Well, for the NBA at least, the minimum salary you can be paid does scale with your years in the league. There is also a a salary cap (per-team) and a maximum salary for any one player based on your previous salary and years of experience. One of the interesting things to happen because of this is that the top-tier players are underpaid by their teams (relative to their competition), and end up making more money from their shoe and advertising deals than their team salary. It also means that veterans who are willing to take minimum salary may find themselves priced out of the league since their minimum is higher than less experienced players.
>It also means that veterans who are willing to take minimum salary may find themselves priced out of the league since their minimum is higher than less experienced players.
Actually, the NBA CBA has that loophole covered. If a veteran player accepts a minimum salary contract, the team only has to pay the minimum salary for a player with 3 years of experience. The league pays the difference, and the difference does not count against the salary cap.
So, the minimum salary scale for veterans is not a disincentive against hiring a veteran who is willing to play for the minimum.
Yes, the longer a player sits on the bench / stares at code the better they are.
If it takes me 2 hours to do a project and someone else 4, by doing that project they have 2 more hours of experience than me. This is especially important when people do the exact same task every day instead of developing a wide range of skills.
It's really essential to find that great talent that has 15 years of experience in ASP.NET/J2EE.
It tends to make a wider range of teams competitive which is good for fans and the sport as a whole. Arguably this might be better for players as they get more money from endorsements when the sport is more popular.
As to developers, it's really just a question of what the baseline is. If company X pays mid level developers 200k / year then they will have little issue attracting or retaining talent.
Seems like a great way to ensure you have mediocre talent.
Maybe I've just been on my own too long, but I at least want the allusion that I'm being paid a fair market rate, instead of that which is 'fair' to my co-workers or company.
Those of us who are passionate about what we do, and work very hard, should run from something like this. I'm very open to conversation as to why that's not a good idea.
Buffer is an example of what seems like a cool product and company...but their salary system, IMO, favors Buffer. Not the people building the product. I'd never work for them.
I work for a public university so all salaries are public. There's no fixed salary levels, but everyonein the same position ends up very close. It completely eliminates office politics, but I don't see the downsides. If you want more money you work hard for promotion or a move to another unit. And you know exactly what those positions are worth and what it takes to get there.
Does your employer promote based on the results of an objective test? For example, many police departments have tests officers must pass to attain a particular supervisory rank. If that's how things work, then I agree that you know "what it takes to get there". In most organizations, public sector or private, there are a limited number of openings at a particular level, dependent upon budget, need, politics, and whatever other means of decision-making are in use. And those positions are filled based on subjective criteria at the discretion of the supervisor(s) responsible for them. So while you do know what other positions pay, the hard part -- assigning each human a role and compensation -- is every bit as opaque and subject to manipulation and error as anywhere else.
The causality may run the other way. Lock-step salaries work best when the key driver of the firm's success is not the knowledge and skills that individual employees bring to the firm, but rather pre-existing market advantages that the firm itself has that let its employees be more productive and make more money. Otherwise, the firm has a strong incentive to acquire those individual knowledge and skills to make itself more profitable.
The most profitable and selective law firms are such because they've built up very strong brand names and client relationships over time. They make an employee's career; an employee doesn't make the firm's success. As a result, the employee doesn't have the negotiating leverage to demand a higher salary, and the firm has a strong incentive to ensure harmony and control costs among its workers.
I'm not disagreeing, though, I feel the law profession's more of an exception, not a counter-example.
Law uniquely aligns with lock-step (like consulting) because... 1. the workers are homogenous (i.e. 3+ years of law edu)
2. the tasks are homogenous (similar to consulting)
3. the end-game is identical (i.e. partner-track/shareholder)
Of course they are, if you can convince high productivity people to work lockstep, then you make profit.
In law people take lower salary for partnership just as in startups devs buy into equity. Since equity / partnership for a few is cheaper than labour the company makes money hand over fist.
Up or out is a supremely effective way to drive people to meet organizational goals. On the down side, you have to be super careful about what you measure. Enron made extensive use of up or out. It worked really really well, right up until it didn't.
