Taxing wars the american way of war finance and the decline of democracy

TAXING WARS


TAXING WARS
The American Way of War Finance and the Decline of Democracy
SARAH E. KREPS


Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research,
scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and certain
other countries.
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America.
© Oxford University Press 2018
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any
means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by license, or under terms
agreed with the appropriate reproduction rights organization. Inquiries concerning reproduction outside the scope of the above should be
sent to the Rights Department, Oxford University Press, at the address above.
You must not circulate this work in any other form and you must impose this same condition on any acquirer.
CIP data is on file at the Library of Congress
ISBN 978–0–19–086530–6
eISBN 978–0–19–086532–0


CONTENTS

Acknowledgments

1 Introduction
2 Bearing the Financial Burden of War

3 Partisan Politics in the Early Wars: Conflicts of 1798, 1812, and 1898
4 The “Liberty Bond” Approach to War Finance: World Wars I and II
5 From Taxation to Borrowing: Declining Fiscal Sacrifice in Korea and Vietnam
6 “Hide-and-Seek” Wars: The Afghanistan and Iraq Wars
7 Cross-National Survey Evidence from the United States, the United Kingdom, and France
8 Conclusion
Appendix
Notes
Index


ACKNOWLEDGMENTS

During graduate school, I worked as a US Air Force reservist at the National Reconnaissance Office
(NRO). The NRO had obtained mythical status in my mind because it dealt with space, satellites,
oodles of secret money, and spooks from the Central Intelligence Agency.
One of the first people I met working at the NRO was a former non-official cover officer. As he
quickly told me, he was the kind of CIA agent who would be killed without question if anyone in the
field discovered his identity. And as with most people in the intelligence business, he was
dispositionally paranoid. One clear manifestation was that he came to work every day with ten $100
bills rolled and squirreled away in various parts of his body and clothes. It was a habit, he said, that
he had picked up in case he needed to bribe his way out of a dangerous situation. He had been unable
to shed the practice even though we worked in the leafy suburbs of northern Virginia.
A somewhat more useful nugget he dispensed to his amateur colleague was how to evade questions
about the classified nature of his work. He said that when people asked what he does, he said he is a
tax specialist. The response would shut down further questions because no one would ever want to
talk more about any intricacies of taxes. I nodded in agreement, because I thought for a long time that
people didn’t want to discuss taxes. I no longer agree, and I hope that the reader of this book does not
either.
Taxes, I realized, are not just an economic tool. They are inherently political. Of course, I had

known this as early as grade school when we were taught about the Tea Party Massacre and the
origins of the American Revolution, but as an adult, it seemed like taxes were simply something
onerous we dealt with in April of every year.
I started to reevaluate this belief one day when I was talking to a Cornell colleague of mine in
comparative politics, Gustavo Flores-Macías. He was studying the security taxes levied in Colombia
to fight the guerrilla war, and I started to dig into the United States experience. It turned out the United
States had not levied a war tax since the Vietnam War, and I wondered why that practice had changed
and what the consequences might be.
My colleague and I collaborated on a couple of articles on war taxes, one of which appeared in the
American Political Science Review and another in the Journal of Conflict Resolution, but it seemed
like a book manuscript was in order that would allow me to write more systematically about the
American experience with paying for its wars. This book is the product of that research. It was a long
time in coming, as I briefly detoured to write a couple of books on drone warfare, which I came to
realize was actually a related story. The United States has increasingly worked to shield its
population from the costs of war not just in blood, through the use of drones, but also in treasure, by
avoiding war taxes and financing its conflict through debt. A long-established tradition of democratic
theory suggests that a key difference between democracies and non-democracies is that a democratic
populace bears the direct costs of war in blood and treasure. The more directly they bear those costs,
the more incentives they have to pressure leaders to keep wars short and low cost.
My normative concern with this project was the inverse of this logic. If individuals no longer saw
the costs of war, would they be less politically engaged with the cost, duration, and outcome? The


study suggests that the answer is yes. Beginning in 2001, the United States began a war against AlQaeda that morphed into a war against unrelated militants in east Africa, north Africa, and Yemen.
President Obama campaigned on winding down the wars in Afghanistan and Iraq, only to wind them
back up. In 2014, he began a war against the Islamic State in Syria and Iraq. Legislators have been
relatively silent on each of these fronts because their constituents are silent. The constituents are
silent because they are shielded from the costs of war. Accountability linkages have correspondingly
unraveled. Debt levels rose as the sense of fiscal propriety that characterized prior wars failed to
emerge from the public, quite a different attitude from the one expressed by Woodrow Wilson:

“Borrowing money is short-sighted finance. We should pay as we go. The industry of this generation
should pay the bills of this generation.”
As I worked through the research on the book, I incurred a number of—pardon the pun—my own
debts. I am grateful for funding from the Institute for Social Sciences at Cornell, which funded an
early experiment I conducted with Gustavo Flores-Macías. The Koch Foundation provided a
generous grant that allowed me to do fieldwork in the United Kingdom, Italy, and Germany to expand
my cross-national perspective. The generous Appel fellowship allowed me to spend my sabbatical
year in Spain while I finished drafting and editing the manuscript.
I also appreciate the feedback I received on various aspects of the research, including comments
from participants in seminars at the University of California-Berkeley, Binghamton University, Duke
University, North Carolina State University, Rutgers University, and the University of North Carolina
at Chapel Hill. I would also like to thank participants at the Costs of War workshop at Cornell, as
they helped move along the ideas for the book in its early stages, and my own book workshop, also at
Cornell, whose participants helped finesse the argument and empirics toward the final stages.
A number of individuals provided helpful feedback along the way. In particular, I would like to
thank those who took the time to provide written or verbal comments: Mariel Barnes, Richard Bensel,
Marc Blythe, Jon Caverley, Malcolm Chalmers, Debak Das, Nisha Fazal, Peter Feaver, Gustavo
Flores-Macias, Ben Fordham, Lawrence Freedman, Aaron Friedberg, Jonathan Kirshner, Peter
Katzenstein, Tamir Libel, Paul MacDonald, Paul Newman, Dan Reiter, Condoleezza Rice, Sten
Rynning, Elizabeth Saunders, Ken Schultz, John Schuessler, Nic Van de Walle, Karin von Hippel,
Kyle Wolfley, Amy Zegart, and Micah Zenko. My thanks also go to Dave McBride at Oxford who
heard an early pitch of the book several years back, had thoughtful suggestions about the argument,
and then helped shepherd the work to its final publication.
Last, none of this would have been possible without the support and love of my family: my parents,
Gustavo, and my kids, Luke and Sebastian. I dedicate this book to them, but especially to my dad,
who died suddenly as I finished final edits and until that last day thought his taxes were too darn high.


