Tradeoffs: What Happens When Economic Assumptions Don’t Reflect the Real World
In a new column, economist and businessman Phillip S. Coles explains why models and theories don’t tell the whole story
By Phillip S. Coles
Welcome to the first installment of a new column dedicated to avoiding the unintended consequences of decision-making. Credit for naming this column goes to my good friend, colleague and marketing authority, Jim Maskulka, Ph.D., from Lehigh University, who immediately suggested “Tradeoffs.” In this column, I intend to investigate economic decision-making from a broad perspective that goes beyond economics as a discipline and speaks to the pragmatic compromises necessary to achieve successful business and policy outcomes. My experience lends itself to this broad perspective: Before my time in academia, I worked internationally in the mushroom and produce industries, where I had to deal with such tradeoffs routinely.
In economics we talk about “ceteris paribus”—“all else being equal.” When running an experiment, we want all “other” factors, the ones in which we have no interest, to be constant so we can measure the relationship between those factors in which we do have an interest. In other words, we want to know how a change in one thing—like tax rates, exchange rates or the stock market—is related to another thing—like possibly tax rates, exchange rates or the stock market—excluding everything else.
Unfortunately, in business, economics or any social science, it is rarely possible to perform controlled experiments that actually allow us to hold all else equal. As an example of a controlled experiment, we might use different fertilizer rates on randomized plots and compare them to a control plot with no fertilizer, to see which rate provides the optimal yield. In economics we need a different approach. For instance, to investigate what might be the best tax policy, we can’t randomly assign Americans different tax rates and establish who fares best. We can only observe things as they occur and attempt to conclude what transpired through analysis. This makes it difficult to tease out what factors are causing particular changes and what factors are just background noise. Statistical tools can “adjust” for other factors, but they are not as reliable as controlled experiments.
In addition, fundamental to the causality problem is the slew of fields that economists know little about but still have economic impacts. Things such as weather, health, education, technology and a multitude of human factors, if not accounted for in an economic model, can render understanding extremely difficult. Despite this, there is a tendency to consider what we hear from experts as gospel. But what about the impact of all those fields in which the expert is not the expert? Are they being considered? Drilling down into one area risks falling victim to the saying, “When the only tool you have is a hammer, everything looks like a nail.” Other disciplines and the tradeoffs among them can easily be discounted because everything looks like a nail, but this narrow perspective can have unintended consequences.
Economist Thomas Sowell pointed out that often we think experts can instruct us in how to create a utopian society, but in fact there are no solutions, only tradeoffs. The tradeoffs are an attempt to avoid the unintended consequences that arise from the things we may not consider because of our particular expertise or worldview. We may see the world through only one lens, the one with which we are the most comfortable. To avoid this, contributions from other “experts” are required. Then generalists must take experts’ input, weigh the tradeoffs and attempt to optimize results. Finally, there must be something to compare the results to—the control. If we compare possible outcomes with the perfect scenario, they will always lose, but if compared with actual alternatives, results can be optimized. Finding that balance is what this column will attempt to address.