Wednesday, June 30, 2010

For my anniversary, some statistics on marriage and divorce

This week marks the third anniversary of my wedding to my beautiful wife Shanna. To honor our commitment to one another, I dug up some data to investigate how unusual our 3-year marriage is. Although my quick search didn't turn up much reliable duration data, I found some interesting tables on marriage rates and divorce rates (# of each event per thousand people) from the Division of Vital Statistics.

To me, the most striking thing about the divorce data is how -- for all but three states -- the divorce rate has declined from 1990 to 2007. What happened in the other three states? In New Jersey, Maine, and Connecticut, the divorce rate was the same in 2007 as it was in 1990.

Does that mean that more marriages today are successful marriages? Not so fast.* A quick scan of the marriage data reveals a similar (maybe even more dramatic) decline. Only two states had higher marriage rates in 2007 than in 1990: West Virginia and Hawaii. West Virginia was virtually the same in the two years (7.2 in 1990, 7.3 in 2007), while Hawaii was much higher. The increase in Hawaii is probably because Hawaiian beach weddings have become more popular.

So, are we an unusual couple? Yes. In an era when marriage is on the decline regardless of the state, we actually tied the knot. As for the three-years-and-counting duration, it still feels like we are newlyweds. Here's hoping that we can follow in my parents' footsteps and get to 40 happy years together and beyond!

*One would have to compute something like "divorces per marriage" and track this over time to get a sense for whether marriages are becoming more or less successful (if "not getting divorced" is a measure of success). The table provides enough information to do this in a crude way. Dividing the divorce rate by the marriage rate yields:

(Divorce Rate)/(Marriage Rate) = (# of divorces/1000s of people)/(# of marriages/1000s of people).

This measure isn't perfect (as a low divorce/marriage ratio today says very little about the quality of today's marriages, and it might just represent changes in the number of marriages over time), and I am sure that other better data sources exist on the state of marriage.

Sunday, June 27, 2010

Follow up: Examples of Different Types of Advertising

Last Wednesday, I wrote about different types of advertising: informative, signaling, and complementary goods. Just for fun, I dug up some examples of the different types of advertising.

Under informative (to the extreme), here's a clip from The Invention of Lying (click through because embedding is disabled on the video). My favorite quote: "It's Coke, very famous. Everyone knows it. I'm Bob. I work for Coke, and I'm asking you to not stop buying Coke."

This next one is a pure signaling advertisement.

The signal is supposed to be, "We are so good at managing money that we can waste two million on a commercial." To me, it seems like they sent a different signal: "We waste money and would like to manage yours." Yet, E-trade seems to have passed the market test. They're still around while the many other businesses that popped up around 2000 are not. Maybe the advertising worked

Next up, here's an obvious example of a commercial that is a complementary good.

Ironically, the Bud Light Tailgate Companion is itself an advertisement for Bud Light, and this commercial is an advertisement for the Tailgate Companion. Bud Light is actually charging a positive price for some of its advertisements (just like Nike does by putting a swoosh on every product).

Here's a commercial that is harder to classify:

It isn't exactly informative, and the signal isn't clear. But, Friskies spent loads of money on that advertising campaign. Do you think it works? If so, what about this commercial increases demand for Friskies? I'd like to know.

Saturday, June 26, 2010

Poll: How do you perceive jury duty?

Last night, I was excited to discover that one of my favorite movies -- 12 Angry Men -- was in the free movies section of On Demand. If you haven't seen the film, it is a great film that provokes thought about the role of jury duty in American society. Even if you are not fond of black and white films, the acting is expressive enough to bring color to your imagination.

I won't spoil the movie, but I'll give the premise. Almost all of the movie takes place in a jury deliberation room. The jury is deliberating the fate of an 18-year-old boy who allegedly killed his father. If convicted, the kid gets the electric chair. The twelve jurors have strong opinions about the case, and the deliberations reveal both good and ugly sides of human nature.

After watching the movie (and observing how the different characters portray attitudes toward jury duty), I thought of this week's poll question.

How do you perceive jury duty?

