Section 1
First, the receipts
Claims about "regulation crushing small business" are usually vibes. Here the field data is unusually clean, so let me anchor before I argue. The IfM Bonn and the IMPULS-Stiftung studied bureaucracy costs in Germany's machine and plant engineering sector and found the burden is sharply regressive by firm size. For a smaller firm, compliance ran to roughly 6.3 percent of annual revenue. For the largest firm in their sample, it was about 1.3 percent. That is close to a five-to-one gap in the share of revenue eaten, between a small firm and a large one, for the same regulations. In one modelled case a smaller firm's bureaucracy burden reached 2.18 million euros a year, which the researchers pointed out equals the full-time salary of 34 employees. The state's own development bank agrees on the shape. KfW Research's Mittelstandspanel (Fokus Nr. 495, April 2025, drawn from around 10,000 firms) found that solo self-employed people spend 8.7 percent of their working time on bureaucracy, the highest of any group, and that the relative burden falls steadily as firms grow larger. Across the whole Mittelstand it is about 7 percent of working time, roughly 32 hours per firm per month, and around 61 billion euros a year. And the macro cost of the whole apparatus: the ifo Institut estimated in November 2024 that excessive bureaucracy holds German economic output down by about 146 billion euros a year, relative to what the country would produce at Sweden's level of administrative burden. Hold two of those numbers next to each other. Small firm: 6.3 percent of revenue on compliance. Large firm: 1.3 percent. Every rule that produces that spread is, in competitive terms, a subsidy to the large firm paid by the small one. Not in cash. In survival odds.
Section 2
The moat is real, but nobody built it
Here is where most commentary goes wrong, and I want to get it right because the solution depends on it. The compliance moat is not a conspiracy. There is no back room where incumbents commission regulations to strangle upstarts. Each rule was written for a real and often good reason: worker safety, tax fairness, data protection, environmental limits, consumer rights. The people who wrote them were not thinking about competitive dynamics at all. The moat is emergent. It is what you get when you apply uniform, fixed-cost obligations across firms of wildly different sizes. No intent required. The arithmetic does the entrenching on its own. This distinction matters enormously, because if you believe it is a conspiracy you will waste your energy raging at a villain who does not exist, and you will miss the fact that a mechanism with no author can still be routed around by anyone who sees it clearly. To see it clearly you need three models. One for why the wall exists (mechanism design). One for how tall it is at your size (comparative statics). And one for how to actually play against an opponent who is standing on top of it (game theory). Then the flag that says the wall is currently being rebuilt.
Section 3
Model 1: Mechanism design, reading the effect off the equilibrium
Mechanism design works backward. You look at the outcome a system reliably produces and ask what set of rules would produce it if someone had designed it on purpose, whether or not anyone did. The outcome German regulation reliably produces: incumbents carry compliance lightly, entrants carry it heavily. What rules produce that? Uniform requirements applied without regard to size. Documentation duties that assume a professional back office already exists. And, critically, thresholds. The IfM researchers noted that bureaucracy costs often arrive in discrete jumps, because some obligations only apply above a certain company size. Below the line, exempt. Above it, a whole new lump lands at once. Those thresholds are the most interesting part, because they are where the moat is most visible. A firm approaching a size threshold faces a step-change in cost the moment it crosses. That creates a real disincentive to grow past the line, which economists call a threshold effect and which every ambitious small owner feels as a strange reluctance they cannot quite name. The rule that was meant to scale obligation with capacity ends up teaching small firms to stay small. A system optimizing for incumbent stability could not have designed it better, and no one designed it at all. The tax layer sharpens the picture. The IW Köln attributes about 45 percent of total bureaucracy cost to tax bureaucracy alone. Tax compliance is the purest fixed cost there is: understanding the rule costs the same whether you file for six employees or six thousand. So the single largest slab of the moat sits in the one domain where the fixed-cost logic is strongest. That is not a coincidence. It is the mechanism showing you where it lives. Mechanism design: the "why is it shaped this way" lens • Assumes: you can read intent-of-effect off the equilibrium, even with no designer present. • Fits because: uniform fixed-cost rules plus size thresholds reliably produce incumbent advantage and a growth disincentive. • Breaks when: you slide from "emergent effect" into "deliberate cartel." The moment you claim intent you cannot prove, you lose the argument and misread the fix. • Counteracts: the belief that regulation is competitively neutral. Fixed costs are never neutral. • May reinforce: a temptation to treat every rule as illegitimate. Most rules have a real purpose; the problem is the size-blindness, not the existence.
