1.4 The Economist’s Tool Kit
Learning Objectives
Explain how economists test hypotheses, develop economic theories, and use models in their analyses.
Explain how the all-other-things unchanged (ceteris paribus) problem and the fallacy of false cause affect the testing of economic hypotheses, and how economists try to overcome these problems.
Distinguish between normative and positive statements.
Now that we have a sense of what economists do, let's talk about how they do it. Economists are in the business of creating theories about how the world works, and then testing those theories to see if they do a good job of describing what we actually see. We already know that economics differs from other social sciences because of its emphasis on opportunity cost, the assumption of maximization in terms of one’s own self-interest, and the analysis of choices at the margin in explaining decisions. But despite those differences, much of the basic methodology of economics—theorizing and then testing—is common to every science. This section explores how economists test and refine their knowledge of the choices people make.
Thank the plague! Pulled home from college in Cambridge to shelter in place as the bubonic plague ravaged England, Isaac Newton made fundamental discoveries in physics, astronomy, and calculus. But nobody can resist the world of economics: Newton finished his career at the royal treasury; his life’s biggest challenge was tracking down a serial counterfeiter! Here, this little guy sits under the very apple tree that inspired Newton’s theory of gravitational attraction.
Source: Photo courtesy of Alan Grant
Researchers in the sciences are often interested in the relationships between variables. A variableSomething whose value can change. is something whose value can change. So, for example, a nutritionist might be interested in the relationship between zinc intake (one variable) and the length of the common cold (a second variable). By contrast, a constantSomething whose value does not change. is something whose value does not change, such as the number of minutes in an hour. While that makes constants sound less important in science, nothing could be further from the truth—just ask Isaac Newton, whose constant of gravitational attraction appears in hundreds of thousands of research articles in physics and engineering.
Scientific research is generally conducted within a framework called the scientific methodA systematic set of procedures through which knowledge is created., a systematic set of procedures that we use to advance our knowledge of the world. In the scientific method, hypotheses about how the world works are suggested and then tested. A hypothesisAn assertion of a relationship between two or more variables that could be proven to be false. is a claim about the relationship between two or more variables that could potentially be proven false. A statement is not a hypothesis if no conceivable test could show it to be false. For example, the statement “Plants like sunshine” is not a hypothesis; there is no way to test whether plants like sunshine or not, so it’s impossible to prove the statement false. The statement “Increasing exposure to sunlight causes plants to grow more quickly” is a hypothesis because it’s possible to test the relationship between sunlight and plant growth with some kind of experiment, and it’s possible the results of your experiment might show your claim to be false.
Oops . . . too much sunlight. Theory = falsified!
Source: © Shutterstock, Inc.
“So, I’ve developed my theory and conducted a test, and darned if the test didn’t prove my theory wrong! What now?!” Well, there’s good news and bad news. Here’s the bad: If a test reveals that your hypothesis is false, then your theory is either flat-out wrong or needs work. Here’s the good: You now have new knowledge—data—that you can use to help improve your theory. For example, let’s suppose you wanted to test the hypothesis about sunlight and plant growth. You might carefully pot some plants and expose each plant to different amounts of sunlight. If you found that your theory was true over some range, but that too much sunlight eventually caused your plants to wither, that information could be used to create a different (and possibly more nuanced) hypothesis about the relationship between those two variables.
A hypothesis that has not been rejected after widespread testing and that wins general acceptance is commonly called a theoryA hypothesis that has not been rejected after widespread testing and that wins general acceptance., like the big bang theory. A theory that has been subjected to even more testing and that has won virtually universal acceptance becomes a lawA theory that has been subjected to even more testing and that has won virtually universal acceptance., like Newton’s laws of motion . . . or economists’ law of demand, which you’ll explore in Chapter 3 “Demand and Supply”.
No matter how strong a theory or a law might appear, there is always a possibility that someone, someday, might find a case that invalidates the hypothesis. That possibility means that nothing in economics, or in any other social science, or in any science at all, can ever be proven true. We can have great confidence in a particular proposition, but it is always a mistake to say that it’s been “proven.”
