Dead
The blog is officially dead. A new blog will be coming soon: Necessary But Not Sufficient.
PERMALINK Posted at 10:32 AM
Wednesday, August 15, 2007
Friday, April 13, 2007
Rapper Names
Rappers have made up names. This should come as no surprise; you can't have really thought that it said LL Cool J on the guy's birth certificate. Still, some people never consider quite what makes a great rapper name. I've identified a few themes:
For more information on rapper names, see Cracked's 25 Worst Rapper Names of All Time.
- Incorporating your real name or initials into your rapper name. eg. k-os (Kevin Brereton), Eminem (Marshall Mathers), Ludacris (Christopher Bridges)
- For white rappers, the use of imagery which indicates whiteness, eg. Vanilla Ice, Snow.
- Intentional misspelling of words and phrases, eg. Ludacris, k-os, Eminem, Mystikal.
For more information on rapper names, see Cracked's 25 Worst Rapper Names of All Time.
PERMALINK Posted at 12:39 PM
Tuesday, April 10, 2007
Strange but True
You know what the main problem with Cadbury's Creme Eggs is? They look too much like real eggs. The yolky appearance of their gooey, delicious insides turned me off for years as a child. But apparently, I'm not the only one who was fooled, as demonstrated by the cockatoo that adopted a clutch of the chocolate eggs as her own.
What happens when you take one of the world's best classical musicians, put a $4 million Stradivarius in his hands, and tell him to play some of the greatest classical pieces ever written, for change in a subway station? Hardly anyone notices that he isn't just your ordinary street musician. It really kind of makes you wonder about a lot of things. My favorite part, though, was the musician's reaction to pulling in $32.17 in 43 minutes (keeping in mind that his talents typically cost over $1000 a minute): "Actually, that's not so bad, considering. That's 40 bucks an hour. I could make an okay living doing this, and I wouldn't have to pay an agent."
And on a more personal note, I just found out that my best friend from high school, who mysteriously dropped off the radar about seven months ago, decided quite spontaneously to move to South Korea to teach English, without telling hardly anyone. Kind of a surprise, but very cool.
Labels: randomness
PERMALINK Posted at 1:00 PM
Tuesday, March 06, 2007
Facts
- I am happy to report that I've been accepted to law school, and I'll be a student of (probably) Queen's University Faculty of Law starting in September. Oh, did I mention I'm giving up engineering to become a lawyer? No? Well now you know.
- Today I have my Automatic Speech Recognition midterm.
- The weather today in Montreal: Overcast, blowing snow, -23℃, windchill -38 (that's -9℉ and -36 to my American readers).
That said, I'm at the library right now studying, but longing for the day that I'm finished my master's.
PERMALINK Posted at 9:47 AM
Wednesday, February 28, 2007
Sunday, February 25, 2007
Rail-ing against bus transit
Public transit, especially in North America, is dominated by buses. Whereas once, streetcar systems were commonplace (even in small communities such as my home town of Kingston, Ontario, which has a streetcar system from 1877 to 1930), they are now virtually nonexistent (Toronto being one notable exception). Heavier rail (i.e. rapid transit) systems are limited to very large centres, and even then, express buses are often used instead (such as in Ottawa). Whether or not you believe the conspiracy theories about bustitution, there are a few aspects of urban rail that I'd like to discuss. For clarity, I will refer to streetcars, light rail systems, etc. collectively as urban light rail (ULR).
The most often considered advantages of ULR have to do with how they function mechanically. ULR is typically electrified. This means it is zero emissions (not counting emissions used to generate the electrical power, but this can typically be done at substantially greater efficiency than in a diesel engine--and from a more diverse range of possible sources). Further efficiency gain comes from running on rails, since the rolling resistance is substantially less than that of bus wheels on asphalt. Both of these factors typically also make ULR quieter than the equivalent bus system.
Buses have a couple of substantial advantages though. The biggest is that the capital costs are a lot less. Buses don't require rails, of course, and buses are also a lot cheaper to buy than rail cars, due mostly if not entirely to economies of scale. Also, buses are a lot more flexible than ULR, since they don't need rails, making it much easier to alter routes and such.
