I am not sure how this happened, but it looks like the second quiz is giving 100 points maximum, when it should be 50 points maximum. Given our discussion in class about how students react to changing the means on exams, this is very unfortunate. However, the Syllabus clearly says that the first and second quiz are each worth 5% of the grade. So out of 1000 points they each should give 50 points.
After the quiz is over I will adjust the scores so that you get the intended amount of points based on the questions you have answered correctly. I'm sorry for this error. Unfortunately, it is the type of bonehead play I am prone to make.
Professor Arvan
Thursday, November 30, 2017
A tournament for students in the class (for fun)
For next Tuesday's quiz to start class, I would like to offer a student generated quiz (joke). The winner, chosen by me, will get 5 Bonus points for the course and the privilege of having their question showcased for the class. Please make your submission by email.
Time permitting, the other submissions will also be shown and we might have a mock election of which one the class prefers. (I'm quite willing to believe the class will select a different submission as the most preferred. Your tastes and mine likely don't agree on much.)
Time permitting, the other submissions will also be shown and we might have a mock election of which one the class prefers. (I'm quite willing to believe the class will select a different submission as the most preferred. Your tastes and mine likely don't agree on much.)
Excel Homework Due 12/6 at 11 PM
This is the last Excel Homework and it is on the Shapiro-Stiglitz model.
This homework is on extensions of the basic Shapiro and Stiglitz model, with a particular focus on making the monitoring intensity a choice variable.
You should watch the video presentation of the basic model first. This is a full lecture on the math of the model. It takes about a half hour. (If you want the PowerPoint file, a link to it is in the description of the video.)
The actual Shapiro and Stiglitz paper.
The video that derives the equations in the paper.
This homework is on extensions of the basic Shapiro and Stiglitz model, with a particular focus on making the monitoring intensity a choice variable.
You should watch the video presentation of the basic model first. This is a full lecture on the math of the model. It takes about a half hour. (If you want the PowerPoint file, a link to it is in the description of the video.)
The actual Shapiro and Stiglitz paper.
The video that derives the equations in the paper.
Today's Quiz
There was the Bigger family with members Mr. Bigger, Mrs. Bigger, and baby Bigger.
Q: Who was bigger?
A: Baby Bigger, because he was a little Bigger.
This book is the source and was used for variety of other jokes over my childhood that I imposed (with some frequency) on my friends.
And this is the inscription from my dad.
Q: Who was bigger?
A: Baby Bigger, because he was a little Bigger.
This book is the source and was used for variety of other jokes over my childhood that I imposed (with some frequency) on my friends.
And this is the inscription from my dad.
Finally, this is the page with all the jokes about the Bigger family.
Wednesday, November 29, 2017
This shouldn't affect us, because we don't have class on Wednesday, but.....
From: Martin-Roy, Tera Nichole
Sent: Wednesday, November 29, 2017 4:50 PM
Subject: Primary Heat Outage - DKH West Side Only - 12/6/17
Sent: Wednesday, November 29, 2017 4:50 PM
Subject: Primary Heat Outage - DKH West Side Only - 12/6/17
Good Evening,
On Wednesday, December 6th from 8am-4pm the Primary Heating Water will be turned off and drained from the west side of DKH only. They will repair a
heating issue, refill the system, and remove the air. The high for Wednesday, December 6th is 34°F
(which could change by then, for better or worse). If you need to spend
significant time on the west side of the building that day, plan
accordingly.
If
you’ve had issues with the valve on your hot water heat register,
please let me know as soon as possible. If a repair is needed, it could
be handled on 12/6
preventing another heat outage.
Stay warm at heart,
Tera Martin-Roy
Department of Economics
University of Illinois at Urbana-Champaign
214 David Kinley Hall, MC-707
1407 W. Gregory Dr.
Urbana, IL 61801
Ph: 217-300-3684
F: 217-244-6678
Tuesday, November 28, 2017
Schedule for Final Quiz
Our final exam is scheduled for 8 AM to 11 AM on Monday, December 18th. I plan to do an online quiz instead, in similar manner to the other quizzes. That is, it will be given in Moodle. I have posted this in the class calendar to start at 11 AM on Sunday December 17th and close on December 18th at noon. This gives you a full day when to take it and part of the time overlaps the scheduled time of the final.
I hope, very much, that this does not create a conflict for anyone. If it does, however, note that I will not give a conflict exam earlier than this. If you need to take a conflict, it will have to be after the 18th. In this case, you should notify my asap so I can make arrangements.
I hope, very much, that this does not create a conflict for anyone. If it does, however, note that I will not give a conflict exam earlier than this. If you need to take a conflict, it will have to be after the 18th. In this case, you should notify my asap so I can make arrangements.
Plagiarism and Moral Hazard
Nobody responded to this in class today about my question of whether you intended to go do graduate school in economics, but consider this (very low probability) scenario where every student in the class raised their hands, either earnestly or because they misrepresented their preferences in a way that they thought would advantage them. How would I have reacted in that case? (Note that to get into graduate school you need letters of recommendation from your professors.) This would have created some moral hazard for me!
Here are the links to the papers I showed about plagiarism. This essay by Jonathan Lethem is really a great read, so I encourage you to have a go at it after the semester is over. It will get you to reconsider your own beliefs about what "original thinking" means and how much of a debt we have to contributions that have come before our own. Also, it might help you understand that one joy I have as a professor is when some former student plagiarizes me, and then lets me know about it.
And here is the piece of mine called Do we plagiarize inadvertently? The thought behind the piece is that we (professors, librarians, campus administrators) read students the riot act about academic integrity, but don't attend to the practical education or how to deal with citation (or lack thereof). I do think as a good citizen you should make some efforts to cite properly, even in informal communication, but before too long there are diminishing returns, so there is no doubt we are less careful about citation in informal communication.
There is one topic I had wanted to discuss in class today but forgot to do so. This is about citing works that you haven't read, or have read but didn't understand. What is the right thing to do in this case? What is the common practice?
On a team writing effort where there is some comparative advantage among the authors, it is my view that as long as one member of the team has read the work, and it is that person who authors the text where the work is cited, that is sufficient. Other team members can trust their teammate on this one. (This usually works, although you can imagine the pitfalls that are possible.) In a solo writing effort, if you cite a work that you haven't read, and you do so incorrectly, you run the risk of discrediting your own piece. If you recall our little discussion today about detection, you might very well get away with it. But then again, you might not. If the purpose for the citation is to show that you've done your homework, then do your homework. That is the simple message.
