2021 Nobel Prize for Economics
nobelprize.org
nobelprize.org
https://marginalrevolution.com/marginalrevolution/2021/10/th...
Check out Angrist's book: "Mastering 'Metrics: The Path from Cause to Effect" https://press.princeton.edu/books/ebook/9781400852383/master... https://www.masteringmetrics.com/
Edit: downvotes suggest some context may be needed. Pearl win the Turing award for casual inference but has been critical of the current Nobel recipients' methods.
> overlook[s] the two Fundamental Laws of causal inference.
1. The law of Counterfactuals and 2. The Law of Conditional Independence
You can find preprints from Pearl online or even some presentations that cover these definitions. His books as well.
I don't know how this applies to the work by the economists. I think Gelman at Columbia talks about Pearl and SCMs and the whole thing.
"2021 Economics Nobel: Natural experiments help answer important questions" (an in-page header)
...or...
"2021 Economics Nobel: Card, Angrist, & Imbens"
...or even if it fits...
"2021 Economics Nobel to Card, Angrist, & Imbens for natural experiments"
> Despite the enormous increase in labour supply, Card found no negative effects for Miami residents with low levels of education. Wages did not fall and unemployment did not increase relative to the other cities.
> while people who immigrated at an earlier time are negatively affected. One explanation for this is that the natives switch to jobs that require good native language skills, and where they do not have to compete with immigrants for jobs.
If this isn't terrible methodology I don't know what is.
Literally using bad data to say immigration doesn't affect the labour market, then saying it does for same-language jobs. So it obviously does but language is a barrier to entry for some jobs. Literally invalidates the first point.
> Card found no negative effects for Miami residents with low levels of education. Wages did not fall and unemployment did not increase relative to the other cities.
Additionally, the claims about specific types of effects on the labour market that you quote were from follow up studies and also don't conflict with the results of the earlier study.
Low-income residents on average, had no negative effects but follow up studies split up that low income group and found that one subgroup benefits and another subgroup suffers.
The line above the first I quoted: > Despite the enormous increase in labour supply, Card found no negative effects for Miami residents with low levels of education.
Then a few lines down (my first post), it said there were negative effects for previous immigrants with low levels of education. They made the assertion, then disproved it in the same paragraph, supposedly this is a highlight of the award?
It literally says an increase in immigration negatively affects the labour market for previous immigrants/workers who speak the same language. This != no negative effect...
The previous conclusion ignores a bunch of other factors.
You seem to be asserting that you can't make a statistical claim about a larger group unless that claim applies to every single possible subgroup? That is a rather ridiculous standard that would invalidate pretty much every single statistical claim ever made...
> The previous conclusion ignores a bunch of other factors.
Which ones?
Edit: You significantly edited your comment during my response. My response to your edits follows:
> They made the assertion, then disproved it in the same paragraph, supposedly this is a highlight of the award?
There is nothing disproving the original claim. The follow up studies performed more detailed analysis that looked at effects on subgroups to increase our understanding of the effects of immigration beyond what the original study found. This is a highlight because the original study was valuable not just for its results but for the methedology that facilitated a bunch of further research.
> To examine how this huge influx of workers affected the labour market in Miami, David Card compared the wage and employment trends in Miami with the evolution of wages and employment in four comparison cities.
> Despite the enormous increase in labour supply, Card found no negative effects for Miami residents with low levels of education. Wages did not fall and unemployment did not increase relative to the other cities. This study generated large amounts of new empirical work, and we now have a better understanding of the effects of immigration.
Look at the conclusion: "Card found NO negative effects for Miami residents with low levels of education." NO negative effects is quite the statement.
Now, the follow up studies:
> or example, follow-up studies have shown that increased immigration has a positive effect on income for many groups who were born in the country, while people who immigrated at an earlier time are negatively affected. One explanation for this is that the natives switch to jobs that require good native language skills, and where they do not have to compete with immigrants for jobs.