That's the story everyone knows. If it just came down to these factors, I'd probably side with the status quo and say that buses are the best way to go, despite my personal preference for rail. However, I've recently thought of a couple of other advantages that ULR has over buses.
My first insight came when I was in Toronto almost a year ago. It goes something like this: if you get on a subway train or a streetcar, you know exactly where it's going. If you get on a bus, it might be going anywhere if you're not careful. Rail vehicles are very intuitive; they travel along the rails. Even in a hypothetical complex system in which multiple rail routes use the same track, there are still a limited number of places you might end up. With buses, if you don't know all the routes, you could literally end up anywhere! Thus, ULR has a certain appeal--especially to inexperienced users--that buses can't replicate.
My second insight took the same idea and extended it to a much more relevant level. A few days ago I was on the metro here in Montreal, talking about subway systems, and we got on to the phenomenon whereby dense development springs up along subway lines in order to take advantage of the convenient transport. As a result, after a sufficient period of time, demographics change in such a way that the routes actually serve the population in a nearly optimal fashion, even if they were poorly laid out initially. Thinking about Kingston (which has now gone the better part of a century with only bus service, and a poor one at that), I realized that this phenomenon doesn't occur with buses. The reason is obvious: because there is so little fixed physical capital, bus routes are easily changed to accommodate usage trends, rather than staying fixed and dictating those trends as rail systems do. Since the routes might change at any time, people and businesses avoid making long-term arrangements with those routes in mind, and as a result, development cannot go forward in a transit-oriented way. (By analogy, would you build a house on a street if you thought the city could tear up the street any time they wanted?)
So my conclusion? If we want to do away with auto-centered development in favour of transit-centered (which we should), replacing buses with more permanent systems is something we should give a bit more consideration.
P.S.- Before any smarty-pants economists get on my case about not accounting for discounting, don't. I know what you're going to say and I disagree. Some day I'll write about how the whole concept of discounting has been brutally abused by economics, and how it's been used to justify policies that are ruining the world. Er... I mean... comments are welcomed.
The most often considered advantages of ULR have to do with how they function mechanically. ULR is typically electrified. This means it is zero emissions (not counting emissions used to generate the electrical power, but this can typically be done at substantially greater efficiency than in a diesel engine--and from a more diverse range of possible sources). Further efficiency gain comes from running on rails, since the rolling resistance is substantially less than that of bus wheels on asphalt. Both of these factors typically also make ULR quieter than the equivalent bus system.
Buses have a couple of substantial advantages though. The biggest is that the capital costs are a lot less. Buses don't require rails, of course, and buses are also a lot cheaper to buy than rail cars, due mostly if not entirely to economies of scale. Also, buses are a lot more flexible than ULR, since they don't need rails, making it much easier to alter routes and such.
That's the story everyone knows. If it just came down to these factors, I'd probably side with the status quo and say that buses are the best way to go, despite my personal preference for rail. However, I've recently thought of a couple of other advantages that ULR has over buses.
My first insight came when I was in Toronto almost a year ago. It goes something like this: if you get on a subway train or a streetcar, you know exactly where it's going. If you get on a bus, it might be going anywhere if you're not careful. Rail vehicles are very intuitive; they travel along the rails. Even in a hypothetical complex system in which multiple rail routes use the same track, there are still a limited number of places you might end up. With buses, if you don't know all the routes, you could literally end up anywhere! Thus, ULR has a certain appeal--especially to inexperienced users--that buses can't replicate.
My second insight took the same idea and extended it to a much more relevant level. A few days ago I was on the metro here in Montreal, talking about subway systems, and we got on to the phenomenon whereby dense development springs up along subway lines in order to take advantage of the convenient transport. As a result, after a sufficient period of time, demographics change in such a way that the routes actually serve the population in a nearly optimal fashion, even if they were poorly laid out initially. Thinking about Kingston (which has now gone the better part of a century with only bus service, and a poor one at that), I realized that this phenomenon doesn't occur with buses. The reason is obvious: because there is so little fixed physical capital, bus routes are easily changed to accommodate usage trends, rather than staying fixed and dictating those trends as rail systems do. Since the routes might change at any time, people and businesses avoid making long-term arrangements with those routes in mind, and as a result, development cannot go forward in a transit-oriented way. (By analogy, would you build a house on a street if you thought the city could tear up the street any time they wanted?)
