Fun with Pricing!
I’ve been lately pretty interested in the issue of price transparency. Prices in most consumer-facing industries tend to be presented as WYSIWYG – “What you see is what you get”. Haggling is generally confined either to business-to-business transactions, small service businesses, and the informal sector. It nominally has no place in large businesses selling to consumers. This is not really true, and there are two great articles I read today explaining how it actually works.
The first is Timothy B. Lee’s on Comcast’s pricing structure. The ups and downs are rather confusing if not totally indecipherable, but one key takeaway is that despite what Comcast publishes, its actual pricing structure is unpublished. The list prices are their starting point in negotiations with customers, and they are completely open to haggling. Mr. Lee found it incredibly frustrating, which makes sense, as it totally defies expectations for a large customer-facing business and seems pretty unprofessional.
The other extreme is an article about the failure of JC Penney’s new “fair and square” pricing strategy. The author attributes the failure to the inferior value prop of “everyday low prices” against the “discounted value” of many department stores. JC Penney’s “fair and square” is about as far away from Comcast’s as you can get, not even offering the sales that distinguish most department stores. However, the frequent sales and discounting its competitors engage in is different still.
Department store pricing really does come out to haggling, but rather than the Comcast bilateral negotiations the negotiation is entirely unidirectional. Prices are set by corporate, or by the store manager, and the consumers respond by either buying or not buying. Frequently they are willing to buy a given item but waiting for a better offer that may or may not come. In most respects, this is how price discovery is supposed to work – businesses try selling their products at a range of prices and they settle on what the market will bear.
Except at least in retail, this tends to work extremely inefficiently. SKUs (stock-keeping-units) tend to turn over very slowly, and often a single store will turn over one SKU or less per item a week. This doesn’t really generate the information needed to set the price at an optimum level before seasonality kicks in and everything ends up getting slashed to low, low prices. Big chains turn over a lot more SKUs in aggregate…but that’s not really useful data in setting prices at a store level, as there are a host of reasons a given SKU might move for different prices in Manhattan vs. Fresno.
This is a market inefficiency hurting both the retailers and the customers. Many customers buy too high – you need to devote a lot of time and effort to arbitraging retailer’s non-transparent pricing moves. Retailers generally can’t price at the revenue-optimizing price, since they lack the volume-enabled data to do so despite regularly tweaking the sale price. I don’t really have a solution here, other than that retailers need to figure out a way to abstract across SKUs to figure out what can move what volume at what price. Optimum pricing is, I think, overall a positive-sum solution – except for the current minority of customers who are “winning”.