretty price sensitive -- grad student -- and I buy a lot of generics; I also buy a lot of items on sale (for certain items I have a pretty good idea how often I'll see it discounted how much). One consequence of the latter behavior is that I'm always prepared for a blizzard because the last time pasta was cheap I bought a bunch of it, and the last time tuna was cheap I bought a bunch of that. If the people who buy generics are also the people who stock up on sale items, and the people who buy brand names are too busy to do intertemporal arbitrage, that could lead to the effect you're seeing.

Stefan writes:

How about turn rate is higher on popular brands? Stores need to restock daily and they restock a higher fraction of the best selling brand. Per unit of sales of the more popular brands have less shelf space even if they have more total shelf space. Indeed, that's one of the economies of scale for brands: you need to pay for shelf space and popular brands sell per unit of shelf space paid for.

Bill N writes:

They may simply get more frequent shipments from the faster sellers.

Paavo Ojala writes:

The store probably normally doesn't want to run out of any product, so they have more of some product that they can sell before it expires.

For popular products the demand is more predictable, so they can have smaller overcapacity of that product.

And for apples and bananas the explanation is simple: apples have longer shelf life than bananas.

Twba writes:

The Eggo waffle shortage is unrelated to the snow storm.

rjs writes:

people are thinking short term:

apples have a longer shelf life than bananas...

Mark writes:

It's the Washington, DC area - status is all that matters. I can't have snowflake's friend, kitten, spending the night and serve her Wegmans waffles in the morning. How gross would that be?

Eric writes:

I think this may be a form of pricing. In a period of short term high demand you would expet reatilers to raise the price until the substitutes were equally valued and purchasing evened out.

However, Wegmen's is concerned about being labled as a "price gouger." Therefore insted of raising the price of demanded items they limit the quantity avaliable forcing late shoppers to the alternatives.

Robin Hanson writes:

This is a great question. I agree with others commenters above; my guess this is about sloppy vs. careful shoppers. Careful ones buy cheaper stuff and already have plenty of food at home, and know they don't need that much more food now. Sloppy shoppers go wild both with quantity and quality.

Dave writes:

Relatedly: why is the sale rack always full of small/medium men's shirts and size 8 men's shoes?

Most of the comments seem to attribute these phenomena to a quirky demand side. I like the just-in-time story, but maybe there's a minimum stock needed to carry a prodcut and it's still efficient to run it at a lower turnover to the popular brands?

aub writes:

"if X is ten times more popular than Y, then you'd expect stores to simply carry ten times as much X as Y"

I think there are a few assumptions that may not be valid.

1. You're assuming that the stores can obtain all they want of a popular item. If all the stores in an area try to buy much more of an item for an unplanned, one time event, the distributors and producers may not be able to supply that.

2. You're assuming the stores have room to store the additional items if they can get them.

3. You're assuming that the stores want to sell more of a popular item. For example, they may get higher margins on the generic goods. Or they may just want to reduce their inventory of low-turnover items.

4. You're assuming the stores can afford to purchase more of an item. Supermarkets have very low net margin. Even if the store has enough cash/credit to purchase more of an item, keeping enough stock on hand to keep from running out during this one event means that they will have higher spoilage. Why take a good (cash) opportunity and turn it into a bad (overstocked) situation?

SB7 writes:

My vote is with John Thacker, Stefan and Bill N: it has to do with restocking. All you can observe is how much shelf space each item takes up, not how often they are delivered to the store and restocked.

It doesn't explain the fruit, but brands also pay for shelf space in stores, which may skew your experience.

David writes:

I would imagine that it has a lot to do with people's mentality when "stocking up". They're not looking to try something new. They are looking to quickly get all the food they think they'll need for the next few days. I would imagine that all else being equal, this increases people's propensity to buy big name brands. These brands also tend to have more advertising, which breeds comfort and familiarity in their consumers. Of course, that doesn't explain apples vs. bananas. For that, my best guess is that people are more likely to put bananas in cereal and pancakes (or peanut butter and banana sandwiches) than they are apples, so when stock up on these sorts of staples, bananas are more complimentary. Apples are primarily eaten by themselves or as part of comparatively complex dishes--which are again not what people are looking for during a "crisis".

Sam writes:

Congratulations, you've discovered brand loyalty.

