The Forging of Bureaucratic Autonomy: Reputations, Networks, and Policy Innovation in Executive Agencies, 1862-1928

by Daniel Carpenter

Overview

Until now political scientists have devoted little attention to the origins of American bureaucracy and the relationship between bureaucratic and interest group politics. In this pioneering book, Daniel Carpenter contributes to our understanding of institutions by presenting a unified study of bureaucratic autonomy in democratic regimes. He focuses on the emergence of bureaucratic policy innovation in the United States during the Progressive Era, asking why the Post Office Department and the Department of Agriculture became politically independent authors of new policy and why the Interior Department did not. To explain these developments, Carpenter offers a new theory of bureaucratic autonomy grounded in organization theory, rational choice models, and network concepts.


According to the author, bureaucracies with unique goals achieve autonomy when their middle-level officials establish reputations among diverse coalitions for effectively providing unique services. These coalitions enable agencies to resist political control and make it costly for politicians to ignore the agencies' ideas. Carpenter assesses his argument through a highly innovative combination of historical narratives, statistical analyses, counterfactuals, and carefully structured policy comparisons. Along the way, he reinterprets the rise of national food and drug regulation, Comstockery and the Progressive anti-vice movement, the emergence of American conservation policy, the ascent of the farm lobby, the creation of postal savings banks and free rural mail delivery, and even the congressional Cannon Revolt of 1910.

Product Details

ISBN-13:	9780691070100
Publisher:	Princeton University Press
Publication date:	09/09/2001
Series:	Princeton Studies in American Politics: Historical, International, and Comparative Perspectives , #78
Edition description:	New Edition
Pages:	504
Product dimensions:	6.00(w) x 9.25(h) x (d)

About the Author

Daniel P. Carpenter is Assistant Professor of Political Science at the University of Michigan. He has also taught at Princeton University and the University of Chicago. This book is based upon his dissertation, which won the 1998 Harold Lasswell Award of the American Political Science Association, and includes a chapter that won the 1995 Herbert Kaufman Award of the APSA.

Read an Excerpt

COPYRIGHT NOTICE: Published by Princeton University Press and copyrighted, © 2001, by Princeton University Press. All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from the publisher, except for reading and browsing via the World Wide Web. Users are not permitted to mount this file on any network servers.

Introduction

"UNCLE JOE" Cannon was fuming.

It was 1905, and Cannon, the Republican Speaker of the U.S. House of Representatives and arguably the nation's most powerful politician, had just been beaten by a bureaucrat. Gifford Pinchot, chief forester in the U.S. Department of Agriculture, had quietly and methodically persuaded Cannon's House colleagues that the nation's forest reserves should be transferred from the Department of Interior to his own Agriculture Department. Pinchot's aim was to put the reserves under a much tighter regimen than Cannon and his Republican allies would allow, introducing user fees and grazing restrictions. Just two years earlier Cannon had taken control of the House floor to denounce the transfer scheme, tarring Pinchot's bureau as a "cheat and a fraud." But now Pinchot had turned the tables. By leaning on his friends in professional and scientific circles and by massively publicizing his bureau's accomplishments, he had convinced much of the nation's press that his organization possessed unique and unparalleled expertise on forestry matters. Moreover, Pinchot had built himself a coalition. Hehad wedded numerous groups together in his transfer crusade, including the General Federation of Women's Clubs, the Audubon Society, the Sierra Club, western ranchers' organizations, the National Board of Trade, and even pivotal members of the House Public Lands Committee who just two years earlier had applauded Cannon's tirade and had voted with the Speaker. Such diverse groups had little in common before and after 1905, but that year they were united in their zeal for land regulation.

Cannon's anger at Pinchot was hardly misplaced. Republicans (especially in the West) would soon rue the day that Pinchot's bureau took over the reserves. Pinchot imposed numerous user fees on logging, mining, and ranching interests, fees that were never contemplated in the 1905 act. Cannon's allies angrily denounced Pinchot and his power as a lawmaker. "History is challenged to instance anything," cried one, "approaching such audacious Departmental assumption of power in a legislative capacity." One senator marveled at how Pinchot had gotten the "right to legislate as to how lands shall be preserved." Another politician complained of Pinchot's hold over public opinion, denouncing the "publicity machine" he had erected in the Forest Service, which mailed out more than 9 million circulars annually. "Congress usually undertakes to ascertain what the people want and legislate accordingly," he grumbled. "Pinchot reverses the proceeding." Against the publicly recognized expertise and political clout of Pinchot's bureau, Congress could do little but stand by and watch. Thus, one of the central acts of American environmental history came from the political entrepreneurship of a bureaucrat.

