4653 stories

Nobody likes a freeloader—including four-year-old kids

1 Share

It may come as a shock to parents of young children, but preschoolers are more cooperative than we realize.

In a novel study to find out how early our instinct for cooperation begins, Yale researchers performed an experiment with kids between the ages of four and 10. The goal was to find out how kids felt about “free riders”—people who fail to contribute to a common project, but reap the benefits from it.

The result? Starting as young as four, kids turn out to dislike free riders intensely, “punishing” those who freeloaded even if they had good reason not to contribute.

“Kids have a pretty strong set of pro-social intuitions around fairness and cooperation, and the need to contribute to larger public goods,” said Yarrow Dunham, assistant professor of psychology and senior author of the paper published in Psychological Science.

Humans are typically cooperative beings, often making sacrifices small and large for the common good—from housesitting for neighbors when they’re out of town to volunteering at schools or stopping at pedestrian walkways. Psychologists have long debated whether we are born this way or socialized to adopt a team-oriented mentality.

Dunham and his colleagues set up six experiments to see how kids felt about freeloaders, all scenarios in which the kids would not suffer directly the effects of freeloading directly but would observe and evaluate the impact it had on others. They did this to isolate whether kids hate freeloading because it hurts them in particular, or whether kids think freeloading is inherently a bad thing.

The kids were presented with some variety of two scenarios: There are four members in a group, and if they all contribute, they get a bigger reward. In one scenario, they contribute plants and the reward is a basket of tomatoes; in another, they contribute chocolates in a “piggy bank,” with the reward being a cake. Children uniformly disliked those who did not contribute and were willing to give up their own rewards to punish the non-contributors. And younger children disliked the freeloaders even more intensely than the nine- and 10-year-olds.

The researchers altered the conditions to test whether kids might dislike freeloaders not because they took advantage of others, but because they were non-conformist; and again to see if the reason for kids’ distaste might be because they were worried freeloaders would force the group to have a worse outcome. No matter what, the kids still dislike the quality. In other words, the researchers found that negative attitudes toward free-riders was because children dislike the concept of freeloading, rather than the outcomes it’s associated with.

“Even young children expect cooperation and are willing to work to sustain it even at a cost to themselves,” Dunham said. “I find this very positive. The seeds that sustain cooperation seem to emerge early on, and while as a society we need to sustain and nurture these values, we may not need to instill them in the first place.”

Dunham said he ran the study because he was curious about the origins of “the incredible cooperative feats of humans.” He left optimistic. “These studies provide evidence that young children have a strong normative expectation of collaborative group behavior and are willing to pay costs to enforce it, and thus from early in development take costly action to facilitate multiparty cooperation even when their own payoffs are not involved,” the study said.

So we may think of our preschoolers as pretty selfish, since in life, they tend more toward freeloading. (So many demands for toys and cookies!) But small children want to be helpfuland they want to cooperate, too. “If you construct these scenarios, they do have strong intuitions about the right thing,” Dunham said.

Read the whole story
Share this story

An oncologist explains how to deliver bad news

1 Share

Everyone has to deliver bad news now and then, but as a radiation oncologist, doing so is a core function of my job.

Many published pieces in the medical literature describe effective strategies. A number of them have catchy mnemonics like SPIKES or ABCDE (physicians love mnemonics). And they all generally come down to the same pieces of advice: Build a relationship. Speak frankly and honestly, but with compassion. Show empathy.

I agree with all of these concepts, but I have also learned more practical steps for delivering bad news throughout my 15 years in practice.

Telling a patient that her PET scan shows cancer progression or that she will need further radiation after surgery is worse news than most people deal with at work, but I believe that the best practices for doing so could also be applied to delivering other types of bad news compassionately. Whether announcing layoffs, telling an employee he was passed over for promotion, or delivering bad quarterly sales results—some thought can help make the experience less painful for all involved.

Plan ahead: Before an appointment with a patient, I review the case, films, and facts. I speak to any other physicians involved whose opinions I need to make a final decision. All of this needs to be done before the appointment, so that the meeting is smooth and the advice is clear.

I also make sure that the meeting takes place in an environment where it’s comfortable to talk and have tissues ready in case of tears.