TAXING WARS

1
Introduction

the attacks of September 11, 2001, President George W. Bush famously stood in
New York City in the rubble of the destroyed World Trade Centers, held a bullhorn, and said, “I can
hear you, the rest of the world hears you, and the people who knocked these buildings down will hear
all of us soon.”1 In the days that followed, he also urged Americans to “get down to Disney World in
Florida . . . take your families and enjoy life, the way we want it to be enjoyed.”2 Americans, he
urged, should not hunker down and make gestures of sacrifice but return to normalcy. Consistent with
this tone, the government cut taxes; and efforts to introduce a “share the sacrifice” war tax to pay for
military action in Iraq in 2007 and Afghanistan in 2009 in response to the attacks were restricted to
political theater of renegade anti-war members of Congress.3 Responding to these war tax proposals,
Representative Jerry Lewis (R-CA) said that Americans were “already being taxed to death” 4 and
would be hostile toward any type of new tax, including a war tax. Democrats themselves “ran away
from this idea as fast as you can say the words ‘Republican majority.’  ” 5 The result is that the
government has financed the wars in Iraq and Afghanistan entirely through borrowing.
Financing these recent wars through debt could not have been more different from earlier
experiences when wars meant taxation. In 1914, three years before America’s direct involvement in
World War I, President Wilson urged war taxes as a way to fund defense preparations: “Borrowing
money is short-sighted finance. We should pay as we go. The industry of this generation should pay
the bills of this generation.”6 Despite trying to gain the American people’s acquiescence to enter the
war and later to maintain their support throughout the war, the Wilson administration levied a series
of war taxes before, during, and after the war, amounting to about one-third of the war’s costs. 7 The
public was no less deterred from the war or war taxes at that time than they had been in the SpanishAmerican War, when editorials proclaimed that the populace would “cheerfully pay the cost” in war
taxes.8
Why, when Wilson was aiming to recruit support for the war from a reluctant public, did he
introduce measures such as a hefty war tax that recent leaders have considered politically toxic? Why
was the public so magnanimous in its willingness to contribute its own resources? By contrast, why
did recent leaders not use the crisis of war, 9 often employed as the entrée for introducing war taxes,

in the aftermath of 9/11 to extract resources from the populace in a way that had been customary in the
past? More generally, what explains shifting attitudes toward bearing the financial burden of war,
moving away from war taxes—and the consequences of that shift?
In this book, I argue that the starkly different approaches are the result of public attitudes toward
wartime fiscal sacrifice that vary depending on the underlying type of war and state-society relations,
in particular the role of taxation in the nation’s social and political life (discussed in more detail in
chapter 2).10 As Scarlett O’Hara said in Gone with the Wind: “Death, taxes and childbirth! There’s
never any convenient time for any of them.” But there are less inconvenient times than others. That
innate antipathy toward the inconvenience of taxation can be dislodged by certain types of wars and
IN THE DAYS AFTER


state-society relations. When these factors combine to make the public cost sensitive, leaders have
pursued forms of war finance that anticipate opposition and minimize constraints on the way they use
force. In the post-1945 world, the public has become almost uniformly unforgiving of fiscal sacrifice,
which explains leaders’ increased tendency to rely on less visible forms of finance such as
borrowing. Leaders have, in turn, increasingly operated without the decision-making constraints that
were in play in many earlier wars when individuals were more directly involved in the costs of war.
It is no surprise that American wars since that shift in the latter half of the twentieth century have
become increasingly costly and protracted, an observable implication of the unraveling link in
accountability between legislators’ decision making and the public’s financial burden that democratic
theorists have long believed distinguishes democracies from non-democracies when it comes to the
conduct of war.
War Finance and Democratic Accountability
The direct connection between the populace and the burdens of war underlies many of the theoretical
checks and balances thought to characterize democratic restraint in conflict. Immanuel Kant suggested
that
if the consent of the citizens is required in order to decide that war should be declared (and in this constitution it cannot but be the
case), nothing is more natural than that they would be very cautious in commencing such a poor game, decreeing for themselves
all the calamities of war. Among the latter would be: having to fight, having to pay the costs of war from their own resources.11

As bearing the costs of war should constrain the public and in turn its leaders, this lies at the heart
of institutional arguments associated with the democratic peace: Since “the people ultimately pay the
price of war in higher taxes and bloodshed,”12 they are more sensitive to the costs, in turn curbing
leaders’ ambitions in conflict. The statement about the costs of war and the populace is grounded in
several assumptions: that the direct, visible costs of war are passed along to the citizenry in a
democracy; that bearing the costs of war is generally unpopular and will make the people judicious
about the use of force; and that they have electoral recourse and can register their displeasure with a
particular policy—for example, if a leader carries out a costly war—at the ballot box. The
implication is not that democracies will not engage in war. It is that since individuals bear the burden,
they are sensitive to those costs and put pressure on leaders to keep wars short and low cost. Coupled
with the electoral checks embedded in a democracy, the costs of war contribute to, as Michael Doyle
labeled the Kantian claim, “republican caution—Kant’s ‘hesitation’—in place of monarchical
caprice.”13 The cost-sensitive democratic populace will tend to moderate leaders’ ambitions in the
conduct of war.
Such institutional constraints are not a given, however. Democratic leaders have found ways to
sidestep or minimize cost-related public opposition that would check their conduct in war. For
example, although the public is thought to be averse to casualties, 14 other arguments suggest that
people are willing to tolerate high casualties for the prospect of a successful war. 15 Elite discourse,
then, can go some way toward mitigating the adverse effects of casualties, emphasizing the prospect
of victory and potentially relaxing some of the checks and balances that were thought to be embedded
in democratic institutions.16
Perhaps more fundamentally, leaders have found ways to minimize the cost in blood altogether.
They have eliminated conscription, thus creating far more localized rather than diffuse consequences


of casualties.17 They have turned toward greater reliance on airpower and unmanned aerial vehicles
(drones) to reduce dramatically the likelihood of casualties suffered by allies.18 Leaders have also
shifted away from a labor-intensive military in favor of a capital-intensive military that is financially
costlier but poses a lower risk of casualties.19 As evidence of this shift, while the Vietnam War

incurred over 58,000 fatalities at a financial cost of about $750 billion in 2010 dollars, the combined
wars in Iraq and Afghanistan—also roughly a decade in duration—resulted in around 6,000 fatalities
but at a cost of about $1.5 trillion.20
If democratic accountability mechanisms require individuals to bear the burdens in blood and
treasure, shifting the cost from blood to treasure need not affect the theoretical linkages between
leaders’ conduct in war and the populace. The reason is that taxation, for example, confronts
individuals with the financial burden of war more directly than debt. As Charles Tilly argued,
taxation “constitutes the largest intervention of governments in their subjects’ private life.” 21 Taxation
is onerous, and when citizens bear the burden of war in taxation, this creates tighter institutional
linkages between the public and leaders’ conduct of war, as taxpayers have more incentives to hold
leaders accountable for how the resources are being used.
In contrast, borrowing shields the public from the direct costs and insulates leaders from heavy
scrutiny; in turn, it undermines the institutional checks and balances that separate a democracy from
the caprice of a non-democracy. To the extent that leaders turn to borrowing as a form of war finance,
rather than taxation, the gap between their actions and public scrutiny—an accountability gap—
widens, reducing the decision-making constraints in ways that affect the allocation of domestic
resources as well as initiation and termination of interstate wars.22
The Contemporary Shift from Taxation
War taxes were once a mainstay of American war finance. As Thomas Paine said back in 1787, “War
. . . has but one thing certain, and that is to increase taxes.”23 As Table 1.1 shows, for a long time, and
during Paine’s time, this was a true statement. Taxes featured prominently in every major American
war through World War II, as well as many of the minor wars such as the 1798 Quasi-War and the
1898 Spanish-American War, producing a finance strategy I call “Liberty Bond” wars in which
leaders resorted early and often to war taxes as a form of war finance. Liberty Bonds, of course, were
a way in which the public financed the government’s debt and were not a tax. 24 The label I use here
draws the spirit of the initiative—asking the public for fiscal sacrifice—rather than the technicality in
which the government accrued debt.
Contrary to Paine’s assertion, since 1945, the United States has increasingly engaged in “Hide-andSeek” wars. Leaders have shied away from asking the populace for fiscal sacrifice, thereby
anticipating and sidestepping public constraints on their conduct of war by avoiding war taxes and
seeking less obvious forms of war finance, especially borrowing. As Chapter 5 discusses in detail,

