Brian Flaxman: Yes! Economics Did Sway Obama Voters to Trump

October 26, 2017 by Miles Kimball

Source: Brian.Flaxman@colorado.edu — *Source: Brian.Flaxman@colorado.edu*

Brian Flaxman is a student in the Economics PhD program at the University of Colorado Boulder where I am now a professor. He identified a very interesting set of facts about the presidential election that put Donald Trump in the White House. Email Brian with any questions (Brian.Flaxman@colorado.edu). From here on, I'll let Brian speak for himself:

Tuesday, November 8, 2016 is the day that Donald Trump was elected 45th President of the United States. Many people across the country believed that Hillary Clinton’s victory was all but inevitable, adding to the shock and surprise of this momentous event. Because of this, there has been a massive effort underway to try to understand how he could have won. One main school of thought is that Trump’s victory can be mostly contributed to right wing identity politics. The other is that there have been many people in the country who have struggled economically, have legitimate grievances with the current political establishment, and found Trump’s populist rhetoric very appealing.

I in no way would make the claim that right wing identity politics didn’t play a role in this election, and doing so would be foolish to say the least. However, not acknowledging the role that the rise in populist fervor played in Trump’s victory would be equally as foolish. There is plenty of circumstantial evidence to support this. Bernie Sanders received over 44% of the Democratic primary vote when facing Hillary Clinton, a mainstay in Democratic politics for the three decades. Now he is the most popular politician in the country according to most polls, including a Harvard-Harris poll released this past week. The Rust Belt, a part of the country that supported Bernie Sanders and Donald Trump in their respective primaries, was likely the part of the country that contributed to Hillary Clinton losing the electoral college.

When discussing the role that economics played in Trump’s win, the main rebuttal is that “The people who voted for Donald Trump never would have voted for a Democrat, regardless of who it was!” Well, that’s a hypothesis that can be scrutinized empirically by looking at election, demographic, and economic data. We can analyze the effect of the Obama to Trump voters, and more importantly, how right wing identity politics and poor economic outcomes contributed to this effect. With the proper data in hand, we can analyze the contributing factors to this shift.

The data I will be analyzing is county level data from both the 2012 and 2016 elections, as well as economic and demographic data from various points, from 2010-2016. I have data from all counties and equivalents not in Alaska. There are several measures that can approximate the support that moved from Obama in 2012 to Trump in 2016, but the one I will focus on I will call Trump’s “share shift”. This is devised from taking the percentage of votes from Donald Trump, and subtracting off Romney and independent and third-party votes. If we were to hold voting populations constant across the 2012 and 2016 elections, this would be the amount that remains would be the Obama voters that switched to Trump. Even though this is almost certainly not the case, it does describe how a county’s overall support moved from Obama in 2012 to Trump in 2016, and provides a good proxy for Obama to Trump voters.

I will be testing the following four indicators of poor economic outcomes:

1. Unemployment year annually from 2016

2. Unemployment change over the past 20 years.

3. Percent living below the poverty line in 2014

4. Loss in overall population from 2010-2014.

I will test these variables after controlling for white population, Hispanic population, the log of income, female population, population that are veterans, population density, change in voter participation from the 2012 election, and for state fixed effects. Sure enough, negative economic indicators contributed significantly to the shift in votes from Obama to Trump.

When running all four variables at once, they are all positively correlated with share shift, and are all significant with 97% confidence or higher. When each variable is regressed alone with the controls, they are all significant and positively correlated. The cross terms for 2016 unemployment and population loss were both positive and significant in their respective regressions. Furthermore poor economic outcomes are significant when dividing the population into certain subsets, like counties Obama won and the 500 top densely populated counties, and even remained when weighting each county by 2014 population[1].

The conclusion I have reached based on the empirical evidence? Negative economic indicators almost certainly contributed to people shifting away from Obama in 2012 to Trump in 2016, and Trump’s populism was almost certainly one of the factors that helped him win the election.

So how large was the impact of negative economic indicators on per county shift share of votes from Obama to Trump? 2.71% of the vote per county on average. You could say that this doesn’t control for the vastly different sizes in population, except this number only drops to 2.29% percent of the vote when you weight the counties by population, meaning it had a huge impact even when you control for population[2]. The bottom line is that while some of the country’s darker tendencies contributed significantly to Donald Trump being the 45th president of the United States, so too did negative economic outcomes for many voters, contributing in part to the anti-establishment and populist sentiments growing in popularity both here at home and abroad.

Note: If you have trouble reading the tables below, here is a link to the tables in a Word document.

[1] I can provide specific empirical results by request. My email is Brian.Flaxman@colorado.edu

[2] Calculated by taking the coefficients from the main regression with all four negative economic indicators, multiplying by that county’s data, summed together.

Related Reading:

Noah Smith's and Matthew Yglesias's Unpopular Opinions That I Mostly Agree With →

October 25, 2017 by Miles Kimball

Roundtable Discussion of the Reproducibility Crisis and the Proposal to Make Half a Percent the Standard for Statistical Significance

October 24, 2017 by Miles Kimball

If you were interested in my post "Let's Set Half a Percent as the Standard for Statistical Significance," you might be interested in this online roundtable discussion. Here is the announcement:

This Friday, October 27th at noon Eastern time, the International
Methods Colloquium will host a roundtable discussion on the
reproducibility crisis in social sciences and a recent proposal to
impose a stricter threshold for statistical significance. The
discussion is motivated by a paper, "Redefine statistical
significance," recently published in Nature Human Behavior (and
available at https://www.nature.com/articles/s41562-017-0189-z).