If that's really going to happen, executives are going to need to be at least as precise than an assembly programer when setting metrics.
Bad things happen when the "competition knob" is turned to 0 (everyone free-rides) and 11 (everyone cheats).
> you have to be super careful about what you measure.
Do you really think you can design metrics so perfect that they won't be gamed? Even if it were possible (I'm pretty sure it isn't), could you do it without undermining cooperative team dynamics that create organizational value? I'm not saying all metrics are bad, but I am saying that turning up the heat with up-or-out is the equivalent of turning the "competition knob" to 11 and that it has a large number of obvious failure modes which make me puzzle over your declaration of its efficacy (unless you meant it in the tautological sense, but that would be silly?).
I'm further puzzled that you brought Enron up as an example of incentives done correctly. Most people think of them as a case-study in runaway incentives and how turning the competition knob to 11 inevitably causes, well, exactly what happened to Enron. When the choices are "cheat or get fired," your company will inevitably fill itself with cheaters. Unless your metrics are perfect, of course, but are you willing to bet your company on it?
> Do you really think you can design metrics so perfect that they won't be gamed?
Nope. I think it's super super hard. I think a lot of executives will say, oh we'll just add "Up or out" to dial up the competition knob, and then they'll self destruct.
Executives make great decisions all the time. Unfortunately most of those decisions involve "close enough" thinking. Up or out introduces razor sharp edges. Everything appears better than fine, hell it appears fantastic. Until it isn't. and when it isn't it's terminal.
I used assembly specifically to highlight the care that must be taken. I wanted to evoke images of bricking devices. Very few executives have the mindset that if i'm not certain i'm right, then i'm wrong. Often their thinking is squishy at the edges, and that's just the cost of doing business.
I do think if it's possible to do "Up or out" effectively, that corporation would be formidable. If gaming the metric advanced the corporation's goals, that would be amazing. But, super hard. Folks selected for being the smartest guy in the room are most likely to fool themselves into thinking they can handle it.
I think where we fundamentally disagree is that you believe it's generally possible to design an incentive system which is well-behaved in the asymptote of extreme competition whereas I do not, except under select conditions where the sum total of value provided by an individual is trivially quantifiable and separable from the contribution of his peers (e.g. assembly line work).
"Care must be taken" simply doesn't cut it; the comparison to assembly programming is ridiculous. For most devices it's hard to screw up badly enough to brick them and even if you do you are out, what, a few hundred dollars? Also, the feedback is unambiguous and instantaneous. Screw up a big organization and you are out millions or billions with only vague, ambiguous feeback that arrives years after you institute the policy. Even a comparison to rocket science doesn't come close to capturing the difficulty involved because rockets are designed to be separable into subsystems which obey well-characterized physical laws whereas most creative processes can't be broken down so rigorously.
The hypothetical perfect incentive system is like the hypothetical analog computer which gives you infinite accuracy and precision for free: it only appears to be an exciting possibility if you neglect all of the factors that limit its efficacy in practice.
I think it works really well in adversarial contexts. I'd guess the best is probably law. Win cases, bill hours. Lack of ethics are exploited ruthlessly by the opposition - that's grounds for appeal.
Somewhat less effectively, sales. Car salesmen have horrible reputations, but lots of technical sales are smart, knowledgeable and helpful. I think the long term relationship aspect helps.
Even less effective, but still clearly in use is finance. Insider trading and collusion are tough to prove, and there's not much incentive to keep the system in check. But finance clearly takes advantage of up or out.
Bricking a device is a fine analogy. Sometimes the slippery slope argument is not a fallacy. I think this is one of those times. You seem to agree, we're just not clear on stakes. I think the slippery slope exists in a 2 person startup all the way up to microsoft or google.
Perfect anything is a pipe dream. "It has been said that democracy is the worst form of government, except all those others that have been tried". If somebody wants to try a social experiment with their investors money, best of luck to them. It's unlikely i'd invest or be interested in working there though.
Whereas Stack Ranking uses greed to estimulate achievement on projects, Lockstep uses greed to solely achieve a job title.
Therefore, Lockstep salaries are more vulnerable to office politics and employees pretending to achieve, in order to 'look good' and get a better salary.
