I think a nice, short definition of Slack would be, “A small gradient of the cost function in the vicinity of the optimum.” In other words, the penalty for making sub-optimal decisions is not very high, because the region of acceptable behaviors around the best one is fairly large.
And yet here we are, where the penalties for sub-optimal decisions are indeed quite high. The obvious question is, “How did it get to be this way?” Well, typically non-smooth cost functions indicate more complexity, so in some sense the decision landscape has gotten more complicated than it used to be. Intuitively, this would seem to be because there is some sort of adversarial process going on where everyone’s strategies have adapted to become more effective.
The narrative that I usually tell myself to explain this goes something like, at some point along the way, firms realized (especially firms that accomplish cognitively demanding tasks) that the majority of their value was being created by a small number of extremely talented people. But these people were not just talented, but also mono-maniacal in their dedication to their jobs. They essentially filled most of their waking hours doing one thing, because they genuinely love it (and they love being hyper-focused). My assumption is that the ratio of the value these people created for their firms over the cost of hiring them was much greater than one. So firms realized that they could cut down on costs heavily by reducing overall hiring and focusing on locating just these people. Therefore they start to look for the signs of these people: Heavy activity outside of work dedicated to their field, a genuine love of it (maybe shown by writing articles, blogs or textbooks, giving talks, etc.), and willingness to work long hours. People adapt by trying to signal these things too even if it means cutting into the time they spend doing things they actually enjoy, but the job market is much tighter now because firms have reduced hiring. In short, firms are winning, people are losing.
But it still leaves the question of why the hyper-talented people are being paid less than they’re worth? That seems to be the crux of it all, some kind of market inefficiency that allows firms to win big by getting super-people at a bargain. My naive guess is that it has something to do with the employees themselves, by the nature of their personalities they are actually willing to work for less than they are worth, but I am pretty uncertain about that. I would appreciate any answers from someone with more economic expertise if they have some ideas.
My model of this problem is that firms cannot pay the hyper-talented what they are worth because the other employees would not stand for it. We have constructed norms whereby hyper-productive CEOs and other executives can get pay that is many times that of their employees, and still this causes massive resentment and even calls for government intervention. If programmers that were 100x as valuable as the average got even 10x the pay, the other employees would strike or riot. Equity does seem to be one way around this trap in some circumstances.
I have personal experience with this. I got [company I used to work for] to hire a very good programmer for a contract, he got the job done ahead of schedule and made us tons of money, and we couldn’t renew for another project, because other programmers found out what he was paid and threatened to quit. So he took a higher-paying job elsewhere.
Implicitly I’ve been assuming that we’re talking about tech workers here, but that may not be totally fair. I should probably extend this model to include most millenials that are highly career-focused, probably based in major coastal cities. In which case, you could probably extend this to sales/marketing and finance people as well.
I think if we do this, then it may be true that the “hyper-talented” people are still the issue here, but it could also be due to more general labor surplus issues as well. Then you get a lot of highly educated, highly motivated young people crowding into major hubs into a few high paying fields, under the belief that these are the primary, if not the only pathways to success.
Then the question becomes, is this actually true? Are they correctly identifying where their best chance at happiness lies, and it just happens that there truly are fewer opportunities for financial stability for most people these days?
If yes, then I think we could probably reduce the “slack” problem to general economic problems, but I’m not sure if that’s the direction you were intending to go with these posts. One thing I don’t know is whether or not Maya is essentially acting rationally under her incentive structures, and the incentive structures are the main culprit, or whether she is valuing certain things too highly, such as prestige and social approval.
My model says that Maya is acting the way humans react to things in similar situations, so whether or not her actions are “rational” the system is still the problem. Maya’s actions seem like they could be rational for certain utility functions, and you can’t play such games by halves (hence the lack of Slack) but my guess is that most Mayas are making a mistake playing the game at all. She should probably quit.
It does seem like financial stability is in practice much harder to achieve than it used to be, for most people. Especially when student loans and health care get involved. There’s less margin for error even with a “good” job and good jobs are harder to find in a pinch for most people. It is unclear to me how much of that is increased needs (letting consumerism, signaling and positional goods eat all your Slack, general high standards, and not putting up with boring or physically demanding work like people used to) versus how much is a real crisis (student loans, health insurance, child care and other forced expenses, lack of jobs with good pay and job security, being treated like dirt due to that, and taking of Slack in order to control people).