Section 4
Model 2: Comparative statics, measuring the wall at your height
Mechanism design tells you the wall exists. Comparative statics tells you how tall it is specifically for you, by moving one variable and holding the rest still. Move firm size down. Watch the compliance-share-of-revenue climb, because a fixed cost spread over a smaller base is a bigger fraction of it. The IMPULS data is the curve made visible: 1.3 percent of revenue at the top, 6.3 percent at the bottom. A four-to-five-times heavier relative load, purely as a function of size, holding the actual rules constant. The strategic implication is the part owners miss. If compliance is 6.3 percent of your revenue and 1.3 percent of your larger competitor's, then before you have quoted a single job you are carrying a five-point cost disadvantage that has nothing to do with how well you run the business. The large firm can price five points lower at the same margin, or bank five points more margin at the same price, funded entirely by an efficiency you cannot access at your size. This is why "just compete on service" is necessary but not sufficient advice. You are not competing on a flat field. You are competing uphill, and the gradient is set by arithmetic you did not choose. Comparative statics: the equilibrium-shift lens • Assumes: you can move size alone and that compliance is genuinely fixed-cost. • Fits because: the cost is a share of a shrinking base as the firm gets smaller. • Breaks when: parts of compliance actually scale with activity (some payroll and transactional tax work do), which flattens the disadvantage somewhat at the very bottom. • Counteracts: the false comfort that small firms and large firms face "the same rules, so it is fair." Same rules, wildly different burdens. • May reinforce: despair. The gradient is real, but the next model is about playing anyway.
Section 5
Model 3: Game theory, playing against an opponent on the wall
Now the useful part. You are a small firm. Your larger competitor sits on a compliance advantage you cannot match by shrinking your own paperwork below theirs, because theirs is already lower as a share of revenue. Game theory asks: given the payoffs, what is your best move, knowing the other player's position? The naive move is to fight on the axis where you are structurally weak: cost. Try to out-cheap a firm that carries a five-point compliance edge and you lose, slowly, then all at once. That is playing their game on their wall. The better move comes from a simple game-theoretic idea: do not contest the front your opponent has fortified. Contest the front they cannot defend. The large firm's compliance advantage is inseparable from the thing that creates it, which is size, and size buys efficiency at the cost of intimacy, speed, and trust. A big firm cannot have the owner personally answer the phone, because the owner is four layers away from the phone. It cannot turn a quote around in an afternoon, because the quote goes through a process. It cannot be the person the customer actually knows. So the equilibrium strategy for a small firm is not to fight the moat. It is to compete on the dimensions the moat's owner had to give up to build it. Trust, responsiveness, and being a known human are not consolation prizes. They are the exact axes on which a fixed-cost efficiency machine is structurally weak. (This cluster has a whole separate piece on trust as the only defensible moat in a low-formality market; the logic rhymes.) There is a second game-theoretic move, and it is about the threshold. Because obligations jump at size thresholds, a small firm has a real choice most never make consciously: stay deliberately below a threshold, or plan the crossing so the new cost lands on a base large enough to absorb it. Crossing a threshold with barely enough revenue to cover the new lump is the worst position. Either stay clearly under it while you build, or cross it with room to spare. Drifting across it by accident is how a growing firm suddenly finds its margin gone and blames the market. Game theory: the strategic-actor lens • Assumes: identifiable players with stable payoffs, each choosing a best response to the other. • Fits because: the large competitor's fortified front (cost) and undefended front (intimacy) are both direct consequences of the same size. • Breaks when: the large firm learns to fake intimacy convincingly (good local branding, a well-run CRM) and closes the undefended front. Then the small firm's edge thins and it needs a new one. • Counteracts: the instinct to fight on price, the one axis you are built to lose. • May reinforce: an over-reliance on "we care more," which is true until a competitor systematizes caring. Trust has to be operationalized, not just felt.
