Runaway

What Gregory Bateson might have said about California's fiscal crisis

In the 1970s, the anthropologist and systems theorist Gregory Bateson became preoccupied with a particular class of failure. Not the ordinary kind (bad decisions, corrupt actors, insufficient resources), but the kind where a system’s own corrective responses make things worse. Where the feedback loop that should damp a problem instead amplifies it. Where the harder you push back, the faster you fall.

He called this a runaway. And he believed it was the characteristic pathology of complex systems that had lost touch with their own structure, systems that were intervening on symptoms without understanding the dynamics producing them.

California looks increasingly like a textbook case.

It’s worth being precise about what made California’s innovation economy work, because the current policy debate tends to treat it as a backdrop, as if Silicon Valley were a geographic fact rather than an complex achievement.

What California built, over roughly six decades, was a functional ecosystem in the biological sense: a web of interdependent relationships in which the outputs of each element became inputs for others. Stanford and Berkeley produced talent and research. That talent attracted venture capital. Venture capital funded experiments. Successful experiments generated wealth that recirculated into angel investments, new funds, philanthropy, and taxes that funded more university research and public infrastructure. The network density was self-reinforcing: the more high-quality actors clustered, the more valuable the cluster became for each of them.

This isn’t a novel observation. What’s less often noted is that the system had an important property Bateson would have recognized: it was complementary. The relationship between the state and its most productive residents was mutually constitutive. The state enabled wealth creation through its universities, infrastructure, and rule of law. Wealth creation funded the state. The state reinvested in the conditions for further wealth creation. Each party needed the other; each enhanced the other’s function. A healthy complementary loop.

Complementary relationships, Bateson observed, are stable when both sides perceive the exchange as reciprocal, and they become unstable when one side begins to experience the relationship as purely extractive.

California’s fiscal situation is genuinely difficult. The state has significant inequality, severe affordable housing shortages, strained public services, and a budget that swings violently with the fortunes of a small number of high earners. These are real problems that create real political pressure.

The response: progressive taxation pushed to increasingly aggressive levels, culminating in the proposed 5% annual wealth tax on net assets above $1 billion, reflects a particular implicit model of the system. In that model, concentrated wealth is a reservoir: accumulated, static, extractable. Tax it and you can fund services. The wealthy will remain because they are embedded here; their networks, their identities, their businesses are Californian.

The problem is that this model is wrong in a specific, Batesonian way. It correctly identifies the output of the system (concentrated wealth), while misidentifying the structure that produces it. It treats the fruit as separable from the tree.

What the model misses is that the wealth is not just located in California; it is, in significant part, produced by California’s ecosystem. Strip out the people whose activity generates and recirculates that wealth, and you don’t have the same ecosystem minus some rich people. You have a degraded ecosystem that generates less of everything: less innovation, less employment, less tax revenue, less of the network effects that made the cluster valuable in the first place.

Bateson’s framework gives us a precise vocabulary for this error: the intervention is operating at the wrong logical level. It acts on the symptoms of the system without a model of the system’s generative structure. And interventions that operate at the wrong logical level don’t just fail to solve the problem; they tend to accelerate it.

Runaway

The mechanics of the resulting runaway are now visible in the data.

High-net-worth individuals are leaving, not in catastrophic numbers yet, but in a pattern that compounds. Venture capital investment, while still substantial in absolute terms, has been declining as a share of national totals. Texas and Florida now regularly appear in founder surveys as primary alternatives to the Bay Area. The number of new business formations in California relative to other states has been trending downward for over a decade.

Each departure is individually rational. A founder with $2 billion in paper wealth, facing a proposed annual levy of $100 million regardless of liquidity, has a clear incentive to establish residency in a zero-tax state before a liquidity event. The calculation is not close. And because the people most likely to leave are precisely the most mobile (i.e., wealthy, successful, globally networked), the exits are disproportionately concentrated among the nodes that provide the most connective tissue to the ecosystem.

This is where the runaway logic kicks in. As the highest-value participants exit, the network becomes less dense, less valuable, slightly less worth staying in for the next tier. The tax base narrows. The state, facing fiscal pressure, maintains or increases rates on those remaining. The cost-benefit calculation for the next tier shifts. More people reconsider. The loop accelerates.

