Noah Smith on David Card and the credibility revolution in economics
October 25, 2021 11:22 AM Subscribe
Noah Smith: The Econ Nobel we were all waiting for. "David Card, Joshua Angrist, and Guido Imbens have made economics a more scientific field." David Card is best-known for his research with Alan Krueger showing that a minimum-wage increase in New Jersey caused employment to rise rather than fall. A profile from the Globe and Mail. Bonus: This year's econ critics make a few good points.
From the first link:
posted by russilwvong at 11:29 AM on October 25, 2021 [4 favorites]
It would be great if we could use our brains for more than a second and not automatically spit out "economics isnt science" or "fake nobel!" (as if they are real things) or "econ is stupid because they assume rationality!".
posted by MisantropicPainforest at 11:30 AM on October 25, 2021 [11 favorites]
posted by MisantropicPainforest at 11:30 AM on October 25, 2021 [11 favorites]
okay, as long as I don't have to accept that economics is science.
posted by philip-random at 11:56 AM on October 25, 2021 [13 favorites]
posted by philip-random at 11:56 AM on October 25, 2021 [13 favorites]
Its a social science. Literally no one is saying its a science.
posted by MisantropicPainforest at 11:58 AM on October 25, 2021 [3 favorites]
posted by MisantropicPainforest at 11:58 AM on October 25, 2021 [3 favorites]
MisantropicPainforest
I think the problem is, and Noah Smith has discussed this on his Twitter account, that economics has a terrible PR problem. For most people, "economics" means (i) the Milton Friedman Chair of the Institute for Economic Liberty being trotted out every time any reform is proposed to spout "regulations bad, taxes bad"; (ii) the simple stuff we learn in high school which makes empirical findings like the ones described above hard to understand ("It's simple supply and demand, as wages go up hiring goes down, idiot!"); or (iii) pop Freakonomics-style "this counterintuitive thing will blow your mind" stuff. It's no wonder most people have a dim view of the field, and interesting stuff like this either never disseminates to a wider public audience or is often hidden behind seemingly-impenetrable academic jargon.
posted by star gentle uterus at 12:07 PM on October 25, 2021 [32 favorites]
I think the problem is, and Noah Smith has discussed this on his Twitter account, that economics has a terrible PR problem. For most people, "economics" means (i) the Milton Friedman Chair of the Institute for Economic Liberty being trotted out every time any reform is proposed to spout "regulations bad, taxes bad"; (ii) the simple stuff we learn in high school which makes empirical findings like the ones described above hard to understand ("It's simple supply and demand, as wages go up hiring goes down, idiot!"); or (iii) pop Freakonomics-style "this counterintuitive thing will blow your mind" stuff. It's no wonder most people have a dim view of the field, and interesting stuff like this either never disseminates to a wider public audience or is often hidden behind seemingly-impenetrable academic jargon.
posted by star gentle uterus at 12:07 PM on October 25, 2021 [32 favorites]
From the last link:
If you go to an empirical econ seminar, you’ll find that it consists mainly of economists challenging authors to justify their assumptions, and authors trying to convince the crowd.
This. Imagine an hour of: "But how do you know that assumption X is true?" "Couldn't your sample be biased because of Y?" "Did you test to see if your conclusion would still hold if you relaxed assumption Z?"
Q: How do you know the audience at the econ seminar is feeling unusually polite?
A: You get through the title slide without being asked a question.
posted by Mr.Know-it-some at 12:09 PM on October 25, 2021 [20 favorites]
If you go to an empirical econ seminar, you’ll find that it consists mainly of economists challenging authors to justify their assumptions, and authors trying to convince the crowd.
This. Imagine an hour of: "But how do you know that assumption X is true?" "Couldn't your sample be biased because of Y?" "Did you test to see if your conclusion would still hold if you relaxed assumption Z?"
Q: How do you know the audience at the econ seminar is feeling unusually polite?
A: You get through the title slide without being asked a question.
posted by Mr.Know-it-some at 12:09 PM on October 25, 2021 [20 favorites]
At some point they need to find another term to divide the science from the mystical bullshit. It is a bit like Alchemy and Chemistry in that regard.
posted by interogative mood at 12:44 PM on October 25, 2021 [6 favorites]
posted by interogative mood at 12:44 PM on October 25, 2021 [6 favorites]
it's cool and good how every discussion of the academic field of economics on metafilter basically always goes the same way regardless of the topic of the post
posted by dismas at 12:52 PM on October 25, 2021 [17 favorites]
posted by dismas at 12:52 PM on October 25, 2021 [17 favorites]
At some point they need to find another term to divide the science from the mystical bullshit. It is a bit like Alchemy and Chemistry in that regard.
Care to point to an academic economics article that's been published in, say, the last decade that is 'alchemy'?
The point is that Card and Imbens & Angrist won the Nobel for their work on empirical micro. They put casual explanations at the forefront of applied empirical research in econ. We went past X and Z are correlated, and into, X causes Y. This is a huge shift and all the graduate training in Econ and in related fields is influenced, sometimes quite heavily, by their work.
posted by MisantropicPainforest at 12:56 PM on October 25, 2021 [11 favorites]
Care to point to an academic economics article that's been published in, say, the last decade that is 'alchemy'?
