This is not the first of Spolsky's essays that's left me simultaneously amazed at his brilliance and clarity and dismayed at something around the edges that just completely misses. Maybe it's just the title that's misleading. His emphasis on openness is great but he's really just moved the discussion from "how much money" to "how should I rate your contributions." Most salary discussions come down to this question and it's a concept that's very difficult to define precisely. The guy whose lunchtime conversations shape his colleagues' ideas may have made a profound contribution but if his own direct efforts don't match "Has consistently had major success..." then I'm guessing there's still a lot of room for negotiation - particularly if the employee realizes that making this case will give him a lot more money. Focusing on output is great, but that's what most salary negotiations end up being about.
I think the point is that Joel tackles the problem of assessing each employee's contribution straight on and unambiguously. I'm sure there is opportunity for negotiation if you believe you have made more contribution than your position on the ladder would suggest. Most other companies won't even attempt to explain how they determine salaries because they are not brave enough to tell their employees "you're good, but not great". Equally, most employees aren't willing to discuss it objectively - we all think we're above average, right? By attempting to put more structure around the process, Fog Creek are making it a lot easier for them and their employees to talk openly about their (and their colleagues') contributions.
"Because salary information is viewed as particularly sensitive, employers often go to great lengths to keep it under wraps."
In Norway, salary information is viewed as public information. Each year the government releases the income, assets and amount of payed tax of every citizen, so that the media can make it public. It is then made searchable online, for example here: http://skattelister.no/
In the U.S., companies that want to use H-1 visa programs must release specific salaries for specific titles.
Warning: Look at these databases at your own risk. You may be shocked at what others at your company are probably earning. Google H1-B salary database.
I'll make the same comment I made when this was posted last year:
Fog Creek has secret profit-sharing plans. All Joel has done is move a portion of employees' compensation from the "salary" column to the "profit-sharing" column.
If Joel were to begin posting the annual profit sharing of all employees, including Michael and Joel, then this article would make sense. But until then, it's just flair.
I work at Fog Creek, and this is just incorrect. Profit sharing is a simple formula known to everyone in the company, so you can easily figure out how much everyone made in profit sharing at the end of the year.
Honestly, I know it sounds too good to be true, but all compensation is completely open and honest.
Since the formula is based on number of days worked, I've always assumed that they just use the same formulas for themselves (and they get a lot more because they've worked at Fog Creek much longer than anyone else), but I guess I've never asked.
But I think the point is more that I know I'm not getting screwed over compared to my peers in the company -- it wouldn't really bother me to find out that they take extra for themselves since it is their company and all...
Profit in this context is whatever your accountant says it is. You can easily shuffle expenses backwards and forwards between quarters if you want to tweak the figures slightly (so, for example, your profit for this quarter is 99.9% of what it would need to be to pay bonuses). Cynical, moi?
I read Spolsky's essay again to see if he mentioned making public the employee ratings. For all his talk of openness in salaries, does he really publish salary information or just the salaries at each tier of the rating system? Are employees allowed to know their co-workers' scores in the tier system? As helwr mentioned below, a lot like Soviet Russia... or the US Government.
Joel said in a stackoverflow podcast that he publishes the formula for the pay scales and profit sharing amounts so if an employee really wanted to know someone else's salary they could figure it out based on their position and how long they have been with the company.
I like the idea of transparency with salaries as long as the company pays well. If an employee decides he/she wants a higher salary than is afforded by a system like that they can look for another job or start their own company.
You may be right about the profit sharing. I thought I remember him saying that it was divided according to how long you have been with the company, but maybe it's not apart of the same "formula".
I don't know what fog creek does. Usually some % of profits is reserved for distribution to employees. say, 10 dollars in profits. 10% to profit sharing, that dollar is divided among the employees, proportional to their salaries. The other 9 dollars go to other stuff, like growing the company, or distribution of profits to shareholders.
So, joel may make a lot of money, but i seriously doubt he's half, or even 10% of the profit sharing money. I also wouldn't count out the idea that some employees (other than Joel and Michael) have stock, so they get paid again when profits are distributed.
Say what you want about the article, but there's one really nice thing about his system: because there's a formula instead of negotiation, the good hackers who are often poor negotiators don't have to worry they have lower salary than the average or poor programmers who just happen to have better people skills or however you call it.
"I found a Seattle-area software-consulting firm called Construx that had published on its website the outline of a decent professional ladder system" It sounds like it is just another consulting firm. Construx founder and CEO is Steve McConnell, he wrote some great books about software development back in the day.
I find the experience idea their very arbitrary, so no experience from when your in school counted and 1 year no matter how many years in a "menial" position.
It kinda implies that while in school you couldn't have possibly picked up any experience that would define you above someone who has a year of industry experience.
I was going to read this article, but halfway through the first paragraph some stupid ad blocked my view. I'll definitely think twice before clicking through on anything pointing to inc.com again.
