Hacker News new | comments | show | ask | jobs | submit login
G4S is just the latest in a long line of outsourcing disasters (newstatesman.com)
32 points by irv 1983 days ago | hide | past | web | favorite | 38 comments

Here in the UK, a lot of activities that were traditionally the responsibility of the state (health, education, policing) are increasingly being outsourced. Conventional wisdom is that the private sector is just much more efficient than the state can ever be, but I have never been able to find evidence-based studies for or against this. It always seems to taken as axiomatic. Can anyone point me to unbiased work that has been done on outcomes in outsourcing of state functions?

As someone who works in the private sector, I find the idea that it is just intrinsically more efficient than modern state organisations to be pretty hard to believe. But thats just anecdotal, and I'd like to know more and maybe have my opinion modified.

> Conventional wisdom is that the private sector is just much more efficient than the state can ever be, but I have never been able to find evidence-based studies for or against this.

This is hard to measure empirically because of selection bias - you can only study things which have already been privatized, and the first things to be privatized are the things you think will privatize successfully. For example, there are quite a few countries that have successfully privatized their telephone networks. But that doesn't mean you could privatize the police or military with the same success!

And because you asked for evidence-based citations, here you go: http://www.quora.com/Is-there-empirical-evidence-to-support-...

I think this is a case of the usual issues of in-sourcing vs outsourcing. I know a large insurance company that flip-flops on their decisions every 3-5 years. When in-sourced, staff are "lazy, cost too much, can't be fired due to HR rules, etc, etc" so they outsource, at which point the outsourcer becomes "they are not responsive, too hard to work with, every little items needs to be tallied against the contract, every little change requires too much time and requirements specification, etc, etc" so they turn around and in-source again, to begin the cycle over again.

In theory, in-sourced allows for better flexibility on what work is being performed (you are already paying the dollars for hours, so you can allocate as you desire) but you may lack expertise, less flexibility to expand then reduce/contract the workforce (which is especially true with governments and their unions), too hard to get rid of people if they do not grow skills and adapt as the organization changes, etc. Outsourcing works well when there are well defined boundaries, and requirements to expand/contract the workforce (within reasonable limits, the problem here is that the Olympics are too big for an outsourcer to deal with because there are not enough staff in reserve).

None-outsourced government IT projects aren't any better.

We decide to install system X so we need to hire some experts in system X. But system X experts are expensive and if we hire them as staff they are paid according to fixed internal pay scales - or there is a hiring freeze - or we have to hire 'internal candidates" from other government branches first.

So we hire consultants. These are a lot more expensive. Then part way through the project there is an investigation into how much "external consultants' are costing. So there is a ban on external consultants which the government shows has saved us $$$ million.

The result of course is that we have a half-implemented system X that costs us $$$$million and doesn't work.

> Conventional wisdom is that the private sector is just much more efficient than the state can ever be, but I have never been able to find evidence-based studies for or against this.

The train of logic is that if coprorations become efficient to maximize returns on investments and products sold, then since government programs have neither a need for a return on investment and exist outside of market economics because all the associated costs are a forced payment by the "customers" that they have no reason for themselves to be efficient. The government as a whole might want them to be efficient, but there is no intrinsic benefit for a government program to maximize efficiency because that has no influence on its short run continued existence.

Not saying it couldn't be as efficient, but large programs end up with too many people in positions of power which means there's probably at least one bad apple in there trying to milk the lack of demand economics or operating costs.

It's the same as buying any other sort of goods and services. Do you think the NHS makes its own paperclips and grows its own potatoes? No, the private sector really is massively more efficient, for two absolutely crucial reasons: 1) there are actually many sellers with interchangeable products and 2) paperwork and catering aren't actually core to the NHS mission.

If 1 and 2 aren't true, if there are a limited number of sellers of the service, and the service is core to the mission, then outsourcing will certainly fail.

The majority of government outsourcing deals are ideology or to hide the numbers.

Take PFI, if I offered to take over the rest of your mortgage on your house. But you had to pay me equivalent of the mortgage payment for 100years and you had to buy my services for any cleaning or maintenance of your house for the 100year term - then even though you were now mortgage free (ie. you had reduced your deficit) - you would be mad to take the deal.

If however your only interest was being able to claim that you had reduced your deficit going into the next election - and it wasn't your money anyway - you might look at it differently

During my time in the USA, I saw that most people think the private sector is inherently more efficient than the government. As you said it seems like an axiom.

However, I disagree it's people+procedures+technology that make a great organization and none of those concepts are exclusive of either private or public sector.