The point is that Card and Imbens & Angrist won the Nobel for their work on empirical micro. They put casual explanations at the forefront of applied empirical research in econ. We went past X and Z are correlated, and into, X causes Y. This is a huge shift and all the graduate training in Econ and in related fields is influenced, sometimes quite heavily, by their work.
posted by MisantropicPainforest at 12:56 PM on October 25, 2021 [11 favorites]
Perhaps the reason it took this long was that Card’s conclusions in his famous minimum wage paper were so hard for many in the field to swallow. Card and Krueger (who sadly died before he could receive the prize) examined a 1992 minimum wage hike in New Jersey, and found that it didn’t result in a loss of jobs. They compared New Jersey to neighboring Pennsylvania, and found no job loss. They compared high-wage restaurants in New Jersey to low-wage restaurants — still, no job loss. Maybe even a little bit of job gain, actually.
Perhaps I'm misunderstanding, and I am very happy to have someone explain if I am, but am I correct that the belief previously is that raising the minimum wage would result in fewer jobs because of an implicit assumption that employers effectively had a big, fixed money pile called "salaries" and that paying more meant that they would need to fire other people because there was only so much money in the pile? And that what they discovered is that in fact employers hire the minimum number of people they need to hire and pay them as little as possible because there is not, in fact, a big "salaries" pile from which they pay but in fact they are trying to extract as much work as possible for as little money as possible so as to maximize their profits and if they could make the same amount of profit with fewer employees they would already be doing so? Because the first line of thinking feels a little...naive to me but I am VERY FAR from an economist so there is probably more to it that I'm not getting.
I am trying to engage with this in good faith and not make "har-dee-har economics" jokes but this is way outside my wheelhouse so I'm sorry if I just look like an idiot
posted by an octopus IRL at 1:04 PM on October 25, 2021 [16 favorites]
Perhaps I'm misunderstanding, and I am very happy to have someone explain if I am, but am I correct that the belief previously is that raising the minimum wage would result in fewer jobs because of an implicit assumption that employers effectively had a big, fixed money pile called "salaries" and that paying more meant that they would need to fire other people because there was only so much money in the pile? And that what they discovered is that in fact employers hire the minimum number of people they need to hire and pay them as little as possible because there is not, in fact, a big "salaries" pile from which they pay but in fact they are trying to extract as much work as possible for as little money as possible so as to maximize their profits and if they could make the same amount of profit with fewer employees they would already be doing so? Because the first line of thinking feels a little...naive to me but I am VERY FAR from an economist so there is probably more to it that I'm not getting.
I am trying to engage with this in good faith and not make "har-dee-har economics" jokes but this is way outside my wheelhouse so I'm sorry if I just look like an idiot
posted by an octopus IRL at 1:04 PM on October 25, 2021 [16 favorites]
as some additional reading on the so-called credibility revolution from about 11 years ago, the journal of economic perspectives (an ungated journal that is meant to be accessible to non-PhDs) had a symposium with a main article by Angrist and his frequent coauthor Pischke and responses from other economists. Angrist is now in the club of laureates with Chris Sims, who is also featured in that symposium, and who referred to the A&P article as containing "mainly nonsense" about macroeconomics.
posted by dismas at 1:09 PM on October 25, 2021 [4 favorites]
posted by dismas at 1:09 PM on October 25, 2021 [4 favorites]
and for those who are more interested in technical details, this ungated preprint of an article by a former professor of mine discusses some of what Angrist and Imbens's approach to econometrics formally means and places it in the larger context of what applied economics is usually trying to achieve.
posted by dismas at 1:12 PM on October 25, 2021 [1 favorite]
posted by dismas at 1:12 PM on October 25, 2021 [1 favorite]
On minimum wages: The general assumption is and has always been that employers are profit maximizing. So to hire someone, you must think that the benefits (increased revenue) would be greater than the cost (the wages). If you have to pay someone $8/hour, then you should think about whether their productivity is greater or less than $8. You will continue to hire until the productivity of the marginal worker - the last hired - is just above $8. In general, the productivity of each additional worker will decline. If the minimum wage then increases to $15, you will fire anyone who's productivity is between $8 and $15.
This seems (at least to me) intuitive and straightforward. It's actually much more complex to understand why it wouldn't be true. (There's a huge ongoing debate, but I even economists who are strongly in favor of higher minimum wages acknowledge that the traditional view is not wrong, but rather that it can be outweighed. If the minimum wage were $50, employment would definitely decline.)
For example, if the market is not completely competitive, the employer may have been making excess profit. If you're running the only mine in town, you can pay miners $8 even if they produce $16.
posted by Mr.Know-it-some at 1:24 PM on October 25, 2021 [7 favorites]
This seems (at least to me) intuitive and straightforward. It's actually much more complex to understand why it wouldn't be true. (There's a huge ongoing debate, but I even economists who are strongly in favor of higher minimum wages acknowledge that the traditional view is not wrong, but rather that it can be outweighed. If the minimum wage were $50, employment would definitely decline.)
For example, if the market is not completely competitive, the employer may have been making excess profit. If you're running the only mine in town, you can pay miners $8 even if they produce $16.
posted by Mr.Know-it-some at 1:24 PM on October 25, 2021 [7 favorites]
Perhaps I'm misunderstanding, and I am very happy to have someone explain if I am, but am I correct that the belief previously is that raising the minimum wage would result in fewer jobs because of an implicit assumption that employers effectively had a big, fixed money pile called "salaries" and that paying more meant that they would need to fire other people because there was only so much money in the pile
Not precisely, I don't think. The notion was more that the most basic model of labor markets implied that fast-food firms' demand for labor was decreasing in the price of labor. So if minimum wages increased that price, all else equal, we would expect that those firms (and the industry) would hire fewer workers. (Perhaps some fast-food restaurants would close down because they couldn't operate profitably, for instance, or have fewer workers on a given shift and those workers would have to do more work/fewer customers would bother to wait in line). To the extent that didn't seem to happen, it pointed to a different set of models as being more plausible explanations for how the industry operated. Card and Krueger discussed this a bit in section 8 of their 1994 paper. (One of the questions I often stick on intro macro exams is trying to get students to come up with a way of rationalizing the no-employment-response finding of Card and Krueger in the basic supply-and-demand model; it's possible, but relies on some things happening that Card and Krueger tried to rule out).
