It's time to reinvent management. You can help.

Bureaucracy’s $3 Trillion Price Tag

by Gary Hamel on November 21, 2016


gary-hamel's picture

Bureaucracy’s $3 Trillion Price Tag

What’s the least efficient activity in your organization? What sort of work delivers the least value per dollar? We think it’s management and administration. The fault isn’t with any particular leader, but with the top-heavy, bureaucracy-infused management structures that predominate in most organizations. We believe that an excess of bureaucracy costs the U.S. economy more than $3 trillion in lost economic output, or about 17% of GDP. Here’s the data.

According to the U.S. Bureau of Labor Statistics, in 2014 there were 23.8 million managers, supervisors and administrators in the American workforce. (This figure does not include individuals in IT-related functions). That works out to one bureaucrat for every 4.7 employees. Overall, bureaucrats comprised 17.6% of the U.S. workforce and received nearly 30% of total compensation. (See Figure 1.)

Figure 1. Employment and compensation share of U.S. managers, supervisors and administrators.

The question is, how many of these 23.8 million overseers do we actually need? We can get an answer by looking at the management practices of a small but growing number of post-bureaucratic organizations. Their experience suggests it’s possible to run large, complex businesses with less than half the bureaucratic load found in typical companies.

Among the vanguard are Morning Star (the California-based tomato processor), W.L. Gore (a $3 billion high-tech company famous for its Gore-Tex® fabrics), Nucor (America’s most profitable steel maker), Svenska Handelsbanken (a Stockholm-based bank with more than 800 branches across Northern Europe and the UK), Sun Hydraulics (a class-leading manufacturer of hydraulic components), Valve Software (a pioneering developer of online games), and General Electric’s Durham, North Carolina jet engine plant.

The case of Svenska Handelsbanken is illustrative. Its return on equity has surpassed that of its European peers every year since 1971. In an organization of 12,000 associates, there are only three levels. Operating decisions are almost entirely decentralized. Each branch makes it own loan decisions, sets its own pricing on loans and de- posits, controls its own marketing budget, runs it own website (on a shared platform), and serves all customer segments, from individuals to multinationals, within its catchment area. Nearly all of these practices run counter to conventional banking wisdom, which holds that to be efficient a bank must consolidate operational activities and centralize decision-making on matters like pricing and lending. Though a rebel, Svenska Handelsbanken has consistently posted industry- beating cost-to-income and loan-loss ratios.

The average span of control in these and other vanguard organizations is more than double the US average. GE’s Durham plant, to take a dramatic example, employs more than 300 technicians and a single supervisor—the plant manager. Not surprisingly, the facility is more than twice as productive as its sister plants within GE’s Aviation division.

The experience of the vanguard suggests it should be possible to double the ratio of employees to managers and administrators from 4.7:1 to 10:1. Doing so would free up 12.5 million individuals for other, more productive work.

In addition, there would be indirect savings. Bureaucrats have a penchant for wrapping their colleagues in red tape. In his anti-bureaucracy screed, The Utopia of Rules, David Graeber bemoans the fact that “no population in the history of the world has spent ... so much time engaged in paperwork.” A myriad of studies have documented the time lost to low-value management processes—from budgeting to the performance review. On the basis of this evidence (further discussed here), it’s reasonable to assume that as much as 50% of all internal compliance activity is of questionable value. According to our estimates, this translates into a waste of 9 million worker-years.

In total, then, there are 21.4 million individuals in the US workforce—12.5 million managers and the equivalent of 9 million bureaucracy-hobbled subordinates, who, through no fault of their own, are creating little or no economic value. This means the US could achieve current levels of economic output with 15% fewer people in the the labor force. That would boost GDP per capita from $120,000 to $141,000.

The goal, of course, isn’t to put 21.4 million people out of work, but to redeploy them into wealth-creating activities. If every one of these individuals was contributing $141,000 to economic output each year, rather than adding nothing, U.S. GDP would grow by $3 trillion—and the figure could be higher. Managers and administrators tend to be better educated than the workforce at large, so we should expect them to deliver better than average output per capita once reassigned to more productive work. Critically, our figures don’t include the productivity benefits that would come from a newly-empowered workforce less entangled by red tape.

Three trillion dollars represents 17% of US GDP. If the bureaucratic burden on the U.S. economy was cut in half over the next ten years, productivity growth would increase by a compounded rate of 1.3% percent annum, essentially doubling the post-2007 productivity growth rate.