Our panelists are:

Daniel Benjamin, Associate Research Professor of Economics at the
University of Southern California and a primary co-author of the paper
in Nature Human Behavior as well as many other articles on inference
and hypothesis testing in the social sciences.

Daniel Lakens, Assistant Professor in Applied Cognitive Psychology at
Eindhoven University of Technology and an author or co-author on many
articles on statistical inference in the social sciences, including
the Open Science Collaboration's recent Science publication
"Estimating the reproducibility of psychological science" (available
at https://dx.doi.org/10.1126/science.aac4716).

Blake McShane, Associate Professor of Marketing at Northwestern
University and a co-author of the recent paper "Abandon Statistical
Significance" as well as many other articles on statistical inference
and replicability.

Jennifer Tackett, Associate Professor of Psychology at Northwestern
University and a co-author of the recent paper "Abandon Statistical
Significance" who specializes in childhood and adolescent
psychopathology.

E.J. Wagenmakers, Professor at the Methodology Unit of the Department
of Psychology at the University of Amsterdam, a co-author of the paper
in Nature Human Behavior and author or co-author of many other
articles concerning statistical inference in the social sciences,
including a meta-analysis of the "power pose" effect (available at
http://www.tandfonline.com/doi/abs/10.1080/23743603.2017.1326760).

To tune in to the presentation and participate in the discussion after
the talk, visit http://www.methods-colloquium.com/and click "Watch
Now!" on the day of the talk. To register for the talk in advance,
click here:

https://zoom.us/webinar/register/45c571249ce37c8d8c34be5db4a05ad8

The IMC uses Zoom, which is free to use for listeners and works on
PCs, Macs, and iOS and Android tablets and phones. You can be a part
of the talk from anywhere around the world with access to the
Internet. The presentation and Q&A will last for a total of one hour.

My Most Popular Storify Stories, as of October 2017

October 24, 2017 by Miles Kimball

Most of the literal Socratic dialogues never happened that way. Plato made them up. But in the 21st century, I have had many genuine (though not literal) Socratic dialogues on Twitter. I have arranged the tweets for many of these Twitter convos into Storify stories. You can see the full archive of my storify stories here. Below, I give links to all of my stories with 200 or more pageviews, starting from the most popular story. The number of pageviews is listed beside each link. (I did this once before, for my August 5, 2014 post.)

Ensuring Safety from Rape and Sexual Assault is Beautiful →

October 23, 2017 by Miles Kimball

John Locke on How Things That Are No One's Property Become Someone's Property

October 22, 2017 by Miles Kimball

Who owns Mars? Despite the common ownership principles of the Moon Treaty ("Agreement Governing the Activities of States on the Moon and Other Celestial Bodies") that nations that make space launches have declined to ratify, if a private company were the first to reach Mars and set up a community of humans on Mars that persisted over time, it would almost surely be treated as owning the land the community sat on and a radius of a few kilometers around that community that the community regularly used. This is especially true if it took the precaution of having its settlement of and claim to that piece of Mars officially sponsored by a small Caribbean island recognized as a sovereign nation by the UN that had not signed the Moon Treaty.

John Locke explains the relevant part of natural law in section 30 of his 2d Treatise on Government: “Of Civil Government” (Chapter V "Of Property"):

Thus this law of reason makes the deer that Indian’s who hath killed it; it is allowed to be his goods, who hath bestowed his labour upon it, though before it was the common right of every one. And amongst those who are counted the civilized part of mankind, who have made and multiplied positive laws to determine property, this original law of nature, for the beginning of property, in what was before common, still takes place; and by virtue thereof, what fish any one catches in the ocean, that great and still remaining common of mankind; or what ambergrease any one takes up here, is by the labour that removes it out of that common state nature left it in, made his property, who takes that pains about it. And even amongst us, the hare that any one is hunting, is thought his who pursues her during the chase: for being a beast that is still looked upon as common, and no man’s private possession; whoever has employed so much labour about any of that kind, as to find and pursue her, has thereby removed her from the state of nature, wherein she was common, and hath begun a property.

The step of getting sponsorship from a small Caribbean island nation that had not signed the Moon Treaty may seem as if it makes it a matter of ordinary international law rather than just natural law. But what it does is simply to ensure that the law of nature applies. As John Locke writes in section 14, even in his day and ours, rulers are still in the law of nature:

It is often asked as a mighty objection, where are, or ever were there any men in such a state of nature? To which it may suffice as an answer at present, that since all princes and rulers of independent governments all through the world, are in a state of nature, it is plain the world never was, nor ever will be, without numbers of men in that state. I have named all governors of independent communities, whether they are, or are not, in league with others ...

In the more immediate future, if a private company mined a metallic near-earth asteroid and brought some of the more valuable metals back to earth, there would be a great deal of sentiment that the company ought to be able to claim title to those metals and sell them. The way US law is going, it should be easy for a company to get sponsorship of the US for this. John Locke would call that sentiment the law of nature.

