Yes, demand curves can slope upwards

A paper with the title "Giffen Behavior: Theory and Evidence" may not get your heart racing with excitement, but here is one of the most interesting pieces of economic research I have seen in a while.  Many years in the making, this paper by my two Kennedy School colleagues Rob Jensen and Nolan Miller demonstrates for the first time the existence of something that economists knew was theoretically possible, but had never been observed to date: demand for a commodity that is upward sloping. That is, a good for which demand goes down when you lower its price, or for which demand goes up when you increase its price! Such a good is called a Giffen good, and anyone who has had to suffer through introductory economics has heard of it at least once as a theoretical curiosum. Economists' urban legend has it that potato was a Giffen good during the Irish potato famine of the 1840s, but as Jensen and Miller note in their paper, this is highly unlikely.

How can demand curves slope upward? Imagine a very poor household who spend a very large part of its income on some subsistence commodity such as rice. Now suppose that the price of rice goes down for some reason. How does the household alter its pattern of consumption? One response is to increase consumption of rice at the expense of other things because rice is now cheaper. This is the standard substitution effect that ensures demand curves are downward sloping. But another is that the household's overall purchasing power--its real income--has increased and therefore the family may choose to consume things, such as meat, that richer families tend to consume. If this decrease in demand for rice outweighs the substitution effect, we have a Giffen good. Put differently, the lower price of rice allows the family to satisfy its nutritional needs by a tastier combination of products, one that relies less on rice and more on meat.  

As this example illustrates, a Giffen good is more likely to exist in a situation where households spend a large part of their income on a subsistence good. So Jensen and Miller went to Hunan and Gansu provinces in China, randomly subsidized the price of rice and wheat, respectively, for poor families for five months, and then observed the results. They found that the poorer families in both provinces did indeed exhibit Giffen behavior with respect to the subsistence crop. The price subsidy led to reduced consumption of rice or wheat, and its removal to more consumption.     

OK, in case you were tempted to extrapolate from this result, let me be clear that it doesn't imply that the way to reduce demand for energy is to subsidize the price of oil, or that improving the U.S. trade balance will require an appreciation of the dollar.

UPDATE: Nolan Miller tells me that a 2005 article in The Nation, referring to an earlier version of the Jensen-Miller paper, makes the extrapolation to gasoline that I said in my last paragraph would be dicey. 

Comments

[James V. Koch]

What's new here? Didn't most of us learn about the possibility of Giffen goods in our micro principles class? The real question is how often the income effect drowns out the substitution effect. (Not very often, it seems.) Hence, other than substituting rice for potatoes, I'm uncertain why the hubbub.

[Walt]

James, it's been controversial as to whether "not very often" is really "never". This is the strongest evidence to date the answer is not "never".

[W Razzak]

Many housing agents in New Zealand tell stories about new expensive homes that were advertised for sell at very high prices did not sell for months during the past few years, where the housing market has been boiling. Then they decided to increase the price. These homes were sold in one week. I thought these high price homes are examples of a Giffen good.

[Leo B]

W - goods of this sort, where utility depends directly on price (perhaps because price signals quality), are known as Veblen goods, and are different from Giffen goods. (Other examples are wine and fine art.) As they say in a footnote to the paper, Giffen goods only assume an income and substitution effect -- prices only matter through their effect on budget sets.

[Mike Lazzaro]

Perhaps I'm missing something here, but this seems like a simple case of an 'inferior good' situation -- as real incomes rise, buyers shift away from inferior products (rice) to superior products (meat).

Again, I may be misapplying some of the basic assumptions about Giffen goods here, but, to me, it would be more impressive to see an actual case where the value of a product is inversely related to its price, as opposed to simply having an upward sloping demand curve.

For another example of upward sloping demand curves that really have nothing to do with Giffen goods, one could look to Pankaj Ghemawat's "Order Backlogs and Strategic Pricing: The Case of the U.S. Large Turbine Generator Industry" -- in this case, the effect is caused by an umbrella pricing strategy being employed by one of the players. The result? An upward sloping demand curve.

[Barkley Rosser]

Mike,

All true Giffen goods are inferior goods, very inferior goods, so much so that their inferiority outweighs the substitution effect.

George Stigler argued that there were no true Giffen goods. Giffen himself was referring to bread in Britain in his original work, not Irish potatoes, as so many misinformed textbooks have long repeated. A problem with the historical data is identification, separating out supply and demand shifts, with changes in supply affecting incomes of farmers as well as price.

Of course another area where we see something that some think is evidence of Giffen goods is in markets with speculation. We see it all the time with bubbles and their ending. Most would argue those are not Giffen goods because demand is shifting. But, they are cases where the only exogenous thing changing is the price, with change in expectations engogenous to the change in price, which does make things a bit messy. I even had a paper on bubbles rejected from a journal once because the referee quoted Stigler on how there are no Giffen goods, and this business of demand rising when price rises in a bubble looks like a Giffen good.

[Ali Sohail (pakistan)]

The piece of research presented above is extremely insightful and it reaffirms the point of view, which exemplies the existense of a diverse base of the so-called facts and thoughts present in the subject field as there is no one, uniform policy, theory or think tank in the field. As always, phenomena's keep changing,reality keeps reshaping itself, nothing is stagnant, nothing is definitive, everything is relative and a fact based on a certain time line, condition, fruit of thought, perspective and paradigm.
Therefore a pessimist would say, in a broader sense it simply implies that, the use, study and development of science whether it be natural or social is another mode we use to divert our enerigies and give meaning to our lives hence useless in true terms.However, an optimist would state that this provides a stronger,deeper hope and feeling to explore the underlying natural forces which define reality and the dynamisim in the field provides it a rich, evolving and never ending base to work on!

[Isaac Crawford]

I'm a little confused. Surely, at best, a demand curve can rise for a part of the overall curve, not for the entire thing. After all, as something becomes more expensive, the less you can buy given a finite budget. I'm interested in feedback on my posting about demand curves here:http://www.isaharr.com/ratchoice/demand/page9.html

I think I've got this pretty much right, don't I?

Isaac

[Michaela]

Hello,I would have one question. How to explain that demand for flats it is raising when prices of flats rise as well? Upward sloping curve is sign of inferior goods, but flat is not inferior?thank you