This productivity bonanza would be even larger outside the US. In 2014, the combined GDP of the 35 countries that comprise the OECD amounted to $49.7 trillion, of which the non-US share was $32 trillion. If bureaucracy is as ubiquitous in these economies as it is in the U.S., and there’s little reason to believe it’s not, there’s another $5.4 trillion to be saved by eliminating the bureaucratic tax—an amount that exceeds the value of the entire Japanese economy.

In recent years, productivity growth in the U.S. and other “advanced” economies has been anemic. Around the world, stagnating incomes and diminished economic hopes are feeding a growing appetite for protectionism and spawning divisive, us versus them, political forces.

While some hold out hope that robots, genomics and the internet of things will one day yield a productivity bonanza, we believe a concerted effort to reverse the rising tide of bureaucracy offers a more immediate and less speculative route to enhanced economic performance.

So what do you think?

  1. Should busting bureaucracy be a national economic priority?
  2. If so, what steps could policy-makers and CEOs take to accelerate progress?

To learn more about the $3 trillion price tag of excessive management, you can download our research paper here.

You need to register in order to submit a comment.

jim-mcgriff-jr's picture

1. Should busting bureaucracy be a national economic priority?

It starts where it starts, not where it is found. Sen. Richard Shelby told me in 1980 that congress passes a law and we do not know what the bureaucratic agency’s do with it. What I mean is that the Federal Government starts most the bureaucratic process when it passes a law. Sarbanes-Oxley cost our company, it was rumored, over 2m in the first year of implementation. Also, it fundamentally changed the record keeping for all our departments. There are substantial check list to make sure all processes complied with the law. I know that this year with the expansion of the coverage of title IX the University of Alabama went from one person to a three persons to manage its implementation. Now, look at the Affordable Care Act and the Americans with Disabilities Act… Compliance cost us and business millions of dollars. Of course the consumer pays for all of this in the price we pay for goods and services. We have to get federal and state regulations under control. This would make our products and services more affordable for us and the world.

2. If so, what steps could policy-makers and CEOs take to accelerate progress?

It starts at the top. Each department must make sure each process and employee adds value to product produced. In the later nineties, a senior vice president at our company told this story. When he took over his department at the corporate headquarters he called in each employee and asked a simple question. What do you do? Some employees could not tell them what they did. He also stopped delivery of the daily state newspapers, this caused such a problem he was asked by the CEO to rescind his decision. About this time our senior executives decided to review our employees. We went from over 11,000 employees in our company to a little more than 6,000. Most came from the corporate headquarters. We had a couple of floors at the headquarters building that were vacant. It was weird to see these floors after we had the purge.

As a front line manager I never asked for more employees. At one business office I asked for the removal of an Accounting Supervisor. He was one supervisor I did not need, since I had accounting experience. I started the last job I had with nine employees; when, I left that job I had seven. The key is to eliminate what is being done that does not add value to the customer or company. This frees up employees to focus on the primary functions of the department.

Liked by Thei Geurts
bev-johnson's picture

Lynda Gratton talks about the "hollowing out of the workforce" where all jobs that can be automated are being automated. This leaves menial jobs at the bottom and executives at the top with the jobs currently being undertaken by the middle class being outsourced to low paying countries or done by computers and artificial intelligence. In my State there are many highly qualified people with university loans working in checkouts and coffee shops. I am not clear about the "wealth creating activities" that senior executives and bureaucrats are to be employed into.

willy-a-sussland's picture

So, traditional organizations are authoritarian - bureaucratic - complacent. Of course, the cost of bureaucracy can be huge, and it is insightful to put a $ 3 trillion price tag on it. However, the cost of bureaucracy may be the least of the problems that bureaucratic organizations face. Times of seismic shifts in the business environment require agility and innovativeness that are stifled by bureaucracy among others.

As Prof. Jay Forrester pointed out at MIT, management is a system. Trying to fix one part of the system can at the best sub-optimise the results that can be obtained, but most likely will introduce new dysfunctions. So, a holistic approach is needed, a holistic approach to management innovation, which among other will eliminate unnecessary activities, including bloated bureaucracy, The problem is that management cannot innovate itself. People can innovate their management, but not the people with the mind-set that created that management system in the first place. So, we have to create innovative people.

