Hacker News Reader

I think the public focus on cost has paradoxically led to this situation. If you’re a good engineer, why work for the government? The compensation is extremely poor compared to industry, especially in tech. The lazy/incompetent people at the top have remained there because they’re too lazy/incompetent to leave for much greener pastures. If you fired them, their replacements will be at least as bad (on average, there are some passionate and competent people), since now the pay is poor and evidently there isn’t even job security.

We’ve built a system where we refuse to pay public servants. We shouldn’t be surprised when a system designed to attract below average people leads to projects with below average results.

Is it really true that government jobs pay less than their private counterparts? In Canada government jobs are actually provide far more compensation than private ones but I guess it might be different for USA.

In the US, actual public employees (not contractors!) are generally required to be paid according to the GS payscales: https://www.opm.gov/policy-data-oversight/pay-leave/salaries...

A GS-9, which is roughly a freshly-minted graduate with a Master's degree, starts at $45k/year, growing to about $60k/year (although this is also adjusted by locality). This is well below prevailing tech sector rates.

Thanks for the data, yep that’s pretty bad.

It's not the amount of pay. It's that results (good or bad) don't really impact your salary. Nobody missed a bonus for the project failing and nobody would have gotten one for it succeeding.

The consultants have even worse incentives. The better they do the less they get paid. Meaning, it's better for them if the project drags on than delivering something on time.

This is a very common myth in business.

Bonuses in most situations lead to a decrease in productivity.

https://hbr.org/1993/09/why-incentive-plans-cannot-work https://smallbusiness.chron.com/bonuses-affect-employees-wor... https://hbr.org/2017/03/research-how-incentive-pay-affects-e...

There's a ton of research out there on this. To simplify it basically if someone is doing mindless work like putting a cap on a bottle on an assembly line, then bonuses increase performance in the long term. But if they are doing complex work like writing software they decrease performance in the long term.

Have you ever worked on a federal contract? The compensation really is not poor, unless you’re comparing it to say FAANG compensation, which is not really normal. They make as much, if not more, than current market rates.

There is significant job security (contract with three year period of performance is essentially guaranteed work for 3+ years), annual bonuses, etc.

The idea propagated by many here that engineers who work as government contractors are somehow less than is nauseating and untrue. Believe it or not, a whole world of talented people exists outside of the bubble you live in.

Yes, there are bad apples, as with every industry, and we are all just throwing out anecdotes, but the vast majority of people I worked with in government were A) hard workers, and B) overqualified for their positions, and C) highly educated/talented.

I don't doubt they are hard workers and intelligent. But if OP is judging them by SV standards of talent then it's not exactly fair. Building a successful company requires a certain drive, risk appetite, and revolutionary zeal that are not easily found in government (and to a lesser extent, academia). There are smart people everywhere but most would never make a significant impact.

It requires a difference of priorities and ambitions. What draws people to the DC area is not the same as what draws them to SV. It’s not a disparity of capability. Aside from that, your point really is not accurate. There are plenty of companies and entrepreneurs users in the DC area that generate FAR more value and revenue than SV startups. Take a look at the wealthiest counties in the US. Loudoun County and Fairfax County have been the top of that list for a while. Or Montgomery County for that matter.

Would you truly consider old money, government regulatory capture/lobbying and defence to be in the same category as startups? They are completely different types of innovation. There job is done once a specific policy is achieved. Overarching changes to the system is often not their priority.

Most tech employees in even in the valley aren’t founders or early employees risking it all. There is probably some element of drive for commercial products or something but I think you’re vastly overestimating the degree to which that is a factor vs compensation.

Government is competing with FAANG for employees. There isn’t a good reason to exclude them from comparison. Even so, working for a startup without equity is often more lucrative than working for the federal government. The situation is even worse at the state and local level where you also have to fear your pension being raided via government bankruptcy.

I’m talking about actual government employees, not contractors. The actual government employees who manage those contracts are not paid well. You can see the results of this too in things like the F35 program where the government had no idea what it wanted and repeatedly redirected and added crazy requirements (to be fair, Lockheed is also incompetent).

You’re right though, painting even a majority of public employees as incompetent isn’t fair.

Government employees who manage contracts can be paid very well, depending on their grade. A GS-14 makes over $120k base, with pension, significant healthcare contribution, etc. A GS-15 is $152k base. I get that that may not seem like much, but it is 2-2.5x the median household income in the US.

Aside from that, a contract manager/Contract Officer is typically not an engineer and if you're lucky, may be someone that can fill more of a product owner role. It's not really comparable in those terms.

F35 really isn't a great example of poor engineering talent, that is an example of a crap acquisition strategy in general and was pretty much destined to be problematic from the start due to politics, conflicting/competing requirements, etc. You could have hired the best engineers in the world and they would have likely ended with the same results, because the majority of the issues with the F35 had little to do with engineering.

/[...]not really normal[...]/

Well, there's probably 100-200k engineers combined working at FAANG jobs, who have the option of the FAANG salary or the government salary. FAANG pays so /many/ paychecks that the expectation really is that high for people who can pass a FAANG interview. There are a lot of people who /can/ pass the interviews but choose not to, effectively taking a pay cut to go work on things they're passionate about; only a few of these end up working in 'boring' parts of government. The US Digital Service was meant to make it a broader/better path, though.

Your people description (B) is worth considering, too: I (personally) am unlikely to take a job I think I'm 'overqualified' for unless it's in a specific problem area I really, really care about. And 'overqualified' sounds near synonymous with 'few growth prospects': if the opportunities are there, why haven't they been taken? And if they aren't, why should I take a pay cut to work in a space with fewer opportunities for growth?

The level of compensation is not normal for engineers, that was the point I was making. There are a lot of talented engineers out there who A) have zero interest in working for a FAANG, B) are perfectly talented, C) have jobs that pay just as well. Yes, that applies to those in government too.

In terms of qualifications, people have different things that motivate them to work in certain fields/jobs. People value things other than money as well, such as power and influence, ability to impact change at scale few organizations can match, feelings they are contributing to a meaningful mission/cause, etc. It's not just about money and overqualification is many times because of the way career ladders work in government and the educational, training, internship opportunities the government pays/provides for that you are unlikely to experience in the private sector. Government will sponsor employees to work an internship in the U.S. Senate for 2+ years, or attend a postsecondary education at Georgetown/Harvard/GWU, etc. Not just for senior level employees, for mid-career ones.

> Have you ever worked on a federal contract? The compensation really is not poor, unless you’re comparing it to say FAANG compensation, which is not really normal. They make as much, if not more, than current market rates.

I think there is a disconnect here. Public employees--people who are actually employed by the government--have fixed, low salaries (relative to the tech sector). Contractors for the public sector--people employed by private companies--actually have pretty decent salaries. These two sets of people are different categories, but a lot of people hear "government" and may be failing to realize the distinction between actual government employees versus contract employees.

It is very rare that the government hires engineers outside of contractors. I would argue compensation really is not that bad even for public employees (I was a GS federal employee, work in private sector now). A couple of years in, you essentially have a job for life, more than fair compensation, pension, 401k matching, etc. BLS' estimate of average salary for a software engineer is 107k per year. It is not all that challenging to make that much as a government employee in the DC area (GS-13 step 2 is about what is comparable).

GS-13 is a significant amount of experience and 107k isn’t a competitive wage for an engineer of that experience level, especially in tech.

How can the government provide adequate technical oversight of its contractors if it doesn’t hire engineers? Answer: it can’t, and you can routinely see the results.

GS-13 is possible within two years of experience and depending on the agency, sometimes quicker. Yes, I can personally attest to this and it will depend on the agency. To reiterate though, outside of a few agencies, most engineers in government are contractors, not civilian employees. Their compensation is much higher.

Just to be clear, GS-15 is the absolute highest pay grade for regular federal employees. A GS-13 is supposed to require at least a Masters and a GS-14 a PhD. If you got 13 after a year, you have very little potential growth for the rest of your career. $107k really is not a very high wage relative to the labor market for high end engineers, especially in tech.

Again, why would a good engineer work for the government? If the government can’t and won’t hire engineers, how can they create informed requirements for contracts or provide technical oversight of their contractors? You can see this playing out in boondoggle after boondoggle. The government poorly specifies a contract, a contractor bids on the contract, the government changes requirements over and over because the lack of technical expertise implies they have no idea what they want, the contractor rakes in tons of money due to direction changes, and eventually the contractor delivers something shitty because the people at the government also had no idea how to evaluate whether the contractor was meeting milestones.

Ok, just to be clear, a GS-13 is comparable to a masters and a GS-14 is comparable to a PhD IF you are using education as your sole means of qualifying for the position. I was hired as a GS-11, promoted to GS-12 after one year, GS-13 after another, received two qualified step increases the year after, plus a regular step increase. So in a matter of three years, I was effectively promoted six times. Once you are on a career ladder there are no education requirements.

In terms of GS-15, their pay goes up to 170k, so its not like GS-15 step 1 is the end of the road. Federal employees have also been receiving above-inflation pay raises and COLA adjustments the last few years.

So why would a good engineer work for the government? Honestly, if pay is your only metric you use to choose your place of employment, then I would not expect them to. However, as I have mentioned, there are plenty of people in the world who choose their place of employment based on many factors other than compensation.

Is government work frustrating? Sure. I'm sure the same can be said of for the engineer developing a CRUD application and has a shitty senior engineer overseeing them. Government work can also be incredibly rewarding and unfortunately, the majority of the time you only hear about the government failures instead of its successes. The federal government issues in excess of 500 billion in contracts per year...I assure you, there are successes in there and I have seen some federal agencies that have implemented acquisition processes and development practices that are competitive with some SV companies.

Not everyone is driven by money, but I think its fair to say that most people would take a job in industry where you can make 2-3x over a government job. I think this is born out in your own admission that it is very rare for government to hire engineers. Above inflation / COLA pay raises are table stakes for good engineers. If a good engineer in industry has only managed to get COLA pay raises over the last 5 years, they should quit.