So my conclusion? If we want to do away with auto-centered development in favour of transit-centered (which we should), replacing buses with more permanent systems is something we should give a bit more consideration.
P.S.- Before any smarty-pants economists get on my case about not accounting for discounting, don't. I know what you're going to say and I disagree. Some day I'll write about how the whole concept of discounting has been brutally abused by economics, and how it's been used to justify policies that are ruining the world. Er... I mean... comments are welcomed.
Labels: environment, transportation
PERMALINK Posted at 7:00 PM
Saturday, February 10, 2007
The Cost of Education
I know the blogging has been slow recently, in my attempt to get my Masters done by the spring, but in honour of last Wednesday's Student Day of Action, I thought I'd say a few words about the cost of postsecondary education.
The primary purpose of the Student Day of Action is to protest rising tuition fees, and to urge governments to reduce or freeze tuition. I'm tempted to write it off simply as rent seeking (and certainly, to some of the protesters, it is), but to be fair, there's a little more to it than that. I'd like to examine some of the issues.
First, let's consider who benefits from university education. Obviously, the individual receiving the education benefits, primarily from increased status and future income. The status part is sort of zero-sum though, so I'll focus on direct economic benefits. Society as a whole also benefits from education. A more educated population creates more economic opportunities for everyone. In both these respects, extending education to more and more people has diminishing returns: very smart, creative, ambitious people benefit immensely, and create immense benefits for others, from education, while at some point the cost of educating less academically oriented people clearly begins to outweigh the total benefits.
The funny thing is, the people who protest high tuition are students. If the tuition is too high, why are you paying it? Shouldn't the people protesting be the people who choose not to pay it? So clearly tuition isn't too high for the people who choose to pay it. The question remains though: are tuition fees so high that we are producing a suboptimal number of graduates? (Or conversely: are they so low we are producing too many graduates?)
I've also heard the argument that university education makes people better citizens, and that this is another source of external, public good that comes from education. I think this is very hard to prove. It's more likely that people who are good citizens tend to go to university, but they would be good citizens even if they didn't. I don't even think this holds for many fields: most engineering students I know don't much know or care about politics or their communities. I think it may be true that social sciences do encourage good citizenship (through understanding of society), but I'm not sure about "liberal arts" either. I think widespread liberal arts education is probably a massive waste of resources, in which most of the benefits accrue to the student, who obtains no economically useful skills or knowledge but some degree of "status" and a great deal of satisfaction at thinking him- or herself more cultured than holders of science degrees.
Ideally, we want the tax-paying public and the student to pay in proportion to the benefits they receive at the economically efficient level. How do we calculate this? Well, it's tricky. Any estimate would have to be rough, but we could probably come up with something which is reasonable within a factor of two. That's not very good though. It's a hard problem. We can say for certain, though, that the correct balance is definitely not one in which the public bears the full cost.
The difference between finding the optimal price for a university degree and finding the optimal price for some other product, such as a litre of gasoline, is that because the price is so high (tens of thousands of dollars directly, maybe hundreds of thousands when opportunity cost is taken into account), the some of the standard assumptions about economic decision making don't apply. (Incidentally, last semester for my Behavioural Economics class I wrote a paper on the effects of price scales on risk preferences, so I can happily say I know what I'm talking about). If you really can't come up with the money, you won't be able to pay your tuition, even if you know it is a good investment in the long-term.
Fortunately, we have a partial solution to that: student loans. The student loan system is good, because it means people can afford university even when they can't afford university. Sure, it sucks to be starting your adult life with thirty grand in student debt, but hey, you decided it was worth it. I say it's only a partial solution though because it doesn't account for one thing: risk. There's always a risk that your university degree isn't going to pay off in the way you thought, especially since a lot can happen to the economy in the four years you're in university. On average, this should produce too many people with "safe" degrees, because potential students have to compensate for the risk.