To an extent, I think the "nobel prize for showing minimum wages don't hurt employment" is not really what's going on here; it's more about the methodology that was pioneered by Card, his late coauthor Krueger, and others and formalized by Angrist and Imbens (and others). I'm not an applied microeconomist, so someone who is might have a different take.
posted by dismas at 1:28 PM on October 25, 2021 [13 favorites]
Not precisely, I don't think. The notion was more that the most basic model of labor markets implied that fast-food firms' demand for labor was decreasing in the price of labor. So if minimum wages increased that price, all else equal, we would expect that those firms (and the industry) would hire fewer workers. (Perhaps some fast-food restaurants would close down because they couldn't operate profitably, for instance, or have fewer workers on a given shift and those workers would have to do more work/fewer customers would bother to wait in line). To the extent that didn't seem to happen, it pointed to a different set of models as being more plausible explanations for how the industry operated. Card and Krueger discussed this a bit in section 8 of their 1994 paper. (One of the questions I often stick on intro macro exams is trying to get students to come up with a way of rationalizing the no-employment-response finding of Card and Krueger in the basic supply-and-demand model; it's possible, but relies on some things happening that Card and Krueger tried to rule out).
To an extent, I think the "nobel prize for showing minimum wages don't hurt employment" is not really what's going on here; it's more about the methodology that was pioneered by Card, his late coauthor Krueger, and others and formalized by Angrist and Imbens (and others). I'm not an applied microeconomist, so someone who is might have a different take.
posted by dismas at 1:28 PM on October 25, 2021 [13 favorites]
That was helpful, thank you! My assumption tends to be "employers, especially those paying minimum wage, will hire as few people as they possibly can to get the job done, sometimes even fewer, because they are shortsighted assholes" but I recognize that even if it really is that simple, and I know it's not, it's still important to be able to model that behavior.
posted by an octopus IRL at 1:33 PM on October 25, 2021 [3 favorites]
posted by an octopus IRL at 1:33 PM on October 25, 2021 [3 favorites]
Publicly traded companies are run by stockholders, and they demand maximum return; there is no consideration given to profitability five years from now or even one year from now, stockholders demand maximum return in the upcoming quarter: therefore; pay employees as little as possible, not because of cost, but to maximize return.
The price of a burger has nothing to do with how much a fast food chain is paying it's employees; the price of a burger is what the market will bear. Raising wages does not cut into the margins necessary to set those prices, it only cuts into stock returns.
posted by kzin602 at 1:44 PM on October 25, 2021 [1 favorite]
The price of a burger has nothing to do with how much a fast food chain is paying it's employees; the price of a burger is what the market will bear. Raising wages does not cut into the margins necessary to set those prices, it only cuts into stock returns.
posted by kzin602 at 1:44 PM on October 25, 2021 [1 favorite]
as few people as they possibly can to get the job done
IANAE, but there isn’t "a" job for a real company, yes? They might have more or fewer customers for more or less money each. So the marginal hire is there to work for a next possible customer.
posted by clew at 1:48 PM on October 25, 2021 [3 favorites]
IANAE, but there isn’t "a" job for a real company, yes? They might have more or fewer customers for more or less money each. So the marginal hire is there to work for a next possible customer.
posted by clew at 1:48 PM on October 25, 2021 [3 favorites]
Publicly traded companies are run by stockholders, and they demand maximum return; there is no consideration given to profitability five years from now or even one year from now, stockholders demand maximum return in the upcoming quarter:
This isn't really true. Stockholders are usually at least somewhat rational, just like managers and executives, and they have rational incentives to not just look at the short term.
When you own a share of a company, its intrinsic value comes from the fact that you own a fraction of both (a) the company's assets and (b) all of the profits/returns it expects to make in the future. The stock price fluctuates largely because of disagreements, differences of opinion, and changing information about how much those future profits will be worth.
In theory, a company could at any time choose to liquidate itself, sell off all its assets, and distribute the resulting cash to investors. But most investors wouldn't want to buy shares in a company that planned to do that, because its expected future profit would be zero. By the same reasoning, a company that decided not to care about long-term profitability would lose value now, because people would take into account the lower expected future dividends when deciding whether to buy it.
posted by teraflop at 2:01 PM on October 25, 2021 [5 favorites]
This isn't really true. Stockholders are usually at least somewhat rational, just like managers and executives, and they have rational incentives to not just look at the short term.
When you own a share of a company, its intrinsic value comes from the fact that you own a fraction of both (a) the company's assets and (b) all of the profits/returns it expects to make in the future. The stock price fluctuates largely because of disagreements, differences of opinion, and changing information about how much those future profits will be worth.