President Truman introduced three war taxes early in the Korean War (two in 1950, another in 1951)
but was rebuffed later in the war, withdrawing plans for additional war taxes in 1952. Leaders since
then have either reluctantly imposed or altogether eschewed war taxes. The last war tax levied in the
United States was a Vietnam surtax in 1968, which President Johnson reluctantly endorsed almost
five decades ago.25
To put this trend in context, while the costs of recent wars tend to be lower today than many earlier
wars, conflicts comparable in size to those of today were once financed through war taxes. For


example, in 1798, the United States levied a war tax for tensions with France—the Quasi-War—even
though the tensions never actually reached the threshold of a war and defense spending was just
0.83% of GDP. In 1914, the first year of the World War I tax, defense spending was 0.71%, where it
roughly remained until 1918, after three waves of war tax legislation. In 1940, the year the United
States passed two sets of war taxes, defense spending was just 1.55% of GDP. Compare these figures
with contemporary spending, which has approached 5% for recent wars, as Table 1.1 shows. In the
past, such revenue demands were grounds for war taxation. Today, leaders finance those wars
through borrowing.
To be sure, major wars produced major taxes, with the largest historical increases in income taxes
associated with the largest revenue needs: during the Civil War, World War I, and World War II. But
I submit that paying for wars is analogous to what Aaron Friedberg argues about financing defense in
general: “In terms of simple bookkeeping arithmetic, there is obviously no single solution” but
instead, it “is therefore a matter of societal choice; it is a political issue rather than, in any sense, a
‘purely’ economic one.” 26 It becomes political since a nation can generate revenues in a number of
ways: through borrowing, printing money, or taxing. Which of these it pursues becomes inherently
political, a product of interactions between the war and the society in which those choices are made.
Tax increases that would and indeed did, lead to revolt during peacetime have sometimes been
tolerated during wartime. The reason for acceptance of taxes during conflict is that “wars seem to be
capable of generating whole new political universes”27 and are able to open “windows” for major
policy shifts.28 While some wars in the past opened these policy windows, causing individuals to
“accept levels of taxation in a period of war that would be regarded as intolerable in quieter

times,”29 leaders in contemporary wars have shunned war taxes altogether, opting instead for less
visible forms of finance, such as borrowing. At issue, then, is why the politics of wartime taxation has
changed. The next section outlines the basis for that change, speaking to the sources of public
willingness to bear the financial burden of war and the way those attitudes have shaped leaders’
strategies toward how they pay for war.
The Argument in Brief: When Leaders Tax
Just because wars are capable of generating new political universes—by raising the public’s
tolerance for taxation—does not mean that they invariably can or do. The Stamp Act of 1765 that
King George III levied on the American colonies to pay for the French and Indian Wars became the
raison d’être for the Revolutionary War and American independence. Alexander Hamilton’s tax to
pay for the 1798 Quasi-War led to the Fries Rebellion, extinguished by the militia that the Federalists
deployed in response. Conclusions about wars opening policy windows derive, in part, from the
propensity to focus on large-scale wars.30 But the modal war, at least for the United States, has not
been wars of a large, existential nature, but rather smaller-scale wars such as the War of 1812 or the
Afghanistan War, neither of which reached the intensity of a World War II that consumed nearly 50%
of GDP at its peak, suggesting that theories of democratic accountability and state-building should
register the impact of these more frequent yet smaller military overtures.
Taking into account how the broader experience of conflict affects democratic accountability, I
first argue that individuals do harbor the sensitivity to war costs that democratic theorists have
identified, but that the degree of sensitivity is shaped both by the type of war and underlying statesociety relations. The second aspect of the argument is that leaders, seeking to minimize constraints in
war, pursue strategies that both anticipate and deflect opposition that would arise from the public’s


sensitivity. That means they turn to less visible or onerous approaches such as borrowing rather than
imposing war taxes when individuals express little appetite for fiscal sacrifice in wartime.
To elaborate further, the first part of my argument acknowledges that individuals are likely to hold
some baseline-level of antipathy toward any taxes. The direct effect of taxation on individuals can
produce dissatisfaction among the same people whose support is needed for the war (and for the
leaders’ political longevity). I identify two main factors that can ease that dissatisfaction. One is
engaging in a major war versus a limited war. Large-scale wars create new political universes

because they clearly delineate the choice between fiscal sacrifice and total defeat. The trappings of
major war—including mobilization, rationing, repurposing civilian industries for military effort, and
clear markers of success and failure—help impress upon the public the impact and stakes of the war.
By contrast, limited wars are less likely to convey the same palpable stakes, since by definition, they
have narrower objectives that allow the state’s politics, economics, and society to carry on without
interruption.31 Indeed, typical objectives of such wars—goals in Afghanistan, for example, included
reducing corruption, educating larger percentages of women, and making it easier to obtain a driver’s
license32—are often more likely to resemble foreign aid than military objectives. The more removed
those goals are from Americans’ sense of security, the less likely Americans are to think that fiscal
sacrifice is warranted as it was in major wars that presented clear and important stakes.
Another factor that affects the public’s propensity for fiscal sacrifice in wartime is the set of statesociety relations underlying the war, a consideration that refers both to the level of taxation and the
portfolio of services individuals expect from their government.33 After World War II, levels of
taxation never returned to previous peacetime levels, which has made the public loath to pay specific
war taxes, expecting peacetime taxes to be sufficient for both the expanded social welfare state and
also for defense. Thus, the postwar politics of taxation has generally favored borrowing over
taxation, implying that individuals have had a lower propensity for fiscal sacrifice for the more
contemporary wars. How these two factors interact, detailed in Chapter 2, affects the tendency for
fiscal sacrifice among the populace.
The second part of the argument suggests that these attitudes shape leaders’ approaches to war
finance: Leaders will favor the revenue approach that mirrors anticipated public attitudes in order to
minimize their decision-making constraints. In the context of passing along the costs of war to the
public, this means that leaders shy away from war taxes when they know the citizenry has no desire
for personal fiscal sacrifice and they will turn instead to borrowing as a way to shield the public from
the costs of war. The costs of borrowing are both diffuse and deferred compared to the direct and
immediate impact of taxation. Taken together, the argument visualizes leaders as informed by public
attitudes but seeking to avoid opposition, and therefore acting to limit public exposure to the costs of
war as a way to minimize constraints on their potential decision making.34
That cost sensitivity is not static but conditional on underlying circumstances challenges two sets of
alternative arguments about the connection between bearing the costs of war and democratic
accountability in war. The first consists of two relatively immutable views of cost sensitivity. Adam

