Saturday, March 21, 2015

The Hardcopy of the Third Edition is Available on Amazon

With a beautiful cover by David Aiello, based in part on an idea by Anarchei.

And it has just been the subject of two posts by my favorite blogger.

Thursday, March 19, 2015

For Law Schools: A Modest Proposal

American law schools suffer from two serious problems, one old, one new. The old problem is the inconsistency between their practice, driven by institutional self-interest, and the moral beliefs of their faculty. The new problem is how to deal with the sharp decline in law school applications over the past few years.

How many students are willing to come to a law school depends on its reputation. That reputation depends, in part, on the performance of its students, how many pass the state bar and how many get jobs, as reflected in the school's rating in the annual U.S. News and World Report rankings. Nobody in the law school business, at least nobody I have met, regards the ratings as a reliable measure of school quality, but everyone watches them. A further reason to care about the performance of the school's students is that if they perform well enough after graduation, they may make lots of money and donate some of it to the school.

The performance of students depends in part on how good a job the school does, in part on how able the students are. Thus every school has an incentive to try to attract good students in order to raise its ratings in order to get more students to apply in order to get the tuition revenue to pay the cost of operating the school. One way of attracting good students is by offering the best students scholarships that pay part or all of the cost of going to law school, and law schools routinely do so. The result is that the best students, the ones who are smart, hard working, and likely to end up with high paying jobs, are  subsidized at the expense of the students lower down in the class who are paying full tuition.

Most law professors have political views that favor benefiting poor people at the expense of rich people. The actual practice of the schools they teach in and help run has precisely the opposite effect. It subsidizes the future rich at the expense of the future poor, the students who, having spent three years and a lot of money getting a law degree, face very uncertain chances of being offered the sort of job that degree is supposed to qualify them for.

That is the old problem. Everyone in the law business knows it, although not everyone chooses to talk about it. 

The new problem is that law school applications have fallen sharply in the past few years, with the result that many law schools face serious budget problems. The only way they can keep enrollment up is by lowering their standards for admission, but lower standards for admission will eventually result in lower ratings, which will make it even harder to maintain enrollment. Schools can try to cut costs, but a large part of the cost is personnel, and a large and expensive part of that is for professors with tenure. It looks like a downward spiral to bankruptcy, at least until enough schools have shrunk or gone out of business to restore the balance between the number of students who want to enroll and the number of seats  law schools want to fill.

I have at least a partial solution to both problems:

Consider a school with a target enrollment of 200. It currently plans to set the lower limit for accepting applications at a level, defined mainly by LSAT score and undergraduate grade point average, that will result in accepting 400, half of whom it expects to enroll.

It instead lowers the cutoff far enough to get an entering class of 250. Fitting them in is no problem because it has sufficient classroom space and teaching staff for more than that, due to the decline in enrollment over the previous several years.

At the end of the first year it sends a message to the fifty students at the bottom of the first year class, warning them that on the basis of their grades so far they are at serious risk of failing to pass the bar. The school offers to refund their first year tuition in full if they choose to drop out. If only thirty accept the offer, a similar message goes to twenty more students. Once the process has been going for a year or two, the school should be able to make a better estimate of the acceptance rate for their offer and so reduce enrollment to 200 in one step.

What is the result?

1. The school ends up with the same revenue as if it followed its original plan and admitted 200 students. Costs are only increased by a little, because the school has excess resources, physical and human, due to past enrollment decline.

2. Since the students least likely to succeed have been warned and offered their money back, the professors may legitimately feel less guilty about taking the money of students who are ultimately not going to make it.

3. First year grades are a considerably better predictor of bar passage rates than the information available at admission, so the school's bar pass rate goes up.

4. In the long run, more students will be willing to apply, because they know if that if law school turns out to be too hard for them they will have an opportunity to leave and get their money back. 

In an earlier post I offered a different approach to the first problem.  

Why Does Highway Construction Take So Long?

I live near the connection of two major interstates, 880 and 280, and frequently take it. Work has been going on to improve the connection for quite a long time; the project was approved in 2011 and under construction as least as early as 2013. Most of the time the resulting constriction is only a minor inconvenience, but during rush hour it can result in a significant delay in getting from one highway to the other.