Ideally, we'd like to balance this risk by shifting it to the something like the government, which can afford to be fairly risk neutral due to scale. One possible solution to this problem that I considered is, rather than making people pay tuition in advance, have them agree to pay an increased income tax once they graduate. More thinking leads me to believe this is a dumb solution because (1) all the deadweight loss of income tax would still apply, (2) there are serious enforcement issues for people who move out of the jurisdiction, and worst of all, (3) it encourages people to pursue degrees with little economic value, since they would only pay in proportion to what they earn, not to the cost of their education.
So I have to conclude that the present system is actually pretty good--or, at least, it's hard to come up with a much better system. A few moderate reforms would be good (such as not tying student loans to parents' income--I know a lot of people who have been seriously put off by that one), but overall, I don't think there's much to complain about.
Disclaimer: I was one of those lucky people whose education was entirely paid for by his forward-thinking parents. So far, my formal university education has been a massive misallocation of resources.
The primary purpose of the Student Day of Action is to protest rising tuition fees, and to urge governments to reduce or freeze tuition. I'm tempted to write it off simply as rent seeking (and certainly, to some of the protesters, it is), but to be fair, there's a little more to it than that. I'd like to examine some of the issues.
First, let's consider who benefits from university education. Obviously, the individual receiving the education benefits, primarily from increased status and future income. The status part is sort of zero-sum though, so I'll focus on direct economic benefits. Society as a whole also benefits from education. A more educated population creates more economic opportunities for everyone. In both these respects, extending education to more and more people has diminishing returns: very smart, creative, ambitious people benefit immensely, and create immense benefits for others, from education, while at some point the cost of educating less academically oriented people clearly begins to outweigh the total benefits.
The funny thing is, the people who protest high tuition are students. If the tuition is too high, why are you paying it? Shouldn't the people protesting be the people who choose not to pay it? So clearly tuition isn't too high for the people who choose to pay it. The question remains though: are tuition fees so high that we are producing a suboptimal number of graduates? (Or conversely: are they so low we are producing too many graduates?)
I've also heard the argument that university education makes people better citizens, and that this is another source of external, public good that comes from education. I think this is very hard to prove. It's more likely that people who are good citizens tend to go to university, but they would be good citizens even if they didn't. I don't even think this holds for many fields: most engineering students I know don't much know or care about politics or their communities. I think it may be true that social sciences do encourage good citizenship (through understanding of society), but I'm not sure about "liberal arts" either. I think widespread liberal arts education is probably a massive waste of resources, in which most of the benefits accrue to the student, who obtains no economically useful skills or knowledge but some degree of "status" and a great deal of satisfaction at thinking him- or herself more cultured than holders of science degrees.
Ideally, we want the tax-paying public and the student to pay in proportion to the benefits they receive at the economically efficient level. How do we calculate this? Well, it's tricky. Any estimate would have to be rough, but we could probably come up with something which is reasonable within a factor of two. That's not very good though. It's a hard problem. We can say for certain, though, that the correct balance is definitely not one in which the public bears the full cost.
The difference between finding the optimal price for a university degree and finding the optimal price for some other product, such as a litre of gasoline, is that because the price is so high (tens of thousands of dollars directly, maybe hundreds of thousands when opportunity cost is taken into account), the some of the standard assumptions about economic decision making don't apply. (Incidentally, last semester for my Behavioural Economics class I wrote a paper on the effects of price scales on risk preferences, so I can happily say I know what I'm talking about). If you really can't come up with the money, you won't be able to pay your tuition, even if you know it is a good investment in the long-term.
Fortunately, we have a partial solution to that: student loans. The student loan system is good, because it means people can afford university even when they can't afford university. Sure, it sucks to be starting your adult life with thirty grand in student debt, but hey, you decided it was worth it. I say it's only a partial solution though because it doesn't account for one thing: risk. There's always a risk that your university degree isn't going to pay off in the way you thought, especially since a lot can happen to the economy in the four years you're in university. On average, this should produce too many people with "safe" degrees, because potential students have to compensate for the risk.