Smith observed that leaders looking for ways to generate revenue for war would always find it “more
convenient to defray this expense by misapplying the sinking fund than by imposing a new tax. Every
new tax is immediately felt more or less by the people. It occasions always some murmur, and meets
with some opposition.”35 Smith’s suggestion about the onerous nature of taxation actually led him to
advocate taxes as a way to pay for war, as it would close the distance between individuals and the
war’s burdens, but he nonetheless predicted that exposure to the costs of war through war taxes


would have the tendency of “offending the people, who, by so great and so sudden an increase of
taxes, would soon be disgusted with the war.” 36 The implication is that individuals are naturally cost
sensitive, and bearing the fiscal burden would invariably tarnish their support for war.
A contrasting but equally immutable view of the public’s relationship with the financial burdens of
war suggests that the public is relatively cost insensitive. To the extent that victory is a function of the
resources the state can commit to the war, 37 citizens will be willing to make economic sacrifices
during wartime, seeing higher taxes as a vehicle for strengthening the military and increasing the
likelihood of winning. Neither the view that individuals are cost sensitive or cost insensitive
considers the possible variability of cost sensitivity and the circumstances that would give rise to
such conditionality.
The argument also challenges both bottom-up and top-down views of democratic accountability.
The bottom-up view portrays leaders as having hard and fast constraints in the form of public
opinion.38 Instead, my argument gives agency to leaders who can strategically pursue policy
instruments that anticipate public opinion consequences and minimize the prospect of political fallout
with respect to their tenure or the war. I argue that leaders may both be responsive to public opinion
and also seek to reduce the way this opinion constrains their foreign policy options. The need for
public consent is exactly what causes leaders to craft their policies in ways that anticipate public
attitudes so as to deflect opposition and prop up support for wars.
It similarly challenges the top-down notion that leaders can manufacture public consent through
elite discourse,39 outright deception or lying,40 or simply through information asymmetries that enable
them to sell the public on the use of force, all of which effectively allow leaders to sidestep
institutional constraints.41 The public does find ways to push back against unpopular policies, and

leaders have learned to preempt opposition by structuring policies in ways that minimize the
constraints that come from dissent. The second part of the argument thereby stakes out a middle
ground between those who suggest that leaders are captives of the public’s institutional checks 42 and
those who believe that leaders can altogether shunt major decisions about war and peace away from
the public’s gaze.43
Research Design
The book examines the argument that public attitudes about fiscal sacrifice depend on the type of war
and state-society circumstances underlying the war and that these attitudes have influenced how
political elites pay for wars. In developing the argument, I draw on the democratic accountability
literature but focus my theoretical and empirical analysis primarily on the United States. While it is
just one state, it is one that subscribes to many institutional characteristics that mark it as a
democracy, such as electoral checks and freedom of expression. My empirical evidence therefore
focuses primarily on the American experience with war, although the last empirical chapter examines
cross-national evidence from the United Kingdom and France, with briefer reference to Israel and
India, to examine whether similar patterns hold in other democracies that have different types of
threats and fiscal traditions.
The first part of my empirical strategy involves case studies, which are especially helpful in
tracing the argument about changing attitudes toward fiscal burdens of war and the way leaders
structure their war finance decisions with those preferences in mind. Each case study briefly
discusses the historical background of the war and the form of war finance. The analytical part of the


chapter evaluates the source of public attitudes toward war and the degree to which public attitudes
affected the turn to taxes or less visible alternatives.
To assess public opinion, I probe “the public mood,” which is defined by “the notion that a rather
large number out in the country are thinking along certain common lines.”44 While public opinion
polls are most helpful in assessing the mood, they are not widely available before World War II.
Thus, in addition to polls, I also use other measures: newspaper editorials, individual or collective
movements lobbying for a particular policy position, and political conventions that make public
declarations in favor of particular policies. As these measures show, there is not always one public

mood but rather different constituencies with different attitudes toward war and war taxes.
To interrogate the linkages between those attitudes and war finance strategies, I turn to primary
source communications between political elites. I also use congressional hearings and debates on war
tax legislation, and secondary source material such as biographies of wartime presidents or diaries of
aides to these presidents. Another way of exploring the linkages between public fiscal sacrifice and
approaches to wartime finance is through the design of tax legislation itself, since historical evidence
may not always point to the effect of public opinion on policy decisions. Of course, an absence of
evidence that public opinion weighed on leaders is not evidence of its absence. Indeed, in leaders’
conduct of foreign policy, “attention to public attitudes [is] so ingrained in their working habits that it
was unnecessary to make constant references to it”45 in historical records. To the extent possible, I
access the ways leaders referenced the explicit effect of public opinion, but I also analyze congruence
between the structure of public opinion and the design of particular policies in a way that, given
public attitudes, would generate maximum consent and minimize opposition.46
The attention to policy design follows the assumptions embedded in arguments about dynamic
representation. According to this perspective, “elected politicians . . . sense the mood of the moment,
assess its trend, and anticipate its consequences. Changes in opinion . . . will lead politicians to
revise their beliefs about future election opportunities and hazards. Such strategic adjustment will . . .
drive policy through rational anticipation.” 47 Leaders, perceiving the national mood, will bring
policy into line with these preferences to minimize the political risk. This means not just electoral
risk to their tenure but in the case of wars, declining consent that is at the root of legitimate conduct of
conflict. Particular policy changes then, based on these assumptions, are the effect of perceptible
changes in public attitudes. Since leaders may not make explicit reference to the constraining effects
of public opinion and how their proposed policies intend to anticipate opposition, I look to the
peculiarities of tax legislation itself. To the extent that particular legislation creates winners and
losers that map onto the individuals or groups expressing particular preferences on tax legislation,
then we can draw inferences about the connection between the public inputs and policy outcomes.
Indeed, in a number of instances, the peculiarities of particular tax legislation were so congruent with
public attitudes that alternative explanations are difficult to surmise.48
A second part of the empirical strategy brings survey evidence to bear on the argument. This part of
the study explores contemporary attitudes in the United States both through surveys probing support

for a war tax to finance the 2009 surge in Afghanistan and through experiments that help draw
inferences about the support for war conditional on its being financed through taxes. The surveys also
use open-ended questions to probe the reasons for contemporary hostility toward war taxes. I also use
surveys to investigate whether the trends observed in the United States apply to other advanced
industrialized democracies such as the United Kingdom and France. An original experiment
conducted concurrently in the United Kingdom and the United States allows for these direct