I have addressed those issues in my "The Innovative Enterprise" Create Space 2014, which has been translated and published in China in 2016, and Routledge will publish in March 2017 my "The Platform of Agile Management and the Program to Install It.", which does not start with the problems enterprises may be carrying forward, but with the needs they have to meet going forward. Among those needs, alertness-adaptability-agility are a must. Addressing those needs leads to build an organization that will neither be authoritarian or bureaucratic or complacent.

Hockey star Wayne Gretzky said " I skate to where the puck is going to be, not where it is now". I submit that organizational design should start with understanding tomorrow's needs not today's problems.

cathy-downes's picture

While I am all for dealing with bureaucracy that does not generate benefit and which sucks resources (who wouldn't be), I am deeply concerned about the bandering about financial numbers in this piece. I can see how you can add up your $3trillion hyperbole price-tag. However, there are critical assumptions buried in them to get to this number that sounds so on the surface attractive. Yes, who wouldn't want to save $3trillion for the US economy. However, and there are some big howevers here.

Having been on the receiving end of many of these types of articles with government leaders being directed to "cut by 50%", etc., the devil in is the detail. This is not an excuse. It is not a cop-out. It is a reality. And reality wins.

Presumption 1: that in 2014 there were 23.8 million managers, supervisors and administrators in the American workforce according to Labor Department categories - assumption that you can just get rid of these managers, or substantially reduce them. Labor Department categorization does not incorporate every task that the 23.8 million folks do. Presumption - they add no value. When you start unraveling this string you find that in the informal side, there are critical roles that many (obviously not all) of these people play in keeping their organizations functioning at least at the level that they are functioning.

If you undertake any meaningful social network analysis of the roles and tasks of people in an organization that are undertaken that don't register on the "Administrators, Managers and Supervisors" job description. This is where you truly start to get an understanding of what people do. I have been part of so many "great launch" change management projects which involved "clearing out bureaucracy", and now of course, we have "drain the swamp". 100% of them have failed miserably to go beyond the org chart in terms of cutting only to find that the left-over organization is missing some critical people that...well...yes...actually made things work in spite, and despite the old industrial organizational structures, which were just too determinedly bound in legal concrete to change. So six months after the glorious cutting effort, the organization is actually worse off.

Cuts usually involve the lowest ranks and rungs on the ladder, and those that have not made a case for their indispensability, many of which are the increasingly vulnerable "supervisors, managers and administrators - Public Enemy #1. The true "internetworkers" (a la Katzenbach) tend to be poor in making their case for indispensability, and their job descriptions rarely reflect the true interconnectivity role they serve in organizations. As a result when they get fired, the organization is internally disconnected and fails, while everyone remaining tries to work out who knows what, where it is stored, how to access it etc.

Another assumption that is presumption is that there are no legal constraints that would limit, prescribe and quite bluntly stop the sorts of cuts in people that this article proposed (regardless of any fly-away line to redeployment in more creative and productive enterprises.

Particularly in government, (Federal, State and Local) there is a ream of regulations, many of which contradict other regulations. This is NOT an appeal for cutting regulations. It is a recognition that the way the US Congress enacts legislation is in many respect illogical and irrational. Political choices frequently outweigh reasoned policy and most legislation ends up being crafted to deal with circumstances where previous legislation has been circumvented by creative people. If you are relying on Congress to improve the way it functions, will wait a very long time. So just going ahead and saying "cut out supervisors, managers and administrators" is too simplistic.

Another assumption is that the issue of second wave process automation, likely to affect a significant amount of work tasks of supervisors, administrators, and managers even in the mainstream Service sectors of the US economy over the next decade, and more and more artificial intelligence apps reduce the requirement for humans in these sectors. Again, we are not putting any resources aside to retrain.

Another assumption - unexplored - that people are retrainable to be re-employed in other "productive" sectors. This is a huge mountain to climb - the presumption that people in their forties and fifties (mostly when it comes to managers, administrators and supervisors) can suddenly start from scratch to learn new knowledge and skill sets without any scaffolding from their existing knowledge and skill sets. This is the false presumption that was applied for manufacturing workers who have had their jobs replaced by robots as much as any outsourcing to China and Mexico). Taking a car plant worker and changing them into a financial service sector worker....well, now we have the prospect of taking a financial service sector worker and making them into the creative economy? We have an educational system that is designed primarily around quantitative, linear, reasoning and thinking, and we need creative, parallel thinkers etc. If we are not educating our young with these different balance of skills, but now being driven to focus intently on Science Technology, Engineering and MATH, seen as the Industrial Age competitive edge - all of which are "left" brain thinking habits (which is a fallacy in itself, as the competitive edge is not in Industrial Age, but in Digital Age creativity), then how are we supposed to "retrain" these people to be more "right brain" skilled (yes, I know, I am only using the left and right brain as metaphors, not as neuroscience).

