Demand and Suuply curves

This morning I discovered UnlearningEconomics blog, whilst I was looking for information on how labour markets work. As part of the blog, UnlearningEconomics wrote a series of posts summarising the work of Steve Keen in his book, Debunking Economics.

The first chapter, and its associated blog post, is on the demand/supply curve. To help me understand what I was reading, I took notes - effectively summarising the summary - and arranged them into an order that helped me see the flow of logic. As it helped me, I thought it might be of itnerest to others.

There are three problems that cause serious issues for neo-classical economics.

Firstly, demand curves assume that income remains constant. For example, A increases in price.

This renders B unaffordable. By not buying B, the person now has more income (i.e., income has not remained constant) and consequently more of A can be bought (or more of C).

So an increase in price of A can lead to an increase in purchase of A, and a decrease in purchase of B.

Or, it could lead to the same amount of A, less of B and more of C being bought.

Secondly, it isn’t possible to separate changes in demand from changes in income. One person’s spending is another person’s income.

Thirdly, consumers aren’t equivalent to one another. Peter doesn’t necessarily spend his pound in the same was as does Paul. There are a myriad of different ways to spend money. A person could buy more of the same type of good, but could also switch to a higher quality type, buy a type of good not previously bought, pay off a debt, or put the money into savings.


Having established these real-life provisos, let’s look at their impact on market demand curves.

Firstly, market demand curves rely on the assumption that incomes remain constant, and that therefore it is possible to study changes in price independent of other effects.

As our first proviso showed, this is not true in real life. In order to correct for this, the Hicksian Compensated Demand Function is used. This makes it possible to separate out the income effect (that incomes aren’t constant) from the substitution effect (the ‘pure’ effect of price on demand).

Unfortunately, as soon as there is more than one person in the economy, it becomes impossible to separate changes in demand from changes in income, as per our second proviso. This is known as the Sonnenschein-Mantel-Debreu theorem.

To get around this, it is necessary to assume that all individuals spend any additional income in the same way. However, as proviso three shows, this is not true in real-life.


The consequence is that the market demand curve is not equivalent to a single individual’s demand curve. Individuals have different demand curves, based on their personal choices of what to purchase with any additional money. Combining the different individual curves together, as is done to form the market demand curve, does not give a simple line with one equilibrium point. The line is complex, and may have multiple equilibrium points. Some of these equilibria may be more desirable, from the point of view of social welfare, than others. A market, without appropriate management,  may settle on an equilibrium which does not maximise social welfare.

FacebookMySpaceTwitterDiggDeliciousStumbleuponGoogle BookmarksRedditNewsvineTechnoratiLinkedinMixxRSS FeedPinterest
Pin It