Tuesday, July 18, 2017

Synthetic Control: Ur (probably) doin' it rong!

People, I've been refereeing a lot of synthetic control papers lately, and I have to say that I don't like what I'm seeing.

I'm seeing 3 big mistakes that people are sometimes trying to sell as features.

Let's discuss.

1. Large, indiscriminate donor pools are not advisable. Look at the Godfather Abadie's papers. The donor pools are 20-40 units. People seem to have the crazy belief that more is better.  It's not.

Let's let Abadie et al. AJPS 2015 explain it:

“Constructing a donor pool of comparison units requires some care. First, units affected by the event or intervention of interest or by events of a similar nature should be excluded from the donor pool. In addition, units that may have suffered large idiosyncratic shocks to the outcome of interest during the study period should also be excluded if such shocks would have not affected the treated unit in the absence of the treatment. Finally, to avoid interpolation biases, it is important to restrict the donor pool to units with characteristics similar to the treated unit. Another reason to restrict the size of the donor pool and consider only units similar to the treated unit is to avoid overfitting. Overfitting arises when the characteristics of the unit affected by the intervention or event of interest are artificially matched by combining idiosyncratic variations in a large sample of unaffected units. "

Got that? pick your donor pools with smarts and with care.

2. Throwing away available pre-intervention outcome data is not advisable.  SC is subject to the same critique as matching, that unobserved factors are not being accounted for. For this reason, the Godfather stresses that the pre-intervention period should be long. From the same AJPS paper:

Critics of Mill’s Method of Differences rightfully point out that the applicability of the method may be limited by the presence of unmeasured factors affecting the out- come variable as well as by heterogeneity in the effects of observed and unobserved factors. However, using a linear factor model, Abadie, Diamond, and Hainmueller (2010) argue that if the number of preintervention periods in the data is large, matching on preintervention outcomes (i.e., on the preintervention counterparts of Y0 and Y1) helps control for unobserved factors and for the heterogene- ity of the effect of the observed and unobserved factors on the outcome of interest. The intuition of this result is straightforward: Only units that are alike in both observed and unobserved determinants of the outcome variable as well as in the effect of those determinants on the outcome variable should produce similar trajectories of the outcome variable over extended periods of time. Once it has been established that the unit representing the case of interest and the synthetic control unit have similar behavior over extended periods of time prior to the intervention, a discrepancy in the outcome variable following the intervention is interpreted as produced by the intervention itself.“ 

I've seen papers discarding pre-intervention data to make the sample "more reasonable". It actually makes the experiment less credible. Now if your are studying a post-Soviet country, sure, you are not going to have a long pre-intervention period. But you should realize that your results are not going to be super robust.

3. Using all the possible lagged outcome variables as predictors is not a good idea! I know, I know, people have done it in good journals and argued in favor of it.

But, "using all outcome lags as separate predictors renders all other covariates irrelevant. This finding holds irrespective of how important these covariates are in order to accurately predict post-treatment values of the outcome, threatening the estimator’s unbiasedness."

To quote Lizzy Warren, "holy guacamole"!!!

Here is a link to the relevant paper.

Here's a bit longer and even scarier quote from it,

"Consequently, in the SCM application we mainly focus on throughout this paper—Billmeier and Nannicini (2013), who analyze the impact of economic liberalization on GDP—the covariates taken from the literature do not affect the synthetic control. The authors obtain the very same counterfactual that would have followed if they had used economically meaningless covariates—or even none at all.3 We further discuss that solely optimizing the pre-treatment fit of the dependent variable and ignoring the covariates can be harmful: the more the covariates are influential for future values of the outcome, the larger a potential bias of the estimated treatment effect will become, possibly leading to wrong conclusions."

So don't use an indiscriminate donor pool. Don't use all the possible lagged outcomes as predictors. Don't throw away pre-intervention data. Unless you want me to go all "reviewer #2" on your asses!