It seems as though what they are doing should take no more than a few weeks. Four possible explanations of why it is instead taking years occur to me:

1. I may be missing important elements of the process that make it much more time consuming than I would expect. Perhaps there are many steps that have to be taken in sequence. Perhaps some part of the process, such as the drying and curing of concrete, takes much longer than I realize.

2. Perhaps doing it slowly is a little easier than doing it fast and the people making the relevant decisions, the construction firm and the California Department of Transport that employs them, have little or no incentive to take account of the inconvenience to drivers of having a major intersection under construction.

3. Perhaps the construction company has persuaded friends in the Department to agree to contract terms that pay by time rather than by project, making it in their interest to stretch out the process.

4. Perhaps the government officials and the politicians above them believe that longer is better, that the political benefits of keeping drivers aware of their highway dollars at work more than outweigh the political cost of continued delays.

There are probably other possible explanations that have not occurred to me. Does anyone know what the right explanation is? Any evidence on how long such projects take in other states or other countries?

My First Novel: A Belated Comment

Reading occasional Amazon comments on Harald, my first novel, I am struck by the tendency of readers familiar with my nonfiction to assume that the novel is intended as an argument in favor of libertarianism. It is not. 

In the novel, I portray three different societies in the novel, all based on historical models. One, the Vales, where my protagonist comes from, is loosely modeled on saga period Iceland. Another, the kingdom of Kaerlia, where most of the book happens, is loosely modeled on early Norman England. The third, the Empire, the antagonist against which the first two are allied, is a mix of Byzantine, Roman, and Abbasid institutions—the relation between the Emperor and his sons was inspired by the relation between Haroun al Rashid and his, although my emperor is considerably more competent than al Rashid was.

Each system has strengths and weaknesses. The most obvious weakness of the Vales is that it has neither a professional army supported by taxation nor a feudal levy. Harald, my protagonist, faces the continuing problem of raising an army of volunteers. The historical models I was thinking of were the Norse armies that ravaged England in the period leading up to Alfred the Great. As best I can tell, they were not national armies but entrepreneurial projects, warriors following a leader with a good reputation in search of land and loot.

One result is that Harald's military strategy largely consists of logistic warfare, of maneuvering the opposing army into a position where it must either surrender or die of hunger or thirst. He cannot afford to fight battles where a significant number of his troops get killed because if he does, nobody will show up next time. He is fighting a defensive war, so cannot offer land to his troops, but does offer loot, captured from the armies he is fighting or paid as ransom for legionaries forced to surrender. Also excitement, glory, and training in useful skills—when his people are not fighting the Empire, some of them are serving it as caravan guards and the like. And it helps that, in addition to his own troops, he has under his command the feudal army of his ally the king of Kaerlia.

The Empire has, from a military standpoint, several significant advantages. It has a highly trained professional army, the legions, which Harald describes as the best infantry in the world. It has the sort of culture that keeps written records and uses them. Harald has been fighting the Empire for more than twenty years and his current opponents have read detailed accounts of his past battles, which means that he gets to use any particular trick only once. To balance that, it has a culture with less room for originality and initiative than his. Artos, the best of the Imperial commanders, is technically as good a general as Harald. But he is less original, less unconventional, because an imperial officer with Harald's approach to solving problems would have been fired, possibly hanged, long before he reached high rank. The only place for that approach in the imperial system is at the very top, in the contest for the throne, and the one example we see of it is due not to Artos but to his employer, the younger and abler son of the Emperor.

As I try to make clear, the reason the Empire has not succeeded in conquering the Vales and the Kingdom is not that its system is inferior to theirs. It is that Harald, his ally the king of Kaerlia (who dies just before the novel starts), and his other ally the Lady Commander of the Order, happen to be extraordinarily talented people and close friends—two very able military commanders and a sovereign sufficiently wise to not only get and retain allies, but give his allies command over the combined army on the grounds that they are better at the job than he is.

I have written and published two novels. Both reflect my political interests and my professional interests as an economist. Neither of them is intended as an argument in favor of libertarianism. 

