Bryan Caplan  

What Will the Supermarket Look Like After the Blizzard?

Mood and Macro... Becker on Adverse Selection vi...
Since there was a lot of interest in explaining supermarket shortages the night before the blizzard, I'm back for round 2.  Here goes:

I'll probably go to the supermarket tomorrow.  What will the supermarket look like?

a. Still stripped bare.  Due to the blizzard (and "fair" pricing), demand will still far exceed supply.

b. Business as usual.  Demand disruptions (plus pre-storm precautionary demand) will roughly balance supply disruptions.

c. Over-stocked.  Since the main roads are clear, but many residential streets remain unplowed, demand will be reduced much more than supply.

Extra points if you can derive your prediction from your initial theory about the pre-blizzard shortages.

Update: (a) holds for the local Giant.  (c) holds for the local Costco.

Comments and Sharing

COMMENTS (18 to date)
ChrisA writes:

My theory is that in times of emergency the supermarkets like the visual cues that the empty shelves give as this encourages further panic buying. I have seen this at Christmas time. Higher end stuff will go first as the shoppers that are more prone to panic buying are also the ones that have less interest in shopping efficiently. Efficient shoppers buy the non-brand names (by definition) and either already have stocks bought at lower prices, and/or are less prone to panic buying, as they weigh up the cost of potential waste versus the cost of going without a staple. This is certainly the difference between me and my wife anyway (she is the efficient shopper not me).

In normal times, supermarkets like full shelves, this gives the impression of plenitude, signalling that prices will be competitive. It also signals quality, akin to the signalling that banks do with
their grandiose buildings.

Therefore whether or not the shelves are stacked tomorrow will depend on the prevailing mood, if the stores sense that it is still "an emergency" then they will leave the shelves still partly empty to encourage further panic buying, if the mood is everything is back to normal, then the re-stocking will happen quickly.

Joe Torben writes:

Tyler Cowen explained this well in a post: stores have a huge variety of goods because the marginal products may get marginal customers to visit just because of those items. Tyler even listed the three or four products that were essential to him.

An occasional blizzard doesn't change anything, so assuming no delays in the delivery of goods, the supermarket should look exactly like before.

Local conditions make it impossible to predict with certainty how any one store will look today, but in a week's time, things will be back to usual.

Adam writes:

Missing items on shelves as long as roads are messed up. No residual demand effects. Rapid recovery with full stocking shortly after roads clear.

dave schutz writes:

My wife went yesterday (on skis, to the Va Square Giant, from which many George Mason staff get fed...) and it was fully stocked, nearly empty of customers.

My guess is, fully stocked. For one thing, people are still working off the huge amounts of food they bought before the storm, for another, it's still work to get there from houses on the side streets.

Lauren writes:

I agree with many of the comments here, and on your previous post, including comments about restocking in the face of suddenly-predicted emergencies and about Wegmans' overall incentives.

I'm surprised that no one mentioned above or in your previous post that last weekend was also the known weekend of the Super Bowl.

The SuperBowl was an event predictable well in advance. It was a weekend for which supermarkets anticipated, pre-ordered, and stocked up on certain kinds of foods related to this annual home-party, wings/pizza/dip/cheese/bread/sandwich/family-coming-to-town-for-the-weekend event. Those pre-ordered foods filled a lot of shelves and advance orders. Balancing existing shelf display and inventory usage between predictable and unpredictable events is probably one of the hardest tradeoffs that supermarkets make.

That consumers who may already have been planning to shop for these staples for the Super Bowl might have flooded or been nervous about and timed their trips to the stores a day or so early when the storm prediction was added on top of that, should be no surprise.

No doubt Wegmans and other savvy supermarkets left room for other last-minute possibilities when they stocked for the Super Bowl--including their standard tradeoffs for weather-related crises. Contrasting food demand for events known months in advance, versus food demand for events known a few days at most in advance, is a trade-off I think supermarkets like Wegmans do pretty well and learn from for the next time. The blizzard in the Mid-Atlantic area last weekend was a very unusual weather event. It was not something a supermarket could--or should!--plan for on a regular basis.

A contingency plan by supermarkets such as Wegman's that was probably a plan for handling anything weather-related within 2 standard deviations was apparently enough to accommodate a weather event that was outside the 2 standard deviation range even on Super Bowl weekend. Two--not one--major events came down on the same weekend. Some brands people may have planned to buy for the one expected event ran out in the face of private contingency plans for an additional unexpected event.

Yet, no one ran out of food. Neither will you run out of food tomorrow. That's the really big news.

What inferences might we draw from this extreme confluence of events, events outside 3 standard deviations in weather combined with the plans by supermarkets to stock for the food demands of the Super Bowl? Broadly sweeping inferences, perhaps none. But narrower inferences, yes, we can draw. You will neither run out of food nor find the shelves so overstocked that the stores are undepletable. Some middle ground will be there for you today and tomorrow. Maybe not every brand you want, but you will have food in every category you want and food galore. Every supermarket, restaurant, and their suppliers have been working overtime since this snowstorm was predicted--and are working still while you have been unable to get out to the supermarket, much less plow your own driveway or shovel your own sidewalk, slept, or may even had power outages--to make sure you have food, water, and supplies if you can get there by foot, car, ski, or sled. Super Bowl, Haitian horrors, blizzard, or whatever may happen.

Marc writes:

D) stocked as normal.

Demand will be normal since there will be people that don't need to buy, because they purchased prior to the storm; there will be people that need to buy, because they bought prior to the storm, lost electricity and their food perished; there will be that need to buy, because they did nothing prior to the storm.

These will balance out and no business owner is going to purchase more than they can sell, so it will look the normal.

Paul writes:

Fully stocked.

Distributors want to bill as soon as possible. The minute they drop off the goods, the clock is ticking.

