Pro Writing Aid: Ready for Prime Time?

There's no shortage of online "readability checkers" that claim to show how readable your text is via this or that metric. But few online tools (especially free ones) attack the readability problem with as much gusto, or in as many ways, as Pro Writing Aid.

Sadly, that's pretty much where the good news ends. My test-drive of Pro Writing Aid didn't find it to be much of an aid. But I give the creators an 'A' for effort. They're on the right track, at least.

The way the tool works is, you paste a bunch of text into PWA's online form and click the Analyze button. About ten seconds later, you'll see a summary report, with a column of links down the left side having names like:

• Overused words
• Word cloud
• Sentence variation
• Grammar
• Adverbs/passive
• Sticky sentences
• Clichés and redundancies
• Repeated words and phrases
• Phrases summary
• Diction
• Vague and abstract words
• Complex words
• Alliteration analysis
• Pacing
• Consistency
• Sentiment
• Time
• Dialog
• Homonyms

You can click on any one of these to see potential problems highlighted in your text. Unfortunately, "potentials" outnumbered actuals quite a bit in the testing I did. (I used several pieces of my own writing for testing, as well as sample chapters from The Adventures of Huckleberry Finn. You'd think the latter would've thrown a lot of flags and warnings, but oddly enough, "it warn't that terrible bad.")

Different writers will get different mileage from this kind of tool, so go ahead and try it yourself: You may very well find it useful. For me, it was like using a spellchecker in that I spent the vast majority of the time dismissing things that a computer would flag as wrong but that a human would know were right.

The Adverbs/Passive tool was puzzling. It flagged every adverb (including every occurrence of "only"), but highlighted few or no instances of passive voice in any of the five writing samples I tried (each sample averaging 1,700 words).

The Word Cloud feature makes pretty pictures but is otherwise useless.

The Overused Words report flagged 47 instances of "it" in Chapter 3 of Huck Finn, saying that about 26 instances could be removed. In point of fact, I couldn't find any instances that warranted removal.

That's not to say the Overused Words report is useless. But as I say, it tends (like many of the other tools) to report far more false positives than it reports good catches.

A fundamental problem with utilities of this sort is that they don't make allowances for the (huge) differences between dialog and narrative, in a piece of text that contains both (such as a chapter from a novel). The reason this is a big problem, obviously, is that spoken English is quite a bit different from written English: It's different as to vocabulary, diction, syntax, word length, sentence length, sentence variety, pacing, use of clichés, constructions based on slang, and probably two or three dozen other particulars. You can't treat dialog and non-dialog text as one and the same thing. They're distinct. What works for one won't necessarily work for the other. I saw this when I passed a piece of dialog-intensive sample text through the Pro Writing Aid analyzer and noticed many more flags in areas of spoken English than in areas of expository English.

Another potential problem with utilities of this type is that they make no distinction between writing aimed at adults and writing aimed at children or young adults. (Or for that matter, writing aimed at a professional audience vs. writing aimed at a lay audience.) It would be nice if there were a way to specify the intended age group for the writing sample in question, so as to get an age-appropriate readout of things like diction and "sticky sentences." 

Long story short: Pro Writing Aid was a disappointment, for me. But I recognize that it might well be a boon to others. So by all means, try it out yourself. And let me know what you think.

reade more... Résuméabuiyad

Evil Writing Prompts

• Write a query letter and ten sample pages for a novel about the dismal state of the publishing industry and send it to two hundred literary agents.

• Write a synopsis for a time-travel vampire romance set in the fantasy kingdom of Twillador and hand it out to every speaker at a writer's conference.

• Write a query letter for a young adult novel called H. Finn, about "a boy, a raft, and a runaway nigger." Sign it S. Clemens, and send it out to 100 literary agents. Publish the rejection letters in a blog.

• In your latest manuscript, do a global search and replace, putting the name of a well-known literary agent in place of your novel's villain's name, then send it to publishers that deal with that agent.

