Robert Caro’s The Power Broker: Robert Moses and the Fall of New York is a 1,246-page biography of a man who was never elected to anything and who, for forty-four years, built more of New York than any elected official before or since: the parkways and expressways, the bridges, Jones Beach and hundreds of parks and playgrounds, Lincoln Center and the United Nations site, and the public housing and urban renewal that, with his highways, displaced hundreds of thousands of people. Published in 1974 and awarded the Pulitzer Prize, it is usually read as a study of power; for an internal auditor it is something more specific, the most detailed account ever written of how a public authority can be designed to escape accountability, how control of information, money and expertise substitutes for a mandate, and how the governance apparatus of a city, state and press came to defer to an official none of them could remove. This guide summarises the book without reproducing it, sets out Moses’s methods as governance mechanisms, maps them to the audit of public bodies and authorities, and ends with a worked example of a regional authority’s audit function testing itself against the book’s questions.
In this guide
- The book in one page
- Who Robert Caro is, and how the book came about
- The structure: seven parts
- The rise: how an unelected official acquired power
- The methods, read as governance mechanisms
- The public authority: the instrument
- The fall, and why it took so long
- The nine ideas worth keeping
- The vocabulary: fourteen terms defined for practitioners
- What it means for internal audit: auditing authorities and the people who run them
- The criticisms, and what the book leaves out
- Editions, reading time and how to read it
- A one-hour discussion guide for an audit team
- Worked example: Millbrook Regional Water Authority asks the Moses questions
- Frequently asked questions
- Related guides
The book in one page
| Question | The answer, in our words |
|---|---|
| Who was Robert Moses? | A reformer who entered public life in the 1910s as an idealist for merit-based civil service, became Governor Al Smith’s most effective lieutenant in the 1920s, built the Long Island parks and Jones Beach, and from the 1930s to 1968 held as many as twelve state and city positions at once, including the chairmanship of the Triborough Bridge and Tunnel Authority, through which he controlled the money, the engineers and the projects that shaped the modern city. |
| What did he build? | Thirteen bridges, more than six hundred miles of parkways and expressways, 658 playgrounds, hundreds of parks, Jones Beach, Lincoln Center, the Flushing Meadows site of the 1939 World’s Fair and the 1964 World’s Fair, which he ran, and the Title I urban renewal and public housing projects of the 1950s. |
| What is the argument? | That Moses acquired power that no elected official could check, by drafting the laws that created his positions, by controlling the revenue of the authorities he ran, by monopolising the engineering expertise the city depended on, by managing the press, and by keeping files on everyone; and that he used it to build what he wanted, where he wanted, at a cost that fell on the people with the least power to object. |
| How did it end? | In 1968 Governor Nelson Rockefeller folded the Triborough Authority into the new Metropolitan Transportation Authority and removed Moses from power, in a manoeuvre that depended on the bondholders’ consent Moses had always said no one could obtain. |
| What is the method? | Seven years of research, more than five hundred interviews, and a prose style that reconstructs decisions from the inside; the book is as much about how power works as about one man. |
| Why read it as an auditor? | Because the public authority is the instrument through which much of government and infrastructure now operates, and the book is the clearest study of how its features, dedicated revenue, bond covenants, appointed boards, technical monopoly, can be turned against accountability; and because every mechanism Moses used has a corporate equivalent. |
| Reading time | About fifty hours for the full text; the parts on the authority, on the methods and on the fall, perhaps a third of the book, carry what an auditor needs. |
Who Robert Caro is, and how the book came about
Caro was a reporter at Newsday on Long Island when, in the mid-1960s, he watched the state Assembly vote overwhelmingly for a bridge across Long Island Sound that Moses championed and that Caro’s own reporting had argued against, and he could not explain how. The book was his attempt to find out, and it took seven years, during which he and his wife ran out of money, sold their house, and he feared he would never finish it. Alfred A. Knopf published it in 1974; it won the Pulitzer Prize for biography and the Francis Parkman Prize, and it has never been out of print. Caro went on to write the multi-volume biography of Lyndon Johnson, built on the same question, how power is acquired and used, and the same method, which he has described in a short book of his own, Working. Moses cooperated with Caro for seven interviews and then stopped; the book’s portrait is drawn from his colleagues, his victims, his records and his works.