Marcus writes:

In the previous post Bryan wrote, "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."

This is where your reasoning is flawed.

Just because product A is 10 times more popular than product B does not mean that the store has on hand at any given moment 10 times as much of product A as product B. This is because a store has to order product by the case combined with the fact that trucks come in usually twice per week.

So, if product A is 10 times more popular than product B then all that means is while I might order 2 cases of product A per truck I might only order 1 case of product B every 5 trucks.

This is especially true when it comes to the amount of product on the shelf because stores try to carry a wide variety of products and stock crews come in every night. Because of these two factors there will not be 10 times more of product A on the shelf than product B. Everything is guaranteed at least one facing and a popular product might get two facings (an insanely popular product like Miracle Whip will get several facings). Grocery stores have to do this so they can have a large variety of products.

My prediction is, assuming the trucks continued to arrive on time, the popular products will be fully stocked and the less popular products might now be under-stocked. This is because the popular products will have been on order for the next truck where as the less popular products might not have been.

tjames writes:

Marcus has pretty much nailed it.

I would offer a small variation. Supermarkets generally order product under the assumption that the upstream supplier can supply any quantity they might reasonably ask for at any time. In the event the storm caused sudden buying that exceeds the ability of the upstream supplier to replace immediately, then you might still see bare shelves for some items. Such are the vagaries of just-in-time production and delivery.

zbicyclist writes:

Demand disruptions hit particular categories more than others: people want to buy batteries, they don't want to buy frozen food.

Supply disruptions (due to trucks not being able to deliver, etc.) will more evenly affect various categories.

If trucks can deliver, the stores will be normally stocked with only spot shortages -- they might have only 1 or 2 SKUs of D batteries instead of the normal full stocking. It may be hard to find a snow shovel.

spencer writes:

The stores will be stocked very close to normal levels.

But the interesting thing is that this will happen without any significant changes in prices, completely contradicting the theory about price gouging that Bryan and other libertarians teach in their class.

Where is the guy in the pick-up truck that their theory makes so much of?

Marcus writes:

Spencer, the stores did in fact sell out, supporting 'price gouging' theory as you put it.

I wonder how much of it will now rot in cupboards and refrigerators?

spencer writes:

Marcus, don't you know that Bryan teaches that there is not such thing as a price gouger.

Marcus writes:

Spencer, *you* are the one who called it 'price gouging' theory. I was using *your* words.

Stores didn't raise prices. Stores sold out.

Presumably, Bryan would argue that stores should raise their prices when demand shoots up. Other people, people like you apparently, call that 'price gouging'.

You said what has happened contradicts 'price gouging' theory. I'm saying that it does not, evidenced by the fact that stores did in fact sell out.

Which means that in all likelihood, resources were not distributed to their most valued uses.

I further speculate that much of those resources will now sit in cupboards and refrigerators and rot.

All because stores didn't raise prices (i.e. price gouge).

alex s. writes:

Based on my assumption from the last post (that much of the abnormal (and inefficient) demand was due to shoppers replacing meals normally eaten outside the home), I would think the shelves would be relatively bare. On the other hand, people that shop normally and are able to plan a little bit have been consuming their own safety stocks, reducing short-term demand. On the third hand, the supply chain may have been running more slowly and inefficiently because of the storm.

In the end, I pick (b), that things will balance out, with the exception of storm-demand items (snow shovels, firewood) and things with the shortest shelf life (which might have perished even when the stores weren't open). I would also think that a two or three day snow storm in winter would not be the biggest shock to the grocery store supply chain.

Having seen the update, it doesn't surprise me that Costco has over-supply, since I would guess their typical customer is more likely the "buy a pallet, keep a spare in the garage" type than the "get just enough to make it to my next meal" shopper. Some demand is probably transferred as local stores substitute for a (potentially longer) Costco trip.

Also, I think you might get a lot of variance from store to store based on very localized logistics (main streets might be open, but some stores might depend on alley or hill access for supply trucks that would take longer to clear).

BrittanyF writes:

I think that if you know a storm is coming you should prepare before hand, not wait until after it happened. Of course there is not going to be any food left because everyone has already gotten all the good stuff. My friend decided to wait the last snowstorm out and get food later that day and she went and there were only two cans of soup left and all the bread was gone. This was in a very small town in NC at a Wal-mart. So I think that people who wait around should expect to not have anything at the store waiting for them after the storm. Not to mention that the workers and trucks delivering the food can not make it to the store to stock up due to the storm.

spencer writes:

Marcus, than you believe that consumers would be better off if the store raised the price to the point where enough people would not buy so that the stores would remain fully stocked.

The customers welfare is maximized by keeping the store fully stocked. Care to explain how that works.

The Cupboard Is Bare writes:

I live on Long Island. We were supposed to get 4-8 inches, but instead got nada, zip, zilch. Since it was supposed to snow during the overnight, no one knew the storm was going to be a non-event until Saturday morning, so you still had people freaking out and running to the stores on Friday.

Knowing that we were on the edge of the storm, we understood that a small shift either way could result in us getting anywhere from nothing to a foot. We decided to fill up one car that was on "E" late Friday night, and the place was busy, but not crazy. Since we actually like driving around in the snow, we decided to shop on Saturday, because we knew that the stores would be empty.

As predicted, the stores were empty, because everyone shopped the night before, but since the stores were stocked for the weekend and the SuperBowl anyway, there were no shortages. Things might have been different if the storm had hit in the middle of the week, as is the case for us with the current storm that is now going on.

As for charging more for food when demand increases...we have so many supermarkets with a few miles of our house, that the stores would have to be crazy to raise their prices. Everyone would simply get ticked off and go to the next store. New Yorkers are like that. :)

Comments for this entry have been closed
Return to top