However, I will confess to, for example, taking a concept I know already, finding the Wikipedia page on the concept or some other generalist source, and then linking to it without reading the whole page thoroughly. Sometimes Wikipedia has errors in it, so this practice is not without risk. However, it is mostly benign. I do it to show I'm not making the concept up, that I learned it ahead of time.
Regarding works that you don't understand, there are some famous works (Keynes' General Theory comes to mind) that are very difficult and hardly anybody understands them, but they were seminal in a line of thinking that followed. Those works should be cited, with reverence. Otherwise, it is quite a risky think to cite a paper you don't understand that others do understand, unless you are asking help from them to explain it to you.
Here are the links to the papers I showed about plagiarism. This essay by Jonathan Lethem is really a great read, so I encourage you to have a go at it after the semester is over. It will get you to reconsider your own beliefs about what "original thinking" means and how much of a debt we have to contributions that have come before our own. Also, it might help you understand that one joy I have as a professor is when some former student plagiarizes me, and then lets me know about it.
And here is the piece of mine called Do we plagiarize inadvertently? The thought behind the piece is that we (professors, librarians, campus administrators) read students the riot act about academic integrity, but don't attend to the practical education or how to deal with citation (or lack thereof). I do think as a good citizen you should make some efforts to cite properly, even in informal communication, but before too long there are diminishing returns, so there is no doubt we are less careful about citation in informal communication.
There is one topic I had wanted to discuss in class today but forgot to do so. This is about citing works that you haven't read, or have read but didn't understand. What is the right thing to do in this case? What is the common practice?
On a team writing effort where there is some comparative advantage among the authors, it is my view that as long as one member of the team has read the work, and it is that person who authors the text where the work is cited, that is sufficient. Other team members can trust their teammate on this one. (This usually works, although you can imagine the pitfalls that are possible.) In a solo writing effort, if you cite a work that you haven't read, and you do so incorrectly, you run the risk of discrediting your own piece. If you recall our little discussion today about detection, you might very well get away with it. But then again, you might not. If the purpose for the citation is to show that you've done your homework, then do your homework. That is the simple message.
However, I will confess to, for example, taking a concept I know already, finding the Wikipedia page on the concept or some other generalist source, and then linking to it without reading the whole page thoroughly. Sometimes Wikipedia has errors in it, so this practice is not without risk. However, it is mostly benign. I do it to show I'm not making the concept up, that I learned it ahead of time.
Regarding works that you don't understand, there are some famous works (Keynes' General Theory comes to mind) that are very difficult and hardly anybody understands them, but they were seminal in a line of thinking that followed. Those works should be cited, with reverence. Otherwise, it is quite a risky think to cite a paper you don't understand that others do understand, unless you are asking help from them to explain it to you.
Today's Quiz
Monday, November 27, 2017
Some Ideas for the Virtual Elevator Speech
I made a template Presentation for you to download and review. It can be played in slideshow mode and then plays automatically, with background music. The slides are meant as instructions on how to make your presentation. So that covers the layout part.
Content-wise, the big deal is image selection. Some thinking should go into what sort of image best represents the idea you want to communicate. I expect that to be the bulk of the work, with there also some effort put into your slide titles to tie into the images and establish connections with your paper. The images should be taken from the Internet, but don't use one which has a copyright notice on it. Otherwise, everything is fair game, though do note we are a G-Rated class and your content should fit that requirement.
As to the music, it is best for it to be purely instrumental, no vocals. This way the music complements the content on the slides and doesn't compete with it. One possible source of such music is to find Karaoke versions of popular songs. Here popular means for students, not for me. Remember, other students in the class are your primary audience. If you find such music on YouTube, there are many converters out there to make an mp3 (audio) file from the video. I've used this one with good success.
You need to put in timings/transitions for the slides. This requires a bit of arithmetic like this - if you have 20 slides and you want the presentation to run for 2 minutes then that is 10 slides per minute or 6 seconds per slide. You don't need to have it completely uniform that way, but it is certainly easier. Slides with a lot of text (you won't have many of these) should have some more time to allow the viewer to read what is on the slide.
Finally, the music should autoplay. There are some tools for doing that.
If you need help with this I can certainly provide that, but my hope is that at least one team member will have these skills and this sort of things is something that each of you should learn, if you don't already know how to do it.
Content-wise, the big deal is image selection. Some thinking should go into what sort of image best represents the idea you want to communicate. I expect that to be the bulk of the work, with there also some effort put into your slide titles to tie into the images and establish connections with your paper. The images should be taken from the Internet, but don't use one which has a copyright notice on it. Otherwise, everything is fair game, though do note we are a G-Rated class and your content should fit that requirement.
As to the music, it is best for it to be purely instrumental, no vocals. This way the music complements the content on the slides and doesn't compete with it. One possible source of such music is to find Karaoke versions of popular songs. Here popular means for students, not for me. Remember, other students in the class are your primary audience. If you find such music on YouTube, there are many converters out there to make an mp3 (audio) file from the video. I've used this one with good success.
You need to put in timings/transitions for the slides. This requires a bit of arithmetic like this - if you have 20 slides and you want the presentation to run for 2 minutes then that is 10 slides per minute or 6 seconds per slide. You don't need to have it completely uniform that way, but it is certainly easier. Slides with a lot of text (you won't have many of these) should have some more time to allow the viewer to read what is on the slide.
Finally, the music should autoplay. There are some tools for doing that.
If you need help with this I can certainly provide that, but my hope is that at least one team member will have these skills and this sort of things is something that each of you should learn, if you don't already know how to do it.
Sunday, November 26, 2017
Where I am on the Second Drafts/Moving on to the last phase of the project
So far I have received submissions from 7 teams. I have read and returned 5 of them and I am done for today. If I didn't get to yours yet, I apologize. Something else came up today and that took a chunk of time. I will get it back to you tomorrow. If your team hasn't yet submitted the second draft, you better get it to me tomorrow, early afternoon by the latest. Otherwise, your team will be put in jeopardy for not doing the work.
