ExpertsFMWhat I learned about power from LBJ and Robert Moses
Dwarkesh Podcast · October 29, 2026
5 Selected Lessons:
heuristic · 2 framework · 2 case · 1
Building a coalition of beneficiaries as a political insurance policy
Source
Dwarkesh Podcast — What I learned about power from LBJ and Robert Moses (October 29, 2026) · at 10:27 ▸
When this applies
when you need durable protection against political or organizational opposition to your projects/decisions
The takeaway
Convert potential political enemies and neutral parties into stakeholders with a direct financial interest in your continued success.
Why it works
By distributing concrete financial benefits (discounted bonds, jobs, fees, brokered contracts) to multiple powerful constituencies, you create a standing group whose own self-interest compels them to defend you whenever you're challenged, with no need for ongoing persuasion.
One level up
Durable power is sustained by aligning other people's self-interest with your survival, not by argument or popularity alone.
Where it stops holding
This stops working when the benefits distributed are discovered to be corrupt/illegal and trigger scandal, or when the coalition members' interests diverge from yours over time.
Detail
procedure:
- Identify the powerful actors who could obstruct your projects (banks, unions, legal/political brokers).
- Give each a direct, recurring financial stake in your projects continuing (discounted bonds, guaranteed jobs, fees, brokered contracts).
- Let this network mobilize on your behalf automatically whenever someone — a borough president, even the governor — stands in your way.
rationale: A self-interested coalition defends you more reliably and persistently than public popularity does.
In their words
Moses could then rely on these beneficiaries to call up anyone who stood in the way SPEAKER_00: of one of his projects, SPEAKER_00: whether that was a borough president or the governor himself.
Three tactics for institutional leverage: control resources, set the agenda, broker every deal
Source
Dwarkesh Podcast — What I learned about power from LBJ and Robert Moses (October 29, 2026) · at 16:16 ▸ · at 19:29 ▸
When this applies
when you lead a group whose members each have independent power and your formal title grants you almost no authority over them
The takeaway
Real power in a body of formally co-equal members comes from controlling scarce resources members depend on, controlling what options get voted on, and being the mandatory broker of every deal.
Why it works
By controlling the scarce things members need (committee seats, bill scheduling), controlling which version of any decision reaches a final vote, and inserting yourself into every negotiation as the necessary intermediary, a nominal leader converts an empty title into real control over outcomes.
One level up
Formal authority and real power are different things; real power is built by making yourself structurally indispensable to everyone else's success.
Where it stops holding
This stops working when members have alternative channels to get what they need (other leaders, bypassable rules), or when transparency requirements prevent private deal-brokering.
Detail
procedure:
- Gain leverage: control committee assignments and bill scheduling so members need your approval to advance their careers or priorities.
- Set the agenda: use mechanisms like unanimous consent agreements to control which version of a bill reaches the floor and limit surprise amendments or debate that could shift outcomes.
- Broker every deal: insert staff into every committee to track what's being drafted and what each side would trade, so that you personally instigate and control most of the deal-making.
benchmarks: {"johnsons_age_at_leadership":44,"senate_average_age":60,"formal_power_before_johnson":"right to be recognized first on the floor"}
rationale: Converts a role the founders deliberately designed to be powerless into the second most powerful position in the country.
In their words
The real political genius of Lyndon Johnson was identifying and then gaining control SPEAKER_01: over every possible source of leverage over his colleagues.
Hold multiple simultaneous, independently-appointed posts to become unfireable
Source
Dwarkesh Podcast — What I learned about power from LBJ and Robert Moses (October 29, 2026) · at 7:12 ▸
When this applies
when you want to make yourself resistant to removal by any single stakeholder or boss
The takeaway
Diversify the sources of your authority so that no single principal has unilateral power to revoke all of it.
Why it works
If removal requires coordinated action across multiple independent decision-makers, no single actor can unilaterally fire you, and each one bears a disproportionate political cost for trying alone.
