Econometricians wear many hats, according to Peter Kennedy, author of A Guide to Econometrics. As economists they use economic theory to analyze problems; as mathematicians they formulate economic theory appropriate for statistical testing; as accountants they collect economic data; as applied statisticians they estimate economic relationships and predict economic events; and as theoretical statisticians they develop statistical techniques for solving economic problems.
With Nathan Associates since 2000, Dr. Tamara is praised by colleagues and clients alike for the soundness of his applied economic analysis and the empirical rigor of his econometric analysis. We recently talked to Dr. Tamara about his work at Nathan.
As an economist at Nathan I apply economic theory to build models that provide solutions to questions that our clients want answered. A model is simply a formal expression of economic theory tailor-made to analyze the unique characteristics of a particular problem. Once a model is constructed, I use data and econometric techniques to determine how well the model performs and how closely it relates to reality. I use econometric models to explain the behavior of consumers and of firms.
A project we did for Greyhound Lines Inc. is a perfect example. Greyhound, a provider of inter-city bus transportation in the United States, was facing a proposal in Congress that would have required the company to install handicap-access doors on all its buses. The company wanted to know if it could pass the installation costs on to its customers without hurting its bottom line.
Greyhound’s ability to increase fares to cover the cost of installing handicap access doors depends on how its customers will react to a price increase. According to the economic theory of demand, consumers tend to buy less of a product or service when the price they have to pay for it increases, when nothing else has changed. This theory also helps predict whether revenues will increase or decrease when prices increase.
Using the theory of demand as the framework, I created a model that allowed me to analyze the demand for Greyhound’s inter-city service, and to show how sensitive Greyhound’s passengers were to changes in ticket prices. We concluded that Greyhound could not increase ticket prices to pass on the cost of the handicap access doors without losing riders and revenue.
To model demand for Greyhound’s inter-city bus service, I identified factors that influence demand using the theory as the guide. At this point I personalized the problem. I imagined myself to be a potential Greyhound passenger. I asked myself: What would influence my decision to travel by Greyhound? What alternative modes of transportation would I consider as a substitute for traveling by bus? What time of the year would I likely be traveling?
What does my profile look like? For example, I probably don’t own a car or I am unable to drive a car because I can’t afford one, because I’m elderly, or because I’m too young to drive. I probably want to go to a place that does not have an airport or a train station. I probably have more time on my hands than a business traveler.
If I’m a business traveler, I might travel by bus if the time it takes to travel by bus is the same as travel by train or airplane. I am more likely to travel on weekends, holidays, or during summer months when the weather is conducive to travel by road.
This technique of personalizing helped me identify factors that affect a passenger’s decision to travel by bus. Remember, the closer a model is to reality the better it explains phenomena and forecasts the future.
Guided by economic theory, modeling is simply an exercise in personalizing a problem to identify what influences the decisions of individuals or firms. When we try to measure phenomena that cannot be quantified but that are critical to constructing the model we must exercise imagination and creativity. So we develop techniques that mimic reality. Personalization is one of them.
Yes. The study for the Irving Zuckerman Trust required a totally different approach. We were asked to analyze whether legislation to limit the amount of debt that a nonfinancial corporation can hold in its capital structure was feasible. The intellectual challenge arose from the absence of any formula to guide us. To analyze this issue we could not take the same approach that we would, say, in evaluating the economic feasibility of an investment project. Instead, we had to answer the question of feasibility by first setting up and testing multiple hypotheses. The debt feasibility study is a perfect example of how economic consultants come up with a creative solution when there is no formulaic approach.
Almost all of the litigation projects that I have worked on have been interesting in terms of product markets, complexity of the issues involved, the econometric modeling, and the opposition—nothing is more refreshing than a good fight. I mean an intellectual battle—I am a strong believer in nonviolence. But confidentiality issues prevent me from discussing these projects.
I started out in accounting and soon discovered it was much too confining. At Clemson University I entered the graduate program in economics, which is very applied. All the graduate students were encouraged to work with data. Our term papers were data intensive and so were our theses.
My dissertation advisor, Mike Maloney, Professor of Economics at Clemson, is an excellent applied economist. I would have to say he influenced me the most. I worked with him on projects and developed useful skills, such as SAS (a statistical software). I think the environment and training at Clemson complemented my own inclination to always verify theory with evidence. I think that was the key.
Krithi, my sixth-month old daughter!