魔鬼经济学

这个片子最大缺点就是太散，总共有6个导演，5个制作团队来完成:
1. Intro & Transitional Segments (including Real Estate, Parenting, Cheating, Cause & Effect, Incentives) 1 director
2. A Roshanda by Any Other Name 1 director
3. Pure Corruption 1 director
4. It’s Not Always A Wonderful Life 1 director
5. Can you bribe a 9th grader to succeed? 2 directors

下面为主要内容及观感，有兴趣的可以挑相应电影片段来看：
0. Real Estate
Does real estate agent really represent your interest?
Incentives matter!!! If you can figure out what people’s incentives are, you have a good chance of guessing how they are going to behave.

1. Parenting (4’26’’-)
Levitt explains what he does in research: to figure out tricky ways to get a causality. There’s a difference between causality and correlation.

2. A Roshanda by Any Other Name
（莎翁名句：A rose by any other name would smell as sweet.）
（出自原书第六章）
Does name matter? Does name define your destiny?

Resume experiment: African American names vs. White names
View point: Name may not define your destiny but it can dictate the ways other ppl perceive you.

3. Cheating
(3&4出自原书第一章：教师和相扑运动员有何相似之处？)
Teachers could cheat.

4. Pure Corruption (24’20’’-47’48’’)
Corruption in the Sumo World with reference to Bernie Madoff (corruption in the realm of high finances)
        *误译：S.E.C (not S.C.C)
        很像犯罪纪实片，有点故弄玄虚之感。本可以独立成片的题材，过短而语嫣不详。日本文化对于西方观众/读者本身具有神秘吸引力。也提到了日本和美国媒体的self-censorship，有一定自我批判的意味。

        Levitt: The only way to combat corruption is to change rules and undo corruption incentives, unleash more investigative reporters and to develop strong protections for whistle blowers.

5. Cause & Effect （47’49’’-49’10’’）
Does ice cream cause polio?

6. It’s Not Always a Wonderful Life (出自原书第四章)
Reasons why crime rates drop considerably in the 1990s?
观感： 全片最有意思和最令人惊奇的发现。

6 most popular explanations (with authors’ verdict): More innovative policing (0%), harsher criminal policing (30%), changes in the crack market (15%), increase gun control, a strong economy, more police on the street (the last three about 10% in total).
(具体怎么得到这些数字的没有解释，这也许是有经济学家批评两位作者是在描述计量经济学的原因，原书中是不是也没有解释？)

Almost half of the drop left unexplained. Levitt’s answer: Roe vs. Wade (the legalized abortion in 1970s)
Unwanted children were not born…
“Unintended consequence of legalized abortion”

关于这个结论的争论（译自维基百科）：
书出版之后Boston Fed的经济学家指出Levitt原论文中的统计错误，改正这些错误后不能得到堕胎合法化是犯罪率降低原因的结论。Levitt回应，Boston Fed经济学家的改正由于存在严重的衰减偏误（attenuation bias）也是有缺陷的，改正自己原文的错误后，流产和犯罪之间的联系较弱，但仍然统计显著。Boston Fed的那两位经济学家很快提供了反驳：即使使用Levitt建议的方法来分析数据，该数据不显示流产率和犯罪率呈正相关关系。他们很快指出，这并不否定Levitt的论文，并强调由于该数据是如此混乱和不完整，很可能甚至是无法证明或反驳Levitt他们的结论。

7. Incentives
How Levitt failed to potty-train his daughter with her favorite M&M chocolates within three days.

8. Can a 9th grader be bribed to succeed?
Interesting (but costly) experiment again!

观感：This is the most touching part of the movie and it sort of resonates with the segment on Parenting. But again the authors’ view on parents’ role in a child’s growth is ambiguous.