Tuesday, November 27, 2012

Making sense of statistics

Morton Jerven has a fascinating post up at the Guardian, "Lies, damn lies and GDP", in which he discusses the striking unreliability of GDP figures for African countries. Best line:
The knowledge that currently there are 40 "Malawis" unaccounted for in the Nigerian economy should raise a few eyebrows.
The low quality of development statistics is well known to those in the field. Jerven's account reminded me of an apocryphal story about an aid official tasked with analyzing trends in economic growth in a particular aid recipient. After much work, he finally found a powerful pattern, noting that economic growth was almost exactly 1.25 times population growth. Proud of this insight, he traveled to the recipient country's economic ministry in order to share his new-found knowledge. As he was waiting at the desk of the official he had come to brief, however, he noticed a faded post-it note taped to the official's desk: "when unable to get growth data for the annual report, just use population growth times 1.25".

Jerven (whose book on the quality of these statistics, Poor Numbers, has just been published by Cornell University Press) correctly points out that
governments, international organisations and independent analysts need these development statistics to track and monitor efforts at improving living conditions on the African continent.
However, the problem is less serious for at least one particular use of these statistics. As I argue in my book, if we wish to study how a country's GDP factors into decision-making on the part of aid donors, international financial institutions, etc., what matters is not some "true" GDP, but rather what these international actors believe the GDP to be. For this purpose, then, the flawed statistics these actors had access to at the time they made their decisions are, in fact, preferable to retroactively "corrected" GDP estimates, regardless of how much better those new estimates reflect "true" GDP.

Monday, November 19, 2012

Let the pandering for 2016 begin...

Obama has only just been re-elected, and Florida senator Marco Rubio has already started pandering to the "my ignorance is just as good as your knowledge" crowd. Here he is in an interview in the Dec. 2012 issue of GQ:

GQ: How old do you think the Earth is?
Marco Rubio: I'm not a scientist, man. I can tell you what recorded history says, I can tell you what the Bible says, but I think that's a dispute amongst theologians and I think it has nothing to do with the gross domestic product or economic growth of the United States. I think the age of the universe has zero to do with how our economy is going to grow. I'm not a scientist. I don't think I'm qualified to answer a question like that. At the end of the day, I think there are multiple theories out there on how the universe was created and I think this is a country where people should have the opportunity to teach them all. I think parents should be able to teach their kids what their faith says, what science says. Whether the Earth was created in 7 days, or 7 actual eras, I'm not sure we'll ever be able to answer that. It's one of the great mysteries.

Of course, the reverend John Lightfoot already solved this mystery in a sermon in the early 1800s, determining that the earth was created in 3960 BC, or, to be even more precise: "That the world was made at equinox, all grant — but differ at which, whether about the eleventh of March, or twelfth of September; to me in September, without all doubt." (from Whole Works, 1822, vol. 7, p. 322).

Since Rubio thinks there are multiple theories (after all, bishop Ussher's more famous calculation has creation in 4004 BC), he is not, apparently, interested in any of the scientific knowledge accumulated in the past two centuries. Ignorance for president in 20165. Yeah!!

Sunday, October 28, 2012

Robert Samuelson embarrasses himself

Robert Samuelson thoroughly embarrasses himself in a Washington Post op-ed published on Thursday. His column attacks a New York Times editorial he claims is so wrong-headed as to move him to write a column "refuting" an editorial for the first time in 35 years. Unfortunately for him, the only thing the column refutes is the idea that he understands what he is talking about.

Others have already done a fine job tearing apart his argument, including Josh Bivens (at EPI), whose data the New York Times cited, as well as Dean Baker (at CEPR) and Paul Krugman. I just want to make one additional point that has not yet been made.

Samuelson's key claim is the following assertion: "What the Times omits is the money to support all these government jobs. It must come from somewhere — generally, taxes or loans (bonds, bills). But if the people whose money is taken via taxation or borrowing had kept the money, they would have spent most or all of it on something — and that spending would have boosted employment."

From this he concludes that government job creation must substitute "public-sector workers for private-sector workers," and hence does not count as real job creation. Note the faulty logic: if he is right, then the exact same argument can be made in the opposite direction — not taking money via taxation increases private jobs only by reducing government jobs. Depending on your starting point, then, the private sector doesn't create jobs either.