Ideally, we'd like to balance this risk by shifting it to the something like the government, which can afford to be fairly risk neutral due to scale. One possible solution to this problem that I considered is, rather than making people pay tuition in advance, have them agree to pay an increased income tax once they graduate. More thinking leads me to believe this is a dumb solution because (1) all the deadweight loss of income tax would still apply, (2) there are serious enforcement issues for people who move out of the jurisdiction, and worst of all, (3) it encourages people to pursue degrees with little economic value, since they would only pay in proportion to what they earn, not to the cost of their education.
So I have to conclude that the present system is actually pretty good--or, at least, it's hard to come up with a much better system. A few moderate reforms would be good (such as not tying student loans to parents' income--I know a lot of people who have been seriously put off by that one), but overall, I don't think there's much to complain about.
Disclaimer: I was one of those lucky people whose education was entirely paid for by his forward-thinking parents. So far, my formal university education has been a massive misallocation of resources.
PERMALINK Posted at 4:34 PM
Monday, January 15, 2007
On the Subject of Beer Bottle Deposits and Sunday Wages
As you may know, I'm not a big fan of capitalism in general; but I do think that it can have a lot of good points, if it's done right. Unfortunately, it's not done right, due to both well-meaning but stupid market intervention on behalf of citizens, and due to downright selfish and hypocritical interventions on behalf of capitalists.
The inspiration for this post comes from something that happened to me yesterday. I finally got around to returning some empty beer bottles. After hauling my load of 36 bottles (worth $3.60 in deposits--much less than they should be worth if you ask me) two blocks to the grocery store, I find out that they don't accept bottle returns on Sundays. I was sort of pissed.
Two things bother me about this. One is that I had to haul the bottles two blocks to the nearest grocery store and not just half a block to the nearest dépanneur. The reason for this is that stores only buy back bottles of brands of beer that they sell. Why? It doesn't make a lot of sense, really; the bottles are all the same. What it must come down to is that the stores lose money on accepting returns, and therefore have an incentive to turn away any returnees that they can, legally. The incentives are all wrong. They need to re-engineer the system so that accepting bottle returns is profitable for vendors.
The other thing, of course, is that they don't accept bottles on Sunday. I suspect this is due to another stupid Québec law. Of course, even if it were legal to return bottles on Sunday, it would be even more costly than usual for the vendor to accept them, due to another stupid Québec law that restricts the number of employees on the floor on a Sunday.
If I were an employee of somewhere (which I'm not--I'm a freeloading student), I'd be asking myself: why is the government preventing me from working on a Sunday if I want to? And furthermore, if I can only find a job for which someone is willing to pay me $5 an hour (less than the minimum wage), why is the government preventing me from taking that job?
The answer, I think, is that the government means well. Lots of people don't want to work on Sundays, so by restricting the number of employees, we can ensure that not too many people are inconvenienced. The minimum wage prevents employees from paying unfairly low wages. These measures work, but not very well. Market economics would suggest, instead, that fair wages would be determined by what people are willing to work for (i.e. as long as employees choose to work for that wage, it's fair). If people prefer not to work on Sundays, then they will demand higher wages on Sundays, and the number of workers on the floor on a Sunday will reflect the economic optimum according to this preference.
That's the theory, and in theory it works. In practice, however, markets are not free. Employees are not paid the market rate for their labour; at best, their wage can be thought of as an aggregate, while at worst it can be thought of as having little relation to the market value of their labour, due to the severe restrictions placed on them. Employees in, say, supermarkets, can't just negotiate their wages from shift to shift unless they want to be fired.
Paradoxically, the minimum wage, an institution that is meant to protect workers, is actually one of the institutions that prevents workers from negotiating their own wages, and thereby shifts the balance of bargaining power to the management. It means that at times when fair market labour prices would be low (high supply and low demand), employers must pay at least the minimum wage, and thus when natural labour prices are high, paying the market price would lead them to pay an average of more than the market aggregate.