On the other hand, if a company that was mining a small portion of an asteroid tried to claim the whole asteroid, or even more clearly, if the company that had set up a community at one spot on Mars tried to claim the whole planet, that would seem like overreaching. John Locke explains why: if only a small part of the asteroid is being mined or only a small part of Mars has been settled, then the rest seems still to be a part of the commons.

Of course, people and groups of people often do claim ownership of something in the commons that is much larger than they are actually able to use. But that ends up seeming unfair. The exception is if something close to the entire set of people currently exploiting the commons resource become part of a formal or informal agreement about the use of the resource. Elinor Ostrom won a Nobel prize for studying how communities of exploiters regulate the exploitation of commonly held resources. One important incentive for setting up such a system is to convince oneself and others that additional people who want to join the group of exploiters can be excluded from use of that resource.

All existing title to something that was not made by humans in the first place goes back to some original moment of appropriation. For most things on the Earth, the moment of appropriation from nature was many years ago. But for the asteroids, moons and planets, many of those alive will witness the moment of appropriation from nature.

Don't miss other John Locke posts. Links at "John Locke's State of Nature and State of War."

Are Rape and Sexual Assault About Power and Lust or Only About Power? →

October 21, 2017 by Miles Kimball

Megan Garber: Celebrities Have Gotten Away with Many Crimes Against Women in the Past; Will They in the Future? →

October 20, 2017 by Miles Kimball

On this topic, also see "An Open Letter from Dylan Farrow."

The Relative Citation Ratio

October 19, 2017 by Miles Kimball

A key problem with citation counts is that they disadvantage fields or subfields that are either small or have a culture of short reference lists. I am intrigued by the solution offered by B. Ian Hutchins, Xin Yuan, James M. Anderson, George M. Santangelo in their paper "Relative Citation Ratio (RCR): A New Metric That Uses Citation Rates to Measure Influence at the Article Level." This is a concept defined at the article level. At a senior hiring, tenure or promotion meeting, I would love to see annotation of the candidates curriculum vitae with relative citation ratios for each paper.

The key idea behind the relative citation ratio is defining a target article's subfield by the articles cocitation network: the set of papers that appear together in some paper's reference list along with the target article. This seems a great way to flexibly define the target article's subfield in a data-based way. It is also a way of defining the target article's subfield in a way that can evolve over time.

One advantage of taking an article's cocitation network as the comparison group for a relative citation ratio is that for articles that get cited at all, the cocitation network quickly becomes quite large, so there isn't a small numbers problem.

Conceptually, the idea of a relative citation ratio is to compare citations per year (not counting the calendar year of publication) for the target paper to average citations per year for the articles in the cocitation network. To make the denominator of the ratio even more stable, the authors substitute the average citations per year in the relevant journal for citations per year of a comparison article in the cocitation network.

The authors B. Ian Hutchins, Xin Yuan, James M. Anderson and George M. Santangelo provide code for the calculation of the relative citation ratio, as well as the values of the relative citation ratio for every article in PubMed (where the National Institutes of Health requires all articles it supports to be posted). The image below illustrates their iCite tool for the relative citation ratio.

Screen Shot 2017-10-14 at 7.43.41 AM.png

In addition to the intuitive appeal of the relative citation index, B. Ian Hutchins, Xin Yuan, James M. Anderson and George M. Santangelo report the results of several analyses that show the relative citation ratio is doing what one would hope it would do.

In addition to annotating CV's by relative citation ratios for each publication, I would like to see overall relative-citation-ratio-based measures of a scholar's overall productivity. These rankings can be used in various ways. For example, just as I pay attention to the pageviews on my blog, I also watch the evolution of my author rank by different measures on REPEC. And I occasionally cite the REPEC data for a relatively impartial measure of someone's scholarly prominence, as I did explicitly for Jeremy Stein in "Meet the Fed's New Intellectual Powerhouse" and implicitly for Andrei Shleifer in "Adam Smith as Patron Saint of Supply-Side Liberalism?"

I would love to see REPEC add measures of productivity based on the relative citation ratio to the many measures they have now.

The sum of relative citation ratios for all of a scholar's publications
The sum of (relative citation ratio/number of authors) for all of a scholar's publications
The sum of relative citation ratios for all of a scholar's publications in a five-year window, say from calendar years t-6 through t-2, where t is the year of publication.
The sum (relative citation ratio/number of authors) for all of a scholar's publications in a five-year window, say from calendar years t-6 through t-2, where t is the year of publication.

Measures 3 and 4 are related to the tough goal I hope tenured professors (who aren't working towards an even more important objective) strive for: to have a tenurable record for each five-year window after getting tenure as well as for the five-year (or sometimes longer) window that actually won them tenure.

I honestly don't know how I personally would fare by these measures. So this wish is driven by my curiosity, not by thinking it will necessarily make me look better. I tend to think I personally write in subfields with relatively high rates of citation and so would probably look a little worse by these measures than by some of the other citation-based measures. But maybe there are other subfields that have even higher rates of citation, so correcting for subfield could make me look relatively better.

There are limitations to any citation-based measure of scholarly productivity. But to the extent we end up looking at citation-based measures of scholarly productivity, those of us who love data can't help but want to stare at the best possible citation-based measures.