• Include the first ten pages of Flowers for Algernon as a writing sample in a query letter. Send to 100 agents, claiming you are a mentally challenged adult seeking representation, and mention that the Americans with Disabilities Act requires them to write you a detailed, personalized letter back, lest you sue.

• Create a fake newspaper clipping about your unfinished novel, scan it, and paste the image directly into a fake bio. Hand it out at writer's conferences.

• Using Google Translate, translate the first ten pages of your novel into Hindi, then have Google Translate translate it back into English. Use the resulting hysterically mangled text as your writing sample. Send it under an Indian pen name to your least favorite literary agents.

• Find the personal e-mail addresses of ten literary agents who accept snail-mail proposals only. Send each one a Word attachment containing the full text of Moby Dick and include in the subject line "REQUESTED MATERIAL."

• Write 25 queries containing 25 random sentences and send them all to one literary agent, from 25 fake e-mail accounts.

• Write an e-mail marked "URGENT" to fifty literary agents, claiming you have gotten offers of representation from multiple agencies for a work you never submitted to anyone. Use weird fonts and font colors.

• Write a fake interview with yourself in which you talk about having won fake awards. Include the interview's URL in queries sent to fifty overseas agents.

• Come up with 25 fake "Praise for" quotations to put in the front of your book. Be sure to quote dead authors. Include with your manuscript.

• Write your own first page for a (real) bestselling book that's on the shelves now. Print it out. Go to the bookstore, find that book on the shelves, and paste your first page over the first page of the book.

Disclaimer: Folks. Folks. This all meant in jest. For heaven's sake don't actually do any of this shit.

reade more... Résuméabuiyad

Asian-Americans destroy the "maker/taker" narrative

The absurd outburst of Nate Silver Denialism was just the most egregious example of the alternate reality bubble that the American conservative movement has been constructing over the last decade-and-change - the phenomenon that used to be called "epistemic closure." In this devastating analysis, Conor Friedersdorf lays out just how monumentally stupid is is for a movement to divorce itself from extant reality:

I see a coalition that has lost all perspective, partly because there's no cost to broadcasting or publishing inane bullshit. In fact, it's often very profitable. A lot of cynical people have gotten rich broadcasting and publishing red meat for movement conservative consumption.

Since the election, I've been reading The Corner, the group blog of the National Review, a publication which probably still represents the intellectual forefront of the conservative movement. I've seen some very intelligent and thoughtful discussion there, and also some absolute head-in-the-sand denialist comfort-food. But one thing I don't see anyone challenging is the master narrative of modern American conservatism: the "maker/taker" story.

The "maker/taker" story is exactly the "47 percent" story that Mitt Romney told at that fund-raiser. It's the idea that the Democrats' core constituency is a bunch of lazy and/or untalented "takers" who want to use the government to steal from the hard-working "makers". Here's The Corner's David French summing up the idea:

To tens of millions of American voters, a conservative message of self-reliance and individual economic freedom is, quite frankly, terrifying. 

First, each of Obama’s core constituencies (single women, African-Americans, and Latinos) is seriously — and disproportionately — economically disadvantaged compared to the classic paradigm of the white, college-educated Republican voter. The rates of poverty and near-poverty among these groups are much greater, thus causing a critical mass of both populations to suffer — even if they’re technically middle class — from a greater degree of economic insecurity...Ideologically and historically they are pre-disposed towards statism as the means of alleviating economic insecurity and distress.

As you can see, the "maker/taker" narrative has a strong ethnic angle; the "takers" are supposed to be mostly minorities and single women. White men and their wives produce things; blacks, Hispanics, and sluts single women live on the dole. Naturally, this ethnic angle plays well with the conservative "base", i.e. Southern and exurban working-class whites for whom politics is ethnic and tribal. Here's Fox News' Bill O'Reilly reinforcing that racial version of the narrative:

The white establishment is now the minority. And the voters, many of them, feel that the economic system is stacked against them and they want stuff. You are going to see a tremendous Hispanic vote for President Obama. Overwhelming black vote for President Obama. And women will probably break President Obama’s way. People feel that they are entitled to things and which candidate, between the two, is going to give them things?