So where does the logic go wrong? Well, Samuelson's simplistic analysis fails to take into account 2 1/2 considerations. First, not all jobs cost the same amount of money, so if we're just moving money back and forth between the government and the private sector, job creation takes place in whichever sector jobs cost less. Second, jobs are not the only thing money is spent on in either sector, so job creation/destruction may depend on which sector spends a greater proportion of its money on jobs.

Third (and this is the 1/2 consideration above), "real" job creation happens, of course, not by shifting money back and forth between the public and private sectors, but rather when certain jobs and certain spending have a multiplier effect and generate further jobs. Samuelson knows this, but simply asserts  that only private sector jobs & spending can have this effect.

Of course, Samuelson does not offer even the tiniest scrap of evidence for this claim, and others who have responded to his column (especially Dean Baker) have already highlighted its flaws.

Wednesday, October 24, 2012

Grumpy word usage post: Pushing ... the envelope

Right now, the New York Times online front page has as its central headline "Pushing the envelope in attacks on Obama" (the article itself has a different title).

This popular phrase originated in the world of aviation, and more specifically the interaction between engineers and pilots. The "envelope" of a plane refers to the calculated extremes of a plane's performance capabilities.

The phrase was popularized most famously in Tom Wolfe's The Right Stuff ("pushing the outside of the envelope ... seemed to be the great challenge and satisfaction of flight test"). The point, obviously, is that an envelope has an outer surface, and that is what pilots were pushing against. Pilots were not trying to redesign the plane to give it a different envelope (i.e. they were not pushing the envelope) — that's what engineers did — but rather to push against the existing surface.

Wolfe's usage is correct, as is the more commonly seen "pushing the edge of the envelope." Incorrect, however, is "pushing the envelope", which is by far the most common form (in Google, 16 times more common than "pushing the edge").

Even in the aviation world, the phrase was soon shortened to "pushing the envelope". Be that as it may, I still find this usage somewhat irritating. I'm similarly annoyed by the usage of the now-common "I could care less" to mean "I couldn't care less": it just seems wrong to change a phrase in such a way as to make its intended meaning explicitly at odds with what it actually says.

Friday, October 19, 2012

Random Virginia presidential factoid of the day

I just learned (via Ta-Nehisi Coates), that John Tyler, who was elected president in 1840 and was the 10th president of the US, still has a grandson alive today! A reminder of how young the United States really is, as a country.

(In fact, the grandson apparently still lives on the ancestral estate, Sherwood Forest, not too far from Williamsburg.) Tyler is of special interest to the William & Mary government department, because at some point in the not-too-distant future we are scheduled to move into Tyler Hall, which is named after the Tyler family.

Thursday, September 20, 2012

Recent links of general interest

• The emoticon turned 30 yesterday. Read about it at the Atlantic (by Megan Garber).
What I find interesting, in a meta way, is that although we can apparently trace the emoticon itself
to a specific date and person, there does not appear to the same kind of clarity about
the word emoticon.

• Thoughtful article by Erik Bleich at Al Jazeera English: "How much free speech do we need?" For me it brought to mind an argument put forward by Michael Ignatieff in Human Rights as Politics and Idolatry: that rights are best thought of in instrumental terms, not as absolutes to be put on a pedestal. This is a difficult insight to translate into any practical prescription, but an important one to keep in mind nonetheless. (Note: Erik also has a very good book, published last year, on this and related issues: The Freedom to Be Racist?)

Ezra Klein points out that Romney really just doesn't understand what responsibility means. And Dante Chinni points out that the people Romney so blithely dismisses are disproportionately likely to be hid voters. Oops! Fortunately for him, they are probably also disproportionately unlikely to be swayed by Romney's now-notorious speech.

Tuesday, September 4, 2012

Poe's law

Today I discovered that there is an internet "law" about the similarities between parody and fundamentalism (of any sort): Poe's law. I had to look it up (here) because I couldn't quite figure out from the context what it meant. My main problem was that I was thinking in terms of Edgar Allan Poe, but that turns out not to be the right Poe. :-)

Anyway, here is the law:

"Without a winking smiley or other blatant display of humor, it is impossible to create a parody of Fundamentalism that SOMEONE won't mistake for the real thing."