Models in Economics
The real world is a messy place, with lots of things all happening at once. In fact it’s far too complex for the human mind—or the most powerful computer—to account for all of the possibilities, contingencies, combinations, and permutations that might affect our understanding of the relationships between variables. So, to simplify the world to a manageable scale, scientists use models. A modelA set of simplifying assumptions about some aspect of the real world. is a set of simplifying assumptions about some aspect of the real world.
Which looks more like a real airplane? Which flies more like a real airplane? Sometimes our models don’t have to perfectly mimic the real world to be useful in explaining complex ideas!
Source: © Shutterstock, Inc.
Have you ever used a road map (or the Maps app on your phone)? That map is a model—it doesn’t show every house, every stream, every fire hydrant, and every driveway, even though knowing where those things are might help you navigate from A to B. Instead, it filters those things out, simplifies the world, and only shows the big picture—roads, rivers, major landmarks. That keeps the size of the map manageable, and makes it more useful to people trying to navigate to a destination. Like the map, the economic models we’ll use are always simplifications of the real world. But in generalizing about the world, we gain the power to explain how the world generally works, without having to account for every possible contingency.
You will encounter your first economic model in Chapter 2 “Confronting Scarcity: Choices in Production”. That model assumes, for example, that an economy can produce only two goods. (Economists often use graphs to represent economic models, and the model in Chapter 2 “Confronting Scarcity: Choices in Production” is no exception. Appendix A “Graphs in Economics” provides a quick refresher course on graphs if you need one.) What economists have discovered is that even though the simplest real-world economies can produce hundreds or thousands of goods, trying to incorporate those extra products into our model makes the mathematics really hard, but doesn’t shed any additional insight into the choices people make. And remember that understanding people’s choices is what economic science is all about!
Testing Hypotheses in Economics
Models in economics help us generate hypotheses about the real world. In this section, we will examine some of the problems we encounter in testing those hypotheses.
Here is a hypothesis suggested by the model of demand and supply, which you’ll see in Chapter 3 “Demand and Supply”: A decrease in the price of gasoline will lead consumers to buy more gasoline. How could we test this hypothesis? One way would be to go out into the real world and gather data about people’s actual behavior.
Let’s look for some convenient data: a time when the price of gasoline was dropping sharply. For example, between May 2014 and January 2015, the average retail price of gasoline in the United States plummeted from $3.68 per gallon to $2.06. Over that same few months, the number of gallons of gasoline consumed by U.S. motorists rose 16.7%. So, real-world data from that time period seems to be consistent with our theory.
Just for fun, let’s gather some more data from a more recent time period and see if the theory still holds. Between February and April 2020, the retail price of gasoline fell from $2.45 per gallon to $1.77. Over the same time period, gasoline consumption fell by 52%! Can you say, “Oops!” This data is inconsistent with our hypothesis that lower gas prices lead to increased gas consumption.
Does that mean that we should dismiss our original hypothesis? Not so fast! There are lots of pitfalls in interpreting any set of economic data. One potential problem is that at any time, several things other than just gas prices may be changing—and some of those other things might be affecting consumers’ choices about how much gas to buy. Pitfalls like these can sometimes make economic analysis difficult, but being able to recognize such pitfalls and explain the real-world data is what makes economics so useful. The next two sections examine these potential pitfalls in detail.
The All-Other-Things-Unchanged Problem
The hypothesis that a decrease in the price of gasoline produces an increase in the quantity of gas desired by consumers carries with it the assumption that there are no other changes that might also affect consumers’ desire for gasoline. A better statement of the hypothesis might be: “A decrease in the price of gasoline will increase the quantity consumers want to buy, ceteris paribus.” Ceteris paribusA Latin phrase that means, "all other things unchanged." is a Latin phrase that means “all other things unchanged, or held constant.”
While our data from early 2020 seems to invalidate our theory, it’s worth noting that we didn’t test our theory while holding all other things constant. In fact, lots of very important things changed between February and April 2020. During that time, you may recall, businesses were being shuttered and people ordered to stay home to prevent the spread of the coronavirus. So, even if rapidly decreasing gasoline prices made consumers want to drive more, the order to shelter in place forced them to drive less! The ceteris paribus assumption we made when formulating our hypothesis wasn’t cooperating in the real world.
L.A. traffic is notoriously bumper-to-bumper . . . but during the coronavirus lockdown even cheap gas couldn’t get drivers back on the road.