In theory, a company could at any time choose to liquidate itself, sell off all its assets, and distribute the resulting cash to investors. But most investors wouldn't want to buy shares in a company that planned to do that, because its expected future profit would be zero. By the same reasoning, a company that decided not to care about long-term profitability would lose value now, because people would take into account the lower expected future dividends when deciding whether to buy it.
posted by teraflop at 2:01 PM on October 25, 2021 [5 favorites]
I mean, it could be true. We could write down competing models of 'behavioral' short-termist shareholders/managers versus the canonical model of 'rational' equity holders who care about the present discounted value of dividends, and then try to characterize how firms would respond differently to different sorts of changes in the environment (e.g., a binding change in minimum wage laws). The difference in qualitative predictions isn't obvious to me, but I also haven't thought about it very hard.
posted by dismas at 2:06 PM on October 25, 2021
posted by dismas at 2:06 PM on October 25, 2021
As a physical scientist and anti-capitalist, making fun of the priors and methods in econ papers is something of a hobby. But, even snarky assholes have to admit that economists often do important and interesting things. Cheers!
posted by eotvos at 2:18 PM on October 25, 2021 [8 favorites]
posted by eotvos at 2:18 PM on October 25, 2021 [8 favorites]
Ok but isn't this default assumption of rationality (or some nearby approximation to it) the very thing that people make fun of economists for? Rather than assuming all markets are competitive and rational until a perfectly crafted study proves otherwise, the burden of proof really ought to go in the other direction.
posted by Pyry at 2:18 PM on October 25, 2021 [5 favorites]
posted by Pyry at 2:18 PM on October 25, 2021 [5 favorites]
Anyhow, as Philip Mirowski has pointed out, there is no Nobel Prize in Economics. There is, however, the "Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel". What it lacks in Nobel-ity it makes up for in length.
posted by sneebler at 2:20 PM on October 25, 2021 [2 favorites]
posted by sneebler at 2:20 PM on October 25, 2021 [2 favorites]
This seems (at least to me) intuitive and straightforward. It's actually much more complex to understand why it wouldn't be true.
I'm not an economist, but I'd suggest there are some other elements involved that looking at wages alone doesn't account for. With businesses that aren't competing with an online market, a rising minimum wage increases the cost of doing business for all business owners, thus minimizing any competitive advantage one business has over the other in employee cost for the same products or services.
To the extent these businesses are desirable to the community they inhabit, the added costs can be either passed on to consumers or factored in as the price of doing business, much like rent. The additional cost of a few dollars and hour for a small group of employees doesn't necessarily require much in the way of added pricing to be made up to maintain the same profit rate as before the wage increase, split among many items it may not be noticeable to the customer or make much difference in their purchasing to a certain point.
Higher wages also can mean more money being spent in the local economy, so if all employers are paying more, they don't necessarily suffer from added costs but some might benefit from additional sales, and some will potentially lose, but that's always the case.
There are businesses and industries where the additional costs might provide a competitive disadvantage if they are competing outside the area of higher labor costs, and there's been some studies suggesting rent increases follow wage increases, which can cause a host of additional problems for employees and employers, so there's more to the issue than in studying business costs and hiring for businesses dependent on customer locality. I'd also suggest there are things to consider about who it is that chooses to run a small to medium sized business along the lines of why and how they make that choice as it informs a lot about how they'll operate that goes beyond the basic assumption of rational choice models I think.
posted by gusottertrout at 2:41 PM on October 25, 2021 [2 favorites]
I'm not an economist, but I'd suggest there are some other elements involved that looking at wages alone doesn't account for. With businesses that aren't competing with an online market, a rising minimum wage increases the cost of doing business for all business owners, thus minimizing any competitive advantage one business has over the other in employee cost for the same products or services.
To the extent these businesses are desirable to the community they inhabit, the added costs can be either passed on to consumers or factored in as the price of doing business, much like rent. The additional cost of a few dollars and hour for a small group of employees doesn't necessarily require much in the way of added pricing to be made up to maintain the same profit rate as before the wage increase, split among many items it may not be noticeable to the customer or make much difference in their purchasing to a certain point.
Higher wages also can mean more money being spent in the local economy, so if all employers are paying more, they don't necessarily suffer from added costs but some might benefit from additional sales, and some will potentially lose, but that's always the case.
There are businesses and industries where the additional costs might provide a competitive disadvantage if they are competing outside the area of higher labor costs, and there's been some studies suggesting rent increases follow wage increases, which can cause a host of additional problems for employees and employers, so there's more to the issue than in studying business costs and hiring for businesses dependent on customer locality. I'd also suggest there are things to consider about who it is that chooses to run a small to medium sized business along the lines of why and how they make that choice as it informs a lot about how they'll operate that goes beyond the basic assumption of rational choice models I think.
posted by gusottertrout at 2:41 PM on October 25, 2021 [2 favorites]
The price of a burger has nothing to do with how much a fast food chain is paying it's employees; the price of a burger is what the market will bear. Raising wages does not cut into the margins necessary to set those prices, it only cuts into stock returns.
I don't think this follows as neatly as you imply: if the price of a meal cannot be raised and must come from profits, there is a cliff beyond which raising wages doesn't result in higher wages but zero wages from zero hours worked. But not every restaurant has equal margins; some will shut down first, and every time that happens the ones remaining have a bit more room to raise prices. And it will probably be the chain restaurants that survive the Franchise Wars, not the owner operated locations.
But the entire point of this Nobel award is that empirical data matters. We can't just play the Laffer curve reductio ad absurdia card, theory crafting about what would happen if we required McDonalds pay 100 dollars an hour and expect to have any predictive power about realistic minimum wage law increases.
posted by pwnguin at 2:49 PM on October 25, 2021 [6 favorites]
I don't think this follows as neatly as you imply: if the price of a meal cannot be raised and must come from profits, there is a cliff beyond which raising wages doesn't result in higher wages but zero wages from zero hours worked. But not every restaurant has equal margins; some will shut down first, and every time that happens the ones remaining have a bit more room to raise prices. And it will probably be the chain restaurants that survive the Franchise Wars, not the owner operated locations.
