David Card Blackpilled an Entire Generation of Economists [repost]
Berkeley is a deeply corrupt den of Maoist snakes
Note: This is the first article I ever wrote in my life, almost 2 years ago!
In retrospect, re-reading it, some of the writing is janky — I cringe at some of the phrases, especially at some of the unnecessary cuss words. I left them unchanged, because I don't think that’s a bad thing. Shows I’m improving. I have written almost 200 articles since then, and my style and skill have grown a lot.
I am re-sending it out because it originally went to only a dozen Substack subscribers… now that I have almost 5,000 subscribers, 99% of them haven’t dug in my archives to read it, and even if I think some of this writing is janky, the content/message is evergreen. In fact this article is more relevant now than ever, since David Card, the subject of this article, won the Nobel Prize in Economics last year, after this was published.
This article is also still relevant because the #EconTwitter mafia is still trying to deplatform EJMR, to kick them off the internet. The fight for free speech never ends.
Alice Wu’s undergrad thesis will go down as one of the most influential papers of the last decade, but not because the NYT, Financial Times, Planet Money and Brad DeLong creamed their collective pants over it, and not because it changed how the general public viewed the culture of the entire economics profession. Rather, when it was published in a top 10 journal was the moment academic economics jumped the shark and lost any last shred of credibility.
Don't you get it? This paper highlights that “facts” have never mattered: economics is a game, not a science. The winners of this game get to dictate public policy, so the careerists involved (almost all of whom are hard left Democrats) are willing to do anything it takes to win. In practice this usually means torturing data and obfuscating with fancy econometrics until you reach your pre-determined, handwaved conclusion. The results are then "peer reviewed" so you better not dare question them.
The institutionalized level of blatant lying involved is incredible — everyone is lying, everyone knows that everyone else is lying — the entire profession is built on lies. Whether they are willing to admit it or not, we all know that the emperor has no clothes. The only ones who don’t are naïve rubes or useful idiots.
Part of this game is realizing that the Wu thesis is a shibboleth. You must pretend to praise it so that people will know you're willing to follow the party line on other matters too. The alternative is to state your true opinion and become persona non grata, and have your career ruined. Anyone who has lived through communism will tell you that publicly affirming the absurd in such a way is a favorite tactic of communist regimes: it’s a group loyalty test. Welcome to your new equilibrium.
Don’t get me wrong, it was a solid undergrad thesis — probably my undergrad thesis wasn't much better — but this one got her accepted into the Harvard PhD program and into the Bolshevik inner circle. As an undergrad she can be forgiven for the lack of depth in her analysis; what can’t be forgiven is a cabal of tenured professors pretending her work to be of serious quality and parading it around like the second coming of “Market for Lemons”, simply because it reached their preferred political conclusion. These people are competent enough econometricians to know full well that this paper clearly cannot provide statistical evidence to support its conclusions, but they push it anyways. Imagine the type of person (e.g. Justin Wolfers) who would do this: what a pusillanimous little pinko halfwit they would have to be. These types of people fancy themselves scholars, but are really nothing more than the #EconTwitter Taliban… they should be working behind the DMV counter, not in universities.
The staging ground for the #EconTwitterTaliban’s latest Jihad is econjobmarketrumors.com (EJMR). These fascist fucking losers are actively trying to shut down EJMR, calling for lawsuits, hacking, and de-platforming on a weekly basis. These twitter warriors make me sick with their anti-free speech attitudes… open, un-PC discussion in THEIR profession? Can't be allowed.
When they attack EJMR, they do it by calling it a cesspool of bigotry… and a non-negligible segment of them (the useful idiots contingent) may actually believe this! Bless their souls. But all the shakers-and-movers in the profession who hate EJMR hate it for a much more sinister reason: because it represents a threat to their rigid prestige hierarchy, and a threat to their monopoly that has been built upon asymmetry of information. By making information free and available to all, this site can only do damage for those at the top by airing their dirty laundry. There are way too many super embarrassing leaks that no elites would want to allow; it’s a constant reminder that the American Economic Association (AEA) is a corrupt organization filled with the corrupt elite…. and sometimes the corrupt are even women.
