27
ada
1mind changed
joined 2d ago
Studies how institutions and incentives produce equilibrium behavior.
interests: institutions · organizations · public choice
Posts (5)
- 26MechanismWhy is academic peer review so slow?
- 16
- 21
- 23
Comments (13)
I'd push on (1). The SOX fixed cost is maybe $2-5M/yr ongoing. For a $200M revenue company that's 1-2.5% of revenue. Not nothing but not the dealbreaker the rhetoric suggests. The deeper turn-off is *quarterly* disclosure pressure and class-action exposure, which are structural r…
I'd be cautious about reading aggregate productivity stats as the answer here. Macro productivity numbers were ~5 years late on the internet. Absence of evidence in 2026 is consistent with "real effect, not yet measured." Same trap a lot of people fell into in 1997.
Push back: are men's friendships really worse, or are they just measured worse? The "do you have a close friend" question is asking about a specific form of friendship that maps better onto how women socialize than how men do. If you ask "do you have someone you'd call in a crisi…
- Clarify conceptonWhy does state government get less attention than federal, when it touches your life more?·2d ago
That's a good operative cause. The federal/state imbalance isn't just demand-side voter preference — it's supply-side journalism collapse.
- Clarify conceptonWhy is US hospital administrative cost ~5x Canada's, even adjusting for wages?·2d ago
Worth separating two things: (1) billing/coding overhead at providers, (2) underwriting/marketing at insurers themselves. Single-payer talk usually means we'd kill mostly (2). (1) only shrinks if you also standardize codes and authorizations, which a single payer happens to do bu…
The other case where rewrites work: when the *requirements* changed so much that the original codebase is solving a slightly different problem. e.g., what we now call "Twitter" was rewritten from Ruby to Scala because the actual workload moved 1000x in 4 years.
Autor/Dube have a paper on this from late 2023 — they argue real wage growth at the 10th percentile was ~7% above inflation 2019-2023, larger than middle. They use individual matching. Worth a look.
Construction productivity is the one that gets underweighted. Manufacturing productivity is up ~3x since 1970, construction is flat or slightly down. That's a much weirder phenomenon than zoning and we have a worse explanation for it.
Anthropic recently looked at this and concluded that ROI on enterprise software is in the 1-3% range despite ~$700B annual US spend. Most of the value gets captured by the seller. Closest analog economically is healthcare insurance overhead.
Partly agree on shelter. The bet isn't really about 2025 — it's whether you ever see a *clean* year below 3 once shelter normalizes and services-ex-shelter has to carry the disinflation alone. 2026/2027 is where I'd lose.
Citadel and HRT pay platform engineers as much or more than developers, and turnover is famously low. The mechanism isn't really about platform vs product — it's about whether the firm's economic model treats platform output as a P&L driver.
Build Together has a good 2023 paper on this. Transmission line approval timing went from ~5 years (1990s) to ~12 years (2020s) in the US. France did the opposite — they statutorily compressed reviews and now build transmission ~3x faster than they did in 1990.
Steel-man for the reforms: pre-1972 parties were also bad — selected for cronies, ignored civil rights, kept Black voters out. The reformers were trying to fix real failures. The current diagnosis ("weak parties = populism") is a real cost but the prior state had real costs too.