But the entire point of this Nobel award is that empirical data matters. We can't just play the Laffer curve reductio ad absurdia card, theory crafting about what would happen if we required McDonalds pay 100 dollars an hour and expect to have any predictive power about realistic minimum wage law increases.
posted by pwnguin at 2:49 PM on October 25, 2021 [6 favorites]
Ok but isn't this default assumption of rationality (or some nearby approximation to it) the very thing that people make fun of economists for? Rather than assuming all markets are competitive and rational until a perfectly crafted study proves otherwise, the burden of proof really ought to go in the other direction.
Who is assuming that all markets are competitive and rational here?
posted by MisantropicPainforest at 2:50 PM on October 25, 2021
Who is assuming that all markets are competitive and rational here?
posted by MisantropicPainforest at 2:50 PM on October 25, 2021
Anyhow, as Philip Mirowski has pointed out, there is no Nobel Prize in Economics. There is, however, the "Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel". What it lacks in Nobel-ity it makes up for in length.
It's always weird to me to see this trotted out.
I don't think "Alfred Nobel didn't think economics would make a useful distraction from the millions of people he made the weapons to kill" is the cutting criticism people think it is?
Nobel wasn't some sort of expert in which fields are worthy and which ones aren't. He was an explosives chemist who invented dynamite, which has civilian and military uses, and then made shitloads of money selling ballistite and guns to anyone who would pay. As far as we can tell, the prizes exist because when his brother died, at least one newspaper published the pre-written obituary for the wrong Nobel brother, so he got to see MERCHANT OF DEATH DIES and was embarrassed. AFAICT, not embarrassed enough to actually stop selling weapons or otherwise amend his life Scrooge-style, but enough to set up the prizes in his will.
posted by GCU Sweet and Full of Grace at 3:18 PM on October 25, 2021 [19 favorites]
It's always weird to me to see this trotted out.
I don't think "Alfred Nobel didn't think economics would make a useful distraction from the millions of people he made the weapons to kill" is the cutting criticism people think it is?
Nobel wasn't some sort of expert in which fields are worthy and which ones aren't. He was an explosives chemist who invented dynamite, which has civilian and military uses, and then made shitloads of money selling ballistite and guns to anyone who would pay. As far as we can tell, the prizes exist because when his brother died, at least one newspaper published the pre-written obituary for the wrong Nobel brother, so he got to see MERCHANT OF DEATH DIES and was embarrassed. AFAICT, not embarrassed enough to actually stop selling weapons or otherwise amend his life Scrooge-style, but enough to set up the prizes in his will.
posted by GCU Sweet and Full of Grace at 3:18 PM on October 25, 2021 [19 favorites]
Care to point to an academic economics article that's been published in, say, the last decade that is 'alchemy'?
Check out the syllabi used in high school level economics classes. It is basically like they are teaching medicine from the perspective of balancing humors and conducting seminars in the case for using leaches. It’s bad.
posted by interogative mood at 3:27 PM on October 25, 2021 [5 favorites]
Check out the syllabi used in high school level economics classes. It is basically like they are teaching medicine from the perspective of balancing humors and conducting seminars in the case for using leaches. It’s bad.
posted by interogative mood at 3:27 PM on October 25, 2021 [5 favorites]
Ok but isn't this default assumption of rationality (or some nearby approximation to it) the very thing that people make fun of economists for?
Sometimes, people make fun of economists for the half-remembered definition of rationality from the econ class they were required to sit through in their first semester of college, yeah, but I don't feel that this is really an indictment of the field of inquiry. For example, I wouldn't say that 'markets' are rational, or that markets that have rational agents in them must necessarily be competitive. They're different concepts.
Rationality is, in some sense, a term of art. There are stronger and weaker definitions of it, and economics research has, for decades at this point, explored and tested aspects of the theory. Psychologically-informed "behavioral" economics is pretty mainstream. We could have competitive markets without rationality, or rationality without assuming perfect competition.
Anyhow, as Philip Mirowski has pointed out, there is no Nobel Prize in Economics.
ah, dang, pack it in fellas, they know the cheat code to beating economists
posted by dismas at 3:27 PM on October 25, 2021 [10 favorites]
Sometimes, people make fun of economists for the half-remembered definition of rationality from the econ class they were required to sit through in their first semester of college, yeah, but I don't feel that this is really an indictment of the field of inquiry. For example, I wouldn't say that 'markets' are rational, or that markets that have rational agents in them must necessarily be competitive. They're different concepts.
Rationality is, in some sense, a term of art. There are stronger and weaker definitions of it, and economics research has, for decades at this point, explored and tested aspects of the theory. Psychologically-informed "behavioral" economics is pretty mainstream. We could have competitive markets without rationality, or rationality without assuming perfect competition.
Anyhow, as Philip Mirowski has pointed out, there is no Nobel Prize in Economics.
ah, dang, pack it in fellas, they know the cheat code to beating economists
posted by dismas at 3:27 PM on October 25, 2021 [10 favorites]
I remember sitting through Macroeconomics in University, and watching the prof draw simple curves to "explain" economic processes; and I'm sitting there thinking the economy is this complex, thousand-variable, self-modifying feedback engine, and you've just reduced it to a line on a two-dimensional plane.
It seems to be way more interesting to attack those simple curves head on. Here's the classic supply and demand graph. Give four examples of when it does *not* hold true.
posted by storybored at 6:03 PM on October 25, 2021
It seems to be way more interesting to attack those simple curves head on. Here's the classic supply and demand graph. Give four examples of when it does *not* hold true.
posted by storybored at 6:03 PM on October 25, 2021
Two years ago I had an economics professor at a major Canadian university tell me I was wrong about what certain major multinationals were going to do on pricing in a particular situation because "price elasticity proves I was wrong". When they went ahead and did it anyway, I called him up and asked whether he had revised his thinking. His response was that their actions proved that they were not profit maximizing, therefore microeconomics did not apply and they were not worth studying.