It was not the last time that Speaker Cannon would find himself outflanked. A year after Pinchot's triumph, Harvey Wiley, chief of the USDA's Chemistry Bureau, successfully completed a twenty-year campaign for a national pure food and drug law. Wiley's bill had also been voted down by Republican Congresses in the late 1890s and early 1900s, again with opposition from Cannon. In 1906, however, Wiley's bureau reversed this pattern and recorded a political triumph that probably outranks Pinchot's. Like the chief forester, Wiley and his associates deluged Congress with data from his studies and lectured around the country. By 1905, in fact, "Dr. Wiley" was a household name in America. Yet Wiley's coalition was much larger and more varied than Pinchot's, and it stands as perhaps the first national "consumer protection" coalition in American political history. Backing Wiley's bill was a stunningly diverse league of more than one hundred organizations, including the General Federation of Women's Clubs, the American Medical Association, the Grange, the Women's Christian Temperance Union, a new coalition of state officials, the Association of Official Analytic Chemists, and the National Board of Trade. With this broad coalition, Wiley and his USDA associates turned to the legislative process, where they broke decisively from nineteenth-century precedent by writing the 1906 act. Cannon and other opponents saw their allies drop away, one by one, into a numbing approval of the Wiley bill. After the act passed, the constituents of Wiley's coalition defended his bureau at every turn, beating back attempts to restrict the USDA's range of activity. Like the 1905 Transfer Act, the Pure Food and Drugs Act of 1906 bears an immense institutional legacy. It authorized the regulation of food and pharmaceutical products now carried out by one of the nation's most powerful federal agencies, the Food and Drug Administration (FDA).

The Department of Agriculture was not the only federal agency to forge new policies in this way. The Post Office—that supposedly tradition-bound, patronage-dripping behemoth—authored immense shifts in national policy from the 1870s to the First World War. Postal officials were the prime movers behind the "Comstock" anti-pornography law of 1873 and the Anti-Lottery Law of 1890. Under Anthony Comstock and his associates, the Post Office became the most feared and powerful moral police agency the nation had yet known. With a varied collection of moral reform groups backing him—prohibitionists in the Anti-Saloon League, social elites in vice-suppression societies, and abortion opponents in the American Medical Association—Comstock became a legislative force unto himself, pressing numerous extensions of his authority through Congress. When President Theodore Roosevelt and congressional Republicans tried to turn Comstock's force on their political opponents after 1905, postal officials brazenly refused, wielding their own power on their own terms.

The Post Office also inaugurated the rural free delivery system, which began in 1891, the Postal Savings Act of 1910, and the parcels post plan of 1912. In the early 1890s Congress had initially balked at these ideas, citing the department's high operations deficit at the time. As postal officials reduced the deficit, and as the department fortified its ties to the media, Progressive moralists, agrarians, and business interests, the momentum for policy change grew. In postal savings, the department capitalized on the Panic of 1907 and European immigrants' distrust of private banks, framing its campaign in terms of the economic virtue of increased savings for the lower classes. When opposition arose, the Post Office joined with Progressive welfare advocates, journalists, and organized agrarians to beat back the attempts of bankers' associations to limit the scope of its discretion. From 1910 onward, the department operated savings banks out of virtually every post office with money order services. Postal officials also geared postal savings institutions toward European immigrants, gaining their participation and cementing legitimacy among a growing portion of the American electorate. In the parcels post system, established two years later, Congress gave the postmaster general unprecedented pricing authority over package delivery. By 1913 the department was able to make deep inroads into markets once dominated entirely by private enterprise.

Although the Post Office and Agriculture Departments triumphed at bureaucratic policymaking, the Interior Department remained an enigmatic failure. Given hold over vast reserves of the public domain and empowered by the Arrears Act of 1879 to distribute billions of dollars in pensions to Civil War veterans, the Interior Department seemed better equipped for autonomous policy-making than any agency of the time. Its policies had beneficiaries across classes, of both sexes, and in every state. Yet for all its formal authority, the Interior Department was perhaps the most politically feeble agency in American government before the New Deal. In managing Civil War pensions, the Interior Department was confined to do the bidding of congressional Republicans and the Grand Army of the Republic. Department proposals to centralize the pension system met, like its other ideas, with silence. The Reclamation Act of 1902—which gave the department more than $100 million in discretionary spending to develop arid lands in the West—promised to reenergize the Interior Department. Yet just a decade after its creation, the department's Reclamation Service had squandered its funds, forcing Congress to subject all of its projects to external review and tight appropriations control. Forest policy, meanwhile, marked Interior's truly crowning embarrassment. The department had been the principal regulator of the public domain in the nineteenth century, but in 1905 the Interior Department lost control of the national forests to the USDA and Pinchot.