The structure: seven parts
| Part | What it covers | What it contributes |
|---|---|---|
| I and II. The idealist; The reformer | Moses’s upbringing, Yale and Oxford, his doctoral work on civil service reform and his first defeat by the political machine | The reformer who learned that ideals without power accomplish nothing |
| III. The rise to power | Al Smith, the drafting of the laws that created the state parks and the parkways, the Long Island State Park Commission | How legislative drafting became an instrument of personal power |
| IV. The use of power | The 1920s to the mid-1930s: the Long Island parks and Jones Beach, the city parks department under La Guardia, the Triborough Bridge, the 1934 race for governor and the defeat of Roosevelt over Order Number 129, the methods perfected | The machinery at work |
| V. The love of power | The later 1930s: the West Side Improvement, parks and playgrounds that favoured some neighbourhoods over others, the break with his brother, the Triborough Authority made his base | Power beyond any single office |
| VI. The lust for power | From 1939 into the 1950s: the Brooklyn-Battery crossing fight, Title I urban renewal, the expressways through the Bronx, the displacement of neighbourhoods, the end of public transit investment | The cost, borne by those without a voice |
| VII. The loss of power | From the late 1950s: the press turns, the Rockefeller years, the 1964 World’s Fair, the MTA merger | How the mechanisms were finally defeated |
| Closing chapter, in part VII | Moses after power | The coda |
The rise: how an unelected official acquired power
Moses’s first lesson, and the book’s, was that reform fails without power; his second was that power could be written into law by the person who drafted it. In the 1920s, as Al Smith’s aide, he drafted the legislation creating the Long Island State Park Commission and wrote into it provisions nobody noticed: that the commission could appropriate land by simply declaring it taken, that its powers were defined through cross-references that sent a reader from clause to clause. The parks he built with those powers were real and popular, Jones Beach above all, and the popularity became his protection. When the Triborough Bridge Authority was created in the 1930s to build a single bridge with federal money, Moses drafted its enabling act so that the authority could issue new bonds whenever old ones were retired, could build anything the bonds could finance, and could not be dissolved while bonds were outstanding; the tolls, which were supposed to end when the bridge was paid for, became a permanent revenue stream that answered to no legislature and no mayor. By the 1950s the authority’s surpluses funded a staff, an engineering capability and a patronage network that the city itself could not match, and officials who wanted anything built had to come to him.
The methods, read as governance mechanisms
| Method | What Moses did | The governance mechanism it defeated | The corporate equivalent |
|---|---|---|---|
| Drafting the enabling law | Wrote the statutes creating his positions, with powers buried in cross-references | Legislative oversight; the separation of the rule-maker from the rule-follower | The executive who writes the policy, the delegation and the committee charter |
| Dedicated revenue | Toll income that belonged to the authority, not the treasury, and could be pledged to new bonds indefinitely | Budget control; the power of the purse | A business unit whose revenue is ring-fenced from group oversight |
| Bond covenants | Promises to bondholders that the authority’s structure and tolls could not be changed while bonds were outstanding | The ability of elected officials to restructure or abolish | Contracts and financing terms that lock in a structure beyond the board’s reach |
| Monopoly of expertise | The only engineering staff in the city able to design and deliver large works; the only data on traffic and costs | Independent assessment of proposals | The function or vendor whose figures nobody else can check |
| Control of information | Project costs and plans disclosed when it was too late to stop them; files kept on allies and opponents | Informed decision-making; a free press | The board pack prepared by the people it is supposed to oversee |
| Patronage and retainers | Legal, insurance and contracting work distributed to the politically useful | Procurement integrity; conflict-of-interest rules | Advisers and contractors selected for loyalty rather than value |
| Popularity and the press | Decades of favourable coverage, cultivated personally | Public scrutiny | The executive whose reputation makes challenge look like ingratitude |
| Resignation as a weapon | The threat to resign, used repeatedly against governors and mayors who depended on his delivery | The principal’s ability to replace the agent | The indispensable executive |
The public authority: the instrument
The chapter an auditor of any public body should read is the one on the Triborough Authority, because it explains an organisational form that has since spread across the world. A public authority is created by statute to do a specific thing, usually to build and operate infrastructure, and is given the power to borrow against its own revenues so that the project does not burden the taxpayer. The design has merits: it insulates capital projects from annual budget politics and lets them be financed over their useful lives. Moses saw what else it could do. An authority with revenue of its own, a board appointed for fixed terms, bond covenants that prevent its restructuring and a staff that answers to its chairman is a government that cannot be voted out; and an authority whose enabling act lets it build anything its bonds can finance has no scope limit either. The book’s account of how the authority’s growing surpluses were spent on projects nobody elected had approved, and how its bond covenants were invoked to make the arrangement permanent, is a classic text for the modern concern with the governance of arm’s-length bodies. The corporate governance audit guide treats subsidiaries and special-purpose entities in the same frame: who appoints the board, who controls the revenue, who can wind it up, and what has been promised to lenders that the owner cannot undo.