The situation is not as black-and-white. The private sector is more efficient in some cases, but that doesn't mean _everything_ should be privatized.

A trip to the DMV (Department of Motor Vehicles) will probably leave anyone with the impression that government offices are dens of inefficiency and bureaucracy. And that may as well be.

I think the sweet spot lies in the middle. There is no reason, for example, for the government to put a (say) landscaper on its payroll to maintain a public park. The overhead of a FT government employee is just too much; the job can be done well by a myriad of private companies. Same goes with bus drivers, janitors, painters, etc.

If the job requires domain knowledge and other such intangibles, then it usually can't be outsourced. This is why many IT projects designed to "upgrade" or replace existing IT systems at government agencies suffer: there are a lot of intangibles that just can't be all laid out on the contract form.

I think people often confuse the axiom...

Stated more correctly, smaller, more focused, organizations outperform larger, more bureaucratic(more management) organizations for particular tasks.

It's not that governments are outperformed by the private sector every time for everything (some things require size and longevity). Conceptually, it has nothing to do with whether an organization is private vs public. It's about incentive, regulation, and consequence horizons.

Incentives Generally, private organization's incentives (if tied appropriately to the desired outcome) are more aligned and offer higher rewards. Government organizations, if completely transparent, have a hard time justifying high rewards to their constituents. All other things being equals, people prefer higher rewards. So, the talent tends to accumulate in the private organizations. There are some exceptions to this (political influence), but I'm generalizing. :)

Regulation The smaller the organization, the fewer rules. Not only do laws tend to work this way, but internal organizational rules work this way as well.

As organizations grow, managers try to scale their ability to influence direction by legislating organizational rules. As more and more of these rules come into play, the overhead and unintended consequences of these rules grow. Since these managers are unable to directly participate in the execution of the tasks (the reason they created the rules at the start), they are slow in detecting changes that should require rule changes.

The overhead and lack of agility created by both size and expanded regulation(rules), makes large organizations far less efficient than smaller organizations. Unfortunately, governments have disproportionally larger sets of rules. Not only is their ruleset larger, and their employment not aligned to performance, but they also have unionization that has a differing set of goals. This creates even more operational friction making it very difficult to allow people to do what they do best. Tragically, this makes efficiency and productivity in large government organizations nearly impossible.

Consequence Horizons Smaller organizations tend to have better alignment between their existence and their ability to execute. This is because they are paid for execution and are typically unable to structure longer term contracts. As a result, if the people paying the bills are not happy, they cease to exist quite quickly.

As organizations grow in size, they start to manage for risk avoidance and not for execution. The larger they get, the more they are able to influence the market to avoid direct competition, to change the criteria by which they are judged, and to structure contracts such that they are far more stable over longer periods. This 'cushion' allows these organization to ignore or tolerate consequences to a higher degree. As a result, they are less sensitive to market judgements about their efficiency, productivity, or ability to execute. They loose sight of that as a goal and start to engage in higher order goals (happiness, employee well being, brand, reputation, influence, etc). These are not bad necessarily, it just makes them less efficient at particular tasks.

In particular, large government organizations are disproportionally disassociated from consequence. This is directly related to election cycles, politics (spinning the message to create success out of failures), and constantly getting inexperienced leadership. The first two are points are obvious, but let me explain the last. New politicians are elected and they appoint the people that they feel are best to run these large organizations. Typically, neither the politician or the person appointed has never spent a day working in that organization or executing the tasks for which they are responsible. With large organizations, it takes truly special managers to make an organization efficient at something you have never done yourself. This is even more difficult when you are not answering to your boss for your ability to execute, but instead are answering based on the ebb and flow of the political landscape.

In general, my point is, that you are right. Large private organizations do not have a significant advantage over large public organizations. That advantage can be widened based on political systems. However, the real advantage that people reference when making this argument is the enormous advantage small to medium private organizations "enjoy".

Understand that I am only talking about efficiency and execution and not societal value (differing goals). I also recognize that there are many tasks/goals that require larger organizations and more longevity.

Very informative post. Thank you!

Sometimes outsourcing works, sometimes it doesn't. Highlighting a few examples in which it has failed does not lead to the conclusion that it never works.

So the author seeks to bridge the gap by identifying a fundamental flaw in the concept of outsourcing, in this case by noting the divergence of objectives between the two parties. Although true, this isn't the fundamental flaw the author is looking for. It simply identifies the importance, for outsourcers, of making their suppliers' objectives coincide with their own.