Why Bureaucratic Autonomy?

Narratives of this sort—bureaucrats building reputations for their agencies, erecting coalitions behind their favored policies, and securing the policies that they favor despite the opposition of the most powerful politicians—are not isolated occurrences in American political development. Stories like these abound in the Progressive Era. Such cases raise important questions of democratic governance, not least the specter of unelected officials with broad policymaking power. Yet the primary query that these narratives raise is one of empirical bewilderment: Why and how does such autonomy happen? Why are some agencies autonomous, whereas others lie dormant? Why did the USDA and the Post Office Department succeed where the Interior Department failed? Why did the USDA and Post Office Department succeed in the Progressive Era and not before?

Bureaucratic autonomy occurs when bureaucrats take actions consistent with their own wishes, actions to which politicians and organized interests defer even though they would prefer that other actions (or no action at all) be taken. (A more extended definition appears in chapter 1.) Bureaucratic autonomy so defined is a common feature, though far from a universal one, of American government in the twentieth century. Agencies have at times created and developed policy with few, if any, constraints from legislative and executive overseers, and they have frequently coordinated organized interests as much as responded to them. To suggest that bureaucracies have policymaking autonomy may strike some readers as a controversial if not outlandish claim. Surely agencies lack the ability to take any action they desire in our system of representative government and rule of law. Yet I contend here that bureaucratic autonomy lies less in fiat than in leverage. Autonomy prevails when agencies can establish political legitimacy—a reputation for expertise, efficiency, or moral protection and a uniquely diverse complex of ties to organized interests and the media—and induce politicians to defer to the wishes of the agency even when they prefer otherwise. Under these conditions, politicians grant agency officials free rein in program building. They stand by while agency officials do away with some of their cherished programs and services. They even welcome agencies in shaping legislation itself.

Reputations, Networks, and Autonomy

A bureaucracy is an organization, and its autonomy (or lack of it) is premised on its organizational reputation and the networks that support it. When bureaucracies in turn-of-the-century American politics gained a lasting esteem for their ability to provide unique services, author new solutions to troubling national dilemmas, operate with newfound efficiency, or offer special protection to the public from economic, social, and even moral hazards, bureaucratic autonomy usually followed. The reputations that autonomous agencies established were diverse. Some agencies became known for their ability to conserve the nation's natural resources. Others were celebrated for protecting American consumers from the hazards of adulterated food and medicines. The Post Office earned esteem for moral guardianship (a powerful role in the culturally conservative Progressive Era), that is, for shielding American families from the evils of pornography and gambling. Whatever their specific content, organizational reputations had two common traits. First, they were grounded in actual organizational capacity. Agencies with strong reputations possessed greater talent, cohesion, and efficiency than agencies with reputations for weakness, corruption, or malfeasance. Second, organizational reputations were not ethereal but socially rooted. They were grounded in diverse political affiliations maintained by career bureaucratic officials. Reputations that were embedded in multiple networks gave agency officials an independence from politicians, allowing them to build manifold coalitions around their favored programs and innovations.

Two features of bureaucratic legitimacy—reputational uniqueness and political multiplicity—are crucial in the pages that follow. Autonomous agencies must demonstrate uniqueness and show that they can create solutions and provide services found nowhere else in the polity. If politicians can easily find compelling policy alternatives to an agency's plans, then agency autonomy will not be stable. Autonomous agencies also have a legitimacy that is grounded—not among the voters of one party or one section, not in a single class or interest group, but in multiple and diverse political affiliations. Agencies are able to innovate freely only when they can marshal the varied forces of American politics into coalitions, coalitions that are unique and irreducible to lines of party, class, or parochial interest. Network-based reputations as such are the very essence of state legitimacy in modern representative regimes.

Autonomy and the State: A New Look at American Political Institutions

The phenomenon of autonomy is fascinating not merely for academic purposes but also because the form of autonomy I consider in this book—independent policymaking power—has forcefully shaped the political institutions of the twentieth-century United States. Chief among these is what is called (with affection or disgust) the modern state. The advance of the bureaucratic state is one of the most wrenching and controversial changes of the twentieth century. However considered—by its capacity for war-making and policing, by its regulation of commerce and private life, by the enormity of its public expenditure—our national bureaucracy has grown ever more formidable and complex. Yet there is an entirely different characteristic of our modern state, one that heralds its arrival more than any other, one unique to the last century, for nothing so distinguishes twentieth-century bureaucratic government from its predecessors as its ability to plan, to innovate, and to author policy.