Monday, July 10, 2017

Listen up people, the Body of Christ AIN'T GLUTEN-FREE!

.....But it can contain GMOs.

This is an actual ruling from an actual religion in the actual year or our Lord 2017.

Since neither Tony Gill or Phil Magness are Pope, you may wonder why this bizarre edict has been issued.

Well, "The new rules are needed because the bread is now sold in supermarkets and on the internet, the cardinal said."

LOL, thanks for clearing that up.

So there you have it people. If you want to pretend that a piece of bread turns into a 2000 year old body, you better be damn well sure that bread has gluten. You know, just like Jesus!

Monday, July 03, 2017

NOT an actual email from a book author...

To be quite clear....

I did not receive the following email from any Duke colleague, or any professor of history. But I did receive the email, and I thought I would repost it.  It pretends (falsely) to be a summary of my review of Democracy in Chains. It isn't. But it is fair to say this is what the author might have said.


Thanks so much for the very positive review and affirmation of my book Democracy in Chains by my colleague Duke Professor of Political Science Dr. Michael Munger. Please find below salient extracts from Dr. Munger’s review.

Democracy in Chains: The Deep History of the Radical Right’s Stealth Plan for America, by my Duke University colleague, Nancy MacLean, a professor in our distinguished Department of History… is… a remarkable book. MacLean has argued persuasively throughout her career for the historical method…in this book… MacLean recounts an exchange, a conversation really, between two conservatives…intent on reverse-engineering a …political order in America…using shadowy methods and discredited theories. Democracy in Chains is a work of …historical...research underpinning …facts … from a much larger record…drawing reliable conclusions about history. 

Democracy in Chains is a great story… of … James M. Buchanan, the winner of the 1986 Nobel Prize in Economic Sciences. MacLean is able to decode the true meaning of his …writings which…sought … to bring down.. America and replace it with a plutocracy. …MacLean’s excellence as a writer,…careful sifting of evidence and respectful encounters with opposing points of view…reveal … that …Buchanan … wanted to establish a …society …for racial segregation… 

MacLean’s book…is admirably academic and careful. … MacLean …found …the attempt by segregationist forces to support vouchers. MacLean says, “The economists made their case in the race-neutral, value-free language of their discipline, offering what they depicted as a strictly economic argument—on ‘matters of fact, not values.’” MacLean … support the claim that Buchanan advocated vouchers for the purpose of achieving segregation. … Buchanan’s support for vouchers and for school choice arose from a deeply held concern for …a …repressive apartheid society where African-Americans were …murderous and … must be forcibly suppressed… MacLean has discovered a number of important documents from the history of Public Choice, and other aspects of the history of the 1960s and 1970s in academic economic circles. There is a terrific example on pp. 115–117, where the “glee” of Buchanan and others about their conspiracy, gathered around a roaring fire in the remote mountains of Virginia, is documented.

 … MacLean has…written that history, using … public documents that … destroy...the conspiracy; …that … would sweep the nation, and the world… When summarized in this way, MacLean’s thesis really does read like a … narrative thread connecting the documents and discussions that …strategize about how to win back the White House and rejuvenate the conservative movement… The contribution of Democracy in Chains, then, is to do two things…Identify James Buchanan as the focal point of the revolution, and identify the content of Public Choice research and teaching as anti-Constitutional and anti-democratic… As I hope has been clear, as a book Democracy in Chains is well-written, and the research it contains is both interesting and …illuminating…as an actual history…of the work of James Buchanan …to end democracy in America. 

My thanks to the actual author of this email, Steve Spearman. And he is right: every word of the above actually appears in my review, and in precisely this order!

Saturday, June 24, 2017

NCGA Does the Right Thing. Sort of.

A Senate bill (SB656), already passed by the NC Senate, is now out of committee in the NC House, and scheduled for a vote on Tuesday, June 26.  The "committee substitute" bill is here, if you want to look at it.....

