How to reduce bureaucracy, the mafia and other forms of free loading

Most people in modern times consider bureaucracy to be a necessary evil in the administration of their complex societies. Complex tax returns, business environmental regulations, vehicle and driver licenses, business licenses, IRS non-profit approval paperwork, food selling regulations, DEI workshops, stock trading regulations, and regulations concerning even the administration of surveys that might trigger the participants (this is a recent incident for CARE that inspired this essay), require either personal frustration and wasted time, or the hiring of professional bureaucrats (CPAs, attorneys, and Institutional Review Board members), an example of niche construction.

Bureaucracy was present in ancient empires, including Babylon, China, India, Egypt, Greece, Israel, and Rome. As far as I can tell, complaining about bureaucracy is as ancient as bureaucracy itself. In order to deal with his bureaucratic frustration, Kafka wrote several surreal books exposing the absurdities of his eastern European bureaucracy. Turchin is both right and wrong about bureaucracies: he is right that in all known examples they are necessary to maintain cooperation and stability, once the number of people per tribe exceeds about 2000, and that higher unifying levels (cities and states) are necessary to prevent inter-tribal warfare among these tribes (of less than 2000) . But he is wrong that the downsides are inevitable.

Iuval’s Substack is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.

To understand why the downsides are not inevitable, we need to understand the downsides, and then diagnose their root causes. One downside is that the bureaucrats are often more self-serving than serving their constituents. Unlike Mafiosos, the regulations they are maintaining and enforcing might have started out with public- serving intentions by law-makers, such as facilitating trade, obtaining funds for public goods projects, and offering protection. But inevitably, these motives are forgotten, except to be used as excuses for giving the bureaucrats more money and power, allowing them to insinuate themselves as middlemen in most transactions and extracting a cut, becoming more and more like Mafiosos. Another problem is that often the bureaucratic procedures are inefficient.

Perhaps upstream of both of these downsides is that the feedbacks to the bureaucrats from their constituents are too weak and of the wrong kind. Information mostly goes from the top of the hierarchy to the bottom, while resources go mostly the other direction. The bureaucrat usually is not in contact with the people she is supposed to serve (there are too many of them), and does not have any reason to care about them, except superficially due to common humanity. The main incentive is not to improve efficiency of procedures or the lives of constituents, but to maintain one’s job and salary (or to climb up an institutional ladder to increase one’s salary). In other words, the government bureaucrat is naturally selected to free ride, for 2 reasons. First, standard multi-level selection dogma would say that free riding is selected because there is no higher level competition between governments (except during times of war, or when migration is easy) to punish free riding and increase cooperation within a country. But I don’t think this is quite correct. Cooperation at lower levels is not started by competition at higher levels, though better cooperators may fare better in competition, once cooperation gets started by other means. Nor is it started just by random mutations or their sociological equivalents of random behaviors and ideas. The incentive for cooperative behaviors has to be synergy between parts (in this case people and their institutions), good feedbacks to reinforce the synergy.

There is a synergy for sure: the bureaucrat gets money from constituents, and constituents get a service from bureaucrats. But the feedbacks are weak and indirect between a bureaucrat and the people she is supposed to serve. The bureaucrat does not get more money (or more time) by getting her constituents better services, and the constituents do not get a better service by paying more money, because there are too many intermediates that weaken the feedback and make it noisy. It is here that higher level competition comes in, not to create the synergy in the first place, but to improve the feedbacks.

Corporate bureaucrats fare better (compared to government bureaucrats), because corporations are competing with other corporations for the satisfaction of their constituents (aka customers), whereas governments are not usually competing with other governments (exceptions noted above). Corporate bureaucracies have less waste and less incentive to free ride. I’m not advocating for unregulated global markets and unbridled competition, because as I’ve discussed in previous posts, this leads to all kinds of catastrophes of the kind dubbed “chaos attractor” by Daniel Schmactenberger. Nor am I proposing the kind of neo-liberal government regulation of capitalism that leads to the bureaucratic problems above (a variant on Schmactenberger’s totalitarian attractor). What else can be done?