Bureaucratic policymaking is the hallmark of modern American government. Our national bureaucracies routinely collect and analyze information, and they systematically forecast economic and social outcomes. Our agencies write regulations and draft legislation based on this information. At times, they apply influential political pressure for the passage of the laws they draft. They administer with considerable discretion the resulting rules and statutes, in accordance with their own standardized routines and procedures. This pattern holds sway in programs ranging from pharmaceutical regulation to defense procurement, from highway construction to welfare assistance, from counterterrorism to grant allocation. We may celebrate or bemoan the expanded role of our bureaucracies, but the story remains the same. The brute fact of modern politics is that myriad national programs begin and end in the hands of federal agencies.

The advance of the policy state is a narrative of organizational evolution and bureaucratic entrepreneurship. Operating within the rigid confines of the American institutional order—the primacy of elected officials, the constraints of American political culture, and the dominance of parties—administrative leaders in the USDA and the Post Office Department slowly carved out pockets of limited discretion by starting small experimental programs. By nurturing local constituencies and by using their multiple network affiliations to build broad support coalitions among professionals, agrarians, women's groups, moral crusaders, and congressional and partisan elites, they won for their young programs both political currency and administrative legitimacy. Fledgling experiments with dubious survival prospects at the turn of the century became, by the close of the 1920s, established policies. At almost every step in the development of these programs, the institutional authorities of the American order—Congress, the president, the parties, the courts, and organized interests—assented to greater and greater administrative innovation. Through reputation building, federal agencies won the capacity to innovate. In American political development, bureaucratic autonomy was not captured but earned.

This argument joins a time-honored dialogue on the evolution of American bureaucracy, a dialogue enriched by scholars as diverse as Stephen Skowronek, Theda Skocpol, Richard Bensel, Terry Moe, Richard John, Martin Shefter, Elizabeth Sanders, Samuel Kernell, Ronald Johnson and Gary Libecap, Scott James and Brian Balogh. My argument also departs subtly but significantly from these accounts. In its focus on bureaucratic development in "pockets" of the American state—and with its analysis of the varied evolution of capacity and reputation across agencies and over time—this book offers a different view of institutional change from that advanced by Skowronek.¹ An emphasis on organizational reputations in addition to organizational capacities distinguishes this narrative from the capacity-based arguments of Skocpol, John, and Kenneth Finegold.² With its argument that strong states are not simply embedded in society but must have durable reputations that are seated in multiple and diverse networks, this book diverges from numerous accounts that connect American state formation to sectional conflict and social movements (Bensel and Sanders), partisan coalitions (Shefter, Kernell, and James), organized federal employees (Johnson and Libecap), professions (Balogh), presidents (Moe, James), or some combination of these forces (Moe).³

Sovereign Stories of the Policy State

In important and novel ways, this story differs from the two sovereign narratives of American state building. For casual observers and scholars alike, the first story is very simple. According to received historical wisdom, the policy state we know today in America is a legacy of the New Deal. In response to the mass economic crisis of the Great Depression and unified under Democratic rule, the government launched a wide array of programs designed to counter the disastrous consequences of the business cycle and to shore up an aging national infrastructure. To stabilize industrial production, President Franklin D. Roosevelt offered the National Recovery Act, the most far-reaching attempt at government-coordinated industrial regulation our nation has known. To boost employment and build public works projects, Roosevelt created the Works Progress Administration and the Civilian Conservation Corps. To ease farmers out of the price depression of the 1930s, Roosevelt tendered a massive price support program in the Agricultural Adjustment Act. And to solidify the retirement income of elderly Americans, the government offered the first entitlement-based guaranteed income program, Social Security. In addition, Roosevelt and the Democrats created and extended numerous regulatory agencies responsible for moderating industries as diverse as banking, communications, and pharmaceuticals. Floating massive programs and creating agency upon agency from scratch, the story goes, Roosevelt and the New Deal Democrats launched the discretionary bureaucratic state we know today.⁴

As a story of the accretion of bureaucratic power, the New Deal narrative is persuasive. As an account of the origins of the modern policy state, however, it suffers from two critical flaws. The first is that many New Deal creations lacked institutional permanence. Social Security, much of the Agricultural Adjustment Act, and several New Deal regulatory agencies endure today, but few of the programs and bureaucracies created in the New Deal survived to 1950. This is particularly true for the more ambitious attempts to plan and coordinate economic production or mass public construction—the National Recovery Act and the public works programs.

The genuine poverty of the New Deal narrative, however, consists in its lack of accuracy and nuance. In America, our bureaucratic state did not evolve all of a piece. Some bureaucracies developed the capacity to innovate and plan decades before 1933, whereas many agencies that were created anew in the New Deal never exhibited this ability. Consider, for instance, the Agricultural Adjustment Act, the foundation of twentieth-century farm policy. The durability of this act, as Finegold and Skocpol have shown, was due largely to the exceptional administrative and organizational capacities of the U.S. Department of Agriculture in the 1930s. Yet few factors in New Deal politics can account for why the USDA developed its organizational capacity before 1930. Alternatively, the demise of the National Recovery Act may be traced in large measure to the inability of the agency implementing it—the National Recovery Administration—to achieve administrative stability. What requires explanation is why one agency was possessed with greater capacity and political legitimacy than the other. Stark differences across agencies such as these abound in American administrative history.

Others have alighted on the fact that the critical steps in American state building took place before the Depression. During the last two decades, numerous scholars—Skowronek, Skocpol, Bensel, Sanders, and James—have hewn out a rough consensus over the proposition that between the end of the Civil War and the New Deal, a national state was constructed in America that was genuinely distinct from the feeble structure of the nineteenth century. To the federal government there accrued a variety of new formal powers ranging from taxation of income to regulation of railroads and foods to the programmatic delivery of benefits to veterans and poor mothers. The American military was transformed from a loose, federated collection of state militia into a centralized bureaucratic machine. Reform movements and political leaders struggled and converged to overhaul the prevailing mode of governmental operations, placing the civil service under a merit standard and transforming the federal budget process. And, of course, the volume and expanse of national governmental operations exploded.⁵

Three features are common to these Progressive narratives, features that also mark their critical flaws. First, these stories focus principally on the rise of a new creature of bureaucracy: the independent commission. In the Interstate Commerce Commission (the ICC, created in 1887) and the Federal Trade Commission (the FTC, created in 1914), scholars of the Progressive narrative see the genuine state-building achievement of the early twentieth century. As Skowronek proclaims, the ICC

emerged in 1920 as the signal triumph of the Progressive reconstitution. Here, the old mode of governmental operations was most completely superseded, and the reintegration of the American state with the new industrial society most clearly consummated. The agency . . . acquired the responsibility for supervising all aspects of the national railway system in accordance with the most advanced precepts of scientific management It was all the promises of the new American state rolled into the expansion of national administrative capacities.⁶

The independent commissions marked a departure from the large-scale departments of the nineteenth century. In theory, at least, they smacked less of bureaucracy and more of expertise. Born after the Pendleton Act of 1883, they were insulated from patronage, freed in principle from party control. In landmark pieces of legislation, they were charged with governing vast segments of the American economy—from the transportation sector (under the ICC) to antitrust and fair industrial competition (the FTC). And they were a uniquely American state-building achievement. In a literature still wedded to American "exceptionalism," the fact that the independent commissions were not institutional copies of the European state but homegrown remedies to industrialization has been their most powerful lure.

Although certainly worthy of the immense historical attention they have received, the independent commissions figured little in the development of the American policy state. First, commissions have occupied a relatively diminutive place in twentieth-century state formation. Over the last one hundred years, the vast share of nonmilitary state activity, government employment, and public expenditure in the United States has been borne by executive departments.⁷ This brute fact of administrative politics was no less true in the Progressive Era and the New Deal. Compared with the impact of executive departments—the regulation of American trade and financial markets by the Treasury Department; the governance of agriculture, pharmaceuticals, and forests by the Department of Agriculture; the management of western lands, Indian affairs, pensions, and patent law by the Interior Department; the persistence of a multibillion dollar economy in the Post Office; voting rights and antitrust enforcement in the Department of Justice; and numerous activities in the Departments of Commerce and Labor—the political, economic, and social implications of ICC and FTC activity in the early twentieth century were small.

Beyond this, the Progressive commissions never exhibited the core properties of the twentieth-century policy state—planning and discretionary innovation. As historians have repeatedly shown, neither commission was capable of sustained policy creation. Both the ICC and FTC took their orders from the dominant Progressive and Republican coalitions in Congress. And the ICC was particularly ill equipped to carry out the aims of the Progressives. Where new programs required it to plan or to innovate, the commission failed miserably. In 1913 Congress passed the Valuation Act, charging the ICC with a massive study of railroad assets and liabilities. The act doubled the size of the commission, but it only exposed its feeble organizational structure and lack of analytic talent. Although Progressive politicians expected completion of the valuation project in a decade, the crudest asset estimates were not available until 1933. An even greater failure of planning followed upon the Transportation Act of 1920, Skowronek's "signal triumph" of Progressive state building. The Transportation Act charged the ICC with setting forth a broad national plan for railroad consolidations. The commission began consolidation planning in May 1920 but did not adopt a plan until 1929. So hapless was its administrative organization that for three successive years, from 1926 to 1928, the commission appealed to Congress to remove the consolidation burden from its shoulders. Two students of the commission summarize its experience with planning in the 1920s.