What would this mean? NC has had some really restrictive ballot access laws, since 1983. In particular,
it now takes about 110,000 (individually validated) signatures to get on the ballot ("access") and 2% of the vote in either the Prez or Gov races to stay on the ballot once on ("retention").

The bill, if it passes, and it's expected to pass, would change those rules as follows:
Access: 10,000 signatures
Retention:  Either the 2% rule for Prez or Gov votes in NC in previous election OR have "a candidate" on the ballot in 80% of the states.

Now, the Libertarian Party has been on the ballot continuously since 2008, when (ahem) I got more the 2% of the vote for Gov.  In 2012, Barbara Howe did it, and in 2016 Lon Cecil did it.  A number of Republicans blame the LP for the Republican loss of the Gov race in 2008 (implausible) and in 2016 (extremely plausible).

But the point is that the LP is already on, again, for 2020. My man Gary Johnson actually got 2.74%, 130k votes, the first time an LP Prez candidate has EVER secured ballot access in NC.  Whoo-HOO!

Okay, so here's the thing. The Republicans don't have many chances to look like they are for ballot access, or expanding electoral freedoms (which include voting for the candidate of your choice). AND, they are likely tired of having to lose some votes, enough in close races to change the outcome, to the annoying "third party" Libertarians.

You can have a double win, if you are a Republican strategist, by cutting the restrictions on ballot access! You get credit for having a more sensible and less draconian set of rules, more in line with other states (NC regularly makes, on merit, those "Worst Laws in the US" lists, when it comes to ballot access).

And you get the Greens on the ballot, diverting votes from the Dems! Is there a downside?

I don't really see one. An interesting question is whether Gov. Cooper will sign it.  If a goofball like me can figure this out, I expect Mr. Cooper will not need to have the effects explained to him. But it will be difficult to veto the thing, because it really does just bring NC back into compliance with pretty normal rules for party access. This is hardly a radical bill, it just also "happens" to help the Republicans at the ballot box.  (Here is a story from the right-leaning Carolina Journal, which doesn't mention the political angle--adding Greens takes away Dem votes--or, oddly, the new 80% retention rule added in the committee version).

EDIT: The reason the 80% (40 states) is important is that Stein was on the ballot in 44 states in 2016.  That means that, depending on the way the law is interpreted retroactively (and it seems to be retroactive, since it says "previous election"), Greens would be on the ballot for the 2018 midterm elections in NC.

Tuesday, June 13, 2017

Taxation without Representation; or Adam Silver wants to lower the Black Youth employment rate

The NCAA and NBA are back at it again, with the NBA making noises about further tightening restrictions on who's allowed to work in their industry.

And of course, he's doing it for the kids!!

Silver said he's "rethinking" his position on the age limit. He called the process "half and done" for freshmen who turn pro after their first seasons on campus, suggesting the players in that pool fail to make academics a priority.

"I don't think it's fair to characterize them as going to one year of school," Silver said on Cowherd's show.

So some old white skeleton dude thinks young black kids shouldn't get paid. OMG, can we really still be doing this in 2017?

I was happy to see some college coaches quoted as approving kids going straight from High School to the Lig:

"I think they should be able to go right from high school if they want to," Iowa State coach Steve Prohm said. "I think it's working fine on our end. Just a part of the business. I would love to have a couple [one-and-dones]."

Seven Kansas players have turned pro in the one-and-done era under Self. Most contributed to his current streak of 13 consecutive Big 12 championships. Self says most also were ready to play in the NBA after high school.

"Andrew Wiggins, Josh Jackson ... I would say they were ready to be paid out of high school," he said. "They were certainly ready to be paid, without question."


It's a small scale tragedy, but a tragedy nonetheless what happens to these kids. Can't go make money, so they play in college where the vast majority are not getting any realistic education, until Adam and the players' association say they can play in the NBA.

Friday, June 02, 2017

Books Cost Too Much! So I wrote a blogpost.

