Executive Belocracy: Current Options

This post on Best Of A Great Lot is a part of a series on the subject of designing a new form of governance. Each piece aims to stand alone, but fits together on the Table of Contents.

Previous: A Sketch of Belocracy, Evaluation as Feedback Cycle, Idea Generation and Sifting, The Belocrat, Policy Design. Next: Executive Belocracy: The Societal Benefit Organization

In previous sections we've gone over the legislative or rule-making aspects of belocracy. Now let's talk about the rest of governance: primarily the enforcement of rules and the running of societal infrastructure. We'll call this executive belocracy, including the wide variety of program creation, project management, enforcement, support and service offerings that are currently in the purview of a democratic executive branch, as well as a variety of things which are currently run by private entities. Legislative belocracy, because it relies on subject experts rather than bundled governance, can handle administrative rulemaking that’s currently devolved to agencies. We'll also see how belocracy handles the pervasive problem where for-profit society generates new infrastructure.

There are, broadly speaking, 3 kinds of organizations right now

One useful categorization of our current system breaks things into 3 types of organization: government agencies, for-profits and nonprofits. Government agencies exist because of (and often have their structure and functioning defined by) laws passed by the legislature. For example, the FDA is authorized and described by the Food, Drug and Cosmetics Act of 1938. For-profits are organizations which exist to earn their owners profit. Nonprofits are, broadly speaking, everything else, though in casual usage, the term is often conflated with 501c3 nonprofits, which receive a special status and tax benefits by persuading the IRS that their mission is charitable.¹ There also exists something of a new hybridization called a public benefit corporation which is sort of a more nonprofit-ish for-profit: an organization which can earn profit and also, according to the rules, pursue a mission benefitting the public, but this is almost a footnote because of how new, uncommon and untested it is.

Executive belocracy needs to be run by organizations. Neither enforcing rules and regulations nor delivering infrastructure is something that can generally be done by individuals alone, nor something that can be automated: humans, working together, must do it. Each piece of executive belocracy will have a specific mission we need it to pursue, and we want each organization to be as effective as possible at achieving its mission. Unfortunately, our current options for organizations all have flaws that need to be overcome if we’re to achieve this goal.

Agencies rely on good intentions since they lack competition

Executive agencies suffer from a variety of normal organizational problems, but they're also unique among these types of organization in that they experience little to no competitive pressure. Some agencies are well run and some are badly run. Many civil servants are people who genuinely want to make the world a better place and bring their best intentions to the job, as I argued before. But if an agency is badly run, sclerotic or ineffective, it's enormously difficult for anyone - citizens, employees, leaders, politicians - to effect real improvement. Dysfunctional organizations tend to stay dysfunctional. Reorgs, reforms and replacing leaders can occasionally help, but just like the stock market "can stay irrational longer than you can stay solvent", organizations can stay dysfunctional longer than you can stay alive.

Most agencies aren't even good or bad overall, they're ambiguous. This ambiguity offers a lot of room for incompetence. Some of their programs are great, many are less good, but how well any one of their programs could run isn't easily determinable. The way we determine this in the private market is whether the competition is winning. But there's no competition among agencies to compare against. Some really effective-seeming programs may have only just scratched the surface of what's possible. Other seemingly ineffective ones may be the best we can do given the universe we live in. We can sometimes point to agencies in other states or countries that are equivalent and run more effectively as a comparison, but agencies don’t have to justify themselves in these comparisons, and a more effective state agency isn’t allowed to take over for a less effective state or federal one. Bureaucrats can say, fairly or unfairly, that the different population and regulatory scheme makes makes fair comparisons impossible, and who can say differently? The status quo bias is strong.