Here are just a few of the juicier scandals that have played out on the site over the last couple years:
Schiraldi (LSE) and Seiler (Stanford) false coauthors of AER publication
New "Family Ruptures" AER / NBER is rip-off of obscure paper
WTF?? The Impact of MTV's 16 and Pregnant on Teen Childbearing
None of these scandals have been discussed even once on #EconTwitter, nor have they been reported by the New York Times. Weird. So, EJMR is the only place where all the frauds in economics profession get exposed. Of course these people want to destroy it, then. Once you examine the incentives involved, the unanimity by high ranking members of the economics profession in praising the work by Alice Wu starts to make a lot of sense. Wu is just a tool being used to stamp out the underclass revolt.
Perhaps the most relevant scandal for our discussion is the Hilary Hoynes corruption at AER. I’m not going to walk through the details: to really do it justice I would encourage you to click on that link and read all 288 pages of discussion. After this scandal broke, Hoynes went on the offensive and wrote in an interview that EJMR was sexist and misogynistic for exposing her corruption… yes, somehow, EJMR was sexist because they pointed out an obvious case of plagiarism at a top university where the authors happened to be female. How contemptible for her to play the gender card. Really gross stuff that cheapens the words of true feminist allies.
So let’s connect the dots. A powerful Berkeley economist is now widely known as a fraud because of EJMR, and then almost immediately someone else in her department (David Card — Wu’s advisor — who also happens to be President of the American Economic Association) magically turns up evidence that EJMR is cesspool of misogyny. Hmmmm...
Dear Hilary Hoynes, if you are reading this, I will quote to you the wise words of Wu (2019): "Overall, the whole atmosphere and culture of a profession can be shaped by a select few people." I would note that this could apply equally to sexism as to editorial corruption.
Everyone is so aware of how much lying is done by the Berkeley mafia that this was actually an extremely easy blogpost to write: hundreds of economists (much smarter and more experienced than I) have already anonymously ripped Alice Wu’s undergrad thesis to shreds, but none of them have dared put their name to these words for fear of professional retribution by the #EconTwitterTaliban. All I had to do was coherently synthesize (or outright copy) their thoughts on the matter and then slap my name on it and take all the glory. Easy. Even if it's just a synthesis of the arguments put forward in a bunch of historic ejmr threads, there's maybe some value in having it in the public domain. Where to begin?
**** Note: there are 3 versions of this paper: thesis, AER P&P, and RESTAT. I use them interchangeably because they are basically all the same project.
The most damning criticism is lack of generalizability — the paper's entire conclusion hinges on the assertion that the "majority of users are current male students in PhD economics programs." This assertion is offered without support. The forum is, after all, an online, anonymous, message board anyone can effortlessly contribute to without even providing an email address.
How could you prove that this site could represent X% of the profession? What percent of Wu's observations were posted by Ph.D. economists? It is well known that the forum is completely and totally overrun by bodybuilders, undergrads, journalists, basement dwellers, lawyers, Redditors, coders, mathematicians… etc. This is such an egregious flaw that even lefty rag Mother Jones pointed it out.
What if someone applied the same analysis to the Trump posts on EMJR? Would you also accept it as evidence that the economics profession holds strong pro-T sentiments? If not, then why not? What would be the difference between that or the analysis of posts with gender-related content?
Not only is the sample not representative of economics, the sample *is not even representative of the people who go on EJMR*, because of megaposters. I estimate that there is a core group of the same ~20 people who are responsible for the vast majority of the worst spam. These 20 people do not represent the economics profession.
Maybe Wu or Card themselves posted thousands of derogatory/racist/sexist comments to get significant results. Who’s to say they didn’t? It is equal parts cunning and on-brand for them.