> the ecommerce startup launched by Marc Lore, which uses a form of the lockstep model. Lore described his company’s system in an interview. “Everyone at the same level makes the same comp. Everyone knows that no one is getting paid more at a similar level,”
revealing this could easily lead to having staff hired away once the levels are figured out.
How about software developers form a programmers association and learn to negotiate and hire agents and lawyers and all that to make sure they're fairly compensated?
If you're going to have lockstep salaries then you also better have solid onboarding and training programs and an actual promotion process other than "if you aren't bored and stay a while you'll get promoted".
Because it's really difficult to get a whole bunch of people together who collectively pretty much have the best deal of anybody in the current economy and try to get them to collectively negotiate an even better deal.
Maybe y'all in Silicon Valley can be under the delusion that programmers somehow have a terrible, terrible job, but I live in Michigan, and that delusion is not available to me. I try not to talk about the details of my job too much with our friends... it's... discourteous.
That is not to say that it's all hunky-dory or that if one could get over the step of organizing that a better deal might not be available; see the recent lawsuits over salary collusion, for instance, which is strong evidence there is at least some "better deal" theoretically achievable. I mean exactly what I said in my first paragraph, not as an attempt to slide in another point quietly... it's really difficult to get anything organized under these circumstances.
I've been appalled to find out how much more I make than some of my colleagues, even though we have the same job title. Now, I feel like I produce more than they do (and higher quality, to boot), so maybe my higher pay is justified, but really, I don't think it's that justified.
I would much rather have levels with merit bonuses than have to rely on being a good negotiator. I could be the best programmer in the world sitting next to the worst programmer in the world, but if she negotiates better than I do, she makes more. Or vice versa. And I don't think that's particularly fair.
I did complain to my immediate boss, and then when I became a manager, I went to HR fought to make the pay more equitable. Not a whole lot happened, as far as I can tell.
To be fair to them, it's not quite so easy though. You can bring underpaid people up to "fair", but you can't really bring overpaid people down, so you still wind up with some unfair (imo) compensation going around.
Incidentally, the last time I got a raise was when I became a manager and became more aware of the exact compensation numbers for the rest of my team. My plan for the next time I get a raise was to ask them to give it to some of my underpaid team members instead, but I'm leaving this company soon, so I won't be able to see if that strategy would work or not.
It's very fair, if your company similarly does a poor job of highlighting benefits of purchasing your product it will similarly be killed in the market.
Negotiation is primarily about your value prop. Workers who understand this ensure they are contributing the most to the bottom line of the company and have the data to back it up.
Changing the color of buttons,etc generally adds far more to the bottom line than making the code run a little bit faster / use less hardware. (obviously there are exceptions, but you can only cut costs to zero where as revenue can grow infinitely)
eg. I bet the invention of the Like button vastly outweighs the invention of HVVM for FB in terms of adding to the bottom line.
Sounds like a great way to collect a company of low performers. Since all the people who can do great things will go elsewhere, where they will actually be paid for being above average in their skill & efficiency.
So instead of negotiating pay, candidates will negotiate titles (except for women, presumably, who will continue putting themselves at a disadvantage in a new way). And instead of salary envy you get title envy. And instead of uncertainty and negotiation over raises, you get uncertainty and negotiation over promotions. And by the way, if you have ever worked for a large corporation with salary bands, you know that this is exactly what happens there.
I think that is a good point, and I'm sure approaches like this don't solve the issue, but there is a difference in transparency.
Pay isn't public, but titles are. So if you are in the company you can see the landscape and know if you should be in the group of people with the better title. This would mean that although women might not be as inclined to negotiate, it would become clearer to people where themselves and others should be with their titles, so discrepancies between performance and title and pay would be more obvious.
And will get a half dozen more titles to keep salaries diversified enough. This will not motivate workers at the same title who are twice as productive of course
It's hard to deny titles based on work, however. A tech lead acts as a team leader and has final say on development decisions. A project manager manages a project, stakeholders associated with the project and tracks external dependencies of the project. A senior developer writes and checks in code without too much oversight, and other team members come to him/her for advice.