For programmers, especially top programmers, this isn’t an excuse. If I’m a superstar programmer who is worth 100x normal programmer, I can’t get 100x or even 10x, but I can get a new job any time I want and likely earn 1.5x with good perks. That really should be good enough to have Slack.
I like to look at energy usage per capita over time as an indicator of whether we can do more stuff than we could before. E.g. can people own cars. Technology changes this a bit, it makes cars more efficient/cheaper to make so more people can have cars. But looking at money etc makes things too easy to fudge.
So getting energy and people , the energy consumption per capita in 1980 it was 64 million BTU per person. In ~2014 it was 73 million BTU per person. So things are getting better energy wise for the globe. I won’t try to break down per country to see if things are getting better for the US, things get complicated via global trade (exporting energy consumption etc). You could expect that the majority of the increase in energy consumption has happened other places than the US (Asia had 300% energy consumption growth) during that time period though so that there might be a net negative in the US.
It is late, so I’m not going to continue. I think we have a rising inequality from various things, so even if there was an increased energy usage over all, people may feel worse off.
Or if you’re paid by results not the hour, as a contractor, you can earn the same in less time. Or even as an employee, you can just be paid to waste most of your time (though this is fairly unsatisfactory). Eg a friend of mine worked with an excellent programmer who would do nothing for months—literally spend most of the time in the pub or messing around with things that interested him—and occasionally spend a weekend programming furiously to produce what was presented to the (crappy) management as what the entire team had been working on for months.
I think that that sort of fixing is possible because a programmers value is based on the scale they work at. If you fix a bug that effects 1 million people per year and leads to 10 cents more per transaction, you have earned the company 100k per year. If your company only has 10k customers, then the same bug in the same time is only 1k per year. So you are worth different amounts to different businesses. As there are only so many very high scale businesses, they can collude.
Similar things might apply to lawyers, getting an X% better deal? I wonder at the social factors involved though.
I think a nice, short definition of Slack would be, “A small gradient of the cost function in the vicinity of the optimum.” In other words, the penalty for making sub-optimal decisions is not very high, because the region of acceptable behaviors around the best one is fairly large.
This doesn’t seem right to me. I think you could describe slack formally in terms of machine learning, but I don’t think this is a correct description of it. I was going to propose a better one, but I don’t know what I think it is yet.
I think a nice, short definition of Slack would be, “A small gradient of the cost function in the vicinity of the optimum.” In other words, the penalty for making sub-optimal decisions is not very high, because the region of acceptable behaviors around the best one is fairly large.
And yet here we are, where the penalties for sub-optimal decisions are indeed quite high. The obvious question is, “How did it get to be this way?” Well, typically non-smooth cost functions indicate more complexity, so in some sense the decision landscape has gotten more complicated than it used to be. Intuitively, this would seem to be because there is some sort of adversarial process going on where everyone’s strategies have adapted to become more effective.
The narrative that I usually tell myself to explain this goes something like, at some point along the way, firms realized (especially firms that accomplish cognitively demanding tasks) that the majority of their value was being created by a small number of extremely talented people. But these people were not just talented, but also mono-maniacal in their dedication to their jobs. They essentially filled most of their waking hours doing one thing, because they genuinely love it (and they love being hyper-focused). My assumption is that the ratio of the value these people created for their firms over the cost of hiring them was much greater than one. So firms realized that they could cut down on costs heavily by reducing overall hiring and focusing on locating just these people. Therefore they start to look for the signs of these people: Heavy activity outside of work dedicated to their field, a genuine love of it (maybe shown by writing articles, blogs or textbooks, giving talks, etc.), and willingness to work long hours. People adapt by trying to signal these things too even if it means cutting into the time they spend doing things they actually enjoy, but the job market is much tighter now because firms have reduced hiring. In short, firms are winning, people are losing.
But it still leaves the question of why the hyper-talented people are being paid less than they’re worth? That seems to be the crux of it all, some kind of market inefficiency that allows firms to win big by getting super-people at a bargain. My naive guess is that it has something to do with the employees themselves, by the nature of their personalities they are actually willing to work for less than they are worth, but I am pretty uncertain about that. I would appreciate any answers from someone with more economic expertise if they have some ideas.
My model of this problem is that firms cannot pay the hyper-talented what they are worth because the other employees would not stand for it. We have constructed norms whereby hyper-productive CEOs and other executives can get pay that is many times that of their employees, and still this causes massive resentment and even calls for government intervention. If programmers that were 100x as valuable as the average got even 10x the pay, the other employees would strike or riot. Equity does seem to be one way around this trap in some circumstances.