Source: Hyperlapse Media/Shutterstock.com
In laboratory sciences such as chemistry and biology, it is relatively easy to conduct experiments in which only selected things change and all other factors are held constant. But the economists’ laboratory is the real world, and as we mentioned before, it’s a messy place. It’s hard to test theories in such an environment—after all, we can’t ask the world to stand still while we collect our data. So most of the time, economists end up testing their theories with messy real-world data, and then use special statistical methods to help them isolate the impact of a change in one variable (like the price of gasoline) on another (like gas consumption), even while other factors are waving their hands in the background.
The Fallacy of False Cause
Here’s another pitfall economists often encounter when testing theory with real-world data: Just because two things happen at the same time doesn’t mean that one causes the other.
Here’s an example: Do you remember the discussion in Chapter 1, Section 3 “The Field of Economics” about how well economics majors did on the Law School Admission Test (LSAT)? Does the strong performance by economics majors mean that training in economics sharpens analytical skills, enabling ordinary students to become LSAT superperformers? Or, might it be possible that the type of person who is predisposed to do well on the LSAT possesses the keen analytical skills that draw her to study, and enable her to succeed in, economics? Truthfully, both are probably at work: Economics tends to attract students with good analytical skills—and studying economics helps to develop those skills.
Here’s why this situation poses such a problem for economic science: Hypotheses in economics typically specify a relationship in which a change in one variable causes another to change. We call the variable that causes (or induces) the change the independent variableA variable that causes or induces a change in another variable., and the variable that responds to the change the dependent variableA variable that responds to change..
Sometimes, though, when two variables (like major and LSAT score) display some relationship (or, are correlated), the fact that the two variables move together might falsely suggest that one of the variables causes changes in the other variable. Consider the following (ridiculous!) hypothesis: People wearing shorts cause warm weather. Even though we know our apparel choices can’t possibly influence Mother Nature, the data we see in the real world certainly seems to support our hypothesis. This is a case where our (misguided) theory probably gets it backward: Rather than shorts causing warm weather, warm weather likely causes shorts.
Reaching the incorrect conclusion that one event causes another because the two events tend to occur together is called the fallacy of false causeThe incorrect assumption that one event causes another because the two events tend to occur together.. You may have heard the fallacy of false cause expressed in a different way: Correlation doesn’t imply causation. Because of the danger of the fallacy of false cause, economists test their theories using special statistical techniques that are designed to determine whether changes in one thing actually do cause changes observed in another. Those tests, however, are not perfect, and don’t always offer convincing evidence that one thing does, in fact, cause changes in another.
False Cause: Has there ever been a Nicolas Cage film that didn’t cause despair? The correlation presented here, from Tyler Vigen’s magnificent Spurious Correlations website, is simply a coincidence—an artifact of the time period Vigen chose. Adding earlier or later data would have destroyed the (coincidental) pattern!
Source: Tyler Vigen. “Number of people who drowned by falling into a pool correlates with Films Nicolas Cage appeared in (1999-2009).” Retrieved from: https://tylervigen.com/spurious-correlations. Reproduced via Creative Commons Attribution 4.0 International (CC BY 4.0): https://creativecommons.org/licenses/by/4.0/.
Pop! Goes the Econ: Friends and the Fallacy of False Cause
In this clip from Friends, Joey’s fridge gives out just as Rachel moves in. View the video, and then try your hand at related Concept Problem 27 at the end of this chapter.
Transcript| 6.8100000000000005 to 8.43 | - Oh, that thing's clearly in the way. |
| 10.92 to 15.39 | Alright. Aha. Stop, rich. |
| 17.185 to 21.36 | Hey Joey. How you doing? Great, Rumi. |
| 22.83 to 24.96 | Yeah, I guess we are roommates now. |
| 24.96 to 28.65 | Yeah, well now that you bring it up, our fridge is broken. |
| 28.65 to 30.75 | We have to get a new one and I checked around |
| 30.75 to 32.46 | and your half is $400. |
| 32.46 to 36.9 | Thanks a lot. I'm not paying for half of that. |
| 36.9 to 39.39 | I'm only staying here until my apartment gets fixed. |
| 39.39 to 41.61 | Look rich. My parents bought this fridge |
| 41.61 to 43.11 | just after I was born. |
| 43.11 to 46.65 | Okay, now I have never had a problem with it. |
| 46.65 to 51.27 | Then you show up and it breaks. What does that tell you? |
| 52.92 to 55.74 | That refrigerators don't live as long as people. |
| 58.29 to 61.17 | Right now you know that the ATM only lets you take out 300 |
| 61.17 to 63.39 | at a time, so I'll take a check for the other a hundred. |
| 64.83 to 69.03 | You're joking, right? Of course. I'm joking. |
| 69.03 to 69.84 | I don't take checks. |
Normative and Positive Statements
Two kinds of claims can be tested. One is the hypothesis, which theorizes about the relationships between events, and which we’ve already discussed at length. The second is an assertion of fact, such as “It’s raining outside” or “Microsoft is the largest producer of operating systems for personal computers in the world.” Like hypotheses, such assertions can be shown to be false. But unlike hypotheses, they can also be shown to be true! Economists call any testable statement—whether an assertion of fact or a hypothesis—a positive statementA statement of fact or a hypothesis..
Although people often disagree about positive statements, those disagreements can ultimately be resolved by testing. There is another category of assertions, however, where testing is useless in resolving differences. A normative statementA statement that makes a value judgment. is one that makes some kind of value judgment. Here are some examples of normative statements:
There is too much income inequality.
People should save more for their retirement.
Tiger King is the greatest TV series of all time.
These statements aren’t testable; they can’t be proven right or wrong. They’re judgments, or opinions, that depend on the values of the person who asserts them.
Because people have different values, normative statements often provoke disagreement. An economist whose values lead them to conclude that we should provide more help for poor people will disagree with one whose values lead to a conclusion that we should not. Because no test exists for these values, these two economists will continue to disagree, unless one persuades the other to change values.
The good news for you as a student of economics is that your values are your own, and we’re not going to try to change them: This book focuses on positive, rather than normative, economics. We won’t indoctrinate you, but we will help inform you. And as you gain a more sophisticated understanding of how the world works and become acquainted with what real-world data shows to be true, you may see your values changing, or becoming more nuanced. By the end of the term, you may well feel differently about policies like the minimum wage, universal basic income, or taxes on imported goods.
Key Takeaways
Economists try to employ the scientific method in their research.
Scientists cannot prove a hypothesis to be true; they can only fail to prove it false.
Economists, like other social scientists and scientists, use models to assist them in their analyses.
Two problems inherent in tests of hypotheses in economics are the all-other-things-unchanged (ceteris paribus) problem and the fallacy of false cause.
Positive statements are factual and can be tested. Normative statements are value judgments that cannot be tested. Many of the disagreements among economists stem from differences in values.
Try It!
Consider this hypothesis, based on information in Chapter 1, Section 3 “The Field of Economics”: “Majoring in economics will result in a higher LSAT score.” Data from the American Bar Association show that economics majors do, indeed, score higher than students from other majors. Is that evidence consistent with this hypothesis? Does the evidence prove that this hypothesis is correct? What fallacy might be involved in accepting the hypothesis?
Refer to related Concept Problem 18 at the end of this chapter.
Case in Point: Does Baldness Cause Heart Disease?
A patient walks into his doctor’s office and asks the following: “I seem to be going bald. I’ve read that this means I’m more likely to have a heart attack. If I take a drug to prevent hair loss, will it reduce that risk?”
How might the doctor reply? Do you suppose the correlation between baldness and heart disease indicates that a lack of hair causes your heart to fail (those winters can be cold!)? Or is it more likely that both baldness and heart disease are affected by underlying factors, such as hypertension, high cholesterol, and smoking?
Source: © Shutterstock, Inc.
Science seems to point to the second explanation as the most likely. The good news for balding men (particularly for those whose baldness begins at the top of their head) is that their baldness serves as an early-warning system that might alert them to cardiovascular problems down the road.
Refer to related Concept Problem 15 at the end of this chapter.
Answer to Try It! Problem
The data are consistent with the hypothesis, but it is never possible to prove that a hypothesis is correct. Accepting the hypothesis could involve the fallacy of false cause; students who major in economics may already have the analytical skills needed to do well on the exam.