Brian Shelley writes:

Also interesting is consumer behavior before hurricanes and tropical storms. Living in Houston I see this phenomenon at least a couple times a year. Go to Wal-Mart and all bottled water will be ravaged. Right next to it will be crate after crate of Gatorade, sodas, Propel, etc..., which need no refrigeration. People will go home empty handed instead of buying something other than water.

Joe writes:

Could it be related to the fact the grocery stores sell their shelf space. A product might have a lower turnover rate, but with revenue from selling the shelf space it might be as profitable as more popular products.

Ben Hughes writes:

It's possible that Wegman's overstocks their own brand to convey a sense of importance or quality. They probably know that if they only stocked proportional to demand, the miniscule amount of their product would, in and of itself, make it less in demand to prospective customers. It's a similar to how when walking around for a restaurant people tend to shy away from restaurants with no one in them.

This explanation of course only makes sense if their profit margin on their own items is higher than other brands, which isn't unrealistic.

John Thacker writes:

"Relatedly: why is the sale rack always full of small/medium men's shirts and size 8 men's shoes?"

Virginia Postrel has discussed this phenomenon in the context of women's sizes.

The short version of the story is that during a sale odd sizes are always left over, but yet people who wear odd sizes are still less likely to find fitting clothes at any time, including sales.

That's because while a lot of people wear odd sizes (the modal size isn't all *that* popular), they tend to wear different odd sizes. So if you wear a common size, you'll notice during a low inventory situation that the only sizes left are these random small and large sizes. However, if you actually wear a unusual size, say 29-30 pants or a 38 short suit jacket, you'll notice that stores never have much of your particular size in stock, and are often out, whether they overall have a lot of stock or not.

alex s. writes:

I was going to go with the "bad shopping" argument, but I like Scott Wentland's idea better: a significant percentage of the buyers in this situation are buying to replace meals eaten out, so they are less price sensitive. It's also likely that people who don't shop that often are more swayed by brand names.

I suspect the apples/bananas issue is due more to supply than demand (i.e. bananas are just harder to keep, so the store didn't start with as many in stock).

floccina writes:

My guess is that the stores devote a minimum of space to each product. For example a store might want 7 days supply of each item in the store but if they only sold one of an item per week they might stock more than a weeks worth because it looks bad and is bad to be completely out. The lower the volumne greater the variance in percent.

Matt writes:

Maybe the people who buy off brands have an economist streak in them. They don't buy name brands because they have accepted that name brands won't justify their marginal satisfaction, and they don't go to the store before a blizzard because they know that going to the store after the blizzard is much more pleasurable and probably less dangerous (fewer cars on the street).

I'm speaking about my own personal experiences. We had a few inches of snow in Raleigh last week and I went to the store and they were empty (but they were all open)which was fantastic. I've never felt safer driving because I mostly had four lane highways all to myself.

beamish writes:

People prefer going to stores that have more variety. Stocking less popular brands is a way of luring shoppers to come buy the more popular brands.

topgun writes:

This could simply be because supermarkets tend to display the popular brands in their inventory more prominently in the store shelves. The suppliers of the products are also probably paying for that prominent spot on the shelf . Careless shoppers in a hurry to buy up stuff simply pick up the first items they see on the racks.

Shnugi writes:

It's because of the kids. You have to stock-up on the things that they'll eat, and children are some of the most brand conscious monsters alive.

Art Carden writes:

Hmmm. A couple of thoughts:

1. Price discrimination. Inelastic demanders (likely represented by panic buyers) go for brand names.

2a. Bananas might be a complement to panic goods.

2b. People treat the goods on the shelf as part of their risk-adjusted available inventory in the short and medium runs. (I think) Apples have a longer shelf life than bananas and are more likely to still be good (and available) if people are snowed in for several days.

2c. Treating goods on the shelf as part of the risk-adjusted short or medium run inventory might also explain why the name brand waffles went first. The perceived decline in quality for store-brand waffles is likely to be smaller than the perceived decline for name-brand waffles. There might be a very large difference between "excellent" and "mediocre" but a smaller difference between "mediocre" and "barely edible." Given that the waffles were probably produced during the Clinton administration, I'm not sure how much this will matter for blizzard conditions. Twinkie inventories might provide a falsification test, but we would probably expect to see the same thing you saw with cheese and chocolate.

3. Just as US silver eagles trade at a premium relative to generic silver rounds, Eggo waffles will probably trade at a slight premium relative to Wegman's waffles in the event that society as we know it collapses and we're reduced to barter. The wise choice for the survivalist might therefore be the name-brand waffles.