(a) A waste of time.
(b) An honorable thing to do.
(c) A fun sideshow.
(d) A way to earn a few extra bucks.

There are other types cast in the movie. It really is a wonderful film. If you feel that these other types capture your view on jury duty, feel free to share them in the comments. As with all polls, this one is open for a week. Vote early, often, and on the sidebar. Tell your friends to vote, and if you're called for jury duty this week, tell your fellow jurors to vote as well. But, don't worry. Unlike the verdict of guilty, the answer to this poll need not be unanimous.

Wednesday, June 23, 2010

Why advertise? Notes from an economics class

Today, I am studying for my field exam (which is next Friday). As I was scouring my notes, the first few lines in my notes on advertising struck me as blog-worthy. It turns out that economists have given quite a bit of thought to the question of "Why advertise?"

Here's what is in my notebook:

There are different notions of advertising.

1. Informative (Stigler '61, Butters '77). This type of advertising is socially beneficial, and it more important for search characteristics (observable before consumption like price or color) than for experience characteristics (product has to be consumed first like taste or suitability).

2. Signal (Kihlstrom and Riordan '84). By advertising, can separate from others, but it has to be costly to signal etc.

3. Complementary Goods (Becker and Murphy '93)...

Then, my notes launch into some math that explains how Becker and Murphy approach the problem. Instead of presenting the math, here's what I wrote about Becker and Murphy's argument in January:
In a world where everyone thinks that the pre-movie advertising is a nuisance, why show any advertising after the lights dim? In 1993, Gary Becker and Kevin Murphy wrote a paper that answers this question. From their analysis, it turns out that advertising can be a good or a bad from the perspective of the consumer, but it can work in either case. Take two examples of advertising that works:

1. Beer advertisements on television usually work by giving us a good laugh that causes us to remember the product the next time we show up at the store. In fact, one of the reasons that people watch the Super Bowl is to see the new advertisements. This is an example of advertising as a good. To most people, it isn't surprising that this works.

2. The beggar on the street with the cardboard sign makes you feel terrible (it enters negatively into your utility function to walk past him), so you put money in his box. People will cross the street (or change their walking route) to avoid beggars. They'll rarely cross the street to happen across the beggar to give him money. This is an example of advertising that is a bad, but works nonetheless.
It's time for me to return to my studying. Next topic: Search.

Saturday, June 19, 2010

Poll: Do you remember Jamiroquai?

I saw this video linked on Jeff Ely's Sordid Links (via BoingBoing), and I couldn't resist linking to it.

Apparently, in all this virtual insanity, Jamiroquai is still kicking out albums and commercials (for an update on latest projects, see the Wikipedia article). It is too bad that their commercials resemble the LOST character Charlie's commercials.

I have my Jamiroquai CD somewhere in my pile of stuff. In high school, I loved their songs Virtual Insanity, Alright, and Deeper Underground. Now, I wonder what I was thinking.

That brings me to my poll of the week.

Do you remember Jamiroquai?

(a) Of course!
(b) I'm trying to forget
(c) They seem oddly familiar
(d) No

As with all polls, this one is open for a week and needs your input. Please tell your friends to vote. If there is someone you haven't talked with since the 1990s, this could be a good ice breaker. And, after you break the ice, tell them to vote on this poll. I'm interested in hearing what you have to say.

Wednesday, June 16, 2010

Do Students Reward Low Standards?

The economics blog-o-sphere is abuzz about this new study in the Journal of Political Economy about the content of teaching evaluations (Cowen, Mankiw, Ely, Sandeep Baliga). In this study, the authors claim to have found evidence that "teaching to the test" degrades long-term student performance. Implicitly, the authors claim that this is because:

(1) Students reward receiving high grades with high evaluations.
(2) Professors are induced by this reward to inflate grades by "teaching to the test."
(3) The grade inflation resulting from "teaching to the test" harms long-term student achievement.

The authors base their conclusions on an interesting data set from the U.S. Air Force Academy. In the data set, students are randomly assigned to professors (eliminating bias from good students picking good professors), exams are standardized (allowing for direct comparisons across courses), and students are randomly assigned to follow-up courses (allowing for an assessment of long-term learning).

