it's not just an issue of fairness / employee morale. Difficulty in productivity measurement probably plays an even bigger role - proof is that in jobs where productivity is easily measured (e.g. sports, show-biz, investment management, arguably even CEO roles - if one accepts stock price as a productivity measure - there are much wider gaps in comp). In jobs where productivity cannot be accurately/directly measured, having big comp range would just cause everyone to spend all their time jockeying for position / trying to game the system / arguing about whether their own productivity measurement is fair. This in itself would cause a ton of wasted time and productivity loss, so pretty obvious why companies don't want to go there.
I agree. Morale is just one factor. Another is wasted time in trying to measure productivity. Another is eliminating the inefficiency of bartering over salary with every hire. Also, there is the minimally acceptable quality factor. Below that is disaster for the firm. If we need to come up with a set salary range (for any of the above reasons), then it is essential to weed out rotten fruit. The easiest way to do this is to select for fruit that is most ripe, but within the range of salary permitted to the hiring manager.
The thing I find most intriguing about this study is that it also suggests that inequality of income is compressed in markets. The system is more unfair, not to the low wage earners (who are helped by compression), but to the higher earners more harmed by compression. I demand more inequality! It would be more fair!
Having been a software developer and managing other developers for my career, I can vouch for the empirical evidence. Every place I've worked from startups with 3 employees to companies with many thousand employees have paid their developers a compressed salary in the sense that developers who could do 10x what others could do (both from a speed and quality perspective) would get paid maybe 2x what the lower capable developers made.
Thanks for re-upping this. Certainly a profound question.
I think part of the answer is agency. Firms do not absolutely prefer more able workers, they typically hire in bands (you can get hired at entry-level, as a senior engineer, as the CEO, etc). Within those bands, there is a pay range. The hiring committee are generally given a fixed budget (for agency reasons). They have no incentive to come in under budget, and will be blamed for hiring a dud, so they typically hire the best person they can find in that budget. Senior leadership, who do have overall financial accountability, oversee overall recruitment targets, such as how many entry-level employees are we hiring, how many experienced professionals, as well as the pay bands, which implicitly sets the employee quality level. They definitely aren't (warm words aside) targeting the very best employees, they are sensitive to the trade-offs.
I think this is a big part of it, although one has to consider the interplay of job role and individual ability when it comes to productivity. If you are hiring someone to sweep up the shop floor you might well turn away a lot of college grads, as even if they are more able it is kind of pointless to hire them as their productivity will be super low. For more productive positions you are probably budgeting for the average person's productivity in the role, and if you get someone better, yay, but by the time you know they are better they have already accepted the salary, so you either keep them there with bonuses or give them a promotion.
Interesting: I had always assumed that firms preferred more able workers because less able workers would incur higher health expenses, which can be highly variable with large outliers. It would be interesting if the data allow one to separate between jobs which included health benefits and those that didn't.
EDIT: Walmart was reported to engineer as many as jobs as possible to require ability to lift heavy weight, presumably to exclude less able workers.
A bit off topic but the font and layout of academic papers really look terrible. If I type things in double spaced “academic font” (I am not sure of the actual font) does that automatically make me look smart. I know this paper is from the early 90s I am hoping things have evolved since then to at least make papers look more attractive to read.
My cynical take on the issue is that they just try to add more math. I understand that math is important in economics, but somewhere along the line it has shifted from math in support of theory to "mathterbation" for appearances’ sake. Along the way, the best theorists have been replaced by the best mathematicians, and the two domains don’t necessarily overlap all that well.
Still, this is probably better than the future of economics, where the mathematicians are replaced by properly diverse social justice warriors dedicated to endlessly proving how inequity harms their politically favored groups.
Robin Hanson recently had a post on Mutual Admiration Society in terms of journal articles and one of the things in that was making your maths look hard. I’m a bit on the left and take your point that if we end up going down the path of finding injustice in the world and then using that to come up with some vacuous theory that doesn’t stack up it will be bad, although there are some lefties who try to do good economics. I’m a maths, engineering type person so economics is always a bit puzzling to me, in my area of the world either the building stands up or falls down and either 1+1=2 or it doesn’t, in economics it seems like you can kind of make any argument you want and it is hard to prove or falsify it.
This paper in conjunction with Rob Axtel's paper on the cycles of the labor market I think explain the life cycle of most companies. Due to the wage compression, eventually big companies get dominated by lower skill workers, as higher skill workers leave for greener pastures and lower skill ones stay put. The firm eventually can't compete and collapses. Axtel's model assumed that firm revenue was distributed evenly amongst workers regardless of productivity, and Romer's shows why that is a good assumption even though it seems counter intuitive.
Yes, and this is part of an even larger and longer term process for bureaucrats, rent seekers, dinosaurs and cartels (Ala Mancur Olson) to dominate economies, thus leading to stagnation, sclerosis and eventual collapse.