I wouldn't say the G4S case is one in which the government has failed in making objectives coincide. G4S will, in all probability, be found in breach of contract for failure to fulfil their obligations, and the government have already stated they will be seeking to recover costs from G4S, such that there will be no additional cost to the taxpayer for this debacle.

Outsourcing is seen as a panacea because it is believed you can offload worrying about the details. All too often this means the outsourcing process itself is handed over to people who don't (or don't want to) know the details. Hence we get fiascos like this.

The result is we are likely to have an Olympics with an overt military presence. Still this should make a lot of overseas visitors feel at home.

Oh, and the New Statesman is explicitly left wing in its politics, I guess for many Americans this may be the first time they have been exposed to left wing opinions as distinct from the Republican Party definition of left wing which seems to be "less right wing than I am".

> Outsourcing is seen as a panacea because it is believed you can offload worrying about the details. All too often this means the outsourcing process itself is handed over to people who don't (or don't want to) know the details. Hence we get fiascos like this.

You're absolutely correct, but it does not follow from instances such as this that outsourcing is itself the root of all evil, as this author would have us believe.

Furthermore, the case in question is hardly an example of poorly managed outsourcing. G4S were, by all accounts, the most suitable firm for the Olympics security contract. That said, it is a largely unprecedented event, and not unsurprisingly they underestimated the task, and couldn't get enough staff trained in time.

Taxpayers don't stand to lose out as a result of this. G4S are to blame, and G4S will have to eat the cost of sorting the matter out; taking a loss of up to £50m on the contract, it has been reported.

> The result is we are likely to have an Olympics with an overt military presence. Still this should make a lot of overseas visitors feel at home.

13,500 military personnel were already scheduled to provide security for the Olympics, and the government has only committed an additional 3,500 in response to the G4S revelations.

> Furthermore, the case in question is hardly an example of poorly managed outsourcing. G4S were, by all accounts, the most suitable firm for the Olympics security contract. That said, it is a largely unprecedented event, and not unsurprisingly they underestimated the task, and couldn't get enough staff trained in time.

Yes, it's not as though this event happens every 4 years...

In fact, G4S managed to spectacularly mess up the security of the recent Jubilee celebrations, as was widely reported. When you add that to the utter mismanagement or the (despicable) asylum detention centres, and on and on... they're as qualified to provide services in the same way Capita are qualified to provide IT consultancy.

But on the poorly managed part, Teresa May claimed the contract was routinely monitored. and yet this is still a surprise to her, 3 weeks from the start of the event!

Recovering costs from G4S will only make G4S recover the recovered costs from other areas of their business, so I expect we will see more escaped prisoners and more security lapses at nuclear facilities, etc.

G4S isn't a non-profit.

Besides, the firm will have contractual obligations in other areas of the business (e.g. prisoner transportation and security at nuclear facilities), and if it reneges on those, it will face similar consequences.

Which could eventually lead to bankruptcy and still stick the taxpayer with the bill and poor service provision.

In any case I agree with your OP that these types of cut and dry presentations are never helpful, and we should be seeking to understand the appropriate context for different solutions instead of generalising.

> Which could eventually lead to bankruptcy and still stick the taxpayer with the bill and poor service provision.

Precisely. Thus G4S have a clear incentive not to let their standards slip elsewhere in the business and face further damage to their finances, or for that matter their reputation.

G4S is a massive multinational with ~£7.5bn annual revenue. They're not going to spiral into bankruptcy over this Olympics debacle.

or for that matter their reputation.

That they still have a reputation worth defending is fairly surprising given their history.

While I'm no fan of G4S or PFI projects, this article is just empty wind.

Where is the evidence that "cash [...] is redirected into shareholders' pockets"? This would be an ideal topic for a scientific study: do outsourced projects cost less for the buyer, or do they cost the same/more and does that extra money go to shareholders? Unfortunately reading this article will not tell you the answer to this or other interesting questions.

I can't really see how you could spend £21k per person on security for an event lasting a couple of weeks.

Do you really think the excess after expenses is going to wages rather than to shareholders? How can they not have enough people if they're paying £8k+ per week. Have they asked TA members if they'd take 3 weeks off work to earn a years wages?

[I read in the 'paper' theyre getting 268M for supplying 13,700 staff and that they are using some volunteers in that number]

The article is too one-sided for sure, but the fact that the cash that doesn't go to the workers goes to the shareholders' pockets is practically a truism: That cash is called profit, and shareholders (rightfully) expect it from their investment.