For example, the GSA has built a centralized login service called login.gov so that each federal agency doesn't have to manage its own authentication and citizens don't have to make multiple accounts. They've spent hundreds of millions of dollars on this between 2020 and 2025, and they have a plan to spend additional hundreds of millions as they keep trying to add departments. Is that a reasonable amount? How would we know? Authentication can have some really tricky edge cases, especially if you want to make it both secure and accessible for all citizens. When faced with enormous numbers for the cost of a government program we often shrug and declare that it’s reasonable because government scale is just hard. But ID.me is a private competitor that provides login services to the government and they managed to get going with a lot less. They raised what was for them a huge round of funding in 2017 at $19m after they had gotten and started delivering on a contract with the GSA. While anyone who wants to claim that the GSA should have been able to get their system running for under $1m is probably an idiot, it's very possible that the GSA is spending twice as much or more as an effective government could have gotten it for. In the for-profit world, the only way we find out what the actual limits of possible are is when multiple teams are given real incentives for winning and compete.

It's easier to evaluate for-profit and nonprofit organizations simply because for any particular niche, there are usually several of them. There's still plenty of ambiguity - perhaps this nonprofit is better because it's only focused on a tiny part of the problem and that one is solving the harder bits; perhaps this for-profit makes a better product because they sell directly to consumers who have that rare thing we call taste, while that one sells to enterprises, who by necessity do not. But it's still much easier than with government agencies.

Even when we can't evaluate the borderline cases, we can evaluate the worst ones: when organizations do particularly badly in competitive ecosystems, they stop existing and new ones can take their place. Government agencies almost never face this pressure.

Nonprofits often focus more on donors than effectiveness

At first glance, nonprofits seem like a good fit for delivering the kind of societal benefit and infrastructure we're after. Most people assume that nonprofits are better at pursuing a mission than for-profit companies, and there are frequently nonprofits that are involved in mission-critical areas of society.

Nonprofits do sometimes have real competitors. But nonprofits pursuit of mission is all over the place, and competition within the nonprofit space doesn't provide a strong incentive for it. To boil it down, nonprofits make money when their donors like them, a process more built on marketing and feelings than cold hard numbers and effectiveness. The Executive Director's whims and ability to manage the politics of the board are often vastly more influential in terms of strategic direction than how effective that direction is, since it likely won’t dramatically change the donor pool. There's little truly external oversight. Donors mostly don't deeply investigate nonprofits for their mission alignment. Most donors spend more time thinking about the details of a major appliance purchase than they do which nonprofit to donate to.

The only place where effectiveness of nonprofits is consistently given serious research effort and paid attention to by donors is the effective altruism movement, which is a rounding error in terms of both nonprofits and donors.

In many nonprofits, Executive Directors pursue prestige and scale (more people, more money) for a variety of reasons. They can say with a straight face that more prestige and more donations will help them serve the mission better, while also knowing that they'll help them hand themselves higher salaries and enjoy more of the perks of high status. The forces pushing back on this are too weak to drive money to a better cause. Nonprofit hospitals would probably do better to put enormous executive compensations toward improving quality or funds for the indigent, since executive comp is mostly tied to hospital size, not outcomes. Popular, highly visible nonprofits that bring in lots of donations don’t stop fundraising. They sometimes regrant to other organizations, but they have also been known to waste their excess.

There are lots of ways that nonprofits do poorly. An extreme example is The Red Cross, which at one point spent half a billion dollars to build 6 houses in Haiti. Most nonprofits don't waste nearly that much money, but more because they rarely get the opportunity than because they're so well run that they couldn't. Then there are other bad behaviors: funnelling "profits" to the executives' personal political goals, claiming to be poor when they are not (e.g. Wikipedia's constant fundraising drives, even though they have so much extra money they have to find non-mission purposes to spend it on), spending a lot of money on bringing in more money (the March of Dimes has always been notable for their fundraising costs, but this chart claims that the Wounded Warrior project is the winner at this, spending an impressive 34% of their gains on fundraising), and overpaying executives.

[Customers] voted 1,452-30 to sack the board [of a rural electrical coop] that had paid itself more than three times the national average.

Nonprofits are hard to create

Woody Allen’s famous joke about life goes:

The food in this place is terrible.

I know, and such small portions!

