Monday, June 14, 2010

All Other Things Being Equal

From my earliest days as an economic major, almost at the same time as I was studying supply and demand, I learned the phrase ceteris paribus which translates into “all other things remaining the same” (or remaining equal).  Almost every concept in economics was learned by manipulating one variable so that you could measure the impact of that variable while other variables were kept constant.  Going back to supply and demand, you would gauge the demand for a product (be it a car or a coat or a croissant) by keeping the price and the preferences for all other products exactly the same.  In other words, what happens to the demand for a croissant (my preference would be for a chocolate croissant) if the price of a brownie, a chocolate chip cookie, and a smore stayed exactly the same.  Certainly a food for thought example of how economics works.  In most cases, the rule is simply as price declines, demand increases.  There are important distinctions even in a concept as basic as demand.  If you are describing the demand for a life-saving drug, the demand would remain exactly the same within a wide range of prices.  You need a drug to save your life; you need a certain dose to achieve that result; and likely you will pay whatever needs to be paid (within reason) to get that drug.  Or the opposite example, you want “take out” pizza for dinner and in most neighborhoods, there are many sources of pizza.  Assuming you feel that most of the pizza is equally good, a small increase in price (above the norm) (all other things being equal) could result in a huge decrease in demand for this particular pizza.

But what does this have to do with higher education besides the fact that pizza is always popular on a college campus.  Economic modeling is all the rage on many, many college and university campuses.  Typically this modeling is in the financial aid area and it is a leveraging model.  Given a student with a particular profile (SAT score and high school GPA), how much financial aid will it take for that student to register at a particular college or university.  More complex models adjust for whether the student is local or lives at a distance, or is attending a highly rated high school, or is interested in a particular major.  These models make use of regression analysis to predict the future based on patterns of the past and as an economist, I appreciate the information provided.  And the presentation of these models is often dazzling.  Instantly, as you change an assumption regarding the financial aid offer for a particular cohort, the model will change its prediction.  Not surprisingly more money yields more students (and much of the time more net tuition income) and less money yields less students (and often less net tuition income).

All of this is very impressive and can at times be amazingly accurate.  However, building a future scenario based on past performance while in a serious economic recession may require more than any such model can accurately provide.  If the foundation of the model is replicating to some extent the past, the validity of the model may not be there if the future varies significantly from the past.  We should all keep in mind that ceteris paribus foundation is most likely to fail just when we most need it to succeed.

No comments:

Post a Comment