In August 1971, Lewis F. Powell Jr.—not yet a Supreme Court justice—circulated a confidential strategy memorandum for the U.S. Chamber of Commerce. It argued that business was losing the “public relations” war across campuses, media, courts, and politics, and that institutions needed sustained, professional investment—not ad hoc lobbying.
Two months later, President Nixon nominated Powell to the Court (see Court historical materials). However you feel about the memo, the sequence is the point: a document about institution-building met a life-tenured branch.
Powell’s 1978 majority opinion in First National Bank of Boston v. Bellotti helped establish corporate speech rights in a form that later mattered to money in politics. The through-line is not that every later outcome was “planned.” It is that institutional design can outlive its operators.
The working hypothesis: Powell’s memo read like a campaign. What it actually described was an incentive machine—endowments, training, career pipelines, legal doctrine—that keeps producing compatible outcomes without anyone holding a central control room. The 2025 push to open 401(k) menus to private markets is the same class of move: change the default path of middle-class capital, and you change the political economy of what “safety” means.
The story everyone is telling
The pop version of the Powell story is a cabal: business wrote a memo, funded think tanks, captured courts, and got what it wanted. That version is not empty—money did move, and doctrine did shift. It is also a bad manual for the present, because it trains you to look for villains instead of gears.
The more useful version is mechanical. The Heritage Foundation, AEI’s modern scale, the Federalist Society’s law-school network, and the vast litigation infrastructure of the last half-century are not a single season of television. They are compounding infrastructure: talent identification, placement, shared language, repeat play.
What “machine” means in practice
A campaign ends. A machine amortizes fixed costs—clerkships, bar culture, bar review, judge selection, repeat litigators—across decades. That is why arguments about the Federalist Society (or any network) so often misfire: one side names coordination; the other side names emergent selection. Both can be partially true; the test is whether outcomes stay stable when the public mood flips.
By the 2020s, the Court’s center of gravity and lower-court placement patterns do not look like a temporary election story. They look like a mature pipeline—one that can also produce public breaks between political leadership and sitting judges, because life tenure is designed to make operators not need day-to-day instructions.
That independence is not a “bug” in the network model. It is the point of building a bench instead of buying an election.
The 401(k) layer: access as architecture
On August 7, 2025, the White House issued an executive order titled Democratizing Access to Alternative Assets for 401(k) Investors—formally published in the Federal Register with a stated aim of reducing practical barriers to including private market, real estate, digital asset, and similar sleeves in defined-contribution plans.
The policy debate is not only “more choice.” It is where retail risk is allowed to sit when assets are not continuously marked, not cheaply liquidated, and not transparent the way public index funds are. The Private Equity Stakeholder Project and related pension researchers have long documented opacity and fee loads in alternative channels moved through retirement systems; extending access does not automatically import institutional diligence to every small plan menu.
If public equities tie many households to corporate earnings through passive participation, private sleeves tie them—through managers—to fee-generating structures that are harder for an unsophisticated allocator to audit.
None of that requires a conspiracy—only the quiet math of defaults + menus + marketing language.
What would change my mind
-
Net-of-fee outcomes: If, over a full cycle, median retail target-date implementations with private-market sleeves match or beat comparable public-only glidepaths by a durable margin after all-in fees and illiquidity, the extraction critique weakens. Show me the cohort data—not the marketing deck.
-
Governance: If DOL transparency rules (or plan-level disclosures) make private sleeve economics legible enough that participants routinely see dollar-weighted net returns versus a simple public benchmark, the information gap I’m worried about closes materially.
-
Judicial/political feedback: If operator-judge fractures produce sustained doctrinal outcomes that contradict the long-run preferences of the pipeline that elevated those judges—across multiple terms—that would be evidence the “machine” is less stable than the incentive story assumes.
Related: Consensus Machine — another lens on how institutional defaults persist without anyone updating them each morning.
If you found this useful, the best thing you can do is forward it to one person who would push back on it. I’d rather be wrong in public than right in private.