Deliver the news with minimal preamble: Don’t keep someone waiting for bad news. A short, few-word phrase to prepare people to hear bad news is appropriate, but do not give a long preamble before coming to the point. That preamble is just torturous time for the patient, and subconsciously or consciously, you are using it to stall.

Be direct: Try not to couch it in jargon or technical language that may cause the patient to stumble. Choose your words carefully and do not equivocate where it’s not appropriate.

Pause: In my experience, allowing for a pause after the delivery of bad news is a wise practice. People will need a few moments to compose themselves and their thoughts. A short period of silence can be helpful. Additionally, most people will not hear anything you say for at least a few seconds, and up to a few minutes, after hearing significant bad news. Allow them a moment for that to pass. Realize your desire to move on quickly after giving bad news is the flipside of the desire to give a long preamble. It is a way to get your own painful experience (being the deliverer of bad news) done with as quickly as possible, when at that moment the focus needs to be on what’s best for the patient.

Express empathy: There are many ways to express empathy both verbally and physically after giving bad news. My advice is to keep your expression simple at this time. Don’t presume you know what someone is thinking or feeling. A simple “I’m so sorry,” a gentle hand on the back or leg, and a long, deep hug are all meaningful. Know yourself and the patient, and be genuine. This is the time to really, deeply mean what you say or do, and to let that feeling show through. Give the patient a chance to express their emotions, feelings, and fears. Listen closely to what he or she says.

Answer questions: Give an opportunity for the patient to ask questions, and answer those questions with honesty.

Know the next step: People usually handle bad news better than you think they will, but they want to know what’s next. Try to have concrete plans in place. If a patient needs to see a surgeon, I’ll let him know I’ve already spoken to the surgeon ahead of time and have set up a visit. If there are no more good treatment options and hospice is the next best step, I’ll let him know we will have the hospice ready to meet him in the next 48 hours. Whatever it may be, don’t give bad news and then send the person away. Give bad news, and then provide a plan or options for how to address the future.

Once you’re alone, give yourself time to process: While you initially need to resist doing things that will ease the burden on yourself at the expense of the person receiving bad news, you should embrace those impulses after the person has left. Remember that anger is the second Kubler Ross stage of grief and accept that if the recipient of bad news got angry or even expressed hatred towards you, that’s often a normal reaction and not a reflection on you. Take time to reflect and sort through your own emotions of fear, guilt, and pain.

Andrew Neuschatz is a physician and partner with Arizona Oncology in Tucson, AZ.

Read the whole story
Share this story

Factor Investing Insights You Won’t Hear from Fama and French

1 Share

Factor investing research has a long storied past. Fama and French’s 1992 and 1993 papers arguably put factor investing “on the map,” but truth be told, factor investing is an old topic with roots grounded in the so-called arbitrage pricing theory. We have a longer piece on the history of factor investing here.

There is a monster empirical research effort to determine which factors describe stock returns in the historical data (many consider this a data-mining effort).  A criticism of the “let’s see what the data says” approach is that these empirical models often lack a clear economic foundation or theory.  In other words, if sunspots predict stock returns in the data, we should not claim victory. Or for a more realistic example, instead of focusing on SMB (size) and HML (value) because they “work,” we should focus on using SMB and HML in our factor models because they are tied to economic theories that are grounded on first principles. Many people have gone this direction, but the theoretical models they’ve developed, while sensible, do a terrible job describing reality (e.g., CAPM).

Bottom line: factor models that describe the movement of expected stock returns should be tied to economic foundations AND they should do a reasonable job explaining the patterns in historical data. This combination helps alleviate data-mining concerns. Simple in theory, but building a reasonable theory and having it roughly match historical data is quite difficult!

Lu Zhang and his colleagues have put in a good faith effort to build factor models that are backed by economic models and do a great job describing historical stock return patterns. One can argue that their theory is wrong, but at least they are generating a theory that motivates their factor model. And because we believe Lu Zhange et al. are arguably on the cutting edge of factor investing research, we’ve focused on covering their research (here, here, and here) and interviewing Lu to get his insights.

The piece that follows related to their latest research papers, “Which Factors” and “q5.

Two New Papers: Which Factors and q

Wes: Lu, you and your colleagues have been busy. Really busy. Since we covered your last research paper, “Replicating Anomalies,” which was a monster research effort already, you and your team have managed to grind out two new papers: “Which Factors” and “q5.” Where do you get the time?

Lu: We have a very efficient team. Kewei Hou, Chen Xue, and I have been working closely together since 2011. Our q-factor paper is the most cited article published at Review of Financial Studies in 2015. Our work on “Replicating Anomalies” has also been well received, in both industry and academia. Haitao Mo joined us a couple years ago. All three of my collaborators are superior empiricists. We go through internal, independent replications every time we report new results in a working paper. We model after Fama and French on the quality of empirical execution. We feel comfortable with our intellectual product.

Wes: Walk us through the key differences between the two new papers.

Lu: We have just circulated the updated drafts for both papers, now dated July 2018. Check them out on my research page or SSRN page. The title for “Motivating Factors” has changed to “Which Factors?” to reflect the content of its revised draft. The two papers are complementary.

The key message for “q5” is to augment the q-factor model with a new expected growth factor to form the q5 model. Using the vast set of 158 significant anomalies from “Replicating Anomalies” as testing deciles in a large-scale empirical horse race, we show (i) the q5 model is the best performing model; and (ii) the q-factor model already compares well with the Fama-French 6-factor model and other models.

The focus of “Which Factors?” is on spanning regressions and on the conceptual foundation of factor models. We reproduce and replicate the Stambaugh-Yuan (2017) 4-factor model and the Daniel-Hirshleifer-Sun (2018) 3-factor model. Reproduction means we follow their exact procedures, and replication means we follow the standard approach per Fama and French (1993). We find that the Stambaugh-Yuan and Daniel-Hirshleifer-Sun models are both sensitive to their construction. Using the standard approach, which ensures that we compare apples with apples, weakens their performance, often substantially.

Conceptually, the paper on “Which Factors?” raises some concerns on motivating the Fama-French 5-factor model from valuation theory. In particular, consistent with the investment CAPM, valuation theory also implies a positive relationship between the expected investment and the 1-period-ahead expected return. As such, CMA can only be motivated from the market-to-book term in the valuation equation via the economic linkage between investment and value, which is, in turn, a key insight from the investment CAPM.

Digging into the “q5” paper

Wes: Let’s start with q5. I got to ask a non-quant question first – what’s with the funky name and the superscript character? Any backstory on that?

Lu: The funky name is Kewei’s idea. We played with “Q5” for a while like the Audi SUV, which I used to drive. But “q5” seems cool. I like its nerdy feel.

Wes: Okay, digging back into the details. Tell us a little more about your expected growth factor. How’s it measured?

Lu: We measure the expected investment growth as the fitted component from monthly cross-sectional regressions of the 1-year-ahead investment-to-assets change on the log of Tobin’s q, operating cash flows, and the change in return on equity in the prior 120-month rolling window. Then we construct the expected growth factor from monthly, independent 2-by-3 sorts on size and the expected growth.

Wes: Is there any way to make a simple version of the measure that doesn’t require a rolling regression estimate? For example, for the forecasted investment growth variable, could someone use recent TTM (trailing twelve months) ROE growth and be in the ballpark?

Lu: The change in return on equity is included when we forecast investment growth in the latest draft of the q5 paper. We include this variable to capture some of the short-term dynamics of investment growth. However, we show that the ratio of operating cash flows to book assets measured as in Ball, Gerakos, Linnainmaa, and Nikolaev (2016) has stronger predictive power for investment growth.

This result makes economic sense. The Ball et al. measure takes away accruals, but include R&D expenses. All else equal, high accruals mean low cash flows available for investments going forward, and thus low expected growth. High accruals also mean high past growth, which implies low expected growth, due to the strong mean version in growth rates. Finally, high R&D expenses mean low current return on equity, due to the standard accounting practice, but induce high expected growth going forward. In fact, cash flows also take care of (to some extent) the accounting problem for other intangible investments such as advertising, employee training, information technology, organizational capital, etc.

Wes: You mention that firms with high expected investment growth should earn higher expected returns. Can you elaborate?

Lu: Sure. The intuition is analogous to (but a bit more involved than) that behind the positive relation between profitability and the expected return in our q-factor paper.

With only two periods, the investment CAPM implies that the discount rate equals the expected profitability divided by the marginal cost of investment (which rises with current investment). As such, high profitability relative to low current investment means a high discount rate, which offsets the high expected marginal benefit of current investment (expected profitability) to keep the current investment low.

In a multiperiod framework, the expected marginal benefit of current investment also increases with the expected next period investment. At the end of the next period, a firm is left with the capital generated from current investment, net of depreciation. The market value of that leftover capital is the present value of all future cash flows that it can generate (what economists call marginal q). The presence of that capital saves the firm the exact amount that equals the marginal cost of next period investment (which, again, increases with next period investment). In all, high next period investment relative to low current investment (high investment growth) means a high discount rate, which counteracts the high expected marginal benefit of current investment (the part from the expected next period investment) to keep the current investment low.

Hope the intuition is penetrable. The math equation is very clear. Words can be ambiguous.

Wes: Barillas and Shanken try and identify the “best” factor model via a Bayesian lens. They find that the 6-factor model of Mkt, SMB, ROE, IA, HMLm, and UMD works the best. In other words, value and momentum are not redundant. The Fama French 5 Factor model gets demolished, and while your HXZ 4-factor model does better, their 6-factor version does better. How do you address this research?

Lu: Sure. First, the Barillas-Shanken model is purely statistical, with little economic foundation. Our q and q5 models stand out in the factors debate in that our models are the only ones with close ties to first principles in economic theory. The anomalies/factors literature is heavy-duty empirical, but we show there is a lot of economics as well.

Second, Barillas and Shanken end up picking our investment and Roe factors over Fama and French’s CMA and RMW. In fact, the truth is even stronger. Figure 1 reports page 6 in Jay Shanken’s discussion slides on our working paper on “A Comparison of New Factor Models,” from which our paper “Which Factors?” descends, in the 2015 Arizona State University Sonoran Winter Finance Conference. Shanken said in the conference that in his work in progress (at the time) with Barillas, they assigned a 50-50 diffuse prior to the q-factor model and the Fama-French 5-factor model, but found that the posterior probability of the q-factor model being true is 97%, and that of the 5-factor model is only 3%. I was happy to learn about the evidence and was hoping that Shanken’s credibility could give us a boost. However, when their working paper started to circulate in May 2015, I could not find the 97% vs. 3% evidence reported anywhere. I pointed out to Shanken, when he gave a seminar at Ohio State in November 2015, that this evidence should be reported, especially in a paper titled “Comparing Asset Pricing Models.” Still, this evidence is nowhere to be found in their published papers in 2017 Review of Financial Studies or 2018 Journal of Finance.

Figure 1. Slide 6 in Jay Shanken’s Discussion on “A Comparison of New Factor Models” at the 2015 ASU Winter Finance Conference

Finally, the Barillas-Shanken tests are performed with only 11 factors, and they even argue that testing portfolios are irrelevant. This is not what we see in our tests. Using the set of 158 significant anomalies, by far the largest among the papers on competing factor models, we show in the q5 paper that the Barillas-Shanken 6-factor model does not perform well. In particular, HMLm and UMD are strongly negatively correlated, pushing up the UMD loadings in factor regressions of annually sorted value-minus-growth anomaly returns. In contrast, this empirical difficulty is entirely absent from the Fama-French 5-factor model and the q-factor model.

Wes: So your expected investment growth factor seems to capture the higher frequency HML factor and the momentum factor? Can you explain the intuition behind that? What are the practical implications?

Lu: Not exactly. The Roe factor in the q-factor model already subsumes UMD, see Table 1 in “Which Factors?” which we first circulated back in September 2014. Momentum has never been a problem for us. Table 1 in “Which Factors?” shows that from January 1967 to December 2016, UMD earns on average 0.65% per month (t=3.61), but its q-factor alpha is only 0.12% (t=0.5).

However, neither the q-factor model nor the q5 model can explain Cliff’s monthly formed HML factor, HMLm. Table 4 in “Which Factors?” duly reports this evidence. However, Cliff’s insight on using more up-to-date information in monthly sorts can also be applied to the q-factor model, in which the size and investment sorts are annual, and only the Roe sort is monthly. Nothing in the investment theory says that we cannot use monthly sorts on size and investment. Once we reconstruct the q-factors with monthly sorts on size, investment, and Roe, we find that the q-factor model and the q5 model both explain HMLm. Panel C of Table 7 in “Which Factors?” reports this evidence.

The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees, and one cannot invest directly in an index.

Wes: What are the practical implications of this finding? How could an investor use this research to form smarter investment portfolios? Let’s say I’m trying to compound my wealth at the highest rate possible – what’s my portfolio look like?

Lu: The q5 paper shows that the q5 model is the best performing model in explaining a vast set of anomalies, and the q-factor model already compares favorably with the Fama-French 6-factor model. This means that the q-factor model and the q5 model can be used to control for “risks” in the sense of common variations in practice.

In terms of compounding wealth, buy stocks with low investment, high Roe, and high expected growth (measured with high cash flows and fast improving Roe).

“Which Factors?” paper

Okay, we’ve covered a lot already. We won’t rehash a lot of territory on your q5 paper in this section, but I did have a few questions related to the “Which Factors?” paper.

Wes: You state that the Fama and French 5-factors are not tied to valuation theory. Can you explain? Also, why should practitioners care? Also, let’s talk about CMA, the investment factor used in the Fama and Fench model. Basically asset growth for all intent and purposes. You highlight that using past investment as a proxy for expected investment doesn’t work. Explain. And why should an investor care?

Lu: These two questions are closely related, so I address them together. There are at least two difficulties with motivating the Fama-French 5-factor model from valuation theory. First, valuation theory, as formulated by Fama and French (2015), is only about the internal rate of return, which is different from the 1-period-ahead expected return, both theoretically and empirically.

More important, Fama and French motivate CMA via two steps. First, they argue that the expected investment is negatively correlated to the expected return. Second, they use current investment (asset growth) as the proxy for the expected investment. We raise concerns on both steps. Reformulating valuation theory with the 1-period-ahead expected return, we show that the relation between the expected investment and the expected return is positive per the valuation equation, consistent with the investment theory. In fact, this positive relation from the investment theory motivates our expected growth factor in the q5 paper. Moreover, we document that unlike profitability, firm-level investment is not persistent at all, casting doubt on using current investment as a proxy for the expected investment. In all, in terms of theory, only the investment CAPM can give rise to the investment and profitability factors.

Why should investors care about economic theory? Well, to guard against data mining. The investment, Roe, and expected growth factors are motivated from the first principles of real investment. As long as public traded firms are maximizing their market value of equity in their capital budgeting, at least to the first order importance, we should expect the q and q5 factor premiums to persist in the long term.

For a long time, the theory has a bad rep among investment managers. The likely reason is that the consumption CAPM, which is the dominating framework in academic finance, does not work in practice. The investment CAPM is a new breed of theory, which seems to explain the data well.

Wes: This paper seems to focus on the Fama French and the Stambaugh-Yuan factor models. One of the newer “factors” out there seems to be the low volatility or beta factor. Care to comment on this vein of research?

Lu: In the latest drafts on “Which Factors?” and “q5,” we also include the Daniel-Hirshleifer-Sun 3-factor model. We show that the Stambaugh-Yuan and Daniel et al. models’ performance is sensitive to their nonstandard factor construction. In particular, Stambaugh and Yuan use the NYSE-Amex-NASDAQ breakpoints with the 20 and 80 percentiles, and the Daniel et al.’s breakpoints are likely even more extreme, especially their financing factor. In contrast, the standard approach calls for the NYSE breakpoints with the 30 and 70 percentiles.

In other words, Stambaugh and Yuan, as well as Daniel, Hirshleifer, and Sun, are comparing apples with oranges. Their testing portfolios are formed with NYSE breakpoints and value-weighted returns, following our q-factor paper. But their factors rely on stocks in the more extreme tail distributions of the underlying characteristics.