Going further down to the root, we see that we need hierarchical levels to maintain order, increase cooperation/synergy feedbacks, reduce internal harmful competition and free riding while increasing beneficial competition, but for this to happen, the topology of these networks needs to be of a different nature than existing topologies of either government or corporate bureaucracies. The quality (amount of resources flowing) of synergizing connections at each level needs to be stronger than at higher levels, flow in both directions between nodes, and it needs to be small enough to manage for the cognitive capacity of humans, and such that the costs of maintenance are also smaller than the synergistic benefits.

3rd attractor networks have all these features and ultimately will lead to people (including bureaucrats) genuinely caring about the people they are mostly interacting with. Are we ready to try them? We don’t need to try them at a global level yet. We can start small with ourselves, our families and our communities, and then go on to higher levels. This is not an easy feat within the current networks, and hence we must also get some breathing space from them to accomplish this feat.

Let’s get more concrete. First, there are bureaucracies that are necessary in the current network topology, that would just disappear in third attractor networks, or even chaotic networks like the internet. The Institutional Review Board for example, protects people very little when it comes to social science experiments. Especially with voluntary surveys that have trigger warnings (telling people if they are triggered they can leave the survey any time). It would have also not allowed some seminal psychology experiments like Milgram and others in attachment theory testing (thankfully didn’t exist back then). Money flows to the IRB members and the other bureaucrats who deem them necessary (legislators and journal editors) but very little protection flows back to anyone. It creates a barrier for research that only big academic fish have deep enough pockets to overcome, and selects against the little non-academic fish who might do good research without it. It also ties into our legal bureaucracy where suing runs rampant, with the IRB supposedly protecting researchers against lawsuits from overprotected Gen-Zers who can claim they were traumatized by a survey. But this sort of thing thrives when there are people who don’t have a real productive contribution to make to their families and communities. They have no choice but to parasitize other people for their very livelihood. In a third attractor network everyone is gainfully employed primarily in their families and in their communities, providing basic, needed goods and services, and only once these are produced can people take on other tasks, like enforcing norms for a watershed, doing science or producing art.

A second example where bureaucracies would still persist, but be better in 3rd attractor networks is the legal system. We need stronger family norms, with families embedded in small communities that also have their norms, preferably written down. Both families and communities also need enforcement mechanisms when people break the norms. The people in a family need to need each other, as in goods and services, which could be emotional support. Right now it’s mostly outsourced and the kids mostly get food and shelter from their parents, with almost nothing back for parents.  Families need to need each other in a community in a similar way, instead of just being neighbors. Much crime would disappear if at the family level people had direct responsibilities and benefits towards each other, instead of towards employers. A judicial council could abjudicate any norm violations at the community level. When people travel between communities, there is the potential for breaking another community’s norms and this would need yet another higher level of a judicial council of communities in a watershed. A higher level could be abjudicating different watersheds. And so on. In such a network, there would be a good chance that one of the judicial bureaucrats knew somebody close to someone (with only a few intermediates) who is accused of a crime, than in our chaotic network, or China’s totalitarian network.

A third example is ecological bureaucracy. In a third attractor network, each entity, from a person to a nation, feels the direct impacts of its actions. There are no externalities, as each entity is embedded and inside another like Matryoshka dolls. All the agencies that have sprung up to regulate corporations and make them be less harmful to nature (and ultimately to people) have not helped much (and might have made things worse because they take more of people’s time and resources) because the feedbacks between people and their actions are too abstract, and the selection for externalization is stronger than the selection against it by punishment. If I poison my neighbor’s air, he will let me know and if we can’t come to an agreement, there is a community council to facilitate. If my town pollutes the waters of those downstream from our river, they will let us know and if our towns can’t come to agreement, there is a federation of towns that will facilitate. And so on.