Thursday, November 29, 2007

Market Failure in Christmas

A model:

Denote the degree to which advertisers push christmas by 'Push'. This includes seasonal music playing in malls, Santa set up in said malls well before christmas, advertisements mentioning the number of shopping days remaining, etc, etc, etc.

Partition the set of all consumers into two groups. Group 'Elf' includes those whose utility is an increasing function of Push. Group 'Humbug' includes those whose utility is a decreasing function of Push. (Full disclosure: I fall into the latter category.)

Claim: Elves spend more money on christmas than do Humbugs. This seems a reasonable assumption; and is stated without proof. (If there is survey data on this, I would love to see it.)

Assume advertisers set Push in order to maximize profits, which we will take as analagous to sales. Consumers react to this advertising as we would expect, given their utility functions. Elves purchase more holiday merchandise, Humbugs purchase less. However, since Elves spend more, is it not reasonable to suspect that an increase in Push would, in absolute terms, elicit more spending from Elves than it would dissuade from Humbugs? This is true, for example, if each reacts equally to Push in terms of percentages. I further suspect that this could true because below some minimum standard, it's very difficult to decrease one's holiday spending; e.g. your significant other still expects presents, even if you can't stand venturing into the Push-filled malls.

But, just because spending contains an asymmetry does not imply that utility does. The amount of Push will be set higher than optimal because those financing the Push are responding disproportionately to the desires of Elves because Elves control a disproportionate amount of spending in the market place.

For example, assume we have 5 Elves spending $1000 on Christmas and 5 Humbugs spending $500 on Christmas. Each has utility of 100. Normalize Push to 1. Assume Push rises by 10%. The Elves increase their spending 10%, to $1100, and the Humbugs decrease their spending to $450. But the utility of the Elves rises, say, 5 utils to 105, and the utility of the Humbugs falls 5 utils to 95. In this manner, there is an incentive to continue increasing Push until the cost of the increase is equal to the spending induced.

In this manner, retailers will not only increase inequality among society, but expend productive resources on increasing Push, which does not improve the utility of society. Note that even if the Elves outnumber the Humbugs - irrespective of by how much - there still exists a point at which more Push has no net effect on the utility of society, but at which advertisers will increase it, given certain indifference curves, or given pretty much any indifference curve, if you believe my argument about it being very hard to substitute away from christmas spending below a certain point.

NOTE: This argument is severely hamstrung by the lack of equation ability inherent to blogger. I'll have to LaTeX it formally over the holidays.

UPDATE: The more I look at this the more I feel that utility should be removed from the equation, and we should simply assign monetary benefits to the gains/losses from exposure to Push, which prevents one from wading into the which-welfare-function-is-right debacle. I'm fairly utilitarian, but ceteris paribus, I prefer less inequality to more - to a certain point.

Advertising Experiment

For those who have been following this blog for some time, you will be aware of the purpose of the ads placed on the right column: Solely to examine the business structure of Google AdSense, since there are roughly eleventy billion of said ads on the Internet.

Total revenue recently jumped (since June, I haven't yet earned enough to purchase a medium coffee at Tim Hortons, so 'jumped' is the wrong word), but the most recent click produced 25% more than prior clicks (which were closely spaced).

I wonder if the AdSense algorithm is sensitive enough to lower payments when it suspects click fraud? Or is advertising payout an increasing function of site traffic? Maybe AdSense rates have just increased.

Wednesday, November 28, 2007

Difficulties Not Faced By Statistics Canada

Zimbabwe's chief statistician has said it is impossible to work out the country's latest inflation rate because of the lack of goods in shops.

Article here, HT to MR.

Sadly, this story still has a long way to run before it gets better. Inflation, to the best of my knowledge, never kicked anyone out of office.

Tuesday, November 27, 2007


For those wondering where I've taken off to, try figuring the cost-benefit analysis between blogging and writing exams. It is indeed that time of year.

The one thing I would like to mention is this article (paywall) from last saturday's Globe describing how the booming Polish economy is providing all sorts of jobs to East Germans.