We are not exactly sure how reliable the low volatility anomaly is, see the latest draft of our “Replicating Anomalies” (dated July 2018). About the beta anomaly, we have tried the betting-against-beta factor from the AQR Web site on the q-factor model at one point. It turns out that our Roe factor largely subsumes it. In general, as shown in our “Replicating Anomalies,” fundamentals seem to be much more important than frictions in the cross-section.

Wrapping up

Wes: A final question on interpreting factor models. Shri Santosh and colleagues highlight that it may be difficult to differentiate between the success of a factor model based on neoclassical factors or something as simple as sentiment. What do you think about this line of thought?  

Lu: I agree. In fact, I wrote an earlier paper titled, “The Investment Manifesto,” published in Journal of Monetary Economics in 2013, in which I have made similar points. Basically, I view factor regressions and cross-sectional regressions are two different but largely equivalent ways of summarizing correlations in the data. No causal inferences should or can be drawn on whether the empirical relations are due to risk or sentiment. In fact, I even wrote in that paper that the concept of “risk” is an outdated notion from the consumption CAPM, which ignores the supply side of asset pricing altogether. In particular, using characteristics to forecast returns is perfectly consistent with the investment CAPM, which provides the supply theory of asset pricing.

Relatedly, I do not think “sentiment” is simple at all. It is nebulous, lacking specifics. In contrast, I think the investment CAPM is very concrete. A single equation tells us to focus on investment and Roe. The expected growth is in the equation, but less specific, alas, because one has to take a stand on what instruments to use to forecast future investment growth.

Wes: Final question: Do you know if anyone maintains a taxonomy of factor models? It would be great to have a one-stop shop on some infographic that outlines all these models.

Lu: I think the q5 paper fits the bill. Our paper on “Replicating Anomalies” takes care of the left-hand side of factor regressions, and together with “Which Factors?” the q5 paper takes care of the right-hand side.

Wes: Lu, this was awesome. Thank you very much for the insight and education. Lot’s to think about!

  • The views and opinions expressed herein are those of the author and do not necessarily reflect the views of Alpha Architect, its affiliates or its employees. Our full disclosures are available here. Definitions of common statistics used in our analysis are available here (towards the bottom).
  • Join thousands of other readers and subscribe to our blog.
  • This site provides NO information on our value ETFs or our momentum ETFs. Please refer to this site.

Read the whole story
Share this story

Thursday Solution

1 Share

Here are the answers to yesterday’s puzzle. The first correct solution came from our commenter Leo (comment #18 on yesterday’s post).

The assumptions of the problem were: Everything I say out loud can be deduced from my axioms. My axioms include the ordinary axioms for arithmetic, among other things. And I recently said out loud that “I cannot prove that God does not exist”.

The questions were: Can I prove there is no God? Can I prove there is a God? And is there enough information her to determine whether there actually is a God?

The answers are yes, yes and no: Yes, I can prove there is no God. Yes, I can also prove there is a God. And no, you can’t use any of this to determine whether there is a God.

To explain, I’ll use the phrase “logical system” to refer to a system of axioms sufficiently strong to talk about basic arithmetic (and perhaps a whole lot of other things), together with the usual logical rules of inference. It’s given in the problem statement that I am a logical system.

Here are two background facts about logical systems:

A. An inconsistent logical system can prove anything at all. That’s because it’s tautological that if P is self-contradictory, then any statement of the form “P implies Q” is valid. If I’m inconsistent, that means I can prove at least one statement (call it P) that’s self-contradictory. Then if I want to prove, say, that the moon is made of green cheese, I note that:

  • I can prove P
  • It’s tautological that “P implies the moon is made of green cheese”
  • Therefore I can conclude by modus ponens that the moon is made of green cheese.

B. No consistent logical system can prove its own consistency. This is Godel’s celebrated Second Incompleteness Theorem.

Now here’s the argument:

1) I’ve asserted that I can’t prove that atheism is true.

2) I only assert things I can prove, so I must be able to prove that I can’t prove atheism is true.

3) Therefore I can prove that there’s something I can’t prove.

4) Therefore I can prove that I am consistent (because if I were inconsistent, I’d be able to prove everything — that’s background fact A).