One level up
Structural redundancy in your sources of authority converts dependence into independence.
Where it stops holding
This stops working when the appointing authorities coordinate against you, or when a single constitutional/legal change can strip all posts at once.
Detail
rule: Accumulate roles appointed by different, independent principals so that removing you from one role doesn't remove your overall power, and no single principal can remove you from all roles at once.
threshold: Moses held 12 simultaneous posts across mayoral and gubernatorial appointment
rationale: Makes you a 'hydra' — cutting off one head doesn't kill the body, and each principal fears the political cost of making an enemy of someone who remains powerful regardless.
In their words
Well, at his peak, Moses was a hydra with a dozen heads. SPEAKER_00: He held 12 different government posts at the same time.
Deliver visible wins within a single election cycle
Source
Dwarkesh Podcast — What I learned about power from LBJ and Robert Moses (October 29, 2026) · at 7:12 ▸ · at 10:27 ▸
When this applies
when you need a sponsor or patron to keep renewing your mandate or funding
The takeaway
Match the delivery timeline of your work to the decision cycle of whoever grants you resources or authority.
Why it works
Sponsors who grant authority need to see tangible results before their own next accountability moment (election, board review, etc.); delivering within that window creates a feedback loop that reinforces continued support.
One level up
Power is sustained less by the scale of what you build than by whether your patrons can personally claim credit for it in time.
Where it stops holding
This stops working when the underlying project genuinely requires a longer timeline than any political cycle allows, forcing a mismatch between patron incentives and project reality.
Detail
rule: Deliver popular, visible results before the patron who authorized you faces their own next re-election.
threshold: If the mayor gave Moses the power to build something, he could expect to be at its ribbon cutting before the mayor was up for re-election
rationale: Politicians will keep empowering someone who reliably produces election-cycle-compatible wins they can take credit for.
In their words
If the mayor gave Robert Moses the power to build something, SPEAKER_00: he could expect to be standing at its ribbon cutting before he was up SPEAKER_00: for re-election.
Bundling unrelated interests to engineer passage of the 1957 Civil Rights Act
Source
Dwarkesh Podcast — What I learned about power from LBJ and Robert Moses (October 29, 2026) · at 21:54 ▸
When this applies
when a deal is deadlocked because each side's minimum requirement is the other side's dealbreaker
The takeaway
When a negotiation is deadlocked on its own terms, expand the scope to include unrelated issues where you can create new value to trade, rather than trying to resolve the original conflict directly.
Why it works
By bringing in entirely unrelated issues that different subsets of stakeholders care about, a broker can create new trade space that didn't exist within the original issue alone, letting each faction get something it wants elsewhere in exchange for conceding on the original sticking point.
One level up
Deadlocks dissolve not through better arguments on the original issue, but by changing what the negotiation is actually about.
Where it stops holding
This stops working when you lack control over the unrelated issues needed to create new trade space, or when combining issues is seen as illegitimate logrolling that delegitimizes the whole deal.
Detail
situation: Northern liberals and Southern senators were deadlocked over whether jury trials would apply to voter-suppression cases — any version acceptable to one side was unacceptable to the other.
decision: Johnson spent seven months building a deal linking unrelated issues: a federal dam for Midwestern senators, jury trial rights for striking railroad workers for Midwestern Republicans, and a voting-rights clause for moderates — while keeping the jury trial provision for Southern election officials to avoid a filibuster.
outcome: The Civil Rights Act of 1957 passed, Johnson's first major civil rights win, achieved by trading across issues that had 'nothing to do with voting rights in the South' on their face.
lesson: A deadlock on a single issue can be resolved by expanding the deal to include unrelated concessions each faction separately values, requiring deep knowledge of everyone's true priorities.
In their words
A dam in Idaho and protections for union workers and federal jury rules had nothing SPEAKER_01: to do with voting rights in the South, SPEAKER_01: at least not to anyone who lacked the inside knowledge and political creativity SPEAKER_01: of Lyndon Johnson.