Many teams didn't do the references correctly. While I try to be causal about many matters, references in papers are something that academics take seriously. In the body of the paper, the reference should read Last Name of Author(s) (year of publication). Then at the end of the paper there should be a references list with a full citation of the paper. You ultimate submission of the your paper has to have the references done correctly.
In addition, I marked up papers, as my way to read through them again seriously. I want to note that I enjoyed the cover letters and really it is the best thing I did about the project - requiring them. One thing I hope you get from my class is that thinking is conversation. The cover letters are conversation with your thinking in response to mine. I wish I could do more assignments like that.
However, I didn't mark them up. From the point of view of production, the cover letters are intermediate product. The next version of the paper that you will submit is closer to final product.
Once I receive those and verify the references are done okay, I will post your papers to a folder in Moodle, so if you are interested you can see the papers of other teams. Having your paper posted in Moodle means your team is cleared to do the last phase of the project, the virtual elevator speech.
Many teams didn't do the references correctly. While I try to be causal about many matters, references in papers are something that academics take seriously. In the body of the paper, the reference should read Last Name of Author(s) (year of publication). Then at the end of the paper there should be a references list with a full citation of the paper. You ultimate submission of the your paper has to have the references done correctly.
In addition, I marked up papers, as my way to read through them again seriously. I want to note that I enjoyed the cover letters and really it is the best thing I did about the project - requiring them. One thing I hope you get from my class is that thinking is conversation. The cover letters are conversation with your thinking in response to mine. I wish I could do more assignments like that.
However, I didn't mark them up. From the point of view of production, the cover letters are intermediate product. The next version of the paper that you will submit is closer to final product.
Once I receive those and verify the references are done okay, I will post your papers to a folder in Moodle, so if you are interested you can see the papers of other teams. Having your paper posted in Moodle means your team is cleared to do the last phase of the project, the virtual elevator speech.
Wednesday, November 22, 2017
Happy Thanksgiving/About the Projects
I want to wish everyone in the class a happy Thanksgiving. I hope the break has been good and relaxing for you.
---------
At present 4 teams have submitted their second drafts and cover letters. I will start looking at this stff on Friday, and I'd like to be done no later than Sunday afternoon, so be advised that was a soft deadline will become increasingly hard over the next few days.
You might be amused by one of my rhymes on this point, written early last Saturday, when only one submission had come in.
---------
At present 4 teams have submitted their second drafts and cover letters. I will start looking at this stff on Friday, and I'd like to be done no later than Sunday afternoon, so be advised that was a soft deadline will become increasingly hard over the next few days.
You might be amused by one of my rhymes on this point, written early last Saturday, when only one submission had come in.
Thursday, November 16, 2017
A little puzzle about gift exchange and pay for performance from our class
Sometimes there are natural experiments done that produce data and then let economists (and other social scientists) try to explain the findings with some plausible theoretical explanation. In our class we have had such a natural experiment.
Early on there were class session evaluations that students would receive 5 points for completing. There were closed ended questions that were required and an optional paragraph question for comments, meaning students didn't have to comment to receive the 5 points. Yet many students did, in fact, post a comment, some with high frequency. In the language of the class, those comments were gifts (to me). You would know better than I would about the motivation for writing such a comment.
Since we ended the class session evaluations, students have had the option of posting a comment on the Feedback Tab at the class site. This is one click away from the homepage. So, regarding effort, I would interpret that as pretty minimal. It would seem that if you were inclined to give a gift in the form of a comment, you'd still be doing that. Yet nobody has submitted such a comment. It's as if by going from session evaluations to this other method, I shut down the comment pipeline.
That observation offers a puzzle. You might explain things by outside factors, such as where we are in the semester. (The number of people completing the class session evaluations had been declining.) Another possibility, however, is that you have to initiate the activity before you get into gift giving mode and the 5 points bonus as incentive was needed to get you to initiate. This is one where students reflecting on their own motivation might shed some light on what explains things. You know what drives you better than I do.
If the need to initiate offers the better explanation, then you might take this finding and ask what it tells a manager. How often must the manager provide some carrot for employees so they reinitiate? Also, what carrots make sense here? We talked about employees on a salary not bearing income risk. Are there other carrots that might be used if such an employee seems stuck in rut and needs to resume working at a high level?
Early on there were class session evaluations that students would receive 5 points for completing. There were closed ended questions that were required and an optional paragraph question for comments, meaning students didn't have to comment to receive the 5 points. Yet many students did, in fact, post a comment, some with high frequency. In the language of the class, those comments were gifts (to me). You would know better than I would about the motivation for writing such a comment.
Since we ended the class session evaluations, students have had the option of posting a comment on the Feedback Tab at the class site. This is one click away from the homepage. So, regarding effort, I would interpret that as pretty minimal. It would seem that if you were inclined to give a gift in the form of a comment, you'd still be doing that. Yet nobody has submitted such a comment. It's as if by going from session evaluations to this other method, I shut down the comment pipeline.
That observation offers a puzzle. You might explain things by outside factors, such as where we are in the semester. (The number of people completing the class session evaluations had been declining.) Another possibility, however, is that you have to initiate the activity before you get into gift giving mode and the 5 points bonus as incentive was needed to get you to initiate. This is one where students reflecting on their own motivation might shed some light on what explains things. You know what drives you better than I do.
If the need to initiate offers the better explanation, then you might take this finding and ask what it tells a manager. How often must the manager provide some carrot for employees so they reinitiate? Also, what carrots make sense here? We talked about employees on a salary not bearing income risk. Are there other carrots that might be used if such an employee seems stuck in rut and needs to resume working at a high level?
Further thoughts on today's class session
I wanted to note something more general about the first part of the session. We had fun. And I believe students learned something. My ideal is to have "educative play" where participants grow from the experience and enjoy it at the same time. This is distinct from "dissipative play" where there is fun but no personal growth. You almost surely need some of that too, to recharge your batteries, but the thought is that if you can combine work and play then life is good. For a few minutes today, I thought we achieved that.
A couple of other points about schmoozing I should have made. It is a generalist skill. If you are a good schmoozer you should be able to do that with lots of quite different people. Thus, it benefits from good generalist reading, I made a suggestion to do this a while back (it is in the last section of this post) where I said it would help you climb the ranks of management. It will also definitely help you in schmoozing. It is especially good if you can relate the discussion to some reading and stay in context the whole time. Talking with people is one way to stay informed. Reading good an interest stuff is another. You really should be doing both.