I'm generally pessimisstic about development, but not for the reason posited: Post-reunification, the West Germans sent over to fix the other half of the country were the ones who couldn't manage to get a decent job in the West. Ergo, they were stuck with the East. With the scrapings from the bottom of the barrel running things in the East, no wonder it didn't get fixed.

Anyway, that's a theory that I thought was fairly interesting. Might even have some merit.

Thursday, November 22, 2007

Quebec Gas Legislation

Clearly, the wrong sort of economists are inside the Quebec government:
The Quebec government has moved to force oil companies to justify gasoline and diesel fuel price increases, causing the industry to fume that it could be forced to file thousands of reports a week.

Now, it's not government's role to dictate what prices an entrepreneur or firm wants to charge. This, sadly, is not the irritating part. I apologize for the libertarian bent. Let me zoom in on what Quebec's Minister for Natural Resources is saying:
"We want to be able to identify if it's always the same company or the same service stations that trigger the price increases," Mr. Béchard said. "If on a Friday afternoon there is a price increase that is more or less justified we will be able to go and see where the movement started from."
"At the request of the energy board, if, for whatever reason, there are price hikes that are unjustified in a certain region, the minister can directly by decree determine the maximum sales price," Mr. Béchard explained.

Of course, the fact that such things as 'unjustified prices' do not exist has not stopped this populist. Further, does anyone think that their is widescale collusion among service stations? If there is, that's a job for the Competition Bureau, not price controls. Finally;
"At the request of the energy board, if, for whatever reason, there are price hikes that are unjustified in a certain region, the minister can directly by decree determine the maximum sales price," Mr. Béchard explained.

Notwithstanding the fact, of course, that their are very excellent reasons for disparity in gas prices between different parts of the province. Let's hope this legislation dies in short order.

POSTSCRIPT: This is obviously, to some degree, the pot calling the kettle black because Newfoundland has a regulatory board for gas prices - though at least that board contains a number of gasoline representatives, and is generally not a very binding constraint, from what I hear.

Wednesday, November 21, 2007

10% Target

A simple back-of-the-envelope calculation in health economics:

Assume you live eighty years (365.25 days/year), working 35 of them for 50 weeks a year, 40 hours a week. This means that almost precisely 10% of your life will be spent working.

During Christmas, I'd like to hunt down data to see if this is a realistic calculation for Canada. I'd then like to compute the numbers for other countries, and, more interestingly, historically.

I'd bet that a few centuries ago, people dreamt of 10%.

Genius Level!

This blog has been certified genius level! I'll have to keep using big superfluously extravagant words. Either that, or I need to quit while I'm ahead.

cash advance

HT: Worthwhile Canadian Initiative

Tuesday, November 20, 2007


Following this morning's CPI data, I guess there's a little room for the central bank to lower rates. Were they so inclined.

What's bothering me though is that I'm seeing a lot of press constructing an argument something like this: 'Well, the core was below 2%, so that gives the Bank of Canada room to open the throttle.' While the target is 2%, that's not the floor - we try for between 1 and 3%. I'd like to take this chance to revise that target down to between 1 and 2%, but I am well aware it won't happen.

Those interested can also see the effects of the dollar. Fresh produce was down 14.6% y/y, and somewhere in the vicinity of 5% in October alone - note that in Newfoundland we're not seeing any of this decrease, according to relatives that do more grocery shopping than I. Women's clothing was down 2.3% in October. Categories that predominately consist of imports.

Peculiarly enough, the release also shows the booming of the St. John's housing market. Property taxes were up 8.8% annually, on the basis of lower rates and much higher assessments. At least I'd bet it's the capital; I certainly can't think of any reason why people would suddenly bidding up the market in Heart's Content/Delight/Desire/End.

Thursday, November 15, 2007

Export Sophistication

The Bank of Canada('s automated mailing list) was kind enough to (im)personally send me a powerpoint presentation that accompanied Paul Jenkins' speech yesterday, which had some bland title.

However, if you care sufficiently to skip to p.16-17, you get to see a pair of fairly interesting graphs comparing the changes in export sophistication between China and Canada from 1985-2001. Now, while I have no idea of how the numbers were arrived at, and am a big believer in regarding skeptically any statistics claiming to accurately depict China as a whole, it's fairly interesting.

