Wal-Mart and Disaster Supplies

A few days ago I commented on an article about how Wal-Mart provides emergency supplies to disaster areas at their regular everyday low prices. I pointed out that libertarians like those at Café Hayek should note these developments and see that some guy in the back of a pick-up truck selling supplies at high prices was not the only way to provide disaster victims additional supplies.

Russ Roberts took exception to what I said and we went though a long discussion of it over at his blog. He started out claiming that Wal-Mart had to face higher costs because of the higher cost of carrying inventories. He finally conceded that my explanation of how Wal-Mart had revolutionized inventory management and that when you looked at how the computer revolution had altered the way big-box retailers managed inventories that I was right in arguing that the cost of inventory management did not mean that Wal-Mart had to charge higher prices.


But he was still unwilling to concede that Wal-Mart did not have to raise prices, claiming that this is what theory shows and we have a long history of the world working this way.

I think the difference stems from the way academic economist view supply curves. They look at supply – demand diagrams and see supply curves as continuous curves rising from the lower left to the upper right of the diagram. It is a beautiful theory and is a great tool for explaining to students how the system works in the long run, especially when suppliers have to create new capacity to increase supply. Moreover, continuous curves are required to do the advance math modern academic economist do. So someone seeped in this methodology obviously and correctly believes that higher prices are needed to attract additional supply. If this is always true the guy in his pick up truck charging storm victims very high prices is doing a good thing.

But in the real world in the short to intermediate run supply curve are not always continuous curves sloping from the lower left to the upper right – especially in the short run. In the short run most industries do not operate a full capacity and do not have to bring new capacity on line to increase supply – they normally have excess capacity.

Because of this most businesses in the US economy typically operate on some form of posted price approach. For example if you want to buy a pizza you call the pizza shop and they tell you the price of a pizza is $10. Moreover, the price will be $10 per pizza if you want to buy one pizza or ten pizzas or anything in between . Over this portion of the supply curve it is flat, it is not a continuous curve with an upward slope. If you said to the pizza owner next Wednesday I am having a kids party and want 20 pizzas delivered at three in the afternoon, what kind of a discount would you give me. The owner would think, at that time most of my ovens are sitting unused and my employees are idle. So sure, I’ll give you a discount and sell you 20 pizzas at $8 each. Just don’t try this for the pizzas to be delivered at half-time of the Super Bowl. Actually, if might even work then because Super Bowl Sunday is one of the worse days of the year for restaurants. But anyway, what we have here is a supply curve that actually slopes downward over portions of the curve.

Think of the typical small retailer. They look at the catalog or price list the vendor provides and finds that if they want one or two cases of something the price is set at one level. But normally, if they buys 20 or thirty cases they will get a volume discount. So again, we typically have a supply curve that over certain portions of the curve actually slopes downward. It is not a continuous curve with a constant upward slope.

Imagine you are the regional manager for Wal-Mart and as you do your planning you estimate that next year you should sell 1,000 crates of six-ounce bottled water each week. So you go to the manufacturer and negotiate a contract. But you do not negotiate a contract to buy 1,000 bottles a week. What you do is negotiate a contract with minimum and maximums. You commit to buying at least 800 bottles per week and up to potentially 1,200 bottles per week at the agreed on price. The contract provides a great deal of detail about exact specifications and how the contract can be changed. But it will also call on Wal-Mart to tell the bottler each Tuesday how many bottles from 800 to 1,200 they want next week and calls for them to ship about 20% of the order each day from Monday to Friday. The manufacturer now has several days to arrange his supplies and production schedule. Wal-Mart is not the vendors only customer and typically Wal-Mart will buy from multiple vendors. So it is a good deal for both. But it means that between a volume of 800 to 1,200 bottles a week Wal-Mart faces a flat supply curve. It is not upward sloping.

If you are a academic economist and long accustomed to thinking about supply curves having a continuous upward slope and you see that a natural disaster creates a temporary surge in the demand for bottled water at some location it is perfectly natural to apply your theory and believe that it will take higher prices to attract the additional supply. The big retail chains like Wal-Mart, Home Depot, 7-11, CVS Drug, etc, sell several hundred thousand bottles of water around the country every day and they each have contracts similar to the one Wal-Mart has with bottled water manufacturers . Compared to this total national supply the few thousand bottles needed at a natural disaster site is insignificant. And the amount the proverbial guy in the pick-up truck could supply is infinitesimal . It is very easy for big-box retailers to redirect a small portion of their supply and remain well within their existing contracts where they face a flat supply curve.

When you look at the world like this and realize that supply curves can be flat or actually downward sloping over certain portions of the curve it is very easy to understand why and how modern retailers can quickly increase supplies to a disaster area at regular every day low prices. This does not mean there is anything wrong with the economic theory. It is a good first approximation of how the world works. In the next year if Wal-Mart wanted to sell 1,500 six ounce bottles of water rather then 1,000 bottles the theory based on an continuously upward sloping supply curves should work very well.

Well, maybe not. Ever year since 1992, retailers like Wal-Mart, Home Depot, etc, have actually experienced deflation. The deflator for GAO type products – think department store products or roughly what Wal-Mart sells – has been falling at a 2% to 3% annual rate every year since 1992 . So actually, there is a very good chance that the per bottle price for 1,500 bottles of water may be cheaper next year. So even the long run supply curve may be downward sloping.

So what evidence do I have that this is the way the world works? Experience is the best evidence I have. But there is also a body of research that finds that most firms tend to change prices only once a year. Firms do work on a posted and/or contract price approach and find that in a low inflation world it does not pay to frequently change prices. Most firms have the flexibility built into their systems to moderately change their supplies in the short to intermediate run without changing prices. In the short run the overwhelming bulk of the items sold and the vast majority of firms in the US economy face a flat supply curve. This is not inconsistent with economic theory. Upward sloping supply curves are a long run phenomenon, not a short run phenomenon.

So I will stick with my argument that modern retail chains like Wal-Mart can and do quickly and easily provide extra supplies to disaster areas at regular every day low prices. Consequently, the proverbial guy in his pick-up truck charging the victims higher prices is not doing them a favor. Academics teaching this to college undergraduates are teaching both bad economics and bad morality.