So, here's the problem: The narrative is wrong. Completely, utterly, wrong. The Democrats' appeal is not based on people "wanting stuff". And how do I know this? I know it because Asian-Americans voted for Obama by a 3-to-1 margin. Check this out:

Much has been made of the Latino vote and its crucial role in boosting President Obama to victory, but it was Asian Americans who made the most dramatic shift in support for the president Tuesday. 

Exit polls show that 73% of Asian Americans backed Obama, an 11-point increase since 2008.  Asian Americans came out in such force for Obama that they topped Latinos as his second-most supportive ethnic group, behind African Americans... 

While Asians accounted for just 3% of the electorate – up from 2% in 2008 – their overwhelming support made them a key component of the Obama coalition, especially in swing states like Virginia, Florida and Colorado.  

And their numbers are increasing rapidly. They were the fastest-growing ethnic group from 2000 to 2010... 

73% [of Asian-American voters] supported Democrats in congressional races.

So, for those of you who don't know this, Asian-Americans make more money than white Americans. Thus, they pay more income tax. And Asians are half as likely as the average American to be on welfare.

Thus, Asian-Americans, by the Romney/O'Reilly/French definition, are "makers", not "takers". Even more than whites. They're also more likely to be married. And to start businesses.

And yet Asian-Americans broke for Obama 3-to-1. David French should definitely be including them in his list of Obama's "core constituencies". The fact that he doesn't do so is a telling sign of "epistemic closure" - of conservatives not seeing what is plainly in front of them, preferring instead to repeat to themselves a pleasant, soothing, but false story.

If they're not "takers", why did Asian-Americans break so strongly for Obama? The answer is pretty clear: Conservative ethnic identity politics. The American conservative movement has made it abundantly clear that it sees America as a "white people country", and views Asians - like blacks and Hispanics - as guests (at best) or interlopers (at worst). The blood-and-soil white ethnocentrism of the conservative movement makes Asians feel like permanent foreigners in their own country, and they don't like feeling like that. Who would?

I'm pretty sure this is the right answer. Why? Because I myself am a member of a group that is demographically and electorally similar to Asian-Americans - namely, Jewish-Americans. Jews are America's second-richest religious group (behind Hindus), and yet Jewish voters broke almost 3-to-1 for Obama this year, and more than 3-to-1 in 2008. Why do Jews vote Democratic? Simple: for all their talk of "Judeo-Christian values" and support for Israel (as if anyone cares about that!), the Republicans make it clear that they think the ethnic core of America is not just white, but Christian as well. Jews don't want to have their schools lead them in prayers to Jesus. We don't like it when Bill O'Reilly demands that stores put only "Merry Christmas" on their holiday banners, and not "Happy Channukah". 

We Jews are not stupid; we know that the grassroots of the conservative movement, at least in its present incarnation, will never accept us as true "sons of the soil". The liberal movement will. It's as simple as that. And I'm pretty sure Asian-Americans are thinking along similar lines.

So if conservative intellectuals really want to turn their movement away from the self-destructive path that they're on, they need to face up to one overwhelmingly important piece of reality: White (Christian) ethnocentrism is turning everyone else against them. Getting rid of that ethnocentrism will take more than putting some Hispanics, or some Asians, or some Jews on the speaker's podium. It will take more than pretending to vote for - or even actually voting for! - Herman Cain or Marco Rubio or Bobby Jindal.

What will it mean? It will mean no longer talking about a "culture war". It will mean no longer talking about the "real America". It will mean rejecting race-baiters like Rush Limbaugh and Glenn Beck and Andrew Breitbart - not grudgingly, but wholeheartedly. It will mean rejecting nativist groups like the Minutemen. It will mean admitting that the Civil War was all about slavery, and that the Confederacy were the bad guys. It will mean disavowing the whole stupid narrative that blacks and Hispanics are a bunch of lazy "takers".