52 comments
[ 2.5 ms ] story [ 69.4 ms ] threadActually, the NBA CBA has that loophole covered. If a veteran player accepts a minimum salary contract, the team only has to pay the minimum salary for a player with 3 years of experience. The league pays the difference, and the difference does not count against the salary cap.
So, the minimum salary scale for veterans is not a disincentive against hiring a veteran who is willing to play for the minimum.
If it takes me 2 hours to do a project and someone else 4, by doing that project they have 2 more hours of experience than me. This is especially important when people do the exact same task every day instead of developing a wide range of skills.
It's really essential to find that great talent that has 15 years of experience in ASP.NET/J2EE.
As to developers, it's really just a question of what the baseline is. If company X pays mid level developers 200k / year then they will have little issue attracting or retaining talent.
Maybe I've just been on my own too long, but I at least want the allusion that I'm being paid a fair market rate, instead of that which is 'fair' to my co-workers or company.
Those of us who are passionate about what we do, and work very hard, should run from something like this. I'm very open to conversation as to why that's not a good idea.
Buffer is an example of what seems like a cool product and company...but their salary system, IMO, favors Buffer. Not the people building the product. I'd never work for them.
I have met a few academics that would definitely bet their paycheck on academia having much more politics than companies.
Yeah, I agree that I don't think equal pay eliminated politics.
Can't tell if you are being sarcastic.
The most profitable and selective law firms are such because they've built up very strong brand names and client relationships over time. They make an employee's career; an employee doesn't make the firm's success. As a result, the employee doesn't have the negotiating leverage to demand a higher salary, and the firm has a strong incentive to ensure harmony and control costs among its workers.
Law uniquely aligns with lock-step (like consulting) because... 1. the workers are homogenous (i.e. 3+ years of law edu) 2. the tasks are homogenous (similar to consulting) 3. the end-game is identical (i.e. partner-track/shareholder)
In law people take lower salary for partnership just as in startups devs buy into equity. Since equity / partnership for a few is cheaper than labour the company makes money hand over fist.
I don't. I want to be paid way above market rates. "Fair market rates" in some industries is breathtakingly low.
If that's really going to happen, executives are going to need to be at least as precise than an assembly programer when setting metrics.
> you have to be super careful about what you measure.
Do you really think you can design metrics so perfect that they won't be gamed? Even if it were possible (I'm pretty sure it isn't), could you do it without undermining cooperative team dynamics that create organizational value? I'm not saying all metrics are bad, but I am saying that turning up the heat with up-or-out is the equivalent of turning the "competition knob" to 11 and that it has a large number of obvious failure modes which make me puzzle over your declaration of its efficacy (unless you meant it in the tautological sense, but that would be silly?).
I'm further puzzled that you brought Enron up as an example of incentives done correctly. Most people think of them as a case-study in runaway incentives and how turning the competition knob to 11 inevitably causes, well, exactly what happened to Enron. When the choices are "cheat or get fired," your company will inevitably fill itself with cheaters. Unless your metrics are perfect, of course, but are you willing to bet your company on it?
Nope. I think it's super super hard. I think a lot of executives will say, oh we'll just add "Up or out" to dial up the competition knob, and then they'll self destruct.
Executives make great decisions all the time. Unfortunately most of those decisions involve "close enough" thinking. Up or out introduces razor sharp edges. Everything appears better than fine, hell it appears fantastic. Until it isn't. and when it isn't it's terminal.
I used assembly specifically to highlight the care that must be taken. I wanted to evoke images of bricking devices. Very few executives have the mindset that if i'm not certain i'm right, then i'm wrong. Often their thinking is squishy at the edges, and that's just the cost of doing business.
I do think if it's possible to do "Up or out" effectively, that corporation would be formidable. If gaming the metric advanced the corporation's goals, that would be amazing. But, super hard. Folks selected for being the smartest guy in the room are most likely to fool themselves into thinking they can handle it.