The other point is about how instrumental the class seemed at first on why to engage in the schmoozing. I hope it will be obvious to you that if you are overtly instrumental about that or any other behavior, then you will not appear genuine. In other words, being instrumental can block your ability to listen. So, I believe, the primary reasons to schmooze is because it is enjoyable. The connections you make as a consequence should be considered byproduct only. If you reverse the order of priority, you won't get as much out of doing it.
---------------
I forgot something not class related, but that I wanted to announce to the class. Bur first here is an economic theory of predictions. If you are going to make a prediction, it is better for it to be outrageous. In the rare even that you are proven right, you'll be considered a genius, with outstanding insight. In the complementary event where you are proven wrong, you hope that people will quickly forget bout your prediction and, after all, life is a random walk, so it can't be forecast.
Now with that as background, I'm predicting that Carlos Beltran will be the next manager of the New York Yankees. ESPN says this is a long shot.
You heard it here first.
-----------------
The Sherry Turkle piece I briefly showed is called Stop Googling. Let's Talk. You should definitely read it. Soon after it appeared I had a little email thread with her about it. Since she responded to my query I was emboldened to send today's rhyme qua lesson plat to her. At least she acknowledged receipt. That is below.
----------------
A couple of other points about schmoozing I should have made. It is a generalist skill. If you are a good schmoozer you should be able to do that with lots of quite different people. Thus, it benefits from good generalist reading, I made a suggestion to do this a while back (it is in the last section of this post) where I said it would help you climb the ranks of management. It will also definitely help you in schmoozing. It is especially good if you can relate the discussion to some reading and stay in context the whole time. Talking with people is one way to stay informed. Reading good an interest stuff is another. You really should be doing both.
The other point is about how instrumental the class seemed at first on why to engage in the schmoozing. I hope it will be obvious to you that if you are overtly instrumental about that or any other behavior, then you will not appear genuine. In other words, being instrumental can block your ability to listen. So, I believe, the primary reasons to schmooze is because it is enjoyable. The connections you make as a consequence should be considered byproduct only. If you reverse the order of priority, you won't get as much out of doing it.
---------------
I forgot something not class related, but that I wanted to announce to the class. Bur first here is an economic theory of predictions. If you are going to make a prediction, it is better for it to be outrageous. In the rare even that you are proven right, you'll be considered a genius, with outstanding insight. In the complementary event where you are proven wrong, you hope that people will quickly forget bout your prediction and, after all, life is a random walk, so it can't be forecast.
Now with that as background, I'm predicting that Carlos Beltran will be the next manager of the New York Yankees. ESPN says this is a long shot.
You heard it here first.
-----------------
The Sherry Turkle piece I briefly showed is called Stop Googling. Let's Talk. You should definitely read it. Soon after it appeared I had a little email thread with her about it. Since she responded to my query I was emboldened to send today's rhyme qua lesson plat to her. At least she acknowledged receipt. That is below.
----------------
From: Sherry Turkle [sturkle@media.mit.edu]
Sent: Thursday, November 16, 2017 10:04 AM
To: Arvan, Lanny
Subject: Re: I thought you might enjoy this.
Sent: Thursday, November 16, 2017 10:04 AM
To: Arvan, Lanny
Subject: Re: I thought you might enjoy this.
Thank you!
S
Teaching students to schmooze
Over coffee rather than booze
And to enjoy the doing
Even if misconstruing
Good face to face conversation is news.
#ItTurnsOutTalkIsntCheap
Found here: https://twitter.com/larvan3/status/931120054854352897
Wednesday, November 15, 2017
For Our Discussion on Learning to Be a Leader
This posts is pretty short and I believe students will find it interesting. There is a link to a column by Sherry Turkle a little later in the piece. That is a must read.
Thursday, November 9, 2017
Second Quiz in Moodle - Date Set
We agreed in class today that the second quiz should be on Thursday Nov. 30, starting right after class, and going through Friday December 1, with it closing at 5 PM. The format might be a tad different than the last time, but not too much. After it is created and after Thanksgiving break, I'll let you know.
I also committed to the Final Quiz to be online in Moodle on and around when our Final Exam is scheduled in the timetable. (Somebody in class said that was December 18 but I haven't verified that yet.) So, if Econ 490 is your last exam, you can leave town before the final and do it from home.
We should all be thankful for online technology.
I also committed to the Final Quiz to be online in Moodle on and around when our Final Exam is scheduled in the timetable. (Somebody in class said that was December 18 but I haven't verified that yet.) So, if Econ 490 is your last exam, you can leave town before the final and do it from home.
We should all be thankful for online technology.
Excel Homework Due 11/29 at 11 PM
Please watch this video first. It gives the algebra of the principal-agent model and will make you comfortable with the notation and the basic ideas.
Then do the Excel homework, which does the same analysis but this time with a graphical approach.
------
This is the simplest possible model of the principal-agent, with low or high effort and low or high output. I am going to describe possible extensions of the model for you to consider (and which we might discuss in class) but which we won't model.
The first is to make output a continuous variable. This, it turns out isn't conceptually that much harder, as long as higher output makes it more likely that the agent chose high effort. In that case, the reward scheme should also rise with output. In many real world reward schemes, the scheme is linear. For example, it entails a base pay and then some percentage share of the output. A lot of sales people work under schemes of this sort. Linearity is hard to explain from the model, but it is easy to understand in that it can be readily communicated. More complex schemes would be harder to articulate and to understand.
The second is to make effort a continuous variable, with higher effort making higher output more likely. This too is not that hard to conceptualize, but then you lose the notion of shirking in the model, unless you establish a cutoff effort level below which the agent is said to shirk.
Further extensions make both output and effort multi-dimensional. This is definitely harder to consider theoretically, but it is far more realistic, especially for knowledge work. As I have said in class before, professional knowledge workers typically don't shirk (play solitaire on their computers all day or do Facebook, though they might do it as a diversion from time to time to refresh themselves). Such employees have a strong sense of autonomy. The issue for the manager is to align what these workers are doing with organizational objectives. Monitoring happens to achieve alignment, not to prevent shirking.