The Cupboard Is Bare writes:

Many people change their eating habits when there's a snowstorm or hurricane. For example, Walmart has discovered that prior to a hurricane, people buy more strawberry pop tarts and beer.

Ted Frank writes:

After all, if X is ten times more popular than Y, then you'd expect stores to simply carry ten times as much X as Y.

Actually, this is false. Given resupply costs, the optimal level of inventory for X that sells N times as much as Y is to carry the square root of N times as much.

Joe Cushing writes:

This is very simple. Inventory turns are greater on more popular items. Just because a store sells 10 times more of item X than Item Y, doesn't mean Item X needs to have 10 times more shelf space in the store to maintain those sales. Under normal conditions Item X is just replaced faster than Item Y. Item X only needs enough shelf space to make it through a typical Saturday because stores get deliveries and restocked every day.

Scott writes:

[Comment removed for supplying false email address. Email the to request restoring this comment. A valid email address is required to post comments on EconLog.--Econlib Ed.]

Adam writes:

Getting snowed in is no fun, so what do people do--pamper themselves. They know their going to have to shovel a lot of snow, so what do they want--some great food! Start with breakfast, all the favorite things. The best waffles (Eggos, not Wegman's), eggs, bacon or sausage, cereal with bananas, jam, and toast. Lunch? How about some of the best brats or hero sandwiches?

Why do the bananas get sold out and the apples left behind? Bananas work well at breakfast by themselves or with cereal, but apples are for bag lunches. Nobody's going to need a bag lunch when they're snowed in for the weekend.

tjames writes:

Speaking as someone who worked in retail grocery for 10 years, this is relatively easy to answer.

The first point to note has already been touched upon by several posters, which is the amount of inventory on-hand is not always correlated to popularity. For high sales volume items, stores generally keep enough to get to the next delivery truck; more is not needed, and has to be stored somewhere if kept on hand (plus excess inventory ties up useful capital). Perishable and/or refrigerated goods pose additional storage burdens so you really want to minimize backstock of these items. Low sales volume items may have weeks of inventory on the minimum shelf space they can be allotted.

There is an additional problem too. What is on the shelves may be only enough to get, not to the next delivery truck, but to the next time someone is on staff to re-stock the shelf from the backroom. Stores often do have excess quantity of their most popular items on-hand in backroom storage, but they are often re-stocked on a schedule (stores I worked at did dry goods nightly, and most perishables once early in the day). When a store is hit by unusually high sales during the day, the shelves may go empty, even while stock sits in the back rooms, waiting for someone to get it onto the shelves.

Brian writes:

"Relatedly: why is the sale rack always full of small/medium men's shirts and size 8 men's shoes?"

Actually this has nothing to do with this issue. Stores are required by most wholesalers and brands to buy multiple sizes. For example one can not by 5 smalls 2 mediums and 1 large, one has to buy a small, medium, and large. So if you want more smalls you have get buy more mediums and large sizes.

Not sure why this is common idustry behaviour, other than the fact maybe they have a hard time projecting size demands for a specific style? It is easyer to make standered ratios of sizes across all/catigories of styles and estimate demand for that style than do both.

Jonathan Sessoms writes:

All these extremely popular brand names such as Eggo and other things like that get all the advertising and tv time and things like that. I have never even heard of the other kind of waffles. So most people are going to go with the item that they are most familiar with. That is why everything has sold out so fast. I live at the beach so I don't understand why people would choose a different fruit from another. We eat everything.
However, If I was getting stuck in a major blizzard, I wouldn't care what the difference was between the name brands. My first priority would be to get as much food as I needed to make sure that my family was safe and had the supplies needed. Anyhow, it still puzzles me why stores wouldn't make a raise in prices. They could have made a fortune with all major supplies running out.

Joe writes:

Bananas & Apples. I think the explanation for the shortage of bananas is simple: (a.) there were more apples to begin with since they are round and hard and stack easily. The odd shape and easy-to-bruise nature of bananas means they aren't typically stacked high in the store display so there are comparatively fewer of them per unit of space. (b.) people buy more bananas at a time because they're sold in bunches of about 6-8 and shoppers are reluctant to break the bunches apart (maybe one time to buy 3-4 bananas, but no one pulls off an individual banana in the grocery store). For my pre-blizzard shopping I bought 2 apples, 2 navel oranges, and 6 bananas. I also passed on the 40 oz. Budweiser and went straight for the 12 pack. So there.

Comments for this entry have been closed
Return to top