The fall, and why it took so long
Moses lost power in 1968, at seventy-nine, and the book’s account of how is a case study in the defeat of entrenchment. Three things changed. The press, which had praised him for four decades, turned from the late 1950s as reporters began to examine the displacement his projects caused and the costs that had been hidden; the narrative that had protected him became the narrative that exposed him. The public, which had experienced his parks, began to experience his expressways, and the neighbourhoods he destroyed produced the opposition the parks never had. And a governor, Nelson Rockefeller, with his own ambitions and a family whose bank was the largest holder of the authority’s bonds and agent for the other bondholders, found a way to obtain the bondholders’ consent to the merger that Moses had always said was impossible. The lesson Caro draws is not that the system corrected itself but that it took forty years, a change in public opinion and an adversary with resources comparable to Moses’s own, and that in the meantime the mechanisms worked exactly as designed. An audit function reading the book should note how little any of the formal oversight bodies, the legislature, the city council, the comptroller, the courts, contributed to the outcome.
The nine ideas worth keeping
- Whoever drafts the rules holds the power. The enabling act, the charter, the delegation and the policy are instruments, and the auditor should ask who wrote them and what they contain that nobody noticed.
- Dedicated revenue is autonomy. Money that does not pass through the budget does not pass through oversight.
- Lenders can be a governance lock. Covenants that prevent restructuring protect lenders and entrench management; the board should know what it has promised away.
- Expertise monopolies defeat challenge. Where only one source can produce or check the numbers, there is no check; independent capability is a control.
- Information timing is control. A proposal disclosed when it can still be stopped is subject to oversight; one disclosed when commitments have been made is not.
- Popularity is not accountability. Delivery earns deference, and deference disables the questions that would have found the cost.
- Indispensability is a risk. The official or executive whose resignation would halt the organisation has acquired a veto over its governance.
- The costs fall where the voice is weakest. Moses built where resistance was least; an auditor should ask who bears the costs of the organisation’s decisions and whether they are heard.
- Formal oversight can coexist with its absence. New York had every institution of accountability and none of them worked for forty years; the test is operation, not existence.
The ninth idea deserves emphasis because it is the one most likely to be dismissed. New York in Moses’s decades had a mayor, a council, a state legislature, a governor, a comptroller, courts, a press and an electorate, and every one of them was formally capable of stopping him. The book’s evidence is that none did for forty years, not because the institutions were corrupt but because each had been given a reason not to act: the mayor needed the projects delivered, the legislature had passed the laws without reading them, the comptroller could not audit money that never entered the treasury, the press admired the parks, and the electorate was not asked. An audit function that reports on the existence of governance structures has described New York in 1950. The corporate governance audit guide and the Corporate Governance summary both insist on the operation test for the same reason: the question is never whether the oversight body exists but whether, in the last three years, it has stopped anything.
The vocabulary: fourteen terms defined for practitioners
| Term | What it means in the book | What it looks like in an audit |
|---|---|---|
| Public authority | A statutory body with its own revenue and borrowing power, run by an appointed board | Arm’s-length bodies, subsidiaries, special-purpose entities |
| Enabling act | The statute that creates a body and defines its powers | Charters, delegations and constitutions: who wrote them, what they allow |
| Bond covenant | A promise to lenders restricting what the issuer may do | Financing terms that bind the board |
| Dedicated revenue | Income pledged to a body rather than to the general treasury | Ring-fenced revenue beyond budget oversight |
| Appropriation | Taking land or resources by declaration | Powers exercised without approval |
| Patronage | Distributing positions and contracts for loyalty | Procurement and appointment integrity |
| Retainer | A standing fee to a professional, often for little work | Adviser relationships to be examined |
| Title I | The federal urban renewal programme of 1949 | Programmes whose costs fall on third parties |
| Condemnation | Compulsory purchase | Powers that transfer cost to those without voice |
| Technical monopoly | Sole control of the expertise a decision needs | Independent assessment as a control |
| Indispensability | The claim that only one person can deliver | Key-person risk at the top |
| Oversight body | A legislature, council, comptroller or court | Formal governance; test whether it operates |
| Narrative capture | Press and public opinion aligned with the official | Reputation as a shield against challenge |
| Merger by consent | Restructuring that required the bondholders’ agreement | The exit that financing terms permit |
What it means for internal audit: auditing authorities and the people who run them
The book’s lessons apply directly to the audit of public bodies, authorities, subsidiaries and any entity with its own revenue and an appointed board, and by analogy to the audit of any powerful executive. The corporate governance audit guide, the authority and limits guide and the culture audit guide carry the method.