The last extension I will mention is considering what happens when output has a large random component to it so it is not a very informative signal. A little bit of randomness doesn't change the story much. A lot of randomness does. (The agent is risk averse and if you make a risk averse individual absorb a lot of risk, you need to pay a substantial risk premium to compensate for that.) So instead, one looks for other solutions and then not rely on rewards for high output.
Then do the Excel homework, which does the same analysis but this time with a graphical approach.
------
This is the simplest possible model of the principal-agent, with low or high effort and low or high output. I am going to describe possible extensions of the model for you to consider (and which we might discuss in class) but which we won't model.
The first is to make output a continuous variable. This, it turns out isn't conceptually that much harder, as long as higher output makes it more likely that the agent chose high effort. In that case, the reward scheme should also rise with output. In many real world reward schemes, the scheme is linear. For example, it entails a base pay and then some percentage share of the output. A lot of sales people work under schemes of this sort. Linearity is hard to explain from the model, but it is easy to understand in that it can be readily communicated. More complex schemes would be harder to articulate and to understand.
The second is to make effort a continuous variable, with higher effort making higher output more likely. This too is not that hard to conceptualize, but then you lose the notion of shirking in the model, unless you establish a cutoff effort level below which the agent is said to shirk.
Further extensions make both output and effort multi-dimensional. This is definitely harder to consider theoretically, but it is far more realistic, especially for knowledge work. As I have said in class before, professional knowledge workers typically don't shirk (play solitaire on their computers all day or do Facebook, though they might do it as a diversion from time to time to refresh themselves). Such employees have a strong sense of autonomy. The issue for the manager is to align what these workers are doing with organizational objectives. Monitoring happens to achieve alignment, not to prevent shirking.
The last extension I will mention is considering what happens when output has a large random component to it so it is not a very informative signal. A little bit of randomness doesn't change the story much. A lot of randomness does. (The agent is risk averse and if you make a risk averse individual absorb a lot of risk, you need to pay a substantial risk premium to compensate for that.) So instead, one looks for other solutions and then not rely on rewards for high output.
Wednesday, November 8, 2017
I'm Back from Michigan and Catching Up
If you'd like to see what I attended, take a look here. In some sense it was like a return to being an undergraduate. I sat through a full lecture where more than 90% of the presentation was over my head! But when it is your brother who is giving the lecture it has interests for other reasons. I didn't check my phone once during the entire thing.
I've done a quick scan of YouTube, where I see the video meant as a substitute for yesterday's class has gotten some hits (the data about minutes watched hasn't been processed yet) and I saw in Moodle that 16 students completed the softball quiz associated with the video. We'll discuss this a bit tomorrow as we try to catch up.
I want to note here the two additional obligations before the Thanksgiving break. There is a blog post due tomorrow on conflict. In some sense this post is on the opposite end of the spectrum for the one for last week. By contrasting the two I believe there will be more meaning for you in each case. Then, of course, second drafts of your project papers are due a week from Friday.
I've done a quick scan of YouTube, where I see the video meant as a substitute for yesterday's class has gotten some hits (the data about minutes watched hasn't been processed yet) and I saw in Moodle that 16 students completed the softball quiz associated with the video. We'll discuss this a bit tomorrow as we try to catch up.
I want to note here the two additional obligations before the Thanksgiving break. There is a blog post due tomorrow on conflict. In some sense this post is on the opposite end of the spectrum for the one for last week. By contrasting the two I believe there will be more meaning for you in each case. Then, of course, second drafts of your project papers are due a week from Friday.
Monday, November 6, 2017
A linkage between altruism and intrinsic motivation
In the Excel Homework due after the Thanksgiving break, we will take on the principal-agent model, which you can think of as giving a basis for performance based pay. Some type of work, notably jobs in sales, typically entail performance based pay. So we need to understand why that is.
Yet the latest blog post you wrote on teamwork and gift exchange included reacting to a piece on altruism The column by David Brooks argued that altruistic motives are inhibited when economic incentives are present. The lesson from Brooks' piece is that when altruism is important for organization productivity (as in the gift exchange model, the Okun implicit contracts view, or the Akerlof and Kranton view where identity is important) then economic incentives need to move to the background. It is my view that you can't eliminate economic incentives entirely and shouldn't try. But performance based pay moves economic incentives front and center so the incentives are always present. Type A personalities may want such an environment. The B personalities prefer something that is more low key. Yet the research which Brooks reports on argues that either type of personality will be less altruistic in the presence of performance pay.
There is a different reason to depart from performance based pay that is intellectual in nature and is particular to knowledge work. This other view is articulated by Daniel Pink in this RSA Animiate video and focuses on psychological explanations for productivity. There can be performance anxiety or, if you prefer, writer's block. If this persists, evidently it lowers productivity. High performance is achieved when intrinsic motivation is strong and the individual becomes so involved in the work as to entirely lose a sense of self. Making extrinsic rewards overt then moves the individual's focus away from the intrinsic motivation and thereby lessons productivity. Better to have the economic rewards provided up front, so that one can put them out of mind when the real work commences.
I have critiqued this video on how it represents the economics. You can't put the economic incentives away forever after a single, up-front contracting process. There will be subsequent re-contracting to enlist the employee again in the future, especially after word about the previous work has gotten out and the employee's reputation has been enhanced as a consequence. Yet I concur with its representation of the importance of intrinsic motivation. So my view is that the manager tries to achieve a balance between the two types of motivation. You might consider time broken up into episodes. Each episode begins with extrinsic rewards to enlist participation. After that the work begins and the rewards fade into the background. Then intrinsic motivation takes over for a time. As the episode reaches conclusion the extrinsic motivation returns and the next episode begins.
This system is by no means perfect. (The economic rewards can reappear too early and interfere with intrinsic motivation when the work is not yet completed. Alternative, the manager might pretend the work is not yet done so as not to make an appropriate upward adjustment in the employee's pay, creating hard feelings in the process and inadvertently encouraging turnover.) But it is better, in my view, than the alternative where rewards for performance are omnipresent. An environment with strong pay for performance incentives is likely to inhibit both altruism and intrinsic motivation.
