Why Government is Not a Business

Simon Wardley, the creator of the always useful “Wardley Maps” has a quip he shared a few years ago about the speed that the IT departments of large enterprises react to technological change.

The Enterprise IT Adoption Cycle. Simon Wardley, CC 3.0 BY-SA

While funny, the humor is pretty close to how we all see enterprise computing. Many would argue that government is the same. It is a common litany that government is slow and unresponsive. We often hear comments about how government workers are lazy, or that governments should be run like a business. In this article, I will use a pace layer model to push back against this argument, and then offer a more nuanced point of view.

Pace Layers

In the late 1990s, Stewart Brand postulated that what we now call complex adaptive systems can stratify into temporal layers that move at different speeds. The higher layers in the model move at a faster pace than those below. The layers coexist, but there is friction at the interfaces between each layer. The friction causes conflict, change, innovation and development.

Great fleas have little fleas upon their backs to bite ’em, And little fleas have lesser fleas, and so ad infinitum. And the great fleas themselves, in turn, have greater fleas to go on; While these again have greater still, and greater still, and so on. [Augustus De Morgan. A Budget of Paradoxes]

The image below is the reference layering model from Brand’s work. I imagine that with some thought, you can imagine smaller systems and their pace layers. The smaller systems are a part of the model below, giving us layers within layers.

Brand’s canonical pace layer example. Image by Rosenfeld Medi CC BY 2.0

As we can see, commerce moves faster than government (which normally resides in the governance layer). If we think about the dynamics of markets, then this assertion makes sense. Companies learn how to exploit a new resource (or an old resource in a new way), and markets quickly form. Governments neither understand nor care about the new market until societal risks need to be mitigated. At that point, laws will be passed and regulations enacted. This constrains the market and, if done correctly, reduces the impact of the risk.

Shearing Strains

Looking around at the world as it is today, we can see some shear points between the layers. I am going to mention three here and then get back to the point.

Culture Wars and the Impact on Higher Layers

If we look at the Culture layer and think about sub-layers within it, we can imagine fast-moving popular culture, the slower moving progressive layers below it, and the even slower conservative and religion layers under that. As popular culture became purified by the commercial pressures of broadcast and social media, the pace of change in popular culture increased. We can safely say that the faster progressive layers cause friction with the slower conservative layers below (think about gay marriage.) The shearing effects can partially be seen as political polarization. This polarization spills into the governance layer above with effects like gridlock in the US Senate, the Trump Presidency, Brexit.

COVID-19

When the lower layers move, it is often discontinuous and causes massive changes in the layers above.

Brand’s canonical pace layer with COVID-19 additions. Image by Rosenfeld Media CC BY 2.0, modified by Robert Hart

The emergence of the new coronavirus caused governments to respond quickly. They enacted some drastic measures, like social distancing, restrictions on commerce, mask mandates. These measures caused a cultural backlash with, for example, the rise of anti-maskers. They also caused supply chain stoppages which in turn created shortages of goods.

These are not the only effects. Think about the vaccine programs and more.

Web 2.0 and Cloud Computing

In the early 2000s, the build-out of the Web 2.0 infrastructure began, enabling a richer e-commerce experience. The maturing of cloud computing formed a catalyst for ubiquitous internet commerce, social media, and the ascendence of massive internet empires.

Governments have not regulated much of internet commerce yet. They have stuck to what they understand, like regulating the wires. It is becoming clear that government needs to decide how to regulate commerce, social media, and rethink intellectual property. When this happens, it will have ripple effects through both the infrastructure and commerce layers.

The Speed of Government

As we have explored, government exists at a lower, slower temporal layer than fashion, commerce, and infrastructure. This gives it the ability to remember what the higher layers learn, and the power to regulate the higher layers.

Commercial services last as long as they are profitable. In the accelerating world of modern capitalism, the length of time you can be profitable is shorter by the decade. Customers do not rely on a commercial offering to be available forever.