comparisons, and follow-up personal interviews help tease out some of the peculiarities of the
British experience despite the experimental evidence pointing to the similar ways in which American
and British respondents dampen their support when faced with the prospect of paying for war through
taxes versus borrowing. In addition, to see whether there might be an exception to the public’s
unwillingness to be taxed in the service of war, I carried out an original survey in France in
November 2015. Conducted days after the deadliest attack against France since World War II—
declared “un acte de guerre” (“act of war”) by President Hollande—the context should have been a
most likely case for fiscal sacrifice. Yet the evidence further corroborates the argument that
contemporary audiences in advanced industrialized democracies view their peacetime taxes as
adequate for covering a range of services, including spending on defense, and they see as little reason
for the war taxes implemented in previous wars as they would for an “air tax.”
In short, through a combination of historical case studies and survey evidence, I provide empirical
support for the argument that compared to the past, when either the war or state-society relations
could make taxation a palatable approach to paying for wars, contemporary audiences are decidedly
averse to bearing the direct financial burden of wars. The turn to borrowing, essentially now a
permanent feature of contemporary wars, serves as a way to minimize public opposition and reduce
constraints on leaders’ decision making in these recent conflicts. Shrouding the cost of war may give
leaders institutional slack but it also unravels the mechanism underlying wartime restraint in a
democracy. Without bearing visible burdens, individuals will have fewer incentives to scrutinize the
way leaders conduct wars, and democratic constraints in wartime will no longer operate in the way
theorized from Adam Smith to Immanuel Kant all the way up to contemporary scholars of
international relations. The reason that a democratic populace scrutinizes leaders’ decisions in war is

because they as taxpayers bear the burden. This analysis suggests that the unraveling of those
accountability linkages explains why wars beginning in the latter half of the twentieth century are
longer, culminating with the ongoing wars in Iraq and Afghanistan fifteen years after their inception.
Given the trends identified in the book, the recent past—the two longest wars in US history—is also
likely to be prologue.
TABLE 1.1 The

Cost of American Wars*

War

Years

Total
War Cost
Cost
% GDP in
(Constant Peak Year
FY2001$) of War

American
Revolution
FrancoAmerican
War
War of
1812
MexicanAmerican
War

1775–

1783
1798–
1800

$2.4 billion

N/A

Total
Tax-toDefense % GDP
GDP in Peak Ratio
Year of War Prior to
War
N/A
N/A

War
Tax
Years

Finance
Strategy

1.93%

1789,
1790
1798

Liberty

Bond
Liberty
Bond

NA

N/A

1.22%

1812–
1815
1846–
1848

$1.5 billion

2.2%

2.7%

1.23%

1813

1.9%

1.45%

None

Hide-andSeek
Hide-andSeek

$2.4 billion

1.4%


Civil War
(Union)

1861–
1865

$59.6 billion

11.3%

11.7%

0.90%

SpanishAmerican
War
World War
I

1898

$9 billion

1.1%

1.5%

2.24%

1917–
1918

$334 billion

13.6%

14.1%

1.84%

World War
II

1941–
1945

$4.1 trillion

35.8%

37.5%

6.88%

Korean
War

1950–
1953

$341 billion

4.2%

13.2%

13.4%

1965–
1975
Persian
1990–
Gulf War
1991
Afghanistan 2001–
2010
Iraq
2003–
2010

$738 billion

2.3%

9.5%

16.2%

$102 billion

0.3%

4.6%

17.79%

1966,
1968
None

$321 billion

0.7%

4.9%

19.35%

None

$784 billion

1%

4.3%

15.99%

None

Vietnam

1861,
1862,
1864
1898

Liberty
Bond War

1914,
1916,
1917,
1919
1940
(2),
1941,
1941,
1944
1950
(2),

1951

Liberty
Bond War

Liberty
Bond War

Liberty
Bond War

Liberty
Bond
War≈
Hide-andSeek War
Hide-andSeek War
Hide-andSeek War
Hide-andSeek War
Hide-andSeek War

* For a complete coding of war taxes, see Gustavo Flores-Macías and Sarah Kreps, “The Political Economy of Death and Taxes: A
Study of American War Finance, 1789–2010,” American Political Science Review 107, no. 4 (November 2013), 833–848.
Source: Based on Stephen Daggett, Costs of Major US Wars (Washington, DC: Congressional Research Service, 2010), 1–2.


2
Bearing the Financial Burden of War

Linda Bilmes published a book called The
costs of the Iraq War but clearly had an

advocacy agenda: to expose the costs of war to the public. The implication was that if the American
people knew the war’s costs, they would turn against the war. While the logic nominally built on
democratic theory—that bearing the burdens of war will cause individuals to put pressure on leaders
to keep the costs low and bring wars to an expedient end—the assumption was that individuals would
be sensitive to costs in debt the same way they would if those costs were passed along in higher
taxes.
This chapter shows that, all things being equal, bearing the financial burden in debt is unlikely to
confront the populace with the costs of war nearly as directly as bearing the burden in higher taxes.
Even with exposure to taxes, however, the effect is not a constant. As I argue theoretically, first,
individuals’ sensitivity to the costs of war varies as a function of the nature of war and prevailing
state-society relations, which have changed over time. The interaction of those factors affects
individuals’ tolerance for bearing the burdens of war.
Second, in making decisions about how to pass the costs of war to the public, leaders are sensitive
to the public’s attitudes and look for ways that will have the least consequences and constraints on
them, the leaders. They turn to borrowing when the public has limited appetite for fiscal sacrifice.
By way of advancing the two related propositions associated with the argument, the chapter first
broadly discusses the way wars introduce revenue demands and how passing along those costs
creates theoretical accountability linkages between leaders and the populace. It then turns to a
discussion about why the public’s sensitivity to costs manifests itself differently as a function of the
nature of war and the state. Last, the chapter shows how changes in those conditions have shaped the
political process through which leaders have chosen to pass along the fiscal costs of war to the public
either directly through taxes or indirectly through borrowing.2
IN 2008, NOBEL LAUREATE JOSEPH STIGLITZ and his co-author
Three Trillion Dollar War .1 The book documented the

The Costs of War and Democratic Accountability
Wars create enormous revenue needs. Studies of government spending shocks in the United States
invariably point to the impact of wars,3 which are characterized by large, urgent revenue needs rather
than the smoother expenditures experienced during peacetime. Figure 2.1 illustrates this point,
mapping periods of war onto US government expenditures since 1789 and showing the often-massive

increases in spending that accompany war.
Driven in part by these revenue needs, the history of taxation bears a close resemblance to the
history of war. 4 The United Kingdom, for example, levied its first income tax in 1799 to support the
Napoleonic Wars. Prior to 1799, “the cost of war had drained Britain’s resources, and run up a
considerable national debt. The army was starving, and poor conditions in the navy in 1797 had led to
mutiny.” The government levied an income tax to provide “aid and contribution for the prosecution of


the war.” 5 Although this tax was quickly repealed after the war, the government again levied an
income tax in the Crimean War. In World War I it increased the top rates from 8% to 50%, then by
World War II it had tripled the tax base and created a permanent tax system. War taxes were seen as
efficient and equitable, and as Chancellor William Gladstone stated, they were “an engine of gigantic
power for great national purposes.”6 The British experience of wars as the basis for taxation was
mirrored in a number of other democracies, with mass warfare creating mass, progressive taxes,7 a
condition that included the American experience until recently.8
Liberal theory views bearing the financial costs of war as an asset of democracies. The logic is
that if the populace bears the burden of war in higher taxes, it will be discriminating in terms of the
consent it gives to leaders in initiating and conducting war. 9 Recognizing the directness of taxation
and the legal relationship between the populace and its government, Adam Smith endorsed
progressive taxes as an equitable way to generate revenue in a democracy and impose necessary
decision-making constraints on leaders in wartime: “The people feeling, during the continuance of the
war, the complete burden of it, would soon grow weary of it, and government, in order to humour
them, would not be under the necessity of carrying it on longer than it was necessary to do so.”10
Austrian economist Joseph Schumpeter drew a more subtle link between the fiscal demands of war
and attitudes toward war, suggesting that the costs of war caused states to go “begging to the estates”
for tax revenues, but in return for the unpopularity of extraction, those estates earned more
accountability from the princes undertaking the wars.11 Paired with his work on capitalism,
suggesting that individual entrepreneurs are “democratized, individualized, rationalized,” and
“vigorously anti-imperialist,”12 Schumpeter’s conclusion was that the liberal state, one where
individuals bear the burdens of war and have levers for registering disapproval, exercises powerful

restraint in its foreign policy.
By contrast, autocracies are sometimes unable to borrow because potential lenders do not trust
them to repay (either because the borrowers are seen as less likely to win the war or honor the
debt).13 In these cases, the autocracy might pass along the direct costs of war through taxes, but the
taxpayer-government relationship would be quite different from that in a democracy. In his study of
Nazi war finance, Otto Nathan writes that one Nazi Treasury official said that “the obligation to pay
taxes was based not upon a legal relationship between the taxpayer and the Treasury but upon a
power relationship.”14 Taxpayers had no influence on the nature of tax legislation passed, and
politicians were not beholden to constituency interests. The double-edged sword of a democracy was
that these interests could capture the state, but the state had some obligation to be responsive to those
interests. In wartime, this would translate into clearer accountability between the taxpaying public
and the leaders who were deciding how the tax money was being used, particularly in conducting
military operations.
The democratic accountability logic has intuitive appeal. The reason is that taxes represent “a
permanent transfer of purchasing power by the taxpayer to the government.”15 It is for this reason that
the “history of state expansion is a history of violent struggles over taxes,” according to Tilly. 16
Taxes are typically onerous. They are also visible. For example, the Washington Post described
World War I taxes as bringing “daily, almost hourly, reminders to the people of the United States of
the burden that is entailed in the prosecutions of a just and victorious war. The average citizen feels
the effect of the war tax when he arises in the morning . . . he is reminded of it the last thing at night
when he puts on his tax-assessed pajamas.” 17 At the least, these reminders would give individuals


incentives to scrutinize the use of those resources, constraining leaders’ freedom of action. 18 At the
extreme, the reminders would cause individuals to reevaluate their support for the war and for the
leader himself. As one editorial concluded pithily at the beginning of World War I, “increased taxes
are never a good political expedient.”19
By comparison, borrowing creates a more indirect connection between leaders’ conduct of war
and the populace. First, borrowing reduces the need to impose higher tax rates contemporaneously, 20
instead transferring the cost to a future cohort that does not affect the current leader’s prospects of

reelection. Second, wars are just one contribution to the debt. Thus, even if public is concerned with
debt issues, the war will not be a single target, as it sits alongside numerous other debt sources.21
Debt repayment of American wars illustrates both points. For most wars, repayment is a decadeslong process accompanied by many other changes that muddle the role of war finance policies. For
example, the United States began to repay its World War I debt during the 1920s, a period of
deflation that increased the real value of the debt. The 1930s further complicated the debt picture, as
the gap between expenditures and revenues grew because of the Depression.22 Rancorous political
debates about how to pay down the peacetime debt made no specific reference to World War I or
President Wilson—by then only part of the debt problem.23
To be sure, borrowing may increase the level of debt so that legislation is required to raise the
debt ceiling, and this can introduce a contentious set of debates. In practice, two factors minimize any
political costs associated with these debates. One is that despite the fanfare, Congress invariably
passes legislation to increase the debt ceiling. As a former director of the Congressional Budget
Office testified, “Most analysts view the statutory limit of federal debt as archaic. . . . [V]oting
separately on the debt is hardly effective as a means of controlling deficits. . . . [B]y the time the debt
ceiling comes up for a vote, it is too late to balk at paying the government’s bills.” 24 Indeed, the
legislation often becomes “must pass” legislation and a vehicle for passing other measures.25 A more
important reason that the political costs of this legislation are relatively low is that the issue often
reaches such a level of complexity that the public is uncertain whom to blame. During the 2011 debt
ceiling debates, only 18% of Americans claimed to understand the issue. Among those who did
understand, political costs did not have a clear directionality; almost as many people were concerned
about not raising the debt limit (42%) as raising it (47%).26 Moreover, for about two-thirds of
people, the negotiations had no effect or even increased their support for the president and the
Speaker of the House, political figures at the center of the debate.27
In short, borrowing obscures the cost of war relative to taxation. While it adds to the overall debt,
it is one of many sources, and the ultimate repayment takes place long after the leader who initiated
the war has stepped down and the war has ended, reducing the political costs to the leader and of the
war. In the run-up to the Spanish-American War, the Chicago Daily Tribune anticipated that such
political expediency would drive decisions about war finance. The editors noted derisively,
“Legislative demagogues always favor the borrowing method. They think high taxes will be
unpopular with their constituents.”28 As the later discussion will show, different types of taxes can

favor particular sections or constituencies, creating heterogeneous political costs and benefits, but in
general, it is a well-founded proposition that taxes are likely to be unpopular.
As a result of the burden of war, individuals are sensitive to the costs of war and more judicious in
how they consent to the use of force. The democratic accountability logic suggests that rather than
absorb untold and unnecessary costs, a democratic populace tends to scrutinize the way its resources
are used and withdraws support when the costs mount. Leaders are sensitive to how they allocate the


populace’s resources, knowing that a dissatisfied populace can threaten their tenure as well as their
ability to carry out a war.29
The logic could plausibly explain decision-making constraints in wartime as well as the postwar
dividends paid out to the populace for the sacrifices made during war. James Sparrow points to a
corresponding sense of “fiscal citizenship” during World War II where individuals were willing to
contribute tax payments not just out of a sense of patriotism but as a quid pro quo for the benefits that
the government conferred through the New Deal (and later the Fair Deal) entitlement programs.30 The
notion of the “extraction-coercion” cycle—in which leaders extract resources for war and must
provide accountability and public goods in exchange for the coercive act of taxation—is based
entirely on the assumption that individuals are “recalcitrant” and generally not easily persuaded to
give up their own resources.31 Leaders have electoral reasons to heed the public’s preferences for
economizing on those resources by keeping wars short and low cost, but they also have to compensate
citizens later for the sacrifices they made, co-opting public loyalty in the wake of war by providing
additional political and economic rights—for example, broader enfranchisement, advances in public
health, and children’s rights—to the populace. 32 Leaders need not even wait until the end of war to
confer those rights. In June 1944, President Franklin Roosevelt signed the Servicemen’s Readjustment
Act that offered the GI Bill for veterans of the war, providing them with funds for college,
unemployment, and housing.33
An alternative view of accountability also considers costs but suggests that leaders and the
populace are more interested in the outcome and are willing to bear the costs if doing so produces a
more favorable result. Since democratic polities see the benefits of war distributed in the form of
public goods, they have incentives to pay the higher taxes that would strengthen defense in wartime

and increase the odds of victory. 34 Some strands of this argument go further and turn the unpopularity
of taxes into a democratic advantage. Benjamin Goldsmith, for example, suggests that taxpayers are
likely to believe that leaders “will only levy greater taxes for legitimate reasons.” 35 The willingness
of democratic leaders to suffer the political costs of introducing war taxes is a signal of the
legitimacy of the war, which in turn should boost public support and provide leaders with the
resources to put forth an “all-out effort” in war.36
If it is the case that the people are prepared for fiscal sacrifice based on the prospect of greater
benefits from victory, then these prospects should also increase their willingness to absorb costs.
Greater expected payoffs mean greater propensity to expend resources and greater propensity for
fiscal sacrifice. Applying the logic to the relationship between bearing the financial burdens of war in
higher taxes and support for conflict, this argument would expect war taxes to have little effect on
people’s attitudes toward war, with the public rallying around the fiscal flag in times of need.
Both of these perspectives focus on the relationship between the public and the costs of war but are
based on two different assumptions. The first assumes a cost-sensitive, fair-weather public
discouraged by high-cost wars, therefore exerting pressure on leaders to keep the costs low. The
second takes the public to be relatively cost insensitive, more interested in winning than how much it
costs to do so. Both perspectives are relatively static, assuming instead that cost sensitivity is
constant even if the two perspectives diverge in terms of what that sensitivity looks like. Instead, I
argue that at least in the American context, public attitudes toward fiscal sacrifice are quite
contingent, sensitive to the nature of conflict and conditioned by underlying state-society relations.
The Visceral, Calculating Public


Before turning to the argument about how both the nature of war and the state shape the public’s
propensity for fiscal sacrifice in wartime, I reflect on what makes citizens tick when it comes to their
attitudes about public policies. The public plays a central role in this story because leaders are
invariably looking for the public’s endorsement and finding ways to shape policies that will help
them sidestep or at least loosen public opposition and constraints. Outlining what moves the public is
therefore foundational for the argument as a whole.
Americans have long been criticized for being ignorant of politics. Decades ago, journalist Walter

Lippmann portrayed the mass public as uninformed, wrong at every turn, and, unfortunately in his
view, “a critical veto upon the judgments of informed and responsible officials,” with the public often
too pacifist when war was needed and vice versa.37 Adding to this somewhat cynical perspective,
Gabriel Almond pointed to the public as having unstable moods that would simply distort the
decision-making process were leaders to heed the public’s perspectives, creating “cyclical
fluctuations” that would “stand in the way of policy stability.”38
Scholars have since provided many correctives to this view of an unreflective public and seem to
have settled on the assumption that at the least, “voters are not fools.”39 This view does not hold any
illusions about the public’s political knowledge—for example, whether the average member of the
public can find Ukraine on a map.40 Rather, it suggests that the public is rational and has preferences
that are generally stable, coherent, and logical in that its attitudes reflect underlying values and
respond in predictable ways to international events, embracing a fairly consistent decision-making
calculus.41 Attitudes about taxation lend credence to this perspective.
As Larry Bartels suggests, “Most ordinary citizens are remarkably ignorant and uncertain about the
workings of the tax system and the policy options under consideration, or actually adopted.”42 But
they do not need to know these intricacies of the tax system to fundamentally feel that they would like
their taxes to be lower. Since World War II, Gallup has asked the question “Do you consider the
amount of federal income tax you have to pay as too high, about right, or too low?” Figure 2.2 shows
responses to that question plotted against top marginal tax rates, which generally reflect trends in tax
rates for other brackets. As the figure shows, the public has consistently seen its taxes as too high
almost irrespective of tax rates, consistent with the view that people have a visceral aversion to
taxes, much as Adam Smith had suggested.
General antipathy toward taxation does not mean that the public’s attitude cannot be moved. 43 In
fact, while individuals will invariably believe that their taxes are too high, they are nonetheless
always engaging in a cost-benefit calculus—even if a crude one—that weighs what they are receiving
in return for their financial contributions. Indeed, the whole premise of the “fiscal contract”—in
which citizens provide their money to the state in exchange for services—speaks exactly to that costbenefit relationship. Studies suggest that the more heavily states tax the rich (as a percentage of GDP),
the more they protect property rights, which are more likely to affect wealthy segments of the
population; the more states tax lower-income populations, the more public services they provide.44 In
exchange for taxation, states are expected to provide social prerogatives, a quid pro quo between the

fiscal investment and the benefit derived from that tax, a finding that is well established in the public
administration literature.45 The dynamic intimates that the relationship between taxation and political
behavior may be more complicated than the simple expectation that taxation provokes hostility.
When it comes to fiscal sacrifice in the context of conflict, the public can be persuaded to “pay any
price, bear any burden,” as President Kennedy suggested in his inaugural address,46 but its sense of


sacrifice is conditional. Two main factors shape the ledger in ways that can nudge the public away
from its instinctive aversion to taxes: the nature of the war and the nature of state-society relations.
The Nature of War
The first factor builds on the dynamic of a fiscal contract in which in exchange for benefits, the public
is willing to bear some costs. Intuitively, the more significant those benefits, the higher the costs they
are willing to bear—again, this is Kennedy’s “bear any burden” formulation. When it comes to the
“benefit” of war, not all wars are created equal, and the differences between and among them has
important implications for the differences in fiscal sacrifice they elicit from individuals.
The study of war tends to dichotomize wars into limited wars and absolute wars, a distinction that
is relevant for this discussion. Limited war is “one in which the belligerents restrict the purposes for
which they fight to concrete, well-defined objectives that do not demand the utmost military effort of
which the belligerents are capable,” and typically one that “demands of the belligerents only a
fractional commitment of their human and physical resources,” meaning that the basic functioning of
the state can “continue without serious disruption.”47
Absolute war observes no limits and intends to achieve the complete destruction and disarmament
of the opponent.48 To this end, states engage in the massive mobilization of their political system,
society, and economy, to the extent that civilians and their infrastructure become part of the overall
war effort, contrary to the limited war where the civilians are divorced from these experiences.
Articulated in this way, the wars that Adam Smith referred to may have been limited wars, as he
references civilians reading about the wars in the newspaper, with amusement. Total, or absolute,
war (the terms are often used interchangeably) would unlikely leave civilians on the sideline (or
amused) in this same way.
The differences in the types of wars states fight are important from the standpoint of fiscal