For establishing whether students scores improved because of the students assignment to professors, this data set is a dream. There is no selection into courses, there is direct comparability across courses, and the academy randomly assigned students to professors, both in the intro courses and follow up courses. In other words, the data set is an experiment. As any introductory statistics course should have taught you (if you had a good professor), experiments allow researchers the ability to assign causation.

In his post on this topic, Jeff Ely points out that assigning causation is tricky, even if we get to assign it:
I am not jumping on the bandwagon. I have read through the paper and while I certainly may have overlooked something (and please correct me if I have) I don’t see any way the authors have ruled out the following equally plausible explanation for the statistical findings. First, students are targeting a GPA. If I am an outstanding teacher and they do unusually well in my class they don’t need to spend as much effort in their next class as those who had lousy teachers, did poorly this time around, and have some catching up to do next time. Second, students recognize when they are being taught by an outstanding teacher and they give him good evaluations.
I think Ely is right. Even though we can determine that the assignment caused scores to increase when evaluations are low, Ely's explanation points out that we cannot sort out what features of the environment implied that the assignment causes scores to increase. Ely's conclusion seems perfectly plausible to me.

Moreover, there is potentially a more troublesome issue with the conclusions of the study. Let's suppose that the authors were right to conclude that "teaching to the test" harms long-term student performance in their sample. Their data apply only to courses taught at the U.S. Air Force Academy, yet they conclude generally. There are a couple of reasons why this is troublesome:

First, there is significant variation across institutions of higher learning with respect to attributes of students (quality, attitude, etc.), the role that student evaluations play, and the form of the evaluations. I have experience with two universities (Montana State University and University of Chicago), and the students I have encountered at the two universities are considerably different. They would likely give high marks on evaluations for very different reasons.

For example, in my first encounter with a UChicago undergraduate, I told him that the course "is going to be a challenge." He responded, "Great! I would expect nothing less." That was just my first encounter, but based on my experience as his teaching assistant, he and most of his classmates probably would have given bad evaluations if we hadn't challenged them. This obviously isn't true at all universities, but some universities have developed a special culture where students long to be challenged.

Given this personal observation, I expect student evaluations to have a different relationship with long-term performance depending on the institution and student population that generates the data. The U.S. Air Force Academy isn't a representative institution of higher learning, so making general conclusions on the basis of a sample generated by the Air Force isn't valid.

Second, the experiment was conducted on a very peculiar type of teaching-learning environment. The introductory course was a common course with multiple sections taught by different professors. The professors taught out of a common syllabus and gave identical exams during a common testing period. This sounds great for the prospect of learning something deep, but for two reasons, this environment is problematic for making inference to the general nature of student evaluations and long-term learning:

(1) The standardization of the course encourages "teaching to the test," even among instructors who would not otherwise teach to the test. In this setting, instructors know that their students' performance is directly comparable across sections. This may induce additional instructors to give hints to students merely so they do well on the common exam. Depending on the type of the instructor who would switch to giving hints, you would expect to find different results if the data were generated from standardized classes than from classes that were not standardized.

(2) More importantly, the standardization of the course strips away important aspects of teaching. Good professors write comprehensive exams that require intense and comprehensive studying. They also set an ambitious agenda (represented in the syllabus) that holds students to a high standard and encourages deep learning. These are are unobservable aspects of teaching in the study's data set because the course has a common syllabus and a common set of exams. Therefore, the only aspect of teaching that the study can purport to describe is lecturing ability.

Because setting the course agenda is the most important instrument instructors have to encourage deep learning, this is a bad data set to use to uncover the truth about how and why student evaluations correlate with deep learning.

In spite of these flaws, it is an interesting data set and an fascinating study. In fact, I completely agree (on intuitive grounds) with what the authors conclude about how universities should view student evaluations:
Since many U.S. colleges and universities use student evaluations as a measurement of teaching quality for academic promotion and tenure decisions, this latter finding draws into question the value and accuracy of this practice.
I agree that it sometimes seems that students reward low standards, but I am not sure the consequences are so dire for long term performance. As I always tell students before they fill out the evaluations, it is the comments that matter most.