On the other hand, government programs must be relied on for generations. Whether it be a levee or Social Security, we expect them to be there and functioning for our lifetime. This long-term dynamic is at odds with the quarterly and annual return focus of commerce. This is why a government should not be run like a business. You cannot move fast ad break things with Medicaid. Lives are at stake!

A Pace Layer Model of Government

Let’s keep looking because there is more to see.

The workings of government can also be separated into pace layers. This is my interpretation of how they are.

The pace layers of government. Robert Hart

At the bottom are societal risks, the thing that a government responds to. The response is legislation, which can be in statute for generations. From the legislation, federal code or state administrative rule is created. For active programs, federal code/administrative rules can change on an annual basis. Cases are individual interactions with private entities who are availing themselves of the program. This could be an economic assistance application, an alleged rights violation, an inspection of a regulated business process, or any of the myriad things government does. Programs may handle hundreds to hundreds of thousands of cases per year. A case is the application of administrative rules to an individual situation and each situation is different. This need to be flexible causes friction with federal code/administrative rules. Administrative rules change in response to changing legislation, changing interpretations of legislation, and changes of practice when handling cases. The interpretation of legislation and hence the spirit of the administrative rules are often driven by the politics of the elected and politically appointed officials. Ultimately, the admin rule must adhere to legislation.

Could it be that when folks talk about running government like a business, they are thinking about the fast-moving cases layer? It is certainly true that the more cases completed, the more the societal risk is mitigated. Working to a fixed budget means that the more efficient you are, the more cases you complete. Therefore, some process improvement techniques could make a difference to the number of cases completed. That does not necessarily mean the methods of tracking revenue, return on investment, and strategic planning that the private sector uses are either appropriate or effective. Why? Because the administrative rule and budgeting processes will subvert them.

So, hopefully, I have established that government is not like business and does not operate like one. It does not operate at the speed of commerce and probably never will. However, there are things that government can and should do to improve the effectiveness of its programs.

The Goal of Government

In Goldratt’s classic The Goal, there is a scene where the hero, Alex Rogo, is sitting in his car on the hill overlooking his factory. He is eating pizza and drinking beer while thinking about the question his mentor, Jonah, posed. What is the goal of your company? Alex considers the question and eventually concludes that the goal is to make money.

One of the features of the chapter is that his internal dialog forms the initial thinking of what eventually becomes a quantitative and qualitative framework for measuring and improving the performance of the factory. The frameworks are Theory of Constraints and Throughput Accounting.s

Over the last generation, many have tried to lift and shift management practices from manufacturing and corporate contexts to government. Those who have worked in government for some time have seen ISO9000, TQM, Business Process Reengineering, Management by Objectives, Lean, Six Sigma, ITIL, Objectives & Key Results, and more come and go. They were tried and found lacking. I would argue that these practices fail for two reasons. Managers apply them beyond the bounds of their applicability, and they ignore the fundamental contexts of government.

If we are going to improve how we manage the public sector, we need to first explore the purpose of government. If it helps, get some beer and pizza.

What is Government?

First, we need to understand what we are considering. What is government? It is multi-level. We have international, national, regional, provincial, state, county, and city governments. It does many things. It runs elections, legislates, governs, educates, regulates, delivers services. It uses its immense power to levy taxes, send people to war, prosecute and incarcerate. That is a lot of disparate responsibilities, but there must be a common thread.

We the People of the United States, in order to form a more perfect Union, establish Justice, insure domestic Tranquility, provide for the common defence, promote the general Welfare, and secure the Blessings of Liberty to ourselves and our Posterity, do ordain and establish this Constitution for the United States of America. [Preamble to US Constitution]

The United States Constitution lays out a list of reasons the U.S. Government was formed. They too seem like a broad set of aspirations, but there is one thing that is common in both lists.

The Goal is Managing Risk

He’d [Max Stier] explain that the federal government provided services that the private sector couldn’t or wouldn’t: medical care for veterans, air traffic control, national highways, food safety guidelines. He’d explain that the federal government was an engine of opportunity: millions of American children, for instance, would have found it even harder than they did to make the most of their lives without the basic nutrition supplied by the federal government. When all else failed, he’d explain the many places the U.S. government stood between Americans and the things that might kill them. “The basic role of government is to keep us safe,” he’d say. [The Fifth Risk, Michael Lewis]

If the basic role of government is to keep us safe, the obvious question is safe from what?