"Care must be taken" simply doesn't cut it; the comparison to assembly programming is ridiculous. For most devices it's hard to screw up badly enough to brick them and even if you do you are out, what, a few hundred dollars? Also, the feedback is unambiguous and instantaneous. Screw up a big organization and you are out millions or billions with only vague, ambiguous feeback that arrives years after you institute the policy. Even a comparison to rocket science doesn't come close to capturing the difficulty involved because rockets are designed to be separable into subsystems which obey well-characterized physical laws whereas most creative processes can't be broken down so rigorously.
The hypothetical perfect incentive system is like the hypothetical analog computer which gives you infinite accuracy and precision for free: it only appears to be an exciting possibility if you neglect all of the factors that limit its efficacy in practice.
Somewhat less effectively, sales. Car salesmen have horrible reputations, but lots of technical sales are smart, knowledgeable and helpful. I think the long term relationship aspect helps.
Even less effective, but still clearly in use is finance. Insider trading and collusion are tough to prove, and there's not much incentive to keep the system in check. But finance clearly takes advantage of up or out.
Bricking a device is a fine analogy. Sometimes the slippery slope argument is not a fallacy. I think this is one of those times. You seem to agree, we're just not clear on stakes. I think the slippery slope exists in a 2 person startup all the way up to microsoft or google.
Perfect anything is a pipe dream. "It has been said that democracy is the worst form of government, except all those others that have been tried". If somebody wants to try a social experiment with their investors money, best of luck to them. It's unlikely i'd invest or be interested in working there though.
Instead of negotiating for salary, just negotiate for title, if you have a title that no one else has all the better.
Therefore, Lockstep salaries are more vulnerable to office politics and employees pretending to achieve, in order to 'look good' and get a better salary.
revealing this could easily lead to having staff hired away once the levels are figured out.
If you're going to have lockstep salaries then you also better have solid onboarding and training programs and an actual promotion process other than "if you aren't bored and stay a while you'll get promoted".
for example:
* screen actors guild * nhlpa - national hockey league players association
Maybe y'all in Silicon Valley can be under the delusion that programmers somehow have a terrible, terrible job, but I live in Michigan, and that delusion is not available to me. I try not to talk about the details of my job too much with our friends... it's... discourteous.
That is not to say that it's all hunky-dory or that if one could get over the step of organizing that a better deal might not be available; see the recent lawsuits over salary collusion, for instance, which is strong evidence there is at least some "better deal" theoretically achievable. I mean exactly what I said in my first paragraph, not as an attempt to slide in another point quietly... it's really difficult to get anything organized under these circumstances.
I would much rather have levels with merit bonuses than have to rely on being a good negotiator. I could be the best programmer in the world sitting next to the worst programmer in the world, but if she negotiates better than I do, she makes more. Or vice versa. And I don't think that's particularly fair.
Have you done anything about this?
To be fair to them, it's not quite so easy though. You can bring underpaid people up to "fair", but you can't really bring overpaid people down, so you still wind up with some unfair (imo) compensation going around.
Incidentally, the last time I got a raise was when I became a manager and became more aware of the exact compensation numbers for the rest of my team. My plan for the next time I get a raise was to ask them to give it to some of my underpaid team members instead, but I'm leaving this company soon, so I won't be able to see if that strategy would work or not.
Negotiation is primarily about your value prop. Workers who understand this ensure they are contributing the most to the bottom line of the company and have the data to back it up.
Changing the color of buttons,etc generally adds far more to the bottom line than making the code run a little bit faster / use less hardware. (obviously there are exceptions, but you can only cut costs to zero where as revenue can grow infinitely)
eg. I bet the invention of the Like button vastly outweighs the invention of HVVM for FB in terms of adding to the bottom line.
Boo-fucking-hoo.
Grow the fuck up. If one cannot learn how to effectively communicate, then one will face the consequences.
I fail to see any material difference.
Pay isn't public, but titles are. So if you are in the company you can see the landscape and know if you should be in the group of people with the better title. This would mean that although women might not be as inclined to negotiate, it would become clearer to people where themselves and others should be with their titles, so discrepancies between performance and title and pay would be more obvious.
I never knew that this was called "lockstep salary", but it's essentially how all of the unions work.
It hurts many employees because it takes away all your negotiating power.