Anne Victoria Clark—The Rock Test: A Hack for Men Who Don’t Want To Be Accused of Sexual Harassment →

October 18, 2017 by Miles Kimball

Thanks to Claudia Sahm for flagging this link in a tweet, as well as telling about another life hack it remindered her of in the next tweet:

reminds when I was on job market, @mileskimball encouraged me to channel my 'inner Kerwin Charles' smart, confident economist, presents well

Matthew Shapiro, Martha Bailey and Tilman Borgers on the Economics Job Market Rumors Website

October 17, 2017 by Miles Kimball

Disgusted with the widespread misogyny, racism and more random viciousness on Economics Job Market Rumors that is becoming more widely appreciated because of Alice Wu's research, I wrote "Signalling When Everyone Knows about Last-Place Aversion: An Application to Economics Job Market Rumors."

As an Emeritus Professor of the University of Michigan Economics Department, I am still on its email lists. I am grateful to Matthew Shapiro, Martha Bailey and Tilman Borgers for giving me permission to share the thoughts about Economics Job Market Rumors that they had expressed to University of Michigan Economics faculty.

Matthew Shapiro (September 8, 2017 email to UM Economics graduate students with the faculty cc'd)

Dear Colleagues:

I am writing to call your attention to a statement on the EJMR site by Olivier Blanchard, President-elect of the American Economic Association.

I endorse this statement and urge our students to shun social media activities that propagate sexism, racism, or bigotry.

Matthew Shapiro

Martha Bailey (October 5, 2017 email to UM Economics faculty)

Dear Colleagues,

As you may (or may not) be aware, the AEA has thus far chosen *not* to make a statement about EJMR. This refusal of the association to take action is, in part, what prompted the eloquent (personal) statement by Olivier Blanchard, President-elect of the American Economic Association. CSWEP [Committee on the Status of Women in the Economics Profession] (where Justin [Wolfers] and I are both board members) has recommended that the AEA take action. Other associations in economics are starting to. One organization that has succeeded is the Economic History Association’s statement (https://eh.net/eha/15371), which Paul [Rhode, Chair of Economics at the University of Michigan] and I both endorsed.

Here is something easy you can do today: sign this petition to encourage the AEA to take action: http://www.iaffe.org/forms/Petition-AEA-EMJR/ (link to petition webpage -- http://www.iaffe.org/petition-aea-emjr/) (See below)

I haven’t ever sent an email like this to my colleagues, but this is an issue very personal to me and one I think deserves more attention. To be clear, I do not believe this is a freedom of speech issue. EJMR has been used as a platform to bully and attack prominent economists—especially women economists. Sexist, racist, homophobic and anti-Semitic statements, particularly when those comments target and bully particular scholars, present a large barrier to the diversity of the economics profession and limits our openness to new ideas.

Our department’s hiring strategy notes that, “we wish to create a department that reflects diversity in life experiences, including diversity of race, gender, sexual orientation, religion, and family background. A diversity of representation of faculty and students will encourage the highest quality research and teaching to address the needs of a diverse world.” Achieving this goal requires that our department and the economics profession promote an intellectual environment and academic discourse that is respectful and collegial for *all*.

The profession has remained silent for too long on this subject, tacitly encouraging the persistence of this harassment and refusing to name it the hate speech that it is. Taking no action is a political statement that can only serve to encourage the perpetuation of harassing and exclusionary discourse. I hope that you will consider speaking to your students and colleagues about this issue, bringing this up in your respective associations, signing a petition, and doing other things to encourage respectful and collegial academic dialogue.

With warm regards,

Martha

Tilman Borgers (October 5, 2017 email to UM Economics faculty)

Dear Colleagues,

I am sure that you will not welcome an extended exchange of comments on the “Economic Job Market Rumors” (EJMR) website on the department email list. I had this in mind when I nonetheless decided to follow up on Matthew’s and Martha’s recent emails. I want to add two thoughts to the debate. Thank you for your patience, if you read them. I believe they are important.

1) In addition to being a forum for misogynist comments, EJMR has been a forum for discriminatory remarks towards Asian faculty and students. This hasn’t been emphasized enough.

2) EJMR has, at times, fulfilled a "whistleblowing" role, giving a forum to legitimate concerns, for example about publishing practices, expressed by those who otherwise would not have a voice. As people think about alternatives to EJMR, I believe it would be good to find a forum that allows “whistleblowing” outside of the context of racism and misogyny.

Regards,

Tilman

Update: Three days after this post appeared, the American Economic Association sent out an email to its members with this:

Statement of the AEA Executive Committee

October 20, 2017

To: Members of the American Economic Association
From: Peter L. Rousseau, Secretary-Treasurer
Subject: Statement of the AEA Executive Committee

Many members of the economics community have expressed concern about offensive behavior within our profession that demeans individuals or groups of individuals. The American Economic Association strongly condemns misogyny, racism, homophobia, antisemitism and other behaviors that harm our profession.

AEA President Alvin E. Roth has charged an ad hoc committee on professional conduct to formulate a set of guidelines for economists to be considered by the Executive Committee. The ad hoc committee is charged with evaluating various aspects of professional conduct, including those which stifle diversity in Economics. It will submit a report in time for discussion in January. There will be a period for comment by the AEA membership on that report following its release.

The Association is also exploring the possibility of creating a website/message board designed to provide additional information and transparency to the job market for new Ph.D.s, and will be surveying departments to assess what information about their search processes might be shared.