Indeed a very good reason to not make any effort to prop up or bail out failing organizations or companies. They are failing because they are no longer functional, and nothing worth saving. Prolonging it only increases the losses.
Think about this from the point of view of a hiring manager: would you pay a single superstar who could replace your entire team? First of all, his salary would greatly exceed yours. Second, why does a one-man team even need a manager?
There are counter-examples. When "the company" (the ppl with the task to hire) do say: This guy is over-qualified, he would soon get unhappy with our pay-level/work-demands and be looking for a better job anyway. Plus bad vibes in the dept (the new guy studied in Oxbridge and holds two PhDs, most of us just did a local college!). Quote: “Overly ambitious people are a mood killer and only cause unrest” read the whole thing (short, fun, in German) https://www.zeit.de/2010/01/Martenstein-01 - Also, a Porsche(seller) does not pay for ads aimed at low-income groups. A company hoping to attract an "at least level 40 worker" will not spend much energy to appeal to "level 90" workers - and will takes its picks from the level 25-45 applications they get.
An employee’s productivity in another company is not exactly equal to their productivity at your company. The market thus has a different estimate of your employer’s value than you do. Plus there are transaction costs to fire and hire even if a new employee is identical. I think those are sufficient to explain why the “market price” for a good employee is usually lower than their value to you.
Indeed a very interesting question. It is related to: Why do workers want to go to a firm with other talented workers or with talented managers, if pay is no higher? If they go to that firm, it seems they would have a worse chance of being promoted, and tougher competition.
Do we need to go to fairness, in the Why Hire Good Workers? question. A simpler answer is that the high-ability workers who accept such a job have worse opportunities for some reason-- maybe higher search costs, maybe needing a job quick (similar), maybe liking the job tasks better-- and so the employer can get away with paying them less than their ability.
if wage differences hurts morale of lower-paid (less able) workers, then wage compression certainly should hurt the morale of the higher-paid (more able). I’m guessing the former is preferred cause of relative population size.
I'm just not sure that's true in the same way. Most people have some level of fairness norm, and being on top always feels good. There may be some people who think to themselves "sure i get paid more than all my coworkers, but I should be paid MUCH more than all my coworkers, therefore I am disatissfied" but I would guess they are unusual.
it's not just an issue of fairness / employee morale. Difficulty in productivity measurement probably plays an even bigger role - proof is that in jobs where productivity is easily measured (e.g. sports, show-biz, investment management, arguably even CEO roles - if one accepts stock price as a productivity measure - there are much wider gaps in comp). In jobs where productivity cannot be accurately/directly measured, having big comp range would just cause everyone to spend all their time jockeying for position / trying to game the system / arguing about whether their own productivity measurement is fair. This in itself would cause a ton of wasted time and productivity loss, so pretty obvious why companies don't want to go there.
I agree. Morale is just one factor. Another is wasted time in trying to measure productivity. Another is eliminating the inefficiency of bartering over salary with every hire. Also, there is the minimally acceptable quality factor. Below that is disaster for the firm. If we need to come up with a set salary range (for any of the above reasons), then it is essential to weed out rotten fruit. The easiest way to do this is to select for fruit that is most ripe, but within the range of salary permitted to the hiring manager.
The thing I find most intriguing about this study is that it also suggests that inequality of income is compressed in markets. The system is more unfair, not to the low wage earners (who are helped by compression), but to the higher earners more harmed by compression. I demand more inequality! It would be more fair!
Having been a software developer and managing other developers for my career, I can vouch for the empirical evidence. Every place I've worked from startups with 3 employees to companies with many thousand employees have paid their developers a compressed salary in the sense that developers who could do 10x what others could do (both from a speed and quality perspective) would get paid maybe 2x what the lower capable developers made.
Thanks for re-upping this. Certainly a profound question.
I think part of the answer is agency. Firms do not absolutely prefer more able workers, they typically hire in bands (you can get hired at entry-level, as a senior engineer, as the CEO, etc). Within those bands, there is a pay range. The hiring committee are generally given a fixed budget (for agency reasons). They have no incentive to come in under budget, and will be blamed for hiring a dud, so they typically hire the best person they can find in that budget. Senior leadership, who do have overall financial accountability, oversee overall recruitment targets, such as how many entry-level employees are we hiring, how many experienced professionals, as well as the pay bands, which implicitly sets the employee quality level. They definitely aren't (warm words aside) targeting the very best employees, they are sensitive to the trade-offs.
I think this is a big part of it, although one has to consider the interplay of job role and individual ability when it comes to productivity. If you are hiring someone to sweep up the shop floor you might well turn away a lot of college grads, as even if they are more able it is kind of pointless to hire them as their productivity will be super low. For more productive positions you are probably budgeting for the average person's productivity in the role, and if you get someone better, yay, but by the time you know they are better they have already accepted the salary, so you either keep them there with bonuses or give them a promotion.