There are, of course, the primary dangers early governments addressed: attack by other nations, anarchy and mob rule, crime, and violent behavior. Over time, the responsibilities have grown. Consider everything government does and you start to see it is to protect us from something:

• Social Security and economic assistance programs protect us from poverty.
• Medicaid, Medicare, public health, veterans health services, water and waste management protect us from disease.
• Dams and levees protect communities from flooding.
• Road building and maintenance, driver and vehicle licensing, and traffic police protect us from the harmful downsides of traveling in fast-moving metal boxes on wheels.
• Child protection services, childcare licensing, and daycare regulations protect children from harm.
• Agriculture, food, and restaurant regulations protect us from being poisoned.
• Market, labor, and housing regulation protect us from extortion.

The types of risks that government manage or mitigate are societal. They range from existential, like nuclear threats, to infrastructure, to personal risks that affect large or vulnerable groups.

We can now see that what government does is managing or mitigating societal risk. Politicians campaign and debate what risks government should manage and by how much. Legislation lays out the risk mitigation policy goals, policies and rules. Government agencies and commissions do the risk mitigation.

The Problems With This Argument

While this argument seems compelling, there are a couple of issues with it.

Firstly, libraries and education are hard to justify as risk mitigation. Justifying roads as a safety issue alone is hardly complete or satisfying. The highway system is also a platform for commerce, which could not be called a risk. Roads, libraries and education are opportunities that benefits society.

Part of the intention of this exercise is to build an epistemological framework around which we can create strategies and measurements. The current lingua franca in government is money, which is problematic because it compels managers and policy-makers to cut costs. Such tactics are both restrictive and demoralizing.

Mitigating and managing societal risk is more positive and aspirational. The ability to justify investment is more apparent. However, it is still a justification to reduce something rather than grow it.

Risk is difficult to measure and hard to normalize across the many disciplines. Some areas already do this, like modeling the cost to life and property if a levee were to break. Other programs have never considered how to measure the risk they manage. If we could identify, quantify, and normalize the performance and potential of risk mitigation, then we have a new framework for managing government.

Shared Service Centers: This Emperor Has No Clothes

Whether it be one-stop-shop call centers or unified intake through a shiny website, without dealing with the core issues of throughput and quality, all your customer experience efforts make no difference.

Those of you who have spent a little time reading John Seddon will know that he likes to rail against call-centers. He explains how they did not achieve what they were designed to do, and that most of the increase in usage by customers is “failure demand”, customers having to call again because they were not satisfactorily served the first time. What Seddon is weak on is why the service centers do not accomplish their goal. In this article, I will give my explanation using the Theory of Constraints.

An hour lost at a bottleneck is an hour out of the entire system. I say an hour saved at a non-bottleneck is worthless. Bottlenecks govern both throughput and inventory

Eliyahu M. Goldratt

Setting the Scene

Let’s imagine a scenario where we have three separate licensing processes. They all follow the same major steps, but the details differ. The diagram below shows the three workflows along with how many license-requests each step can handle per unit time.

How many licenses can each process produce? Intuitively, we know that itis the same as the slowest part of the chain. The next diagram shows the slowest step or the bottleneck, in red and the speed of each process.

Adding a Unified Intake

In the name of process improvement, let’s add a nice unified intake process in front of these licensing processes. It could be a call center, or a website with chatbots and pretty forms to fill out. It can be the best planned, designed and executed project ever. The diagram below shows how it might work and the capacity of this step. What effect will this new intake process have on the number of licenses created?

The Impact

We added a new intake step that can handle 40 license requests per unit time. The capacity is higher than the combined capacities of each subsequent step. It must speed up the system, right?

Well no! The speed of each process is set by the slowest step. We have done nothing to improve the slowest part, so we have not improved the whole system.