He refused to be willing to consider that there were more considerations that could drive a pricing decision than price elasticity. That experience did not leave me particularly impressed with how economic theorists matched up with reality.
posted by sabraonthehill at 6:51 PM on October 25, 2021 [4 favorites]
He refused to be willing to consider that there were more considerations that could drive a pricing decision than price elasticity. That experience did not leave me particularly impressed with how economic theorists matched up with reality.
posted by sabraonthehill at 6:51 PM on October 25, 2021 [4 favorites]
It's funny how the example of McDonalds always raises itself. That's where I first learned about applied economic concepts, at 15, when my manager—studying econ at uni—used to wander about shouting 'productivity, productivity'! (He meant 'faster, faster', of course).
It's not the concept of rationality that's in question here or the rigour of teaching economic principles as mathematical relationships, because every discipline has to start explaining its metaphors over-simplified somewhere, it's that economics has a unique relationship to power that not many others do. (Not many people have been made redundant after a philological-semiotic analysis). There are honestly relatively few actual trained economists in the world, but there are plenty of untrained people using economic language in self-interest: 'Fiasco, we need you to be more productive, so clock off before you clean up and pack the kitchen'. That's an economic argument, and it's of the sordid kind most ordinary people associate with economic thought, and it's no wonder that we automatically pat our wallet pocket in when we hear one like it. Not being trained in economics, I didn't have a response in the same language that would answer, and 'I don't really want to work faster for the same or no money' doesn't have an empirical or theoretical basis.
I mean of course it's true that economics has hundreds of years of history of trying to understand and explain human behaviours relating to the things we value and experience as plenty or scarcity, that it's a series of metaphors to understand the way people experience value and desire in collective ways, and that it's got a deep interrelationships with other sciences, social sciences and humanities, yes including history and literature. But it's also a moral social science about what should be as well as what is, it was moral since long before Adam Smith used the word, and there's really no separating the good and rigorous economists theorising/empiricising about human welfare using economic concepts, from the McDonalds managers using economic language to steal from teenagers.
Noah Smith (last link):
posted by Fiasco da Gama at 6:53 PM on October 25, 2021 [8 favorites]
It's not the concept of rationality that's in question here or the rigour of teaching economic principles as mathematical relationships, because every discipline has to start explaining its metaphors over-simplified somewhere, it's that economics has a unique relationship to power that not many others do. (Not many people have been made redundant after a philological-semiotic analysis). There are honestly relatively few actual trained economists in the world, but there are plenty of untrained people using economic language in self-interest: 'Fiasco, we need you to be more productive, so clock off before you clean up and pack the kitchen'. That's an economic argument, and it's of the sordid kind most ordinary people associate with economic thought, and it's no wonder that we automatically pat our wallet pocket in when we hear one like it. Not being trained in economics, I didn't have a response in the same language that would answer, and 'I don't really want to work faster for the same or no money' doesn't have an empirical or theoretical basis.
I mean of course it's true that economics has hundreds of years of history of trying to understand and explain human behaviours relating to the things we value and experience as plenty or scarcity, that it's a series of metaphors to understand the way people experience value and desire in collective ways, and that it's got a deep interrelationships with other sciences, social sciences and humanities, yes including history and literature. But it's also a moral social science about what should be as well as what is, it was moral since long before Adam Smith used the word, and there's really no separating the good and rigorous economists theorising/empiricising about human welfare using economic concepts, from the McDonalds managers using economic language to steal from teenagers.
Noah Smith (last link):
By teaching kids that econ needs to be grounded in evidence, CORE will help them understand what we do and don’t know about how the economy worksHere's to a basis of empirical data to inquiry, in the seminar room as in the fast food kitchens!
posted by Fiasco da Gama at 6:53 PM on October 25, 2021 [8 favorites]
I’m not generally one to always grasp at the simplest of explanations, because I’ve seen how they are often wrong.
That said, I’m awfully tempted to surmise that the reason that employment didn’t drop when the minimum wage was increased can easily be explained by the possibility that there was so much exploitation of labor already built into the equation that raising the cost of labor a bit didn’t really threaten any owner’s business model, appreciably.
Maybe owners and shareholders took a haircut on the net profits, but they were still raking it in, so no real need to lay people off and, in so doing, kill the goose that was laying those golden eggs.
I have to believe it was rather more complex than that, though.
posted by darkstar at 8:49 PM on October 25, 2021 [3 favorites]
That said, I’m awfully tempted to surmise that the reason that employment didn’t drop when the minimum wage was increased can easily be explained by the possibility that there was so much exploitation of labor already built into the equation that raising the cost of labor a bit didn’t really threaten any owner’s business model, appreciably.
Maybe owners and shareholders took a haircut on the net profits, but they were still raking it in, so no real need to lay people off and, in so doing, kill the goose that was laying those golden eggs.
I have to believe it was rather more complex than that, though.
posted by darkstar at 8:49 PM on October 25, 2021 [3 favorites]
Nobel wasn't some sort of expert in which fields are worthy and which ones aren't.
How much more pathetic does that make economists, for inventing this prize and trying to cosplay it like one of the others?
There are plenty of fields that don't have a Nobel Prize and make do perfectly well: mathematics has been a contentious omission since the beginning; mathematicians solved this by going and implementing their own prize. There's a compelling case for expanding the scope of the literature prize to include, say, filmmaking, but that field has similarly decided it's content to recognize its distinguished practitioners in other ways.