In other words, it will mean doing a lot more than the conservative movement is currently prepared to do. But guys, at least face up to reality. At least open your eyes and see what the rest of America really thinks about you. The Asian-American vote is an unmistakable sign that your master narrative is wrong.


Update: Lots of other people are saying the same thing. For example, here's Chris Hayes. Also, here's Paul Krugman with a nice graph.

Update 2: In a new article, Charles Murray says much the same thing as this blog post. Which is interesting, given Murray's prominent role in promoting the general white-supremacist overtone of the conservative movement.

Update 3: Richard Posner is with us too.

reade more... Résuméabuiyad

Should I read Casey Mulligan's book?

Casey Mulligan has a new book out, called The Redistribution Recession, advancing his unorthodox ideas about the cause of the Great Recession. Tyler Cowen is a big fan, writing:

To get to the point, [Mulligan's book] is quite good...there are only a few readable books which integrate actual empirical research with a look at the Great Recession.  This is by no means the whole story, but this is a book which anyone seriously interested in the topic should read.

Should I read this book? Well, it does seem interesting, and it comes highly recommended. I'm sure I'd have lots to say about it. On the other hand, I do have limited time; I can't read every pop econ book that comes across my computer screen.

I already know some of what the book will say, since it appears to incorporate many of the ideas that Casey Mulligan writes about on his blog. His central thesis - that unemployment insurance, food stamps, and other government benefits caused the Great Recession - has always seemed highly implausible to me, for three reasons:

1. There was no big policy change preceding the recession, and hence any effect of government policy would have had to be forward-looking; in other words, Mulligan's thesis requires that workers stopped working because they expected Obama to be elected president in 2008 and increase govt. benefits. That seems quite implausible to me.

2. Real wages fell in the crash of late 2008 and early 2009. Later they partially rebounded, but overall their  growth in and after the crash was quite sluggish. Lower wage growth means there is no shortage of labor; if there were a shortage, we'd see wage growth increase. So negative shocks to labor supply, of the kind postulated by Mulligan, can't be the whole story.

3. The recession was global in nature; in many countries, it was worse than in the U.S. Given the wide diversity of policies, pre-recession policy changes, and post-recession policy responses, it seems logically impossible that policy could explain the global phenomenon. Perhaps a policy shock in the U.S. caused contagion that spread to other countries? If so, their recessions should look much different than ours.

I am not the first to make these points; John Quiggin (see also here) and Paul Krugman have already noted these things.

If Mulligan's book explicitly addressed the three critiques, I'd read it. But so far, I have heard nothing to indicate that it does so. On Point 1, Mulligan is silent, pointing only to the levels of benefits after the recession was already well underway. Point 2 has been made before, but Mulligan hasn't really responded to it. And on Point 3, Mulligan simply waves his hand and says "If you think other countries are important, you go study them and tell me what you find."

So without directly addressing these three points, Mulligan's thesis does not pass the smell test. And I don't have time to read books whose theses don't pass the smell test.

Which is a shame, because I suspect that Mulligan has actually uncovered some important and interesting facts about the labor market. The seasonal behavior of employment is weird. The growth of wages over the 2008-12 period, even though it was too sluggish to indicate a labor supply shortage, is interesting. Is the labor market segmented? Is real wage growth sticky? Those are important questions! And after having studied the subject, I'm not at all convinced that we really understand how labor markets work (I'm sure Robert Shimer, for example, who is one of the best and most serious labor-macro economists, would agree with that statement).

But Mulligan seems less interested in pointing out weird facts about labor markets - which would tend to invalidate mainstream theses about how those markets work - and more interested in pushing an alternative grand overarching thesis that doesn't really make sense. Instead of saying "Hey, here are some facts that throw a wrench into the Aggregate-Demand-and-Sticky-Wages theory of unemployment!", he instead says "Unemployment is caused by government benefits encouraging people to take vacations."

Why does he try for the more ambitious goal of explaining the world, instead of the more intellectually defensible but limited goal of disrupting existing theories? It's pretty obvious that politics is involved. Most conservatives naturally want to believe - they desire to believe - that we live in a world in which economic problems are caused by the laziness of the poor and the misguided redistributive instincts of liberal Robin Hoods. Mulligan purports to give them a reason to believe this. The title of the book - "The Redistribution Recession" - should leave no doubt as to the book's target audience.

When I have a strong reason to believe that a book's thesis is overstated for political appeal, I naturally view the author's evidence with more suspicion. That doesn't mean I discount the evidence, or the arguments. But it raises the hurdle for whether or not I decide to commit the time and effort to read a book.

So Casey Mulligan may be right about some things, or have some good points. But unless I hear a lot more of the details, I am going to overlook and ignore those good points, because A) the thesis doesn't yet pass the smell test, and B) the book has a bit too much political appeal for my taste. My apologies.

reade more... Résuméabuiyad

New Atlantic column: Who will be better for the economy, Obama or Romney?

I have a new column up in the Atlantic, which tries to answer the question of "Which candidate will be better for the economy?". The short answer: We don't really know, since there are so many things we don't know about candidates, politics, and how the economy really works. But there are a few issues on which Democrats have consistently proven themselves to be more responsible and sensible than Republicans over the past thirty years, and there is no reason to think these patters will change soon. From the column:

1. The National Debt...Since the national debt began climbing in 1981, a clear partisan pattern has emerged: Republican presidents have tended to expand the deficit more than Democratic ones...Has the pattern been broken [under Obama]?...[There is] a hugely important difference between Obama and his debt-ballooning Republican counterparts - Obama's deficits have come during a historic recession, while Reagan and Bush borrowed money while the economy was expanding...So Obama's deficits at least might be a temporary phenomenon, while the Reagan and Bush deficits were clear indicators of a short-sighted, unsustainable policy...This suggests that if we care about reversing the deficit once the economy recovers (as it now seems to be doing), we should go with Obama over Romney, despite Obama's large deficits... 

2. The Full Faith and Credit of the U.S....In 2011, Congress' near-failure to raise the debt ceiling nearly precipitated a default on a portion of America's national debt....Four days after the debt ceiling deal, Standard & Poor's downgraded their rating of U.S. government debt...Why should we blame Republicans for the debt ceiling debacle, rather than President Obama?...First, congressional Republicans repeatedly rejected Obama's attempts at compromise... Second, Republicans themselves expressed a blasé lack of concern about the prospect of default...Romney's membership in a party that reduced the value of the full faith and credit of the U.S. government is a reason to vote for Obama. The off chance that the GOP would actually tolerate a sovereign default is enough to tip the scales strongly in favor of the Democrat... 

3. Public Goods...Our crumbling infrastructure received a "D" grade from the American Society of Civil Engineers in 2009. Obama's 2009 "stimulus" bill plugged some of the gap, but a more sustained effort is needed. However, Romney's budget plan includes large cuts in the part of the budget that pays for public goods, and Republicans in general have shown very little inclination to repair our aging roads, bridges, electrical grids, and ports... 

I say we focus on the areas where we have good evidence. On the issues of fiscal responsibility, debt default brinksmanship, and infrastructure spending, Democrats have come out looking better than Republicans again and again since 1980. There is no reason to think that Mitt Romney would change the Republican side of that pattern. Barack Obama is the safest bet.

Now, note that I am trying to be objective here, and wear only my economist hat. I personally have a lot more reasons to vote for Obama - social issues in particular. But even when we just talk about the economics, I think Obama comes out looking like the more rational choice.

reade more... Résuméabuiyad

How not to criticize Japan

I will be the first to tell you that Japan has serious problems. Chief among these, in my estimation, are A) a dysfunctional labor market, B) institutional support for unproductive companies and company divisions, C) poor corporate governance, D) a dysfunctional political system, E) underemployment of women, and F) the large national debt. Obviously these tend to reinforce each other, leading to a bad equilibrium.