What if we ditched the minimum wage, and instead introduced legislation that prevented employers from requiring workers to work a given shift at a fixed wage? Employees could, in effect, "bid" on shifts, offering to work a given shift for a specified minimum wage, and then the employer would choose to hire the n employees with the lowest bids at the wage of the nth lowest bid. This would ensure that the economically optimal number of employees work a given shift (according to the employer's decision of how many to employ), and conversely that they get paid the economically optimal wage. Sure, there would be some employees with high standards who might not get many shifts, but they could take other jobs as well (there is a slight problem of combinatorial auctioning here, but it probably wouldn't be too huge).
The government has a lot of power to make markets work better, to make them more free. If only it did so more often.
The inspiration for this post comes from something that happened to me yesterday. I finally got around to returning some empty beer bottles. After hauling my load of 36 bottles (worth $3.60 in deposits--much less than they should be worth if you ask me) two blocks to the grocery store, I find out that they don't accept bottle returns on Sundays. I was sort of pissed.
Two things bother me about this. One is that I had to haul the bottles two blocks to the nearest grocery store and not just half a block to the nearest dépanneur. The reason for this is that stores only buy back bottles of brands of beer that they sell. Why? It doesn't make a lot of sense, really; the bottles are all the same. What it must come down to is that the stores lose money on accepting returns, and therefore have an incentive to turn away any returnees that they can, legally. The incentives are all wrong. They need to re-engineer the system so that accepting bottle returns is profitable for vendors.
The other thing, of course, is that they don't accept bottles on Sunday. I suspect this is due to another stupid Québec law. Of course, even if it were legal to return bottles on Sunday, it would be even more costly than usual for the vendor to accept them, due to another stupid Québec law that restricts the number of employees on the floor on a Sunday.
If I were an employee of somewhere (which I'm not--I'm a freeloading student), I'd be asking myself: why is the government preventing me from working on a Sunday if I want to? And furthermore, if I can only find a job for which someone is willing to pay me $5 an hour (less than the minimum wage), why is the government preventing me from taking that job?
The answer, I think, is that the government means well. Lots of people don't want to work on Sundays, so by restricting the number of employees, we can ensure that not too many people are inconvenienced. The minimum wage prevents employees from paying unfairly low wages. These measures work, but not very well. Market economics would suggest, instead, that fair wages would be determined by what people are willing to work for (i.e. as long as employees choose to work for that wage, it's fair). If people prefer not to work on Sundays, then they will demand higher wages on Sundays, and the number of workers on the floor on a Sunday will reflect the economic optimum according to this preference.
That's the theory, and in theory it works. In practice, however, markets are not free. Employees are not paid the market rate for their labour; at best, their wage can be thought of as an aggregate, while at worst it can be thought of as having little relation to the market value of their labour, due to the severe restrictions placed on them. Employees in, say, supermarkets, can't just negotiate their wages from shift to shift unless they want to be fired.
Paradoxically, the minimum wage, an institution that is meant to protect workers, is actually one of the institutions that prevents workers from negotiating their own wages, and thereby shifts the balance of bargaining power to the management. It means that at times when fair market labour prices would be low (high supply and low demand), employers must pay at least the minimum wage, and thus when natural labour prices are high, paying the market price would lead them to pay an average of more than the market aggregate.
What if we ditched the minimum wage, and instead introduced legislation that prevented employers from requiring workers to work a given shift at a fixed wage? Employees could, in effect, "bid" on shifts, offering to work a given shift for a specified minimum wage, and then the employer would choose to hire the n employees with the lowest bids at the wage of the nth lowest bid. This would ensure that the economically optimal number of employees work a given shift (according to the employer's decision of how many to employ), and conversely that they get paid the economically optimal wage. Sure, there would be some employees with high standards who might not get many shifts, but they could take other jobs as well (there is a slight problem of combinatorial auctioning here, but it probably wouldn't be too huge).
The government has a lot of power to make markets work better, to make them more free. If only it did so more often.
PERMALINK Posted at 2:03 PM