I think what rwmj is saying is:

State total cost = State front line workers' salaries + State overheads Private sector total cost = Private sector front line workers' salaries + Private sector overheads + Profit.

Overheads are everything from paddy wagons to radios to uniforms to back office staff to police horses loaned to Rebekah Brooks to the G4S CEO's £830K/year salary. Private sector firms will presumably only bid on a contract if the profit is positive, I'd agree that's a truism. But is the profit positive because the private sector total cost is higher, or because the private sector front line workers' salaries are lower, or because the private sector overheads are lower?

I think that's what rwmj wants evidence of.

The problem is that usually the people doing all the work get very little whilst the people who're just being rich get made richer. Yes within the current system a small return on investment should be expected but a commensurably larger wage should be paid before a large return is given to investors.

It's by no means a universal truth that an investor deserves profit.

I didn't say that an investor deserves (which is a moral judgement) a profit, I said that she rightfully expects to make one - otherwise she simply wouldn't invest.

It's not a "truism" at all. For a start, companies are going bust taking these contracts[1] -- going bust generally being a good sign that they aren't making a profit. More importantly, the alternative is a state-owned security service (police, army, etc) who are likely to be inefficient, highly unionized with tons of overhanging pension payments. Is that cheaper? I don't know which is why I was saying in my original comment that these things can be studied.

[1] http://www.out-law.com/en/articles/2012/july/special-adminis...

The article said that "the cash not going to the work force is redirected into shareholders' pockets". I don't read anything in that phrase about the service being cheaper or not. I just read that if the contractor gets X cash, and then gives Y cash to its workforce, the difference goes to the shareholders. If labor is what makes the biggest part of the expenses (and I guess the author is referring to such cases) that is almost a truism.

The problem is not outsourcing, it's thinking that you can outsource (or delegate) and important task and not pay attention to it's execution.

Good management breaks outsourced projects down into replaceable, manageable chunks. This is important for oversight and for competition. If you size an outsourced chunk of work correctly, then you should be able to increase competition for that contract. If you structure the contract correctly, your oversight should be able to identify and replace problemed contractors.

These guys are all smart enough to know that. It's just that they are also all smart to know that it is easier to avoid political responsibility for failures when it can be blamed on an outside contractor.

I completely agree with the thesis of the article; that outsourcing leads to inflexibility is performed by a staff that is not necessarily invested in the goals of the organisation. However, how does this apply to providing security for the Olympics, a once-off event that may not return to London for another 50 years? A city can't be expected to constantly maintain a police force capable of covering an event of this size.

I can name SpaceX as a successful example of outsourcing.

The real reason is probably more akin to "incentive matters".

I agree if you mean more than just a financial incentive. Job satisfaction at Space X must be extraordinary. G4S on the other hand is a company whose business plan is to sell 'security theatre'. They hire and train people for the minimum amount possible and resell that for the maximum amount possible, they don't have any incentive to care about their employees. They are just a HR department that processes people applying for and leaving jobs as efficiently as possible and a sales department that sells that time. No one is going to take pride in their work for a company like this. Maybe they made a mistake in assuming that there were an enough desperate unemployed people available to fill these positions?

I think that successful organisations, whether they be state run or private companies, depend massively on the character of the individuals that run them. I expect that Space X is successful because Elon Musk is a very rewarding person to work for. And for an example of a state run organisation like this see Frank Pick[1] who, as manager of London Transport, untangled the mess of privately constructed transport systems and made it into the poster child for public transport systems across the world. Pick's organisational values and culture outlasted him for many years, even today new transport projects in London have to be pretty high quality because they are measured against the legacy of all the great stuff the was designed and built in the 40's and 50's.

[1] http://en.wikipedia.org/wiki/Frank_Pick

SpaceX is an excellent example of outsourcing working well.

But G4S had the incentive of a £300 million contract, and their CEO gets £830K/year, and they still couldn't get it right.

I would argue that SpaceX's success is much more tied to the founder's vision than to any kind of monetary incentives from the government.

I personally suspect that the real mechanism is not private vs public, but rather organization structure and leadership style.

SpaceX is led by Elon Musk, 'nuff said.

SpaceX are well-known for not using subcontractors to make their parts etc. They do almost everything in-house.

Does spaceX even outsource their security guards?

Yeah, unlike policing and health, SpaceX's activities are not necessary. Don't get me wrong, what SpaceX is doing is fantastic, even inspirational, but not an essential service. So, if it goes wrong, even badly wrong, it really does not matter.

I don't see how "unnecessary" or "necessary" matters, as long as we're talking excellence.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact