Lots of comments below are accurate e.g. terrible gov't oversight, extremely large and complex projects, corporations responding correctly to bad incentives, corporations thriving off the complexity of govt contracting
However I will posit an additional layer to this...and it lies with the American people themselves. We just cant handle gov't officials and contractors making a ton of money. The real root of the complexity in govt contracting is FAR (Federal Acquisition Regulations) Which create the complexity to avoid egregious corruption and profits in the federal space. While on paper it sounds great...lets make rules that limit what corporations can make off of the government... what you actually get is tremendous waste. Redundancy, arcane billing practices and auditing, and huge bureaucratic RFP bidding systems that limit competition. A classic case in not doing a complete cost-benefit analysis.
The solution is not palatable to the general American public. Put up with some fraud and corruption in the federal government which would be offset by increased competition from the private sector (thus better products and prices). And dont kid yourself if you think fraud and corruption doesnt still go on...it's just more sophisticated.
IMO, the real solution is to let the government compete with the private sector on 'government' contracts. Basically, do an analysis on how much it would cost to do in house and then put the same contract up for bid. (aka a Sanity check) However, you need to add a companies reputation to their bid so if they say it costs X and they tend to be 30% over budget or have quality issues then you don't just accept X. With that said and done if someone can really deliver high quality and still make a 20% profit then let them.
The government's hiring process makes this impossible. They are too slow, the people who know what needs to be done are completely separate from the centralized offices which hire, and therefore the government only hires people based on arbitrary measures including keywords on resume, number of years in x industry, certifications from Craposoft and Orac-ket. They are fucking fantastic at hiring useless middle-management fucks whose only ambition in life is a paycheck and to stop using their brain when they leave the office (at 4:00 or 3:30 pm on a weekday)
Ambitious people leave in droves, because an organization run entirely on "other people's money" always steers towards complacency and lack of accountability, translating to ambitious people who care about delivering on-time being surrounded by people whose kid's softball game takes precedence over leaving customers out in the cold.
Anyone who has EVER dealt with the federal government in any fucking way knows that what you are saying would never work. The laziest people I have ever encountered in my life are in the Federal government. I'm a huge advocate for the environment, and my first dealings with the EPA made me want to cry. They are horrifically lazy at a cultural level. The DoD is fucking awful. Go to a military base at 3:00 PM on a Friday afternoon and watch as a traffic line builds up at the exit gates as the civillian workers go home 2 hours early, EVERY weekend.
There are some hard workers in government - you see many around DC. But I don't disagree that there are some lazy people in government - I have heard much worse about the EPA than you have stated for example.
The military is some of the hardest working but criminally underpaid.
The whole contracting system is broken - the amount of money government leaks there is criminal.
Military are not criminally underpaid. When you factor in housing benefits, pension, health care, education, etc, the military is a fantastic career option, which is why so many people join it. For a non-college graduate, it is absolutely the best career path you can take in the US.
When you get out, you have access to federally subsidized loans that nobody else gets, as well as the ability to have Tricare (ultra cheap health insurance) for the rest of your life. When you consider that less than 10% of active duty military will ever see combat, this is a bargain.
I am not sure this is feasible in practice. The government is not a business so "competing" isnt a fair rubric. The government has unlimited resources and does not have to "make a profit". I dont know how you could create an apples to apples comparison. In the business world it's easy...profit.
Companies reputation is a fuzzy number too. As the stock market adage goes "past performance does not guarantee future success". Sure brand reputation means something but how do you measure that and create equal comparison...across hugely disparate government contracts. Comparing the US Census contract (yes this is contracted out almost entirely) to the F35 contract would be absurd.
I do not necessarily have the end all solution. I tend to favor smaller government because I am not sure our current political system scales to be able to execute large projects (I tend to think the ideas behind many large scale projects are correct but fail when you consider the costs of implementation within our political constraints).
I think the idea behind having the government compete wasn't that it was a fair competition, it was just a way to spend the least amount of money.
If the government is able to do something internally for $500m, and the disadvantaged private sector is going to cost $750m, then let the government do it themselves. (On the government's books, this is a net win.) If the government costs it at $500m, but a private company can do it for $400m... Why let the government do it? Contract it out and save money.
Okay, but who determines the quality issues? The government could easily fall into the same issue that some of our clients do.
Yes, that competitor provided you a solution for 10% less, however it's horribly bloated and inefficient and will have double the on-going hardware costs, and the solution itself is a fucking mess that is going to increase the cost of any changes or maintenance ten-fold.
Problem is, they don't know the difference. The people who will best know the difference are going to be the people who work with/in the same area - the supplier's competitors. Far from impartial.
So maybe we set up some sort of oversight agency, but then we either have them make snap judgements without the appropriate understanding and penalizing what may be more difficult/complicated projects (meaning no one will want to bid on them), or we create such a huge overhead in documentation that we're pretty much back where we are now - no one but the largest of players can even afford the overhead to participate.
I would also speculate that many of these projects are subject to horrendous scope creep where accounting systems are being used for far more than just accounting needs. These additional requirements need to be evaluated and excluded from accounting and auditing systems if they are beyond the scope of such systems. Instead they should be rolled into other systems that have a narrower scope.
For example, the posted article states:
"while an impenetrable tangle of logistics and personnel systems
can hinder commanders’ ability to know who and what are
available for deployment."
That to me sounds like scope creep. Yes, commanders need that capability, but that is not a feature that should be part of an accounting and auditing system. Of course these is somewhat relationship between a soldier being paid his wages and being available for deployment, but that is a weak relationship that is best handle via a public API for each system that other specialized defense systems can consume.
I would posit that the contracting approach does nothing to prevent scope creep since most contractors participating in this system are glad to let things keep snowballing because it means more billable hours.
When you put it like that, paying for such oversight is well worth it. Would the military request less if it was more efficient or would it keep its budget and just fight (== exchange GDP for the "benefit" of additional limitless risk) on more simultaneous fronts?
Really, I don't see that many private sector companies competing on the big contracts no matter what you do. But for the smaller profit and more general purpose projects, the problem is that their purchasers interfere with the process to get what they want/need as the buyer. For example, they needed our systems and we didn't much care about learning their processes, so an Aerospace seemed to take the bulk of the profits by making a package that depended on us.
If the system were equally complex yet more transparent/published and tamper resistant then I don't think corruption would be so rampant and cost may at least be allocated to regular companies that choose to navigate them alone. But with no system at all, I think the Aerospace that took most of the profits would have taken even more and delivered still less.
One aspect that's missing in these comments is an acknowledgement of the complexity of the projects which are the subject of defense contracts. The Pentagon and its contractors are literally at the edge of the state of the art in fields like aerospace. Very few companies have the expertise to even attempt to take on these contracts, and most projects have ill-defined specifications simply because nobody is quite sure what is technologically possible until the project is well underway. And nobody knows how much money will be needed up front. That's like asking Zuck in his dorm room: "how much money will you need to build Facebook as it exists circa 2013?"
I used to work at a company that did DOD contracts (in the wireless sector). In that area, what the DOD was playing with in the early 2000's is what VC-funded companies are playing with just in the last few years. For the work we did, there were no detailed specs because nobody really knew what was within reach. In response to the question: "what do you want?" the DOD's response is often: "what could we have?" Pointing at how much more efficient private industry is makes no sense. Take an example like Space X. They're a shining example of private industry, and are doing very innovative stuff, but they're treading a lot of the same ground NASA and DOD contractors did in the 1960's. Of course they're getting it done for a lot less money! They know what's possible, they have a general idea of what works. The contractors doing this stuff in the 1960's had none of that to work with.
Obviously the time is due for modernizing the accounting system. But it's important to realize that a lot of the mess dates from the Cold War, where keeping up with the Soviet Union was the priority, not satisfying the bean counters.
This is very well explained at the book moon lander, it's an amazing inside story of the problems they encountered during the design and comstruction of the modules. What has surpriced me is that the design effort was minor compared with the management and quality control. They had to multiply the budgets severaltimes to be able to correct unexpected problems.
I don't think that having a bucket for "Gold Toilet Seats" or "Classified Research" is the key problem. I think that the problem is that there is a gigantic integration effort that fails due to a variety of interlocking political factors and the malincentives to get the job done right.