Saturday, June 12, 2010

Poll: Why would someone work for free?

Last night, I rode the bus into downtown Chicago. One of the patrons in the back of the bus was blaring offensive rap music (i.e., as nearly as I could tell, the lyrics were 1/4 swear words, 1/2 offensive to women, and 1/4 about violence). Not only was this guy blaring the music, but he was singing every word. Other bus riders were clearly bothered by the display, but everyone was too intimidated to confront the guy. So, he kept on rapping to his soundtrack.

After I got past being disturbed by the scene, I had to wonder about why this guy would bother committing those lyrics to memory. The guy could clearly go on for hours repeating the words to the exact rhythm that the recording artist intended. It probably took some serious work to develop the ability to spout off those lyrics. How much time did it take for him to learn all of the lyrics? More importantly, how was he ever convinced to devote so much time to this task? He obviously has some cognitive skill. Why not apply that skill elsewhere?*

As an economist, I view behavior as a response to a system of rewards. These rewards can be intrinsic (e.g., sense of purpose, fulfilling a lifelong goal, etc.) or extrinsic (e.g., money, "learn this rap song or you can't be a member of our gang," etc.). These motivations can be powerful in determining what someone will do. Which rewards work is a matter of context. For some interesting background on this, see this awesome video (HT: Jodi Beggs):

By the way, I was fascinated by how they synchronized the talk with the white board artist's animation.

Another note: My friend Dana Chandler has been looking into what motivates people in a series of experiments on online labor markets. He's trying to isolate the effect of context on productivity. In a broader context, this is quite relevant for how labor markets work (it's not just about paying the employee their reservation wage). Interesting and important stuff.

That brings me to my poll question of the week.

Why would someone work for free?

(a) He/she already has enough to eat.
(b) Sense of purpose
(c) It's not work. It's leisure.
(d) There's no possible reason/ erratic behavior

As with all of my polls, this poll is open for a week. So, vote early and often. Tell your friends to vote. Watch the video and read Dana's paper. I won't pay you, but think of the higher purpose.

*For all I know, this guy could be an investment banker, but he was loudly singing rap lyrics on the back of a public bus. If he is an investment banker, he went out of his way to pursue this hobby.

Tuesday, June 8, 2010

Price Discrimination Everywhere: Baseball Edition

Jodi Beggs recently responded to an interesting question on baseball economics on her blog. Here's the question posed to her by a guy named Seth:

I had a topic for you to address that I thought might be good for someone with a behavioral econ background. Tonight is Stephen Strasburg’s MLB debut (#1 uber prospect) for the Nationals. The Nationals will only let you buy tickets for tonight if you buy tickets to 3 other games during the season. You do get one of those games free, plus a Nats hat. Why don’t the Nats just raise their prices for tonight’s game? My guess is that people hate price increases, but I thought I would see if you have some research to back that up or other ideas?

It's an interesting question, and it just so happens that I was planning to write on something directly related to this topic this week. Seth, if you're out there, I'm giving you a two-for-one deal: Two complementary economics answers for the price of one. First, here's Jodi's main point (she has a lot of side points, too... really, check them out here):
Luckily, that explanation is simple- that kid is cool, and people really want to see him. Regardless of whether markets are competitive or not, increased demand leads to higher prices. But Seth points out that the Nationals are being a bit sneaky- they aren’t explicitly raising the ticket prices for the popular game, but they are bundling it with other items. I can’t be sure of what their reasoning is, but my hypothesis is something along the lines of the following: like I said, the marginal cost of one more person at the ballpark is essentially zero since there are a fixed number of seats. (The opportunity cost of the bundled tickets is also zero since the tickets would likely go unsold otherwise.) In addition, once a person is at the ballpark there is a chance that he will buy the $8 draft beers, the $6 cheeseburgers or, if you’re in Boston at least, the $4 Dunkin Donuts iced coffee. These two facts together imply that there is an incentive to get butts in the seats, in which case the bundling strategy is clearly superior to just raising prices.
Basically, Jodi points out that the Nationals were blessed with high demand, so they did what our supply and demand models predict; they raised prices. But, they did so in a smart way. They bundled the high value product with a different product they were having difficulty selling. As Jodi points out, this is a smart idea because it achieves cost savings, filling seats that would otherwise go unfilled.