Having originally programmed the ICC to reflect congressional bidding, Congress gave the commission authority to determine policy, but the ICC was unwilling to innovate bold plans or to generate power to support its policies. To solve the railroad problem and provide a rational transportation system, the ICC needed to plan, shape, innovate and act, but it continued merely to reflect power and respond to pressure from other sources.⁸

A second form of the Progressive narrative emerges in the work of Skocpol, who has paid the most scholarly attention to state-building developments outside the commissions. Skocpol's Protecting Soldiers and Mothers details the evolution of early forms of social provision in the United States. The explosion of Civil War pensions after Reconstruction provided a far-reaching system of social distribution. Later, the diffusion of mothers' pensions across the states and the creation of the Sheppard-Towner program constituted the earliest forms of governmental aid to single mothers. Arguing from a "polity-centered" perspective, Skocpol credits party structure, reformist professionals, and gender-driven, locally rooted women's movements for forging early American social policy.

Still, Skocpol herself admits that the promise of Progressive social policy lay more in its historical possibilities as a social program than in its bureaucratic achievements. Bureaucratic planning and agency entrepreneurship play a lesser role in Skocpol's narrative than do the coalition strategies of party officials and far-flung women's organizations. As successful as Civil War pensions were, they never engendered any sustained administrative innovation in the Interior Department's Pension Bureau. Instead, pension administration further debilitated the Interior Department from a state-building standpoint, as I discuss in chapter 2. Nor did mothers'pensions inaugurate bureaucratic planning in American social policy. In summary, neither veterans' benefits nor mothers' pensions added materially to the organizational capacities of American bureaucracy.

The third and by far the most prevalent misperception among state-building narratives is that civil service reform was a historically sufficient condition for the emergence of unique bureaucratic preferences. In contrast to Bensel and Skowronek,⁹ I argue that civil service professionalization through the construction of a merit-based system of hiring and promotion does not suffice to render the state autonomous of parties, legislative coalitions, and interest formations. Instead, the development of statist identities in the federal bureaucracy evolved agency by agency, often through network-based hiring that was specific to particular agencies. Skowronek's narrative itself demonstrates the limits of merit reform, insofar as genuine enforcement of the Pendleton Act was completed only from 1908 to 1920, a quarter-century after its enactment in 1883.¹⁰ But the primary reasons for the insufficiency of merit reform lie in well-established arguments about how bureaucracies can be captured by organized interests or dominated by legislatures, for the Pendleton Act did not suppress the opportunities for organized interests to influence agencies by shaping the character of appointments to positions of political leadership in them, appointments made by the president and confirmed by Congress. Nor were agencies that were released from the strictures of patronage any less likely to witness their discretion constrained in acts of federal courts and in rigid legislative specification, as Skowronek's "reconstitution" of economic regulation demonstrates.¹¹

Unpacking the Politics of Bureaucracy

With rare exceptions, analyses of American state building share one other flaw. They study bureaucracy only through the legislation that creates agencies, the presidents who govern them, or the court decisions that check or enable their decision making. Skowronek's narratives of state building place presidents (particularly Theodore Roosevelt) at the center of the story, presaging his landmark book on executive power, The Politics Presidents Make (1992). Bensel, Sanders, and James study bureaucratic development through the passage of laws, focusing their analyses primarily on roll-call votes. Although administrative agencies undoubtedly occupy a smaller place in American political institutions than in other nations, the neglect of bureaucratic organizations in studies of administrative development is unfortunate. First, it reduces political development to institutional creation, to the neglect of institutional transformation. Second, it leaves the most important political outcomes—the impact of policies on citizens—unstudied. Only by focusing on administrative outcomes can transformations in the relationship between state and society be properly analyzed. As the narratives in this book show repeatedly, legislation cannot hard-wire administrative outcomes. Scholars must examine the ongoing relationship between politicians, citizens, and bureaucrats.

From the vantage point of ease of analysis, the neglect of bureaucracy in American political development is easy to understand. Congressional debates, roll-call votes, presidential biographies, and court decisions are readily available to researchers in American politics. Administrative documents are not. Whereas scholars have made rich and ingenious use of congressional and executive materials and refer occasionally to agencies' annual reports, they have almost uniformly ignored other primary sources: agency memorandums, records of administrative decisions, civil service records, and correspondence between agencies, members of Congress, and the president.