Economics is, to my knowledge, the only field that has invented a prize and named it literally after Nobel. They're certainly the only field to have thrown enough money at the Foundation to have it mentioned in the same breath as the original prizes, thereby overcoming nearly unanimous opposition to the idea from Nobel's own descendants. There's really no way of interpreting this aside from as a cynical attempt to hijack cachet from the "real" prizes onto their own field which (as this thread has made clear) still struggles somewhat in the reputability department. That lack of reputability may or may not reflect the field's actual merits (for sake of argument, let's say it does not), but such a transparent attempt to bridge the gap by piggybacking on existing fame does them no favors.
In conclusion, it's possible for all of the following to be simultaneously true:
1) The Nobel system, as originally conceived, is fraught, for reasons including those you mention, as well as things like when half the selection committee worked at Bell Labs and kept fixing it so their buddies would win. Henry Kissinger got a fucking Peace Prize, for fuck's sake.
2) Economics is a worthy field, and eminent contributors thereto are completely deserving of distinction similar to that which attends recipients of the actual Nobel Prizes, Fields Medal, Turing Award, etc.
3) (1) and (2) together do not mean that the Economics prize, as currently construed, is to be taken on the same terms as the original prizes. I'd personally rather it be renamed and reconstituted in a form that can't be in any way mistaken for pretension to the original prizes or to any of their attendant pageantry, but...
4) failing that, the least we can do is to continue pointing out the asterisk: is not ACTUALLY a Nobel Prize. Which is why this happens around this time every year.
posted by 7segment at 9:37 PM on October 25, 2021 [7 favorites]
How much more pathetic does that make economists, for inventing this prize and trying to cosplay it like one of the others?
There are plenty of fields that don't have a Nobel Prize and make do perfectly well: mathematics has been a contentious omission since the beginning; mathematicians solved this by going and implementing their own prize. There's a compelling case for expanding the scope of the literature prize to include, say, filmmaking, but that field has similarly decided it's content to recognize its distinguished practitioners in other ways.
Economics is, to my knowledge, the only field that has invented a prize and named it literally after Nobel. They're certainly the only field to have thrown enough money at the Foundation to have it mentioned in the same breath as the original prizes, thereby overcoming nearly unanimous opposition to the idea from Nobel's own descendants. There's really no way of interpreting this aside from as a cynical attempt to hijack cachet from the "real" prizes onto their own field which (as this thread has made clear) still struggles somewhat in the reputability department. That lack of reputability may or may not reflect the field's actual merits (for sake of argument, let's say it does not), but such a transparent attempt to bridge the gap by piggybacking on existing fame does them no favors.
In conclusion, it's possible for all of the following to be simultaneously true:
1) The Nobel system, as originally conceived, is fraught, for reasons including those you mention, as well as things like when half the selection committee worked at Bell Labs and kept fixing it so their buddies would win. Henry Kissinger got a fucking Peace Prize, for fuck's sake.
2) Economics is a worthy field, and eminent contributors thereto are completely deserving of distinction similar to that which attends recipients of the actual Nobel Prizes, Fields Medal, Turing Award, etc.
3) (1) and (2) together do not mean that the Economics prize, as currently construed, is to be taken on the same terms as the original prizes. I'd personally rather it be renamed and reconstituted in a form that can't be in any way mistaken for pretension to the original prizes or to any of their attendant pageantry, but...
4) failing that, the least we can do is to continue pointing out the asterisk: is not ACTUALLY a Nobel Prize. Which is why this happens around this time every year.
posted by 7segment at 9:37 PM on October 25, 2021 [7 favorites]
Frankly, if I accept premises (1) and (2) - that the Nobel prizes are respectable and that the field of economics is respectable, or that they are both just short of respectable - it leads me rather swiftly to the conclusion that the distinction between the economics prize and the rest of the Nobels is not particularly important to argue about.
posted by atoxyl at 10:49 PM on October 25, 2021 [3 favorites]
posted by atoxyl at 10:49 PM on October 25, 2021 [3 favorites]
Personally, I think it is hilarious that some economists have created a counterfeit Nobel Prize that some then respect, compete for, and defend. It's almost prefect applied economics
posted by srboisvert at 1:02 AM on October 26, 2021
posted by srboisvert at 1:02 AM on October 26, 2021
Mod note: Suggestion: Perhaps folks could get back to discussing the actual work and its effect on the field rather than getting tediously mired in a circular argument about economics' lack of credibility as a science when the actual post already submits this as the basis of the discussion. You may agree or disagree that the Card / Angrist / Imbens work raises the level in this regard, or maybe you have some thoughts on the validity of empirical evidence as applied in economics, or you may have thoughts about the work itself, or you could comment on the thoughts of people here who have already brought up ideas on these points in the thread, or you could just go on to a different post, since the post / article already establishes that economics has a credibility problem, so various repetitions of this sentiment really shouldn't become the whole conversation here.
posted by taz (staff) at 3:33 AM on October 26, 2021 [11 favorites]
posted by taz (staff) at 3:33 AM on October 26, 2021 [11 favorites]
I understand the minimum wage issue now, but how do employers determine the value of work that an employee does? The joke in IT is that it never improves the balance sheet and whether things are either working ("what are we paying you for then?") or broken ("what are paying you for then?") it's always a money loss. I do data analysis these days for a fashion company with its own stores. I couldn't tell you how much money my department makes the company either. The only workers I can point to and say "they definitely bring in money" are the sales people in the stores. Everyone else is a (to steal a term) REMF their front line work (analogy deliberately chosen during the pandemic). And even then, the amount earned by each store employee is in part dictated by who walks in the store that day and if they are the lucky one to snag a whale and create a personal connection with them.