What did the new intake improve? We can see that it did not improve throughput or the quality of the completed product. It made it simpler and more pleasant to request the license. You could argue that we have improved the customer experience and in a simple sense that is true. However, the request is the first part and feeling good about that is fleeting. If confronted with having to wait a long time, or a poor quality product at the end, then customers are going to dismiss the intake improvements.

Fundamentals First

When you plan to improve a process or system of work, you need to be clear headed about what you are trying to accomplish. You also need to take into account that the customer will include other aspects that you are not considering. You will find that without taking throughput and quality into account, your improvements may not be recieved as well as you would like.

Remote Work, Maslow, and Culture Shock

When the COVID-19 pandemic began to spread across the globe, many of us were sent home by our employers. A considerable proportion of us will remain there for at least the next year. Moving home took a lot of effort, but I do not think the transition is over. In this article, I will explain why.

Before we get to the main argument, let’s first think about a couple of theories that I will tie together to make my point.

Maslow’s Hierarchy pf Needs

Maslow’s Hierarchy of Needs by Androidmarsexpress – Own work, CC BY-SA 4.0, https://commons.wikimedia.org/w/index.php?curid=93026655

I’m sure all of you who completed a psychology class at school remember Maslow’s Hierarchy of Needs. As a refresher, the pyramid groups human needs into layers. You start at the bottom of the pyramid. To climb to the next level, you must first satisfy all the needs below. So, to feel a sense of accomplishment with your life, you are not hungry, not worried about money, and have friends who love and appreciate you. The pyramid is a simplification, but good enough for my argument.

Culture Shock

When you relocate to a new country, the psychological journey from novel to normal is called culture shock. The duration and severity are contingent on the person and the differences between the two cultures.

Culture shock has four stages. First is the honeymoon phase, where everything you see is fresh and exciting. You learn how to survive in your new world. Simple activities, like settling into your home, learning where to buy food, exploring the locale, and starting a job is exciting. The challenges are frustrating but rewarding.

The second stage is the negotiation phase. You are learning how to fit in and trying to reconcile the differences between the culture you left and the one you joined. You spend a lot of time comparing and telling anyone who will listen about the differences. You have to sit with the differences for a while to finally see how aspects of life may be different, but they are not wrong.

Adjustment is the third stage. You apply what you learned in the previous phase, and start to let go of your old cultural attachments. You let new routines and relationships emerge, and your new world starts to feel part of you.

The last stage is acceptance. Now you begin to weave yourself into the fabric of society. You can function well in the new culture, and you are now looking forward to getting on with life.

The culture shock transition follows an emotional arc. The initial positive stage of excitement and accomplishment progresses to a longer time of despondence and frustration. When you reach acceptance, the new culture feels normal, and you are satisfied with your new situation.

Culture Shock and Maslow

After experiencing culture shock a couple of times, I see a link with Maslow’s Hierarchy of Needs. At first, you are concerned with satisfying physical needs, like food, housing, and work. Because the pay-offs are immediate and unequivocal, they lead to a sense of euphoria. As you progress to making friends, fitting in, and making a difference, these tougher challenges can leave you discouraged and frustrated. By the time you reach acceptance, your feelings are more neutral.

What About Working from Home?

In the last six months, we have moved from our bustling offices to a spare corner of our homes. Culture shock is not limited to moving to a new country and transitioning to remote work is certainly a change in culture.

Initially, employers concentrated on the physical aspects of pf remote work. They scrambled to supply the technology we need and gave us space to form new work habits. We have met these new challenges and overcome them. We feel good. Remote work isn’t so bad!

Now, we must rebuild those relationships we rely on to be productive. We miss the water cooler gossip, the subtle signaling that tells us if we are performing well, and all the other social cues of employment. Next is the challenge of maintaining that sense of belonging, knowing you are contributing to the mission, professional development, and career progression. The methods of the collocated workplace may not work in the virtual one. If you have not considered these issues and started testing ideas on how to overcome them, then your staff may experience a long deep trough when the remote work honeymoon ends.