Yet the latest blog post you wrote on teamwork and gift exchange included reacting to a piece on altruism The column by David Brooks argued that altruistic motives are inhibited when economic incentives are present. The lesson from Brooks' piece is that when altruism is important for organization productivity (as in the gift exchange model, the Okun implicit contracts view, or the Akerlof and Kranton view where identity is important) then economic incentives need to move to the background. It is my view that you can't eliminate economic incentives entirely and shouldn't try. But performance based pay moves economic incentives front and center so the incentives are always present. Type A personalities may want such an environment. The B personalities prefer something that is more low key. Yet the research which Brooks reports on argues that either type of personality will be less altruistic in the presence of performance pay.
There is a different reason to depart from performance based pay that is intellectual in nature and is particular to knowledge work. This other view is articulated by Daniel Pink in this RSA Animiate video and focuses on psychological explanations for productivity. There can be performance anxiety or, if you prefer, writer's block. If this persists, evidently it lowers productivity. High performance is achieved when intrinsic motivation is strong and the individual becomes so involved in the work as to entirely lose a sense of self. Making extrinsic rewards overt then moves the individual's focus away from the intrinsic motivation and thereby lessons productivity. Better to have the economic rewards provided up front, so that one can put them out of mind when the real work commences.
I have critiqued this video on how it represents the economics. You can't put the economic incentives away forever after a single, up-front contracting process. There will be subsequent re-contracting to enlist the employee again in the future, especially after word about the previous work has gotten out and the employee's reputation has been enhanced as a consequence. Yet I concur with its representation of the importance of intrinsic motivation. So my view is that the manager tries to achieve a balance between the two types of motivation. You might consider time broken up into episodes. Each episode begins with extrinsic rewards to enlist participation. After that the work begins and the rewards fade into the background. Then intrinsic motivation takes over for a time. As the episode reaches conclusion the extrinsic motivation returns and the next episode begins.
This system is by no means perfect. (The economic rewards can reappear too early and interfere with intrinsic motivation when the work is not yet completed. Alternative, the manager might pretend the work is not yet done so as not to make an appropriate upward adjustment in the employee's pay, creating hard feelings in the process and inadvertently encouraging turnover.) But it is better, in my view, than the alternative where rewards for performance are omnipresent. An environment with strong pay for performance incentives is likely to inhibit both altruism and intrinsic motivation.
Blog Posts/Reminder - No Class Tomorrow/Steve Blass Disease
I am now caught up in commenting on the posts from this past week. I'm sorry that it took longer than usual. It was hard to find the time to do this while I'm in Michigan.
* * * * *
There is no class tomorrow as I will be driving back from Michigan during scheduled class time. Here is the information about what you should do instead.
* * * * *
On the drive up, while I tried to listen to classical music on my phone, my wife was listening to some podcasts from This American Life that played on the car's audio system. I couldn't avoid hearing that too and some of it I paid attention to. This segment on Steve Blass Disease is interesting, in itself and also applied to our class. The issue of whether a top performer suddenly can no longer perform well at all, and if there is an explanation it has nothing to do with physical well being nor with shirking, makes one wonder whether it goes beyond sports. In golf it is called the yips. I am not sure whether performance anxiety properly captures the idea because many people do overcome performance anxiety. We are talking about a situation that is so debilitating that it is not overcome.
Act One: Aces Are Wild
* * * * *
There is no class tomorrow as I will be driving back from Michigan during scheduled class time. Here is the information about what you should do instead.
* * * * *
On the drive up, while I tried to listen to classical music on my phone, my wife was listening to some podcasts from This American Life that played on the car's audio system. I couldn't avoid hearing that too and some of it I paid attention to. This segment on Steve Blass Disease is interesting, in itself and also applied to our class. The issue of whether a top performer suddenly can no longer perform well at all, and if there is an explanation it has nothing to do with physical well being nor with shirking, makes one wonder whether it goes beyond sports. In golf it is called the yips. I am not sure whether performance anxiety properly captures the idea because many people do overcome performance anxiety. We are talking about a situation that is so debilitating that it is not overcome.
Act One: Aces Are Wild
Saturday, November 4, 2017
Please Get Your Blog Posts In Today
We are leaving for Michigan in a few minutes. I won't look at posts any more today, but I will tomorrow morning (whenever I get up, East Coast time.)
And for those of you who need this information, Daylight Savings ends tonight. You get an extra hour. I wonder what you'll do with it.
Friday, November 3, 2017
What's the answer?
I have a friend who was a few years ahead of me as an assistant professor at Illinois. His area of specialization is energy economics. (Think big oil.) He left Illinois to go to the University of Houston. He subsequently went to SMU and is a chaired professor there. We stay in touch, primarily via a mutual colleague. Recently the topic is baseball. He is quite an Astros fan now. He wrote:
What is the answer to this question?
"The Yanks look like they could be the Astros/Cubs of 2018. Quick quiz: if the two teams are evenly matched, which is more likely, a 6 or 7 game series?'
What is the answer to this question?
Further thoughts on the economics of insurance
Because the Affordable Care Act has become such a political football, I shied away from discussing it in class yesterday, even though it would be a way to bring in current events into what we are studying. But I will talk about healthcare some in this post and try to do so in as dispassionate a way as I can.
First, however, we should consider that labor contracts have an insurance aspect to them, particularly when employees are on salary or when they work at an hourly wage for a pre-specified number of hours per week. Abstracting from inflation risk and the possibility of losing the job entirely, the income of such workers is predictable and therefore is safe. Productivity (and here we don't just mean the physical output but also want to include the demand for the product being produced) is likely to be risky. If employees have fixed income contracts then it is the employer who bears the productivity risk. There is some argument that employees should share in that risk, to increase their sense of ownership in the firm. But on straight risk aversion grounds, it is reasonable to assume that the employer is less risk averse than the employees. Indeed, one real reason to have large corporations with shares of stock that are publicly traded is that owners diversify their holdings in many such firms, which greatly reduces the idiosyncratic firm risk in their portfolio (though systematic market risk remains). If you treat shareholders as achieving good diversification and/or the company is largely held by a rich person who can achieve diverse holdings and/or self-insure, then flat payments to employees makes sense. The implicit contracts view of the labor market then would suggest that in exchange for the firm absorbing the productivity risk the employees reward the firm with loyalty, which is valuable, especially in tight labor markets.