The opposite effect is known as Poe's corollary (although it is generally just subsumed under the broad mantle of Poe's law):

It is impossible for an act of Fundamentalism to be made that someone won't mistake for a parody.

Some conservative reactions to the Colbert Show illustrate Poe's Law in action; some of the current proposals for the United States return to the gold standard (or even gold coinage) fall under the corollary.

Sunday, August 26, 2012

Generosity (of spirit & money) and personal exposure

Lisa Wade has a nice blog post about "Variation in the Generosity of the Rich", following up on an NPR story which in turn covered a story in the online "Chronicle of Philanthropy" about linking measures of charitable giving to the economic diversity of one's zip code.

That story uses IRS data from 2008 to calculate tax-declared charitable giving as a percentage of post-tax income; for reasons of data availability, only incomes above $50,000 are included. The study website has a good description of how the data were obtained and used. (This is crucial; indeed, if you ever find a study claiming results without telling you where they got their data, you should just ignore it — chances are great it is just crap.)

Anyway, one key finding is that people with lower incomes give a greater fraction of their post-tax income to charity. Another is that people who live in more economically diverse zip codes tend to give more.

Social psychologist Paul Piff, whom NPR interviewed for its story, argued this greater generosity may well be due to the fact that people who live in economically diverse neighborhoods are more aware of how important charity can be to poorer people, and hence give more.

Wade notes that there are at least 2 other explanations: "generous rich people move to diverse neighborhoods and stingy rich people isolate themselves" and "some other variable (e.g., political affiliation, religiosity) is correlated with both neighborhood preference and generosity."

In fact, I think the Chronicle's discussion of their data sources points to a likely alternative explanation that falls into Wade's third explanation: retired people, who have little active income but may give to charity from their savings, will get registered as giving a high percentage of their income to charity. Retired people who have lived somewhere all their lives may well be richer, on average, than younger people in their neighbourhood. This alone might account for a good portion of the findings.

However, Wade's musings about Piff's argument (and Piff's argument itself) also made me think of another "generosity" issue, where awareness due to personal exposure makes a big difference: support for gay marriage among deeply religious people (including pastors). I have lost count of the number of stories about pastors or other religiously active people who were strongly opposed to gay marriage until someone in their immediate surroundings (often a son or daughter) came out. Only then did they change their stance.

Given that pastors, in particular, should be expected to be quite empathetic, as well as aware of what goes on in their community, the fact that even for them having direct personal experience with an issue makes such a big difference suggests just how important direct personal contact can be in changing people's behaviour and opinion.

It appears clear, then, that there is a big difference between rationally knowing how important something is and viscerally feeling it. We might like it to be otherwise, but all evidence shows that it is the case. This has all kinds of important policy implications. (Singer's The life you can save covers a few of them, but by no means all).

Back to the original story: it would be interesting to see 1) whether controlling for age (i.e. retired people) has an impact on the pattern of the findings, and 2) whether the pattern is stronger in really densely occupied zip codes (where people rub shoulders much more directly than in sparser ones) and/or in zip codes where most people live for a long time (as opposed to those where people are more transient).

Tuesday, May 1, 2012

Ah, the reliability of "the cloud"

Yet another illustration of the importance of not trusting your valuable data to someone else's server (and especially not a single other server!):

"Car accident knocks out DMV computer system"

Monday, April 30, 2012

Lots of links I liked

This summer I'm going to gradually start blogging a bit more regularly again. In the meantime, recent days have brought a whole bunch of links I liked, which I pass on with minimal comment.

1. Mitt Romney disingenuously disparages Jimmy Carter, gets schooled by James Fallows.

2. The New York Times asked some literary lights  to give their on U.S. politics these days: E.L. Doctorow, Margaret Atwood, and Martin Amis.

3. Jonathan Chait highlights the difference between perception and reality as regards Paul Ryan's policy ideas and ambitions. (Note: apparently his criticism was too subtle for some, as can be seen here). Also worth (re-)reading in this context: Chait's takedown of the despicable Ayn Rand.

More tomorrow.

