When demand slopes upwards

By Felix Salmon
November 9, 2009
per bottle:

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

At least between, say, $3 and $6 per bottle:

Supplying wine to sell at $5, $4, $3 or even two bucks per bottle is not that difficult once you set out to do it. Cheap surplus grapes, cheap surplus wines, low-cost winemaking processes and economies of scale all contribute to extreme value supply. Nope, supply is easy. The challenge, until recently at least, has been selling the stuff.

Studies have repeatedly shown that wine drinkers are influenced by price – but not in the way you learned in Econ 101. A lower price does not always produce more sales because insecure buyers infer quality from price. They assume that higher price means better wine.

I’d like to see some empirical data here, but intuitively it’s at least possible that raising a wine from $3 to $6 per bottle might increase rather than decrease sales. That seems to be changing, as Mike Veseth explains in the rest of the article. But let’s say it holds true here, or once did. This isn’t a case of Veblen goods: drinking a $6 bottle of wine hardly counts as conspicuous consumption. Is there a name for this phenomenon? And is it found elsewhere?

Update: “Fred Engels“, in the comments, makes the excellent point that there’s another place this phenomenon is often found: the stock market, where demand often rises as the price of a stock goes up.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/

It’s the perceived quality of the goods. For example, Mercedes-Benz let their quality control go to hell back in the early ought-oughts, because of the Chrysler merger. People were still willing to pay the extra for a Benz because of the perceived quality, even though the quality wasn’t there. Or perhaps a better car example was the Japanese versus the American cars in the sixties and seventies. Now we’re seeing that wine can be both reasonably good quality and cheaper. People’s consumption functions do not change as quickly as would be expected for a ‘rational’ consumer.

Posted by datanerd | Report as abusive

What you have here is really two goods, each with their own demand curves. The way I look at this is that there’s a supply and demand curve for cheap wines and another set for expensive wines. Both of these should be downward sloping.

In economics, Giffen goods are those where demand increases as prices increase. http://en.wikipedia.org/wiki/Giffen_good s

Posted by capitalcalls | Report as abusive

capitalcalls: calling something a Giffen good implies a bit of a different mechanism than what’s in play here.

I read of something similar to this happening with “rocky mountain oysters” at a particular restaurant; they tried selling them for 50 cents each and nobody wanted them, but they raised the price to $3 or so and limited them to two per customer and could portray them as a delicacy, and they sold much better. I don’t have a good source on this, though.

I am no microeconomist, but I always assumed it was a given that uninformed consumers (and the seller usually knows more than the buyer in the $2-6 wine market) use price as a proxy for quality. Pretty easy for sales of random-label wine Z to rise as putative price X increase along that range.

Me, I used to buy a lot of $2-5 bottles at the Grocery Outlet. You pour out or cook with most, and buy cases of the finds. If that is everyone’s plan, sales will not increase, but then I always said that if everyone acted like I do capitalism would have collapsed long ago.

I’m not sure there’s a name for it other than, perhaps the “snob effect”. But this kind of thing doesn’t fit very neatly into any macroeconomic model (just like behavioral finance doesn’t square with the Efficient Market Hypothesis). This is a marketing-driven creation, pure and simple — and the biggest and best example I can think of is how Sidney Frank marketed Grey Goose when he launched it (there’s a really good article from NY Mag a few years back about Frank and the power of marketing).

Conspicuous consumption. The higher price demonstrates the wealth and status of the consumer. It’s just like jewelry.

Posted by Neil D | Report as abusive

It is conspicuous consumption to the insecure buyer.

Posted by Neil D | Report as abusive

Your celebrity status is making your head swim, Salmon.

Very funny comments that made my jobless snobless day.

For the maths boffins, from what I recall:- these are supply and demand curve intercepts, not straight lines, and these shift around, quite correct. One could almost call it the ‘Efficiency Frontier’ of Investment Theory 101. Water to wine

Posted by Casper | Report as abusive

So what you call the fact that when prices for financial assets go up demand increases?

Posted by Fred Engels | Report as abusive

either they’re Veblen goods, or quality is uncertain and price is seen as strong enough signal of quality. or some combination.

Posted by Atrios | Report as abusive

Let the record show that you, Felix, were a strong signatory in 2009 to the false belief that the US “cannot default on its debt because its debt is in dollars and it can simply print dollars.”

Now that 6 months have passed since you more habitually spouted this internally flawed argument, do you finally see the problem with this view?

Let me help you out: the US started to willfully default on its debt when the FED made the decision to monetize debt, thus functionally monetizing a portion of the operating budget. The result was fated to be recorded eventually in the currency, and in gold, and you are seeing that now.

Posted by John Brewster | Report as abusive

I have a hunch that significant medical procedures will behave the same way. If it’s more expensive, it must be better/more advanced/more effective/better surgeons. If it’s cheaper, the surgeon must be a hack with a degree from a Caribbean diploma mill, or be using worn-out equipment, or outdated procedures, or whatever.

That’s why I don’t think the answer to controlling healthcare costs is simply to expose patients to the prices and let them choose.

Posted by Jon H | Report as abusive

It’s not conspicuous consumption; it’s a fear of what is in that two-dollar bottle of wine. The same instinct might cause you to steer clear of a 45-cent hamburger: if the price is that low, it’s bound to be wretched. At a slightly higher price, you can at least hope for a better quality good.

Posted by a | Report as abusive

John, where did I write that?

Posted by Felix Salmon | Report as abusive

Way back when, we were taught that this phenomenon is called “reverse demand elasticity” and applies to many luxury goods — in particular, when a consumer doesn’t have much to go on besides the price, s/he will use a price signal as an indicator of quality.

Posted by SelenesMom | Report as abusive