Starting with this as a basis, the role of bonus payments to employees, as well as compensating employees with stock rather than with wages, now fairly common practices, requires some explanation. Bonuses themselves are of multiple types. One sort of bonus is awarded to each employee when the firm itself does well. Another type of bonus is awarded to a particular employee when that employee has performed well. In each of the cases, the employees are absorbing some of the productivity risk. The question is why. One possible explanation is that the employer can't diversify nearly as well as was assumed in the previous paragraph. Another is that these forms of compensation are there for incentive reasons to motivate the employee (this is the most commonly held explanation). A third reason might be to reflect market power in the labor market by the employer.
Let's consider one other aspect of this, which is how wages and salaries change over time. If there is a jobladder, then normally employees are paid more as they climb the ladder. Within rank, payments may go up over time to reflect some return to seniority and/or to serve as implicit cost of living adjustments. Employees do bear some of the risk in climbing the job ladder, particularly of the sort where the next rung is already jammed with people so the firm doesn't promote an otherwise deserving person until the next rung clears out some. This is a form of coordination problem that does induce income risk. A complete safety play for employees doesn't exist; even having tenure at a university is not a completely safe situation income-wise.
* * * * *
I want to turn to the economics of healthcare - just a little bit - so you can relate the theory we did to some real world issues.
First, let me observe that much healthcare is routine and predictable. For example, I am at the age where my primary care physician wants to see me on a regular basis (every six months or so). There is no moral hazard in this sort of thing, yet there is a co-pay with each visit. The question is why. (Or put another way, why isn't the co-pay instead bundled into the premium?) The answer, I believe, is that there is some advantage to the provider in keeping premiums down when it bids for the state contract. Higher premiums make the bid appear less attractive. (Incidentally, this same reason offers some explanation for why much government procurement ends up in cost-overruns.) But the provider still must recover its costs, so user fees for visits emerge as one way to do that. My HMO calls that user fee a co-pay. When it is evident that there is no moral hazard, the co-pay should be zero, according to the theory. But once the provider has a way to charge co-pays in other circumstances where it does make sense, then co-pays can be employed for this other purpose as well.
Second, in our very simple model of insurance the loss was L. For automobile accidents, the adjuster can determine L quite precisely. For a serious health problem, however, determining L can be much harder and interpretation of the evidence can vary (which is why sometimes patients will get a second opinion). Thus, coverage is often not based on determining L. Rather it is based on the treatment that the doctor prescribes or the procedures that the doctor performed on the patient. This sort of relationship between coverage and treatment is sometimes referred to as fee for service. Doctors have some incentive to act opportunistically in this respect to collect more in fees. The moral hazard in this case plays out with the doctor decisions biased towards more treatment than good defensive medicine would indicate.
This issue is coupled with the matter of doctor liability in the event of malpractice. While the insurance company might object to procedures that don't seem necessary, the patient will never do so if there is no adverse consequence from the procedure. Patients are much more likely to object if after the fact the outcome is poor for them and if some procedure that had not been performed might have made the outcome better. This is the general circumstance behind getting sued for malpractice. Getting sued is costly, even when there was no malpractice. So to avoid that contingency, there is an incentive to over prescribe treatments. These factors contribute to the high cost of health care.
Another factor that influences health care costs, much less commented on, is how much doctors communicate with one another when treating a very ill patient. Given our current focus on teamwork, you can think of a high level of communication among the doctors as a well functioning team. All too often, however, specialists act in an uncoordinated way. That can really pile on the costs and make the treatment worse. This piece, well worth the read, illustrates these matters nicely.
I will not comment at all about prescription drug prices other than to note that I believe the main issue is market power.
So I will close on something else, which is depressing to consider but necessary, if we are to get a hold on keeping healthcare costs manageable. This is about expenditure on healthcare during the last year of life. On this I want to separate out care for a child or a teenager with a horrible disease from care for an elderly person. I will consider only the latter. The data show very high expenditure to prolong life just a little bit. The issues are fundamentally ethical. Economics doesn't seem to help here in explaining what is going on. I have lived through this in the case of each of my parents. It is a very emotional thing and the emotions get tied up with the economic decisions. Based on my experience I can say there are no right answers here, just lots of unsatisfactory alternatives.
First, however, we should consider that labor contracts have an insurance aspect to them, particularly when employees are on salary or when they work at an hourly wage for a pre-specified number of hours per week. Abstracting from inflation risk and the possibility of losing the job entirely, the income of such workers is predictable and therefore is safe. Productivity (and here we don't just mean the physical output but also want to include the demand for the product being produced) is likely to be risky. If employees have fixed income contracts then it is the employer who bears the productivity risk. There is some argument that employees should share in that risk, to increase their sense of ownership in the firm. But on straight risk aversion grounds, it is reasonable to assume that the employer is less risk averse than the employees. Indeed, one real reason to have large corporations with shares of stock that are publicly traded is that owners diversify their holdings in many such firms, which greatly reduces the idiosyncratic firm risk in their portfolio (though systematic market risk remains). If you treat shareholders as achieving good diversification and/or the company is largely held by a rich person who can achieve diverse holdings and/or self-insure, then flat payments to employees makes sense. The implicit contracts view of the labor market then would suggest that in exchange for the firm absorbing the productivity risk the employees reward the firm with loyalty, which is valuable, especially in tight labor markets.
Starting with this as a basis, the role of bonus payments to employees, as well as compensating employees with stock rather than with wages, now fairly common practices, requires some explanation. Bonuses themselves are of multiple types. One sort of bonus is awarded to each employee when the firm itself does well. Another type of bonus is awarded to a particular employee when that employee has performed well. In each of the cases, the employees are absorbing some of the productivity risk. The question is why. One possible explanation is that the employer can't diversify nearly as well as was assumed in the previous paragraph. Another is that these forms of compensation are there for incentive reasons to motivate the employee (this is the most commonly held explanation). A third reason might be to reflect market power in the labor market by the employer.
Let's consider one other aspect of this, which is how wages and salaries change over time. If there is a jobladder, then normally employees are paid more as they climb the ladder. Within rank, payments may go up over time to reflect some return to seniority and/or to serve as implicit cost of living adjustments. Employees do bear some of the risk in climbing the job ladder, particularly of the sort where the next rung is already jammed with people so the firm doesn't promote an otherwise deserving person until the next rung clears out some. This is a form of coordination problem that does induce income risk. A complete safety play for employees doesn't exist; even having tenure at a university is not a completely safe situation income-wise.