"if you're the kind of person who likes to use self-important HR jargon"
Wow, talk about self-important. People have a term for something you just took a paragraph to explain. Can you guess why? This comes across as a pointless dig to make yourself feel superior.
this whole Microsoft-style pay scale reminds me of government jobs, or worse - Soviet Union where my parents earned a standard doctor-engineer-truck driver-movie director-nurse salary of 100 rubles a month +/- the bureaucrat defined 'pay scale'. Come on, people, this is a free market!
"is this the UDA? / Or is this the IRA? / I thought it was the USA / Or just another country" (c) Sex Pistols
Capitalism is a series of centralized entities embedded in a larger sea of freedom with no (theoretically) overarching order imposed.
Fog Creek is free to compensate people in this manner. You are free to seek employment elsewhere if this is a problem.
It would not be a free market if either A: the government mandated this approach (or worse, mandated the pay scale) or B: an oligarchy formed where everybody determined compensation the same way. So far, neither have happened.
(Ironically, many people's proposed solution to various problems caused by their not liking one particular centralized entity's choice is to centralize the entire market into one entity. You didn't say anything about that, I just can't resist pointing that out. Unionizing programmers could have this effect, though.)
It might be worth noting that there's a decent amount of debate in free-market-capitalist circles over just how natural it is that these centralized entities emerge, versus how much they're encouraged/enabled by the government. A lot of the structure of the modern corporation is basically state-created: the limited-liability corporate status to begin with, the centralized form of management with a board of directors, etc.
If there was no built-in scaffolding of corporate status and corporate law, and it were just individuals operating in a free market forming whatever arrangements they want via private contract law, it's at least possible that large, centralized entities wouldn't emerge as easily or as frequently. From that perspective, the modern "sea of centralized entities" version of capitalism is itself centrally planned: the government decided that it'd be best if the economy were made up of a sea of centralized entities, and wrote the rules so that it happened.
Companies exist because they reduce the transaction costs of getting things done relative to doing everythig with independent contractors for example (http://en.wikipedia.org/wiki/Transaction_cost).
The libertarian argument is about how much government regulation influences the size of companies, that is how large companies would be absent regulation, the vast majority of which impacts smaller companies more per dollar of income than it does larger one.
Limited-liability is a natural, almost inevitable, requirement for joint-stock companies, such as corporations, to exist. It does NOT limit criminal or tortious liability of the company; all it does is limit the financial liability of investors to what they have invested in that company, so if your retirement account bought stock in a company that went bankrupt, it is only out what it invested, it does not have to contribute toward paying off the company's secured debt.
At the limit, I still see some large stuff coming together unless you rip out the legal system well beyond what I think is optimal. My father works for Chrysler, and I have observed that even as they outsource everything they can, the remaining core is intrinsically huge, as there really has to be an entity that is finally responsible for the car. That means that entity has to have a huge testing component. That further means it needs to do its own design. It also really needs to be the one doing the manufacturing, too, even if that is only a "final assembly" step.
Structuring things so that this wouldn't be necessary anywhere in the economy is probably not currently practical. But ask me again in 20-30 years; technology is rapidly shifting from encouraging centralization to the opposite.
This is an EXTREMELY interesting observation, since most people think "market" and think "corporations," when the corporation is a fairly new thing -- at least, when it wasn't blessed by the govt (e.g. East India Trading 'Company').
It's simple to take it for granted that it's natural, when instead, it's about as natural as bamboo entrained to grow in loops. Made of natural stuff... yes. In natural form... no.
I KNEW all those individual facts, but you know, I never put it together until you put it just so.
What have you been reading that shaped your views/thinking? I'd like to check it out.
The main thing here is that Fog Creek has plenty of margin to use. When there's plenty of profit to go around, of course you can come up with money for everyone.
The key question, then, when applying to a company for a job, is do they have the profit to throw my way?
I feel like my system is simpler. I pay people what they're worth to me (provided I can afford it). Even this system isn't perfect... given a specified experience/skill level, they may be worth more to me 2 months ago than they are now (or vice versa). But it makes it very easy to decide what everyone should get paid.
If they are worth X, but I can only pay Y (where Y < X), then I pay Y (and hope that they'll work for Y) until I can afford to pay them X. At that point, I raise their pay to X.
Also note that, if they are worth X, but you can only pay Y, this should only be a temporary situation. Because, if they are truly worth X, then once you hire them there should be no reason you cannot afford to pay them X eventually (since they are bringing that value to your organization).
Congratulations, you just invented how public school teachers are paid (at least so long as unions remain in power). You have one extra measurement, it's true, but the system is still massively flawed.
Sorry for the inconvenience. I experienced no kinds of popups when reading the article (Firefox+Ubuntu). I would not have had submitted the article otherwise.
"If you worked as a receptionist for six years, for example, you aren't credited with six years of experience; I give you credit for one year." I'm sure Joel's next receptionist will be happy their years of servitude didn't go to waste.
I find it interesting the Spolsky said he was going to stop writing yet keeps coming back. He mentioned on SO podcast that it was because it was such a time suck and that's why he wanted to give it up. Obviously he loves it and just can't...