I've not claimed that government can't do anything right, and I strongly believe in government. I do however believe that current public policy is designed to cause government to fail. Lots of contracts in that yearly $500 billion are successful, but things like the F-35 program, the Littoral Combat Ship, the Space Launch System, Healthcare.gov, the fuck ups in this article, and more are big enough that they eat pretty significantly into that $500 billion budget. The government has some serious issues procuring engineering work. Some pork politics certainly contribute, but lack of technical expertise at federal agencies is, I think, a significant factor.

I don’t agree with you that it is safe to say if people could make 2-3x more money they would take the opportunity. It is equivalent to saying people who become teachers, who make far less than most professions of similar educational requirements, would jump at an opportunity to make 2-3x the pay or are somehow incapable of performing comparable functions. It’s just not true.

I should probably clarify that there are some federal agencies that hire engineers as federal employees. For example, NSA, NASA, JPL, etc. To your point, they do have better success when it comes to engineering. However, they are also agencies that have a primary mission/function which requires it. For other agencies, it’s not all that easy for them to justify career engineering employees when the roles are tertiary functions. It’s not to say they don’t have them, but it is to say that they have far less of that expertise. That is why entities such as 18F and USDS exist though, to help fill that role.

While it seems like you are not keen on giving it a try, I highly encourage you to give government work a shot even if only on a temporary basis with somewhere like USDS or 18F. Only way to help make positive changes to the way government does business is through people with the drive and talent to do so. There is plenty of opportunity to influence those types of changes, even at lower ends of the employment hierarchy.

Appreciate the conversation and you offering your perspective!

> Have you ever worked on a federal contract? The compensation really is not poor, unless you’re comparing it to say FAANG compensation, which is not really normal. They make as much, if not more, than current market rates.

Yes. It was OK pay, but not better than market. Annual bonuses weren't a thing. There was no path for career advancement beyond senior engineer. The supposed stability was not nearly as good as you describe.

There's a bunch of reasons I stopped doing defense contracting.

I had a friend that was a fracking engineer and made big bucks working on drilling sites. Constant travel, awful hours, "culture" of working all the time.

He took a big paycut to start working for the state government but he has good benefits, good retirement, and normal hours that you can actually raise a family with. And while all of his peers have scattered the nation looking for jobs when natural gas prices crashed, he's still sitting pretty with his same steady job.

So there is a lot more than just salary to working for the government.

I don't see how your argument fits the situation, the government is clearly not focusing on costs if it's wasting millions of dollars on obviously not viable projects and paying consultants $200/h to sit around for months waiting for a security clearance...

There's definitely a focus on cost. It is just misguided. Short of not having security clearances, I'm not sure how you can avoid paying people to wait for them. Simple projects, like properly paying people unemployment benefits, have become not viable for the government to execute. There are many reasons why this has happened, but I think the erosion of technical expertise in the government is a big contributor. Rather than paying market rates for engineering talent, the government hires a bunch of individually cheaper non-engineers to oversee technical projects. The government could save a lot of money by hiring a handful of competent engineers rather than a ton of nontechnical management.

Relatedly, I think the government could probably save a ton of money with some in house technical talent to execute basic projects like the systems in this article rather than going through the crazy procurement process. It isn't clear to me that contractors are providing any added value to justify the added cost for their profit.

Step 1: Fire all engineers on government money, to decrease "overhead" and let "industry" do it

Step 2: Parent post

Step 3: Government is inefficient, inept, if not outright corrupt, we must gut it further! Goto Step 1.

The other day I was getting data from the USGS, and it was remarkable how well structured, sorted, searchable the data was, how it came in high quality, with a custom-built downloader applet that worked very nicely actoss multiple machines. I went from zero to complete solution in QGIS in one day. It's not all that bad.

The saddest thing I witnessed were consultants like you describe bragging about how easy it is "to milk the cow" (to quote one). They bragged that they never earned their bonus in such an easy way.

“There are four ways in which you can spend money. You can spend your own money on yourself. When you do that, then you really watch out what you’re doing, and you try to get the most for your money.

Then you can spend your own money on somebody else. For example, I buy a birthday present for someone. Well, then I’m not so careful about the content of the present, but I’m very careful about the cost.

Then, I can spend somebody else’s money on myself. And if I spend somebody else’s money on myself, then I’m sure going to have a good lunch!

Finally, I can spend somebody else’s money on somebody else. And if I spend somebody else’s money on somebody else, I’m not concerned about how much it is, and I’m not concerned about what I get. And that’s government.” -Milton Friedman

I also worked on a state unemployment system, and it was much the same. Took a couple weeks to get my laptop, but I still had to come in and sit in a chair and read the physical newspaper. When it did arrive, it took the better part of an hour to run the build.

That was the last government work I ever did. I lasted about 3 months.

This means that ultimately, all tax-paying citizens are that whale. Even the very consultants/vendors doing it.

Oroboros