I hope that this book eschews the limitations of earlier analyses without neglecting the important role of politicians and courts. I have tried to unpack American bureaucracy by collecting more than fifty thousand pages of primary-source documents, including data on more than seven thousand civil servants and hundreds of official government prints. I have also consulted a rich secondary literature on the development of these three executive departments. In addition, I have incorporated the reports and archival files of the House and Senate committees that oversaw these agencies, as well as the papers of influential members of Congress, of presidents, and of several state archives where the relationship between the executive departments and state-level institutions was illuminated. I admit that the research here does not exhibit the exhaustive specificity of particular studies such as Richard John's treatment of the antebellum postal system, Donald Pisani's analyses of western reclamation, or James Harvey Young's history of the Pure Food and Drugs Act of 1906 (all cited in later chapters). As a comparative study of agencies, however, I hope that it breaks new ground and offers at least a partial template for other social scientists.

Why the Post Office, Agriculture, and Interior?

I chose the three departments I examine in this book with a view to illuminating the dynamics of bureaucratic autonomy in domestic policy arenas. Each of these agencies was created at least a generation before 1900 and was neither disassembled nor merged during the period under study. Although all three organizations experienced fargoing transformations between 1860 and 1930, they kept their titles and a core set of functions. I also include short discussions of the Department of Treasury and the Department of Commerce and Labor (later split into separate departments, as they exist today) in various chapters.

Readers will also note that this analysis excludes the military and the Department of State. I leave these agencies to other scholars, principally for the reason that foreign policy introduces constitutional issues that would complicate my theoretical purposes. Because the president carries greater institutional authority in foreign policy under our Constitution, the relative autonomy of military and diplomatic agencies from politicians is an altogether different matter from its status in domestic agencies. Because the bureaucratic structure of the American military changed materially from the Civil War to the New Deal, moreover, it would be difficult to analyze the military as a single executive department. In addition, a focus on domestic agencies narrows the factors that can explain the divergent historical outcomes observed across agencies and over time. In this sense, my exclusion of military and diplomatic agencies amounts to a form of historical control. For a similar reason, my analysis of the Department of Interior excludes its management of Native American populations, a fascinating episode in bureaucratic policymaking, which I leave to other scholars.

What the three agencies I analyze here do offer are differences in tasks and structure which render their divergent paths of evolution all the more intriguing. Both the USDA and the Post Office, I claim, developed a degree of relative autonomy from politicians and organized interests. Yet these agencies had little in common that differentiated them from the Interior Department. The Department of Agriculture was the dominant scientific agency of the federal government; postal officials exhibited few scientific or professional credentials of note. The USDA was the official representative organ of farmers in the executive branch, though as I show, the department's support from organized agrarians was weaker than has commonly been supposed. Meanwhile, the vast share of postal employment was in cities, and the most earnest lobbyists of the department were moral reformers, merchants, and railroads. If anything, the Interior Department's strong embedment in the West and in rural communities should have disposed it to receiving greater favor from rural interests than it did. Nor can agency size or patronage explain the outcomes. The Agriculture Department continued through the Progressive Era as among the smallest of executive departments, dwarfed by the Post Office, Interior, and Treasury.

Only bureaucratic legitimacy—the evolving belief among politicians and the organized public in the problem-solving capacities of a select few agencies—explains the evolution of conditional autonomy in the Post Office and Agriculture Departments. Rooted in the arrival of a new generation of officials with novel talents and a distinct ideological view of the proper role of their bureaucracies, bureaucratic legitimacy crystallized in the Progressive Era and was cemented during the the First World War and the 1920s. I offer a theoretical account of this process in chapter 1, and in chapter 2 I explain why it did not arise for most departments, especially the Department of the Interior. I organize the remaining eight chapters into three distinct sections. In chapters 3, 4, and 5, I narrate the rise of bureaucratic autonomy in the Post Office Department. In chapters 6 through 9, I discuss the trajectory of autonomy in the USDA. In chapter 10 I show how the Interior Department lost wide-ranging bureaucratic discretion in federal reclamation policy. I offer a concluding note on the political nature of bureaucratic autonomy and consider the meaning of these patterns for the larger interpretation of Progressivism.

Table of Contents

List of Illustrations ix

List of Tables xi

Acknowledgments xiii

Abbreviations xv

Introduction 1

One: Entrepreneurship, Networked Legitimacy, and Autonomy 14

Two: The Clerical State: Obstacles to Bureaucratic Autonomy

in Nineteenth-Century America 37

Three: The Railway Mail, Comstockery, and the Waning of the Old Postal Regime, 1862-94 65

Four: Organizational Renewal and Policy Innovation in the National Postal System, 1890-1910 94

Five: The Triumph of the Moral Economy: Finance, Parcels, and the Labor Dilemma in the Post Office, 1908-24 144

Six: Science in the Service of Seeds: The USDA, 1862-1900 179

Seven: From Seeds to Science: The USDA as University, 1897-1917 212

Eight: Multiple Networks and the Autonomy of Bureaus: Departures in Food, Pharmaceutical, and Forestry Policy, 1897-1913 255

Nine: Brokerage and Bureaucratic Policymaking: The Cementing of Autonomy at the USDA, 1914-28 290

Ten: Structure, Reputation, and the Bureaucratic Failure of Reclamation Policy, 1902-14 326

Conclusion: The Politics of Bureaucratic Autonomy 353

Notes 369

Archival Sources 459

Index 465

What People are Saying About This

Richard Bensel

Whether we regard the modern state as fair as Athena, stepping fully formed from the brow of Zeus, or as foul as Frankenstein, sutured on a scientist's table, there had to be a time of quickening when the limbs began to twitch and the brain began to spark. In a splendid reinterpretation of the classic period of American state formation, Dan Carpenter demonstrates that a self-conscious mentality emerged because career bureaucratic officials created overlapping networks between their agencies and forged public reputations that secured support from the citizenry. Thus freed them from the influence of political parties, these officials then turned on the very politicians who had created them.
— Richard Bensel, Cornell University

Stephen Skowronek

The Forging of Bureaucratic Autonomy is a major work sure to influence future understandings of progressivism, state-building, and American political development. Carpenter delves into the highly variable world of bureaucratic entrepreneurship and innovation in organization to explain the emergence of scattered pockets of administrative autonomy within the executive branch of American government. His carefully crafted analysis of the conditions under which administrators have gained control over the political authorities that ostensibly control them presents a formidable challenge to the assumptions of political scientists, and it should prompt some equally careful rethinking of the operations of American democracy more generally.
— Stephen Skowronek, Yale University

From the Publisher

"The Forging of Bureaucratic Autonomy is a major work sure to influence future understandings of progressivism, state-building, and American political development. Carpenter delves into the highly variable world of bureaucratic entrepreneurship and innovation in organization to explain the emergence of scattered pockets of administrative autonomy within the executive branch of American government. His carefully crafted analysis of the conditions under which administrators have gained control over the political authorities that ostensibly control them presents a formidable challenge to the assumptions of political scientists, and it should prompt some equally careful rethinking of the operations of American democracy more generally."—Stephen Skowronek, Yale University

"Although we tend to discuss the strength, or weakness, of state autonomy as though it were the same for every agency, the fact of the matter is that autonomy varies considerably from agency to agency. In this excellent book, Daniel Carpenter is among the first to make this observation and explore its implications."—Graham K. Wilson, University of Wisconsin-Madison

"Whether we regard the modern state as fair as Athena, stepping fully formed from the brow of Zeus, or as foul as Frankenstein, sutured on a scientist's table, there had to be a time of quickening when the limbs began to twitch and the brain began to spark. In a splendid reinterpretation of the classic period of American state formation, Dan Carpenter demonstrates that a self-conscious mentality emerged because career bureaucratic officials created overlapping networks between their agencies and forged public reputations that secured support from the citizenry. Thus freed them from the influence of political parties, these officials then turned on the very politicians who had created them."—Richard Bensel, Cornell University

Graham K. Wilson

Although we tend to discuss the strength,or weakness,of state autonomy as though it were the same for every agency,the fact of the matter is that autonomy varies considerably from agency to agency. In this excellent book,Daniel Carpenter is among the first to make this observation and explore its implications.

Wilson

Although we tend to discuss the strength, or weakness, of state autonomy as though it were the same for every agency, the fact of the matter is that autonomy varies considerably from agency to agency. In this excellent book, Daniel Carpenter is among the first to make this observation and explore its implications.
— Graham K. Wilson, University of Wisconsin-Madison

Editorial Reviews

Carpenter's book is intellectually arresting—weaving quantitative and qualitative empiricism through an impressive array of theoretical propositions toward an attractive theory of bureaucratic autonomy in the administrative state . . . . [A]dmirably successful in adding to our narrative of the development of the American administrative state.

Public Administration Review - Anthony Bertelli

Carpenter's book is intellectually arresting—weaving quantitative and qualitative empiricism through an impressive array of theoretical propositions toward an attractive theory of bureaucratic autonomy in the administrative state . . . . [A]dmirably successful in adding to our narrative of the development of the American administrative state.
— Anthony Bertelli

Public Administration Review

Winner of the Levine Memorial Book Prize

Winner of the Gladys M. Kammerer Award

From the Publisher