Bundling is smart for another reason that Jodi didn't mention in her post. Bundling can effectively enable price discrimination (or charging different prices for the same good). The obvious good here is a ticket to watch the phenom, but let's focus on the auxiliary tickets to see more clearly how the Nationals are selling at different prices in this example. Effectively, the Nats can exploit the excess demand for "phenom watching" by using a bundle to charge more than they would have gotten for the ordinary tickets.* If the "phenom watchers" attend the three ordinary games, they will likely sit next to someone who paid half the price for a ticket that they bought in the parking lot before the game. (UPDATE: This is backwards for the price discrimination story. All else equal, the high-phenom demand individuals will feel like they paid a low price for the ordinary tickets. See my correction in the comments.)

An additional consequence of bundling the tickets this way is that the bundle acts as a screening device. Some fly-by-night fans might want to go see the phenom's major league debut, just because it is his major league debut (some people like the spectacle). All else being equal, these fans will tend to have less overall demand than those fans who just love their Nationals. In this sense, the four-ticket bundle will be more attractive to "genuine" fans than it will be to spectacle spectators.**

In case you want to know more about price discrimination, I recently recorded three videos on the topic. Formally, the Nationals are engaging in 2nd degree price discrimination (which is -- appropriately -- the second video in my series). Unlike 1st and 3rd degree price discrimination, the formal theory of 2nd degree PD is a bit hard, but the intuition is just what I've said: use bundles to charge different prices for the same good.

Here are the videos:




If you like these videos, I have 45 more videos on microeconomics at my YouTube channel, which follow (so far) the first six chapters of my book. Check it out if you're interested.

*To be fair, Jodi touched on the price discrimination logic in the story. She just didn't emphasize it. With me being a lover of price discrimination, I felt compelled to write a post emphasizing this feature.

** The team could have different reasons for wanting to screen the audience for genuine fans. Maybe they cheer harder or spend more money at the concession stands.

Friday, June 4, 2010

Poll: Why do firms tie in sales?

An interesting topic in industrial organization / antitrust economics is what to do with firms who tie in sales of one product to sales of another. There are plenty of examples of tying in sales. Here are just a few:

1. Getting around regulation. Unfurnished apartments in a rent-controlled market (see my price ceiling article) are not allowed to have rent above some set amount, whereas furnished apartments can charge higher rents. This leads to tying in: An apartment with a dumpy recliner is a furnished apartment.

2. Efficiency Considerations. To tie in a sale, the firm must bundle two or more distinct products into one package to be sold together. If the products are sold together, what does it mean for these distinct products to be products? For concreteness, is a car a product or a bundle? It probably makes sense to think of the car as a bundle of products, but these products are efficiently tied together. After all, who would want to buy a car component-wise?

3. Price Discrimination. Tying in sales allows the firm to effectively charge different prices to different consumers based on willingness to pay. Take an old example, the IBM case, where IBM (then a supplier of tabulating machines) tied the sale of its IBM punch cards to the purchase of their IBM tabulation machines. By so tying the sales and marking up the punch cards, IBM could effectively charge a higher price to intensive users of the machines. These high-intensity users presumably have a higher willingness to pay for tabulation machines. Therefore, charging a higher price to these users makes economic sense.

Another (potential) case of price discrimination through tying is the iMarket. By tying the Apple suite of products to the iTunes and iBooks markets, Apple can effectively charge a higher price to consumers who have the highest demand for all of the iStuff. Thus, even though the iDevice costs the same to an intense iUser as a casual iUser, the intensive consumer pays a higher price (through buying more iStuff that has a supplemental markup built in).