Section 6
The structure-break flag: the wall is being rebuilt right now
Every model above assumes the compliance regime is stable. It is not. Germany is mid-transition on mandatory electronic invoicing (e-Rechnung, via the Wachstumschancengesetz), and a transition is exactly when a moat can be re-cut in either direction. The timeline: from 1 January 2025 every firm must be able to receive structured e-invoices. From 1 January 2027 firms above 800,000 euros turnover must issue them. From 1 January 2028 essentially all firms must. The smallest Kleinunternehmer are exempt from issuing but must receive. Watch what the thresholds do to the competitive game. For a two-year window (2027 to 2028), the mandate applies to larger firms first and smaller ones last. Read cynically, that is the moat working as designed: the burden lands on those best able to carry it, and the smallest get a grace period. But read strategically, it hands small firms something rare, a warning shot. You can watch larger competitors implement the system, learn from their mistakes, and adopt the automation on your own timeline before it is compulsory. And here is the genuine structure-break. Every prior fixed-cost rule made the moat deeper, because it added manual back-office work that large firms absorbed and small firms could not. e-Rechnung is the first major mandate that ships with the automation attached. A structured invoice is machine-readable by definition. The mandate is, in effect, forcing the whole economy onto a data format that lets small firms automate the exact back-office work that used to be their disadvantage. For once, the fixed cost comes with a lever to eliminate the recurring cost underneath it. That means the moat's depth is, briefly, up for grabs. Small firms that treat e-Rechnung as one more burden to survive keep the manual process and stay in the ditch. Small firms that treat it as the state subsidizing their back-office automation climb partway out of it. Same rule. The competitive effect depends on the response, which is the definition of a regime you can still influence.
Section 7
The counter-strategy, in three moves
Seeing the moat is not a plan. Here is the plan, in the same order every piece in this cluster uses: triage, portfolio, base-rate check. Triage the compliance load (GEER), automate-first. For every recurring obligation, ask in order: can a machine do it, should an expert do it, can someone junior do it, and only then, must the owner do it. The point is not to match the large firm's back office. It is to shrink the recurring cost enough that the five-point gradient becomes a two-point gradient, which your intimacy advantage can more than cover. The failure mode: assuming the automation exists for every niche obligation. For some it does not yet, and forcing it just adds a dead subscription. Pull the cheap, reversible levers first so a miss costs a month. Build a dated portfolio (RADAR). Do now: adopt e-invoice-capable software while it is still optional for you, so you learn it on your schedule. Hedge: keep a light relationship with a tax advisor as insurance against the next rule and the next audit. Defer with a trigger: do not build a full back office or chase a size threshold on instinct. Write the number that will make you cross the threshold, and cross it only with revenue to spare. The failure mode: a trigger you never measure never fires, and "defer" quietly becomes "never." Check the base rate (CHAIN). The reference class is unkind: most small firms absorb each new mandate by hand and change nothing structural, which keeps the moat exactly where the incumbents like it. Knowing that is how you beat it, by refusing the default. The second-order prize: once your data is structured and your pipe is built, every future mandate costs you less to absorb, so the moat stops deepening against you specifically. The matrix-break flag: base rates from the paper era may not survive the shift to mandatory machine-readable data. History says small firms lose the compliance race. History did not have the state handing them the automation. Use the base rate as a warning about your habits, not a verdict on your odds.
Section 8
What this ensemble cannot see
The models are blind to three things worth naming. They cannot see genuine regulatory capture where it does exist. I have argued the German moat is mostly emergent, not designed, and I believe the data supports that. But in specific industries, incumbents really do lobby for rules that raise entry costs, and there the moat has an author. The ensemble treats the wall as arithmetic. In some sectors it is also strategy, and you should check which you are in. They cannot price legitimacy. Some of these rules protect real people from real harm. A framework that treats all compliance as pure competitive friction will, taken too far, turn into an argument for cutting protections that exist for good reason. The goal is to survive the moat, not to campaign against safety. And they cannot tell you whether reform is coming. The ifo's 146-billion-euro figure is a reform argument, and the IW Köln notes past bureaucracy-relief laws underdelivered. If serious reform lands, the moat shrinks on its own and part of this strategy becomes moot. If it does not, the moat stands. The models assume the wall stays roughly where it is. Politics is the variable they cannot read.
Section 9
The fitness test
One question, this week. Estimate your compliance cost as a share of revenue, and estimate your largest competitor's. You will not get it perfectly, but the direction is what matters. If your share is meaningfully higher, and by the arithmetic of fixed costs it almost certainly is, then you already know you cannot win on price. The only question left is whether you are spending your energy fighting uphill on the moat's own terms, or spending it on the one front the moat forced your competitor to abandon. The wall is real, and nobody built it. That is the bad news and the good news in one sentence. Bad, because you cannot appeal to a villain. Good, because a wall with no author is just terrain, and terrain can be flanked by anyone who bothers to read the map.