Let’s parallel that with nonprofits. They’re so deeply flawed, and also we have too few of them.

Nonprofits are harder to build to scale than even notoriously difficult for-profits, which means that we get vastly less innovation in nonprofit-run societal infrastructure than we get from for-profit companies.

Successful novel organizations cost a lot of time and money to create, and the more innovative organizations tend to cost more. Every year people start thousands of investor-backed businesses, and some harder-to-quantify number that are self-funded. Many of these organizations cost millions of dollars before achieving significant success. In all of these cases, the founders (and investors) hope that they will earn back the money that they spend on the creation of the business. More than half of new businesses fail within the first 5 years. Of the most visible category of investor-backed businesses, VC-backed ones, 90% of them fail to pay a profit back to investors. Even so, enough of these organizations succeed to create a reinforcing cycle of new money that can be invested into new ventures. Startup founder turned angel investor turned VC is a popular career path for the successful technologist.

Much less money is risked on new nonprofit ventures because there's no reinforcing cycle. Founders have to self-fund or find grants or donations. Donors and founders do not have their self-interest connected to the outcome of the organization. The risk of failure is equivalent — new organizations are tough for consistent reasons — but the successful nonprofits can't make it easier for the next generation. The result is vastly fewer new successful nonprofits.

Grant funding is particularly difficult. Early stage organizations need a lot of flexibility to navigate the path to their success, and grants usually come with the expectation that you've figured the path of success out before you apply. Open-ended grants with few restrictions exist, but are rarer. Many grants come with very specific terms about how the money will be used and reporting and compliance requirements that require extra work to manage. Earmarked funds are much harder to work with than investor funding and add an additional constraint.

Many nonprofits start with extremely minimal funding and build on volunteer labor. Many fewer of them grow to be anywhere near as successful at "making a dent in the universe" (as Steve Jobs put it) as successful for-profits. The fact that so many people want to volunteer their time is a testament to our idealism, but the fact that we can't funnel that in more useful ways is a failure, culturally. We burn out a lot of idealism, get fewer nonprofits than we could, and cement the belief that nonprofits can and should pay less than for-profits. Many talented people leave the nonprofit field for higher salaries. Talent and resource shortages are a constant struggle for most nonprofits. For these and a wide variety of other reasons, nonprofit workers in general find their jobs more stressful even though they're more satisfied with their work.

And meanwhile, those successful for-profits that find their way to scale and create infrastructure do not have a golden path by which they can transition to become nonprofit, so even though many of them obviously should be, they aren’t.

We didn’t buy ID.me and spread it to other government agencies because that’s not something the US government can do. We started Login.gov at enormous expense instead. We didn’t buy TurboTax, we spent years building out a competitor called Direct File while Intuit, the owners of TurboTax were actively lobbying against it, and now we’re probably killing DirectFile because Intuit succeeded.

Altogether the nonprofit ecosystem as it is currently constituted isn’t what we should look to for how to run executive belocracy.

Public Benefit Corporations are new and uncertain

The public benefit corporation, the new form of organization I mentioned, is worth commenting on. It seems to have been created by people who were frustrated with how hard it is to create nonprofits combined with people who were frustrated with being told that their for-profits had to pursue profit at all costs. This parallel system is a legal framework inspired by the privately-run B Corporation certification, though with many fewer hoops to jump through to join. Officially, they are supposed to have legal protection for making choices that pursue their mission at the expense of profit, though this has not yet been much tested in the courts.

The law does not require them to prove that they’re pursuing their goal effectively. Shareholders can sue, and then courts can decide, but because of their newness and lack of case history, it’s really uncertain how those lawsuits will go. It’s possible that executives may be able to use the mission to deflect lawsuits about ineffective pursuit of profit, and use their good intentions to deflect lawsuits about ineffective pursuit of mission. We just don’t know. There’s nothing in the law requiring audits of mission or mission effectiveness, shareholders have limited access to information, board members may not be particularly independent, and executives even less so.