At each step, the state’s corrective response (maintaining revenue by taxing the remaining base more heavily) is the action that drives the next round of exits. This is the signature of a Batesonian runaway: the system’s own error-correcting behavior is producing the error.

Double Bind

What makes this more than just a bad policy cycle is the structure of the trap California is now in. Bateson defined a double bind as a situation in which two injunctions (each apparently reasonable) are mutually contradictory, and in which the subject cannot exit the situation or comment on its contradictory nature without penalty. Applied to systems rather than individuals, it describes a state in which every available response worsens the condition.

California’s version looks like this: the public services that make the state livable and that constitute part of its value proposition (e.g., universities, infrastructure, safety net), require substantial tax revenue to maintain. Without that revenue, they degrade, and the state becomes less attractive, accelerating exit. But the taxation required to maintain those services is itself accelerating exit. Cut taxes and you degrade the public goods that anchor the ecosystem. Maintain taxes and you drive out the people whose activity generates the ecosystem’s productivity. The exits worsen the fiscal position, which intensifies the pressure to tax, which drives more exits.

There is no move available within the existing logic of the system that resolves the contradiction. This is the double bind.

The prior error that created it (the one Bateson would look for) is California’s failure to maintain the housing conditions that could have kept the ecosystem’s middle layers intact. The engineers, designers, researchers, and operators who support and extend the work of founders and investors were progressively priced out of the state over decades. This was the slow erosion of the ecosystem’s base, driven by the same political economy that now drives the fiscal crisis: a state that protected existing asset holders while failing to invest in the infrastructure that would have sustained broad participation. The wealth tax debate is, in this sense, a late-stage symptom of a longer dysfunction. The tree was being damaged at the roots for years before anyone noticed the canopy thinning.

Systemic Response?

Bateson’s concept of calibration (the meta-level correction that can reset a runaway) requires stepping outside the feedback loop and acting on its structure rather than within its terms. For California, that would mean policy designed around ecosystem maintenance rather than extraction.

The difference in practice: rather than taxing accumulated wealth as a static reservoir, the state would focus on the conditions that generate productive activity (e.g., dramatically expanding housing supply to reduce the cost of participation, investing in the research infrastructure that makes the cluster valuable, reforming the regulatory environment that imposes friction on new venture formation). The goal is to make the complementary relationship functional again: the state as active co-producer of the ecosystem, rather than as tax collector of its outputs.

This is not an argument against redistribution in principle. It’s an argument that the method of redistribution matters systemically, that a state which degrades its productive ecosystem to fund redistribution is not actually a more equitable state over time, but a less productive and ultimately less equitable one.

Whether California’s political economy can generate that kind of calibration is a separate question. The pressures that produced the current trajectory (e.g., concentrated homeowner interests, a legislature insulated from the competitive pressures facing founders and businesses, genuine and legitimate anger about inequality) are not easily dissolved. Systems caught in runaways rarely self-correct before the stress becomes acute enough to force structural change.

The founders and investors most likely to be reading this are, in a sense, the protagonists of the Batesonian story. Many are making (or watching peers make) the individually rational decision to relocate. Each of those decisions is defensible on its own terms. Each also accelerates the degradation of the system that produced the opportunity in the first place.

This is not a reason to stay, necessarily. Individual actors cannot unilaterally solve structural problems, and there is no virtue in absorbing a punitive tax burden as a gesture toward a system that isn’t changing. But it is a reason to be clear-eyed about what is being lost and to resist the comfortable narrative that the next cluster, wherever it forms, will automatically replicate what California built. Ecosystems of this complexity and productivity take generations to develop. They can unravel much faster.

Bateson’s deepest point was that you cannot understand a system by looking at its parts in isolation. The wealth that is now dispersing to Miami and Austin and Dubai was not produced in isolation, but in relationship, by a particular density of talent, capital, culture, and institutional investment that accumulated over decades. The question worth sitting with is not just where to domicile, but what it would take to maintain the conditions under which the next generation of that ecosystem could exist.

That’s a systems question. And right now, California isn’t asking it.