* * * * *
I want to turn to the economics of healthcare - just a little bit - so you can relate the theory we did to some real world issues.
First, let me observe that much healthcare is routine and predictable. For example, I am at the age where my primary care physician wants to see me on a regular basis (every six months or so). There is no moral hazard in this sort of thing, yet there is a co-pay with each visit. The question is why. (Or put another way, why isn't the co-pay instead bundled into the premium?) The answer, I believe, is that there is some advantage to the provider in keeping premiums down when it bids for the state contract. Higher premiums make the bid appear less attractive. (Incidentally, this same reason offers some explanation for why much government procurement ends up in cost-overruns.) But the provider still must recover its costs, so user fees for visits emerge as one way to do that. My HMO calls that user fee a co-pay. When it is evident that there is no moral hazard, the co-pay should be zero, according to the theory. But once the provider has a way to charge co-pays in other circumstances where it does make sense, then co-pays can be employed for this other purpose as well.
Second, in our very simple model of insurance the loss was L. For automobile accidents, the adjuster can determine L quite precisely. For a serious health problem, however, determining L can be much harder and interpretation of the evidence can vary (which is why sometimes patients will get a second opinion). Thus, coverage is often not based on determining L. Rather it is based on the treatment that the doctor prescribes or the procedures that the doctor performed on the patient. This sort of relationship between coverage and treatment is sometimes referred to as fee for service. Doctors have some incentive to act opportunistically in this respect to collect more in fees. The moral hazard in this case plays out with the doctor decisions biased towards more treatment than good defensive medicine would indicate.
This issue is coupled with the matter of doctor liability in the event of malpractice. While the insurance company might object to procedures that don't seem necessary, the patient will never do so if there is no adverse consequence from the procedure. Patients are much more likely to object if after the fact the outcome is poor for them and if some procedure that had not been performed might have made the outcome better. This is the general circumstance behind getting sued for malpractice. Getting sued is costly, even when there was no malpractice. So to avoid that contingency, there is an incentive to over prescribe treatments. These factors contribute to the high cost of health care.
Another factor that influences health care costs, much less commented on, is how much doctors communicate with one another when treating a very ill patient. Given our current focus on teamwork, you can think of a high level of communication among the doctors as a well functioning team. All too often, however, specialists act in an uncoordinated way. That can really pile on the costs and make the treatment worse. This piece, well worth the read, illustrates these matters nicely.
I will not comment at all about prescription drug prices other than to note that I believe the main issue is market power.
So I will close on something else, which is depressing to consider but necessary, if we are to get a hold on keeping healthcare costs manageable. This is about expenditure on healthcare during the last year of life. On this I want to separate out care for a child or a teenager with a horrible disease from care for an elderly person. I will consider only the latter. The data show very high expenditure to prolong life just a little bit. The issues are fundamentally ethical. Economics doesn't seem to help here in explaining what is going on. I have lived through this in the case of each of my parents. It is a very emotional thing and the emotions get tied up with the economic decisions. Based on my experience I can say there are no right answers here, just lots of unsatisfactory alternatives.
Materials to substitute for class next Tuesday
We won't be having class on Tuesday, as I will be out of town. I have made a video for you to watch and listen to. In the past when I've made videos of this sort, I took some effort in the production so your eyes would focus on the right stuff on the screen. I didn't do it this time - just not enough time to do something more professional. So this will have to do. The content is pretty much right but the production is not.
The associated PowerPoint can be read on its own and might be useful to download.
Then do the softball quiz in Moodle which will be available from Monday at noon till Wednesday at 11 PM. The purpose of this quiz is not to test you on the presentation. It is to give some bonus points to students who take it and take that as a gift as apology for missing class.
The associated PowerPoint can be read on its own and might be useful to download.
Then do the softball quiz in Moodle which will be available from Monday at noon till Wednesday at 11 PM. The purpose of this quiz is not to test you on the presentation. It is to give some bonus points to students who take it and take that as a gift as apology for missing class.
Thursday, November 2, 2017
The Excel Homework with the Appropriate Formulas but with the Answers Omitted.
I have now completed the stuff to serve as substitute for class on Tuesday 11/7. We won't be having class then. In the next post I will provide the information for that.
* * * * *
As promised, here is the Excel file with the formulas (most of which are in column A). I would appreciate learning from any and all of you whether you try to produce such formulas yourself before answering the questions, or if you do that all in your head without writing any of it down. If you care to respond to that, please do so by commenting on this post, or if you prefer your privacy you can do it by email.
Wednesday, November 1, 2017
Excel Homework Due 11/8 at 11 PM
The homework is on bargaining. Note that M&R have a discussion of bargaining in Chapter 5. You should read that. The model they go through has both the buyer and seller having two types. In the Excel homework, the model has a continuum of types on each side of the bargain. In that sense it is harder but also more elegant. And, I believe, it should give you a better understanding about the relationship between the private information and the inefficiency created.
There is a fiction in the model that helps to understand what is going on. The fiction is that there is a third party - an arbitrator - who provides rules for how the bargaining will happen, when trade will occur, and at what price. The fiction is necessary because the model is static. Real world negotiations happen over time and to model that correctly one needs a dynamic model. There are such models, but they are well beyond the scope of our class. So we will keep the modeling relatively simple and wave our hands about what happens in real world bargaining.
There is a rather extensive discussion in this homework before you log in. I encourage you to read that carefully and not gloss over it. It is about how to bargain well and what happens in real-world procurement, which entails a good deal of such bargaining. There are practical lessons here that you can carry over to your work life, even if you don't engage in procurement. .
There is a fiction in the model that helps to understand what is going on. The fiction is that there is a third party - an arbitrator - who provides rules for how the bargaining will happen, when trade will occur, and at what price. The fiction is necessary because the model is static. Real world negotiations happen over time and to model that correctly one needs a dynamic model. There are such models, but they are well beyond the scope of our class. So we will keep the modeling relatively simple and wave our hands about what happens in real world bargaining.
There is a rather extensive discussion in this homework before you log in. I encourage you to read that carefully and not gloss over it. It is about how to bargain well and what happens in real-world procurement, which entails a good deal of such bargaining. There are practical lessons here that you can carry over to your work life, even if you don't engage in procurement. .
Subscribe to:
Posts (Atom)