If owners, such as acquirers after an acquisition, decide that the organization should stop being a public benefit corporation, this is fairly easy to accomplish. In fact, making it easier to remove public benefit status and protecting directors against liability for failure to deliver on the public benefit are two "improvements" to the law that are credited with encouraging more of them:

An amendment to the law in 2020 has made this option more feasible and attractive for a growing number of companies. For example, directors now have greater protection if a company is accused of failing to deliver on its public benefit goals, and opting in or out of a PBC is now a matter of obtaining a majority of shareholder votes, lowered from 90%.

Despite or perhaps because of the flexibility, this option does sometimes appeal to potential nonprofit founders: they can start a 501c3 nonprofit to receive donations and grants, and a sister public benefit corp to raise money and earn profit. The benefit corp can contract with the 501c3, providing another revenue source and makes it easier, in theory, to build both organizations. But it's not at all clear that it's effective at causing these organizations to stay true to their mission in the long run.

For-profits pursue profit

Finally, we have traditional for-profits. For-profits are great at competition and generating profit: there’s an enormous ecosystem built up to fund and build these organizations and so we generally get multiple competitors for most niches. Setting a mission beyond just profit is an incredibly common boardroom exercise, perhaps because most employees don’t find themselves that motivated by squeezing out another percent of profit.

Over the long term, we've consistently seen the drive for profit corrode the mission even in supposed mission-driven for-profits. This is especially visible during the sale of businesses to larger rollups and competitors, when founders quit or when the for-profit manages to gain a monopoly (the latter being particularly common among companies that stumble onto a piece of societal infrastructure).

Decades of monopolization and resulting poorly managed infrastructure from internet service providers, electricity companies, privately-owned water distribution, and prisons are good examples, but in almost any area of infrastructure you'll find examples of bad actors misusing the position that delivering infrastructure puts them into. Technologists like to point to Google, who originally pledged to prioritize search clearly over ads and have slowly been walking that back because ads is where they earn their money.

Consider the challenge of accountability that privately operated monopolistic power utilities pose. There's been a recent trend in the western United States of suing power companies for the damages from wildfires. The theory goes that they should have been doing preventative maintenance, and because they didn't, their equipment sometimes acted as the spark for the destructive fires. The electric companies have admitted that this is sometimes the case, and so they've been ordered to pay billions of dollars in damages. But that money isn't going to come from the company executives or owners, it's just another cost of doing business passed on to ratepayers. Since the electric companies don't have true competition for the most part, consumers can't choose the less negligent companies, even if they knew enough to do so. Regulators instead attempt to force the utilities to behave in better ways, but it mostly seems not to work.

What for-profits are great at is competitive experimentation. I mentioned above that each year thousands of investor-backed for-profits are created. This is where we get a lot of our new ideas of what services citizens want, and where thousands of those ideas are proven unworkable or bad. Some small portion of those new ideas flower into enormous businesses, and some small portion of large businesses further develop into infrastructure. Electricity, clean water, all forms of communications and transport, banking: all of these started as privately offered services and became things that the entirety of our society relies on. Along the way, plenty of other highly lauded ideas (such as pneumatic tubes for mail delivery) failed.

In part 2, we'll look at belocracy's proposed alternative to all three and a half of these models.

If you find my work interesting, please share and subscribe. It helps tremendously.

1. Nonprofits being dedicated to serving a mission is a polite fiction that we maintain because it's frequently true. In fact, there's not a ton that prevents a nonprofit from doing work that doesn't quite (or at all) support their mission if the executives decide to do so and the board doesn't notice or doesn't object. This happens often enough that the nonprofit world has the term mission creep to describe it. It’s true that in extreme cases the IRS can even revoke 501c3 status if an organization behaves badly enough, though it's rare and usually in response to some procedural technicality. Here's one such case. But in general, it’s not true that the IRS provides any meaningful oversight of whether a nonprofit pursues its mission. This is entirely the responsibility of the board, and while the board is frequently required to include outsiders, there’s no strong incentive for board members to be particularly harsh critics of foolish nonprofits.