Interesting: I had always assumed that firms preferred more able workers because less able workers would incur higher health expenses, which can be highly variable with large outliers. It would be interesting if the data allow one to separate between jobs which included health benefits and those that didn't.
EDIT: Walmart was reported to engineer as many as jobs as possible to require ability to lift heavy weight, presumably to exclude less able workers.
A bit off topic but the font and layout of academic papers really look terrible. If I type things in double spaced “academic font” (I am not sure of the actual font) does that automatically make me look smart. I know this paper is from the early 90s I am hoping things have evolved since then to at least make papers look more attractive to read.
My cynical take on the issue is that they just try to add more math. I understand that math is important in economics, but somewhere along the line it has shifted from math in support of theory to "mathterbation" for appearances’ sake. Along the way, the best theorists have been replaced by the best mathematicians, and the two domains don’t necessarily overlap all that well.
Still, this is probably better than the future of economics, where the mathematicians are replaced by properly diverse social justice warriors dedicated to endlessly proving how inequity harms their politically favored groups.
Robin Hanson recently had a post on Mutual Admiration Society in terms of journal articles and one of the things in that was making your maths look hard. I’m a bit on the left and take your point that if we end up going down the path of finding injustice in the world and then using that to come up with some vacuous theory that doesn’t stack up it will be bad, although there are some lefties who try to do good economics. I’m a maths, engineering type person so economics is always a bit puzzling to me, in my area of the world either the building stands up or falls down and either 1+1=2 or it doesn’t, in economics it seems like you can kind of make any argument you want and it is hard to prove or falsify it.
This paper in conjunction with Rob Axtel's paper on the cycles of the labor market I think explain the life cycle of most companies. Due to the wage compression, eventually big companies get dominated by lower skill workers, as higher skill workers leave for greener pastures and lower skill ones stay put. The firm eventually can't compete and collapses. Axtel's model assumed that firm revenue was distributed evenly amongst workers regardless of productivity, and Romer's shows why that is a good assumption even though it seems counter intuitive.
Yes, and this is part of an even larger and longer term process for bureaucrats, rent seekers, dinosaurs and cartels (Ala Mancur Olson) to dominate economies, thus leading to stagnation, sclerosis and eventual collapse.
Indeed a very good reason to not make any effort to prop up or bail out failing organizations or companies. They are failing because they are no longer functional, and nothing worth saving. Prolonging it only increases the losses.
Think about this from the point of view of a hiring manager: would you pay a single superstar who could replace your entire team? First of all, his salary would greatly exceed yours. Second, why does a one-man team even need a manager?
There are counter-examples. When "the company" (the ppl with the task to hire) do say: This guy is over-qualified, he would soon get unhappy with our pay-level/work-demands and be looking for a better job anyway. Plus bad vibes in the dept (the new guy studied in Oxbridge and holds two PhDs, most of us just did a local college!). Quote: “Overly ambitious people are a mood killer and only cause unrest” read the whole thing (short, fun, in German) https://www.zeit.de/2010/01/Martenstein-01 - Also, a Porsche(seller) does not pay for ads aimed at low-income groups. A company hoping to attract an "at least level 40 worker" will not spend much energy to appeal to "level 90" workers - and will takes its picks from the level 25-45 applications they get.
An employee’s productivity in another company is not exactly equal to their productivity at your company. The market thus has a different estimate of your employer’s value than you do. Plus there are transaction costs to fire and hire even if a new employee is identical. I think those are sufficient to explain why the “market price” for a good employee is usually lower than their value to you.
Take financial advice a quality adviser is one who can bring clients and will probably get paid more than one who can't bring a book of clients.
Indeed a very interesting question. It is related to: Why do workers want to go to a firm with other talented workers or with talented managers, if pay is no higher? If they go to that firm, it seems they would have a worse chance of being promoted, and tougher competition.
Do we need to go to fairness, in the Why Hire Good Workers? question. A simpler answer is that the high-ability workers who accept such a job have worse opportunities for some reason-- maybe higher search costs, maybe needing a job quick (similar), maybe liking the job tasks better-- and so the employer can get away with paying them less than their ability.
if wage differences hurts morale of lower-paid (less able) workers, then wage compression certainly should hurt the morale of the higher-paid (more able). I’m guessing the former is preferred cause of relative population size.
I'm just not sure that's true in the same way. Most people have some level of fairness norm, and being on top always feels good. There may be some people who think to themselves "sure i get paid more than all my coworkers, but I should be paid MUCH more than all my coworkers, therefore I am disatissfied" but I would guess they are unusual.
How does that work forlow budget items, low quality products for a low cost.