| Question from the book | Control the audit function should expect | Test that produces evidence |
|---|---|---|
| Who drafted the governing instruments, and what do they allow? | Charters, delegations and policies reviewed by someone other than their author; scope limits stated | Read the enabling instrument for powers not in the summary; identify the author |
| Where does the revenue go, and who oversees it? | All revenue visible to the oversight body; surpluses allocated by decision, not by default | Trace the largest revenue stream to the body that approves its use |
| What has been promised to lenders? | Covenants inventoried and reported to the board with their governance effect | Read the financing agreements for restructuring, scope and governance restrictions |
| Who can check the numbers? | Independent capacity to assess costs, forecasts and technical claims | For the three largest projects, identify who reviewed the estimates independently |
| When does the board learn of commitments? | Approval before commitment; disclosure of options while they are open | Compare the dates of board papers with the dates of contracts and announcements |
| How are advisers and contractors chosen? | Competitive procurement; retainers justified; conflicts declared | Sample the professional-services spend for selection and work performed |
| Who bears the costs, and are they heard? | Impact assessment and consultation for decisions affecting third parties | Read the consultation record for the last major decision; look for the objections and the responses |
| What happens if the leader leaves? | Succession plans; no single person on whom delivery depends | The Phoenix Project summary‘s constraint test, applied to the executive |
For the audit of a powerful executive rather than a body, the book supplies a reading of behaviour that the function should treat as evidence: the policy the executive drafted and interprets, the information that reaches the board when the decision is made, the adviser who has been on retainer for a decade, the public reputation that makes a question look like disloyalty, and the resignation that is threatened when the board hesitates. None of these is wrongdoing. Together they are the pattern the book describes, and the audit committee playbook describes the committee that recognises it.
The criticisms, and what the book leaves out
The book has been argued over for fifty years, and the criticisms are substantial. Historians and planners have shown that Caro overstated Moses’s uniqueness: highway building, slum clearance and the neglect of transit were national policy, funded by federal programmes, and other cities did the same without a Moses. The portrait of the man is relentlessly dark, and later scholarship, including the 2007 exhibition and volume Robert Moses and the Modern City, argued that his works have aged better than Caro allowed and that the city he left was more functional than the one the book implies. Some specific claims, including the much-cited story that he built parkway bridges low to keep buses and their poorer passengers from the beaches, have been questioned. And the book is long: Caro’s method produces detail that a reader with a purpose must select from. For an auditor the criticisms do not touch the part that matters, which is the mechanisms. Whether Moses was worse than his era or merely its most effective practitioner, the account of how an authority’s revenue, covenants, expertise and information were turned into unaccountable power is accurate, documented and unrefuted, and it is why the book is assigned in public administration and governance courses to this day.
Editions, reading time and how to read it
| What | Details | Best for |
|---|---|---|
| Hardback | Alfred A. Knopf, 1974, 1,246 pages | The original; heavy |
| Paperback | Vintage, 1975, and continuously in print | The edition most readers own |
| Audiobook | Unabridged, about sixty-six hours | For the committed; the parts on the authority and the fall can be sampled |
| Reading plan | The chapters on the Long Island parks legislation, the Triborough Authority’s enabling act and finances, the methods chapters of part four, one Title I case, and the MTA merger | An auditor with a week of evenings |
| Companion reading | Working, Caro’s short book on his method; Robert Moses and the Modern City (2007) for the revisionist case; the Corporate Governance summary for the theory of accountability | Placing the argument |
A one-hour discussion guide for an audit team
- Ten minutes: the instruments. Name the organisation’s charters, delegations and policies and who drafted each. Discuss whether anyone outside the author has read them for what they permit.
- Fifteen minutes: the revenue and the covenants. Identify any revenue stream or entity that sits outside the main budget process, and any financing term that restricts what the board can change.
- Ten minutes: the monopoly. Name the figures the organisation relies on that only one function or vendor can produce or check.
- Fifteen minutes: the pattern. Discuss, without naming anyone, whether any of the book’s behavioural patterns, drafted rules, information timing, retainers, reputation, indispensability, appear in the organisation, and whether the function has ever tested them.