Tuesday, January 10, 2012

Stuff I liked

Very cool stop-motion animation in a Toronto bookstore: The Joy of Books

Creative use of language by Alabama linebacker Courtney Upshaw, after his team's defeat of LSU yesterday: “I’ve got to say, we outphysicaled them today”

Newest religion officially recognized by the Swedish government: the Church of Kopimi. Ctrl-C and Ctrl-V are their sacred symbols.

Friday, January 6, 2012

Education as a principal-agent problem

The Atlantic has an interesting article by Anu Partanen on its website: "What Americans keep ignoring about Finland's schools success." The basic argument is that Finland implemented education reform about a generation ago in order to achieve equity, and that excellence is, in some sense, a fortuitous by-product.

Of course, equity (in this case, equal access to the same quality educational resources for all children, regardless of wealth, ethnicity, location, etc.) is a hard sell in the United States. As a result, Partanen suggests, the dozens of fact-finding missions that come to visit Finland each year almost willfully ignore the key lesson of Finland's model.

Partanen repeatedly cites a top Finnish education official, Pasi Sahlberg, who has written the book Finnish lessons: What can the world learn from educational change in Finland? I have not read the book, so I may be kicking in an open door here, but it appears to me that Partanen's representation of Finland's model risks overlooking its most important attribute.

Consider education as a principal-agent problem. The state would like high educational achievement for all its citizens, and can be seen as the principal. The agents who, in the end, make this goal possible are the teachers. (In between are intermediate principals/agents such as state education boards and the like, but I'm ignoring those for the sake of simplicity.)

Assume that being a good teacher has two components: learnable (a skill set) and exogenous (for example, liking children) and that only a subset of the population of possible teachers has the requisite exogenous characteristics. How do you maximize the chances that your agents are all good teachers?

There are two obvious strategies to adopt: require serious training, and offer a salary (plus non-salary perks such as 'prestige') sufficiently high to attract enough people with the exogenous characteristics needed. This is what Finland appears to have done. It is also what some of the better private schools do in the United States, with pretty good success.

Of course, both training and higher salaries cost money. How might you try to achieve the same goal more cheaply? Well, you could try to specify precisely what students need to learn, and how, and then hire agents to do exactly, and only, that. And make sure they do so by administering ever more frequent standardized tests.

These agents won't need as many skills, and they won't necessarily need the exogenous characteristics, because you have specified exactly what they'll be doing anyway. So you can pay them much less. And you'll get a large number of unmotivated, relatively unskilled teachers. You'll still get a bunch of superb teachers as well, of course, but you risk demotivating them too, since they will not be able to deploy their own capabilities to their fullest extent. This situation would seem to exist in rather too many public school systems in the United States.

However, while Finland's focus on equity may be ignored by American educational reformers, the importance of higher salaries (and prestige) and better training are not. Education reformers have been calling for both for years. In contrast to Partanen, I would bet that you could retain the system of public and private schools the United States has at the moment and still achieve much better results, as long as you were willing to commit to paying for higher teacher salaries and more extensive training.

The problem, of course, is that while the state is the principal in my little model, it is also an agent: it is the agent of the American public. So we would need to convince the American public to spend more (perhaps considerably more) on teachers. And that is perhaps an even harder sell than equity.

Thursday, January 5, 2012

"He's voted for foreign aid repeatedly"

Another reason not to support Ron Paul. His son, Rand, going on the attack against Santorum, said on Tuesday, on CBS's The Early Show: "On economic issues, like foreign aid — he’s voted for foreign aid repeatedly."

Now I don't actually have any idea about Santorum's general stance on foreign aid. (I suspect that, like most of the Republican candidates this election round, he's largely opposed to it.) Regardless, I find it offensive that Paul is trying to imply that any support of any foreign aid at any time is obviously wrong.

Interestingly, Paul's statement also illustrates an argument about the framing of aid I make in my book. Paul clearly sees aid as an economic issue. I don't think he is being strategic here (though he might be): he really thinks of aid in those terms. More importantly, I would argue he thinks of it in those terms not because he is a libertarian (many libertarians in Europe think of aid in humanitarian terms, and support it), but because he is an American. Foreign aid has been framed in economic or security terms in the United States for decades, much more so than is the case in other donor countries. As a result, if Americans think aid serves no economic purpose, they're much more likely to reject it than are their counterparts in, say, Denmark.