I have to agree with you about dealing with the Soviet Union it made accounting an after thought. Now the war has moved to an economic front (if it was not always that). And a fighting force that cannot effectively keep track of its resource usage is a vulnerable one.
Nobody has a problem with funding experimental research and projects. That is not an excuse to not follow basic accounting principles and, quite frankly, have no clue where money has gone.
The biggest problem I see in these articles is the insistence on using off the shelf systems. SAP and Oracle are always at the center of these deals, and they end up being disasters. If you have MULTIPLE billion dollar failed projects using COTS systems, maybe its time to step back and look at a custom system.
The root of the problem isn't using or not using COTS.
The root of the problem is the government keep hiring consultants they pay by the hour. They they act surprised when the consultants respond rationally to the incentives.
Or they agree to a fixed price contract where everyone knows the government will pay you more if you say you can't deliver and threaten to go bankrupt. Then they act surprised when the contractors respond rationally to the incentives.
The government needs to break projects down into smaller chunks, to mitigate the losses if a supplier doesn't deliver, you don't pay them, and they go bankrupt. Then they need to establish and enforce an incentive system where suppliers are punished, not rewarded, for failing to get the job done.
This. The tangled, incestuous web of government contracts and bids is such a colossal failure that is destroying the US government from the inside, out. It's easy to point fingers a politicians and public figures, but in reality you have a much larger network of largely hidden corporations, who's entire industry thrives on overly complex process and success by failure. I really hope the startup disruption trend in the next decade involves completely reworking how government contracts are fulfilled. If you can figure out how to crack that, the market opportunity is vast.
>The government needs to break projects down into smaller chunks, to mitigate the losses if a supplier doesn't deliver, you don't pay them, and they go bankrupt. Then they need to establish and enforce an incentive system where suppliers are punished, not rewarded, for failing to get the job done.
that means government getting an expertise (at least at the PM level) in that specific domain. While it wouldn't be a bad thing (and actually a very positive shift vs. current state of affairs), it would immediately hit the typical "government shouldn't be in the <insert specific domain> business".
Anyway, just to cheer you up - compare US Gov / DOD corruption & waste to the cost (i.e basically plunder) of Sochi Olympics in Russia. While making great strides, USG has still long way to go before making into top corruption league :)
Or, the root of the problem is in treating these companies as too big to fail.
If the big companies were allowed to fail, that would develop a market for others to enter who could succeed.
With gov't contracting, it is unrealistic to leave aside the political connections the companies have. It is simply unacceptable to the political crony class to allow one of their own to fail.
> If the big companies were allowed to fail, that would develop a market for others to enter who could succeed.
If the big companies were allowed to fail, maybe they'd stop underbidding on contracts to get them based on their reputation and smallest bid (knowing that there's no risk), and instead bid actual costing.
To solve problems, first you have to identify the problem, then examine possible solutions to establish how the solutions will solve the problems.
It is not clear to me how a custom system will solve any of the problems of the process. I'm not close enough to know, but the primary problem is probably entrenched bureaucracy that has been given no reason to change and perceives major threats-to-power if they change. A custom system seems more likely to exacerbate the problem than solve it by giving the entrenched bureaucracy even more knobs to demand they can twiddle (and knobs on the knobs, and knobs on those knobs....), though admittedly the fail is flowing so thickly here it would probably be impossible to distinguish the additional fail from the already stratospheric baseline.
These solutions are highly customizable. I'm not a defender of SAP/Oracle, but I don't see how writing your own package would be any more likely to succeed than paying an ERP vendor.
An ancient and creaky administrative regime like what the military has is nearly impossible to manage financially. The GAO has tried and essentially given up on trying to audit.
the military CAN get things working when they can't write checks. They have a system that works for THAT function and has for 50 years. So it's a matter of LEVERAGE to get compliance.
The problem with Oracle and SAP is that they aren't "home builders" they are parts and tools salesmen. They don't know ANYTHING about BUILDING HOUSES... they just sell the tools. even their own businesses aren't actually GOOD businesses with huge amounts of nepotism and politics. The first thing when you sign up with them (or any of the big named software vendors) is that you have to find yourself a "contractor" just like building a house. The "big famous software company" disavow any actual results, your mileage may vary, etc.... They're like the "Music Man" selling boy's bands.