Academic textbooks are really expensive. The current version of Mankiw's principles text is $186 in hardcover.  I'll go out on a limb and conservatively guess that the marginal cost of producing another copy is no more than $25. Of course there are fixed costs (like N. Greg's reported $1,000,000 fee), but.........

Here's what I think is going on.

Textbooks are a type of network good. The more people use a given text, the more others will also want to use it. So the market approaches winner take all with just a few books selling well and making profits. This is probably true of all kinds of books.

Weirdly, the books that don't sell well (the losers) almost have to price high as they are trying to cover their fixed costs over a small sales volume.

The winners don't have to price high to break even, but given that demand is high and relatively inelastic, why shouldn't they? Indeed they do, and with gusto.

So the losers price high to try and stay in business and the winners price high because the network good nature of the market gives them a serious degree of market power. In a market like this, a loser cutting price is quite unlikely to receive much of an increase in market share

So how to fix the problem?

Doubly weird is the tried and true idea of "more competition" in form of a larger number of texts on the market probably isn't really going to help as will just create more losers struggling to stay alive. Competition here is "for the market" rather than "in the market" to use the terms of Cowen and Tabarrok's expensive text.

Another market solution is resale, which happens on a very large scale already. Amazingly, new text prices would likely be even higher without this form of competition, where old copies of the winning books compete against new ones. Naturally, publishers don't like resale and respond by pumping out new editions on an inefficiently fast schedule. But resale definitely is a check on new text prices already. If we want them lower, it's not a new tool to use. It's also true that resale doesn't only work to lower new prices. The fact that there is a resale market for a product likely would increase the demand and raise the price for that product. So, it's complicated.

If increased competition in terms of more textbooks is unlikely to help, and resale has already done what it can do, then what?

Well, Facebook is a network good and it's free to the end users, so maybe allow textbooks to contain paid ads? There's a free market solution for you folks!

There's another type of industry with high fixed costs and low marginal costs; natural monopolies like utility companies. We generally regulate them with the alleged aim of getting the outcome to be close to average cost pricing. Maybe the same should happen in the textbook market?

If we don't like a regulatory response, then it's going to take a big decline in demand to get prices down. There is some evidence that some schools, community colleges in particular are going to open source texts or even abandoning the use of textbooks at all. This kind of makes sense as their customers are the most price elastic ones.

Oh pity the poor publishers. If only they could price discriminate better!

So don't worry, Mankiw's days at the top are numbered (anybody remember Lotus 1-2-3?), but the new best seller probably won't be any cheaper.

Sunday, April 16, 2017

Dear Angus, Dear Angus...

I didn't think crap like this still happened (if crap like this can happen, why was Jared Kushner even born?), but here are actual questions published and answered by "Dear Abby". Only this time, Dear Angus answers instead.

DEAR ABBY: I have been with my husband for 12 years, married for three. I had an affair a little over a year ago that he found out about. He has let me back into the house, but he demeans my character at every opportunity. I don’t fight back because I know I am the cause of his pain.

We have a 3-year-old daughter, and I am now six weeks pregnant with his child. I do not want to argue with him, because if I had been a better wife, he would not be so angry. But the hurt I feel from his words over the past months is weighing heavy on me, especially with my new hormones.

I’m holding it in, but should I leave? Become a single mother? How can I get him to a counselor? — NEEDS COUNSELING

Well it seems that he likes you enough to sleep with your cheating ass. So I'm sure everything will peaches and cream once the baby is born (assuming it looks like him). If not, just get your cheating ass pregnant again!

DEAR ABBY: I have been with the same man for six months. He has been separated from his wife for 10 years — but not legally.

When he finally decided to tell her there is someone else and he’s moving on, she went crazy. She said she wants alimony and half of everything, plus the house will have to be sold because she will not allow “the new woman” to live in “her” house.

It’s been a month since he told her. We talked to a lawyer about a divorce, but all he is worried about is paying alimony and losing the house.