And for those interested:

Harald on Amazon
Harald web page
Harald as free podcasts
Salamander kindle
Salamander hardcopy

Monday, March 02, 2015

Redit AMA planned for this Wednesday

2 P.M. PST.

For the reason see.

Wednesday, February 18, 2015

Exploiting my Thirty Minutes of Fame

Someone recently posted, somewhere online, a picture of me playing World of Warcraft while listening to a talk at the International Students for Liberty conference last weekend. It got quite a lot of attention, including, I am told, "over 5,000,000 click rate and 10k SHARE" in China.

My guess is that there are two things going on. One is that people who already know of my existence see this as evidence that I am a real person, not some academic off in outer space somewhere. The other is that people who have been told by their parents or other authority figures that playing WoW is a juvenile activity and a waste of time see the picture as evidence on their side of the argument.

My son Bill, having observed the phenomenon online, suggests that I ought to find some way of taking advantage of my temporary fame. His only idea was setting up an AMA on Reddit, something I might do if I can figure out how—I have done two AMA's in the past, but both were arranged by other people. 

This post is an alternative approach. If lots of people are (briefly) looking at me, the obvious thing to do is to point them at things of mine that I would like them to read and that they might enjoy reading:

My web page, with lots and lots of stuff on it, including links to the full text of several of my books and most of my articles.

The third edition of my first book, The Machinery of Freedom (Kindle. The hardcopy will be available shortly on Amazon).

Harald, my first novel. EBook version. Free podcasts of it, read by me.

Salamander, my second novel. (Kindle. The hardcopy of that will also be available shortly on Amazon)

And for any who share my interests in recreational medievalism, two books coauthored with my wife:

A Miscellany. Free pdf. Hardcopy.

How to Milk an Almond, Stuff an Egg, and Armor a Turnip: A Thousand Years of Recipes. Free pdf. Hardcopy.


Friday, February 13, 2015

Wanted: A Better Shower Controller

Taking a shower this morning I was struck, not for the first time, by how badly designed the mechanism for controlling the temperature is. Turn it a little to the right and the shower is uncomfortably hot. Turn it just a little back to the left and it is uncomfortably cold.

What is going on is pretty clear. The controller maps its position to the amount of hot water in the mix in a roughly linear fashion. All the way to the left is straight cold, all the way to the right is straight hot, any intermediate position is a proportional mix.

In practice, almost nobody wants a cold shower or, unless the temperature of the hot water is pretty low, a straight hot shower. What almost everyone wants is a mix within a fairly limited range—say from .6 hot to .8 hot—with the exact range varying both with the temperature of the hot and the cold water and the preferences of the person taking the shower.

Imagine, to make life simple, that the controller has a handle that can be rotated through an arc of a hundred degrees. In the simplest version of my improved controller, the zero degree position is still pure cold, since there may be some masochists who like a cold shower to wake themselves up in the morning. The hundred degree position is still pure hot, since the water heater might be malfunctioning and producing only lukewarm water. But the first ten degrees of rotation map into the range from 0 hot water to .6 hot water. The final ten degrees map into the range from .8 to pure hot. The other eighty degrees of rotation cover the range from .6 to .8, giving me much better control over the temperature in the range I care about. A slightly improved model, designed to take account of variations in the temperature of the hot and cold water in different houses and hotels, has an adjustment that shifts exactly where the central range is, used once and then left alone unless something changes those temperatures. 

For the Silicon Valley market we have the intelligent version, which keeps track of what temperature the user actually takes a shower at and adjusts its central range accordingly. Its high end variant allows for up to four users with different tastes in shower temperature. Step into the shower, tell it which you are, and it automatically chooses a mapping that makes it easy to control the temperature over the range you care about.

For the next generation, we have the thermostatic version, which maps the setting of the controller not to the amount of hot and cold in the mix but to its temperature. Turn on the shower and for the first few seconds nothing is coming out, because the "hot water" isn't—it has cooled in the pipe between the water heater and the shower. When the temperature of the hot water reaches the level set on the controller, the shower turns on. Thereafter it adjusts the mix to give the temperature it is set for. As in the previous versions, the position of the controller maps to the temperature of the water in a non-linear fashion, designed to use most of its range to cover the range of temperature that the user cares about.

Do any of these already exist?