But the way to untie this Gordion Knot is not to simply start wringing our hands and flailing about in every direction, blaming any trend or story upon which we happen to lay our eyes. Unfortunately, this approach to Japan analysis is relatively common in the Western business press. For example, consider this article in MarketWatch. "Once a powerhouse, [Japan is] withdrawing from the world stage," the subheading laments. Here are some excerpts:

“That’s about right,” [a friend living in Tokyo] said, “battered by deflation and a overly strong currency, the Japanese have concluded that they can’t compete with a rising China and are withdrawing into themselves.”... 

Japan’s 27-year long dominance in machine-tool production came to an end two years ago. China, where production costs are 40% lower, has seized the top spot even though the Japanese emphasize that they continue to lead in quality. 

Pummeled by a strong currency and two decades of deflation, Japanese companies are shifting production to China and elsewhere. Japan’s industrial core is eroding and threatened with being hollowed out... 

What went wrong? How did a nation with world-class companies, a highly educated work force, and a well-deserved reputation for efficiency and discipline descend into protracted decline? 

Since Japan’s asset bubbles burst over 20 years ago, policy makers have persistently fiddled with the levers of monetary and fiscal policy but their efforts have spectacularly failed... 

Meanwhile, the exchange rate of the yen, regarded as a reflection of the nation’s economic health, has steadily appreciated from the 227 to the dollar that prevailed in 1983 to a mere 78 yen today...At any rate, the currency’s rise has done little to reduce Japan’s chronic balance of payments surplus... 

Meanwhile, the brutal deflation that accompanied the bust persists. Falling prices have translated into massive wealth destruction... 

Japanese officials counter that they’ve tried monetary stimulus, including zero interest rates and quantitative easing, but had meager results... 

The most troubling aspect of Japan’s malaise may be psychological...Japan seems to have its lost its self-confidence along with its drive for economic leadership.

A few points.

1. Why do people think Japan ought to "compete" with China? Just because they are both East Asian countries? Why does the growth of China's economy threaten the health of Japan's economy?

And what does "compete" even mean in this case? Does it mean - as the author hints when he talks about machine-tool manufacturing market share - that Japan ought to try to make sure its companies hold dominant global market share positions over China in every conceivable industry? That would go directly against the strategy advocated by Michael Porter, Hirotaka Takeuchi, and Mariko Sakakibara in their book Can Japan Compete?, in which they identify lack of specialization and blind pursuit of market share as flaws in the Japanese corporate model.

2. Machine tools? Why machine tools? Is a country's economy on the wrong track if it doesn't have the #1 global market share in machine tool manufacturing? Can there only be one successful economy in the world at a given time, defined by who is manufacturing the most machine tools? If so, the U.S. and Germany and Canada and Singapore are all out of luck.

Or is it just because Japan used to be the #1 machine tool manufacturer? Does loss of marketshare leadership in one industry indicate general economic decline? Should be worried that Japan is no longer the world's leading manufacturer of paper doilies?

I just don't get it.

3. Why is a "strong" yen bad? OK, it hurts exporters. But it helps importers and consumers. Don't we always hear people saying that a "strong dollar" is an indicator of America's economic strength? Why should Japan be exactly the opposite?

(Note: although this author cites a study claiming that the yen is overvalued, most studies say that it is fairly valued.)

And if Japan's balance-of-payments surplus is a problem, why should the yen weaken? That makes no sense, since a weaker yen would tend to increase the balance-of-payments surplus.

4. Monetary policy did nothing? Well, maybe. But from where I'm sitting, it sure looks as if Japan's quantitative easing in the early 2000s was quickly followed by a burst of good economic performance (the "Koizumi boom" in 2004-7). Now, correlation does not equal causation, but to conclude that monetary policy is ineffective seems a bit ridiculous to me, especially when sustained low real interest rates will be necessary to erode Japan's mountain of debt.