- Ten minutes: the plan. Decide whether an authority, subsidiary or arm’s-length body in the group deserves a governance review on the book’s questions, and who would sponsor it.
Worked example: Millbrook Regional Water Authority asks the Moses questions
Millbrook Regional Water Authority is a fictional statutory body serving a metropolitan area of 1.4 million people, financed by rates and revenue bonds, governed by a board appointed by the region’s municipalities, with a long-serving executive director and an internal audit function of four. In 2027 the board’s new audit committee chair, who had read the book, asked the function to review the authority’s governance against its questions.
| Moses question | What the function found | Result |
|---|---|---|
| Who drafted the governing instruments? | The authority’s procurement policy, delegation schedule and capital planning procedure had all been written by the executive director’s office and approved by the board in a single session each; the delegation allowed capital commitments up to $25 million without board approval | Finding: delegation reduced and tiered; policies to be reviewed by the committee before approval |
| Where does the revenue go? | Rate revenue passed through the board’s budget; a separate connection-fee fund of $38 million had accumulated outside it and funded projects at the executive director’s discretion | Finding: the fund brought into the budget; its projects reported to the board |
| What has been promised to lenders? | The 2019 revenue bond indenture required rate increases sufficient to maintain coverage and prohibited the board from reducing rates below that level; the board had not been told | Observation: a covenant summary added to the board’s annual papers |
| Who can check the numbers? | Capital cost estimates were produced by the authority’s engineering department and reviewed by a consulting firm on retainer for fourteen years | Finding: independent review rotated and competitively procured |
| When does the board learn of commitments? | Two of the last five major contracts had been announced publicly before the board paper was circulated | Finding: approval before announcement |
| How are advisers chosen? | Legal, bond counsel and engineering retainers had not been tendered in a decade | Finding: tender cycle established |
| Who bears the costs? | A reservoir expansion had displaced a rural community whose objections appeared in the consultation file and nowhere in the board papers | Observation: consultation summaries to the board, including objections |
| What happens if the leader leaves? | No succession plan; the executive director had threatened resignation twice in board minutes | Finding: succession planning; the committee chair’s note that the threat itself was a governance signal |
The review produced six findings and two observations, in an authority that had never had a qualified audit, a regulatory finding or a scandal, and the committee chair’s report to the board observed that the book’s point was exactly that: none of the mechanisms required wrongdoing, each had a reasonable origin, and together they had placed most of the authority’s decisions beyond the board’s reach. The executive director, to his credit, implemented the findings; the board, to its, read the book.
Frequently asked questions
Is The Power Broker worth reading in full?
For a reader interested in power, cities or biography, yes; it is one of the great American books. An auditor with a purpose can read the parts on the authority, the methods and the fall, about a third of it, and get what the profession needs.
What is the governance lesson in one sentence?
That a body with its own revenue, lender covenants that prevent restructuring, a monopoly of expertise and control over when the board learns of commitments can be unaccountable in practice while every formal oversight institution exists, and that the test of governance is operation, not existence.
Does it apply to companies?
By analogy, closely: ring-fenced business units, special-purpose entities, financing terms that bind the board, executives who draft the policies they operate under, advisers on long retainers and indispensable leaders are the corporate forms of the book’s mechanisms.
How accurate is the book?
Its account of the mechanisms is documented and unrefuted; its judgment of Moses and of his era has been contested by historians and planners, who argue that highway building and urban renewal were national policy. Read the mechanisms as settled and the verdict as argued.
What should an auditor of a public body do with it?
Run the eight questions in the audit map against the body’s governing instruments, revenue, financing terms, expertise, information timing, advisers, affected parties and succession. The worked example shows what a small function found in an afternoon of reading and a month of testing.
Where does the book sit in the reading list?
In the governance group of the internal audit reading list, as the case study of unaccountable power beside the textbook on accountability.
Related guides
- How to Audit Corporate Governance — subsidiaries, authorities and the people who run them
- What Internal Audit Can and Cannot Do: Authority, Access and Limits — the function’s own charter
- The Audit Committee Playbook — the committee that recognises the pattern
- Corporate Governance by Monks and Minow: Summary for Auditors — the theory of accountability
- Barbarians at the Gate: Summarised for Auditors — managers against the board, in a company
- How to Audit Corporate Culture — reputation and deference as cultural signals
- The Best Books for Internal Auditors — where this book sits among the others
- All book summaries — the full set of summaries for auditors
Leave a Reply