I have personal experience with this. I got [company I used to work for] to hire a very good programmer for a contract, he got the job done ahead of schedule and made us tons of money, and we couldn’t renew for another project, because other programmers found out what he was paid and threatened to quit. So he took a higher-paying job elsewhere.
Implicitly I’ve been assuming that we’re talking about tech workers here, but that may not be totally fair. I should probably extend this model to include most millenials that are highly career-focused, probably based in major coastal cities. In which case, you could probably extend this to sales/marketing and finance people as well.
I think if we do this, then it may be true that the “hyper-talented” people are still the issue here, but it could also be due to more general labor surplus issues as well. Then you get a lot of highly educated, highly motivated young people crowding into major hubs into a few high paying fields, under the belief that these are the primary, if not the only pathways to success.
Then the question becomes, is this actually true? Are they correctly identifying where their best chance at happiness lies, and it just happens that there truly are fewer opportunities for financial stability for most people these days?
If yes, then I think we could probably reduce the “slack” problem to general economic problems, but I’m not sure if that’s the direction you were intending to go with these posts. One thing I don’t know is whether or not Maya is essentially acting rationally under her incentive structures, and the incentive structures are the main culprit, or whether she is valuing certain things too highly, such as prestige and social approval.
My model says that Maya is acting the way humans react to things in similar situations, so whether or not her actions are “rational” the system is still the problem. Maya’s actions seem like they could be rational for certain utility functions, and you can’t play such games by halves (hence the lack of Slack) but my guess is that most Mayas are making a mistake playing the game at all. She should probably quit.
It does seem like financial stability is in practice much harder to achieve than it used to be, for most people. Especially when student loans and health care get involved. There’s less margin for error even with a “good” job and good jobs are harder to find in a pinch for most people. It is unclear to me how much of that is increased needs (letting consumerism, signaling and positional goods eat all your Slack, general high standards, and not putting up with boring or physically demanding work like people used to) versus how much is a real crisis (student loans, health insurance, child care and other forced expenses, lack of jobs with good pay and job security, being treated like dirt due to that, and taking of Slack in order to control people).
For programmers, especially top programmers, this isn’t an excuse. If I’m a superstar programmer who is worth 100x normal programmer, I can’t get 100x or even 10x, but I can get a new job any time I want and likely earn 1.5x with good perks. That really should be good enough to have Slack.
I like to look at energy usage per capita over time as an indicator of whether we can do more stuff than we could before. E.g. can people own cars. Technology changes this a bit, it makes cars more efficient/cheaper to make so more people can have cars. But looking at money etc makes things too easy to fudge.
So getting energy and people , the energy consumption per capita in 1980 it was 64 million BTU per person. In ~2014 it was 73 million BTU per person. So things are getting better energy wise for the globe. I won’t try to break down per country to see if things are getting better for the US, things get complicated via global trade (exporting energy consumption etc). You could expect that the majority of the increase in energy consumption has happened other places than the US (Asia had 300% energy consumption growth) during that time period though so that there might be a net negative in the US.
It is late, so I’m not going to continue. I think we have a rising inequality from various things, so even if there was an increased energy usage over all, people may feel worse off.
Or if you’re paid by results not the hour, as a contractor, you can earn the same in less time. Or even as an employee, you can just be paid to waste most of your time (though this is fairly unsatisfactory). Eg a friend of mine worked with an excellent programmer who would do nothing for months—literally spend most of the time in the pub or messing around with things that interested him—and occasionally spend a weekend programming furiously to produce what was presented to the (crappy) management as what the entire team had been working on for months.
There was some furore a while back about wage fixing between the big tech companies. Whether it has been fixed or not is another matter.
I think that that sort of fixing is possible because a programmers value is based on the scale they work at. If you fix a bug that effects 1 million people per year and leads to 10 cents more per transaction, you have earned the company 100k per year. If your company only has 10k customers, then the same bug in the same time is only 1k per year. So you are worth different amounts to different businesses. As there are only so many very high scale businesses, they can collude.
Similar things might apply to lawyers, getting an X% better deal? I wonder at the social factors involved though.
This doesn’t seem right to me. I think you could describe slack formally in terms of machine learning, but I don’t think this is a correct description of it. I was going to propose a better one, but I don’t know what I think it is yet.