5. Deflation causes wealth destruction? Well, maybe it does, by restricting economic activity. But I don't think that's what this author is talking about. I think what he's doing is identifying consumer price deflation with asset price deflation. In fact, the two are completely different things. The U.S. experienced low inflation in the 1990s, but asset prices soared. Deflation may cause wealth destruction via macroeconomic effects, but it does not equate to wealth destruction.

6. And now we get to the part at the end when the author finally answers his own question of "What went wrong?". "The most troubling aspect of Japan’s malaise," he declares, "may be psychological."

That would play well in a 1980-vintage Ronald Reagan stump speech. But how the heck does this author (who lives in Washington and works for a Hong Kong newspaper, by the way) have any idea that this is the case? How does he know that Japan's psychological pessimism is a cause of poor economic performance rather than an effect? I'd be pretty bummed if my economy was stagnating! (Actually, that is not exactly an "if" at this point.)

The idea of persistent psychological malaise - a lack of "animal spirits" - is an old one, and a tempting one. Wouldn't it be neat if macroeconomics were all about national will? If all we had to do to bring back robust growth was to turn that frown upside down, chin in, chest out, nose to the grindstone, when the goin' gets tough the tough get goin', get busy livin' or get busy dyin'? Maybe if Japan had a Ronald Reagan to stand in front of a flag and tell them that everything was going to be OK, then their dysfunctional labor markets, corporate governance, and politics would vanish like so much smoke on the breeze?

OK, obviously I kid. But explaining Japan's problems in terms of the "national mood" doesn't seem very helpful to me. (And in fact, some of my own research indicates that happiness probably doesn't exert much of a pull on the economy at all!)

Basically, writers like this analyst are speaking to a very particular audience: foreign investors looking for a simple story to tell them whether to pull their money out of Japan. The dominant narrative coming out of Japan is one of general doom and gloom, and people like to read things that reinforce the dominant narrative, because people like to feel sure about things. So articles like this thrive because they pick out random facts - machine-tool market share! - and buzzwords - psychological malaise! - that seem to support the narrative. The problem is, this sort of reporting doesn't help anyone think carefully about what Japan's problems really are, or how to fix them. All this hand-wringing does is to encourage the very malaise that it decries.

Not that I think that articles like this are taken seriously by members of the policymaking elite, of course. But they still annoy me.

reade more... Résuméabuiyad

Why predict percentages?

"I'll bet you $20 that there's a 70% chance that Obama will win the election."

That's a bet nobody will ever collect on, because it's impossible to verify a percentage chance of something. So why do forecasters like Nate Silver - and any bookie or oddsmaker - say that there's a 70% chance of Obama winning? Why don't they just make an up-or-down prediction?

The answer is: Those percentages give you information to the extent that you believe in the forecaster. If you believe that Nate Silver's model is the best available forecast of the election results, then you believe that the odds he gives are the "fair" odds. Knowing the fair odds will help you hedge properly against the chance that Obama or Romney will be elected president, say for example if you have a business whose livelihood depends on policy. It might also help you make a buck on InTrade, especially if you believe that things like market manipulation can make those prediction markets temporarily inefficient.

But will those odds tell you whether to believe in a forecaster? Surprisingly, the answer is "maybe". The main way to ascertain how good a forecaster is is to observe a repeated sample - just watch the forecaster make 100 forecasts (of who will win, not of what the odds are!), and observe how often (s)he is wrong. This is the main way that you tell how good a forecaster is. But it isn't the only way. If you look at the forecaster's odds and find that they move in predictable ways, then you know that the forecaster could have done a better job. After all, why predict a 70% chance today when you have good information telling you that it will change to a 78% chance tomorrow? Just predict a 78% chance today!

So the odds can be useful in evaluating forecasters, as well as in making use of forecasts.

reade more... Résuméabuiyad