Basically, it illustrates that China's exports are rapidly converging with Canada's in terms of expertise required. Which is contrary to some speculation that China's just getting bigger, not smarter, and probably bad news if you're exporting from Canada.

Of course, China won't grow at 10% forever. I'm fairly sure rising inflation is a classic sign of GDP starting to approach potential.

Wednesday, November 14, 2007

Free the Data

Dani Rodrik alerts me to a website ambitious enough to attempt to catalogue the CO2 emissions of every power station in the world. They're a far ways away yet, but the data looks pretty good for the developing world.

Of course neither St. John's nor Holyrood, where our local power station is technically located, made the cut. If any other person wants to go digging through the data, feel free.

This brings me to the main point of this entry. Dear Statistics Canada, please free the data.

Right now, even as a student at a university affliated with the CANSIM database, retrieving information from statscan is roughly on par with undergoing a root canal. Or something equally painful; I've never had a cavity. But I digress.

I tried poking through some budget records to see how much money our national statistics agency takes in from selling information. My bet is not a ton, and there's nothing that aggravates me more than having a question pop into my mind and running into a paywall on their site, detouring through CANSIM, getting frustrated at CANSIM, etc, etc. The end result is often giving up. I shudder to think of how many other students we have in our universities not seeking truth because we can't get the data in an easy manner.

So, Mr. Harper. I know it's early. But I'd love a Christmas present along these lines.

Saturday, November 10, 2007

Saturdays at MUN

Don't let anyone tell you the economics department takes the weekend off here. Three (of 9) profs are in, the visiting prof is in, all the (2) grad students are in, and all the senior undergrads who aren't internationals (1, me) are in.

Friday, November 9, 2007

Doing Business

Canada ranks seventh in the world for ease of doing business, according to the World Bank's 'Doing Business 2008' study, recently released.

We're particularly poor at enforcing contracts (43rd), which meshes with what I've picked up in the patent/growth literature on Canada's relatively poor protection of intellectual property.

Trading across borders is another stumbling block (39th), which I suspect might have been even worse if they looked at the interprovincial barriers to trade. For example, Newfoundland is currently planning to spend unnecessary billions building an undersea transmission line to the US for hydro electric power because Quebec is a roadblock east.

I have yet to hear any reasonable justification for any interprovincial barriers to trade. Maybe we'll see more movement towards TILMA and related initiatives. I hope.

Thursday, November 8, 2007

Central Bank Independence

There needs to be an economic manual for politicians. One of the things in it would be "Don't tell central bankers how to do their job." Listen up, Mr. McGuinty.
Meanwhile, Ontario Premier Dalton McGuinty said Thursday Prime Minister Stephen Harper "listened intently" during a private, 45-minute meeting they held at a Toronto hotel. McGuinty said he made a case for lower interest rates to curb the soaring Canadian dollar.

Article here. I don't even think a cut in rates would do much to depreciate the dollar, but note "cuts", the plural, so I'm hesitant to say McGuinty's supposed plan would have little impact, because I don't know how deep he'd like to cut.

Feel free to suggest other additions to this volume. Maybe we can get it printed and mailed.

The Intersection Game

Okay, if you're not mathematically inclined, you might want to bail now.

Let's play a game. Take some closed bounded interval of the real line. Partition this interval between two players. The first player chooses a subset of positive length of the original interval, the second player chooses another subset of positive length of this new interval, and this continues. One wins when a subset is reached which is entirely contained within their interval.

An example. Take [0,1]. We partition it such that I have [0,.5] and you have (.5,1]. Obviously, whoever goes first will win: I'll choose my subset of [0,1] as [0,.3], and if you went first, you might choose [.8,.9]. In either case, this ends the game right away.

Another. Suppose I have {0, 1/2, 1/3, 1/4, 1/5,...} and you have the rest of [0,1]. You'll always win this game, regardless of who goes first.

Suppose you go second. What sort of partition of the interval would you demand to have a chance of winning the game? Hint: A dense set won't always do it, e.g. if you're stuck with the rationals versus the irrationals, you'll always lose.

Interestingly enough, if you choose a random interval, your probability of winning is then 1. A question which I do not have the answer to is when given any interval, how much would you play the game from the first-mover position? The second-mover position?

POSTSCRIPT: My distribution of marks in this course is bizarre. An A+, 3 A's, 3 B's, 1 C, and 3 F's.

POST-POSTSCRIPT: If you want to cheat/learn, see here.

Wednesday, November 7, 2007

Exchange Rate Volatility

For some time, price stability has been a major concern of many central banks, Canada obviously included. We have accepted that price volatility entails a significant real cost. Yet, we have also made allowances that relative prices need to adjust.

So, where do we draw the line between curtailing the perhaps not as significant, but definitely real and definitely growing cost of volatility in exchange rates? Now, I'm not advocating bailouts for dying paper mills, but since there is an upper bound for what can be tolerated, there must be some optimal value at which the tradeoff between the costs of volatility and the costs of restraining price adjustment is balanced. I do not know where this is, but I suspect it is not a corner solution.

I mean, the dollar moved four cents yesterday. FOUR. 1.085->1.10->1.075. (Admittedly, the Chinese don't grow bored of mahjongg every night and decide to buy some Canadian dollars.) How many economists of yesteryear would have believed the sheer magnitudes of the currency moves we've seen recently? Probably not a ton. I'm fairly sure someone said we needed more superlatives to accurately describe the situation. I'm inclined to agree.

I mean, Bretton-Woods isn't the answer. But I suspect something will give before we see ten-cent swings in the value of the dollar in a 24-hour period. Maybe that something is a central role for stabilizing currency.

Tuesday, November 6, 2007

The Economist on Newfoundland

I'm a bit late with this (missed it myself), but a roaming correspondent for 'The Economist' recently spent a week (October 8-12) in Newfoundland, producing an eminently readable journal.

Sunday, November 4, 2007

Application Update

If nothing else, the change in the value of the Canadian dollar since I've started my graduate applications process will have saved me almost $90 if I pay all the fees today, compared to in September.

The writing of 'statements of purpose' is almost complete, and most all my references are in, with the exception of one prof who has yet to submit any. Transcripts are all mailed, though another round will have to go out after final exam marks are in.

SSHRC is also out of my hands, and in those of the department head. In sum, the process is going fairly smoothly.

Saturday, November 3, 2007

Moral Hazard and the NHL

For those of you who don't reside in this frozen land, the NHL refers to the 'National Hockey League'. That is ice hockey, not field hockey.

Anyway, a regular concern of mine is whether the powers that be have economists on staff. In any situation where one needs to set rules to prevent injury - financial or otherwise - we have a role for economics.

A friend of mine suggested today that a fund be created to support players who are severely injured while playing hockey. I felt compelled to point out that this might relieve some of the guilt (i.e. lower the cost) of causing an injury in the first place. Of course, I have no idea on whether the net benefits here are positive. They may well be.

In fact, this logic applies to any initiative that reduces injuries without increasing the punishment levied to the offender, including safety improvements to the arena. Beware of unintended consequences in all things.

NOTE: Obviously this post was motivated in part by the hit on Patrice Bergeron. As much as I enjoy a good hockey game, I have real troubles watching it at times.

Friday, November 2, 2007

Jobs Report

The last series of Labour Force Surveys have been strong, (see here for today's release) I remain leery of just how good the news purportedly is.

Specifically, because of the runup in employment over the last month, 60% of it is government jobs. In my mind, these don't stand on equal footing with jobs created in the private market. Government hiring is, to a certain degree, exogenous from the state of the economy. When employment growth relies to that extent on government hiring, it rings hollow.

Thursday, November 1, 2007

Reasonable Knowledge

I have substantially lowered my expectations for the level of knowledge of an average economic agent after finding out today that Ontario does not have harmonized sales tax.

Which would be excusable if I were not an economist who lived there for twenty weeks this summer.

Here in Newfoundland, we have the HST, so the default assumption is that other places are on board with this eminently logical idea.