Wednesday, July 09, 2014

Post-mortem

 I need to work through this the only way i know how: by analyzing it. A friend sent me about 5 commentaries from the Brazilian press but I think they missed some key points.
  1. The first goal followed the form of other goals that have knocked brazil out of the world cup. On a set play, the Brazilian defenders just inexcusably fail to mark the leading scorer of the opposite team (Zidane twice in 98, T. Henry in 2006, G. Muller in 2014).
  2. The next 4 goals require a completely different explanation. I think we have to see this as a collective psychological collapse. A team is suddenly unable to carry out basic operations. Germany was a hot knife and Brazil was warm butter.
  3. One can and should blame Scolari but I don't think you can really fault his team selections--if you think he should have kept to the formation that got Brazil to the semis. Scolari substituted like for like. Bernard dribbles and drives along the left and is not much use on defence (like Neymar). Dante plays the same position as Thiago Silva for a considerably better team (Bayern Munich).
  4. The German midfield of Khedira, Kroos, and Schweinsteiger is really one of the best ever and in hindsight it was a huge mistake to think that Luis Gustavo and Fernandinho would be up to the challenge of matching up with them. The problem is that we didn't have anyone on the bench who could have been predicted to do much better. Probably Fred should have been dropped and replaced by Paulinho so at least we would not have been outnumbered in that sector. But Scolari is far too set in his ways to make such a major tactical shift.

Sunday, June 01, 2014

The future of education is not online



An article in the NYT yesterday with the provocative title "Business Disrupted" outlined the dilemma faced by HBS where star profs like Porter and Christensen have sharply different views on the potential role of online education.

Well, who cares about those guys say, what does Keith Head think? Thanks for asking.

If our “business” is something like selling books or records, or DVD rentals, then the internet might  reasonably be seen as an existential threat. Ask Blockbusters or Borders (or Duthie books here in Vancouver) or any number of record (CD) store chains that don’t exist anymore.

But if our business is more like a yoga class or a music concert even selling lingerie (Victoria’s Secret has had a mail order catalog since 1978 and e-commerce since 1995 but  they still found it worthwhile to take over the huge retail space Virgin had occupied on Robson) then it seems we have less to fear.   In the case of yoga and live music, a core aspect of the business is the shared experience. Those attending get something from the instructor, sure, but they also get energy and other emotional connections from the rest of the audience. My Pilates instructor puts his classes on Youtube but I watched for about a minute before losing interest. And people have long had yoga DVDs but I have a feeling most of them are rarely used after purchase. Music concerts continue to charge hefty prices even though we can find many performances online.

If this analogy is correct, then only a certain type of business education is seriously threatened by the internet: non-interactive, 100+ seat lectures.  We need to be moving not in the direction of "recorded music" (lecture clips videoed and put online) but in the direction of live music, emphasizing a shared experience where you have to physically be there to get it.

One corollary is that just as cellphones need to be off in yoga class or violin concertos, laptops need to be closed in the classroom (related note recent research finds taking notes longhand enhances memory). As someone put it recently in a teaching panel, open laptops "suck the energy out of the classroom."

Wednesday, May 28, 2014

Excel is EVIL

In the news, more reasons why no one, especially not academic economists, should use Excel formulas as a way to calculate anything even marginally important.

Last year we had the Reinhart and Rogoff case of spreadsheet errors and now we have the Piketty case.

Chris Giles of FT writes

In the wealth inequality tables for the top 1 per cent and top 10 per cent in Capital in the 21st Century, there were 142 data points and I found problems with 114 of them. As a problem proportion, it comes in at a rather high 80 per cent.

At the end of the same blog posts he unfortunately has to add a mea culpa about his previous post:

Finally, I am grateful to @kenivers1 for spotting an error in the FT’s spreadsheet I posted online. The constructed European average for the top 10% wealth share in 1810 contained a formula error. The spreadsheet has now been amended. Our mistaken result was 83.4%, while the corrected one is 81.3%.

Making mistakes is easy. My Stata code almost never works the first time I try to run it.  But once it does run, there is code with (some) comments and hopefully it possible to figure out why I did what I did and whether it makes sense or not.   On the other hand excel buries the formulas. Even when you look at them they are written in terms of A1 and D5 (wasn't that supposed to be D1?).  In other words when do data work in Excel you are more or less ensuring that you will make mistakes that will be very hard to find unless someone is looking hard. And that's a bad thing.

Here's Justin Wolfers on how to avoid/identify mistakes.


Sunday, October 14, 2012

What's wrong with Bittman's proposed food label?

Mark Bittman, the NYT food columnist proposes a "dream food label" today. His argument for the label, suggests that only the food industry would oppose his proposal. But I several problems with it. The most important has to do with his example of frozen blueberries. Bittman gives them a score of 5/5 for nutrition but only 2/5 for "Welfare". What's so bad for welfare?  Answer: 
"Although the blueberries are organic, they’re sourced from Chile, where the workers are being paid a dollar a day and little attention is paid to soil quality: 2 for Welfare."  
Stop right there! Before you go any further... Workers paid a dollar per day in Chile? According to World Bank data in 2005 0.7% of Chileans were living on less than a 1.25 per day. And if you take the Chilean minimum wage of 182,000 pesos per month, divided by 30 and multiply by the exchange rate, that works out to $12.74 per day.