International intervention: Why Libya and not Syria?

Philip Gourevitch has an interesting post at the New Yorker on the situation in Syria: The Arab Winter. It is striking to me how much stamina the protesters in Syria have displayed, given how little headway they have been able to make on their own against Assad's security forces, and how little international assistance, or even attention, they have received.

Gourevitch concludes that although the Arab League has taken some action on Syria, and has monitors on the ground now, these are likely to remain toothless. He does not deem it likely that more will happen: "Qadaffi was uniquely reviled, and uniquely disposable, and disposing of him was the easy part of the revolution (as it was with Mubarak in Egypt). With Assad it’s trickier—and the Syrian people remain hostages of that trickiness."

It is also worth thinking about geopolitics here (Gourevitch does so a little, but it is not the focus of his argument): chaos in Libya is much less worrisome to all kinds of key actors in world politics than is chaos in Syria, as a simple look at the map makes clear. What would instability in Syria mean for Israel, for Turkey, for Iraq, and for their various allies?

Still, there cannot be many other countries at the moment where the government is killing its own civilians at the same rate as Assad's troops are doing (an average of about 20 citizens per day since March, Gourevitch calculates). For those who argued for intervention in Libya on purely humanitarian grounds, that ought to mean something.

(Interesting idea for an undergraduate research project: compile data on rates of death-by-government-forces in authoritarian countries and see what kinds of patterns, if any, one can find. There are datasets on total deaths, so rates should be not that hard to deduce. But I can't think, off-hand, of anyone who has looked into them. Be the first!)

Wednesday, January 4, 2012

Understanding war

Interesting op-ed in yesterday's New York Times by John Tirman, of MIT: "The forgotten wages of war". Tirman points out that Americans rarely debate, or even think about, civilian destruction caused by their war efforts. This is problematic because, as Tirman argues, "The consequences of how we fight wars reveals [sic] a great deal about how and why others fight us."

Tirman has an important point, but I think he overlooks a deeper issue, one that bears on why the United States has shown rather less reluctance to go to war than have European countries in recent decades. In the American imagination, a war is something you go out and fight elsewhere; for many continental Europeans, a war is something that comes and destroys your country, and nobody is immune. Different experiences during World War II and the Cold War account for most of this difference, so it may be waning. Perhaps that is too bad: as Tirman points out, a greater understanding of what war means to those who have it visited upon them is salutary.

(By the way, Tirman has apparently written a recent book about the issue: The Deaths of Others: The Fate of Civilians in America’s War, which looks at American wars going back to World War II. The table of contents looks pretty interesting.)

Monday, January 2, 2012

Democracy, hunger, aid

The famine in Somalia this fall received a fair amount of press, which likely helped increase aid flows to the country. Aid efforts appear to have helped forestall a worst-case outcome, which is great. Of course, the fact that a famine erupted at all is due in large part to the absence of a functioning government (and continual fighting between groups aspiring to be the government) in the country.

Amartya Sen famously pointed out in his classic Poverty and Famines that famines are exceedingly unlikely under a functioning, democratic government. Hunger, however, does happen; moreover, being less extreme than famine, hunger is generally under-reported. The New York Times performs a valuable service today, highlighting hunger in the Democratic Republic of Congo (in an article by Adam Nossiter).

Congo is, at least nominally, a democracy. Indeed, elections were held in November, albeit with lots of problems and irregularities. The article gives a striking illustration both of Sen's thesis and of one mechanism that might undermine it.

On the one hand, Nossiter notes that the government is interested in enriching itself, not its citizens, investing nothing in the agricultural sector. In fact, one expert suggests that all agricultural projects undertaken in the country are funded by foreign aid donors rather than the national government. This is the kind of behaviour one might expect from autocratic governments that need not rely on public support. Moreover, it is the kind of behaviour that one could imagine leading to a famine at some point in the future.

On the other hand, Nossiter suggests, the daily struggle for sustenance may make it possible for such blatant disregard for the lives of citizens to co-exist with regular (albeit rather flawed) elections. So is a famine possible in democracies with rulers who have farmed out the "caring for the survival of our citizens" to foreign aid organizations? And if so, what can/should aid organizations do about this?