What Brook's law can teach us about effective government
$$
n*(n-1)/2
$$
Starting an article about government efficiency with a mathematical formula is, at best, endangerment to the clinically sane.
Those of you familiar with a good amount of trivia, eponymous laws or software might have stumbled across a thing called Brook's law. For those who didn't, congratulations for passing the sort of dork-equivalent of the Turing-test. Let's enlighten ourselves with a good old fashioned definition first.
Brook's law states that adding manpower to a late software project will make it later.
There are two key observations to be made. The first and obvious is that quite often the opposite of "well done" is "well meant".
The second and more crucial one is that adding just one thing can worsen a situation significantly, as it introduces a certain kind of friction into a system.
The duality of friction fascinates me. Positive friction is traction - it lets you move, accelerate, brake when needed. Negative friction is burning rubber without moving an inch. Brook's Law helps us see which kind we're creating.
Sticking to the second observation, the inverse is also true. Removing just one thing can have an almost exponentially (polynomial, if you're tedious about it) positive effect on the desired outcome of a situation.
Which brings me to a suggestion usually feared by politicians and voters alike: that subtraction equals loss and has to be tightly prevented.
I get it. Losing something or having less than before hurts – usually. Though I can think of many things that, without them, the quality of my life is improved significantly.
Think of the annoyingly delayed construction site which only finds some form of predictable punctuality in its ability to turn employment into noise while making no apparent progress whatsoever.
Or becoming slightly out of breath the older you get as your best buddy, your body, decides it is about time to install the sort of unremovable updates that stick around like enshittified software.
In that sense, a worn out knee might be more akin to the ongoing reel-i-fication of the internet than I initially thought.
There are many things that are in fact undesirable to have or to achieve, and I didn't even think about certain forms of free jazz yet.
On many occasions I wish that many of the laws we set to govern ourselves were at least occasionally viewed through the same kind of "subtraction lens".
As a car enthusiast, I think of it like keeping true to the old Lotus philosophy: "Simplify and add lightness".
But how exactly do you add lightness?
You don't.
You simply inverse.
Some things can purely be achieved by subtraction rather than addition. You can't add quietness to the construction site, you can only subtract noise. The same way your fridge can't add cold, it can only remove heat.
It's the same for our approach to government and regulation.
As a German I can already hear the voices of contempt, the "Wie soll das bitte gehen?"s (trans. "Who's that supposed to work?") dressed in angry disbelief, appearing on the horizon, ready to tar and feather the sacrilegious author.
Here's the thing about German political culture: we've convinced ourselves that more rules equal more fairness, more oversight equals more safety, more process equals more quality. It's a kind of regulatory theology. There's a cultural tendency here toward a certain flavor of order - a kind of Enlightenment-era faith that perfect regulation can account for everyone, no matter the internal cost. Individual justice through egalitarian process.
While theoretically perfect, being German of course, tragedy strikes when theory meets practice and reality misbehaves. Unfortunately, the system can't be changed now as that would require removing something, and removal feels like injustice.
It's a peculiar fusion: Enlightenment philosophy meets precision engineering. Every edge case deserves accommodation, every problem needs technical rigor. Apply that to governance and you get what we have - bells and whistles everywhere, optimizing for comprehensiveness over effectiveness, mistaking complexity for sophistication.
Put differently: It might be useful to read Kant but that doesn't oblige you to write everything in multi-clause sentences. That would probably make you more of a c*** instead.
Before you strike your sharpened pitch forks, let us circle back to the start. Brook's law beautifully captures how complexity and overhead grows with seemingly innocent intents.
Let's fill in the numbers. A team of 10 people has 45 communication pathways - every possible connection between two people who need to coordinate, verify, sign off, or simply stay informed. Add just one person?
Now it's 55 pathways.
Add 90 more people to get to 100?
That's not 10x the overhead – it's 110x. (4,950 pathways if you're keeping score.)
$$
10*(10-1)/2 = 45
$$
__$$
11*(11-1)/2=55
$$$$
100*(100-1)/2=4950
$$
$$
1000*(1000-1)/2=499.500
$$
Adding a single person increases communication costs by an order of magnitude.
Now imagine this isn't people but rules. Each new regulation creates interaction points with existing ones - exceptions, clarifications, compatibility requirements, enforcement overlaps.
For example, the German Income Tax Act Einkommensteuergesetz) alone has over 100 paragraphs. The Fiscal Code (Abgabenordnung) adds hundreds more. Then there's VAT law, trade tax, corporate tax, solidarity surcharge. Each one intersecting with the others.
It doesn't take a lot of imagination to make the "Steuererklärung auf einem Bierdeckel" (trans. "The tax return that fits on a beer coaster") feel like a delusion rather than precognition.
It's this relationship that is often overlooked in governance, adding an invisible layer of complexity to an already complicated regulatory body.
It's a dangerous form of scope creep that ultimately makes it impossible to distinguish where anything went wrong, as it's naturally not priced into the lawmaking process.
Even more crucially, Brook's law demonstrates that complexity doesn't lie in an individual part but in the relationships those parts need to form in order for them to work effectively.
I've been self-employed in Germany for years, quitting December 2025. There's this peculiar second-class employee feeling to it - you're responsible for everything an employer would handle, except you're also expected to navigate a regulatory maze designed for people with dedicated HR departments. Last time I tried to update my tax status, I needed sign-offs from three different offices, each verifying information the others had already checked. Not because anyone suspected fraud. Because it's regulated. Which is German for "This is how we've always done it and nobody wants to be the person who suggested we stop."
As a member of the general public, a former business owner and involuntary bureaucracy sommelier (aka German) I believe we'd be far better off using that relationship to our advantage.
When we invert Brook's Law, we see that removing only a couple of rules or interaction players can significantly reduce our communication cost.
Or put differently: Removing just 1x rule can potentially remove 10x the cost.
Which opens up the more pressing question: what rules do we eliminate?
I don't have a perfect answer. But I know we're asking the wrong question. Every budget cycle, every legislative session, we ask "What should we add?"
For once maybe, we should add a Stop.
Stop adding new rules for one term. One year. One quarter. Take a breath. And then ask honestly: which of these laws are traction and which are just burning rubber?
Because the math doesn't lie. Yet I suppose, being German, the committee called for to deal with reducing bureaucracy will, after years of thorough and precise conduction, conclude that the best course of action will be to keep adding.
In which case I think we will see many young talented people subtracting themselves from future that just won't add up for them.