4. Strategic Tying. This is the type of tying that got Microsoft into trouble. Tying sales can be a successful anti-competitive strategy. One reasonable interpretation of Microsoft's antitrust harms is that they used tie-in sales to foreclose small companies (like Netscape) from building a viable competing suite of applications to Microsoft's application software suite. Down the road, such a suite could have been the basis for a fully-competitive operating system environment to Microsoft OS.

Anticipating this, Microsoft tied Internet Explorer and other applications to the sale of its operating systems. To the extent that this foreclosed opportunities for rivals to enter, Microsoft's strategy of strategic tying worked. Indeed, Microsoft remains a large and successful corporation. Yet, at this point, one might wonder if Microsoft's position has already eroded to such forces as web-based applications.

Each of these broad topics is interesting in itself, but my question for this week is:

Which of these reasons for tie in sales is most important?

(a) Circumventing Regulation.
(b) Efficiency
(c) Price Discrimination.
(d) Strategic tying.

The poll is open for a week, so vote early and often. Ask your friends to vote. I would love to -- on a condition of reading this post -- tie you into an obligation to vote on this poll, but I don't have that sort of power. Please vote, anyway. I'm interested in hearing what you have to say.

Wednesday, June 2, 2010

What is the effect of a price cap?

Today's post is about a deceptively complicated topic: price ceilings. In an ordinary supply and demand analysis, supply slopes up and demand slopes down. Their intersection define the equilibrium price and quantity.

This is an equilibrium because -- given the market price -- neither the producers nor the consumers have an incentive to change their behavior. Everyone who is willing to pay more than the price buys the product. Every unit that costs less than the market price is sold. And, there are just as many units demanded as are units supplied.*

Turning to today's topic, what happens when the government places a cap on the price below the equilibrium price? Such a cap is called a price ceiling, and the first-order analysis looks like this:

At the regulated price, quantity demanded exceeds quantity supplied, so there isn't enough to go around. Hence, price ceilings lead to shortages. Given equations for supply and demand, it wouldn't be difficult to compute the size of the shortage. We could even tell the story in reverse for price floors (which stipulate that the price cannot go below some level). In that case, price floors cause surpluses. For many, this is where the analysis stops.

This is unfortunate because if you start thinking about it, the analysis does not make much sense if we stop here. There are still interesting questions that we have not yet answered. What happens to the excess demand? Does it just evaporate? Who among the (too-many) consumers gets the (too-few) products available for sale?

The problem is that the graph no longer depicts an equilibrium. Here's another picture to clarify the point:

At the quantity traded, there is a wedge between what buyers are willing to pay (Pd) and the maximum price allowed by the market (Ps). If you are a consumer who is willing to pay more than the maximum legally-mandated price, do you just stand by and let someone else get the good? No! You are willing to expend up to the surplus you would get at the mandated price to get your hands on one of the rare products that will be traded.

One way to resolve the shortage is to assume that individuals waste some value of their time waiting in line to help "pay" for the good. In this world, everyone who waits in line gets the good eventually, but the difference in the maximum price and the willingness to pay is covered by the costly hassle of standing in line.**

In the standard supply and demand equilibrium, whatever the buyer pays the seller receives in exchange for bringing the product to market, but this isn't so for waiting in line. When the buyer pays for a product in time, no one gets those foregone opportunities back. In this sense, waiting in line is pure waste.

There are other ways to resolve the shortage. For example, suppliers could bundle the good with other goods and services (and charge extra for those services). Under rent control, you often see that landlords charge high application fees and key fees. When the U.S. government instituted price controls on gasoline (in the 1970s), many gas stations offered complementary services that included a tank of gas. Get your oil changed (at this high price) and you also get a tank of gas without the worry of being the one who doesn't get to fill up if there is a shortage.

If you think about any price change, there are a host of other possibilities for buyers and sellers to get around price restrictions. I have only scratched the surface, but it is a surface worth scratching.

*If more consumers show up than suppliers, the price will be bid up. If there are fewer consumers than suppliers, the price will fall. The only stable price is when quantity supplied equals quantity demanded.

** In general, you could take "line" to be figuratively. Consumers need not literally wait in line. They could spend time filling out forms and making phone calls, or they could literally pay someone else to deal with the hassle.