Valerie Ramey, Vice President of the AEA, responds to this update in this thread on my Facebook page (which is totally public). This has started an interesting discussion you can see at this link. (Valerie also "liked" my note there that I was posting this link.)

Kim Gittleson: Where are all the women in economics? →

October 16, 2017 by Miles Kimball

Hat tip to Barry Ritholz's excellent Friday, October 13th list of reads.

Travis Bradberry: 10 Habits All Genuinely Confident People Share

October 15, 2017 by Miles Kimball

*Link to "10 habits all genuinely confident people share" on Quartz*

My lack of full confidence is shown by my urge to compare myself against Travis Bradberry's list of the habits of genuinely confident people. These six I do OK on:

They speak with certainty
They seek out small victories
They exercise
They take risks
They aren’t afraid to be wrong
They celebrate other people’s successes

But my insecurities show up in trying to impress other people. I don't let that get in the way of being my own person. (It would be quite hard for me psychologically not to be my own person even if I tried.) But I definitely

Care about what others think of me
Worry about whether I measure up
Seek attention
Talk too much

On talking too much, I have noticed that in addition to a high intercept on talking, I have an anomalously small negative elasticity of talking with respect to group size. Thus, the bigger the group, the higher the ratio of how much I talk compared to 1/n. I get the closest to talking a reasonable fraction of the time when I am talking to just one other person.

They say confession is good for the soul. We'll see.

Jordan Rosenfeld: The Scientific Link Between Boredom and Creativity →

October 14, 2017 by Miles Kimball

Mandë Holford and Lindsay Portnoy: Children Should Be Playing More Games in the Classroom. Here’s Why →

October 13, 2017 by Miles Kimball

Mitch Prinstein on Status and Likability

October 12, 2017 by Miles Kimball

"You can have both. In fact, about 30 percent of those who are high-in-status are also really, really likeable. ... But ... one of the key ways to get status is to be aggressive. And that is the number one predictor of being disliked. Many people get their status by stepping on others and making themselves seem more powerful or important or worthy of attention than others. In doing so, they actually make themselves quite dislikable."

—MItch Prinstein, in an interview with Dave Nussbaum about his book "Popular: The Power of Likability in a Status-Obsessed World"

Psychological Safety vs. Fear: Many People Perform Better in Response to Carrots than Sticks →

October 11, 2017 by Miles Kimball

Meat Is Amazingly Nutritious—But Is It Amazingly Nutritious for Cancer Cells, Too?

October 10, 2017 by Miles Kimball

Link to the Wikipedia article for "Good Calories, Bad Calories: Fats, Carbs, and the Controversial Science of Diet and Health

The idea that red meat is unhealthy has been drilled into the American public for at least half a century. For those of us who have high standards for causal inference throughout the causal chain all the way from eating meat to death or morbidity, there is surprisingly little evidence for this view, though I will talk about some. How well the Inuit and Masai fared on a traditional diet of meat should give one pause in asserting too strongly the harms of red meat.

Let me give two bits of evidence, one positive for meat, suggesting it is very nutritious for the human body; another negative, suggesting it is highly nutritious for cancer cells as well.

First, let's turn to meat being very nutritious for the human body. In Chapter 19 of his excellent history of nutritional thought, "Good Calories, Bad Calories: Fats, Carbs, and the Controversial Science of Diet and Health," Gary Taubes tells about an early 20th Century scientific dispute over whether meat is enough to keep someone alive and in robust health:

The notion of a carbohydrate-restricted diet based exclusively on fatty meat was publicized after World War I by the Harvard anthropologist-turned-Arctic-explorer Vilhjalmur Stefansson, who was concerned with the overall healthfulness of the diet, rather than its potential for weight loss. Stefansson had spent a decade eating nothing but meat among the Inuit of northern Canada and Alaska. The Inuit, he insisted, as well as the visiting explorers and traders who lived on this diet, were among the healthiest if not the most vigorous populations imaginable.

Among the tribes with whom Stefansson lived and traveled, the diet was primarily caribou meat, “with perhaps 30 percent fish, 10 percent seal meat, and 5 or 10 percent made up of polar bear, rabbits, birds and eggs.”

The Inuit considered vegetables and fruit “not proper human food,” Stefansson wrote, but they occasionally ate the roots of the knotweed plant in times of dire necessity. The Inuit paid little attention to the plants in their environment “because they added nothing to their food supply,” noted the Canadian anthropologist Diamond Jenness, who spent the years 1914–16 living in the Coronation Gulf region of Canada’s Arctic coast. Jenness described their typical diet during one three-month stretch as “no fruit, no vegetables; morning and night nothing but seal meat washed down with ice-cold water or hot broth.” (The ability to thrive on such a vegetable-and fruit-free diet was also noted by the lawyer and abolitionist Richard Henry Dana, Jr., in his 1840 memoirs of life on a sailing ship, Two Years Before the Mast. For sixteen months, Dana wrote, “we lived upon almost nothing but fresh beef; fried beefsteaks, three times a day…[in] perfect health, and without ailings and failings.” ...

... If the Inuit thrived in the harshest of environments without eating carbohydrates and whatever nutrients exist in fruits and vegetables, they, by definition, were consuming a balanced, healthy diet. If they did so solely because they had become evolutionarily adapted to such a diet, which was a typical rejoinder to Stefansson’s argument, then how can one explain those traders and explorers, like Stefansson himself and the members of his expeditions, who also lived happily and healthfully for years at a time on this diet?

Nutritionists of the era assumed that all-meat diets were unhealthy because (1) excessive meat consumption was alleged to raise blood pressure and cause gout; (2) the monotony of eating only meat—or any other single food—was said to induce a physical sense of revulsion; (3) the absence of fresh fruit and vegetables in these diets would cause scurvy and other deficiency diseases, and (4) protein-rich diets were thought to induce chronic kidney damage, a belief based largely on early research by Louis Newburgh.

The nutritionists back then were Vilhjalmur Stefansson's opponents in this scientific dispute:

... Francis Benedict, as Stefansson told it, claimed that it was “easier to believe” that Stefansson and all the various members of his expeditions “were lying, than to concede that [they] had remained in good health for several years on an exclusive meat regimen.”

In the winter of 1928, Stefansson and Karsten Anderson, a thirty-eight-year-old Danish explorer, became the subjects in a yearlong experiment that was intended to settle the meat-diet controversy. The experiment was planned and supervised by a committee of a dozen respected nutritionists, anthropologists, and physicians.*

Eugene Du Bois and ten of his colleagues from Cornell and the Russell Sage Institute of Pathology would oversee the day-to-day details of the experiment.

... they began living exclusively on meat, at which point they moved into Bellevue Hospital in New York and were put under twenty-four-hour observation. Stefansson remained at Bellevue for three weeks, Anderson for thirteen weeks. After they were released, they continued to eat only meat for the remainder of one year. If they cheated on the diet, according to Du Bois, the experimenters would know it from regular examinations of Stefansson’s and Anderson’s urine. “In every individual specimen of urine which was tested during the intervals when they were living at home,” Du Bois wrote, “acetone [ketone] bodies were present in amounts so constant that fluctuations in the carbohydrate intake were practically ruled out.”

Du Bois was convinced:

Du Bois, who supervised the experiments, wrote an introduction to Stefansson’s book. After Stefansson and Anderson were living exclusively on meat, he said, “a great many dire predictions and brilliant theories faded into nothingness.” A diet that should have left Stefansson and Anderson deathly ill from scurvy had left them as healthy as or healthier than the balanced diet they had been eating in the years immediately preceding the study. “Quite evidently we must revise some of our text book statements,” Du Bois concluded.

To me, the most surprising thing was that no vitamin deficiencies resulted. Gary Taubes explains:

meat has a full set of amino acids;
meat has all the vitamins except vitamin C; and
the need for vitamin C is much lower when eating meat than when eating a lot of carbs.

There are arguments over whether our distant ancestors, in the environment of evolutionary adaptation ate mostly plants or mostly meat. I suspect that the recurrent ice ages back then made things tough enough that for us to be here today, our distant ancestors had to do OK on a wide variety of different diets. Sometimes, very little meat would have been available. At other times, meat might have been almost all that was available. I don't think it strange that we should be adapted to do well as almost total carnivores, even though we also have the capability to deal well with plant food in its natural form.

Now, to rain on the parade of pro-meat ideas above, let me turn to some interesting evidence suggesting that meat is very nutritious for cancer cells as well. The evidence comes from research by T. Colin Campbell and coauthors, recounted in his book, "The China Study: Revised and Expanded Edition: The Most Comprehensive Study of Nutrition Ever Conducted and the Startling Implications for Diet, Weight Loss, and Long-Term Health."

Link to the Wikipedia page for "The China Study"

Colin Campbell's team predisposed mice and rats to cancer by a range of methods, such as dosing them with aflatoxin and breeding them with genes making them prone to cancer. When these mice and rats predisposed to cancer were fed low-casein diets they usually didn't develop cancer. When they were fed high-casein diets, they usually did develop cancer.

Casein is a protein found in abundance in milk and cheese. Like other proteins from animal sources, it is a "complete protein" with the full range of amino acids needed to build mouse, rat or human cells. My hypothesis for what was going on in the experiments is that if there is a superabundance of complete protein available, it makes it easy for cancer cells to grab some of that protein or the complete set of amino acids from that protein to build strong cancer cells. In other words, my hypothesis for the results Colin Campbell's team got is that a diet low in complete protein is like a very safe form of chemotherapy: having low levels of complete protein is harder on cancer cells than it is on normal cells that don't need the materials for growth as much.

A straightforward way to have low levels of complete sets of amino acids, and thereby inhibit the development of any incipient cancers one might have, is to eat relatively little animal protein and to not have too much in the way of complete combinations of proteins for plant proteins beyond that. If you have had a cancer already been detected, it is probably worth talking with one's doctor about cutting down complete protein even further.

I wonder if eating plant proteins in a way that is complete over time, but incomplete on any one day, could hit the sweet spot of providing enough of all the essential amino acids for the normal cells to do their thing, while still making it hard for cancer cells to get all the key amino acids they need to grow quickly.

In addition to the experimental evidence in mice and rats, Colin Campbell and his coauthors have a monograph Diet, Lifestyle and Mortality in China: A Study of the Characteristics of 65 Chinese Counties that provides 65 very interesting data points with rates of incidence for many different types of cancer and other diseases, together with dietary patterns for a county based on samples of individuals from two villages in each county. Because this data was collected relatively soon after the opening of China after the death of Mao, diets in China had not yet Westernized when the data was collected, and were different in different counties. Though it is hard to separate out different possible explanations with only 65 data points, animal protein levels did tend to be positively correlated with cancer incidence, as one should expect of the experiments in rats and mice were an indication of what happens for human beings. (One thing to keep in mind is that this "China Study," by looking at all reported cancers within a county was sensitive to differences in cancer incidence of modest size. Nevertheless, if the correlations are actually indicative of causal effects as in the mice and rat experiments, the effects are much larger than cancer dangers that get hyped a great deal.)

How much animal protein is too much? Experimentally, in the mice and rates the comparison was between something like 5% of calories as animal protein and 20%. Despite starting out in life as an enthusiastic meat, milk and egg eater, Colin Campbell wants everyone to go vegan, but I am not ready to go there, both because

I enjoy animal-based food, and
I am not 100% sure a level of complete protein equal to zero is fully safe—and to the extent a vegan diet provides a full set of amino acids it might have the same downsides as an animal protein diet by providing great nutrition for cancer cells.

What I am willing to do personally, is to try to keep animal protein down to 1 calorie per day per pound of body weight, or a number of grams of protein equal to one quarter of my weight in pounds. How do I come up with that as a helpful target? In Chapter 16 of The China Study, "Government: Is It for the People," Colin writes:

Relative to total calorie intake, only 5–6% dietary protein is required to replace the protein regularly excreted by the body (as amino acids). About 9–10% protein, however, is the amount that has been recommended for the past fifty years to be assured that most people at least get their 5–6% “requirement.” This 9–10% recommendation is equivalent to the well-known RDA.

I am thinking of a target of around 7% of calories in animal protein on a typical day, without trying to count up the amount of plant protein. Then the complete animal protein alone would be enough to replace the amino acids that are used up and depleted, with plant protein as extra. The footnote at the end of Colin's passage suggests that 2200 calories a day would be typical for someone weighing 70 kilograms. Since a kilogram is about 2.2 pounds, proportionally that is 2200 calories/(2.2 * 70 kilograms) per day= (100/7) (calories/pound) per day. 7% of that is 1 calorie per pound per day. A gram of protein has about 4 calories, so 1 calorie of animal protein per pound of body weight per day is about 1/4 of a gram of animal protein per pound of body weight per day.

I found by trying it out that for generic animal foods, you can google "how many grams of protein does [animal food x] have" and get a ready answer (with the specification of a measure of weight nice but often optional). For less generic animal foods—for example mozzarella cheese of a particular brand—one can look on the package. To provide some idea of magnitudes, a large egg has about 6 grams of protein, a cup of milk has 7.7 grams of protein, while a quarter-pound hamburger patty has about 29 grams of protein.

Although I think Colin Campbell and his coauthors provide some important data on a possible relationship between animal protein consumption and cancer, I don't think Colin reads the full range of evidence appropriately in coming to his overall viewpoint.

In particular, given the evidence he has to worry about animal protein (by my hypothesis, because of its completeness), there is no evidence to speak of against animal fat. Any worrisome correlation that seems to make animal fat suspect could easily be due to negative effects of the animal protein that tends to go along with animal fat. And the US and other advanced countries have done a massive experiment of trying to improve health by subtracting fat—mostly animal fat—from food products. That subtraction of animal fat has done nothing to improve health, as large studies such as the Women's Health Study have indicated. Colin Campbell talks about interactions among the components of whole foods being the key, but only the nutritional interaction among the amino acids in animal proteins is necessary to explain things. Subtracting animal fat doesn't do any good because it is the animal protein that is the problem, not the animal fat!

For more on dietary fat, see "Jason Fung: Dietary Fat is Innocent of the Charges Leveled Against It."

Also, I am quite skeptical of Colin Campbell's view that animal protein is key to the main axis of "Western Diseases." The reason I am skeptical is that most "Western Diseases" or "diseases of affluence" are correlated with obesity, and there is reason to think that, to a first approximation, "Obesity Is Always and Everywhere an Insulin Phenomenon." Disregulation of the insulin system has to do with things that cause insulin to spike. In "Obesity Is Always and Everywhere an Insulin Phenomenon" I give a rule of thumb that a Food Insulin Index over 40 is high. Sugar and refined carbs typically exceed that level of 40 by quite a bit. Animal foods, not so much. Let me reproduce my animal food selections from a food insulin index table that I gave in Obesity Is Always and Everywhere an Insulin Phenomenon" here:

Meat and Eggs

eggs (poached) 23± 8
steak 37 ± 24
skinless roast chicken 17± 8
white fish 43± 26
hot dog 16± 6
bacon 9± 4
tuna packed in water 26± 8
tuna packed in oil (drained) 16± 4

Dairy

strawberry lowfat yogurt . 64 ± 12
skim milk 60 ± 26
1% milk 34 ± 8
whole milk 24 ± 6
cream cheese 18 ± 12
butter 2 ± 2

Skim milk and lowfat yogurt look bad, and whitefish is a little high, but it doesn't look like most animal foods cause big insulin spikes.

In other words, meat, dairy and eggs, by providing an abundance of complete proteins may help build strong cancer cells, but there is no good evidence that they make people fat. And if meat, dairy and eggs don't make you fat, the variance of cancer they would be available to help explain would be the (still substantial) variance of cancer that is not explained by variance in obesity. Statistically, I would be especially interested in multiple regressions of cancer on obesity and animal protein consumption (with measurement error correction for obesity to make sure that one wasn't undercontrolling for obesity).

As I mentioned above, personally, I am convinced enough by the evidence to marginally cut back on my animal protein consumption. If I were diagnosed with cancer, I would cut back further. There are some extra issues with dairy proteins that Colin Campbell points to that worry me some—grist for a future post. There is no evidence whatsoever to convince me to reduce my animal fat or plant fat consumption.

Despite disagreements, it is important to point to areas of agreement. Colin Campbell, Gary Taubes and Jason Fung all argue quite strongly that sugar and refined carbs are very bad. You should cut them out of your diet.

Everyone agrees that non-starchy vegetables are great. Because fruit has both sugar (bad) and antioxidants + other valuable phytochemicals (good), Colin Campbell and Gary Taubes differ on how much fruit to eat. But everyone agrees that whole fruits are much healthier than fruit juice. (The fiber in the whole fruit slows down the digestion of the sugar to moderate the insulin spike.)

More research is needed to definitely answer the key questions. Unfortunately, much of the research effort in the last few decades has been barking up the wrong tree, thinking that fat in general, or animal fat in particular, was a key dietary villain. In my view, research should focus on (a) "How much can be explained by insulin levels?" and (b) "Controlling for insulin levels, what are the effects of animal protein (or complete sets of amino acids)?" If there are some baleful effects, then I would want to know "Is there some level of animal protein consumption that is safe?" I also want to know whether infrequent (say once a week or once every other week) high levels of animal protein from going out to a restaurant are dangerous or not. The part of this question that relates to my hypothesis is the question of if cancer growth depends on a regular superabundance of complete protein or (as I hope is not so) if cancer growth can proceed vigorously even with only an occasional superabundance of complete protein.

In "Obesity Is Always and Everywhere an Insulin Phenomenon," I make the analogy that trying to lose weight without skipping meals is like trying to reduce inflation without a recession. Given the crucial importance of combining a low-refined-carb diet with fasting for weight loss and maintaining a lower weight, research should focus on (c) investigating the effects of fasting. The direct effect of fasting is to lower insulin levels. But "How often can fasting lower insulin resistance?" And "To what extent can relatively frequent 18-20 hour fasts or 42-44 hour fasts mitigate any baleful effects of animal proteins that might be discovered under research agenda (b)?"

Don’t miss my other posts on diet and health:

I. The Basics

II. Sugar as a Slow Poison

III. Anti-Cancer Eating

IV. Eating Tips

V. Calories In/Calories Out

VI. Other Health Issues

VII. Wonkish

VIII. Debates about Particular Foods and about Exercise

IX. Gary Taubes

X. Twitter Discussions

XI. On My Interest in Diet and Health

See the last section of "Five Books That Have Changed My Life" and the podcast "Miles Kimball Explains to Tracy Alloway and Joe Weisenthal Why Losing Weight Is Like Defeating Inflation." If you want to know how I got interested in diet and health and fighting obesity and a little more about my own experience with weight gain and weight loss, see “Diana Kimball: Listening Creates Possibilities” and my post "A Barycentric Autobiography. I defend the ability of economists like me to make a contribution to understanding diet and health in “On the Epistemology of Diet and Health: Miles Refuses to `Stay in His Lane’.”

Brandon Fuller and Sean Rust: Why Letting States Sponsor Immigration Visas Should Satisfy Everyone →

October 09, 2017 by Miles Kimball

Private Property Reduces Decision-Making Costs

October 08, 2017 by Miles Kimball

For some people, there is a certain glow to democratic decision-making. And once obtained, there can be no objection to explicit universal consent. But democratic decision-making, and getting explicit universal consent can be very costly.

For example, the high level of democracy for decision-making by the tenure-track faculty at the University of Michigan is one case in point in my own life. This required a lot of time in faculty meetings. Fewer decisions are explicitly democratic at in the Economics Department at the University of Colorado Boulder where I am now, but the quality and degree of consensus in decision-making is nevertheless high here.

Dividing up goods held in common into private property for individuals is a way to dramatically reduce the amount of costly democratic, or more generally political, decision-making that is required. In section 29 of his 2d Treatise on Government: “Of Civil Government” (Chapter V "Of Property"), John Locke points to this as an advantage of the kind of "labor theory of property" he thinks appropriate to the state of nature:

By making an explicit consent of every commoner necessary to any one’s appropriating to himself any part of what is given in common, children or servants could not cut the meat, which their father or master had provided for them in common, without assigning to every one his peculiar part. Though the water running in the fountain be every one’s, yet who can doubt but that in the pitcher is his only who drew it out? His labour hath taken it out of the hands of nature, where it was common, and belonged equally to all her children, and hath thereby appropriated it to himself.

One doesn't have to accept John Locke's labor theory of property in order to see a benefit of some method of dividing things not yet divided up and making them private property.

When I say this, I am thinking of goods that are rivalrous in consumption. The argument for making private property of goods that are nonrivalrous—that everyone could simultaneously enjoy—is much weaker. There the rule that everyone gets to benefit from the good is a very attractive one; society should be slow, careful and limited in ever making nonrivalrous goods private property.