Finding signal on Twitter is more difficult than it used to be. We curate the best tweets on topics like AI, startups, and product development every weekday so you can focus on what matters.
In 2004, if you asked a technology investor where the best software companies in the world were, they'd give you two answers: Boston and San Francisco.
This is obviously no longer the case. San Francisco has produced $14 trillion in enterprise value over the last two decades. While Boston has produced a measly $100 billion.
If you told that same investor that New York, in all its cocaine and grey pinstripe suit financial glory, would usurp Boston as a regional technology center—they'd think you were nuts.
The question of why Boston lost is worth studying.
From an inputs perspective, the city should have everything going for it. Two of the best universities in the world are based there. Y Combinator was founded there. It is without a doubt one of the most beautiful cities in the country. Mark Zuckerberg went to college there. So did the Stripe founders. So did the Cursor founders. So did the Dropbox founders. So what went wrong?
To understand the scale of the collapse, you have to remember that for decades, Boston's 'Route 128' was the center of the software universe. DEC was the second-largest computer company on earth with 140k employees at its peak. Lotus was the essential app that brought the enterprise to the PC. Akamai built the modern internet. Where did it go wrong?
This is a question worth answering. But any attempt to answer it gets you one of two answers:
The city died when Zuckerberg tried to raise money there, and couldn't and had to go west.
What do you mean Boston is dead? We just lead the Series F of TurboLogs at $15m post.
Of course, neither of these is a useful story. Figuring out what really happened is not only an existential question for Boston, but for the US technology ecosystem broadly.
My answer is simple: Boston is a story of what happens when negative cultural and regulatory feedback loops align. The city, as a technology ecosystem, was killed by three simple forces:
A progressive regulatory state that treated business as a coffer to loot for the benefit of property owners.
For decades, Massachusetts refused to conform to federal QSBS rules. The state finally conformed in 2022. That same year, they passed the “Millionaire's Tax.” A founder who sells for $10 million in Massachusetts owes $860,000. The same founder in Austin owes zero. Massachusetts also charges 6.25% sales tax on SaaS revenue. Most states don't tax software at all.
2. A puritan culture too enmeshed in elite institutions to police itself
Post 2010 main activity of Boston-based venture is not building companies: it is extorting founders and running organized crime cabals. The culture that should police this- the endowments, the large lps, the people who show up to the galas- is too intertwined with the perpetrators and their networks to say anything. This meant a constant “trust tax” on conduct in the city.
3. An inputs-first view of technological progress.
We have the best universities. We built a ton of lab space (even if 40% of it now empty). We have the best talent in the world. So why isn't it working? Can't we build another innovation center? Is our soil alone not magic?
If this three-factor explanation seems simplistic, and maybe even something you've heard before-- it's because it is. It's the very same problem that's facing technology industries across the entire US, and I suspect it will be similarly fatal.
Technology ecosystems are fundamentally fragile networks that generate trillions in tax yields for their hosts who can't keep themselves from killing the golden goose every few decades.
Let's play out what happens when a host rejects the organism:
First, talent networks dissipate. Need to hire a VP of Engineering who's taken a company from 25 to 500 people? There are 600 to choose from in SF. There are five in Boston and soon those five will leave for SF, where they can demand higher comp and better odds of success. On the junior side, new grads no longer stay local. They catch the first flight out every summer.
As the network evaporates, the state sinks its teeth in tighter to capture equivalent yield from whoever's left. As the ecosystem dissipates, seedy market participants enrich themselves: through preferential pricing (who is flying to Boston to win a seed? fine, we'll pay up at $10m) or shadier tactics of extorting founders with off market and often illegal tactics (see Nikita, and others tweets for stories that can legally be told). You see this character even in firms that started in Boston and left (less Matrix, they're good guys), who maintain a certain organized crime complex even after they've moved west.
These are complex, flesh-and-blood problems. They destroy cities, lives, and trillions in taxable enterprise value all because of short-term thinking on the part of state governments.
And the worst part: you can't undo them.
While I am sympathetic to calls to reclaim Boston as a great technology ecosystem—I would love to move back and not deal with New York—I struggle to see how the remaining ecosystem doesn't enter complete free fall.
You cannot legislate your way out of a collapsing network. You cannot cold start a network already folding in on itself.
And yet San Francisco, and the US technology ecosystem as a whole, is lining up for the exact same fate. A regulatory state that treats technology as a cash cow. Prop M. The office vacancy tax.
A culture too enmeshed in its own elite networks to police itself. AI has attracted dozens of bad actors into the ecosystem, and the same stodginess that made Boston impossible to clean up is settling in.
An inputs-first view of progress. We have the best AI labs. We have the most GPUs. Hell, the President even bought us some GPUs. We have the frontier models. So why would anything go wrong?
The difference is the stakes. Boston's collapse cost the US a few hundred billion in enterprise value. San Francisco's collapse erases a third of the country's GDP growth over the last decade.
But the failure here is not just economic. The failure is existential.
Our technology industry has failed to provide a coherent argument for its own existence on the national stage. If this is not solved, 2028 will become a referendum on caging, killing, and looting the technology industry over the water and power libels.
The popular image of the AI boom is not uncertain. Recent polling suggests the average American understands AI is a thing that wastes water, skyrockets power costs, and scams their grandparents in exchange for exposing children to deviant sexual content, sports gambling, and all other manner of sin.
If the best answer to why we should not imprison technology executives, burn down data centers, and kill the american technology sector is “so we can build better chatbots for your sports betting,” voters will vote to do so.
In a zero-sum world, voters do not think long term; they envy, and then they loot. We do not loot the sewage system or the power grid because we understand they are the bulwarks against chaos. We accept their costs because they hold back chaos. Does the median voter believe the same to be true about technology?
Technology is the only mechanism we have for escaping the Malthusian trap. But because we have been too cowardly to articulate this, because we have substituted “rationalism” and “AGI” for a coherent theology of Progress, the state sees the industry as only a parasite to be consumed.
If we cannot articulate why innovation is a moral imperative, we can expect the entire technology industry to end up like Boston. First taxed, then looted, then exhausted. And we'll be stuck wondering where it all went.