Now that's a negative effect. He made the assertion of NO negative effects.
In second year economics my prof would have failed me for such a basic conclusion. Look at influx of people, compare unemployment rates to other cities, make conclusion from that.
When immigration happens there's things happening on both sides of the supply/demand curve. Immigrants bring money, they also are labour. Different groups bring varying amounts of supply/demand. They compete for various jobs depending on their level of education (including language). There's also undocumented employment.
Anyhow, there's no mention of the wealth of the immigrants (often western nations are recipients of capital flight), no mention of aggregate demand, it says wages didn't fall but no mention of inflation and wage growth (ie. real wages), etc...
And if a study is so easily invalidated (sorry, elaborated upon), what's the point? Anyhow, it stuck out to me because it's an oft-cited study despite being incredibly basic and flawed (and many studies since more or less disproving it).
Anyhow, it's bad data (well, bad methodology from start to finish) because it draws a conclusion based on correlation between immigration and unemployment/wages without considering a ton of factors in between. It's like taking a chart of AAPL, GOOG, saying one depends on the other because they both go up during the same timeframe, ignoring everything else that's happening in the economy and tech industry.
Edit - it should also be mentioned, that literally the last ~5 or so years of politics in the west have centred around inequality, the plight of uneducated workers in the western world, globalism's effect of living conditions in the western world, etc... Populism on both the left and right is a thing because of it. It's pretty clear that a significant segment is being left behind. Wage stagnation IS a thing, and we know it because there's a lot of hindsight in the last 30 years. And Covid exacerbated inequality, with educated workers increasing their net worth drastically while lockdowns wreaked havoc on large segments of the economy. The cited study says absolutely nothing about the labour market, except that in one time and place, unemployment and wages didn't go down.
This is a hilarious comment. You are saying that your economics professor would have failed you for making a claim akin to what one of these three Economics Nobel laureates claimed in a very well known, classic article in the labor literature? I'm an economics professor and I would be delighted to see work like this from an undergrad!
>> Now that's a negative effect. He made the assertion of NO negative effects.
What you and later studies have identified is that there may be heterogeneous causal effects across groups. There is a large literature on this topic!
>> if a study is so easily invalidated...
I'm not worried about this. You have not "invalidated" this study with five minutes of thinking!
>>> it's bad data (well, bad methodology from start to finish) because it draws a conclusion based on correlation between immigration and unemployment/wages
1. It is not bad methodology. That is literally what half of the prize which went to Imbens and Angrist is about. This is a methodology which permits making causal claims in the absence of an experiment!
2. It is not drawing a conclusion based on a correlation.
For more information about what they actually did and how it works (and this is a big interesting topic!) you might benefit from the Nobel Foundation's scientific background paper on this prize, here: https://www.nobelprize.org/uploads/2021/10/advanced-economic...
The Nobel Peace Prize is awarded by the Nobel Peace Prize Committee, which is appointed by the Norwegian Parliament. They decided Obama was worthy.
The Nobel Prize in Economics (or, as everyone is fond of pointing out: "the Sveriges Riksbank Prize in Economic Science in Honor of Alfred Nobel") is awarded by the central bank of Sweden.
That's funny because the reason these researchers are so lauded is precisely because they found novel methodologies to avoid doing exactly the sort of purely correlative analysis that you describe.
I don't do reduced-form causal inference work myself, but I recognize that it is widespread and extremely influential throughout the profession.
You can of course argue that someone else should have received the prize. My own preferred candidates did not win. But this work was going to win eventually for sure and to argue that there's a lot of work from the last decade which deserved to win is not credible.
The original nation of the immigrant loses workers and consumers. Some workers in the host nation lost bargaining power. There's lost efficiency as the newly arrived immigrants generally accept a worse job than they had back home (see all the immigrants with degrees working menial jobs).
Free movement is a great idea, but it's essentially colonialism version 2. We take the best and brightest from developing nations and make them servants.