sacrifice. Total wars naturally require more resources, so by sheer revenue needs they are more
intensive. The fiscal sociology literature more general tends to focus on these types of wars, which
may be why it assumes that states invariably engage in taxation in wartime.49 Another key difference
is that these large-scale wars are associated with mass mobilization.50 Mass mobilization means
asking a large percentage of the population to sacrifice in blood. The compensatory theory of taxation
suggests that under these conditions, a sense of fairness kicks in and the populace urges conscription
of wealth to remedy the inequalities that arise from the conscription of men, namely that those with
higher incomes were far less likely to be conscripted than the less affluent.51 While fairness may be a
component of why mass mobilization elicits a sense of sacrifice, mass mobilization is also important
because of what it makes manifest: a sense of emergency in which fiscal sacrifice becomes
instrumentally valuable. Countries only mobilize to that degree under conditions of duress, suggesting
to the public that the alternative to fiscal sacrifice may not be fiscal sacrifice but rather something
graver like one’s country losing a war, independence, or the ability to provide for one’s security.
Advocating taxes in the Civil War, the New York Times also connected taxes with winning the war,
suggesting that “battles are won by the last dollar” that can be extracted.52 By 1864, the Times again
linked its advocacy for taxation with the outcome of the war: “The only plan consistent with safety or
honor, is to tax and raise men.”53
Absolute wars, however, have not been the modal war, but the exception. Indeed, Lawrence
Freedman writes that Clausewitz actually attempted to revise his observations about war—which had


emphasized absolute war because of his experiences witnessing the Napoleonic war firsthand—
having noted that not all wars were total and that Clausewitzian theories might not be relevant for
limited wars.54 In the American experience, absolute or total wars were represented by the Civil
War, and World Wars I and II, which created palpable costs of not engaging in fiscal sacrifice. But
far more typical were the so-called small wars against insurgents. Iraq and Afghanistan are merely
the modern manifestation of a long tradition of such wars, with the early incarnation being wars
against the Barbary pirates in the early 1800s, Haitians in the 1900s, and, roughly concurrently, the
Mexicans under Pancho Villa.55
Whereas these smaller-scale conflicts were previously interspersed with larger-scale ones, the

latter have essentially vanished from the landscape, as great power wars, if not obsolete, have at least
become remarkably rare. One explanation is that modern society has become “kinder and gentler.” 56
Another is that the world wars of the first half of the twentieth century were sufficiently destructive
that they deterred subsequent conflict. 57 Perhaps the most prominent explanation for the decline of
major war is the advent of nuclear weapons, which transformed the character of war by making its
costs so high that the victor would probably suffer as much destruction as the vanquished.58
Deterred by the prospect of high costs, and decreasingly equipped to fight large-scale conventional
wars, democracies such as the United States increasingly avoided the types of high-cost wars of
attrition that had been common in earlier eras. The reluctance of countries to engage in high-cost total
wars that had plagued the preceding century is no doubt a positive development,59 but the secondorder effects may be less salutary. While contemporary wars have been lower cost than great power
wars of the past, they are also by their nature less important to the citizens of these countries and less
likely to unleash any, or at least lasting, rallies around the fiscal flag.
Indeed, because limited wars do not disrupt “economic, social, and political patterns of
existence,” the average member of the public could be excused from not knowing that such a war was
even being conducted.60 Even for a well-informed member of the public, the objectives of these wars
often seem more like exercises in foreign aid or nation-building and are therefore quite removed from
the types of military goals individuals might associate as having significance for national security.
Measures of progress in the Afghanistan war, for example, were often non-military—for example,
rates of capital flight, budget execution, and percentage of officials who buy bureaucratic positions;
all of these may seem like goals with tenuous connection to the security of the American domestic
audience.61 Indeed, the Army’s Counterinsurgency (COIN) Manual itself implies that these types of
goals might not exactly whet a democratic public’s appetite for financial sacrifice. 62 As Stephen
Biddle puts it, the concern dealt with “the willingness of democracies to afford the cost of effective
counterinsurgency (the manual warns that COIN is slow, labor-intensive, and very expensive in lives
and treasure)”63—or rather the public’s unwillingness.
In short, states have been deterred from major war, but wars themselves have not gone away; they
have been replaced with smaller-scale wars, often fought in or over countries that were not clearly
associated with core national security interests. Such wars lack the gravity of stakes that would be
patently obvious in a great power war that requires the vast mobilization of the populace and its
resources. These contemporary wars might see an initial surge of support—for example, even

Vietnam had reasonably high levels of support early in the war—but the sustainability of that support
is likely to be fragile as the populace does not experience the full mobilization that would make the
war seem meaningful and worthy of fiscal sacrifice.64


Changes in State-Society Relations
A second factor that shapes the public’s propensity for fiscal sacrifice in wartime is state-society
relations, which refers to the “interactions and interdependency between the state and society.” 65 In
some respects, this definition does more to confuse than clarify because of its circularity. With an eye
toward explaining what state-society relations actually means, some scholars have focused on the
relationship among political parties, elections, and policymaking; others have examined aspects of
governance such as devolution of authority to state and local governments; and yet others have studied
trends toward privatization or regulation. But another key aspect consists of a government
responsibility that might otherwise seem pedestrian: taxation. Without revenue, governments cannot
satisfy the basic needs of welfare or defense, so taxation is fundamental to both the state and the
citizens that reside therein.66 It was for these reasons that Joseph Schumpeter suggested in his essay
about the tax state that “the spirit of a people, its cultural level, its social structure, the deeds its
policy may prepare—all this and more is written in its fiscal history.” The study of fiscal sociology
thus took root as Schumpeter and his successors shone a light on taxation, specifically the levels of
extraction, the public goods for which those resources are dedicated, and individuals’ attitudes about
both.67
Political science scholars followed but focused more on the state’s tax levels and institutions
erected to help collect the taxes and also administer the public goods that result from extraction.
Viewed in this way, taxes—typically tax revenues as a percentage of gross domestic product—
represent a key measure of state capacity, since they refer to the resources available for the provision
of public goods as well as the apparatus in place to carry out the tax collection. 68 My argument
engages the perspectives of both sociologists and political scientists, focusing on state-society
relations through the study of tax revenues (state capacity) by way of understanding the fiscal contract
between the people and its government.69 I therefore refer variously to taxation, state capacity, and
state-society relations as variations on the same theme.

Wars have an outsized influence on all of these themes. Famously, Charles Tilly linked wars with
state-building: “War made the state, and the state made war.” 70 The mechanism underlying this
relationship had to do with taxes. Historically, wars created revenue needs, which led to increased
extraction that would respond to those revenue needs.71 Extractive capacity then contributed to state
capacity more broadly since it created the need for tax collection agencies, courts, and police that
could enforce extraction policy. 72
If wars created states, then large-scale wars created particularly capacious states. The typical
causal story underlying state-building suggests that “large effects are the product only of large
causes”73 and that those effects from large causes become quasi-permanent. Once governments have
invested large amounts of resources into tax collection apparatus, they are less likely to dismantle the
enterprise, with the observable implication being that tax collection remains high and state capacity
does not return to its antebellum levels.74 The consequence is a so-called ratchet effect.75
Certain aspects of the US experience would be quite compatible with the large effects, large causes
theory of state-building. As suggested in Figure 2.3, extraction increased significantly during the Civil
War, World War I, and World War II, and the consequences of World War II were significant enough
to have represented a ratchet effect in which peacetime tax levels never returned to prewar levels.
By focusing primarily on structure (war), the ratchet effect theory of state-building captures well
the postwar permanence of the state. By leaving out agency, however—in particular, the role of