5) Therefore I am inconsistent (because no consistent system can prove its own consistency — that’s background fact B).

6) Therefore I can prove anything (that’s background fact A again). (More precisely, I can prove anything I can state.)

In particular:

Yes, I can prove there is no God (because I can prove anything!).

Yes, I can prove there is a God (because I can prove anything!).

No, you can’t use any of this to learn anything about whether there is a God (or any other aspect of reality) because you already know that I can prove anything — so my (in)ability to prove something carries no information whatsoever.

More generally, the moral is this: As a matter of pure logic, If there is anything that you can prove you can’t prove, then you can prove anything.


Read the whole story
Share this story

“Eighth Grade” shows the difference between how the US and Europe think about teens and sex

1 Share

Eighth Grade is a highly-acclaimed coming-of-age movie about a 13-year-old American girl enduring the trials and tribulations of modern adolescence. But while teenagers in the US might well relate to the movie’s heroine, they won’t be able to see the movie in theaters—unless they’re at least 17 or accompanied by a parent or guardian. That’s because the Motion Picture Association of America (MPAA) gave the film an R rating for “language and some sexual material.”

There aren’t many other ratings to compare that against. The movie has only come out overseas in two countries–the United Kingdom and Canada. But in Canada, Eighth Grade was given a 14A rating, meaning that everyone older than 14 can see it without an adult. Meanwhile, the British equivalent of the MPAA, the British Board of Film Classification, hasn’t yet rated Eighth Grade, but it’s a good bet that the movie be rated more leniently in the UK.

The discrepancy in Eighth Grade’s Canada and US ratings is symbolic of the difference between the US and the rest of the world, according to the movie’s director Bo Burnham. “There seems to be a strange double-standard between sexuality and violence,” he tells Quartz. “It’s a little weird how much violence you can have in a PG-13 movie.” That’s because, as Charles Bramesco argues in a recent piece for Vox, movie ratings reflect what a culture deems acceptable content for children. And the US and Europe are on very different pages about what they view as child-appropriate.

What is the MPAA?

In the US, the MPAA is the body that reviews and rates most movies and trailers according to a classification grid, ranging from General Audiences (G) to Parental Guidance Suggested (PG), Parents Strongly Cautioned (PG-13), Restricted (R), and No One 17 And Under Admitted (NC-17). In the first half of the 20th century, movie ratings were really a form of censorship, when the fear that movies were a corrupting influence on young children spurred the creation of the “Hays code,” under which filmmakers were obligated to minimize or avoid depictions of everything from “lustful kissing” and interracial relationships to divorce and toilets. The Hays code gained traction when the Legion of Decency, an organization created by the Catholic Church, signed onto it and began assigning ratings to films based on whether they were “morally objectionable” or not.

After World War II, the Hays code started to fade. In 1952, the Supreme Court ruled that movies were protected by the First Amendment. That decision ushered in the movie rating system we know today: in 1968, Jack Valenti, a special assistant to Lyndon B. Johnson, became president of the MPAA, and instituted a voluntary movie rating system based not on moral censorship, but rather on information for parents about what their kids could expect to see onscreen.

Today, the Classification & Ratings Administration, part of the MPAA, issues movie ratings, based on the vote of an independent rating board made up Los-Angles based parents who “have the capacity to put themselves in the role of most American parents.” In effect, the board acts as a kind of moral and ethical arbiter, making judgments limiting the audience for a given movie based on a set of criteria that includes nudity, alcohol or drug use, coarse language, and violence.

American vs. European views of sex and violence

The MPAA is generally considered to be more lenient towards violence in evaluating ratings for children, but tougher on sex and non-sexual nudity (frontal male nudity in particular). As Bramesco writes for Vox, citing the 2006 investigative documentary This Film Is Not Yet Rated, “Sex scenes are picked through with a fine-toothed comb, any detail—a wiping of the chin, a moan too emphatically acted, any maneuver beyond the most vanilla standards—sufficient to bump a film up to the R zone and limit its reach.” It’s hard to predict what “vanilla” will mean for the MPAA; the rules are so opaque that directors like Bo Burnham often have to guess what it is about their movie that earned them an R rating.