But suppose Chilean blueberry farmers do happen to paid as if they lived in Mali. Think about the conclusion Bittman wants you to draw. Because they are impoverished you should buy less from them.  I think Joan Robinson once put it simply: "The only thing worse than being exploited by a capitalist is not being exploited by a capitalist."  If for some strange reason, Chilean blueberry pickers have such lousy opportunities that they are willing to accept a dollar per day, why would it be a good thing to deprive them of what little they have?

Other things bother me on reflection. Why do we care about "foodness" and GMOs except for impacts of these characteristics on nutrition. If a frozen blueberry is just as nutritious as a fresh one (he gives them the full 5 nutrition points), why penalize for freezing? Fresh blueberries aren't available all year so we freeze. What's wrong with that?

Monday, August 29, 2011

Crazy Social Artistic Black Widows

Elizabeth Kolbert uses recent genetic studies of the Neanderthals as the frame for discussing what differentiates humans from our nearest relatives (Neanderthals and apes). The article is behind the New Yorker paywall. Here's my quick summary of her main points about our species' distinctive features.

  1. To paraphrase the tagline of the movie Black Widow, "we mate, then we kill." In the case of both the Neanderthals and Denisovans our ancestors definitely had sex. And it worked out well for us, as the BBC reports it improved our immune systems. It's not clear we actually killed them after having sex, but they disappeared not long after we arrived in the regions they had previously occupied. And we have a record of causing extinctions of other species that have the misfortune to try to share the same space with us.
  2. We painted our cave walls. Apparently the others did not.
  3. We crossed oceans. Partly because we were smart enough to develop the technology to do so. And partly because we were crazy enough to paddle off into the open sea.
  4. We're not especially impressive problem solvers. An adult Orangutan can solve problems that flummox human toddlers. But the human kids show greater ability to engage in social learning. Neanderthals may have been somewhat autistic.

Monday, July 25, 2011

"In Fact"

Don't you find it annoying when people write "in fact" as a way to assert something without providing actual evidence. One of the reasons people do this (here I can speak from experience) is because that asserted fact is not a fact at all, but merely an assertion we wish to be true.

Example.

In a recent blog entry, Krugman writes

Even people who are supposedly well informed believe that there was a vast expansion of government under Obama, when in fact there wasn’t. (emphasis added)

OK, here are the actual facts obtained from the Office of Management and Budget historical tables. In 2008, total government outlays were 2.98 trillion. The 2011 estimate is 3.83 trillion. As a percentage of GDP, federal government outlays rose from 20.7% to 25.1%.

So unless we want to redefine "expansion", "government" or "under Obama", we have to accept that there was in fact an
"expansion of government under Obama." But was it vast? To me, almost a trillion dollars (28.5% increase) over 5 years is on the big side of things, especially when considering that outlays had been at 21% or below as a share of GDP since 1994.

OK, that is the end of the "Republican" portion of this post. Now we have to point out that the increases in government outlays have come in large part from increases in military spending, pensions, and medicare (medical care for the elderly). Spending is not just foreign aid and "bridges to nowhere". In fact that sort thing is very small.

And while government has indeed expanded, it is equally evident that tax receipts have been shrinking as a share of GDP.

Sunday, July 17, 2011

Bees and Blueberries

While at the Kitsilano Farmer's Market, I saw a sign for honey made from the blueberries at G- farms. I asked the honeymaker if he has to pay for his bees to have access to the farm. He said actually they pay him for bringing his bees to their farm.

Why can't the farms rely on wild bees to do the pollination? Well in the Frasier valley, the only wild bees are bumble bees which hibernate in the winter in mouseholes. With the cool spring, the bumble bees were late in getting started. I asked if he could charge extra because of the added need for his bees. Unfortunately for him, prices for pollination services fell this summer due to the arrival of a new competitor from Northern Alberta (of all places!). Where the old established price was $95 (per field? acre? I didn't catch that), this guy went to a farm show and announced his willingness to accept $65.

The newcomer has 20,000 colonies in semis and decided to come to the Frasier valley to nourish his bees here before taking them to the Canola fields of the prairies for the summer. The farmer I spoke to still is able to charge over $65 to some local farms ($75 to G- farm) because of the established relationship which means the farmer can rely upon him to come each year. But the old $95 prices are unattainable.

I bought 250ml of blueberry honey for $4.95. Not only am I looking forward to enjoying it on French Toast, I may use it as a prop for my next lecture on externalities...since now I have my own anecdote to add to the "fable of the bees." (Journal of Law and Economics, 1973, "The Fable of the Bees: An Economic Investigation" by Steven Cheung)