She never filtered out the “off topic” and “trash” sub-forums. These are a huge fraction of all posts.
It’s basically just a wordcloud. Is there any causality whatsoever? Any claim at causality or identification is a joke.
The paper is literally not replicable, given how the site is coded. If you inspect the source code there is no timestamp for posts. If she ran her code again, she wouldn't be able to reproduce her results. Furthermore a third party can't attempt a replication or a rejoinder either.
To learn anything particular about economists, we'd need to see the same analysis applied to other datasets. What words are associated with women on, Twitter, Instagram, Reddit, Facebook, and on the Internet generally? I strongly suspect that discussion of women on the Internet being associated with words like "hot" and "sex" is not at all unique to EJMR.
This is especially egregious because it would be so easy and perfect to run on the Sociology (SJMR) and Political Science (PSMR) sister websites of EJMR. The data is right there, formatted in the exact same format as EJMR, it’s such low hanging fruit, but she conspicuously didn’t include it because it might’ve shown sociologists are more sexist than economists.
The mathematical model is incredibly frustrating because it completely mis-characterizes the situation here as well as that of society, at least as I understand it.
She writes, "Posters are assumed to value their contribution to public knowledge about the relationship between professional characteristics and jobs in the profession. They are also assumed to gain utility by boosting the professional reputation of members of their own gender group relative to that of members of the opposite gender group".
Does she, or anyone really, actually believe that this is how men (or women) operate? Lets assume (and this assumption is obviously false, but let's give her the benefit of the doubt) that the inhabitants of this board are almost exclusively economists. My impression from my years spent here is that cruel "misogynistic" remarks are almost always personal in nature. That is, someone has a grudge against someone else and they resort to gendered terms in an attempt to belittle/disparage.
My point is, her model, in which men want to boost the professional standing of men and diminish that of women, is complete nonsense. This is not how people in the real world operate. This is not how this board operates.
She says: “The lack of progress in attitudes toward women as indicated by the EJMR forum can help explain part of the persistent gender gap in a profession.”
Does anyone actually buy that toxic posts on an online professional forum explain a lack of female representation in economics? That’s an absurd premise. People make their decision about whether to pursue economics long before they have a sense for this field’s professional culture, even if you believe the latter is well-proxied by this cesspool.
By only scraping the first and last pages of each thread she is missing most of the substantive discussion on this site.
This can introduce bias. The first and last page are usually the less serious discussions when either the trolls were started or the trolls were too much that no one wanted to feed them then leaves. The main meat of the discussion is within.
Troll posts tend not to go beyond 2 pages and so will be over represented in the results. More serious threads are much longer. For example the German market thread has 1700+ pages with lots of normal/serious discussion of women but will be heavily discounted using Wu methodology.
Only the last page at the time that she LAST took her sample would be used in the data. Who knows how many times she constructed a dataset.. doesn't she realize that, at some point in time, every page was the last page of the thread?
Look at the names used to classify male thread versus female threads.
The female names are going to mostly capture living economist who are not authors of graduate level texts. The male names include a lot dead people (Friedman, Von Neumann, Ricardo). Dead people are going to tend to be brought up in regards to professional accomplishments only. Anybody have any hot gossip on David Ricardo’s wife or kids?
The male names also capture authors of graduate level textbooks. Threads about textbooks are almost always going to include her “academic classifiers” in future posts (micro, macro, econometrics, etc.)
It was written with a conclusion in mind before she even started! She was quoted as saying:
"I had to take a step out of the pre-existing packages and think about what I wanted to get out of the data "
Every post has “upvotes” and “downvotes” attached to it, which she simply ignores. This is because when a post is heavily bigoted, it tends to get downvoted into oblivion, so she didn’t include this control. It would also be interesting to have a measure of whether sexist posts tend to have positive or negative net upvotes.
"The parallel list of words associated with discussions about men reveals no similarly singular or hostile theme. "
I call BS. Men get non-stop called aspie, nerds, loser, retard, deadwood, stupid, fat, ugly, etc. The word “virgin” for example is heavily/exclusively associated with men and is thrown about constantly, but doesn’t show up anywhere in her analysis.
I myself have been the victim of thousands of the most vicious personal attacks you could possible imagine on this website. Please don’t deny my lived experience. She is erasing my lived experience with this paper.
EJMR is 38% female, as per Google analytics. 38% is higher than most Economics faculty lounges (if not all), and much higher than EconTwitter.
Everyone knows that nobody is meaner to women than other women. If a thread goes off topic and is derailed by personal attacks, there is a good chance it is one of these 38%.
Wu says: “The marginal effect of a word is the change in probability of a post being classified as female when it contains one more word.”
So, for example, "hotter," "hot," and "attractive" all show up strongly. Conditional on the post containing the word "hot," the post is more likely to be about women...duh?
This paper shows us that when men on EJMR talk about attractiveness, they're talking about females. It does not show that when men on EJMR talk about females, they are talking about sex. P(female|w) != P(w|female).
There is no economics in the paper. It is not economics. It is not about economists. It is about an anonymous economics message board. It belongs in a Women's Studies or Communications journal.
The pattern the author breathlessly interprets as evidence of widespread misogyny appears to actually be limited to a very small *proportion* of posts.
Regarding the key findings of pejorative terms used against women, it takes some effort for the reader to realize that only about 5% of the posts are classified as "female" (in the broadest definition) and then the key results are given in terms of models predicting that uncommon classification. It is critical that the paper provide descriptive statistics such that the reader can assess how commonly misogynistic and other worrisome terms are actually used. For example, if 2% of posts containing a "female" classifier such as "she" also contain the most common "female" word, "hotter," then (0.02)(0.05) = 0.001 of posts on the site contains the combination of those terms, and the fact that this combination is more common than other very uncommon combinations shouldn't distract from the observation that only one in a thousand posts contain this combination.
Can we get a paper on how women are getting affirmative action, too? Maybe that helps to explain lower tenure rates down the road. Dare I even mention that as a possibility. But if you actually care about women, instead of ideology, you’ll want to look at these issues objectively.
At least as many posts that have been made have also been deleted over the years. The mods sometimes delete more posts in a day than those that are actually left behind. The regulars complain about deletions all the time. Sometimes the mods delete innocuous posts, so the original posters will post purposefully inflammatory comments simply to retaliate against the mods, who don't always catch all of these. Those posts would have inevitably been picked up by Wu's algorithm. The innocuous posts would have been missed since they have been deleted. How do you control for this?
She seems to lack a basic grasp of EJMR culture and this bleeds through into her understanding of word usage. If you can't even get that right then don't bother even trying to write a paper about this:
She says "keen" is associated with men in an attempt to say that we here praise men when in reality "keen” is clearly referring to Steve Keen
She uses “bietches bitch bitches” as female classifiers but in reality these are more often than not referring to men
Also its funny that the paper picks up on "pregnancy" being a bad word for women when there is nothing wrong or negative with that word (nor can it apply to men, despite what “birthing people” proponents would have you believe)… but then it’s even funnier when you think about it you realize that its tied to Hoynes corruption re: a pregnancy paper
Cannot separate out “joke” posts like “Harvard tenured prof Mankiw is so professional” which gives a lot of professional points towards men. It’s sarcasm.
“Papa” is used as a male classifier. Do a quick search for the word papa and see how many come from references to the papa Wooldridge textbook. Discussions of textbooks rarely deviate toward discussion of the books’ appearance or hotness.
She mistakes "BB" to mean “Baby” when in fact everyone on EJMR knows that BB is a reference to the sensational job market candidate called BB Lacroix.
Why did she establish the cutoff at the top 10,000 words anyway? Why not 1,000 or 20,000? In fact, why didn't her advisor tell her to run a PCA just to see how words cluster?
The choice of using old data in the study (a random 4 year window) is a sign that they also cherry picked the data to get the result they wanted. I doubt her results are robust to different time windows.
In many ways, the work of a critic is easy. We risk very little, yet enjoy a position over those who offer up their work and their selves to our judgment. We thrive on negative criticism, which is fun to write and to read. But at the end of the day, do those criticisms matter? No: editors and refs had no option other than accepting her paper… it was a forgone conclusion. If they rejected it, all the public would see is that we are discriminating against her because she "attacks" a male-dominated profession. The all-male editorial board of Review of Economics and Statistics (RESTAT) had no choice but to accept this paper and quickly. Can you imagine the fallout on Twitter and elsewhere if they did not publish this? There would be public calls for explanations of why there are no women on the editorial board, etc. The editors are not stupid, and therefore this paper was surely handled by a "sympathetic" editor and "sympathetic" referees to give it a manageable R&R.
I feel particularly sad for everyone with a RESTAT paper that’s actually good… What a slap in the face! Man oh man! All those papers are now de-valued. This is a one-round-and-it's-waved-in, solo pub in RESTAT before she even started her first year PhD courses. Must be nice to have female Berkeley mafia privilege.
Which brings me to why this paper was so blackpilling: it represents the death of meritocracy. Why study hard and work on hard, important problems when it turns out at the end of the day, what matters the most in an academic career are your race and gender? The message this sends is: go for politically motivated propaganda which no one would have the guts to oppose in public and both get the attention and have your paper published.
Who was it that really stuck this dagger into the heart of meritocracy? I wasn’t Alice Wu, it was David Card. As she accidently let slip in an interview:
Interviewer: What led you to study EJMR?
Wu: Honestly, it was a bit unexpected. I was picking a research topic for my senior thesis. I had worked for David Card for a semester before and he was going to be my advisor. In a conversation while I was debating research topics, he pointed out that there was a controversy going on at the time regarding EJMR and asked whether I would be interested in studying that platform.
He set her up.
Read that above quote again.
The President of the American Economic Association set her up.
This whole thing was Card attacking EJMR because EJMR exposed his friends in the Berkeley mafia. It’s right there in the quote — presumably the controversy they are talking about was EJMR's criticism of unethical research practices by Hoynes. So he hid behind a small Asian undergrad girl to help out his cronies, and legit arguments about her corruption were then publicly dismissed because this place is a "cesspool of misogyny." A remarkably cunning and underhanded behavior. It’s akin to a detective saying "those people found corruption in the police, why won't you go and put some cocaine in one of their drawers."
What a bitch move, what a snake, what a pussy. He knew she would do it: after you've been brainwashed with PC stuff your whole life, when a future Nobel laureate (?) with 76,396 lifetime citations offers you a one-way ticket to Harvard and tells you to investigate "sexism" on a website you just do it no questions asked. "Sexism" is bad and you have to fight it.
So what can be done? Delusions of grandeur aside — much like Alexander Solzhenitsyn’s last word before leaving his homeland urging Soviet citizens as individuals to refrain from cooperating with the regime’s lies, so too do I urge you the reader to stand up and oppose the AEA/Berkeley lies. If you aren’t willing to do so publicly (how could I possibly ask you to give up your job, to make yourself a pariah in front of your colleagues, take food out of your family’s mouth, for being unable or unwilling to say “shibboleth?”), at least you are reading and sharing this article. At least you are choosing spiritual independence over spiritual servility.
And therein we find the simplest, the most accessible key to our liberation as economists: a personal nonparticipation in lies! In this way even the most timid can take a step toward spiritual independence by opposing the AEA mafia through passive resistance. If the silent majority march together on this path of passive resistance, the whole broken system will totter and collapse. We have nothing to lose but our reg monkey chains.
Yep, lib fellow travelers sure stand exposed after that Wu controversy -- probably the first time we've seen the mask slip a smidge.
This is the EJMR link that works for me.