The MPAA’s reluctance to let children see sex on screen is strange, given that sexual experimentation is a normal and important part of the adolescent experience. That’s part of the rationale behind Eighth Grade director Burnham’s decision not to edit the film down to a PG-13 rating by cutting out certain swear words or a vaguely sexually suggestive scene where Kayla Googles how to give a blow job, and subsequently gets grossed out. “The movie’s not exposing them to anything they’re not already aware of,” Burnham tells Quartz. But movies, at the very least, portray the trials of growing up in a more responsible way than what kids have access to on the internet, he said.

In general, the US does tend to rate sexuality more harshly than violence, and that is pretty much flipped everywhere else in the world,” Betsy Bozdech, the executive editor of ratings and reviews for Common Sense Media, a non-profit that rates and reviews movies to help parents make decisions about the content their kids watch, tells Quartz. For her part, Bozdech says that Common Sense Media gave Eighth Grade a 14+ rating, and that both parents and kids on their site gave the movie a 12+ rating. “I hope that parents will take their kids to see it,” she said.

For many European movie ratings agencies, including the British Board of Film Classification, scenes that depict sex are deemed more acceptable for adolescents. European attitudes hold that sexual exploration is a normal part of growing up, and that kids should be allowed to see it on screen. That’s part of a broader difference between how Americans and Europeans view sex. For example, a 2013 Pew poll found that 30% of US adults still think that sex between unmarried adults is morally unacceptable, but in Europe, only 6-13% of respondents thought it was unacceptable.

What’s behind the MPAA’s movie ratings?

There are some questionable dynamics at stake in the MPAA’s movie ratings. For example, the Classification & Ratings Administration has been accused of being biased against depictions of women’s sexuality and queer sex, much more so than against heterosexual or male sexual pleasure. That bias becomes evident when comparing American and European ratings of movies depicting queer sex: For example, the critically-acclaimed 2013 movie Blue is the Warmest Color had lengthy scenes of lesbian sex between a teenager and her older lover. The film was rated NC-17 in the US, but in France, where the movie was filmed, it was rated acceptable for kids above the age of 12—the equivalent of the American PG-13–by the French National Center for Cinema and Animated Image (CNC). When a Catholic group tried to sue the Ministry of Culture for its rating, saying the movie should not be allowed for children below 16, the country’s highest administrative jurisdiction, the Conseil d’état, ruled in favor of the “12” rating, saying that “Although true that the sex scenes in question, although simulated, present a character of undeniable realism, they are both free of all violence, and filmed without degrading intent.”

Depictions of sexual violence, as in the gay thriller Stranger by the Lake, have garnered more severe ratings from the French agency in the past. But as long as the sex is consensual and violence-free, the French agency typically deems such scenes appropriate for middle-schoolers.

Meanwhile, the MPAA may be prudish about sex itself, but it isn’t at all concerned with the question of how Hollywood perpetuates harmful, sexist ideas about women’s sexuality, according to Peggy Orenstein, a journalist and author of the book Girls and Sex: Navigating the Complicated New Landscape. “The MPAA rating system in no way addresses the sexualization, objectification or marginalization of women in Hollywood,” she said, meaning “to just focus on the depiction of sexual acts and violence is in many ways to miss the forest for the trees.” Orenstein adds, “we have a ridiculously and unquestioned high tolerance for exposure to violence at the youngest ages in media.”

The US rating system’s puritanical attitudes toward sex ultimately does little to protect teenagers—particularly when the MPAA denies adolescents the opportunity to see their own experiences reflected in characters their age, exploring their sexuality in complicated and meaningful ways, onscreen. As Burnham says of the MPAA’s reaction to Eighth Grade, “By not allowing us to portray difficult or strange situations for kids … it’s contributing to the problem.”

Read the whole story
Share this story

Andres Manuel Lopez Obrador Explained for Investors in Mexico

1 Share
Greetings again Macro Man readers! Those that remember my work here will know that I couldn't let the Mexican election pass without some sort of commentary. I decided it would be a little passe to write up 800 words on my opinion. Instead I decided to let a talking rabbit do the work for me. Enjoy!

Read the whole story
Share this story
Next Page of Stories