Considering that the definition of econometrics is economic measurement, theoretical econometrics might seem a contradiction in terms. After all, isn't economics a discipline built on theories? Shouldn't econometrics then be only about measuring economic phenomena?
To a large extent, it is. It also helps to predict future economic conditions, given the data at hand. However, there must still be theoretical applications to determine if the methods of calculation and the selection of variables are correct for a given analysis. That's the job of the theoretical econometrist.
Sounds confusing, right?
In this article, your Superprof explains the theoretical side of econometrics.
Theoretical Econometrics Overview
For as long as humans have been assigning value to things - at least as far back as in Ancient Greece, justifying value and its impact on the human experiment has been the subject of great thought and debate. As human conditions evolved - from agrarian to industrial, and essentially classless societies became heavily ranked, those discussions expanded to include ideas like ownership, wealth and labour.
For centuries, all of those economic ideas were just that: ideas. Theories that were impossible to prove, for all that they were, indeed, based on observable - empirical evidence.
All of that changed about 100 years ago, when the sub-discipline of economics, econometrics, was established.
For the first time in history, economists had a way to prove their theories and even project future economic conditions based on specific variables. Unfortunately, econometrists soon ran into trouble.
Have you ever been invited to ask a question, perhaps by a discussion panel or maybe even upon meeting your favourite music or film star?
It's a rather common phenomenon when finally given license to ask any one question or make any one statement you want. Suddenly, you're plagued with uncertainty over which question or statement will be the right one and the most impactful; the one that will yield the desired response.
That, in essence, is what theoretical econometrics is all about.
Out of the abundance of data to examine and the economic questions that could be answered, theoretical econometrists challenge each econometric procedure to determine if it will lead economists to the right conclusions.
Once that all-important question has been answered, applied econometrists can get to work on estimating coefficients and applying them to linear regression models.
How Theoretical Econometrics Works
To keep this article from getting too heavy, convoluted and difficult to understand, let's use a real-life, commonplace happening to show the need for and purpose of theoretical econometrics. Shopping will serve us well.
You want to go shopping. What you're shopping for and why you're shopping will decide how and where you will shop, and how much you'll spend.
If you're shopping for common, everyday items, you might shop online or head to a store you know carries those items. You may or may not comparison shop to get the best value for your money. You may buy more than you need depending on the bargains and/or promotions you might find or, conversely. decide that now is not the right time to buy.
What, when, how and how much you buy depends on how great your need for that item is and how much time you have to make a purchase. If you're shopping on your meal break at work, you might buy faster than if you could linger over various web offers and/or displays.
Each of these aspects changes the process of shopping, just as a desired outcome would change how economic data is measured. Are you starting to get an idea of how important theoretical econometrics is?
What if you have no impending need for anything in particular; you just want to shop for the fun of it, and maybe you'll buy something in the process? That too changes the equation significantly.
You might bypass online shopping to wander through countless stores, trying on various articles of clothing and shoes and, in the end, buy nothing at all - but still end up with a feeling of satisfaction that retail therapy can bring. You might sample perfumes or try on new makeup, go to a sports equipment shop to see all the new seasonal gear or head to a car dealer to test-drive the latest models, all with no intention of buying anything.
Obviously, if you need retail therapy, utilitarian shopping will not satisfy and, if you need to dream about nice things you could one day own, scrolling online or dashing into a store and emerging a few minutes later, bag in hand, will leave you unfulfilled.
Econometrists have the same conundrum.
They cannot use the same variables, equations and formulae to answer every question or predict every economic phenomenon. They must constantly test their statistical procedures and develop new ones to satisfy the needs of the moment, regardless of the fact that economic data constantly changes.
That's what econometrics is all about, really. Its importance lies in its capacity to make quantitative statements out of qualitative ideas so that policymakers - those who make decisions based on economic phenomena are as informed as possible.
Theoretical econometrists make that happen by constantly testing and revising econometric procedures and techniques.
The Theoretical Econometrist's Toolkit
Statistics, economics and econometrics are all maths-heavy subjects but theoretical econometrists use a variety of tools and methods to test new modelling procedures.
Their primary function is to test whether the already existing procedures and statistical tests are sufficient to bring and/or predict any possible economic unknowns.
To put that into layman's terms, it's about the same as checking your car over before you head out on a long road trip. You'll want to be sure that your engine is in good enough condition to get you there and back, that your tires have enough air and tread, and that your fluids are topped off (and not degraded).
In other words, you want to minimise the risk of unknown calamities by making sure your vehicle is sound and up to the task.
Granted, checking over your car is a hands-on affair and all of the points that theoretical economists verify are on paper (or, as is more likely, digital) but the principle is the same.
If a particular calculation will leave the answer short of its target - maybe it doesn't provide a complete picture of the economic phenomenon under examination, they will alter the equation or formulate a new one.
Theoretical econometrists might also change a model or technique because of a trend their counterparts, the applied econometrists, have reported.
Theoretical econometrists don't work with data so they don't get to see how trends emerge or how aspects of some data no longer fit within an established technique for estimation.
When notified of such an event, 'theoreticals' deploy all of the tools at their disposal to reshape the technique, making it once again effective.
The Challenges of Theoretical Econometrics
This discipline's origins lie in two scientifically-minded individuals who, working independently of one another, arrived at the same conclusion. Both established that economic data must be held to the same rigorous tests and examinations that any other scientific discipline pursues to understand its subject matter.
It should come as no surprise that Messrs Frisch and Tinbergen were passionate about maths and physics. Through their training in those studies, they learned how to follow the meticulous and exacting steps scientific research demands; for them, it was no long stretch to apply those methods to economics.
There is a major difference between the rigidly controlled conditions of a laboratory experiment and the ever-changing variables that drive economic phenomena on the ground, though. That difference makes the science of econometric theory particularly daunting.
So, how do they do it? They rely on data from a complex network of related equations, wherein all of the variables might change simultaneously.
Let's say, for instance, that the desired answer is the economic impact of pandemic-related shutdowns. Variables under consideration include:
- population groups by age, gender, race and geographic location
- industry: hospitality (hotels, resorts and restaurants), travel, manufacturing, entertainment and so on
- religion: charitable donations, memberships in religious organisations, church donations
- politics and civic involvement: favoured versus disfavoured parties in power, levels of social unrest and compliance
- national economy: urban and rural, across a conurbation or in major cities
- global economy: industrialised versus non-industrialised countries, wealthier versus poorer nations
- infrastructure: everything from vaccine distribution and administration to manufacturing and food supply chains
As you can see, each of those aspects of modern economic life has several variables that may, at one time or another, take precedence.
It's the job of the theoretical econometrist to determine whether the existing models and testing procedures have the right properties to fully estimate any unknown factors related to the economic question at hand.
In these pandemic times, they had their work cut out for them...
They also strive to meet the three goals of econometrics by devising new statistical procedures sturdy enough to withstand economic data's constant fluctuations.
That's where the heavy use of math comes in, along with numerical quantities and theoretical statistics.
Once theoretical econometrists have proven their models to be sufficiently durable, applied econometrists can get to work on finding the answers policymakers need to make informed decisions.
Does all of that sound too confusing? Maybe you need a broader introduction to econometrics...