If you want to audit the military, that's where North Korea has the correct idea. Generals don't "retire" until the leader says they do. Start enforcing Sun Tzu style discipline to get this done... knock off a few 5 star general's heads. That'll get things done. At a minimum, strip them of private residences, and put them in boot camp barracks (complete with scrubbing their own floors) till they get this thing done.
When they passed Sarbane-Oxley they gave corporate executives just a few years to implement it. Don't implement it and we DELIST your stock. Implement it WRONG and go to jail. Needless to say, public traded companies worked VERY QUICKLY to get it done and fired people at will if they were in the way.
when the military needed to PAY people.... they got the job done well. that system celebrated 50 years of service getting checks out. So it's just a matter of getting all those disparate systems in line with the check writing system. Just nobody has the political will to accomplish it.
This is where the commander in chief not being able to summarily execute people in the military has it's drawbacks.. cause that's how Sun Tzu got the palace maids to properly military drill... do it right or get killed. Oldie but goodie.
>"The Pentagon alone has never been audited, leaving roughly $8.5 trillion in taxpayer dollars unaccounted for since 1996, the first year it was supposed to be audited."
Wow - that means an average of $500 billion per year of taxpayer money unaccounted since then!
From reading the article, one distinct impression is that the real reason for failed accounting may be simpler and more troubling than all the other obvious ones : because it is so much easier to hide inefficiency, incompetence, and gross abuse/misuse of funds if there are no proper accounting systems in place. The freedom from oversight can be addicting. While there may be no grand conspiracy in place at the top to achieve this - the will to fix what is broken could certainly be weaker if it gives you the freedom to do as you please.
That's ridiculous! These wars are just fronts for money laundering & control of oil. It's only by coincidence that they can align some wars with some human-rights violation. Otherwise, they'll just make stuff up.
We need a revolution, there I said it. Go ahead NSA and add me to the watch-list. Nobody here doubts that you scan this site daily.
Well, there might be good reasons for having that amount of cash - but you can't help wondering what the processes and systems there were for controlling what it was all used for..
Edit: This attitude by a US general doesn't exactly come across well:
Q: "But the fact is billions of dollars have disappeared without trace."
Oliver: "Of their money. Billions of dollars of their money, yeah I understand. I'm saying what difference does it make?"
And not just thievery, I searched the article and found nothing about secret programs. "consequences for failure" isn't the problem, the real problem is "consequences for success"
No tinfoil hat stuff needed, merely look at history, opening the books of the Manhattan project or area 51 in the stealth fighter years would not have been very wise. I suspect as a former .mil guy that a lot more money disappears into stuff like that, than is officially admitted, or is lost to corruption.
If you think a simple IT project is expensive when .mil does it, doing anything secretly insta-quadruples the cost if not more.
If its anything like what happens when Oracle or similar come in its this, they spent so much that it would be a waste to not keep spending.
That and the higher level people who started it, bought into it, put their name on it, are the last people who ever back down and are damn near not approachable by those in the trenches.
Seen it far too many times. Backing down cost reputation which is more expensive than money from someone else
I couldn't imagine the pain involved in being attached to one of these projects for multiple years. I suspect most smart and ambitious people agree with me. So in the end, you add a lack of talent to all the other very real things that others have added.
However I will posit an additional layer to this...and it lies with the American people themselves. We just cant handle gov't officials and contractors making a ton of money. The real root of the complexity in govt contracting is FAR (Federal Acquisition Regulations) Which create the complexity to avoid egregious corruption and profits in the federal space. While on paper it sounds great...lets make rules that limit what corporations can make off of the government... what you actually get is tremendous waste. Redundancy, arcane billing practices and auditing, and huge bureaucratic RFP bidding systems that limit competition. A classic case in not doing a complete cost-benefit analysis.
The solution is not palatable to the general American public. Put up with some fraud and corruption in the federal government which would be offset by increased competition from the private sector (thus better products and prices). And dont kid yourself if you think fraud and corruption doesnt still go on...it's just more sophisticated.