How do economists determine that an the next hired employee will or will not create more marginal value if employers can't? In not arguing against rationality (if they could determine it, they'd use it), but rather a complete lack of information that only a place like Amazon, with its computing power, might be able to strive for.
posted by Hactar at 6:21 AM on October 28, 2021 [1 favorite]
How do economists determine that an the next hired employee will or will not create more marginal value if employers can't? In not arguing against rationality (if they could determine it, they'd use it), but rather a complete lack of information that only a place like Amazon, with its computing power, might be able to strive for.
posted by Hactar at 6:21 AM on October 28, 2021 [1 favorite]
The minimum wage paper and the research of these economists was published in 1994. Yet it doesn’t seem to have changed the debate or the policy at least in the United States. We haven’t raised the minimum wage since 2009 at the Federal level.
posted by interogative mood at 11:39 AM on October 28, 2021
posted by interogative mood at 11:39 AM on October 28, 2021
How do economists determine that an the next hired employee will or will not create more marginal value if employers can't?
Well, economists don't determine that, they rely on employers to, on the tenous rational actor assumptions well discussed already.
So how do employers measure marginal value? Well, one way is to look at store revenue on an hourly basis. You know the cost of goods sold, and you know the hourly wages of staff. It should be pretty simple therefore, to measure whether it's profitable to remain open at 10pm, or if you should close at 9. It's a little harder to look at the data and prove out that staying open an extra hour longer might be profitable, but if your data suggests it might be, you can take the risk and find out. At rush hours, queue lengths and walk-aways are what I would look at. Of course, it's really hard to hire for just lunch and dinner, so there are a number of demand management strategies used to shift demand away from peak. Think telco 'free nights and weekends', matinees, and happy hours, or the oft-maligned 'surge pricing' from Uber.
As you point out, this gets massively harder in engineering, where labor is amortized across sales. In Big Tech, you have such massive installs that the marginal gain from implementing a 1 percent improvement pays your annual wage, even if it only takes a week to identify and deploy. Data analysis teams should be able to justify their marginal hires pretty easily by finding cost savings, same as any manager. Where things struggle IMO is ITsec and internal IT. ITsec has seemingly unbounded costs available, very little data to go on, and the customer impact is basically only black swans. Internal IT typically suffers from a lack of cost accounting, which is kind of the genius of AWS: one team's cost is another team's revenue.
posted by pwnguin at 2:40 PM on October 28, 2021 [1 favorite]
Well, economists don't determine that, they rely on employers to, on the tenous rational actor assumptions well discussed already.
So how do employers measure marginal value? Well, one way is to look at store revenue on an hourly basis. You know the cost of goods sold, and you know the hourly wages of staff. It should be pretty simple therefore, to measure whether it's profitable to remain open at 10pm, or if you should close at 9. It's a little harder to look at the data and prove out that staying open an extra hour longer might be profitable, but if your data suggests it might be, you can take the risk and find out. At rush hours, queue lengths and walk-aways are what I would look at. Of course, it's really hard to hire for just lunch and dinner, so there are a number of demand management strategies used to shift demand away from peak. Think telco 'free nights and weekends', matinees, and happy hours, or the oft-maligned 'surge pricing' from Uber.
As you point out, this gets massively harder in engineering, where labor is amortized across sales. In Big Tech, you have such massive installs that the marginal gain from implementing a 1 percent improvement pays your annual wage, even if it only takes a week to identify and deploy. Data analysis teams should be able to justify their marginal hires pretty easily by finding cost savings, same as any manager. Where things struggle IMO is ITsec and internal IT. ITsec has seemingly unbounded costs available, very little data to go on, and the customer impact is basically only black swans. Internal IT typically suffers from a lack of cost accounting, which is kind of the genius of AWS: one team's cost is another team's revenue.
posted by pwnguin at 2:40 PM on October 28, 2021 [1 favorite]
Also the question of whether firms optimize at the margin when making decisions is an old discussion in economics -- commonly referred to as 'the marginalist controversy' in the history of thought. Milton Friedman's 1953 essay on the methodology of positive economics -- which remains influential among economists -- discusses it. His argument, briefly, is that theory should be judged on its predictions, not whether individual underlying assumptions are realistic. (In his case, he was uninterested in survey evidence about how firms claimed to set prices, but rather whether assuming they maximized profits by setting marginal product equal to marginal cost lead to reasonable predictions about their behavior). Reasonably, some people are unconvinced by this argument on some level, but to an extent we're rehashing arguments our intellectual (great)-grandparents had.
my own view as an applied macroeconomist is that I'm happy to entertain alternative assumptions about how people and firms behave and take them to the data. for example, this paper (I've had difficulty finding an ungated version, sorry to say) discusses relatively recent evidence for 'full cost' than marginal-cost pricing, formalizes a problem where it arises, and then explores what a world where firms set prices that way would mean for macroeconomic volatility. Economics as a field doesn't fall apart if we find evidence against an undergraduate textbook model, or an influential model in a leading journal; we just have to deal with it.
posted by dismas at 12:48 PM on November 2, 2021
my own view as an applied macroeconomist is that I'm happy to entertain alternative assumptions about how people and firms behave and take them to the data. for example, this paper (I've had difficulty finding an ungated version, sorry to say) discusses relatively recent evidence for 'full cost' than marginal-cost pricing, formalizes a problem where it arises, and then explores what a world where firms set prices that way would mean for macroeconomic volatility. Economics as a field doesn't fall apart if we find evidence against an undergraduate textbook model, or an influential model in a leading journal; we just have to deal with it.
posted by dismas at 12:48 PM on November 2, 2021
« Older Blurring the lines between play-acting, ritual and... | Possibly Real Copy Of ‘Fairies’ by Andy Warhol Newer »
This thread has been archived and is closed to new comments