Another assumption is scalability. I too have read about the great examples of Nucor, etc. mentioned in this piece. However, while in comparison with most small businesses, such examples may seem large, they are nothing compared to something like the US Department of Defense. There is a constant struggle the larger the enterprise to force standardization of practice to ensure reliability, consistency, predictability. These are great goals of the industrial age which itself depends for much of its productivity growth on economies of scale.

In a 21st century Information/Digital Age (call it what you will), economies of scale are not what generates productivity gains. Therefore efforts at standardization simply grind down innovation, agility and adaptability. When you seek those from a large organization, the best practices of much smaller organizations do not scale!

I could go on, and my apologies for getting on my high horse, but I think MIX is a great place for working these issues, and I am all in to make a contribution, but if ideas are to be useful (and therefore likely to be taken up), they need to start from a realistic and evaluative basis, rather than a "$3trillion price tag of excessive management" tweet-level statement.

Liked by Thei Geurts
alfredo-bregni's picture

I don't know US figures, and surely I don't want to get into nitty gritty maths.
What I'm sure about is that a serious, properly sponsored redesign, whether against bureaucracy, overhead, just costs, or anything else -- **when one redesigns from scratch he/she doesn't really cares...** -- lands anywhere from 20 to 40% cost reduction, with possible significant improvements in service and/or value delivered.
Whether that is attainable or not in a public environment is a sheer political problem; nothing else. Surely not an estimate issue. Just whether you will actually roll up your sleeves to implement, or not.
Would it be me, and had I to fire somebody, in a EU environment I'd just protect (excess) people, but not all the costs they generate. In a US environment, I don't know: you decide.

onno-geveke's picture

Interesting perspective and attractive potential future. Even more so, because killing bureaucracy will also open up space for employees to arrive at work being fully themselves. Finding more energy in what they love to do most, thus becoming more productive and (more importantly) healthier and happier. Imagine what additional savings and revenues that might create in healthcare etc.

Should it be a national priority? Probably international, linking it into the UN Goals for Sustainable Development.
First step might be to organise a organisational system wide dialogue around 'what would our organisation be when we are vibrant and fully productive with a minimum of bureaucracy'?

alfredo-bregni's picture

Companies / organizations can be viewed as a collection of functions, processes and team-based projects / project implementations:
- Functions hold [past] knowledge;
- Processes deliver [present] value;
- Projects design [future] change, which project implementations bring to present time, by making it happen.

Processes, projects and project implementations involve functional resources, who increase their knowledge by learning from all of them:
- In processes, by doing;
- In projects and project implementations, from teaming (and doing).

Process improvement / (re)design should be the primary project focus, possibly in a creative / innovative / learning-oriented fashion.
Whereupon comes the implementation phase.

Bureaucracy can be reduced:
- In functions, to sheer human resource management;
- In processes, projects and project implementations, in a massive way.

Projects and project implementations can benefit from a learning-oriented teaming approach, capable of:[1]
- Boosting results;
- Maximizing results-effort and results-time ratios;
- Limiting managerial needs / bureaucracy to a minimum, thanks to project team members' self-management (this can be naturally extended to project implementation by involving project team members, thereby granting key success factors like purpose, cohesion and discipline at no cost).

The key link between process improvement / (re)design and processes / process management can a well crafted ICT support, capable of putting (re)design / improvements in a form suitable to be understood both from the humans and machines, thereby:[2]
- Simplifying both the initial design and the succeeding improvement / redesign efforts;
- Completely eliminating the ICT implementation phase, i.e. becoming immediately executable in a distributed environment;
- Getting rid of the traditional ICT implementation drawbacks: costs, risks, delays, incompleteness and rigidities.

Apart from people's knowledge, competences and teamwork, to be managed in a more traditional (but heavily learning-oriented) fashion, all the process and process (re)design workings can be heavily streamlined, with tons of bureaucracy slashed away.


Author's CV: