The New Normal

Economists say good is a “normal” is if an individual’s rate of consumption of that thing tends to rise with income. Most goods fall into this ‘normal’ category (thus the name): gasoline, brand-name clothing, electronics, microbrew beer, housing.

Obviously, the rate of consumption does not necessarily increase linearly with income, thanks to ‘diminishing marginal utility.’ A family who makes $100,000 per year might live in a house with 3 bedrooms, a 2-car garage and an eat-in kitchen, but the family that makes 10 times as much does not necessarily live in a house with 30 bedrooms, a 20-car garage and 10 kitchens.

This relationship could be represented by a power-law equation like this:

So-called “inferior” goods are the opposite. The definition Investopedia gives of an inferior good is: “A type of good for which demand declines as the level of income or real GDP in the economy increases.

Examples might be things like: public transportation (Investopedia’s example), ramen noodles, particleboard furniture, canned cheese spread, off-brand electronics, second-hand clothing. Generally, anything that people gladly trade away or stop buying the second they make enough money to buy something else.

A few things bother me with this classification of goods into ‘normal‘ and ‘inferior‘ categories:

One is that the same product might be ‘normal’ or ‘inferior’ to people in different economic circumstances. To a recent college graduate who just got a job, a shiny new Volkswagen Beetle might be a very desirable vehicle, while a successful salesman trying to impress potential clients might pass right by it in favor of something with more amenities.

Another problem I find is that the definition of ‘inferior’ goods (those where consumption rises as income falls) implies that a person with a low income should consume more than someone with a medium income and a person with no income should consume the most of all.

Of course, people with no income have no ability to buy anything. Economists define ‘demand’ as the desire to purchase a good in the presence of the ability to pay. I assume the Ability to Purchase a good increases linearly with income and the Desire to Purchase a good probably peaks, for a given product, over some income range. That Volkswagen Beetle, for example, appeals to relatively well-off young people, being too expensive for lower-income individuals and too cheap for higher-income ones. You might think of the appeal of the Beetle as a Gaussian (i.e., bell-shaped) distribution, peaking at a certain income level, but being of low appeal to people with significantly lower or higher incomes.

If we multiply a linearly increasing Ability by a Gaussian (i.e., ‘normal’) distributed Desire, we get a consumption pattern that looks like:

This to me seems a much more realistic representation of consumption for most goods. Economy-class airline tickets, for example: you can’t buy any if you have no income. As income increases, so does your ability to buy. But at some point, a higher-class ticket becomes more desirable, so your consumption of economy-class tickets drops off.

In this view, there really are no ‘normal‘ or ‘inferior‘ goods. If your income is such that you are past the peak, the good is ‘inferior’ to you. If you are to the left of the peak, the good is ‘normal’ to you. Some products might have a peak so close to $0 that they are basically inferior to everyone. And some might have peaks at such a high income that they essentially look like the first, blue chart above.

This view addressed a third problem I have with the traditional classification of goods into ‘normal’ and ‘inferior’ categories: For some products, consumption probably has multiple ‘humps’, one for people with lower income and one for people with higher income. Take firewood, for example. Low-income people buy a lot (for heating) and a second group of higher-income people buy some (for ambience). Different groups buy for different reasons (probably related to Maslow’s Hierarchy of Needs), and this might be represented simply by overlaying by multiple Gaussians on top of each other.


  1. Bradd,

    There was some classic analysis some time around 1991, I think by Mercedes I believe, of the distribution of purchases by income level in China, that fell into a normal distribution as you’re showing (I don’t recall the tail to the right but it makes sense).

    The key point that Mercedes took from this was – look how steep the curve is (you drew one very similar to what they found).

    So if the *average* income rose by only 5-10% (for argument, let’s say the center point of the curve moves to the right 5-10%), this corresponded to the curve widening to the right – and due to the steepness, you could have something like a 100% increase in sales (area under the curve) just by a 5-10% increase in average income.

    So they concluded that China had massive potential for increases in sales as people moved to the cities. I think all of this is related to the two classic adoption curves as well, Gompertz and the other one, whatever it’s called.

    – Ted

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: