Science Influential paper linking recessions and left-wing voting patterns retracted - "The retraction marks a rarity among economics papers, which research has shown are infrequently retracted compared to papers on other subjects."

Influential paper linking recessions and left-wing voting patterns retracted

Vox Efx via Wikimedia A highly cited economics paper that suggested people raised during recessions were more likely to vote for left-leaning political parties has been retracted, apparently due to…

retractionwatch.com

https://archive.is/oAn2p

A highly cited economics paper that suggested people raised during recessions were more likely to vote for left-leaning political parties has been retracted, apparently due to a coding error that rendered the results invalid.

The retraction marks a rarity among economics papers, which research has shown are infrequently retracted compared to papers on other subjects. The article appears to be the first in The Review of Economic Studies to have been retracted for a reason other than publisher error.

The study’s authors, Paola Giuliano and Antonio Spilimbergo, are economists at the Anderson School of Management at the University of California, Los Angeles and the International Monetary Fund, respectively. Giuliano is also the Chauncey J. Medberry Chair in Management at UCLA.

The paper, “Growing up in a Recession,” was published in November 2013. It has been cited 222 times, according to Clarivate’s Web of Science. Working papers from the World Bank and the Organisation for Economic Co-operation and Development have also cited the article.


The paper was retracted on January 11. According to the retraction notice:

The authors and editorial team are retracting this article because the original findings cannot be replicated, likely as a result of an inadvertent coding error. While the original codes and data sets are no longer available, new analysis with a markedly similar data set does not support the original results.

Neither the paper’s authors nor the editors of the journal responded to a request for comment from Retraction Watch.

The study has received a fair amount of media attention. An article in Salon, published soon after the final version of the study, was titled, “Has the recession spawned a generation of Democrats?” An earlier, working version of the article available in 2009 sparked an opinion piece in the New York Times from conservative columnist Ross Douthat. More recently, articles have cited the study in the context of the COVID-19 pandemic and the U.S. government’s response.

Retractions from economics journals have been rarer than for some other subjects, such as the biomedical sciences. A 2012 study examining retractions of economics papers for plagiarism found only six going back to 2009. But the Retraction Watch Database shows that around 700 economics papers were retracted between 2013 and December 2022.

Discussion of the retraction on Twitter was mixed. Economist Florian Ederer applauded the authors for retracting the paper.


Others were more critical of the retraction notice, particularly the assertion that the code used in the study no longer exists.

“I don’t really understand the reason,” wrote Tim Martens, an assistant professor of accounting at Bocconi University in Italy, in a reply to Ederer’s tweet. “Did their dogs eat the code?”

OP edit: Attached original study as a PDF.

 

[Kari Kamiya]

A highly cited economics paper that suggested people raised during recessions were more likely to vote for left-leaning political parties has been retracted, apparently due to a coding error that rendered the results invalid.

This is just like my glitchy voting machines!

 

[Tom Myers]

"Coding error" is code for Excel error.

 

[Rei is Shit.]

"apparently due to a coding error that rendered the results invalid."

And this is why you shouldn't trust economist in general. There is no way that 'coding' something like this is difficult.

 

[Weebburner]

So this make lefties look bad?
No its wrong, there was an error.
Okay whats the error?
The code its wrong.
Okay can you post the code so we can see the error?
code's gone sry.

 

[Penis Drager 2.0]

So this make lefties look bad?
No its wrong, there was an error.
Okay whats the error?
The code its wrong.
Okay can you post the code so we can see the error?
code's gone sry.

I'm not sure how, exactly, this study makes the left look "bad."
The conclusion was that growing up in a recession made you more likely to favor left wing politics in adulthood, not that left-wing voting causes recessions. Also note that the study in question was up for nearly a decade, and cited by left wing rags like The New York Times and The Guardian, before this retraction.

It does kinda scream incompetence, though, that both the code and the original dataset they used have gone missing.

 

[Desu Mountain]

I remember this paper. It was cited by lefties constantly as a counter to the "hard times make hard men" argument.

 

[Pixy]

We need to accelerate another recession so we can reattempt this study with a modern dataset.

 

[Dante Alighieri]

Recessions make more lefties, lefties make more recessions.

 

[wtfNeedSignUp]

How do you even apply statistics to cases like this in economics? There are way too few countries to sample from and their history, culture and politics is so vastly different it's impossible to remove bias

 

[Fomo Hoire]

I'm not sure how, exactly, this study makes the left look "bad."
The conclusion was that growing up in a recession made you more likely to favor left wing politics in adulthood, not that left-wing voting causes recessions. Also note that the study in question was up for nearly a decade, and cited by left wing rags like The New York Times and The Guardian, before this retraction.

It does kinda scream incompetence, though, that both the code and the original dataset they used have gone missing.

I'd say this doesn't immediately make them look bad, but because they're such howlers, their reaction to it will make them look bad.

Spoiler: Howlers

No, not the Harry Potter ones, though they sure screech like them and I am still referencing a children's book.

Animorphs' howlers

Essentially an alien race designed to be the ultimate warriors. Physically, they're like Wolverine but with a greater range of motion. (They have claws and regenerative abilities).

Mentally, they're a hive mind

The Howlers operated as a collective unit; the experiences of any one Howler were shared among all of them, so that every single one may become the best possible combatant.

Note that the collective unit comprises of the whole species, including all the Howlers to ever die. It's not bound to a generation or a region or a tribe, the collective memory is huge and it is powerful.

Howlers never have to worry about morale when they go out to casually commit genocide; they think it's fun, and even if they didn't, they don't have any memories of ever having lost. They believe they are unbeatable.

It's not true that they've never lost a battle, however. Their creator just kills off anyone who loses a battle before those memories can get uploaded to the collective memory.

They are 100% certain of the truth of the world: They are invincible. Nobody has ever experienced loss. And this truth is maintained through rigorous pruning of wrong thinkers before they can infect the hive mind. It's essentially a religious cult, except their God might be real (I don't remember if Crayak is a God or just far more powerful than humans).

And when the collective hive mind is finally polluted by wrong think, their entire species is deemed useless and is destroyed by the aforementioned maybe-God.

The degree of ideological purity the left requires resembles that of the Howlers. They are just as certain of the truth of the world as the Howlers are, because they have rejected the scientific method and shut off any possibility that what is Known may be updated with future research. The truth is the truth and it is static and it will never change. Destroy anyone and anything that may force that world view to crumble.

Now, how can the left address this without being forced to walk back their worship of The Experts and admitting that even the experts can be wrong? How can they ignore it without being anti-science, especially if it's something they used to claim was credible in the past? What response can possibly make the left look good after they've gone full cult with their The Science is Settled, Trust the Experts shtick?

 

[Morethanabitfoolish]

It does kinda scream incompetence, though, that both the code and the original dataset they used have gone missing.

Unless someone has been using the original as a template for others you would think they'd have spotted the error on the original so all of that should have been available. If they have been doing that and only caught it now, 10 years on, means there might be some others that need retracting too.

 

[Kheapathic]

Economists are professional niggers like lawyers and politicians; and should be dealt with accordingly.

What response can possibly make the left look good after they've gone full cult with their The Science is Settled, Trust the Experts shtick?

Unironically, mass self-deletion. If they want to look good, accept their howling has caused more permanent damage than imaginable; and if they truly are sorry for being so blind and destructive, than the only answer is to let the void take them. Because they're too dangerous to be left alive.

How do you even apply statistics to cases like this in economics? There are way too few countries to sample from and their history, culture and politics is so vastly different it's impossible to remove bias

You can't, because there are too many variables to consider when trying to apply political affiliation to a person. Where they were born, where they live, where they went to school, family/neighborhood environment and a whole host of other things affect people more than "Yeah, shit sucks right now." Especially when you consider most people don't pay attention to the economy when "being raised" as the article says. You pay attention to the economy when you get a job, realize how much is taken from you in taxes, and then realize where it goes. But even then, you might have a private opinion of how you really feel; but also a public opinion to not draw the ire of your friends/neighbors/etc... and you're not gonna voice your private opinions during a survey.

 

[IAmNotAlpharius]

Well this is odd.

I'm not sure how, exactly, this study makes the left look "bad."
The conclusion was that growing up in a recession made you more likely to favor left wing politics in adulthood, not that left-wing voting causes recessions. Also note that the study in question was up for nearly a decade, and cited by left wing rags like The New York Times and The Guardian, before this retraction.

It does kinda scream incompetence, though, that both the code and the original dataset they used have gone missing.

That’s what I find puzzling. How does the code and dataset disappear? It seems odd that there isn’t a copy of it somewhere. I know if I published something I’d want to hold onto that shit so it doesn’t get retracted. Losing a published paper like that would sting.

 

[duckbutter&toejamsandwich]

I'll play the conspiracy theorist and ask if maybe this got deleted because people are starting to notice the recession we are in seems intentionally produced? That maybe part of the intention is to push people to vote a certain way and globohomo didn't want anyone catching on?

Way more likely there never was any dataset or code, and when new data wasn't supporting this study someone asked for the code. To cover up that it was total bullshit they lied and said it was lost. It would be far from the first time this happened.

 

[Penis Drager 2.0]

That’s what I find puzzling. How does the code and dataset disappear? It seems odd that there isn’t a copy of it somewhere. I know if I published something I’d want to hold onto that shit so it doesn’t get retracted. Losing a published paper like that would sting.

The thing is, and shower me in rainbows for this, you'd think that the publishers would want a copy of the dataset and code to look over and store for archival purposes, even if inclusion in the actual study wouldn't be worth the bloat.
Perhaps this could be a lesson?

How do you even apply statistics to cases like this in economics? There are way too few countries to sample from and their history, culture and politics is so vastly different it's impossible to remove bias

I'd recommend doing a quick skim of the study. What they basically did was ran a correlational study on people's political views over time in the US and the state of the economy at the time, they then evaluated the same data internationally. Finally they address "heterogenous effects" which is basically meant to mean confounding variables. To the latter point, they state:

In order to test for this possibility, we include in our specification interaction terms between recessions during the impressionable years and different initial characteristics or different experiences during the impressionable years. The NLS72 is the most appropriate among our datasets because we can observe individuals at each point in time during the impressionable years.

Also note that "left wing" here is meant to broadly mean "a desire for stronger government control over the economy and/or wealth redistribution." Think "Political Compass" left wing.
Really, the whole thing isn't so far fetched and concluding that you simply "cannot study something" because it's "too complex" is little more than a barrier to scientific study. Confounding variables are, well, confounding. But if you can render those variables to the point of brownian motion by having a sufficiently diverse sample space, you can get a bit of a handle on things and glean some information regarding the one variable you're actually trying to research.

This was, on the face of it, hardly a homerun study that "proved" anything. But, if the correlation actually existed and wasn't evidently an artifact of fudged up code, it would have been informative and warranted future study. The core hypothesis that arises would be that people will generally become more open to government oversight on the economy if their formative years were fraught with economic disparity.

Now, how can the left address this without being forced to walk back their worship of The Experts and admitting that even the experts can be wrong?

Note that the article here is from Retraction Watch. It's literally a newsletter dedicated to reporting on papers that have been retracted.
It's honestly kinda cool that something like this exists. It's a common theme that the media will report endlessly on any "groundbreaking" study only to not say a damn word when the study gets retracted because that's "boring."

 

[Archie_Kimkicker]

Regardless of the validity or invalidity of this particular paper, economic research is not well known for its rigor.

 

[wtfNeedSignUp]

I'd recommend doing a quick skim of the study. What they basically did was ran a correlational study on people's political views over time in the US and the state of the economy at the time, they then evaluated the same data internationally. Finally they address "heterogenous effects" which is basically meant to mean confounding variables. To the latter point, they state:

Also note that "left wing" here is meant to broadly mean "a desire for stronger government control over the economy and/or wealth redistribution." Think "Political Compass" left wing.
Really, the whole thing isn't so far fetched and concluding that you simply "cannot study something" because it's "too complex" is little more than a barrier to scientific study. Confounding variables are, well, confounding. But if you can render those variables to the point of brownian motion by having a sufficiently diverse sample space, you can get a bit of a handle on things and glean some information regarding the one variable you're actually trying to research.

This was, on the face of it, hardly a homerun study that "proved" anything. But, if the correlation actually existed and wasn't evidently an artifact of fudged up code, it would have been informative and warranted future study. The core hypothesis that arises would be that people will generally become more open to government oversight on the economy if their formative years were fraught with economic disparity.

The problem is not that the sample size of people is small, it's that the sample size of countries is very small, and no matter how good your technique is to isolate variables, if it's constrained by low sample size then the results are worthless.

The country difference is usually not that big, but when the main element of the study will almost always be individual to that country (no two economic recessions are alike, ditto politics can vary wildly even with neighboring countries).

It's the same reason whenever there's a poll of "most Americans are pro-gun" and you see they sampled 10000 people from California, New York and other blue states. Even if they did their best to remove variables they still have a massive bias.

 

[pikachudidnothingwrong]

I can't wait for normies to wake up and realize that we are and have been in a place where scientific studies and research are "retracted", ignored, attacked and even punished for finding out the wrong facts or going against the preferred narrative.

Even the hard sciences are sick now. We have grants and approvals for things requiring critical race and letter people bullshit for even unrelated fields. (seriously.. space and physics societies are demanding race crap with the study of stars and space!)