I am getting sick of hearing about it. All he keeps saying is, “I love you, but I don’t want to lose my house or pay her money.” What should I do? — FIGHT OR FLIGHT IN MASSACHUSETTS

I really don't see the problem here. After all, he loves you! I'm positive that any day now he'll pay the money and get the divorce. Oh and can we talk about what a B*(&H the wife is? What kind of jerk want's half the marital assets in a divorce? She is so damn selfish! She's bliss-blocking you.

DEAR ABBY: We recently lost one of our cherished pets, our oldest cat, Mandy. We never had children, so our pets ARE our children.

I get that people who have never had pets don’t understand the joy and unconditional love they can bring. But I don’t understand why people we thought were close to us haven’t acknowledged our loss in any way. Some of them have — or had — pets at one time.

A few did send cards or emails, and they were so appreciated. Their kindness will never be forgotten.

Mandy wasn’t sickly. She just stopped eating one day. When we took her to the vet a few days after trying everything we could think of, the diagnosis was kidney cancer. A couple of days later we had to make the heartbreaking decision to put her to sleep.

My question is, am I expecting too much of people? After all, you wouldn’t ignore the death of a human child.
I’m not only disappointed but resentful that these so-called friends and family don’t seem to care. I suppose to some Mandy was “just a cat.” But to us, she was our beloved furry child and we are devastated.

Please inform people that a kind word or short note would mean the world to people like us who are suffering real grief. — DEEPLY GRIEVING IN ILLINOIS

What a bunch of ASS*&%(S! I would suggest kidnapping their children, mailing them death threats, and then make a point of ignoring their pain when they tell you about it. It's for their own good, really.

Wednesday, April 12, 2017

What I'm listening to

So people, here are the new (at least to me) bands that are floating my boat this spring

1. Pinegrove. I swear I've listened to "Old Friends" like 50 times and still get goose bumps when I hear it. Amazing.

2. Diet Cig. LOL for the name, OMG for the toonz.

3. Chastity Belt. These guys have been around a while but are new to me. Love it!

4. Whitney. Great pedigree on this band (Smith Westerns, UMO). Different from them though and it surprises me how much I like it!

Monday, April 10, 2017

In defense of United Airlines

OK people, before we get started, yes, UA should have raised the compensation until they got enough volunteers. And yes, the cop should have found a less brutal way to get the one pax who wouldn't comply with getting bumped off the plane. That cop should be in big trouble.

Now let's get to it.

Can we talk about the recalcitrant pax? What a jerk! Refuses to comply with an entirely legal request from the flight crew. Holds up the plane and all the other passengers because he is too important to get bumped. Refuses to comply with the cops. Causes hours of delay and screws things up for potentially 100s of people down the line.

He's the villain in this case, not UA.

Again, don't get me wrong. I don't like UA. Don't like flying them. And they should have raised the compensation to clear the market.

But, the noncompliant pax here can only be described as a mega-douche.  This incident has no bearing on my decision of whether or not use use UA when I travel. I know I can get bumped, and I'm prepared to live with it. Everyone should know this: you can get bumped even if you are a "VIP".

Plus I already know the cops these days are likely to do any old thing to us citizens, so in no way would I escalate any airline pissing contest to a cop pissing contest.

That guy screwed over hundreds of people I guess because he believed the law didn't apply to him.

He's the opposite of a hero.

Monday, February 13, 2017

Why I love New Mexico, reason #486

People there like good music!!

 From Forbes, here's the favorite vinyl of 2016 in New Mexico:

New Mexico:

1. Random Access Memories – Daft Punk

2. Violent Femmes – Violent Femmes

3. Kind of Blue – Miles Davis

Plus we are small enough that George RR Martin can literally solve all our problems, so that's cool too......

Compare to Oklahoma where the favorite record of 2016 was Fleetwood Mac's Greatest Hits.

By the way, over 13 million vinyl albums were sold in 2016.